No. 211

ADVISORY COMMITTEE ON PESTICIDES

FOOD AND ENVIRONMENT PROTECTION ACT 1985, PART III

Control of Pesticides Regulations 1986

Evaluation of Fully Approved or Provisionally Approved Products

DICHLORVOS (2): PARTIAL REVIEW AS A PUBLIC HYGIENE INSECTICIDE

APRIL 2004

Prepared by : The Health and Safety Executive Biocides & Pesticides Assessment Unit Magdalen House Stanley Precinct Bootle Merseyside L20 3QZ Available from: Department for Environment, Food and Rural Affairs Pesticides Safety Directorate Mallard House Kings Pool 3 Peasholme Green York YO1 7PX a

ii

IMPORTANT NOTICE TO RECIPIENTS OF EVALUATION DOCUMENTS

PUBLISHED BY THE ADVISORY COMMITTEE ON PESTICIDES In making this information available, the Advisory Committee on Pesticides (ACP) Secretariat is obliged to consider its commercial value. You are therefore required, under regulation 8(4) of the Control of Pesticides Regulations 1986, as amended by the Control of Pesticides (Amendment) Regulations 1997, not to make commercial use of the contents of this publication nor publish any part unless authorised. Should you wish to publish all or part of this document, written requests for authorisation should be addressed to:

ACP Secretariat, Room 202, Mallard House,

3 Peasholme Green Kings Pool

YORK YO1 7PX

The Control of Pesticides Regulations 1986 (SI 1986/1510) and the Control of Pesticides (Amendment) Regulations 1997 (SI 1997/188) are available from the Stationery Office. Public access to raw data underlying these publications can be arranged by contacting:

The Biocides and Pesticides Unit HSE

Magdalen House Stanley Precinct

Bootle Merseyside L20 3QZ

NB: This document reflects the outcome of the 283rd, 284th, 285th and the 290th Advisory Committee on Pesticides meetings held in April , May, July 2001 and March 2002 respectively.

iii

CONTENTS PAGE ABBREVIATIONS iv OVERALL SUMMARY vii 1 INTRODUCTION AND REGISTRATION HISTORY 1 2 PHYSICAL CHEMISTRY 5 3 MAMMALIAN TOXICOKINETICS AND TOXICOLOGY 11 4 OPERATOR AND CONSUMER EXPOSURE AND RISK 112 ASSESSMENTS 5 EFFICACY 146 6 OUTCOME OF THE REVIEW AND DATA REQUIREMENTS 257 7 REFERENCES 263 APPENDICES 292

iv

ABBREVIATIONS ACP Advisory Committee on Pesticides AChE acetylcholinesterase ACGIH American Conference of Governmental Industrial Hygienists ai active ingredient ALP alkaline phosphatase ALT alanine aminotransferase API annual parasite incidence AST serum aspartate aminotransferase BPAU Biocides and Pesticides Assessment Unit bw body weight BS British Standard BSI British Standards Institution 14C radioactive carbon 14

oC degrees centigrade CAS Chemical Abstract Service ChE cholinesterase CHL Chinese hamster lung CHO Chinese hamster ovary Ci Curie CI confidence interval cm centimetre(s) CO2 carbon dioxide COM Committee on Mutagenicity cpm counts per minute CTH controlled temperature and humidity d day(s) DNA Deoxyribonucleic acid DRs data requirements EI electron impact EINECS European Inventory of Existing Commercial Chemical Substances EU European union FAO Food and Agricultural Organisation FID Flame Ionisation Detector FIFRA Federal Insecticide, Fungicide and Rodenticide Act (US legislation) FOB Functional Observation Battery g gram(s) GC gas chromatography GLP Good Laboratory Practice h hour(s) HCT haematocrit HPLC high-performance liquid chromatography ID Inner diameter i.m. intra muscular IMS intermediate syndrome IR infrared IUPAC International Union of Pure and Applied Chemistry i.v. intra venous

v

KD50 concentration required to knock down 50 % of test population KD95 concentration required to knock down 95 % of test population KT50 time required for a given concentration to knock down 50 % of test population KT90 time required for a given concentration to knock down 90 % of test population kg kilogramme(s) l litre(s) LC50 concentration required to kill 50 % of test population LD50 median lethal dose LOQ limit of quantitation m metre(s) max maximum µg microgram(s) µl microlitre(s) µm micron(s) mg milligramme(s) min minute(s) ml millilitre(s) mN millinewton MS Mass Spectra MTD maximum tolerated dose NMR Nuclear Magnetic Resonance No. number NPIS National Poisons Information Service NOAEC no observed adverse effect concentration NOAEL no observed adverse effect level NOEC no observed effect concentration NOEL no observed effect level OECD Organisation for Economic Co-operation and Development 32P radioactive phosphorous

Pa pascal(s) Pow partition coefficient between octanol and water ppm parts per million PVC polyvinylchloride QA Quality Assurance Pow partition coefficient between octanol and water RBC red blood cell r.h. relative humidity s second(s) s.c. sub cutaneous SCE Sister chromatid exchange STP standard temperature and pressure SVC saturated vapour concentration SVR strip:volume ratio t½ half-life TCD Thermal Conductivity Detection TLC thin layer chromatography TOTP tri-ortho-tolyl phosphate

vi

TWA Time Weighted Average UDS unscheduled DNA synthesis US EPA Environment Protection Agency (USA) UK United Kingdom UV ultraviolet v/v volume/volume w week(s) WHO World Health Organisation w/w weight for weight

vii

OVERALL SUMMARY

INTRODUCTION AND BACKGROUND Dichlorvos is an organophosphorus compound currently approved for use as an insecticide against crawling and flying insects in non-agricultural pesticides. The use of dichlorvos in public hygiene and amateur insecticides was previously reviewed by the Advisory Committee on Pesticides (ACP) in 1994 (see ACP Evaluation No. 120). Following the 1994 review, approval for a number of uses of dichlorvos was removed. Data requirements and some restrictions of use were set for those products for which approval was allowed to continue. The recent review took place because dichlorvos is one of the chemicals included in the 1998 review of organophosphorus and carbamate compounds. Three suppliers and 12 approval holders submitted data on the active substance and approved products to the review. This document presents a review of the physical chemistry, mammalian toxicity and efficacy of dichlorvos. Data on the genotoxicity and carcinogenicity of dichlorvos were also reviewed by the the Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM) and relevant members of the Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC) and their opinions have been included in this document. The document also includes assessments of the risks to human health for both amateur and professional users and to consumers present during treatment or entering treated areas. The data submitted in response to the data requirements set in 1994 have been evaluated and incorporated into this paper. It should be noted that this document does not address the environmental fate and behaviour or the ecotoxicology of dichlorvos nor provide an environmental risk assessment. PHYSICOCHEMICAL PROPERTIES OF DICHLORVOS Dichlorvos is the BSI common name for 2,2-dichlorovinyl dimethyl phosphate (IUPAC). It is a colourless to yellow liquid with a molecular mass of 220.98, freezing point of less than -15 °C, relative density of 1.42 - 1.43, vapour pressure 1.44 - 1.6 (at 20 °C) and decomposes on heating then boils at 234 oC (at reduced pressure 133.3 - 399.9 Pa, boiling point of 87°C was observed). Dichlorvos has a water solubility of greater then 15 g l -1, log Pow 1.5 - 1.9 (at 25 °C) and a surface tension of 59.6 - 63.5 mN m-1 (at 1 - 1.012 g l -1). The flash point is 172 oC. Depending on the manufacturer, technical dichlorvos has a minimum purity of 95.8 - 97 % w/w. Analytical methods have been provided for the determination of dichlorvos in technical material and formulations. Storage stability studies on the following types of formulation were submitted: pre-pressurised aerosol, cassette and resin strip. MAMMALIAN TOXICOKINETICS The toxicokinetic profile of dichlorvos has been thoroughly investigated in experimental animals (rats, mice, hamsters, rabbits) following oral exposure; a well-conducted dermal study is available while more limited data is available following inhalation exposure. Following oral exposure dichlorvos is rapidly and completely absorbed. Absorption is not as great following dermal exposure (approximately 10 % of the applied dose). Following inhalation exposure absorption certainly occurs; the extent is unclear but is expected to be extensive (given the good absorption from the GI tract).

viii

Once absorbed dichlorvos and/or its metabolites are rapidly and widely distributed around the body in all species studied (rat, mouse and hamster). Metabolism is both rapid and complete, and appears to be qualitatively similar in all species studied. Dichlorvos appears to undergo transformation via glutathione conjugation or cleavage mediated by cytochrome P450 enzymes. It appears that metabolism leads to a deactivation of the dichlorvos molecule, with evidence that dichlorvos metabolites are eventually broken down and incorporated into endogenous amino acids and proteins. Excretion of dichlorvos metabolites is rapid across all species tested. Radioactivity is mainly excreted as CO2 in expired air (up to 50 %), or excreted in urine (up to 25 %) within 24 h of exposure. Following dermal exposure, a significant amount of the dichlorvos that passes through the skin appears to remain 'fixed' there. Dichlorvos and/or its metabolites can cross the placental barrier and have been detected in fetal blood following maternal exposure. There is no data available to indicate whether or not dichlorvos can appear in breast milk. Human data are limited. In one human volunteer study, 27 % of the administered dose was recovered in expired air (up to 8 h) and 9 % in urine (up to 48 h). The urinary metabolites identified in this study were qualitatively similar to those seen in animal studies. Post-mortem examination of a suicide indicated that dichlorvos was widely distributed. Thus from the very limited information available, the toxicokinetics of dichlorvos in humans appeared qualitatively similar to that seen in animals. MAMMALIAN TOXICOLOGY Acute Toxicity Reliable data in humans are available following acute oral exposure. Two modern volunteer studies are reported in which a single oral dose of up to 1 mg kg-1 dichlorvos did not significantly inhibit erythrocyte cholinesterase activity to values > 20 % of those found in baseline values, and no other clinical signs of toxicity were reported. In less reliable reports, fatalities have been reported following attempted suicide by oral ingestion of large quantities of dichlorvos. In no case was it possible to accurately estimate the amount of dichlorvos ingested. Commonly observed clinical signs (and typical of OP poisoning), consisting of salivation, excessive lacrimation, bronchial secretion, pulmonary oedema, breathing difficulties leading to respiratory failure and coma, were seen both in survivors and before death. Delayed neuropathy is reported in humans following poisoning with dichlorvos; however, the reliability of these studies is uncertain. No reports of dermal exposure to dichlorvos in the absence of solvents were available. No reliable inhalation studies in humans are available from which a NOAEC or LOAEC could be derived. Information is available from reliable animal studies involving exposure to dichlorvos via oral, dermal, and inhalation routes. Dichlorvos is toxic following oral exposure with rat LD50 values of between 46.4 - 105 mg kg-1, which indicate that EC classification criteria for

ix

Toxic if swallowed (25 - 200 mg kg-1) is fulfilled. Oral LD50 values of 87 - 184 mg kg-1 and 22.5 - 74 mg kg-1 in mice and rabbits, respectively, have also been reported. Rat dermal LD50 values of between 210 - 456 mg kg-1, indicate the EC classification criteria for Toxic in contact with skin (50 - 400 mg kg-1) is fulfilled. Two well-conducted OECD guideline inhalation studies in rats indicate that aerosols of dichlorvos are severely toxic with LC50 values of 230 and 485 mg m-3 reported. Given the lower value, the EC classification criteria for Very Toxic by inhalation for aerosols (≤ 250 mg m-3) is fulfilled. In one of these studies a NOAEC of 28 mg m-3 following a 4 h exposure was identified. Signs of toxicity associated with cholinesterase inhibition were apparent generally within 1 h of exposure, though there was some evidence of recovery. There is a report of a single rabbit death following dichlorvos administration to the eye; however, this was not repeated in a further eye irritation study performed to an accepted protocol. Significant reduction of brain acetylcholinesterase activity in rats was rapid, falling to 10 % of control activity within 15 min following oral administration of 50 mg kg-1. In one study no effect was seen at 8 mg kg-1, however, erythrocyte acetylcholinesterase activity was not measured so it is unclear whether this is a true NOAEL. Other studies have reported reductions in erythrocyte acetylcholinesterase activity (> 20 %) in rats and dogs at approximately 40 mg kg-1 and in monkeys at 80 mg kg-1 following oral administration. NOAELs could not be derived from animal studies following exposure via the dermal and inhalation routes due either to toxicity or no measurement of acetylcholinesterase activity (the most relevant/sensitive toxic sign of dichlorvos exposure). A well-conducted, specialised behavioural neurotoxicity study is available in the rat. A NOAEL of 0.5 mg kg-1 was apparent following acute administration based on signs of toxicity at higher dose levels (35 mg kg-1 and above); however, no acetylcholinesterase measurements were made, so no conclusions about the NOAEL for systemic toxicity could be drawn. No histopathological damage was observed. The potential of dichlorvos to cause delayed neuropathy has been investigated in the adult hen. There was no evidence of delayed neuropathy following a single gavage administration at a toxic dose level. In contrast older, non-standard acute studies in hens administered dichlorvos via the sc route have reported delayed neuropathy. However, given the data from the more robust modern study using the oral route, this finding is considered unreliable. A formulation of dichlorvos, insecticide resin strips (18.6 % dichlorvos) is reported to have a rat oral LD50 of 382 - 679 mg kg-1 and a dermal LD50 of 27400 mg kg-1. A rat LC50 value of 0.13 - 0.22 mg l-1 has been reported for Formulation A (50 % dichlorvos). There are no reports of skin or eye irritation in humans that can be attributed to exposure to dichlorvos alone. Skin irritation studies performed to accepted protocols indicate that dichlorvos does not meet the criteria for classification as a skin irritant. A well-conducted eye irritation study indicated that dichlorvos does not meet the criteria for classification as an eye irritant. No clear evidence is available of the potential of dichlorvos to produce skin sensitisation in humans (only one case report and instances of dermatitis in workers from which no firm conclusions could be drawn). There are no good animal skin sensitisation studies available. However, when the available human and animal evidence was deliberated by the EC Pesticide Classification and Labelling Working Group, dichlorvos was considered to match the criteria for classification as a skin sensitiser.

x

No clear evidence is available of the potential of dichlorvos to induce asthma. Only one case showing exposure conditions and symptoms consistent with Reactive Airways Dysfunction Syndrome has been reported. There is no data to show that dichlorvos can induce respiratory sensitisation in animals. Repeat Dose Toxicity The toxicity of dichlorvos following repeated oral exposure has been well studied, with two studies available in rats and dogs consistent with current EC guidelines. More limited information is available following repeated dermal and inhalation exposures. Two well-conducted 13-week gavage study in rats and a 52-week oral study in dogs report NOAELs of 0.1 mg kg-1 d-1 and 0.05 mg kg-1 d-1, respectively, based on significant (> 20 %) inhibition of erythrocyte acetylcholinesterase activity at higher dose levels. In an older 2-year feeding study in rats the NOAEL was considered to be 4.7 ppm (equivalent to 0.6 mg kg-1 d-1), based on a reduction in brain cholinesterase activity at higher doses. However the reliability of this result is uncertain given the results from the 13-week studies. An older 2-year dietary study in dogs a NOAEL of 0.008 mg kg-1 d-1 was derived based on decreased erythrocyte cholinesterase activity at higher doses. The potential of dichlorvos to cause delayed neuropathy has been investigated in the adult hen. Repeated gavage administration for 28 d at a dose level causing death in some birds failed to demonstrate delayed neuropathy. Reliable human volunteer studies are available following oral exposure while less robust data is available following inhalation exposure and no data is available following dermal exposure. Well-conducted, modern ethical human volunteer studies revealed toxicologically- significant reductions in erythrocyte acetylcholinesterase activity in 2 of 6 volunteers following oral exposure to 0.1 mg kg-1 d-1 dichlorvos for up to 21 d. An older study involving volunteer male prisoners indicated a NOAEL of 0.025 mg kg-1 d-1 following 28-day exposure, while no effect on erythrocyte acetylcholinesterase activity was observed in male prisoners following administration of 0.025 mg kg-1 d-1 for 60 d. An older inhalation study reports no effect on erythrocyte acetylcholinesterase activity following a 4 d exposure to between 0.9 - 1.22 mg m-3 for 5 - 7.5 h d-1 (equivalent to 0.09 - 0.19 mg kg-1 d-1). No adequate studies using dermal application were available. However, in a single non-standard study, a reduction in brain acetylcholinesterase activity was seen in the rat at 2.9 mg kg-1 d-1 (40 applications over 117 d); however, it is unclear whether this reduction was toxicologically significant. In a rat 2-year inhalation, virtually continuous (23 h d-1), exposure study the NOAEC was 0.05 mg m-3 (equivalent to 0.016 mg kg-1 d-1) based on brain acetylcholinesterase inhibition, with no other significant findings. A NOAEC of 0.11 mg m-3 (equivalent to 0.013 mg kg-1 d-1) was observed in rats following exposure for 4 h d-1 for 120 d, based on decreased brain and erythrocyte cholinesterase activities at higher dose levels. A NOAEC of 0.14 mg m-3 (equivalent to 0.19 mg kg-1 d-1) was observed in mice following exposure for 23 h d-1 for 5 d, based on decreased brain and erythrocyte cholinesterase activities at higher dose levels. Genotoxicity

xi

Genotoxicity data were reviewed by the Department of Health Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM). The following assessment of the mutagenic potential of dichlorvos in vitro was agreed: ‘Members agreed that dichlorvos is a weak methylating agent (compared to methyl methanesulphonate (MMS)). The COM concurred with the following assessment of the in vitro mutagenicity studies. i) Dichlorvos is mutagenic, both in the presence and absence of exogenous metabolism, to bacteria, yeast cells and in mammalian cell gene mutation assays, chromosome aberrations assay, the in-vitro micronucleus test and sister chromatid exchange assays. ii) Positive results have been reported in in vitro UDS assays using human lymphocytes and human epithelial-like cells. iii). Dichlorvos has been shown to methylate nucleophiles and to induce strand breaks in isolated DNA. Members agreed that DNA methylation induced by dichlorvos contributed towards the mutagenicity reported in in-vitro test systems but noted that other mechanisms might also be involved. Members considered that the positive results obtained in in-vitro mutagenicity tests with dichlorvos in the presence of an exogenous metabolising fraction and in the assay for single strand breakage of DNA also suggested that dichlorvos and/or its metabolites were genotoxic. This might include dichloroacetaldehyde although the available evidence was insufficient to identify all potential mutagenic metabolites of dichlorvos. The COM noted that there were a large number of in-vivo studies available. Dichlorvos was negative in most published in-vivo mutagenicity assays where it was administered as a single dose. These included mouse bone-marrow micronucleus (using the ip route) and bone marrow chromosome aberration studies in mice and hamsters using oral and, in two studies (mice/hamster) inhalation exposure. Negative results were also reported in SCE in mice and UDS assays (liver (rats)/forestomach (mice)). A negative result was also reported in an adequately conducted bone-marrow chromosome aberration study where mice were given daily oral doses of dichlorvos by gavage for 5 days. Studies in germ cells have given negative results including dominant lethal assays and a chromosome aberration study in mouse spermatogonial cells. The COM also noted that there were a number of positive studies. The COM concluded that a consistent pattern of mutagenic effects had been documented in the in-vivo studies in which dichlorvos induced mutagenic effects at high doses in the skin following topical application, and in the liver following repeated intraperitoneal dosing, suggesting a potential site-of-contact effects (i.e. at initial sites of exposure). Carcinogenicity Of the rat studies available only two (one by gavage, one inhalation) are considered to be adequate for a proper assessment of the carcinogenicity of dichlorvos. No evidence of carcinogenicity was seen in the inhalation study in CFE rats. In the gavage study with F344 rats using corn oil as vehicle the increased incidence of mononuclear cell leukaemias (males)

xii

and mammary gland tumours (females) were within the historical control range and not considered to be treatment related. The pancreatic lesions (acinar cell hyperplasia and adenoma) may be related to the use of corn oil as vehicle. An additional adequate drinking water study provided no evidence of treatment-related tumour formation in F344 rats. Four studies in mice are available but only one is considered to be entirely adequate for the assessment of carcinogenicity. In this study, in B6C3F1 mice, there was an increase in squamous cell papillomas of the forestomach in males and females (significant in females only) and carcinomas of the forestomach in females only. In a second study, in which historical control data were not available, squamous cell carcinoma of the oesophagus was seen in one low dose male and one high dose female and an oesophageal papilloma in a high dose female. Relevant members of the Department of Health Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC) were asked by the COM to assess the carcinogenicity studies sumitted for this review. Regarding the studies in rats the COC members concluded that there were limitations in most of the studies (e.g. age of study, numbers of animals used, extent of pathology investigations) and that there was no clear evidence for a carcinogenic effect in rats. Regarding other studies in mice, COC members considered that there were limitations in the conduct of these studies similar to those undertaken in the rat. Members considered that it was not possible to undertake a comparison of the studies in mice where dichlorvos had been administered in corn oil and those where dichlorvos had been administered in the drinking water or as an aqueous solution by gavage. However, COC members reaffirmed, that there was limited evidence for an effect on squamous epithelium of the forestomach and oesophagus in mice. On considering the overall package of carcinogenicity bioassays COC members felt that there was no consistent evidence for a genotoxic carcinogenic effect. COC members noted that there was no agreed mechanism for the forestomach tumours. COM Conclusion On The Genotoxicity And Carcinogenicity Of Dichlorvos The COM conclusion on the genotoxicity and carcinogenicity of dichlorvos is reproduced below. The full COM statement on the genotoxicity and carcinogenicity of dichlorvos can be found at http://www.advisorybodies.doh.gov.uk/Com/dichlorvos.htm ‘The COM agreed that there is clear unequivocal evidence that dichlorvos can induce DNA damage, chromosomal breakage and mutations in mammalian cells from in vitro studies. The compound has been shown to interact with DNA via methylation, however several other mechanisms are theoretically possible. In-vivo, dichlorvos can be rapidly detoxified by hydrolysis before it reaches the systemic circulation. Members noted from the HSE review [undertaken by HSE on behalf of the ACP] that retention of 14C-vinyl-labelled dichlorvos in skin was recorded in a study where radiolabelled dichlorvos was applied to the skin on the backs of male rats. Several non-standard in-vivo mutagenicity assays have indicated that dichlorvos can induce genetic damage when systemic detoxification mechanisms are bypassed, e.g. following exposure to the skin and exposure to the liver following intraperitoneal dosing. The COM agreed that there was a potential risk of mutagenicity at site of contact tissues, i.e. at the initial sites of exposure. The COM felt there was no evidence for systemic mutagenic effects. The COM agreed that until evidence was provided to the contrary and in the absence of appropriate mechanistic data, a precautionary approach

xiii

should be adopted and no threshold could be assumed for the mutagenic activity of dichlorvos. Members were aware that there was some limited evidence for a carcinogenic effect in mice from standard bioassays. This related to an increase in squamous cell papillomas of the forestomach in mice and carcinomas of the forestomach in female mice given gavage doses of dichlorvos together with the finding of squamous cell papilloma and carcinoma of the oesophagus in a small number of mice. Members noted there was no evidence for carcinogenicity from a number of other carcinogenicity bioassays including an inhalation bioassay in the rat, although there were limitations with all of these studies. Members noted that negative results had been obtained with dichlorvos in a single dose UDS assay in the forestomach of mice using gavage dosing. An increase in replicative DNA synthesis had been reported in this study. Members noted that there were a number of proposals regarding the mechanism of dichlorvos tumourigenicity in the mouse forestomach including localised irritancy of dichlorvos in corn oil. The COM agreed that this proposal had not been proven and considered that it was not possible to exclude a genotoxic effect from these data, given the relative insensitivity of the method used as indicated by the response with the positive control chemical; they felt that repeat dosing would most likely be required to identify any mutagenic effect of dichlorvos in this assay. The COM concluded that dichlorvos should be regarded as an in vivo mutagen at the site-of-contact (i.e. at the initial sites of exposure). The COM considered there was no evidence for systemic mutagenic effects. High doses of dichlorvos induced mutagenic effects in the skin following topical application and in the liver following intraperitoneal dosing. The COM noted the limited evidence for a carcinogenic effect of dichlorvos. This related to tumours of the forestomach in mice after gavage dosing and also the oesophageal tumours seen after dietary administration. There was no satisfactory explanation proven for the mechanisms of these tumours and the COM felt, given the available mutagenicity data on dichlorvos, that it would be prudent to assume a genotoxic mechanism. The COM agreed that in the absence of appropriate mechanistic data, a precautionary approach should be adopted and no threshold could be assumed for the mutagenic and carcinogenic effects of dichlorvos.’ Reproductive Toxicity No data in humans are available. A well-conducted 3-generation rat drinking water study is available. Although the standard parameters of fertility remained unaffected by treatment, additional analyses carried out during the study indicated that female animals of the 80 ppm group (7 - 17 mg kg-1 d-1) developed treatment-related abnormalities in oestrous cyclicity. Thus, a NOAEL for effects on fertility of 20 ppm (1.9 - 4.6 mg kg-1 d-1) was apparent. The overall NOAEL for adult toxicity was 5 ppm (0.5 - 1.2 mg kg-1 d-1) based on significant (> 20 %) decreases in acetylcholinesterase activity (brain and erythrocyte). This finding is consistent with the results of a study, which primarily investigated electrical changes in the nervous system in 3 generations of rats following dichlorvos exposure, in which no effects on fertility were observed at the top dose used of 3.92 mg kg-1 d-1. A NOAEL for adult toxicity could not be established as erythrocyte acetylcholinesterase activity was not measured. Few conclusions can be drawn from other less robust studies.

xiv

Two well-conducted gavage studies are available in rats and rabbits. Both studies show no developmental toxicity even at levels, which induce a significant level of maternal toxicity. No effects on development were reported at a dose level of 21 mg kg-1 d-1 (the top dose level in the study) in the rat study for developmental effects, while the NOAEL for maternal toxicity was 3 mg kg-1 d-1 based on tremors and significantly reduced bodyweight gain reported in top dose animals (consistent with dichlorvos acting as an acetylcholinesterase inhibitor). No effects on development were reported at a top dose level of 7 mg kg-1 d-1 in a rabbit study, while the NOAEL for maternal toxicity was 0.1 mg kg-1 d-1, based on deaths and significantly reduced bodyweight gain at 2.5 mg kg-1 d-1 and above. Few conclusions can be drawn from other less-robust studies in different species and by different routes of exposure suggesting that dichlorvos is not a developmental toxicant. Overall, these results indicate that dichlorvos is not a reproductive toxicant. OPERATOR AND CONSUMER EXPOSURE AND RISK ASSESSMENTS Risk assessments have been presented for both professionals and amateur users and for consumers exposed to aerosols, slow release strips and slow release controllable and non-controllable cassettes containing dichlorvos. The TER values for primary exposure and acute secondary exposure to aerosols are unacceptable for both professional and amateur users. It is considered unlikely that protective clothing would reduce the systemic exposure of users of these products to an acceptable level. The TER values for primary exposures and acute secondary exposures from museum uses of slow release controllable and non-controllable cassettes and slow release strips are acceptable. However, the selected reference scenarios for chronic exposures from museum uses of these products are low. However, it is possible that ambient air levels may differ in situations where the number of cassettes deployed and building design differ significantly. Therefore it is proposed that the use of slow release cassettes and strips in museum display cases, specimen cabinets and storage cupboards be assessed on a case-by-case basis on the provision of air level data. TER values for the use of slow release strips in pheromone traps are considered acceptable, however there are concerns about their use in the presence of food. The TER values for chronic secondary exposures from residential uses of slow release controllable and non-controllable cassettes are unacceptable. Further data to refine exposure is not considered appropriate. EFFICACY The data in this review includes both supporting data demonstrating the innate insecticidal efficacy of dichlorvos as an active ingredient, and laboratory, simulated use and field data on the efficacy of both strip products and pre-pressurised handheld aerosols and other spray products. The evaluation includes both data held by individual companies and submitted in support of the review, and data present in the public domain.

xv

Due to the long established use of dichlorvos as an insecticide, a relatively large body of data exists in the public domain. However, many of these data relate to early experiments and sometimes lack details such as the exact formulations, controls and methods of assessment employed, and, for strip products, the type of product (e.g. controllable or non-controllable cassette). Innate Data Although, in the main, the results of the simple screening studies were not always well reported or documented, these data demonstrated the insecticidal activity of dichlorvos when applied topically, as a vapour, or by direct contact against a wide spectrum of test insects from the orders Diptera, Dictyoptera, Hemiptera, Lepidoptera, Coleoptera and Siphonaptera. Slow Release Strips Air Monitoring These data demonstrate that standard strip products/cassette units, when used in domestic premises, produced typical maximum air concentrations of approximately 0.02 - 0.06 mg m-3 dichlorvos for approximately 16 weeks, which is the typical lifetime of a product. As the air concentrations are dependent on the environmental conditions and the air sampling/measurement techniques used in the different test studies, the air concentration values are indicators only of the likely concentrations to be found in practice. Diptera, Dictyoptera and Hemiptera Data on the use of strips against Diptera, Dictyoptera and Hemiptera were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support the use of slow release strips at a strip:volume ratio (SVR) of 0.49-0.80 g m-3 against Diptera. Further data have now been evaluated, and these data support the previous recommendation. Lepidoptera Simulated use and field data were evaluated on the use of strips against Lepidoptera. With the exception of one study, these data provided evidence that strips with an SVR of 0.46 g m-3 are effective against Lepidoptera. The data were supported by the air monitoring results, which indicated that when dichlorvos strips with SVR’s of 0.46-0.49 g m-3 are used in domestic premises in the UK, they typically produce air concentrations shown in the field to be effective against Lepidoptera. Coleoptera Although data have been presented which provide some evidence that dichlorvos strips can be effective against Coleoptera, these data are limited. Consequently, further data are required to fully support a label claim against Coleoptera. Formicoid Hymenoptera

xvi

Although data have been presented which provide some evidence that dichlorvos strips can be effective against Formicoid Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Formicoid Hymenoptera. Hymenoptera Although data have been presented which provide some evidence that dichlorvos strips can be effective against Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Hymenoptera. Niche market use-Lepidoptera and Coleoptera Data on the niche market use of strips against Lepidoptera and Coleoptera were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support this use at an SVR of approximately 6.0 g m-3. No further data have been submitted. Pre-Pressurised Handheld Aerosols (Dichlorvos Only) Data were presented to demonstrate that dichlorvos, when formulated as the sole insecticide in aerosol products, has insecticidal activity against a wide spectrum of test insects from the orders Diptera, Dictyoptera, Coleoptera and Lepidoptera. These data are considered by HSE as useful supporting information. Pre-Pressurised Handheld Aerosols (Dichlorvos and Pyrethrins/Pyrethroids) Diptera Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Diptera, these data are limited. Consequently, further data are required to fully support a label claim against Diptera. Dictyoptera Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Dictyoptera, these data are limited. Consequently, further data are required to fully support a label claim against Dictyoptera. Coleoptera Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Coleoptera, these data are limited. Consequently, further data are required to fully support a label claim against Coleoptera. Hymenoptera Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Hymenoptera.

xvii

OUTCOME OF THE REVIEW At its meetings in April and May 2001, the ACP considered the available physicochemical, toxicological and efficacy data on dichlorvos including reasoned arguments from a number of approval holders and suppliers regarding the interpretation of data. Based on its consideration of anticholinesterase effects, the ACP recommended in May 2001 that Ministers be advised that:

a) approval for all aerosols containing dichlorvos be revoked based on the unacceptable toxicity-exposure ratios (TERs) derived for primary and secondary exposures from professional and amateur use; b) approval for residential uses of slow release controllable and noncontrollable cassettes containing dichlorvos be revoked based on the unacceptable TERs derived for secondary exposures from professional and amateur use; c) approval for the professional use of slow-release strips and controllable and non-controllable cassettes containing dichlorvos in museums be retained subject to the fulfilment of physicochemical, operator exposure and efficacy data requirements (detailed in Section 6); d) approval for the use of slow-release strips in pheromone traps in areas where food may be stored, prepared or consumed be suspended pending the provision of information on food residues; and e) approval for the use of slow release strips in pheromone traps in areas where food is not present be retained subject to the fulfilment of physicochemical and efficacy data requirements (detailed in Section 6).

A review of the agricultural uses of dichlorvos was also considered by the ACP at the same time (ACP evaluation in preparation). The ACP also noted that dichlorvos was under discussion by the Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM) and relevant members of the Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC), and recognised that their recommendations would perhaps require modification after consideration of the COM’s findings. At the July 2001 ACP meeting, the Committee considered the implications of the COM’s possible conclusions, and agreed to advise Ministers as follows:

a) If the COM’s final conclusions were that dichlorvos was an in vivo mutagen, and it could not exclude the possibility that the occurrence of tumours in animal tests of carcinogenicity resulted from a genotoxic mechanism, there should be immediate revocation of all uses (both agricultural and non-agricultural). b) Alternatively, if the COM concluded that dichlorvos was an in vivo mutagen, but that the tumours observed in animal tests did not result from a genotoxic mechanism, or if it could not confirm that dichlorvos was an in vivo mutagen, or if it took the view

xviii

that dichlorvos was not an in vivo mutagen, then the ACP’s previous recommendations would be maintained.

The COM concluded, following a second meeting in July 2001, that dichlorvos should be regarded as an in vivo mutagen (i.e. capable of inducing mutations in living animals) at site of contact and that it could not exclude the possibility of it acting as a genotoxic carcinogen. Consequently the ACP recommended to Ministers that it would be prudent to revoke, with immediate effect, all agricultural and non-agricultural uses of dichlorvos. This advice was given as a precautionary measure, since the possibility of genotoxic carcinogenicity could not be excluded. The ACP considered that any risk of human carcinogenicity was likely to be very small, and would be mainly associated with certain uses in the home and with exposures to some operators in the agricultural sector. Ministers decided that in addition to the revocations and suspensions that were proposed because of concerns about possible anticholinesterase effects of dichlorvos, all uses of dichlorvos should be suspended with immediate effect. Before such regulatory action could be carried out, AMVAC Chemical UK Ltd (an approval holder and manufacturer of dichlorvos) obtained an injunction, which prevented regulatory action. Government agencies were also prohibited from making any announcement to the public about the regulatory action that was proposed. AMVAC also gained permission for a judicial review hearing, which was heard in November 2001. The grounds for the challenge were that AMVAC had not been properly informed of the proposed regulatory action or the basis for it, and had not been given sufficient time to make representations. AMVAC also claimed that Ministers had not given proper regard to the precautionary principle and to the European Convention on Human Rights. The judgement of the Court was issued in December 2001. Mr Justice Crane rejected most of the company’s submissions, including those concerning the precautionary principle and the Convention on Human Rights. However, he ruled that the company had been given insufficient time to respond to the conclusions of the Government’s expert advisers prior to regulatory action being taken. He accepted that the matter was urgent but considered that the claimant had now had full opportunity to present any further material. During the period of the injunction, the ACP was unable to publish the minutes of its meetings. The ACP had concerns that this compromised the openness of the advice given to Ministers, and could thereby have an adverse effect on public confidence in the regulatory process. It was also concerned that speculation about the missing minutes might create unwarranted public anxiety. Notwithstanding these concerns, the ACP agreed that while rapid implementation of regulatory action was desirable once decisions had been made, it was also important that the regulatory process be fair and open to scrutiny. In this case, the ACP’s advice to Ministers had been precautionary (i.e. based on insufficient reassurance that exposures to the compound were acceptable rather than direct evidence that people were being harmed), and the delay caused by the legal action would be acceptable provided that it was not unduly prolonged. 59 Following the Court judgement, approval holders for both agricultural and non-agricultural products were asked to provide any further data relating to the potential genotoxic carcinogenicity of dichlorvos. These data were considered by the COM, relevant members of

xix

the COC and the ACP in February and March 2002. The COM considered that there was insufficient evidence to change their previous opinion i.e. that dichlorvos should be regarded as an in vivo mutagen (i.e. capable of inducing mutations in living animals) at site of contact and that it could not exclude the possibility of it acting as a genotoxic carcinogen (The full COM statement regarding the genotoxicity and carcinogenicity of dichlorvos can be found at http://www.advisorybodies.doh.gov.uk/Com/dichlorvos.htm). As the ACP considered that it could not satisfactorily rule out the possibility that dichlorvos is a genotoxic carcinogen (i.e. causes cancer by damaging DNA), it advised Ministers that there could be a small risk of adverse health effects following prolonged exposure to dichlorvos. Ministers decided as a precautionary measure that in addition to the revocations and suspensions that were proposed because of concerns about possible anticholinesterase effects of dichlorvos, the advertisement, sale and supply of all remaining products should be suspended with immediate effect until further studies were conducted to address the mechanisms of tumour formation in mice.

1

1 INTRODUCTION AND REGISTRATION HISTORY

1.1 INTRODUCTION AND BACKGROUND TO THE REVIEW

Dichlorvos, O-(2,2-dichlorovinyl)-O,O-dimethylphosphate, is an organophosphorus compound currently approved for use as an insecticide against crawling and flying insects in non-agricultural pesticides. It is one of the chemicals included in the current review of organophosphorous and carbamate compounds. The use of dichlorvos in public hygiene and amateur insecticides was previously reviewed by the Advisory Committee on Pesticides (ACP) in 1994 (ACP Evaluation 120 ). Following the 1994 review, approval for a number of uses of dichlorvos was removed. Data requirements and some restrictions of use were set for those products for which approval was allowed to continue. This document presents a review of the physical chemistry, mammalian toxicity and efficacy of dichlorvos. Data on the genotoxicity and carcinogenicity of dichlorvos were also reviewed by the the Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM) and relevant members of the Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC) and their opinions have been included in this document. The document also includes assessments of the risks to human health for both amateur and professional users and to consumers present during treatment or entering treated areas. The data submitted in response to the data requirements set in 1994 has been evaluated and incorporated into this paper. It should be noted that this document does not address the environmental fate and behaviour or the ecotoxicology of dichlorvos nor provide an environmental risk assessment.

1.2 REGISTRATION HISTORY OF DICHLORVOS

The full registration history of dichlorvos is detailed in Table 1.1.

Table 1.1 Registration history of dichlorvos Date Outcome 1962 Dichlorvos was first introduced into the UK as an agricultural pesticide. 1976 Provisional commercial clearance of dichlorvos as a non-agricultural pesticide was

granted under the Pesticides Safety Precautions Scheme (PSPS) in 1976. Clearances were granted for hand-held aerosols containing up to 0.5 % w/w dichlorvos and impregnated resin strips containing not more than 20 % dichlorvos for use in food storage practice and the home kitchen and larder and for controllable and non-controllable slow release units (cassettes) containing up to 20 % w/w dichlorvos for use against woodworm in roof spaces.

1977 An extension of use as a pesticide to household aerosols (up to 0.8 % w/w) for home uses other than the kitchen and larder to control nuisance insect pests was granted. This was extended to cover use of products containing up to 0.8 % w/w on bedding, clothing and carpets.

2

Date Outcome 1978 It was considered that the use of dichlorvos in concentrates or ready-for-use

formulations (up to 1.0 % w/w) in public hygiene was acceptable and that the use by professional operators of both controllable and non-controllable units emitting dichlorvos at a slow rate of release was acceptable. It was also agreed that slow release units, containing no more than 20 g dichlorvos, could be retailed for the control of nuisance pests in the home and that such units should be controllable unless intended for treatment of 12 m3 or less.

1994 In 1994, the Advisory Committee on Pesticides (ACP) reviewed the use of dichlorvos in public hygiene and amateur insecticides. The ACP recommended that approval for the following dichlorvos products be removed: thermal vaporisers (in food storage practice and use in domestic kitchens), concentrates or ready-for-use formulations (public hygiene), aerosols for use on bedding and clothing and strips as wood preservatives (ACP Evaluation 120). It was also recommended that the application rate for aerosol surface sprays should be limited to spraying in bands 6 - 8 cm wide. In addition, the ACP recommended that, subject to data requirements, approval be allowed to continue for slow release products containing a maximum of 20 g dichlorvos, with an application rate such that the ratio of the amount of dichlorvos to the volume of room to be treated (strip: volume ratio) is between 0.4 - 0.8 g m-3. Niche market uses of slow release strips, not involving exposure of consumers may have a strip:volume ratio of approximately 6 g m-3. Aerosol space sprays for amateur and professional use can contain a maximum of 0.8 % w/w dichlorvos, whilst aerosol surface sprays for amateur and professional use can contain a maximum of 0.5 % w/w dichlorvos. The data requirements recommended by the ACP are listed in Appendix 1.1.

1.3 CURRENT PRODUCTS

Dichlorvos is approved for use by amateur and professional users as a public hygiene insecticide. Examples of public hygiene use include application of pesticides in and around hotel kitchens, restaurants, food retailers’ premises, canteens, food warehouses, mills, slaughterhouses, barns, pet shops, kennels, non food animal bedding, pigeon lofts, and stables. In all these situations there should be no risk of pesticides contaminating foodstuffs. At present, (April, 2001) there are 47 current approvals for products containing dichlorvos. These are summarised by product type in Table 1.2.

Table 1.2 Breakdown of currently approved products containing dichlorvos

User Group Product type Professional Amateur Professional

and AmateurAerosol space spray - 1 2 Aerosol space and surface spray - 2 - Slow release strip-museum uses 1 - - Slow release strip-pheromone trap 2 - - Slow release controllable cassette - 26 - Slow release non-controllable cassette - 13 -

3

One niche market product is approved: a slow release strip used in museum display cases, cabinets and small cupboards. Slow release strips are also used in pheromone traps. Further details on the pattern of use of these products can be found in Section 4. Full details of the current maximum approved levels of dichlorvos in each of the above dichlorvos product types are shown in Table 1.3.

Table 1.3 Maximum approved levels of dichlorvos in current products

Formulation Type User Max. % w/w

Container Volume

Aerosol space spray Amateur Professional

0.8 0.8

600 ml 300 g

Aerosol space and surface spray

Amateur 0.5 300g

Slow release strip Professional Professional

20 20

Museum use: strips containing 14.6 g dichlorvos cut according to volume of display case. Maximum strip: volume ratio 6 g m-3. Pheromone trap: maximum application rate: 0.49 g per 100 m2.

Slow release controllable cassettes

Amateur 80 20 g of dichlorvos per unit

Slow release non-controllable cassettes

Amateur 20 2 g of dichlorvos per unit

The maximum precedent for space and surface sprays is set by the lowest precedent concentration of either the space or surface spray.

If the product is applied as a surface spray it is applied by spraying in bands 6 - 8 cm wide around the perimeters of rooms and soft furnishings and/or by spraying directly onto the insects from a distance of approximately 25 cm. For slow release controllable and non-controllable cassettes the strip: volume ratio should be between 0.4 - 0.8 g dichlorvos m-3 and contain no more than 20 g dichlorvos per strip.

1.4 CLASSIFICATION OF DICHLORVOS

The current EU classification of dichlorvos according to the Dangerous Substances Directive (67/548/EEC; 28th Adaptation to Technical Progress, January 2001) is: VERY TOXIC (T+): R26 VERY TOXIC BY INHALATION TOXIC (T): R24/25 TOXIC IN CONTACT WITH SKIN AND IF SWALLOWED SENSITISING (Xi): R43 MAY CAUSE SENSITISATION BY SKIN CONTACT

4

DANGEROUS TO THE ENVIRONMENT (N): R50 VERY TOXIC TO AQUATIC ORGANISMS This classification is compatible with the recommendations of the ACP following the review of dichlorvos in 1994 (Appendix 1.1). The EU Working Group on Classification, Packaging and Labelling (after reference to the appropriate specialised experts) did not consider that dichlorvos should be classified in any category for either carcinogenicity or mutagenicity. Similarly, when dichlorvos was considered by the ACP in 1994, the Committee concluded that the weight of evidence did not suggest that dichlorvos presented a carcinogenic risk to humans. By contrast, the carcinogenic potential of dichlorvos has been classified as 'suggestive' under the USA Environmental Protection Agency, 1999, Draft Agency Cancer Guidelines based on the increased incidence of mononuclear cell leukaemia in the male F344 rat and forestomach tumours in the mouse. However, it should be noted that the EPA Cancer Assessment Review Committee considered the tumours observed in animal studies might not be applicable to human risk assessment.

1.5 USAGE INFORMATION

As part of the dichlorvos review, Approval Holders were requested to supply usage information for their products. This included information on the number of products sold in the UK per annum, or where this was not appropriate (for example, where a product is sold in packaging of various sizes) the total amount of dichlorvos used in the manufacture of products. Information was also requested regarding the points of sale, retail or wholesale, in order to estimate the scale of use by the public. Information on 15 of the 47 currently approved products was provided. In 1999, products containing at least 19699 kg of dichlorvos (~97 % purity) were sold. It should be noted that some non-agricultural pesticide products containing dichlorvos are exported.

1.6 LABELLING OF DICHLORVOS PRODUCTS

HSE have prepared generic draft labels for all dichlorvos product types. These are presented in Appendix 1.2.

5

2 PHYSICAL CHEMISTRY

2.1 PHYSICAL CHEMISTRY OF DICHLORVOS

Supplier A, Supplier B and Supplier C are manufacturers of dichlorvos used in non-agricultural pesticides. Committees have already seen data presented by Supplier A and Supplier B. However, some of the data provided by Supplier B is new, generated in response to the post-review data requirements set in 1994. The data provided by Supplier C has not been seen by Committees before. The data provided by suppliers are summarised in Table 2.1. Full details of the compositions of the technical material were submitted by the 3 suppliers.

2.1.1 IDENTITY OF THE ACTIVE SUBSTANCE

BSI Name: Dichlorvos IUPAC Name: 2,2-dichlorovinyl dimethylphosphate Other synonyms: O-(2,2-dichlorovinyl) O,O-dimethylphosphate, DDVP,

2,2-dichloroethenyl dimethyl phosphate (CAS) CAS number: 62-73-7 EINECS number: 200-547-7 Molecular formula: C4H7Cl2O4P Structural formula:

H

P

O

O

Cl

Cl OCH3

OCH3

Molecular mass: 220.98

6

Table 2.1 Summary of physical chemistry data Company Supplier A Supplier B Supplier C Site of manufacture Provided Provided Provided Batch purity range (%w/w) 97 - 100

(Unpublished, 1993a) 95.8 - 96.5 97 - 99.8

Impurities (each of max > 0.5 % w/w)

Provided (Unpublished, 1991a)

Provided (Unpublished, 1993i)

Provided (Unpublished, 1997b)

Spectral data submitted: IR UV-Vis 1H-NMR 13C-NMR (additional to DRs) 31P-NMR (additional to DRs)

Provided Provided Provided Provided Not provided (Unpublished, undated a)

Provided Provided Provided Provided Provided (Unpublished, 1993j)

Provided Not provided Not provided Not provided Not provided (Unpublished, 1997c)

Mass Spectrum Provided (Unpublished, undated a)

Not Provided Provided (Unpublished, 1995c)

Appearance at STP Colourless to yellow liquid (Unpublished, 1995a)

Colourless liquid (Unpublished, 1999a)

Colourless to pale yellow liquid (Unpublished, 1988)

Freezing Point Not tested Not tested < -15 °C (Unpublished, 1988)

Boiling Point 234 °C at 101.3 KPa (though signs of decomposition noted) (Unpublished, 1993b) GLP

Some decomposition at 190 °C (Unpublished, 1999b) GLP

87 °C at 133.3 - 399.9 Pa (Unpublished, 1995d)

Vapour Pressure 1.6 Pa (at 20 °C) (Unpublished, 1991b)

4.9 Pa (at 25 °C) (Unpublished, 1997a)

1.44 Pa (at 20 °C) (Unpublished, 1990)

Relative Density 1.43 (Unpublished, 1993c) GLP

1.42 (Unpublished, 1999c) GLP

1.42 (Unpublished, 1995e)

Surface tension of aqueous solution, at 20 oC

59.6 - 60.7 mN m-1 (1 g l-1) (Unpublished, 1995b) GLP

63.5 mN m-1 (1.012 g l-1) (Unpublished, 1999d) GLP

60.4 mN m-1 (1 g l-1) (Unpublished, 1995f) GLP

7

Water solubility 18 g l-1 (at 25 °C) (Unpublished, 1993d) GLP

20 g l-1 (at 20.1 °C) (Unpublished, 1999c)

15.2 g l-1 (at 25 °C) (Unpublished, 1987b)

log Pow 1.9 (at 25 °C) (Unpublished, 1993e) GLP

1.5 (at 25 °C ± 2 °C) (Unpublished, 1999c)

1.58 (at 25 °C) (Unpublished, 1987c)

Flash Point 172 °C (Unpublished, 1995a)

172 °C (Anonymous a, 1997)

80 °C (Unpublished, 1988)

Flammability Not Flammable (based on flash point and boiling point)

Not Flammable (based on flash point and boiling point)

Not Flammable (based on flash point and boiling point)

Explosivity Not explosive (Unpublished, 1995a)

Not provided Not provided

Oxidising Not provided Not oxidising (Unpublished, 1999a)

Not provided

Analytical methods: AI in technical material Impurities in technical material AI in formulation AI in water LOQ of 0.1 µg l-1 in water

Provided (GC-FID) (Unpublished, 1992a; 1993f) Provided (GC-FID, titration) (Unpublished, 1993g; 1993h) Provided (GC-FID) (Unpublished, 1987a) Provided (GC-FID) (Anonymous, 1980) Not provided

Provided (GC-FID) (Unpublished, undated b) Provided (GC-FID) (Unpublished, 1993i) Provided (GC-FID) (Unpublished, undated b) Provided (HPLC-UV) (Unpublished, 1999c) Not provided

Provided (GC-TCD) (Unpublished, 1995g) Provided (GC-TCD or MS, titration) (Unpublished, 1997c) Provided (GC-TCD) (Unpublished, 1993k) Provided (HPLC-UV) (Unpublished, 1995h) Not provided

Storage stability data: AI in formulation

Provided (Unpublished, 1992b)

Not Provided

Provided (Unpublished, undated c)

8

2.2 STORAGE STABILITY

There are currently three types of formulations containing dichlorvos approved for use in non-agricultural pesticides: pre-pressurised aerosols (space and surface), slow release strips and slow release cassettes (controllable and non-controllable). Storage stability studies were submitted on a pre-pressurised aerosol formulation (Unpublished, 1992b), a slow release resin strip (Unpublished, undated c) and a slow release cassette (Unpublished, 1994). The test on the pre-pressurised aerosol showed that there is no significant loss of active ingredient after 8 months at 35 °C and 36 months at room temperature. However details on emission parameters were not provided. The slow release resin strip formulation showed no significant loss of active ingredient following storage for 3 years at ambient temperature and humidity, only small losses were seen after a 5-year period of storage. The strips were stored in their foil packaging. Details on the conditions of the study (such as the ambient temperature) and observations on the strip were not provided. The slow release cassette showed no loss of active ingredient after storage for 6 months at 30 °C when stored in their aluminium packaging. The data supplier believes that in their experience this is a guarantee of stability for 2 years at room temperature. This brief single statement is not considered adequate as a reasoned case and observations on the slow release cassette were not provided. The composition of the slow release cassette was not provided. For details of storage stability studies, see Appendix 2.1.

2.3 ANALYTICAL METHODS

Acceptable validated analytical methods for the determination of dichlorvos in technical material, in formulations and in water have been submitted.

2.4 DATA REQUIREMENTS

The following data should be submitted within 1 year except for 2-year storage stability tests, which should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months. Data Requirements to be addressed by all Approval Holders i. Two year storage stability study on representative strip products if not covered by the

slow release resin strip storage stability studies. Protocols to be agreed with HSE.

9

Approval Holders Supplied by supplier A i. Confirmation as to whether the impurities mentioned in the 1991 ‘Purity and by-

products of technical active’ data sheet are present in currently produced technical active dichlorvos, as these impurities are not examined in the ‘Method for the determination of by-products and supplementary tests for dichlorvos’ report dated 1993. There is also no reference to the above-mentioned impurities in the ‘Chemical Analysis of 5 Representative batches of Dichlorvos Technical’ report dated 1993 (Unpublished, 1993k). If the impurities are still present, analytical methods for detection and quantification are required.

ii. An analytical method to estimate with adequate reliability dichlorvos in water at 0.1 µg l-1 (chromatograms and full test reports should be submitted).

iii. Freezing point study (a limit test e.g. down to 0 oC, will be acceptable). iv. Evidence (such as test report) for vapour pressure value. v. Evidence (such as test report) for flash point value. vi. Evidence (such as test report or reasoned case) for explosivity statement. vii. The oxidising properties should be addressed: a reasoned case may be sufficient. Approval Holders supplied by Supplier B i. Mass Spectrum data for the active ingredient (stating purity). This information

should include a copy of the spectrum, with full operating details such as spectrometer, sample inlet system and ionisation mode and assignment of peaks.

ii. Validation for analytical method to determine dichlorvos in technical active ingredient, and chromatograms and typical results for dichlorvos in the technical material.

iii. Validation data for the method of analysis to determine dichlorvos in aerosol products.

iv. Confirmation of the stabiliser mentioned in the ‘Analysis of DDVP’ report. v. Method validation for impurities and reasons to support allocation of peaks to column

cyclisation and β-elimination products.

vi. An analytical method to estimate with adequate reliability dichlorvos in water at 0.1 µg l-1 (chromatograms and full test reports should be submitted).

vii. Freezing point study (a limit test e.g. down to 0 oC, will be acceptable). viii. Reasoned case to support applicability of flash point stated in The Pesticide Manual

to dichlorvos manufactured by Supplier B.

10

ix. The explosivity properties should be addressed: a reasoned case may be sufficient. x. The statement regarding the oxidising properties should be supported: a reasoned case

may be sufficient. Approval Holders supplied by Supplier C i. 1H NMR data on the active ingredient (stating purity). This information should

include a copy of the spectrum, with full operating details such as spectrometer, frequency, solvent and reference material, and assignment of peaks.

ii. Validation for analytical methods to determine dichlorvos in technical active ingredient and to determine impurities in technical dichlorvos.

iii. Chromatograms and typical results for the determination of dichlorvos in the formulation (strip). Validation data for the method of analysis to determine dichlorvos in strips.

iv. An analytical method to estimate with adequate reliability dichlorvos in water at 0.1 µg l-1 (chromatograms and full test reports should be submitted).

v. Evidence (such as test report) should be submitted in support of the freezing point. vi. Evidence (such as test report) should be submitted in support of the flash point. vii. The explosivity properties should be addressed: a reasoned case may be sufficient. viii. The oxidising properties should be addressed: a reasoned case may be sufficient. ix. More information on the ‘normal conditions of ambient temperature and humidity’

(i.e. what was the temperature and the humidity or where was the location of the study and the time of year) during the strip storage stability study. More information on the analysis. Confirmation of the composition of the strip. Observations of the physical state and integrity of the strip, and of the colour and weight of the strip. If these details are not available, a new 2-year storage stability study at ambient temperature is required; protocol to be agreed with HSE.

11

3 MAMMALIAN TOXICOKINETICS AND TOXICOLOGY

3.1 MAMMALIAN TOXICOKINETICS

3.1.1 ORAL

3.1.1.1 ABSORPTION, DISTRIBUTION AND ELIMINATION

A number of balance studies in the rat, mouse and hamster are available. These have used dichlorvos radiolabelled at the phosphorus, methoxy carbon or vinyl carbon atoms (see Figure 3.1). Although there are no differences in the toxicokinetics of dichlorvos the fate of the radioactivity differs according to the position of the label.

Figure 3.1 Dichlorvos showing position of radio-labels

Three studies, from the same laboratory, reported on the administration of dichlorvos to rats, mice and hamsters (Hutson & Hoadley, 1972; Hutson & Hoadley, 1972a; Hutson et al., 1971). Rats were administered [14C-methoxy]-dichlorvos (specific activity 18.3 µCi mg-1, 1 mg per animal) or [14C-vinyl]-dichlorvos (males 0.99 mg per animal, specific activity 14.5 µCi mg-1; females 0.72 mg per animal, specific activity 6.75 µCi mg-1). Mice received [14C-methoxy]- dichlorvos (0.52 mg per animal, specific activity 18.3 µCi mg-1). Hamsters received [14C-vinyl]- dichlorvos (males approximately 4 mg kg-1 or females 1.6 mg kg-1). Animals were maintained in metabolism cages for 4 d and urine, faeces and respired air collected daily. Total recovery of radioactivity was > 85 % in all cases. There was a significant difference in the excretion pattern between 14C-methoxy and 14C-vinyl labels. However there was no significant difference between sexes or between rats and mice administered [14C-methoxy]-dichlorvos. There may be slight quantitative differences between rats and hamsters administered [14C-vinyl]-dichlorvos. Urine was the major route of excretion in animals administered [14C-methoxy]-dichlorvos (rats 62-65 % and mice 59-64 % of total radioactivity). Expired 14CO2 accounted for about 15 % in rats and mice. In rats 5-7 % and mice 3 % of total radioactivity was excreted in faeces. The carcass contained 5 % of total radioactivity in rats and mice, while skin contained 2 % in rats. Excretion of 14C label was rapid; with 93 - 94 % of urinary

12

radioactivity, 82 - 88 % of the total expired as 14CO2 and 50 - 55% of the faecal radioactivity excreted by 24 h. For [14C-vinyl]-dichlorvos, expired 14CO2 was the major route of excretion (37 - 39 % of total radioactivity in rats and 43 - 50 % in hamsters). Urine accounted for 13 - 18 % of radioactivity in rats and 15 - 24 % in hamsters, while faeces contained 3.4 - 4 % in rats and 3 - 7 % in hamsters. The carcass contained 12 - 16 % of total radioactivity in rats and 14.8 - 16.5 % in hamsters. The [14C-vinyl] label was also rapidly excreted; with 80 - 92 % of urinary radioactivity, 70 - 80 % of the total expired as 14CO2 and 47 - 70 % of the faecal radioactivity excreted by 24 h. The same authors also reported results of an inhalation study in rats exposed to [14C-vinyl]-dichlorvos vapour for 1 h (Hutson et al., 1971). Exposure was estimated to be 0.71-1.07 mg dichlorvos. The results obtained (expressed as a percentage of the total recovered as 14CO2) indicated that 39 % of the amount excreted as 14CO2 was recovered from urine and 9.8 % in faeces. Excretion was rapid, with 70 % of expired 14CO2, and 90 % of urinary and 33 % of faecal radioactivity excreted in the first 24 h. The results were comparable with those from oral studies. In an earlier study female rats received 1.1 mg kg-1 [14C-vinyl]-dichlorvos by gavage (Casida et al., 1962). Animals were maintained in metabolism cages for 24 h to allow collection of expired air, or for 7 d for collection of urine and faeces. Over 24 h expired 14CO2 accounted for 16 % of the administered dose. Over 7 d, 27 - 32 % of the administered dose was recovered from urine and 3 % from faeces. Radioactivity was found in tissues 7 d after dosing. The levels were reported as 3.1 ppm of

14C equivalents in the liver, between 0.24 and 1 ppm were present in blood, brain, fat, heart, kidney, muscle, stomach, large and small intestine. Similar results were obtained following ip administration. In one study using 32P-dichlorvos male and female rats received 10 mg kg-1 by gavage and were sacrificed at various times up to day 7 after dosing (Casida et al., 1962). Tissue levels were similar in both sexes. At 15 min after dosing the majority of organo-soluble 32P was recovered from either the GI tract or liver (reported as 126 and 38 ppm dichlorvos equivalents respectively). However low levels (<10 ppm) were found in blood, bone, brain, fat, heart, kidney and muscle. The levels in the GI tract fell rapidly, reaching approximately 15 ppm by 14 h. Levels in the liver were increased at 1 h post dosing (to 59 ppm), falling slowly to 3.2 ppm on day 4. Only two other organs contained significant amounts of organo-soluble 32P. These were the kidney (where levels were 26, 12 and 7 ppm at 1, 4 and 14 h post-administration respectively) and bone (where levels gradually increased over the first 14 h to 12 ppm and remained constant until the end of the study). In another study with 32P-dichlorvos, male and female rats received 0.1, 1.0, 20, 40 or 80 mg kg-1 by gavage (Casida et al., 1962). Urine and faeces were collected for 148 h, before animals were killed. The percentage excretion was stated not to vary greatly with dosage in

13

either male or females. In both sexes, 60 - 70 % of administered dose was recovered from urine and 11 - 17 % from faeces. A study reported that dichlorvos penetrated the rabbit placenta (Maslinska et al., 1979). Rabbits (5 treated and 2 control) received 0 or 6 mg kg-1, by gavage, on the expected day of delivery. Fetuses were delivered by caesarean section 5, 10, 20, 30 and 120 min after the dose and the blood analysed for dichlorvos. Dichlorvos was detected (by GC) in fetal blood at 5 min; rising to a maximum of 7.7 µmol l-1 at 20 min and falling to 0.5 µmol l-1 at 120 min. No dichlorvos was detected in control fetal blood. Male Sprague-Dawley rats (2 per dose) received 0.63 or 6.3 mg kg-1 d-1 by gavage for 3 d (Bradway et al., 1977). Urine and blood were collected during the treatment period and for up to 7 d afterwards. The only metabolite identified was dimethyl phosphate (DMP) in the urine. This accounted for approximately 10 % of administered dose at either 0.63 or 6.3 mg kg-1 d-1. Excretion was rapid, with DMP falling below the level of detection 3 d after administration ceased at either dose. Both doses produced a reduction in plasma cholinesterase. A well-conducted study in rats performed to GLP and consistent with OECD guidelines investigated the toxicokinetic profile of dichlorvos following acute oral exposure (Unpublished, 1989, 1991a). Rats (Crl: CD (SD) BR; 5 per sex per group) were administered a single gavage dose of either 0.8 or 21 mg kg-1 [14C]dichlorvos (purity 100 %; labelled in the vinyl position) in deionised water. Control animals (2 per sex) received deionised water and were sacrificed 24 h later. Animals were observed twice daily for moribundity/mortality or signs of toxicity. Expired air, urine and faeces were collected between 0-6, 6-12, 12-18 and 18-24 h post-exposure and then daily for 7 days and radioactivity levels measured. Blood samples were taken from treated animals on day 7 post-exposure. Treated animals were sacrificed on day 7, and radioactivity measurements were made in the following tissues: bone (femur), brain, fat, ovaries, testes, heart, pancreas, liver, kidneys, lungs, muscle (thigh), spleen, uterus and residual carcass. One female of the 21 mg kg-1 group died 2.5 h post-exposure, while the remaining animals all exhibited tremors and salivation (signs associated with cholinesterase inhibition). Levels of radioactivity in control animals were generally non-detectable. Recovery of administered dose in both treated groups was good (89 - 94 %). The toxicokinetic profile of dichlorvos observed in this investigation was similar between both the sexes and the dose levels. Expired air accounted for the majority of the radiolabel detected (41 - 54 % of administered dose), followed by residual carcass (13 - 24 %), urine (10 - 17 %), faeces (4 - 7 %) and liver (3 – 5 %). Measurements in other tissues accounted for only 1% of the administered dose: of these the highest measurements occurring in order in the kidney, lung, spleen, uterus and bone, fat showing the lowest levels. The majority of radioactivity was excreted in the expired air, urine and faeces within 24 h of exposure; and more than half of that measured in expired air and urine was excreted within 6 h of exposure. As part of the same study described above, rats (5 per sex) were administered 0.8 mg kg-1 d-1 dichlorvos by gavage for 15 days. On the 16th day the animals received a single gavage dose of [14C] dichlorvos. The subsequent methodology was as previously described.

14

No signs of toxicity were reported in the animals during the study. The results following this dosing regime were remarkably similar to those observed following acute exposure. Again the expired air accounted for the majority of the radiolabel detected (56 % of administered dose), followed by residual carcass (18 %), urine (14 %), faeces (5 %), liver (4 %) and other tissues accounting for 1 %. The excretion profile was also similar (Unpublished, 1989, 1991a). Overall, these studies indicate that dichlorvos is well absorbed following oral exposure and able to enter the systemic circulation. Distribution around the body appears to be wide and is certainly to the liver. A significant amount of radioactivity was measured in the residual carcass at the end of the study. This is likely to be a result of the rapid metabolism of dichlorvos and subsequent incorporation into endogenous amino acids and proteins (see metabolism section). The appearance of 14CO2 in expired air soon after dosing indicates that dichlorvos will be rapidly metabolised once it enters the systemic circulation. The main routes of excretion for dichlorvos and/or its metabolites are via the expired air as CO2 or via the urine. A study reported by Majewski et al., (1979) found no evidence of dichlorvos in the muscle tissue of rabbits (sex unknown) administered 5 mg kg-1 d-1 dichlorvos in propylene glycol for 2 weeks, and sacrificed 6, 12, 24 and 48 h post-administration (2 animals per time point; 2 controls). The same authors found no evidence of dichlorvos in fetal tissues, after administration of 5 mg kg-1 d-1 dichlorvos in propylene glycol for 25 days to female rabbits (2 control; 6 test). The animals were sacrificed 24 h post administration.

3.1.1.2 HUMAN DATA

A post-mortem on a 24-year-old male found dead after drinking a liquid formulation of dichlorvos (type unknown) showed that dichlorvos was widely distributed throughout the body (Jadhav et al., 1989). A male volunteer received an oral dose of 5 mg [14C-vinyl]-dichlorvos (Hutson & Hoadley, 1972a). Exhaled air was collected intermittently over 0-8 h; the subject exhaling for 10 min through a respirometer. Urine was collected for 48 h. A significant proportion (estimated as 27 %) of the administered dose was recovered as 14CO2. Over 48 h, 9 % of the administered dose was recovered from urine, although samples taken on day 3 and 4 still contained measurable amounts of radioactivity. Urine (0 – 24 h samples) was analysed for metabolites by paper chromatography. Hippuric acid, desmethyl dichlorvos, urea and a dichloroethanol conjugate were all identified, although quantification of the conjugate was not possible. The others represented 0.4, 0.15 and 0.1% of administered radioactivity respectively.

3.1.1.3 METABOLISM

A proposed metabolic pathway for dichlorvos is presented in Figure 3.2. Urine, from both rats and mice administered either 1 or 0.52 mg per animal [14C-methoxy]-dichlorvos by gavage, was analysed by paper chromatography (Hutson & Hoadley, 1972).

15

The major metabolite isolated from the urine of both rats and mice was DMP (64 - 77 % of urinary radioactivity). A further metabolite, desmethyl dichlorvos, was isolated and constituted 4 and 28 % of urinary radioactivity in rats and mice respectively (indicating that the mouse may preferentially demethylate in comparison with the rat). Minor metabolites were; S-methyl-L-cysteine (rats and mice), S-methyl-L- cysteine oxide (mice), methylmercapturic acid S-oxide (rat), methyl mercapturic acid (mouse). These were all present at less than 2 % of the total urinary radioactivity. Two components, not further quantified, remained unidentified. Hamsters and mice were administered [14C-vinyl]- dichlorvos by gavage and urine samples (collected for 0-24 h) were analysed by paper chromatography (Hutson & Hoadley, 1972a). The metabolites, not fully characterised, included hippuric acid (1 % of administered radioactivity in hamsters and 0.6 % in mice), desmethyl dichlorvos (18.5 % of administered radioactivity in mice, not assessed in hamsters) and urea (0.6 % in mice, not measured in hamsters). Female rats were administered [14C-vinyl]-dichlorvos by gavage and urine samples were analysed by derivatisation and carrier crystallisation (Casida et al., 1962). The major metabolite identified was a conjugate of dichloroethanol (>90 % of urinary radioactivity), tentatively identified as dichloroethyl glucuronide. Other minor metabolites were derivatives that hydrolysed in acid to give dichloroacetaldehyde and acid derivatives (not fully identified). Similar metabolites were obtained following ip administration. Male and female albino rats received up to 80 mg kg-1 32P-dichlorvos by gavage and urine was analysed by ion exchange chromatography (Casida et al., 1962). Between 82 - 91% of urinary radioactivity was present as either monomethyl phosphate (MMP) or DMP (co-eluted in the system used), 2 - 13% as desmethyl dichlorvos and 3 - 16 % as inorganic phosphate. Desmethyl dichlorvos tended to be excreted over the initial 12 h after administration while inorganic phosphate was excreted later (12 - 48 h). DMP and MMP were excreted throughout the study. Urine (0 - 24 h samples), from rats treated with [14C-vinyl] or 36Cl-dichlorvos by gavage (0.99 or 0.72 mg 14C-dichlorvos or 18.8 mg 36Cl-dichlorvos), was analysed by paper chromatography (Hutson et al., 1971). A major metabolite isolated after treatment with 36Cl-dichlorvos was sodium chloride; four major 36Cl-containing metabolites were identical to metabolites seen with 14C-dichlorvos (not further quantified). The major metabolites in urine from 14C-dichlorvos treated rats were hippuric acid (8.3 % of urinary radioactivity), desmethyl dichlorvos (10.9 %), 2,2-dichloroethyl-ß-D-glucopyranosiduronic acid (27 %) and urea (3.1 %). The presence of dichloroacetaldehyde and dichloroacetic acid could not be demonstrated. Five female rats received 0.6 mg [14C-vinyl]-dichlorvos (11.9 µCi) by gavage and were sacrificed 4 d later (Hutson et al., 1971). The livers were isolated and found to contain 5.5 % of the administered dose. On separation the majority of radioactivity was found in the protein fraction (75 % of liver radioactivity) whereas the soluble fraction contained 11 %, the

16

lipid fraction 12 % and nucleic acid fraction 3 %. When the protein was hydrolysed three amino acids were found to contain 14C, glycine and serine (specific activity 2012 and 1966 µCi mol-1 respectively) in major amounts and a smaller amount as cystine (not quantified). 32P-dichlorvos was orally administered to Guernsey cows (1 per dose) at 1 or 20 mg kg-1 (Casida et al., 1962). Urine and faeces were collected for up to 168 h. At 1 mg kg-1, 15 % of total recovered was excreted in urine and 53 % in faeces. At 20 mg kg-1, 40 % of the dose was recovered in urine and 51 % in faeces. Analysis of urine showed the major metabolites were DMP and MMP, which co-eluted in the system used (69 - 98 % of radioactivity). Desmethyl dichlorvos accounted for 30 % of urinary radioactivity 0.5 - 1 h after dosing. Low levels of inorganic phosphate were also identified. When milk was analysed for 32P equivalents after oral dosing, at 1 mg kg-1, measurable amounts were not detected until 2 h after treatment. The maximum amount (approximately 0.6 ppm) was seen 12 - 24 h after dosing. At 20 mg kg-1 higher levels were seen, with a maximum of approximately 10 ppm at 8 - 18 h. The majority of radioactivity was due to hydrolysis products. Amounts of organo-soluble 32P were only above background levels 0.25 - 2 h after administration (maximum 77 ppb at 1 h). A single rat received 500 mg kg-1 32P-DMP by gavage (Casida et al., 1962). Urine was collected for 90 h before the animal was killed and tissue distribution observed. The only radioactivity recovered from the urine was as DMP (approximately 50 % of dose). The tissues contained almost no radioactivity. Following the studies (Unpublished 1989, 1991a) previously described in Section 3.1.1.1, further work was performed with the objective of identifying the 14C-metabolites of dichlorvos in urine and faeces. Techniques used included thin layer chromatography, gas chromatography, glucuronidase hydrolysis, isotope dilution analysis and mass spectrometry. Five radioactive samples were isolated from urine for mass spectrometry. Two were identified as urea (19 - 33 % of urinary radioactivity) and hippuric acid (4 - 20 %), the others remained unknown. However, the absence of characteristic ‘chlorine clusters’ in their mass spectra indicates that the metabolites had been dehalogenated. Enzyme hydrolysis of urine revealed the possible presence of glucuronide metabolites but these were not identified. Low levels of radiolabelled urea and hippuric acid were also detected in faeces. The authors proposed a metabolic pathway for dichlorvos and indicated that the large amount of radioactivity eliminated in the expired air as carbon dioxide, and the presence of urea and hippuric acid suggests the involvement of a 'one-carbon pool pathway' (Unpublished,1991a).

3.1.2 INHALATION

3.1.2.1 ABSORPTION, DISTRIBUTION AND ELIMINATION

CFE rats were exposed by inhalation to 10 or 90 mg m-3 dichlorvos for 4 h (Blair et al., 1975). Animals were killed immediately after exposure and tissues taken for analysis of dichlorvos content. Extraction efficiencies of dichlorvos were reported as low and were not

17

corrected for. At 10 mg m-3 dichlorvos was detected in male kidneys but not in female kidneys or blood, liver, fat or lung (limit of detection in tissues 0.01 µg g-1 and in blood 0.1 µg g-1). At 90 mg m-3, dichlorvos was detected in trachea (both sexes), male kidney and female fat. Male rats were also exposed to 50 mg m-3 for 4 h and killed up to 60 min after exposure ceased (Blair et al., 1975). Kidneys and blood were analysed for dichlorvos (limit of detection 0.01 µg g-1). Dichlorvos disappeared rapidly from the kidney, with a t½ of 13.5 min. No dichlorvos could be detected in blood 15 min after exposure (limit of detection 0.1 µg g-1). Male CFE rats were exposed by inhalation to 0.05 or 0.5 mg m-3 dichlorvos continuously for 14 d (Blair et al., 1975). Animals were killed immediately after exposure and tissues analysed for dichlorvos. No tissues were found to contain dichlorvos above the limit of detection (0.01 µg g-1 in tissues or 0.1 µg g-1 in blood).

3.1.2.2 METABOLISM

Male rats were exposed by inhalation for 1 h to [14C-vinyl]-dichlorvos vapour and urine (collected for 0 - 24 h) was analysed by paper chromatography (Hutson et al., 1971). The major metabolites identified were; hippuric acid (9.3 % of urinary radioactivity), desmethyl dichlorvos (4.3 %) and urea (5.3 %). Some metabolites were not identified.

3.1.2.3 HUMAN DATA

Two male volunteers were exposed in a 20 m3 chamber to dichlorvos vapour (Blair et al., 1975). The first was exposed to 0.25 mg m-3 for 10 h, the second to 0.7 mg m-3 for 20 h. Blood samples, taken 1 min after leaving exposure chamber, contained no dichlorvos (detection limit 0.1 µg g-1).

3.1.2.4 OTHER STUDIES

A study by Wennerberg and Lofroth (1974) indicated that dichlorvos could act as a methylating agent to both E. coli and in mice. Following ip and inhalation exposure of mice to [14C]-dichlorvos (radiolabelled in the methyl position), excretion of labelled 7-methylguanine in the urine was reported. Exposure of the E. coli test system to the same radiolabelled dichlorvos leads to the presence of labelled 7-methylguanine in both isolated DNA and RNA.

3.1.3 DERMAL

3.1.3.1 ABSORPTION, DISTRIBUTION AND ELIMINATION

A study conducted to GLP and contemporary EPA guidelines investigated the dermal absorption of dichlorvos in rats (Unpublished, 1990a). [14C]-dichlorvos (purity > 95 %; labelled in the vinyl position) was applied to the skin on the back of male rats (Crl: CD (SD)BR; 12/group) at target concentrations of 3.6, 36 or 360 µg per rat under a non-occlusive

18

dressing. A 'protective appliance' was placed over the application site to prevent contamination of other portions of the rat or its excreta. The device also incorporated a charcoal impregnated filter to collect any radioactivity lost through evaporation of the test substance. Groups of 4 animals per dose were sacrificed at 10, 24 and 120 h post-application. However, the final sacrifice time for animals of the 360 µg group was 102 h due to an 'impending ice-storm' at the laboratory site (it should be noted that the investigation at this dose level was carried out after completion of the investigation at the lower dose levels). The application site of all animals was washed 10 h post-exposure, and the protective appliance (including the charcoal filter) replaced on animals continuing in the study. Following application the animals were placed in metabolism chambers and urine and faeces collected. The level of 14CO2 in expired air was also measured. The collection periods ended at 10 h, 24 h and each subsequent 24 h period until animals were sacrificed. At terminal sacrifice blood samples were taken and the bladder contents removed to add to the final urine collection. Also, the skin at the application area was removed. Thus, radioactivity measurements were made on urine and cage rinse, the protective appliance, the charcoal filter, the skin wash, blood, faeces, carcass, expired air and skin. It is unclear whether any signs of toxicity in the animals were looked for. The actual doses administered to the animals were on average 96, 98 and 90 % of the target doses (3.6, 36 and 360 µg), respectively. Recovery of administered dose (90 - 93 %), was similar across the 3 groups and was partly accounted for by evaporation from the test site (43 - 50 % of the administered dose). The percentage of the dose absorbed was calculated by summing the radioactivity levels in the carcass, blood, skin at dose site, expired air, urine and faeces. It was similar across all dose groups and sacrifice times (22 - 30 % of administered dose). The percentage of the dose that actually passed through the skin into the systemic circulation was again similar across all dose groups and sacrifice times (6 - 11 % of administered dose). The fact that levels in the skin remained essentially unchanged between 10 and 120/102 h indicates that dichlorvos may have the potential to accumulate in the skin if there was repeated exposure. Measurements in blood never accounted for more than 0.2% of the administered dose across the dose groups. However, the concentrations of [14C]-dichlorvos and its metabolites in blood increased with increased dose and fell in a similar manner over time. As would be expected following the observation of similar absorption profiles excretion via the faeces, urine and expired air were also similar across the dose groups. By 10h expired air accounted for approximately 2.7 % of the administered dose, urine 1.2 % and faeces 0.1 %. At the end of the study these figures had risen to approximately 4.6 %, 2 % and 1 % of the administered dose, respectively. This study indicates that dichlorvos has the potential to cross the skin following acute exposure. Also, it appears to have the potential to bioaccumulate in the skin. Although there is no evidence for release of accumulated dichlorvos or a metabolite from the skin in the short time period investigated in this study, this cannot be ruled out. It should be noted that dichlorvos is well absorbed through the eye, causing systemic toxicity and a significant decrease in plasma cholinesterase activity in rabbits and rats (Unpublished, 1986a).

19

3.1.3.2 METABOLISM

No data are available.

3.1.4 IN VITRO STUDIES

A number of studies on metabolism of dichlorvos in vitro are available (Hodgson & Casida, 1962; Dicowsky & Morello, 1971; Blair et al., 1975; Reiner et al., 1980; Traverso et al., 1989). Some of these studies have investigated specific aspects of metabolic kinetics or tissue differences in metabolism. Studies have identified the following metabolites; DMP (the principal metabolite), MMP, desmethyl dichlorvos and inorganic phosphate following incubation of 32P-dichlorvos with whole homogenates of liver, kidney, spleen and adrenal glands from the rat and rabbit (Hodgson & Casida, 1962). No further metabolism of 32P-DMP occurred. The major metabolite following incubation of [14C-vinyl]-dichlorvos was dichloroacetaldehyde, which was further reduced to dichloroethanol (Hodgson & Casida, 1962). A small amount of an acid, possibly dichloroacetic acid, was also found. The major metabolite found after incubation of [14C-methoxy]- dichlorvos was DMP, with lesser amounts of desmethyl dichlorvos (Blair et al., 1975). Analysis of metabolites from unlabelled dichlorvos revealed dichloroacetaldehyde, desmethyl dichlorvos, DMP and MMP (Dicowsky & Morello, 1971). The demethylation of dichlorvos to desmethyl dichlorvos was shown to be glutathione dependent in rat liver supernatant but cleavage of the vinyl-phosphate bond was glutathione independent (Dicowsky & Morello, 1971). A study has shown that a variety of tissue homogenates (liver, kidney, spleen and plasma) from both the rat and rabbit are capable of hydrolysing dichlorvos (Hodgson & Casida, 1962). Two studies have investigated the identity of the enzymes that hydrolyse dichlorvos, identifying characteristics similar to A-esterases (Reiner et al., 1980; Traverso et al., 1989).

3.1.5 SUMMARY OF TOXICOKINETICS

The toxicokinetic profile of dichlorvos has been thoroughly investigated in experimental animals (rats, mice, hamsters, rabbits) following oral exposure; a well-conducted dermal study is available while more limited data is available following inhalation exposure. Following oral exposure dichlorvos is rapidly and completely absorbed. Absorption is not as great following dermal exposure, data from a modern toxicokinetic study indicating a figure of approximately 10 % of the applied dose. However, data from acute oral and dermal studies indicate that the actual figure may be greater than this. Following inhalation exposure absorption certainly occurs; the extent is unclear but is expected to be extensive (given the good absorption from the GI tract). Once absorbed dichlorvos and/or its metabolites are rapidly and widely distributed around the body in all species studied (rat, mouse and hamster). Metabolism is both rapid and complete, and appears to be qualitatively similar in all species studied. Dichlorvos appears to undergo transformation via glutathione conjugation or

20

cleavage mediated by cytochrome P450 enzymes. It appears that metabolism leads to a deactivation of the dichlorvos molecule, with evidence that dichlorvos metabolites are eventually broken down and incorporated into endogenous amino acids and proteins. Excretion of dichlorvos metabolites is rapid across all species tested. Radioactivity is mainly excreted as CO2 in expired air (up to 50 %), or excreted in urine (up to 25 %) within 24 h of exposure. Following dermal exposure, a significant amount of the dichlorvos that passes through the skin appears to remain 'fixed' there. Dichlorvos and/or its metabolites can cross the placental barrier and have been detected in fetal blood following maternal exposure. There is no data available to indicate whether or not dichlorvos can appear in breast milk. Human data are limited. In one human volunteer study, 27 % of the administered dose was recovered in expired air (up to 8 h) and 9 % in urine (up to 48 h). The urinary metabolites identified in this study were qualitatively similar to those seen in animal studies. Post-mortem examination of a suicide indicated that dichlorvos was widely distributed. Thus from the very limited information available, the toxicokinetics of dichlorvos in humans appeared qualitatively similar to that seen in animals.

21

Figure 3.2 Proposed metabolic pathway for dichlorvos

22

3.2 MAMMALIAN TOXICITY

There is a large body of information concerning the effects of single and repeated exposure to dichlorvos, with most studies having been conducted in the rat. In the majority of the repeated-exposure studies, data have been presented concerning the effect of dichlorvos on erythrocyte and brain acetylcholinesterase activity, as well as on non-specific plasma cholinesterase and tissue carboxyl esterases. This reflects the position that for organophosphates in general the inhibition of cholinesterase enzymes are considered key endpoints. Before summarising each of the available studies, a general position interpreting such data is set out below. The ACP has recently deliberated on the interpretation of such data, and has concluded that the magnitudes of inhibition judged to be toxicologically significant are 20 % or more for both erythrocyte and brain acetylcholinesterase activities. Further, the ACP considers that inhibition of plasma cholinesterase is not an adverse health effect and that measures of plasma cholinesterase inhibition should therefore not be interpreted as toxicologically significant endpoints. Changes in tissue carboxylesterase activity are also not considered an adverse effect, given that these enzymes have a very broad range of potential physiological substrates, and that changes in the activity of these enzymes may not necessarily reflect changes in the acetylcholinesterase enzymes involved in neuronal transmission. However, where available, data on these enzymes are presented in individual studies as supportive information. The fact that some organophosphates are reversible inhibitors of acetylcholinesterase activity introduces the problem of spontaneous reactivation of the enzyme. Improper handling can promote enzyme reactivation and thus result in an overestimation of cholinesterase activity (underestimation of enzyme inhibition). Several procedures to minimise these problems have been proposed by the Consultation, held in Geneva in January 1998, on interpretation of inhibition of acetylcholinesterase activity, which provided guidance following the 1997, Joint FAO/WHO Meeting on Pesticide Residues (JMPR). These include; the use of the Ellman assay protocol, no washing of erythrocytes, diluting samples by no more than 2 fold and storing samples at temperatures between -20 oC and -80 oC.

3.2.1 ACUTE TOXICITY

3.2.1.1 ORAL

3.2.1.1.1 Active Ingredient

Rat In 2 well-conducted studies, consistent with current EC-guidelines, rats (Wistar; 5 per sex per group) were administered single doses of up to 98 mg kg-1 DDVP-technical in propylene glycol by gavage. The studies were carried out by the same laboratory on 2 different samples of DDVP-technical. Within a few minutes of dosing rats showed sluggishness and ataxia followed by coma. Deaths occurred between approximately 0.5 - 2.5 h post-dosing, while

23

surviving animals recovered gradually (Unpublished, 1979a,b). No deaths were reported at the lowest dose level administered (approximately 34 mg kg-1), however, signs of toxicity were reported and therefore a NOAEL could not be derived from these studies. An oral LD50 value of approximately 60 mg kg-1 in rats is reported from these studies. Oral LD50 values of between 46 and 105 mg kg-1 in the rat have been reported from studies (including one to 1981 OECD guidelines) using dichlorvos of up to 98 % purity (Unpublished, 1973; Unpublished, 1969a; Unpublished, 1986b; Sasinovich, 1967). In a recently conducted study, the LD50 was 105 mg kg-1 (Unpublished, 1986b). Fatalities (occurring within 2 h of dosing) were observed at 100 mg kg-1 and above in both sexes. Clinical signs of toxicity, seen at 50 mg kg-1 and above, were dyspnoea, exophthalmus, ruffled fur, curved position and tremors. Additionally, at 100 mg kg-1, chromodacryorrhea and clonic-tonic convulsions were seen. Signs developed within 1 h, and in the case of ruffled fur and dyspnoea were present until days 12 and 9 respectively. All other signs disappeared within 3 d. Necropsy of fatalities revealed oedema in the lung; there were no changes in survivors necropsied at the end of the observation period. In other, more limited studies no mortalities were seen in males receiving 31.7 mg kg-1 (the lowest dose used) but 1/5 females died at this dose. Deaths were generally observed within 1 h (Unpublished, 1973). In addition to the clinical signs of toxicity reported above tachypnoea, lacrimation and prostration were observed (Unpublished, 1973; Srivastava et al., 1989). Recovery, where reported, occurred within 4 - 5 d. Necropsy of fatalities revealed congestion of the liver and bloating of the GI tract (Unpublished, 1969a). In one study investigating tissue enzyme levels (testes, brain, liver and heart), ALT, AST, acid and alkaline phosphatase were all increased 3 h after dosing. However activity returned to control levels within 12 h (Srivastava et al., 1989). Rats received oral doses of up to 50 mg kg-1 (the activity of brain acetylcholinesterase was determined up to 24 h afterwards) (Modak et al., 1974; Teichert et al., 1976; Gadamski and Szumanska, 1977; Brzezinski and Wysocka-Paruszewska, 1980; Pachecka et al., 1975). Reduction of activity was rapid at 50 mg kg-1, to 10 % of control activity 15 min after dosing; although some recovery (to 60 % of control) was observed within 24 h (Modak et al., 1974; Brzezinski and Wysocka-Paruszewska, 1980). In one study no effect was seen at 8 mg kg-1. However, this is not a true NOAEL as erythrocyte acetylcholinesterase activity was not measured (Pachecka et al., 1975). Increased brain acetylcholine concentration (increases of between 48 - 171 %) has also been reported 15 min after an oral dose of 50 mg kg-1 (Modak et al., 1974). Acetylcholinesterase activity in the gastric mucosa was reduced (to 56 % of control) after an oral dose of 40 mg kg-1; this dose also significantly increased HCl production and the activity of histidine decarboxylase (Malinski et al., 1979). Within 4 min of administration of 40 mg kg-1 dichlorvos plasma cholinesterase, acid and alkaline phosphatase activities were significantly reduced (Pachecka et al., 1980). Plasma cholinesterase and erythrocyte acetylcholinesterase activities were reduced (to 23 and 64 % of control respectively) 2 h after Beagles received 42 mg kg-1 of dichlorvos by gavage (Ward & Glicksberg, 1971). Plasma cholinesterase activity recovered rapidly, returning to

24

pre-exposure levels by day 5. Erythrocyte acetylcholinesterase activity took 28 d to return to pre-exposure levels. Doses of up to 80 mg kg-1 did not cause clinical signs of toxicity in Rhesus monkeys (Hass et al., 1972). Plasma cholinesterase and erythrocyte acetylcholinesterase activities were reduced at these doses but there was no evidence of a dose-response relationship. Recovery of activity took 3 and 6 weeks respectively. A non-standard study by Teichert-Kuliszewska and Szymczyk (1979) investigated the effect of dichlorvos exposure on carbohydrate metabolism. Male rats (Wistar; 7 - 8/group) received either 0 or 40 mg kg-1 dichlorvos (approximately half the LD50) in soya bean oil by gavage. Animals were sacrificed 1 or 2 h post dosing, blood samples taken and the livers extracted. Measurements of serum glucose and the liver enzyme activity of glycogen phosphorylase, glucose-6-phosphatase, glycogen synthetase and UDP-glucose pyrophosphorylase were made. Rats were hyperglycaemic 1 h after dosing, serum glucose levels being over twice the level of controls; by 2 h, levels were approximately 50 % greater than controls. No effect was observed on the activity of the liver enzymes glucose-6-phosphatase or glycogen synthetase; however, glycogen phosphorylase activity was increased by 50 % and UDP-glucose pyrophosphorylase decreased by 30 % over controls at both time points. In a well conducted acute neurotoxicity study performed to GLP and to EPA guidelines, rats (Sprague-Dawley; 12 per sex per group) were administered a single gavage dose of either 0, 0.5, 35 or 70 mg kg-1 dichlorvos (purity 97.9 %) (Unpublished, 1993a). The dose levels were selected based on the results of a range-finding study (0.1 - 80 mg kg-1 dichlorvos). Animals were observed twice daily for mortality/moribundity; clinical signs were observed daily; and bodyweights were recorded pre-study and on days 0, 7 and 14. However, it should be noted that no cholinesterase measurements were made. All animals underwent functional observational battery (FOB) testing pre-administration, 15 minutes post-administration (reported by the authors as the approximate time of 'peak effect'), and on study days 7 and 14. This testing was performed blind by the same technicians. The FOB testing included home cage observations (posture, convulsions/tremors, faeces consistency, biting, eyelid closure); handling observations (ease of removal from cage, lacrimation/chromodacyorrhea, piloerection, eyelid closure, red/crusty deposits, eye prominence, ease of handling animal in hand, salivation, fur appearance, respiratory rate/character, mucous membranes/eye/skin colour, muscle tone); open field observations, evaluated over a 2 minute observation period (mobility, rearing, convulsions/tremors, grooming, bizarre/stereotypic behaviour, time to first step (sec), gait, arousal, urination/defecation, gait sore, backing); sensory observations (approach response, startle response, pupil response, forelimb extension, air righting reflex, touch response, tail pinch, eye blink response, hindlimb extension, olfactory orientation); neuromuscular observations (hindlimb extensor strength, hindlimb foot splay, grip strength-hind and forelimb, rotarod performance) and physiological observations (catalepsy, body temperature and bodyweight). Locomotor activity was measured immediately after completion of the FOB, over a 41 minute period using the 'Digiscan Micro Animal Activity System'. A complete necropsy was conducted on all animals found dead during the study, which was reported to include examination of the external surface, all orifices, and the cranial, thoracic,

25

abdominal and pelvic cavities including the viscera. All surviving animals were necropsied after the 14-day observation period. This included measurement of brain weight (excluding olfactory bulbs) and brain dimensions (length and width). Any gross changes, abnormal colouration or lesions of the brain and spinal cord were recorded. Nerve tissues were examined histopathologically from 5 animals per sex of the control and 70 mg kg-1 groups. These included the central nervous system tissues brain (forebrain, centre of cerebrum, midbrain, cerebellum and pons, and the medulla oblongata); spinal cord (at cervical swellings C3-C8, and at lumbar swellings T13-L4); gasserian ganglion/trigeminal nerves; lumbar dorsal root ganglion at T13-L4; lumbar dorsal root fibres at T13-L4; lumbar ventral root fibres at T13-L4; cervical dorsal root ganglion C3-C8; cervical dorsal root fibres C3-C8; cervical ventral root fibres C3-C8; optic nerves; and eyes. Also, peripheral nervous system tissues were examined including sciatic nerves (mid-thigh region and at sciatic notch); sural nerves; tibial nerves; peroneal nerves; and forelimbs. Seven animals (2m; 5f) of the 70 mg kg-1 group died within 4 h of administration. One of the male deaths was reported to be as a result of intubation trauma. No other deaths were reported. No treatment-related effect on bodyweight was reported. The principal clinical sign of toxicity reported was an increased incidence of constricted pupils in males of the 35 and 70 mg kg-1 groups, which occurred sporadically throughout the duration of the study. Although constricted pupils were noted in the all the other treated groups, they occurred with a similar frequency to that observed in the control animals. The main results of the FOB are described below and relate only to animals of the 35 and 70 mg kg-1 groups 15 minutes post-administration. The observations described were dose-related and differed statistically significantly from controls (unless stated otherwise). Results from the pre-administration testing revealed no differences between animals from any of the 4 groups. No toxicologically significant observations were made in the 0.5 mg kg-1 group compared with controls 15 minutes post-administration. When the same observations were made on days 7 and 14, there were no differences observed between treated and control groups. 'Home cage observations' included alterations in posture, clonic convulsions and tremors. 'Handling observations' included increased salivation, and changes in fur appearance and eye prominence (exophthalmus; 70 mg kg-1, males only). Other changes reported included changes in skin colour, muscle tone and respiratory character, however these changes did not differ statistically significantly from control animals. 'Open field observations' included impairment of mobility and gait, convulsions (clonic), tremors, decreased arousal levels and group mean times to first step (70 mg kg-1- males only). 'Sensory observations' included absence of tail pinch response, pupil responses, approach response (70 mg kg-1 - females only), touch response (70 mg kg-1) and air righting reflex (70 mg kg-1). 'Neuromuscular observations' included shorter group mean rotarod performance values, reduced hindlimb resistance (70 mg kg-1) and reduced grip strength (70 mg kg-1). 'Physiological observations' included increased group mean catalepsy values (70 mg kg-1) and decreased group mean body temperatures. Locomotor activity measured pre-administration revealed no differences between treated and control groups. Observations made during the FOB, 15 minutes post administration revealed statistically significant decreases in group mean ambulatory values and overall total motor activity in animals of the 35 and 70 mg kg-1 groups. This decrease was particularly apparent

26

at the start but became less apparent by the end of the 41 minute observation period. No toxicologically significant changes were apparent following treatment with 0.5 mg kg-1. When the same observations were made on days 7 and 14, there were no differences observed between treated and control groups. Gross and microscopic necropsy did not reveal any treatment-related toxicologically significant findings. Overall, this study indicates a NOAEL of 0.5 mg kg-1 for signs of acute neurotoxicity, however, no cholinesterase measurements were made, so it cannot be established whether this is the NOAEL for systemic toxicity. The effects reported at higher dose levels are consistent with effects that would be expected following cholinesterase inhibition appearing within minutes of dichlorvos administration and the surviving animals recovering completely within 7 days. A study in rats (Winstan; 5 – 8 per group) by Sristava & Malik (1988) reported that a single gavage dose of 60 mg kg-1 dichlorvos (as Nuvan, purity 76 %) induced a time-dependent decrease in blood, liver, lung, testes and muscle levels of cholinesterase and carboxylesterase enzymes within 30 minutes of exposure. Generally the maximum decrease occurred after 3 h in all tissues (cholinesterase levels 55 - 66 % of control values; carboxylesterase levels 42 - 59 % of control values) apart from liver cholinesterase (64 % of control values after 60 min) and blood carboxylesterase (44 % of control values after 60 min). In a study reported by Ellinger (1985), rats were administered a single oral dose of 70 mg kg-

1 dichlorvos. The animals were sacrificed 24 h later; the livers were removed and enzyme activities investigated. Liver cholinesterase was inhibited by up to 50 % compared with control values. However, no other liver enzymes analysed (alkaline phosphatase, GIDH, LDH, GOT, GPT, LAP) showed dichlorvos-induced effects. In a poorly reported study by Pacheka et al., (1980) adult male rats were administered a single oral dose of 40 mg kg-1 dichlorvos. Animals were then sacrificed at various times post-administration (at least 6 per time point) and blood collected. No details of control groups were given. Plasma levels of alkaline phosphatase, acid phosphatase and cholinesterase were then measured. They found cholinesterase activity was maximally inhibited at 30 minutes post-administration and remained statistically significantly lower than controls for at least 24 h. alkaline phosphatase activity was statistically significantly lower than control values for approximately 1 h post-administration. Acid phosphatase activity was statistically significantly increased compared to control values after 15 minutes, but reduced from 30 min up to 5 h. The inhibition of cholinesterase activity is what would be expected, however, the authors indicate that the changes in phosphatase activity are a response to the inhibition of the cholinesterase rather than a direct inhibitory effect of dichlorvos. In two studies, rats received single oral doses of 88 mg kg-1 dichlorvos by gavage (Nagymajtenyi et al., 1988; Desi & Nagymajtenyi, 1988). Clinical signs of toxicity and mortalities were seen. Brain and tissue acetylcholinesterase were significantly depressed in both studies. Reported findings were: EEG changes, consisting of a reduction in all

27

components of the frequency band 24 h after dosing; an increase in mean frequency and a decrease in amplitude; a reduction in the conduction velocity in the tail nerve; an increase in the absolute refractory period and a slight reduction in heart rate and an increased amplitude of the Q wave on ECG. A non-standard, poorly reported acute neurotoxicity study in rats by Sarin & Gill (2000) is available. Male rats (Wistar; numbers not given) received a single subcutaneous dose of dichlorvos of 200 mg kg-1, preceded by intraperitoneal doses of 100 mg kg-1 pralidoxime iodide and 20 mg kg-1 atropine sulphate (to prevent interference from cholinergic effects). It is noted that this dose is well above the reported LD50 for sc exposure in rats of 72 mg kg-1. No positive control group is reported. The authors report a significant inhibition of NTE in the platelets and brain tissue; and a severe motor deficit in all animals identified using the rotarod test. The authors concluded that this was evidence of delayed neuropathy, however, no histopathological analysis was performed. Under OECD guidelines, the hen is the most appropriate species for this kind of study (being more sensitive than the rat). Also, a well-conducted study by Ehrich et al. (1995) (see below) indicated that although NTE activity was inhibited in rat brain (at much lower doses and following ip administration) no indication of delayed neuropathy lesions were observed on histopathological analysis. Given this and the fact that a negative acute oral hen study conducted to OECD guidelines is available (there is no guideline for detecting delayed neuropathy in rodents), the evidence presented in this paper for delayed neuropathy induced by dichlorvos exposure is not convincing. Mice In mice oral LD50 values of between 87 and 184 mg kg-1 have been reported for dichlorvos of purity 83 - 100 % (Wagner & Johnson, 1970; Haley et al., 1975; Natoff, 1970; Sasinovich, 1967). Deaths, at unspecified doses, occurred up to 3 h after dosing (Wagner and Johnson, 1970; Haley et al., 1975). Clinical signs of toxicity (muscle fasciculations and ataxia) were reported within a few minutes of administration but the doses at which these occurred were not available (Wagner & Johnson, 1970). Rabbit In the rabbit oral LD50 values of 22.5 (83 % pure) and 74 mg kg-1 (technical dichlorvos, purity not further specified) have been reported (Sasinovich, 1968; Unpublished, 1970a). Fatalities and clinical signs, consisting of dyspnoea, salivation, asynchronism of extremities, clonic-tonic muscle spasms, exophthalmus and lateral position, were observed at 46 mg kg-1 and above. Necropsy of fatalities revealed congestion of the thymus, lungs and liver and haemorrhaging of the thymus and stomach (Unpublished, 1970a). Dog Oral administration of dichlorvos, at up to 22 mg kg-1 (purity unknown), to dogs produced similar signs of toxicity, which subsided within 4 h, to those reported in rats, mice and rabbits (Snow & Watson, 1973). No deaths were observed at 11 mg kg-1. Necropsy of animals that

28

died at 22 mg kg-1 (reportedly due to bradycardia and arrhythmia) revealed haemorrhages and hyperaemia in the GI tract, pulmonary congestion and hyperaemia and cardiovascular changes including extensive haemorrhage (Snow, 1973; Kirkland et al., 1974). In another study, significant changes in liver histopathology (an increase in cytoplasmic area, reduction in nuclear area, increase in number of mitochondria and amount of SER and RER) were seen at 24 h after Beagles received a single oral dose of 44 mg kg-1 (Cockrell et al., 1972). No mortalities were reported. Motor unit irritability was not increased in Beagles receiving single doses of up to 148.5 mg kg-1 dichlorvos (Hazelwood et al., 1978). Dichlorvos had no effect on the vagal tone in female pointers (2 per dose) administered 0 or 60 mg kg-1 dichlorvos (in capsules as an anthelminic formulation) (Dellinger et al., 1987). Prophylactic treatment was started 90 min after administration. This dose reduced plasma and erythrocyte cholinesterase activity. Hen A study, performed to GLP and based partly on a method of the EPA/OPP pesticide assessment guidelines, was carried out to investigate the acute delayed neuropathy potential of dichlorvos in hens (Unpublished, 1988a). Adult domestic hens (10 per group) were administered gavage doses of either 16.5 mg kg-1 dichlorvos (purity 96.5 %; approximately equivalent to the LD50) in distilled water, 600 mg kg-1 triorthocresyl phosphate (TOCP) in corn oil (positive control) or distilled water (negative control). The birds were observed for 21 days. The dichlorvos and negative control groups were then redosed, but the positive control group animals were sacrificed. The surviving birds were observed for a further 21 days. All dichlorvos-treated birds were administered an intramuscular injection of 5 mg kg-1 atropine sulphate concurrent with dichlorvos dosing. Atropine was also reported to be administered in doses of 2 mg kg-1 on an individual basis, as needed, to dichlorvos treated birds. Observations included looking for signs of cholinesterase activity on the day of dosing; and evaluating locomotor impairment twice weekly during a period of forced motor activity. Animals were sacrificed at the end of the study and histopathology performed on tissues including brain, spinal cord and sciatic nerve (right and left). It should be noted that no statistical analysis appears to have been performed during this study. No mortalities occurred during the study. Approximately 30 minutes post-dosing dichlorvos treated birds showed signs of toxicity including lethargy, loss of coordination, lower limb weakness, wing droop, reduced reaction to external stimuli (sound and movement), ruffled appearance, prostrate posture and loss of righting reflex. Most birds recovered by day 4. Similar signs of toxicity were observed following the second dose of dichlorvos, most birds appearing normal from day 25 onwards. Little in the way of signs of toxicity was observed following TOCP treatment, however, by day 21 all birds displayed signs ranging from slight loss of coordination to loss of coordination and lower limb weakness. Evaluation of locomotor activity, revealed no signs of ataxia in dichlorvos-treated birds, however, slight to moderate ataxia was apparent in 6/10 birds administered TOCP by day 21.

29

Histopathological examination revealed minimal nerve fibre degeneration and slight swelling of the axis cylinder of the sciatic nerve in one bird only. Birds administered TOCP showed sciatic nerve changes including minimal to slight nerve fibre degeneration (2/10); minimal to moderate Schwann cell proliferation (3/10); and minimal to slight swelling of the axis cylinder (5/10). No remarkable histopathological changes were observed in the negative control animals. Overall, this study indicates that a single dose of 16.5 mg kg-1 dichlorvos (approximately equivalent to the LD50) does not produce delayed neuropathy. A study by Ehrich et al. (1995) was performed to determine if in vivo inhibition of NTE in 2 species of animals (rat and hen) would be equivalently useful in predicting organophosphate induced delayed neuropathy. A number of compounds were investigated during the study including dichlorvos. Groups of 12 rats (Long Evans) were administered either 0, 5, 10 or 30 mg kg-1 dichlorvos (96 % purity) in corn oil by ip injection; while groups of 12 hens were administered 0, 5, 30 or 60 mg kg-1 in corn oil by ip injection. The highest doses were based on lethality due to cholinergic poisoning or on the ability to inhibit NTE by at least 70 % in pilot studies; the lowest doses were based on the capability to inhibit either NTE or acetylcholinesterase by at least 25 % in the same studies. All animals were pre-treated with 20 mg kg-1 atropine sulphate ip, 15 minutes before dichlorvos treatment. The atropine (10 or 20 mg kg-1, ip) was read ministered to animals showing cholinergic signs within 1 h of dosing; and all groups were given 20 mg kg-1 atropine ip between 1 - 3 h post-treatment. Four animals/group were sacrificed 4 h post-treatment and the brain and spinal cord removed for determination of esterase activities. The remaining animals were used for clinical evaluations and neuropathological assessments. Acetylcholinesterase was determined using the Ellman method (Ellman et al., 1961), while NTE activity was determined using the method of Sprague et al. (1981). The ratio of NTE to acetylcholinesterase inhibition was calculated in spinal cord only. Neurotoxicity indices were assessed in hens by grading ataxia on an 8-point scale; while rats were observed using a functional observational battery, which included endpoints indicative of damage to the autonomic nervous system, motor and sensory systems and central nervous system. Neuropathological examination was performed 3 weeks post-treatment, tissues examined included segments of the medulla, cervical and lumbar spinal cord, and branches of the tibial nerve that supply the gastronemius muscle. In rats 3 animals of the 30 mg kg-1 group died following treatment, and there was also evidence of over stimulation of the cholinergic muscarinic receptors in this dose group. No mortality occurred in hens, however, there was evidence of over stimulation of the cholinergic muscarinic receptors in the 60 mg kg-1 group. Brain acetylcholinesterase activity was biologically significantly inhibited at all doses in the hen compared with control values (74, 43 and 16 % of control values at 5, 30 & 60 mg kg-1, respectively) and at the top two doses in rats (94, 79 and 63 % of control values at 5, 10 & 30 mg kg-1, respectively). Brain NTE activity was biologically significantly inhibited at the top two doses in hens (93, 68 and 52 % of control values at 5, 30 and 60 mg kg-1, respectively) and at all doses in rats (65, 24 and 15 % of control values at 5, 10 and 30 mg kg-1, respectively). The authors presented the results of the spinal cord analysis in diagram form only, so only approximate inhibitions can be reported here. In the rat spinal cord similar profiles of inhibition were observed for both NTE and acetylcholinesterase activity. NTE activity was inhibited to approximately 60 % of

30

control values at the lowest dose and to approximately 40 % of control values at the top two doses; while acetylcholinesterase activity was approximately 70 % of control values at the lowest dose and approximately 60 % of control values at the top two doses. In the hen, NTE activity was only significantly inhibited at the top two doses (approximately 60 % of control values); acetylcholinesterase activity was significantly inhibited at all doses (approximately 60 % of control values at the lowest two doses and 40 % of control values at the top dose). The authors report the NTE/acetylcholinesterase activity inhibition ratios to be 1.3 in rats and 0.53 in hens. No alterations in gait were reported and no evidence of OPIDN was found following histopathological analysis of the spinal cord or peripheral nerves in either rats or hens. Other compounds that were tested in this study and were shown to induce OPIDN (tri-ortho-tolyl phosphate, phenyl saligenin cyclic phosphate, diisopropyl phosphorofluoridate, mipafox) also showed NTE/acetylcholinesterase activity inhibition ratios that were higher in the hen than the rat, whereas with compounds showing no evidence of OPIDN, as in the case of dichlorvos, the opposite effect was observed. The overall conclusion of the study indicated that while rats do develop delayed neuropathy, the doses required lead to a higher incidence of mortality compared to hens, as rats seem to be more susceptible to cholinergic poisoning. Therefore it is suggested that the hen is the most suitable species for the investigation of OPIDN. Overall, although administration was by the ip route, the result in hens confirms the findings in a previously described acute oral study, conducted to OECD guidelines, that dichlorvos does not induce delayed neuropathy. Two non-standard studies in hens administered dichlorvos by subcutaneous injection were reported in the 1994 HSE review of dichlorvos, and are included here as additional information. In the first, non-standard study, 6-12 months old RIR x Light Sussex strain hens received a single subcutaneous (sc) dose of 100 or 125 mg kg-1 dichlorvos with prophylactic treatment and were observed for 10-15 d for development of ataxia (Johnson, 1978). Ataxia was not seen in any hen at either dose. A further group of 8 hens received 100 mg sc followed by 80 mg kg-1 after an interval 48 h. Brain NTE activity was 11 - 18 % of control after the first (2 hens) and 12 % after the second (1 hen) dose. NTE activity in the spinal cord was 47 - 70 % of control after the first dose and 50 % after the second. In the remaining hens mild ataxia was seen in 3 and moderate in a further 2 birds. Another group of 16 hens received 125 mg kg-1 followed by 100 mg kg-1 after an interval of 48 h. NTE activity was 15 % in the brain and 44 % in the spinal cord in the single hen examined after the first dose. In the 4 hens examined after the second dose NTE activity was 11 - 13 % in brain and 22 - 38 % in spinal cord. Ataxia was seen in the remaining hens, graded at mild (5/8), moderate (1/8) or severe (2/8). No histopathology was undertaken. The authors concluded that two doses of dichlorvos (between 10-20 x unprotected LD50) were capable of causing delayed neuropathy.

31

This second study used a commercial formulation of 50 % dichlorvos in hydrocarbons to investigate delayed neuropathy (Caroldi & Lotti, 1981). Adult White Leghorn hens (6 per group) received sc doses of 100 mg kg-1 and either eserine and atropine sulphate or eserine, atropine and 2-PAM. Deaths, 4/6, were observed in hens receiving only eserine and atropine. NTE measurements were performed at 24 h (2 hens per group). Hens (4 which received 2-PAM) were observed for up to 45 d. Mild signs of ataxia developed by day 14 in all hens. The NTE assay showed substantial inhibition in the brain, spinal cord and peripheral nerves (10 - 30 % of normal activity). No histopathology was performed. In this study dichlorvos showed potential to cause mild ataxia after a single sc dose.

3.2.1.1.2 Formulations

Resin Vaporizer Strips In an older study, groups of rats (4 - 5 per sex per group; strain unknown) were administered single gavage doses of 180 - 1400 mg kg-1 resin vaporizer strips (18.6 % dichlorvos). Signs of toxicity developed within 5 minutes of dosing and included salivation, exophthalmus, lacrimation, muscular fasciculation, generalised tremors and tonic spasms. The animals then either died (within 2 h of dosing) or recovered completely. Macroscopic examination revealed no remarkable changes (Unpublished, 1964a). Oral LD50 values of 679 and 382 mg kg-1 for males and females, respectively, were reported from this study.

3.2.1.2 DERMAL

3.2.1.2.1 Active Ingredient

Dermal LD50 values in the rat, of between 210 and 456 mg kg-1, have been reported (Unpublished, 1986c; Fytizas et al., 1970; Unpublished, 1969b). In one study (Unpublished, 1986c), conducted to recent OECD Guidelines and GLP using dichlorvos > 97 % pure, the LD50 was 224 mg kg-1. Deaths (at 74 mg kg-1 and above) occurred within 24 h of dosing. No fatalities were seen at 42 mg kg-1, the lowest dose used. Clinical signs (lethargy, tremors, coma, respiratory difficulties and reduced body temperature) appeared within 30 min of dosing at 74 mg kg-1 and above and were resolved by day 3. Occasionally the treated skin showed erythema followed by scab formation. At necropsy, animals found dead revealed petechiae and/or erosion of the GI tract accompanied by bloody contents. Other findings included petechiae of the thymus, dilatation of renal pelvis, blood in the bladder contents. No treatment related macroscopic findings were seen in survivors.

3.2.1.2.2 Formulations

Resin Vaporizer Strips In an older study, groups of male rabbits (4 per group; New Zealand White) were administered dermally 3200 - 54000 mg kg-1 resin vaporizer strips (18.6 % dichlorvos) for 24 h. The surviving animals were monitored for 2 weeks. Signs of toxicity were reported to

32

occur between 4 - 8 h and included salivation, lacrimation, muscular fasciculation and mild tremors. An LD50 value of 27400 mg kg-1 is reported (Unpublished, 1964a).

3.2.1.3 INHALATION

3.2.1.3.1 Active Ingredient

In a study performed to GLP and contemporary OECD guidelines, rats were exposed to dichlorvos in the form of both vapour and aerosol. Groups of rats (Wistar) were exposed head only for 4 h to 230 - 1926 mg m-3 dichlorvos (purity 98.7 %) as an aerosol (MMAD 1.6 - 6.4 µm; 10 rats per sex per group); or to a vapour of 116 mg m-3 dichlorvos as a vapour (5 rats per sex per group). The rats were observed for a period of 14 d following exposure. Following aerosol exposure 1 female died at 230 mg m-3; at 239 mg m-3 no deaths occurred; at 402 mg m-3 4 animals died (2m, 2f); at 534 mg m-3 15 animals died (7m, 8f) within 4 hours post-exposure; at 1042 mg m-3 all but a single male died within 4 h post-exposure; and at 1926 mg m-3 all animals died within 1 h. Animals of all dose groups displayed signs of toxicity including violent muscular tremors accompanied by muscular asthenia, clonus, lying on the side, ataxia, and respiratory distress, apathy. These animals were also observed to have ruffled/unkempt fur, slightly reduced motility and high-legged gait. These signs had resolved in all surviving animals within 5 days post-exposure. Macroscopic examination of decedents revealed distended, turgid lung; pale and patchy liver with lobular markings; pale spleen and kidneys; hyperaemia of glandular stomach and serosa of small intestine; and blood and mucous inside the GI tract. Surviving animals showed no treatment-related macroscopic abnormalities. An LC50 value in rats following aerosol exposure of 485 mg m-3 (523 mg m-3 and 447 mg m-3 in males and females, respectively) is reported in this study. Following vapour exposure no animals died. Animals displayed ruffled/unkempt fur, reduced motility and high-legged gait, which had resolved 3 days post-exposure. Macroscopic examination revealed no treatment-related abnormalities (Unpublished, 1984a). A NOAEL could not be derived from this study. An LC50 value of 230 mg m-3 in the rat has been reported (Unpublished, 1986d). The study was conducted to recent OECD Guidelines and GLP and used head only exposure to aerosols of up to 240 mg m-3 dichlorvos (97 % pure, mean median aerodynamic diameter of 2.28 µm). Deaths were seen at all doses (170 mg m-3 and above) and occurred during exposure. Clinical signs observed in survivors at all doses consisted of ataxia, tremors, lethargy, hypopnea and bloody encrustation of nose and eye. Recovery occurred within 3 - 4 d. At necropsy of fatalities, haemorrhages were seen in the lungs of a few rats. In addition the trachea from some animals contained a bloody discharge. LC50 values of 14.8 mg m-3 were reported in two poorly conducted studies (Hussain et al., 1982; Sasinovich, 1967). No deaths were observed at 5 mg m-3. As well as the signs of toxicity noted above, loss of muscular co-ordination, balance and body weight were observed. A number of other studies have investigated the acute inhalation toxicity of dichlorvos in the rat, without determining LC50 values. Wistar rats were exposed (head only) to up to 210 mg m-3 dichlorvos (98 % pure) for 4 h (Unpublished, 1982). All rats exposed to

33

210 mg m-3 died within 24 h. A single male died at 142 mg m-3 but no other deaths were observed at up to 206 mg m-3. Signs (lethargy, respiratory difficulties and piloerection) were seen at 85 mg m-3; the lowest dose used. Recovery was observed within 4 days. At higher doses (206 mg m-3 and above), signs were more severe with body tremors, hypothermia and ataxia also being seen. Two more limited studies are available. Fatalities were observed when CFE rats were exposed to 80 mg m-3 of dichlorvos for 6.9 h (Unpublished, 1969c). However no signs were reported after exposure to 28 mg m-3 for 4 h (Pauluhn et al., 1987). A LC50 value of 13.2 mg m-3 in mice, after a 4 h inhalation exposure to dichlorvos (83 % pure), has been reported (Sasinovich, 1967). Deaths and signs of toxicity consistent with OP poisoning were observed at 11.4 mg m-3 and above. Further studies are available in which mice were exposed to dichlorvos, but no LC50 was determined. In one study, no fatalities were reported in mice exposed (head only) to 218 mg m-3 dichlorvos (98 % pure) for 4 h (Unpublished, 1982). Clinical signs (consisting of body tremors, splayed gait, paresis of the hind legs and lethargy) were observed with complete recovery by day 2. In another study all mice exposed to 80 mg m-3 of dichlorvos for 6.9 h died (Unpublished, 1969c).

3.2.1.3.2 Formulations

Formulation A A well-conducted inhalation study was performed to GLP and contemporary OECD guidelines in rats exposed to Formulation A (50 % dichlorvos). The exposures were reported as measured concentrations of dichlorvos in the study report. Nine groups of rats (Sprague-Dawley; 9 per sex per group) were exposed head only to aerosols (MMAD 1.05 - 1.37 µm) containing between 9.8 - 337 mg m-3 dichlorvos for 4 h. Between 9.8 - 28 mg m-3 the aerosols were generated in hexane, and then up to 337 mg m-3 from the undiluted formulation. Deaths occurred in both sexes at 137 mg m-3 and above, and with the exception of one animal during the exposure period. Signs of toxicity characteristic of organophosphate poisoning (tremors, salivation) were observed on the day of treatment following exposure to 90 mg m-3 and above; the majority of these signs resolved the day after treatment. No treatment-related macroscopic findings were reported (Unpublished, 1988b). This study reports LC50 values of 220 and 130 mg m-3 dichlorvos for males and females, respectively, following 4 h exposure to an aerosol of Formulation A.

3.2.1.4 OTHER ROUTES

The intraperitoneal LD50 value in the rat was reported as between 3 and 60.9 mg kg-1 (Fratranska et al., 1978: Naidu et al., 1987; Ryhaenen et al., 1988). Deaths were observed within 15 min of injection; recovery from non-fatal doses was seen within 24 h (Naidu et al., 1987; Takahashi et al., 1991). In one study no deaths were observed at 0.624 mg kg-1 (Takahashi et al., 1991). Typical signs of OP poisoning were observed within 5 min and reversible bradycardia was seen at 30 - 50 mg kg-1 in male Wistar rats (Naidu et al., 1987; Takahashi et al., 1991; Hyde et al., 1978). At lethal doses brain acetylcholinesterase activity

34

was 10 % of control 1 - 5 min after administration (Naidu et al., 1987; Takahashi et al., 1991). Intraperitoneal LD50 values of 24 and 45.7 mg kg-1 have been reported in the mouse (Unpublished, 1970b; Natoff, 1970). Typical signs of OP poisoning were seen 5 min after administration (Unpublished, 1970b). A subcutaneous LD50 value of 72 mg kg-1 in CF rats is available (Unpublished, 1969c). All rats receiving 60 mg kg-1 died, with deaths occurring up to 3 h after administration (Astolfi et al., 1970). Necropsy results, reported for the lung only, showed excess production of mucus in the bronchi and bronchioles. Desquamation of the bronchial epithelium, leading to large areas of atelectasis, was also seen. Anoxemia (caused by atelectasis), hypoventilation, paresis of the respiratory muscles and depression of the respiratory centre were observed. No lesions were found to indicate pulmonary oedema. Subcutaneous LD50's in the mouse of 23.8 mg kg-1 and 28.5 mg kg-1 in Pirbright strain guinea pigs have been reported (Natoff, 1970; Unpublished, 1969c). Several studies are available which used iv, ip, im or sc administration of dichlorvos to rats and mice at 25 - 200 % LD50 (Reiner & Plestina, 1979; Purshottam & Keveeshwar, 1979; Stavinoha et al., 1976; Kobayashi et al., 1986; Soininen et al., 1990; Cohen & Ehrich, 1974; Yoshikawa et al., 1990). Results showed rapid reduction in brain acetylcholinesterase activity in the brain (to < 20% control levels within 15 min). There was also evidence of a significant reduction in brain acetylcholinesterase activity in rats following tail immersion in dichlorvos solution (Borkowska et al., 1989). A study by Hinz et al., (1996) reported that the inhibition of rat brain and serum cholinesterase activity was not effected by pH. The inhibition was mediated by a competitive interaction with the catalytic site of the enzyme leading to irreversible inhibition within a few minutes of incubation.

3.2.1.5 HUMAN DATA

3.2.1.5.1 Oral

A well-conducted human volunteer study, conducted to GLP and to a protocol approved by an independent Ethics Committee, investigated erythrocyte cholinesterase inhibition following oral administration of dichlorvos in 2 phases (Unpublished, 1997a). The first phase involved acute administration and the second repeated administration (see Section 3.2.4.2.1). The acute phase of the study involved 6 healthy male volunteers (determined by routine medical). On the first day of dosing a 35 mg capsule of dichlorvos (purity 97.7%) in corn oil was administered to 4 fasted volunteers; this was followed 1 week later by a capsule of corn oil only (placebo; volunteers); and a further week later a second 35 mg capsule of dichlorvos (6 volunteers, 2 being admitted late to the study). The average weight of the 6 volunteers was approximately 75 kg, thus the dose of dichlorvos administered was approximately 0.5 mg kg-1. Three times in the week prior to the study blood samples were taken for measurement of an individual’s baseline erythrocyte cholinesterase activity (using a modified

35

Ellman method); and subsequent measurements were made on days 1, 3, 5 and 7 (or 8) after each dose. Urine was collected in the 24 h prior to dosing and analysed for drugs of abuse; also volume, pH and creatinine output was measured. Urine collections were made at 12 hourly intervals for 2 days post-dosing and at 24 hourly intervals for a further 3 days. Body temperature was taken pre-dose and 2, 4, 8, 12 and 24 h post dosing. A symptom form was completed for each volunteer 24 h after each dose. No toxicologically-significant changes in erythrocyte cholinesterase activity were found, activities being between 90 - 101 % following dichlorvos administration and 96 - 105 % following placebo administration. No abnormal changes in body temperature were recorded following each dose. No toxicologically-relevant symptoms were reported as a consequence of dichlorvos administration. No urinalysis data was reported. This study indicates that an oral dose of approximately 0.5 mg kg-1 dichlorvos in humans does not inhibit erythrocyte cholinesterase activity. A further well-conducted human volunteer study, conducted to GLP and to a protocol approved by the independent Ethics Committee, by the same authors as above, investigated erythrocyte cholinesterase inhibition following oral administration of dichlorvos (Unpublished, 1997b). A single 70 mg capsule of dichlorvos (purity 98 %) in corn oil was administered orally to 6 healthy, fasted male volunteers (determined by thorough medical). The average weight of the 6 volunteers was approximately 75 kg, thus the dose of dichlorvos administered was approximately 1 mg kg-1. Baseline erythrocyte cholinesterase levels were measured (using a modified Ellman method) in each volunteer 3 times per week in the 3 weeks before dosing and just before dosing; subsequent measurements were made 1, 3, 5 (or 6), 7 and 14 dpost-dosing. Body temperature was taken pre-dose and 2, 4, 8, 12 and 24 h post-dosing. A symptom form was completed for each volunteer 24 h post-dosing. Urine was collected in the 24 h prior to dosing and analysed for drugs of abuse; further urine samples were reported to be collected ‘at intervals after dosing’. No treatment-related signs of toxicity or changes in temperature were reported. Mean erythrocyte cholinesterase activity showed a slight reduction with time following dosing (87 -104 %, 91 - 102 %, 86 - 94 %, 85 - 93 % & 82 - 94 % of control values on days 1, 3, 5, 7 & 14 post-dosing, respectively). However, these reductions in cholinesterase activity up to 14 days post-dosing are unlikely to be of biological significance, especially given the toxicological profile of dichlorvos (rapid absorption and elimination). Therefore, this study shows that an oral dose of approximately 1 mg kg-1 does not significantly inhibit erythrocyte cholinesterase activity (< 20 % inhibition). Groups of male volunteers (4-19 per group, 44 controls), after informed consent, with no history of exposure to OPs in the previous six months, received an oral dose of dichlorvos (97 % pure in PVC resin capsules at 20 % w/w) of between 0 and 32 mg kg-1 (Slomka & Hine, 1981). The total number of volunteers treated was 151. The release rate in vivo was calculated as 27 % over 24 h using collected capsules (time of collection not given). No compound related side effects were reported. Blood samples were assayed for plasma and erythrocyte cholinesterase activities (using a modified Michel potentiometric titration

36

method) and reported as percentages of pre-treatment values. At 24 h a dose related decrease in plasma cholinesterase activity was observed from 50 - 75 % of control levels at the lowest dose (0.1 - 1 mg kg-1; 17 volunteers) to 20 - 30 % of pre-treatment values at 32 mg kg-1. The range in the concurrent control group was 70 - 100 % of pre-treatment values. The range of values reported for erythrocyte acetylcholinesterase activity at 24 h post administration included values which indicated that inhibition was > 20 % of pre-treatment values at each dose level and in the concurrent control group; 40 - 100 % (concurrent controls; 44 volunteers), 50 - 100 % at 0.1 - 1 mg kg-1 (17 volunteers), 60 - 100 % at 2 - 3 mg kg-1 (11 volunteers), 70 - 100 % at 4 - 6 mg kg-1 (11 volunteers), 65 - 100 % at 7 - 9 mg kg-1 (19 volunteers), 50 - 100 % at 10 - 12 mg kg-1 (7 volunteers), 55 - 95 % at 13 - 16 mg kg-1 (16 volunteers), 35 - 85 % at 17 - 20 mg kg-1 (11 volunteers), 40 - 85 % at 21 - 26 mg kg-1 (4 volunteers) and 55 - 75 % at 32 mg kg-1 (11 volunteers). Although the inhibition of erythrocyte acetylcholinesterase activity at the lower doses of dichlorvos administered is within the range of concurrent controls, the wide variation in the concurrent control activities compared with the pre-treatment control activities calls into question the reliability of the study. Given that a decrease of > 20 % in erythrocyte acetyl cholinesterase activity is considered a biologically significant effect, a NOAEL cannot be derived from this study. A study on nerve electrophysiology of patients after OP poisoning is available (Waida et al., 1987). Three cases of delayed neurotoxicity were reported following severe cholinergic poisoning allegedly due to dichlorvos. Initial treatment occurred at a different hospital and case histories were not given. In all cases the neuropathy started as leg pains and paraesthesia 10-15 d after poisoning. EMG studies showed normal conduction velocity in motor neurones of the leg but the action potential was very small (0.05 mV). Conduction in sensory neurones was unaffected. Severe axonal degeneration neuropathy was observed in two cases of dichlorvos poisoning. Recovery was seen in one patient within 12 months. A number of reports detailing suicide attempts (up to 1993) involving oral ingestion of large quantities of dichlorvos by adults are available (Li et al., 1989; Jadhav et al., 1989; Shinoda et al., 1972; Bidanset, 1978; Kraus et al., 1970; Saito & Chiba, 1981; Watanabe et al., 1976). There are no reports of accidental ingestion by adults or children. No further conclusions could be drawn from some reports because of the limited details presented or co-exposure to other pesticides (Bidanset, 1978; Kraus et al., 1970; Saito & Chiba, 1981). In other cases more details were available (Li et al., 1989; Jadhav et al., 1989; Shinoda et al., 1972; Watanabe et al., 1976). Although an estimation of the dose was presented which indicated that fatalities occurred following ingestion of up to 50 ml of product; it was not possible to further define the amount of dichlorvos ingested (Li et al., 1989). The cause of death was considered to be respiratory failure. Symptoms, prior to death or in non-fatal cases, developing as soon as 15 min after ingestion, consisted of excessive salivation, bronchial secretion, pulmonary oedema, lacrimation, respiratory failure and coma. Recovery after ingestion of large quantities was seen following intensive treatment, although no follow up details were presented (Li et al., 1989; Watanabe et al., 1976).

37

A Japanese study reveals 5 deaths from 16 cases of dichlorvos poisoning (no details of amounts was available), mostly from suicide attempts. However the study combines data from all cases of organophosphate poisonings and no other useful information could be extracted (Yamashita et al., 1997).

3.2.1.6 DERMAL

No reports were available documenting exposure to dichlorvos without solvents. A male PCO with 10 years experience was accidentally exposed to dichlorvos (1 % solution in mineral spirit) from a faulty unit whilst knapsack spraying (Bisby et al., 1975). After 10 min he noticed the leak from a faulty seal and changed his overalls, but not undergarments, and placed a plastic sheet under the sprayer. By the end of the shift he reported excessive tiredness, weakness, dizziness and breathing difficulties. After showering, the symptoms improved although the rapid shallow breathing lasted for 2 h. The PCO developed skin burns, consisting of extensive areas of erythema and bulla. When whole blood cholinesterase was measured 3 d after exposure, it was 36 % of normal but reached 78 % after a further month, when the man returned to work. As this was a mixed exposure it is unclear whether the irritant reaction was a result of dichlorvos exposure. A male 52 year old truck driver developed dermatitis on the neck, chest, hands and forearms 24 h following direct skin contact with a liquid containing 5 % dichlorvos, 15 % petroleum distillate and 80 % trichloroethane during an accidental spill (Mathias, 1983). Other symptoms were frontal headache, mild rhinorrhoea, bitter taste and burning sensation in mouth. A patch test using 0.1 and 1 % dichlorvos in petroleum was negative at 48 and 96 h. When tested initially, plasma cholinesterase activity was found to be at the low end of the normal range. Two weeks later activity was at the higher end of the normal range. A study reported a series of five cases of children (4 - 10 years old) presenting after poisoning by aerosol A (2 % propoxur and 0.5 % dichlorvos) in Singapore (Sim et al., 1980). In 4/5 cases the aerosol had been used to treat head lice. Clinical signs observed were: pupil constriction (5/5), vomiting (4/5), excessive sweating (3/5), drowsiness (3/5) and lacrimation (1/5). Recovery was seen in cases within 24 h. In 4/5 cases plasma cholinesterase activity was measured on admission, and was found to be 40 - 65 % of normal. However it was within the normal range after 3 d. Another report noted headache, nausea, vomiting, diarrhoea and difficulty in breathing following accidental dermal exposure to a product containing three active ingredients (including 0.5 % dichlorvos, methoprene and propoxur and the solvents trichloroethane, difluoroethane and dichloromethane) (Sidhu et al., 1989).

3.2.1.7 INHALATION

Six healthy adult males, no ages or weights presented, were exposed (head-only) to dichlorvos vapour ranging from 7 - 52 mg m-3 for between 20 - 240 min (Unpublished,

38

1969d; Unpublished, 1970c). It should be noted that each exposure group consisted of a single individual and in some cases the same individual was exposed to a number of different concentrations. No indication of ethical consent for the study was presented in the report. A total of 12 separate exposures were reported but there was no indication of the concentration to which each individual was exposed. Erythrocyte acetylcholinesterase and plasma cholinesterase were measured following exposure (Michel method). No dichlorvos related clinical symptoms were recorded. Dichlorvos could not be detected in exhaled air until the exposure concentration reached 40 mg m-3. Plasma cholinesterase was depressed after all exposures (26 - 88 % of pre-exposure levels) although individual results were extremely variable. No change in erythrocyte acetylcholinesterase activity was observed compared to pre-exposure levels following exposures to 6.3, 12 and 13 mg m-3 for 120, 120 and 240 min, respectively. Activity was decreased to 93, 81, 74 and 82 % of pre-exposure levels following exposure to 17.8 (90 min), 18.7 (120 min), 22.6 (105 min) and 38.3 mg m-3 (105 min). No change in activity was subsequently reported following exposures to 41.5, 44.1, 44.7, 44.7 and 52 mg m-3 for 24, 20, 20, 90 and 65 min, respectively. Thus, only exposure to 22.6 mg m-3 for 105 min produced a biologically significant decrease in erythrocyte acetylcholinesterase activity. However, no dose-response relationships were apparent, and it is strange that exposure to higher dose levels, particularly 44.7 and 52 mg m-3 for 90 and 65 min, respectively, produced no change in erythrocyte acetylcholinesterase activity. This questions the reliability of the measurements made during the study and so no meaningful conclusions can be drawn from this data. Adult volunteers (26 male and 5 female, aged 19 - 57 and between 54 - 94 kg males and 57 - 93 kg females) were exposed to dichlorvos vapour (0.90 - 1.22 mg m-3) for between 2 - 7.5 h (Unpublished, 1970d). Exposures took place in a large inhalation chamber. No indication of ethical consent for the study was presented in the report. Exposure for more than 6.5 h resulted in depression of plasma cholinesterase activity (to a minimum of 71 % pre-exposure values). There was no effect on erythrocyte acetylcholinesterase activity. Given that this study was performed by the same laboratory as the previous one, concerns over the reliability of the methodology exist and therefore no meaningful conclusions can be drawn from the data. Another study reported effects of dichlorvos (0.9 - 1.1 mg m-3 for 45 min) exposure on 8 human volunteers at operational aircraft altitudes (Smith et al., 1972). The protocol consisted of control periods (ground level and at a pressure equivalent to 8000 ft) and an exposure period (at 8000 ft). No indication of ethical consent for the study was presented in the report. Plasma and erythrocyte cholinesterase (pH-Stat method) activity was not significantly effected, neither was dark adaptation, sweat rate and bronchiolar resistance (both tests for cholinergic function). Erythrocyte acetylcholinesterase activity was not reported on an individual basis and therefore it is unclear whether any decreases to below 20 % of pre-exposure values occurred. Therefore no meaningful conclusions can be drawn from this study. A report is available of a case of a 52 year old woman with asthma, which appeared to be triggered by contact with her cat, which wore a collar impregnated with dichlorvos (Bryant 1985). Hospital tests showed no response to inhaling cat hair extract. However exposure for

39

1 h to dichlorvos released from a cat collar produced a reduction in forced expired volume. The response started 2 h after exposure, reaching a maximum at 6 h. No change in plasma cholinesterase was seen. When the patient removed the cat collar, and several dichlorvos strips from her home, the symptoms disappeared over a 3 month period. The authors considered that the response was indicative, but not diagnostic, of sensitisation. A study recorded the case history of a family (32 year old mother, 35 year old father and 14 year old daughter) exposed to dichlorvos and propoxur when their house was treated (Markowitz, 1992). After an acute exposure (approximately 12 h) the following symptoms were reported: burning of throat, chest heaviness, wheezing, shortness of breath, headaches, fatigue and nausea. Blood samples were taken 96 h after exposure. Erythrocyte acetylcholinesterase activity was at the lower end of normal (2.7 - 3.8 IU ml-1, normal range 3-5 IU ml-1) and serum cholinesterase was low (0.9-2.6 units ml-1, normal range 2.5-7.0 units ml-1). By day 45 activity of both enzymes were within the normal range.

3.2.1.8 OTHER ROUTES

No conclusions can be drawn from a number of fatalities, where the route was unspecified or following self-injection, because of the lack of information or involvement of more than one active ingredient (Moody & Terp, 1988; Kusic et al., 1991). Very limited details of the cases of 6 children, between 2-12 years, having bone marrow failure after exposure to pesticidal products (containing dichlorvos and propoxur) have been reported (Reeves et al., 1981). Exposure was claimed to last for 2 min - 2 d. Bone marrow failure was diagnosed between 4 d to 28 weeks after exposure.

3.2.1.9 CASE REPORTS FROM HSE AND NATIONAL POISONS INFORMATION SERVICE

A report is available from the National Poisons Information Service (NPIS) concerning incidents involving dichlorvos. It provides a breakdown of all incidents reported to NPIS between 1983-88 (Unpublished, 1991b). Further details were included on serious or unusual cases (1981-1990), which had been followed up by the NPIS. A total of 98 individual cases were reported to NPIS involving dichlorvos alone, or in combination with other agents. No information was provided on the concentration of dichlorvos in products, types and uses of products or whether the incidents followed misuse or normal use. Therefore it was not possible to determine from the report whether any of these cases involved the use of non-agricultural products. The majority of incidents were reported in the age group >12 years (69 %), children aged 5 - 12 accounted for 4 % of cases and 16 % of those under 5 years old. The route of exposure was known for 95 of the cases. The most frequent route of exposure was inhalation (42 %), followed by oral (29 %), mixed routes (12 %) and dermal (10 %). A small number of cases (2 %) involved exposure to the eyes. On initial contact, symptom severity was unknown in 14 % of cases, 22 % were considered asymptomatic, 49 % mild and 14 % moderate. The most common symptoms among the 98 cases were: skin irritation (7), cough (7), vomiting (11), nausea (11), malaise (5) and dizziness (6). The pattern of symptoms, in the 17 cases where follow up details were available, was slightly different: headache (6), diarrhoea (4), epigastric pain (4), nausea (3) and dizziness (3).

40

Over the period 1987-1993 one poisoning incident involving dichlorvos was reported to HSE's Pesticide Incident Appraisal Panel. A 25 year old male worker became dizzy and nauseous whilst working in a hut used to prepare a sea lice treatment product (Formulation A: 50 % dichlorvos). A further incident, involving an amateur user, has also been reported to HSE. The user cut up a large dichlorvos containing strip and placed sections in three small boxes. After a week the sealed boxes were opened indoors. Symptoms, including headache, sweating, sore throat, disorientation, giddiness, difficulty breathing, pin-point pupils and heart pains, developed within 24 h and lasted for 2 weeks. Between 1993-1999 there were 107 cases of dichlorvos poisoning reported to NPIS, 90 % of these involved home use products; 90 % were accidental; 21 % involved children; 80 % of those poisoned displayed some symptoms and 8 % of those were severe. There were also 3 cases of supposed dichlorvos exposure reported to PIAP, with a 50 % probability of symptoms being caused by dichlorvos.

3.2.1.10 SUMMARY

Reliable data in humans are available following acute oral exposure. Two modern volunteer studies are reported in which a single oral dose of up to 1 mg kg-1 dichlorvos did not significantly inhibit erythrocyte cholinesterase activity to values > 20 % of those found in baseline values, and no other clinical signs of toxicity were reported. In less reliable reports, fatalities have been reported following attempted suicide by oral ingestion of large quantities of dichlorvos. In no case was it possible to accurately estimate the amount of dichlorvos ingested. Commonly observed clinical signs (and typical of OP poisoning), consisting of salivation, excessive lacrimation, bronchial secretion, pulmonary oedema, breathing difficulties leading to respiratory failure and coma, were seen both in survivors and before death. Delayed neuropathy is reported in humans following poisoning with dichlorvos, however, the reliability of these studies is uncertain. No reports of dermal exposure to dichlorvos in the absence of solvents were available. No reliable inhalation studies in humans are available from which a NOAEC or LOAEC could be derived. Information is available from reliable animal studies involving exposure to dichlorvos via oral, dermal, and inhalation routes. Dichlorvos is toxic following oral exposure with rat LD50 values of between 46.4 - 105 mg kg-1, which indicate that EC classification criteria for Toxic if swallowed (25 - 200 mg kg-1) is fulfilled. Oral LD50 values of 87 - 184 mg kg-1 and 22.5 - 74 mg kg-1 in mice and rabbits, respectively, have also been reported. Rat dermal LD50 values of between 210 - 456 mg kg-1, which indicate the EC classification criteria for Toxic in contact with skin (50 - 400 mg kg-1) is fulfilled. Two well-conducted OECD guideline inhalation studies in rats indicate that aerosols of dichlorvos are severely toxic with LC50 values of 230 and 485 mg m-3 reported. Given the lower value, the EC classification criteria for Very Toxic by inhalation for aerosols (≤ 250 mg m-3) is fulfilled. In one of these studies a NOAEC of 28 mg m-3 following a 4 h exposure was identified. Signs of toxicity associated with cholinesterase inhibition were apparent generally within 1 h of exposure, though there was some evidence of recovery. There is a report of a single rabbit death following dichlorvos administration to the eye, however, this was not repeated in a further eye irritation study performed to an accepted protocol.

41

Significant reduction of brain acetylcholinesterase activity in rats was rapid, falling to 10 % of control activity within 15 min following oral administration of 50 mg kg-1. In one study no effect was seen at 8 mg kg-1, however, erythrocyte acetylcholinesterase activity was not measured so it is unclear whether this is a true NOAEL. Other studies have reported reductions in erythrocyte acetylcholinesterase activity (> 20 %) in rats and dogs at approximately 40 mg kg-1 and in monkeys at 80 mg kg-1 following oral administration. NOAELs could not be derived from animal studies following exposure via the dermal and inhalation routes due either to toxicity or no measurement of acetylcholinesterase activity (the most relevant/sensitive toxic sign of dichlorvos exposure). A well-conducted, specialised behavioural neurotoxicity study is available in the rat. A NOAEL of 0.5 mg kg-1 was apparent following acute administration based on signs of toxicity at higher dose levels (35 mg kg-1 and above), however, no acetylcholinesterase measurements were made, so no conclusions about the NOAEL for systemic toxicity could be drawn. No histopathological damage was observed. The potential of dichlorvos to cause delayed neuropathy has been investigated in the adult hen. There was no evidence of delayed neuropathy following a single gavage administration at a toxic dose level. In contrast older, non-standard acute studies in hens administered dichlorvos via the sc route have reported delayed neuropathy. However, given the data from the more robust modern study using the oral route, this finding is considered unreliable. A formulation of dichlorvos, insecticide resin strips (18.6 % dichlorvos) is reported to have a rat oral LD50 of 382 - 679 mg kg-1 and a dermal LD50 of 27400 mg kg-1. A rat LC50 value of 0.13 - 0.22 mg l-1 has been reported for Formulation A (50 % dichlorvos).

3.2.2 IRRITATION

3.2.2.1 SKIN

In a test performed to contemporary OECD guidelines, dichlorvos (purity 97.3 %; 500 µl) was applied to the skin of 3 rabbits (HC: NZW) for 4 h. Slight erythema and very slight oedema were reported 1 h post-exposure (mean scores 2 and 1, respectively) and remained slight to very slight up to 7 days post-exposure (mean scores 2, 1.67, 1.33 and 1.67 for erythema and 1, 1, 1 and 0.33 for oedema at 24 h, 48 h, 72 h and 7 days, respectively). These signs of irritation had resolved completely by day 14 post-exposure (Unpublished, 1984b). These results would not lead to classification for skin irritation on the basis of current EU guidelines. The dermal irritancy of dichlorvos was assessed in rabbits (Unpublished, 1986e). The study was carried out to a modified OECD Guideline and GLP. Due to the high acute dermal toxicity, 0.2 ml of undiluted dichlorvos was applied under a semi-occlusive patch for up to 200 min only. One rabbit developed convulsions and tremors within 60 min and died 2 h later. In the remaining 2 rabbits, tremors and ataxia were observed 3 h after application. Exposure was terminated after approximately 3 h 20 min. Clinical signs of toxicity had disappeared by day 2.

42

Erythema (maximum grade 1) was observed in both survivors 1 h after patch removal. This persisted until day 7 in both animals (maximum grade 2 at 48 h seen in 2/2 animals and at day 7 in 1/2). Oedema was observed 24 h after patch removal in both animals (maximum grade 1), which persisted to day 7 in one animal. Recovery was complete within 14 d. Although this test is not adequate for classification purposes it indicates a slight potential to cause skin irritation. In another study, essentially to OECD Guidelines and conducted to GLP, using 0.5 ml undiluted dichlorvos under a semi-occlusive dressing, all rabbits died within 1 h (Unpublished, 1986f). No skin observations were made. In a modern dermal acute toxicity study occasional erythema and scab formation were observed (Unpublished, 1986c). In repeated dose toxicity studies, no skin lesions were observed in rats at either 21.4 mg kg-1 d-1 or in guinea pigs at 100 mg kg-1 d-1 in poorly reported studies (Dikshith et al., 1976; Unpublished, 1966).

3.2.2.2 EYE

In a test performed to contemporary OECD guidelines, dichlorvos (purity 97.3 %; 100 µl) was applied to the conjunctival sac of one eye of 6 rabbits (HC: NZW). The treated eye was rinsed with saline 24 h post exposure. One rabbit died 25 minutes after administration of dichlorvos. Very slight corneal opacity was reported up to 72 h post-exposure (mean score 1), which resolved slowly but completely by 21 days post-administration. Only very slight iridial irritation was reported up to 24 h (mean score 0.8), which had completely resolved by 48 h post-exposure. Slight to very slight conjunctival redness was reported after 1 h (mean score 1.2), which resolved completely by 14 days post-exposure (mean scores 2, 2, 1.2 and 0.6 at 24h, 48 h, 72 h and 7 days post-exposure, respectively). Slight to moderate conjunctival oedema was reported 1 h post-exposure (mean score 2.2) but this resolved completely by day 7 post-exposure (mean scores 1.8, 1.2 and 0.8 at 24 h, 48 h and 72 h, respectively) (Unpublished, 1984b). These results would not lead to classification for eye irritation on the basis of current EC guidelines. A preliminary eye irritation study, carried out to GLP and essentially to OECD Guidelines, was conducted using 1 female NZW rabbit (Unpublished, 1986g). Undiluted dichlorvos, 0.1 ml, was instilled in the conjunctival sac of one eye. The animal died within 7 min. Severe signs of toxicity (lethargy, convulsions, muscle contractions and immobility) were seen. No eye observations were made.

3.2.2.3 SUMMARY

There are no reports of skin or eye irritation in humans that can be attributed to exposure to dichlorvos alone. Skin irritation studies performed to accepted protocols indicate that dichlorvos does not meet the criteria for classification as a skin irritant.

43

A well-conducted eye irritation study indicated that dichlorvos does not meet the criteria for classification as an eye irritant.

3.2.3 SENSITISATION

3.2.3.1 SKIN

3.2.3.1.1 Animal Data

The skin sensitising potential of dichlorvos (> 97 % pure) was investigated in a split adjuvant test conducted to OECD Guidelines and GLP (Unpublished, 1986h). The study used female Dunkin Hartley guinea-pigs (10 control and 20 test). The test animals received four 48 h dermal applications (to the shaved back) of 0.2 ml of test substance (10 % dichlorvos in petrolatum) under an occlusive patch on days 0, 2, 4 and 7. Test and control animals also received intradermal injections of 0.1 ml of Freunds complete adjuvant (FCA) on either side of the sensitisation area on day 4. The induction concentration of 10 % was used following an earlier study in which 100 % caused significant systemic toxicity. No skin irritation was seen at either 10 or 100 % in the absence of prior treatment with FCA. Two weeks after last induction, test and control animals were challenged by application of 0, 1, 2.5 and 10 % dichlorvos in petrolatum. Skin reactions were assessed 24 and 48 h after dressing removal. The challenge was repeated using the same control group 7 d after the first challenge, using 0, 0.5 and 1.0 % dichlorvos in petrolatum. A skin reaction graded 2 and above was considered to be positive. At first challenge skin reactions in test and controls at 2.5 and 10 % were severe (> 50 % graded 3 and above), possibly due a lowered irritation threshold by FCA. At 1 % dichlorvos, skin reactions (grade 2) were seen in 10/19 test animals and 2/9 controls. Skin reactions in test animals were also more prolonged. At the second challenge skin reaction (grade 2+) were seen in 20/20 test animals at 0.5 and 1 % compared to 6/10 at 0.5% and 7/10 at 1 % in the controls. The authors considered the skin reactions seen in controls might be due to sensitisation following the first challenge. This calls into question the reliability of the study and together with the non-standard protocol no firm conclusions can be drawn on the sensitising potential of dichlorvos. Limited details of a Magnusson and Kligman maximisation test using 10 female Hartley Guinea-pigs was available (Fujita, 1985). Induction concentrations of 5 % (intradermal) and 25 % (topical) dichlorvos were used. A sighting study indicated that erythema was observed after 24 h topical applications of > 2 % dichlorvos under occlusive patches. The challenge doses were therefore chosen as 0.05 and 0.5 % dichlorvos. Skin reactions were recorded at 24 and 48 h after patch removal. At 24 h after challenge with 0.5 %, 30 % of animals were reported as sensitised, rising to 40 % at 48 h. No sensitisation was observed following challenge at 0.05 % at either time. No details of controls were presented and therefore no firm conclusions can be drawn.

44

In a study investigating the skin sensitising potential of the individual components of a dichlorvos product, dichlorvos (10 %) was negative (Unpublished, 1968a). This study followed the flank/flank procedure of Stevens (1967). Individual results were not presented. The numbers of animals and dosing regime were not clear. Induction consisted of 3 topical applications of test materials (0.2 ml) at 24 h intervals (type of patch not specified). Challenge doses (0.2 ml) were applied to flanks 4 d after last application. The response was assessed 24 h later. Duration of exposure was not known. The control (dinitrobenzene, 0.32 %) gave a positive response.

3.2.3.1.2 Human Data

An analysis of 202 cases of dermatitis, allegedly caused by OPs, and reported by medical staff in Japan between 1972-1979 is available (Matsusuita et al 1985). Dichlorvos was estimated to be the single cause of dermatitis in 11/50 cases where one could be established and was a contributory agent in 55 cases. It is not clear whether the dermatitis was a result of irritation or evidence of sensitisation. A study is available in which 84 tea growers in Japan (between 50 - 70 % of whom were previously exposed to dichlorvos) were patch tested with 0.2 % dichlorvos on the inner side of the forearm (Fujita, 1985). Patches were removed after 24 h and reactions evaluated according to a proposal of the International Contact Dermatitis Research Group. Of the females tested, 29 were negative, 17 equivocal and 17 positive. Of the males tested, 14 were negative, 6 equivocal and 1 was positive. It is not possible to determine whether the positive results obtained were evidence of sensitisation or irritancy as previous exposure to dichlorvos in these subjects was not documented. Although this study may present evidence of skin sensitisation the significance of this finding cannot be assessed. There is a report of a 42-year-old woman who developed erythema, swelling, vesiculation and intense pruritus on the fingers of both hands after working for 3 months as a pest control operative handling a formulation containing dichlorvos and a formulation containing propoxur (Stoermer, 1985). She also developed similar cutaneous manifestations on the soles of her feet when spraying the pesticides wearing sandals or otherwise unprotected feet. The condition regressed and healed completely within one week following non-specific local treatment. Subsequent patch testing revealed clear positive results with the formulation containing dichlorvos, which persisted for 72 h post-administration. Administration of 1 % of the formulation containing dichlorvos produced erythema, oedema, nodules and vesicles at 24, 48 and 72 h; a 0.5 % solution produced erythema with oedema at 24 h, erythema, oedema and nodules at 48 h and erythema, oedema, nodules and vesicles at 72 h; a 0.1 % solution produced erythema with oedema at 24 h, erythema, oedema and nodules at 48 and 72 h; finally a 0.01 % solution produced no reaction at 24 h, erythema at 48 h and erythema with oedema at 72 h. The solvent used in the preparations of the formulation containing dichlorvos, erkosol and a 1 % solution of the formulation containing propoxur gave no reaction in the patch testing. Although these results indicate that dichlorvos is a potential skin sensitiser in humans, no firm conclusions can be drawn on the basis of a single patients patch test results.

45

3.2.3.2 RESPIRATORY

3.2.3.2.1 Animal Data

No data available.

3.2.3.2.2 Human Data

A case report, published in letter form only, describes a 26-year-old man (who smoked cigarettes each day, but had no previous history of respiratory complaint) who developed persistent asthma following inhalation exposure of dichlorvos (Deschamps et al., 1994). The patient was a cook who had worked in a closed room that had been treated the day before with ‘large quantities of dichlorvos diluted with xylene’, however, actual concentrations were not reported. Symptoms, which persisted for 8 days before medical advice was sought included, cough, dyspnoea and wheezing in the workplace, but eye irritation and rhinorrhoea were also noted. It is unclear from the report whether the patient continued working for these 8 days. Examination of the patient revealed ausculatory wheezes and a normal chest radiograph. Serum cholinesterase levels were decreased (5.5 mU ml-1, compared with reported lower normal limit of 8 mU ml-1) but returned to normal 18 days later. Skin tests for various aeroallergens were negative and serum IgE was normal. Respiratory symptoms regressed following 2 days of treatment with inhaled salbutamol. Pulmonary function tests 20 days following exposure were reported to show bronchial obstruction. The patient continued to have weekly asthma attacks when he attempted mild physical exercise and remained unable to do any exercise a year following exposure. Asthma attacks occurred nightly without treatment. Although this is of concern, the extent of the dichlorvos exposure is unclear and therefore other factors cannot be ruled out (smoking; environmental exposures) in contributing to the patient’s condition. Also, this case has many of the hallmarks of Reactive Airways Dysfunction Syndrome rather than classical asthma.

3.2.3.3 SUMMARY

No clear evidence is available of the potential of dichlorvos to produce skin sensitisation in humans (only one case report and instances of dermatitis in workers from which no firm conclusions could be drawn). There are no good animal skin sensitisation studies available. However, when the available human and animal evidence was deliberated by the EC Pesticide Classification and Labelling Working Group, dichlorvos was considered to match the criteria for classification as a skin sensitiser. No clear evidence is available of the potential of dichlorvos to induce asthma. Only one case showing exposure conditions and symptoms consistent with Reactive Airways Dysfunction Syndrome has been reported. There is no data to show that dichlorvos can induce respiratory sensitisation in animals.

46

3.2.4 REPEAT DOSE TOXICITY

3.2.4.1 SUBCHRONIC TOXICITY

3.2.4.1.1 Oral

Rat A 13-week gavage study performed to GLP and consistent with current EU guidelines has been conducted in rats. Rats (Crl: CD(SD)BR albino; 10 per sex per group) were administered gavage doses of 0, 0.1, 1.5 or 15 mg kg-1 d-1 dichlorvos (purity 98.3 %), 5 d wk-

1 for 13 weeks. Blood was taken from all animals during week 7 (3 h post-dose) for measurement of erythrocyte acetylcholinesterase and plasma cholinesterase (method not stated) activity. These measurements were also made at the end of the study in addition to the standard haematological and clinical chemistry investigations. Brain acetylcholinesterase activity was also measured at the end of the study. Necropsy was performed on all animals and included macroscopic examination of the external surface of the body; all orifices; the cranial cavity; the external surfaces of the brain and spinal cord; the nasal cavity and paranasal sinuses; and the thoracic, abdominal and pelvic cavities and viscera. The weights of the following organs were collected from all animals: brain, kidneys, liver, ovaries and testes. Histopathological examination was performed on all tissues of the control and high dose groups; and additionally any macroscopic lesions, lungs, liver and kidneys of animals of the 0.1 and 1.5 mg kg-1 d-1 groups. No treatment-related deaths occurred during the study. Salivation and urine staining were observed sporadically in the second half of the study in some top dose animals 30 - 60 min post-dosing. No treatment-related changes in bodyweight, food consumption or in ophthalmic observations occurred. Haemoglobin, haematocrit and erythrocyte count were statistically significantly decreased in males of the 1.5 and 15 mg kg-1 d-1 groups and females of the 15 mg kg-1 d-1 group. Serum cholesterol was statistically significantly increased in high dose males. Plasma cholinesterase and erythrocyte acetylcholinesterase activity was decreased in a dose-related manner after 7 weeks and the levels in most cases remained similar when the measurements were repeated at the end of the study. Statistically significantly lower values were reported in plasma in both sexes at 15 mg kg-1 d-1 (35 - 65 % of control values in males; approximately 50 % in females) and in 1.5 mg kg-1 d-1 males (approximately 75 % of control values); in erythrocytes in both sexes at 1.5 and 15 mg kg-1 d-1 (approximately 75 and 55 % of control values in males and 75 and 60 % of control values in females, respectively). A significant decrease was also reported in 0.1 mg kg-1 d-1 females, however, this was 92 % of control values and is unlikely to be of biological significance. Brain cholinesterase levels were only toxicologically significantly decreased at 15 mg kg-1 d-1 in females (51 % of control values); and at 15 mg kg-1 d-1 in males (71 % of control values). No treatment-related macroscopic or microscopic changes were reported (Unpublished, 1988c). Overall a NOAEL of 0.1 mg kg-1 d-1 can be established from this study following 13-week oral exposure of rats to dichlorvos, based on inhibition of erythrocyte acetylcholinesterase activity at 1.5 mg kg-1 d-1.

47

In a study performed to GLP and EPA/FIFRA guidelines, rats (Sprague-Dawley; 15 per sex per group) were administered gavage doses of 0, 0.1, 7.5 or 15 mg kg-1 d-1 dichlorvos (purity 97.9 %), 7 d wk-1 for 91 days. These dose levels were chosen on the basis of results from a previous 90-day study in rats. Within each group 10 animals per sex were allocated for cholinesterase determinations and 5 animals/sex for neuropathology examination. Animals were observed twice daily for mortality/moribundity; clinical signs were observed daily and also at the time of peak effect (15 minutes post-dosing) during the treatment period; and bodyweights were recorded weekly. A FOB (see Section 3.2.1.1.1 for details) was conducted on 10 animals per sex per group (5 animals each from the cholinesterase and neuropathology groups) pre-administration and then during the 4th, 8th and 13th weeks of exposure. Locomotor activity was measured immediately after completion of the FOB, over a 41-minute period using the 'Digiscan Micro Animal Activity System'. Plasma cholinesterase and erythrocyte acetylcholinesterase activity (using a method that utilised the Ellman reaction) was evaluated pre-administration, during the 4th and 8th weeks of exposure and at study termination. Brain acetylcholinesterase activity was evaluated at study termination in the following regions: olfactory region, cerebellum, hippocampus, cerebral cortex, brain stem and midbrain. A full necropsy was conducted on all animals killed in extremis, including examination of the external surface, all orifices, and the cranial, thoracic and pelvic cavities including the viscera. Following excision the brain of animals in the cholinesterase group was weighed and dissected into the regions described above. Investigations of the animals in the neuropathology group included measurement of brain weight (excluding olfactory bulbs) and brain dimensions (length and width). Any gross changes, abnormal colouration or lesions of the brain and spinal cord were recorded. Nerve tissues were examined histopathologically from the control and 15 mg kg-1 groups, and included the central nervous system tissues brain (forebrain, centre of cerebrum, midbrain, cerebellum and pons, and the medulla oblongata); spinal cord (at cervical swellings C3-C8, and at lumbar swellings T13-L4); gasserian ganglion/trigeminal nerves; lumbar dorsal root ganglion at T13-L4; lumbar dorsal root fibres at T13-L4; lumbar ventral root fibres at T13-L4; cervical dorsal root ganglion C3-C8; cervical dorsal root fibres C3-C8; cervical ventral root fibres C3-C8; optic nerves; and eyes. Also, peripheral nervous system tissues were examined including sciatic nerves (mid-thigh region and at sciatic notch); sural nerves; tibial nerves; peroneal nerves; and forelimbs. One animal was killed in extremis during the study, a male from the 0.1 mg kg-1 d-1 group, however, this was not considered to be caused by dichlorvos exposure. Signs of toxicity associated with dichlorvos exposure occurred predominantly 15 minutes post-administration in the 15 mg kg-1 d-1 group and to a much lesser extent in the 7.5 mg kg-1 d-1 group. The signs were observed initially in the first week of dosing with 15 mg kg-1 d-1 and persisted to the end of the study, whereas in the 7.5 mg kg-1 d-1 group the findings (tremors) were first observed in the third week of exposure and only occasionally persisted to the end of the study. All animals of the 15 mg kg-1 d-1 group showed tremors (forelimb/hindlimb, whole body and repetitive movement of the jaw) and salivation; while a majority showed exophthalmus, lacrimation and a clear material on the forelimbs. The incidences of rales, chromodacryorrhoea and the finding of red, yellow and/or orange material around the mouth was increased when compared to control animals. Animals of the 7.5 mg kg-1 d-1 group showed some of these findings, most notably tremors and sporadic observations of salivation

48

and exophthalmus, but to a much lesser extent. No treatment-related findings were reported following exposure to 0.1 mg kg-1 d-1. Group mean bodyweight gain of females of the 15 mg kg-1 d-1 was statistically significantly lower than control (79 % of control values) at the end of the study, however, group mean bodyweight was approximately 91 % of control values. No difference was observed in males or in the other treatment groups. The FOB testing revealed no significant differences in any parameter tested between treated and control animals apart from decreased mean bodyweight of top dose females. Measurement of locomotor activity revealed no differences between treated and control groups. Mean plasma cholinesterase activity was reduced in both males and females following administration of 7.5 mg kg-1 d-1 (63, 70 and 58 % and 56, 45 and 44 % of control values, respectively, at weeks 3, 7 and 13) and 15 mg kg-1 d-1 (66, 64 and 51% and 57, 49 and 42 % of control values, respectively, at weeks 3, 7 & 13). These values were all statistically significantly when compared with controls, apart from the value for males of the 7.5 mg kg-1 d-1 group at week 7. Mean erythrocyte acetylcholinesterase activities were not toxicologically significantly reduced apart from a reduction in males of the 15 mg kg-1 d-1 group to 65 % of control values during week 3 (the significance of this result is unclear). Brain acetylcholinesterase activity was significantly reduced in the brainstem of males of the 15 mg kg-1 d-1 group (84 % of control values); and in the cerebral cortex of both males and females of the 7.5 (88 and 89 % of control values, respectively) and 15 mg kg-1 d-1 groups (85 and 90 % of control values, respectively), although these are not levels of inhibition that signify concern. Animals exposed to 0.1 mg kg-1 d-1 did not show any toxicologically significant changes in cholinesterase activity. Necropsy of the animal killed in extremis revealed no treatment-related changes. No toxicologically-significant treatment-related changes were observed in terms of brain weights or dimensions between the control and treated groups. Histopathological examination of nerve tissues revealed no toxicologically-significant treatment-related changes (Unpublished, 1993b). Overall, the signs of toxicity observed in animals of the top 2 dose groups of this study are consistent with what would be expected following exposure to a cholinesterase inhibitor. No toxicologically significant reduction in erythrocyte or brain cholinesterase levels was observed in treated animals. A NOAEL of 0.1 mg kg-1 d-1 can be derived from this study. A study reported by Desi and Nagymajtenyi (1999) was designed to investigate the changes of spontaneous and evoked electrical changes in the central and peripheral nervous systems following exposure to moderate doses of dichlorvos following subchronic exposure (up to 12 weeks), over 3 consecutive generations, or during different stages of development. During the subchronic study male rats (Wistar; 10 per group) were administered gavage doses of either 0, 0.98, 1.96 or 3.92 mg kg-1 d-1 dichlorvos (purity 98 %) in distilled water, 5 d wk-1 for 4, 8 or 12 weeks. Following treatment the rats were anaesthetised, the left side of the brain exposed and electrodes placed on the primary somatosensory (Par1), visual (Oc1M) and auditory centres (Te1). Electrocorticograms (ECoG) were subsequently recorded from these areas over a 5 minute period; and calculations of mean amplitude, mean frequency and power spectrum analysis made. Following the ECoG recordings, parameters of cortical

49

sensory evoked potentials (latency times) were measured following visual, acoustic and somatosensory stimulation (via electrodes in the nasal skin). The conduction velocity, and relative and absolute refractory periods in the tail nerve was recorded. The rats were then sacrificed and the weights of the brain, liver, heart, lung, kidneys, thymus and adrenal glands determined. Brain acetylcholinesterase activity was also measured. No signs of toxicity were reported during exposure and no difference in body weight or organ weights was observed between treated and control animals. The authors report that the ECoG investigation showed dose and time-dependent decreases in mean amplitude; and decreases in slow part and increases in fast part of power spectra. The somatosensory, visual and auditory ECoG mean frequencies and latency times of evoked potentials showed dose and time-dependent increases which were statistically significant after 8 weeks in the top dose group only and after 12 weeks in the 1.96 and 3.92 mg kg-1 groups. Although all dose groups showed decreased tail nerve conduction velocities, these only reached statistical significance compared with controls in top dose animals after 8 and 12 weeks. Absolute refractory period became longer in all treated groups, but only reached statistical significance compared with controls in the top dose group after 8 weeks and after 12 weeks in the 1.96 and 3.92 mg kg-1 groups. Although brain cholinesterase activity was reported to be inhibited in the treated groups, the inhibition did not reach statistical significance. Overall, the results indicate the occurrence of electrical changes in the nervous system following treatment with 1.96 mg kg-1 d-1 and above across all the repeated exposure regimes. A NOAEL of 0.98 mg kg-1 d-1 is apparent from all these studies for the parameters tested. A NOAEL for systemic toxicity could not be established as erythrocyte acetylcholinesterase activity was not measured. The following studies in rats were considered in the 1994 HSE review of dichlorvos, but are now considered of limited value given the better quality rat studies that are now available. They have been included as additional information. In a 15 week feeding study weanling CD rats (15 per sex per dose) received nominal concentrations of 0, 0.1, 1, 10, 100 and 1000 ppm of dichlorvos (93 % pure) (Unpublished, 1964b). Urinalysis, on 5/sex/dose, was conducted at the end of the treatment period. The remaining rats were subjected to gross pathology (but not histopathology), haematology, plasma cholinesterase, and erythrocyte and brain acetylcholinesterase activity determination. No mortalities or signs of toxicity were observed. There were no significant differences in food consumption, body weights, or body weight gain at the end of the study. At 1000 ppm during the first half of the study, male and female rats showed retarded body weight gain compared to the controls. There were no significant changes in relative organ weights, haematology or urinalyses. Plasma cholinesterase and erythrocyte acetylcholinesterase activities were reduced in females at 10 ppm and in both sexes at 100 and 1000 ppm. In both sexes brain acetylcholinesterase activity was reduced at 1000 ppm only (to approximately 40 % of control). No gross pathological changes attributable to dichlorvos were observed. The study was limited by a failure to measure the actual concentration in the diet and the lack of histopathology, and therefore a NOAEL cannot be derived from this study.

50

Two general repeated dose studies are available in the published literature. The information presented was of limited value since neither report provides information on the purity of dichlorvos used nor one study used an irregular dosing regimen with imprecise quantification of dose. Male albino rats (10 per dose) were administered 0, 167 or 333 ppm dichlorvos (purity unknown) in the diet for 45 days (El-Shimy et al., 1983). Haematology was performed after days 21 and 42. At post mortem the weight of the liver, spleen, lungs, heart, kidneys, testes and brain was determined. At 333 ppm, 4 animals died. Clinical signs of toxicity (muscular fibrillation, laboured breathing, diarrhoea and increased frequency of micturation) were observed at this dose only. A reduction in food consumption was observed throughout the study at 333 ppm and on days 14 to 35 at 167 ppm. The only significant change in organ weight was a 75 % decrease in the absolute spleen weight at 333 ppm. Reduced haemoglobin levels were seen in all treated rats on day 42 (approximately 20 % at 167 ppm and 35 % at 333 ppm). The erythrocytes were reported as poorly staining (hypochromia) but details were not presented. A substantial reduction in white blood cells count was seen at 333 ppm on days 21 and 42 with no change in the differential count. Blood glucose concentration was increased in both treated groups at day 35 (by 158 and 162 % respectively). At day 42, glucose concentration was still increased at 167 ppm (141 %) but was reduced at 333 ppm (51 %). The authors attributed this to a loss of appetite. Whole blood cholinesterase was significantly reduced in both groups, reaching about 30 % of control at 333 ppm by day 42. Female Wistar rats (6 per group) were given 0 or 5 % of the LD50 (equivalent to 3.52 mg kg-

1) by gavage every 72 h for 117 d (40 doses) (Gajewski and Katkiewicz, 1981). Rats were sacrificed 1 h after the last dose. Erythrocyte acetylcholinesterase activity was unchanged. Acetylcholinesterase activity was similar in the three regions of the brain examined, the cortex, medulla and hypothalamus (approximately 38 % of control). Muscle and liver acetylcholinesterase activities were reduced (to 85 % and 48 % of controls respectively). Poorly reported histological results suggested that hyperaemia was present in the myocardium (5/5) and liver (5/5). Other findings quoted were 'extravasations (2/5 in myocardium), necrobiosis (3/5 in myocardium) and vacuolar degeneration (5/5 in liver)'. No lesions were observed in 6 control rats examined. In a number of short-term studies, primarily investigating cholinesterase activity after dichlorvos treatment, male rats received up to 40 mg kg-1 d-1 by gavage for 14 d (Teichert et al., 1976; Pachecka et al., 1975; Maslinska et al., 1980). Doses of 4 mg kg-1 d-1 and above were found to significantly reduce brain acetylcholinesterase activity. No effect was observed after dosing at 1.6 mg kg-1 d-1 by gavage for 90 d or at 0.2 mg kg-1 d-1 in the diet for 14 d (Pachecka et al., 1975; Nag & Nandi, 1991). Two poorly reported studies stated that doses of up to 50 % LD50 d-1 for 14 days reduced brain acetylcholinesterase and brain mitochondrial cytochrome oxidase activities with no effect at 2 % LD50 d-1 for up to 90 d (Sitkiewicz & Zalewska, 1975b; Brzezinski & Wysocka-Paruszewska, 1980). Dog

51

A 52-week oral exposure study, performed to GLP and consistent with EC guidelines, has been conducted. Dogs (Beagle; 4 per sex per group) were administered 0, 0.05, 1.0 or 3.0 mg kg-1 d-1 dichlorvos purity 100 %) in the form of gelatin capsules. The low dose group were administered 0.1 mg kg-1 d-1 for the first 4 weeks of the study, but this was reduced to 0.05 mg kg-1 d-1, when an inhibition of plasma cholinesterase was noted, in order to try and achieve a no effect level. Standard haematology, clinical chemistry and urinalysis were performed pre-dosing and during weeks 26 and 52 of the study. In addition, plasma cholinesterase and erythrocyte acetylcholinesterase activities (method not stated) were assessed 3 times in the week prior to treatment and during weeks 2, 6, 13, 26, 39 and 52 (approximately 3 h post-dose). Brain acetylcholinesterase activity was measured in animals following sacrifice. Necropsy was performed on all animals and included macroscopic examination of the external surface of the body; all orifices; the cranial cavity; the carcass; the external surfaces of the brain and spinal cord; the cut surfaces of the brain and spinal cord; the nasal cavity and paranasal sinuses; and the thoracic, abdominal and pelvic cavities and viscera; cervical tissue and organs. The weights of the following organs were collected from all animals: adrenals, brain (including brainstem), heart, kidneys, liver, ovaries and testes. Histopathological examination was performed on a wide range of tissues. No mortality was reported during the study. Treatment-related signs of toxicity included ataxia, salivation and dyspnoea following an overdose in a single top dose male. Incidences of soft faeces and emesis (containing compound or capsule) were reported throughout the study in most animals, and were greater in treated animals compared with controls. No treatment-related changes in mean bodyweight, bodyweight gain, food consumption or ophthalmoscopic abnormalities were reported. Haematology, clinical chemistry and urine analyses revealed no treatment-related changes between treated and control groups. Following treatment with 0.1 mg kg-1 d-1 plasma cholinesterase activity were reduced to approximately 77 % of pre-treatment levels; once the dose was changed to 0.05 mg kg-1 d-1 levels returned to within the range of the control group (93 - 114 % of pre-treatment levels) for the remainder of the study. In the other dose group’s plasma cholinesterase was inhibited in a dose-related manner by week 2 (41 and 34 % of pre-treatment values at 1 and 3 mg kg-1 d-1, respectively) and remained inhibited for the remainder of the study (41 - 59 % and 26 - 39 %, respectively). Erythrocyte acetylcholinesterase activity was similar to pre-treatment values following 2 weeks exposure to 0.1 mg kg-1 d-1; at the 6 week time point levels had activity was inhibited to 74 and 50 % of pre-treatment values in males and females, respectively. Activity had recovered to pre-treatment levels by week 13 and onwards (94 - 108 %). It is likely that the inhibition reported during week 6 was a delayed response to treatment with 0.1 mg kg-1 d-1 rather than a contemporary response to 0.05 mg kg-1 d-1. Again the other treated groups showed a dose-response related inhibition by week 2 (67 and 29 % of pre-treatment values at 1 and 3 mg kg-1 d-1, respectively); a maximum inhibition during week 6 (36 and 8 % of pre-treatment values, respectively); and clear inhibition of activity thereafter (46 - 62 and 13 - 21 % of pre-treatment values, respectively). Brain acetylcholinesterase activity of animals treated with 0.05 mg kg-1 d-1 and females treated with 1 mg kg-1 d-1 was similar to activity in control animals; but was toxicologically significantly reduced in males of the 1 mg kg-1 d-1 group (78 % of control values) and all animals of the 3 mg kg-1 d-1 group (53 and 71 % of control values in males and females, respectively). No treatment-related macroscopic or microscopic changes were reported (Unpublished, 1990b).

52

Overall, this study shows inhibition of acetylcholinesterase activity (brain and erythrocyte) in the absence of any other overt signs of toxicity following oral exposure of dogs to 1 and 3 mg kg-1 d-1 for 52 weeks. A NOAEL of 0.05 mg kg-1 d-1 can be identified from this study. Pure bred Beagles aged 6-12 months (3 per sex per dose), received 0, 0.32, 0.96 or 1.42 mg kg-1 d-1 dichlorvos (93 % pure) in gelatin capsules for up to 90 d (Unpublished, 1962). Animals initially receiving 0.32 mg kg-1 d-1 for the first 21 d then received 2.65 mg kg-1 d-1 for the remainder of the study. A further group (3 per sex) was later initiated which received 0.32 mg kg-1 d-1 for 90 d. Plasma cholinesterase and erythrocyte acetylcholinesterase determinations were made approximately every two weeks. Brain acetylcholinesterase activity was determined at termination. Organ weights, haematology, limited clinical chemistry (BUN and bilirubin levels) and histopathology were determined at the end of the study. There were no deaths. All animals at 2.65 mg kg-1 d-1 and 2/6 at 1.42 mg kg-1 d-1 displayed greater hyperactivity and excitability (increased movement in cages and aggression) than the controls. The authors of the study considered these behavioural effects to be treatment related. There were no significant differences in weight gain, relative or absolute organ weight, haematology, clinical chemistry or plasma cholinesterase during the study. Toxicologically significant reductions in erythrocyte acetylcholinesterase activity were reported at various time points throughout the study at dose levels of 0.96 mg kg-1 d-1 and above. Brain acetylcholinesterase activity was reduced at 2.65 mg kg-1 d-1, to 38 % of control values and slightly reduced at 1.42 mg kg-1 d-1 (88 % of controls). The NOAEL is 0.32 mg kg-1 d-1 based on erythrocyte acetylcholinesterase inhibition at 0.96 mg kg-1 d-1. Hen In a delayed neuropathy study performed to GLP and EPA guidelines (and consistent with current OECD guidelines), adult female domestic hens (21 per group) were administered gavage doses of 0, 0.3, 1 or 3 mg kg-1 d-1 dichlorvos (purity 97.9 %) in distilled water for 28 d. The positive control group of birds was administered 7.5 mg kg-1 d-1 TOCP in corn oil. The birds were observed for up to 77 days following the start of treatment. This included daily examinations for signs of delayed locomotor ataxia, which on treatment days occurred immediately after being dosed. Surviving animals (6 per group) were sacrificed on day 49 and day 77 of the study respectively. Determinations of brain acetylcholinesterase activity, brain and spinal cord neuropathy target esterase (NTE) activity were performed 4h after administration on day 4 and on day 30 in 3 animals per group. Gross necropsy was performed on all mortalities after day 7 and on those animals undergoing scheduled sacrifice. Histopathology was performed on the following tissues from the animals undergoing scheduled sacrifice: forebrain, mid and hind brain, upper cervical spinal cord, lower cervical spinal cord, thoracic spinal cord, lumbar spinal cord, proximal sciatic nerve, distal sciatic nerve (above knee) and tibial nerve (distal nerve). Following this study additional groups of birds (3 per group) were administered gavage doses of 0 or 0.1 mg kg-1 d-1 dichlorvos for determination of brain cholinesterase activities on day 30. Also, the report describes a further additional study carried out to the same protocol as above with birds administered 0, 10, 12.5, 15 and 20 mg kg-1 d-1 TOCP.

53

Five birds died during the study (1 at 1 mg kg-1 d-1 and 4 at 3 mg kg-1 d-1). Toxicologically- relevant clinical signs of toxicity, including subdued behaviour and unsteady gait, were observed during the study only in the 3 mg kg-1 d-1 group. These occurred in the majority of birds from around day 8 of administration, but diminished with time and all surviving birds appeared normal by day 30. There was no evidence of delayed motor ataxia in any of the groups (including positive controls). A treatment-related decrease in group mean bodyweight was only observed in birds administered 3 mg kg-1 d-1. This decrease to 83 % of control values by day 14 was followed by a gradual recovery to control levels by day 35. Brain acetylcholinesterase activity was inhibited following exposure to dichlorvos on both day 4 (93, 56 and 37 % of control values at 0.3, 1 and 3 mg kg-1 d-1, respectively) and day 30 (74, 66 and 46 % of control values, respectively). The additional group tested at 0.1 mg kg-1 d-1 showed no effects on brain cholinesterase activity on day 30. There was no inhibition observed in the TOCP group. No treatment-related effects were observed on either brain or spinal cord NTE activity following dichlorvos exposure. However, both were inhibited following exposure to TOCP (50 and 77 %, respectively on day 4; 39 and 45 %, respectively on day 30). Gross necropsy revealed no treatment-related changes. Histopathological examination of animals sacrificed on day 49 revealed minimal axonal degeneration in 2/6 birds of the 3 mg kg-1 group in two levels of the spinal cord; and in one bird in each of the 1 and 0.3 mg kg-1 groups in one level of the spinal cord. Animals exposed to TOCP showed minimal or moderate axonal degeneration in two levels of the spinal cord (2/6); and minimal axonal degeneration in the cerebellum (2/6) (one animal showing both). No axonal degeneration graded as minimal or above was observed in the negative control animals. Examination of animals sacrificed on day 77 revealed minimal axonal degeneration in 1/5 birds of the 3 mg kg-1 group in two levels of the spinal cord; and in 1/6 birds in the 1 mg kg-1 group in three levels of the spinal cord. No effects were observed in the remaining animals of the 0.3 mg kg-1 group. All animals exposed to TOCP showed minimal to moderate axonal degeneration in one or more levels of the spinal cord, and one also showed minimal swollen axons in the medulla. The additional study involving higher TOCP exposure reported higher incidences of axonal degeneration, and in the top dose group a large proportion of the birds were killed on humane grounds due to ataxia (Unpublished, 1994a). Overall, the absence of delayed ataxia, lack of inhibition of NTE activity and only weak histopathological evidence of nerve damage indicates that exposure to dichlorvos, at dose levels causing death, does not produce evidence of delayed neuropathy. An increased inhibition of brain acetylcholinesterase activity was reported on day 30 in animals exposed to 0.3 mg kg-1 d-1 (> 20 % of control values). The histopathological slides of the above study were independently assessed by a representative of the American College of Veterinary Pathologists who concluded that ‘no neurotoxic effect, including and ability to induce organophosphorus ester-induced delayed neuropathy (OPIDN), was detected for any of the dichlorvos dose levels used.’ (Unpublished, 1994b). This view was subsequently endorsed by a US Pathology Working Group (Unpublished, 1998).

54

The following studies were included in the 1994 review of dichlorvos and are included as additional information. Hens, Hyline White Leghorn pullets 3 per dose, received dermal applications of 0, 0.6, 1.7, 3.3 or 14.4 mg kg-1 d-1 or oral doses of 0, 3.5 or 6.6 mg kg-1 d-1 dichlorvos (99.9% pure) for up to 90 d (Francis et al., 1985). Dermal doses of 14.4 mg kg-1 d-1 or oral administration of 6.6 mg kg-1 d-1 caused severe signs of toxicity, followed rapidly by death. The condition of hens receiving dermal applications of 1.7 or 3.3 mg kg-1 d-1 deteriorated and hens died between days 30-45. Before death all hens showed ataxia grade 2 (staggers) and 5/6 developed grade 3 (rests on hocks). No signs were seen at the lower application rate, or after oral dosing at 3.5 mg kg-1 d-1. No histopathology was performed. The authors stated that characteristic OPIDN symptoms were not observed. Other Species A number of limited studies are available in the rabbit and Rhesus monkey (Desi et al., 1978; 1980; Yamazaki et al., 1975; Maslinska et al., 1981). Two reported a dose related decrease in immune function in rabbits following 6 weeks gavage administration of either 2.5 mg kg-1 d-1 or 2.5 % LD50 dichlorvos (Desi et al., 1978; 1980). A study in juvenile rabbits investigated acetylcholinesterase activity and 5-HT concentrations in specific areas of the brain and spinal cord following gavage administration of 9 mg kg-1 d-1 dichlorvos for 10 d (Maslinska et al., 1981). Reductions in acetylcholinesterase activity were reported in all sections investigated, whereas both increases and reductions in 5-HT concentrations were reported. A report documents electron microscopy findings in a range of tissues (external ocular muscle, liver, kidney, intercostal muscles and sacral nerves) from rabbits and Rhesus monkeys administered dichlorvos (97. 6 % pure) orally for 3 months at up to 5.0 mg kg-1 d-1 or 1.0 mg kg-1 d-1 respectively (Yamazaki et al., 1975). Findings reported were changes in the neuromuscular junction (reduction in synaptic vesicles and disarrangement of the myofilaments), mitochondria (swelling and disordering of cristae with sparsely distributed atrophy), liver (binuclear cells, vacuolisation of the hepatocytes, increase in amount of SER and RER, decrease in glycogen granules, decrease and flattening of villi in biliary canaliculi and Kupffer's cells became swollen with numerous phagocytic granules) and muscle (increase in myelin bodies and atrophy of muscle fibres). The significance of these findings is unclear.

3.2.4.1.2 Summary of Oral Studies

Two well conducted 13-week gavage studies in rats both identified NOAEL of 0.1 mg kg-1 d-1, based on signs of toxicity consistent with dichlorvos acting as an acetylcholinesterase inhibitor or significant inhibition (> 20 %) of erythrocyte acetylcholinesterase activity at higher dose levels. This figure was consistent with a NOAEL of 0.05 mg kg-1 d-1 in a 52 week dog study, based on significant inhibition of brain and erythrocyte acetylcholinesterase activity. A well conducted study in hens revealed no evidence of delayed neuropathy at dose levels producing severe systemic toxicity.

3.2.4.1.3 Dermal

55

In a study by Luty et al. (1998) 0, 7.5 and 37.5 mg kg-1 d-1 dichlorvos in a 20 % water alcohol solution was applied to the tail skin of female rats (Wistar; 10 per group) 4 h d-1 for 28 d. Blood was then removed from the rats to evaluate the effect of exposure on the granulocyte system. Following sacrifice histopathology was performed on the lungs, liver, heart, kidneys, brain, thymus, spleen and lymphatic nodules. Although histopathological changes are reported in the exposure groups, their significance is unclear as they have not been compared with findings in control animals (either concurrent or historical). The granulocyte investigation revealed a dose-related statistically significant increased bactericidal function of neutrophils, and a statistically significant increase in phagocytosis index in the top dose group. A NOAEL could not be identified from the study as the key toxicological endpoint of acetylcholinesterase inhibition was not measured. In a brief summary, lacking full details of dichlorvos application and full histopathology examination, no signs of skin lesions were seen in rats at 21.4 mg kg-1 d-1 for 90 d (Dikshith et al., 1976). In a second study, limited by the unusual dosing regime, brain acetylcholinesterase activity was reduced in rats after dermal applications of 0 or 2.94 mg kg-

1 every 72 h for 117 d (40 doses) (Gajewski & Katkiewicz, 1981). In a short study in guinea pigs, lacking details of application method and histopathology, plasma and erythrocyte cholinesterase activities were significantly reduced at 100 mg kg-1 d-1 for 8 d (Unpublished, 1966). Effects of undiluted dichlorvos on the skin were reported as negligible.

3.2.4.1.4 Summary Of Dermal Studies

No adequate studies using dermal application were available. However, in a single non-standard study, a significant reduction in brain acetylcholinesterase activity was seen in the rat at 2.9 mg kg-1 d-1 (40 applications over 117 d).

3.2.4.1.5 Inhalation

Rat Three studies are available which essentially investigated inhibition of cholinesterase activity (Unpublished, 1968b; Sasinovich, 1967; Schmidt et al., 1975). In the first study, brain acetylcholinesterase activity was significantly reduced in rats exposed (whole body) to 1.2 mg m-3 dichlorvos vapour for 23 h d-1 for 5 d (Unpublished, 1968b). No effects on brain acetylcholinesterase activity were seen at 0.14 mg m-3 (however it is not clear if erythrocyte acetylcholinesterase was measured). In the second study, exposure of rats to 0.11 mg m-3 4 h d-1 for 120 d revealed no treatment-related changes, including no change in brain and erythrocyte acetylcholinesterase activity reduced (method of measurement not stated) (Sasinovich, 1967). Brain acetylcholinesterase was slightly reduced (92 % of control values) after exposure to 1.07 mg m-3 4 h d-1 for 120 d, but liver (70 %) and serum cholinesterase (78 %) and erythrocyte acetylcholinesterase (76 %) activities were reduced. Rats, exposed to 5.2 mg m-3 for 4 h d-1 for 60 d, showed a reduction

56

in activity of brain acetylcholinesterase (78 % of control values), serum cholinesterase (42 %) and erythrocyte cholinesterase (78 %). Exposure to a mean concentration of 8.2 mg m-3 for 54 d resulted in mortalities (2/8) on days 30 and 45, and loss of appetite and weight and apathy in survivors. Overall, a NOAEC in rats of 0.11 mg m-3 following 4 h d-1 exposure for 120 d can be derived. In the third study rats were exposed (whole body) to up to 4.3 mg m-3 for 14 d dichlorvos vapour generated by suspending cut-up Vapona strips in cages (Schmidt et al., 1975). At 0.8 and 1.8 mg m-3 significantly reduced acetylcholinesterase activity in the bronchial tissue (50 - 60 % of control activity) but no changes in whole blood acetylcholinesterase activity were observed. At 4.3 mg m-3 blood cholinesterase activity was also reduced. At 0.2 mg m-3 (stated to be the usual in use concentration for a Vapona strip in a well ventilated room) no significant changes in whole blood acetylcholinesterase activity were seen. It is unclear whether this is a NOAEC as there is no indication if erythrocyte acetylcholinesterase is also inhibited. Mice In a study to investigate cholinesterase inhibition CF1 mice (10 per sex per dose) were exposed (whole body) to average concentrations of 0, 0.14 and 1.2 mg m-3 dichlorvos vapour for 23 h d-1 for 5 d (Unpublished, 1968b). At 1.2 mg m-3 brain and erythrocyte acetylcholinesterase activity was significantly reduced (approximately 70 % of control) as was the activity of plasma cholinesterases. No effects were seen on any acetylcholinesterase activity at 0.14 mg m-3, indicating that this is a NOAEC in mice. Monkeys A group of 8 monkeys (4 per sex) was exposed to 0.051 mg m-3 dichlorvos for 23 h d-1, 7 d wk-1 for 3 months (Unpublished, 1977). A control group (4 male, 1 female) was kept under the same environmental conditions but not exposed to dichlorvos. Haematology and clinical chemistry were performed at termination. No mortalities or signs of toxicity were observed. No changes in haematology or clinical chemistry were seen. Plasma cholinesterase and erythrocyte acetylcholinesterase activities were reduced (to approximately 75 % and 60-70 % of pre-exposure values respectively). Electromyographic studies showed no treatment related changes. Guinea pigs Pirbright Albino strain guinea pigs (10 per sex per dose) were exposed (whole body) to average concentrations of 0, 0.14 or 1.2 mg m-3 dichlorvos vapour in inhalation chambers for 23 h d-1 for 5 d (Unpublished, 1968b). At the end of the study, blood and brain tissue were taken for determination of cholinesterase activity. No clinical signs of toxicity were observed. Brain acetylcholinesterase activity was only reduced at 1.2 mg m-3 (to 87% of control values in both sexes). No effects were seen on brain acetylcholinesterase activity at 0.14 mg m-3 (it is unclear whether this is a NOAEC as there is no indication if erythrocyte acetylcholinesterase is also inhibited).

57

3.2.4.1.6 Summary Of Inhalation Studies

A NOAEC of 0.11 mg m-3 (equivalent to 0.006 mg kg-1 d-1) was observed in rats following exposure for 4 h d-1 for 120 d, based on decreased brain and erythrocyte cholinesterase activities at higher dose levels. A NOAEC of 0.14 mg m-3 (equivalent to 0.21 mg kg-1 d-1) was observed in mice following exposure for 23 h d-1 for 5 d, based on decreased brain and erythrocyte cholinesterase activities at higher dose levels.

3.2.4.2 HUMAN DATA

3.2.4.2.1 Oral

A single-blind, placebo controlled, randomised study, conducted to GCP/GLP and to a protocol approved by an independent Ethics Committee, investigated the effects of repeated oral exposure of dichlorvos on erythrocyte acetylcholinesterase activity in healthy male volunteers. Six volunteers were administered 7 mg d-1 dichlorvos (purity 98 %) in corn oil in the form of a capsule for 21 d; and 3 volunteers were administered a capsule of corn oil daily over the same time period. The mean weight of the volunteers was approximately 75 kg, therefore the administered dose was approximately 0.1 mg kg-1 d-1. Pre-study blood samples for measurement of erythrocyte acetylcholinesterase activity (using a modified Ellman method) were taken 3 times in each of the 2 weeks before the start of the study; further measurements were made prior to dosing on days 1, 2, 4, 7, 9, 11, 14, 16 and 18; and a single measurement was made in each volunteer within 9 days of the final dose. Urine samples were analysed for drugs of abuse pre-study, and also collected at intervals during the study. Symptom forms were completed 24 h following each dose. No symptoms related to dichlorvos exposure were reported. Erythrocyte acetylcholinesterase activity was reduced in a time dependent manner following dichlorvos exposure, falling to approximately 77 - 92 % of pre-dose activity, at the end of dosing, with two volunteers showing a decrease in erythrocyte acetylcholinesterase activity of > 20% at the end of the dosing period as shown in Table 3.1. Also, it is noted that a further volunteer showed a biologically significant reduction in erythrocyte acetylcholinesterase activity 9 days after the final dose. Control values remained in the range 97 - 104 % of pre-dose values, and were statistically significantly greater than the dichlorvos group values from day 7 onwards. No urinalysis data was reported. Overall, the study shows that oral exposure of humans to approximately 0.1 mg kg-1 d-1 dichlorvos for 21 days leads to significantly decreased levels of erythrocyte acetylcholinesterase activity (> 20 % of pre-dosing values) in 2 of 6 volunteers. Thus, a NOAEL cannot be derived from this study (Unpublished, 1997c).

Table 3.1 Individual erythrocyte acetylcholinesterase values as a percentage of pre-dosing values in control and treated volunteers following oral administration

of 0.1 mg kg-1 d-1 dichlorvos for 21 days.

Day Controls Treated

58

A B C D E F G H I 1 94 98 99 98 101 111 95 95 97 2 98 97 100 100 98 89 96 94 93 4 102 99 103 100 94 95 96 93 94 7 98 98 104 89 92 83 98 94 92 9 96 105 100 88 94 87 103 91 101 11 105 106 101 91 91 86 95 88 91 14 103 97 99 84 90 83 90 88 84 16 100 87 99 86 90 83 84 90 81 18 103 92 101 92 90 84 78* 83 77* Post- dose

93 97 100 82 82 79* 89 85 82

The second phase of a study investigating erythrocyte cholinesterase inhibition involved the same 6 healthy male volunteers as previously described (see Section 3.2.1.5.1). Each volunteer was administered a 21 mg capsule (approximately 0.3 mg kg-1 d-1) of dichlorvos in corn oil for either 12 days (volunteers 5 & 6) or 15 days (volunteers 1 - 4). Pre-study blood samples for measurement of erythrocyte cholinesterase activity were as previously described; and erythrocyte cholinesterase activity was subsequently measured on days 3, 5, 8, 10, 12, 15, 17, 19, 22, 24, 26, 29, 33, 40, 47 and 54 after the first dose, immediately prior to dosing were appropriate. Pre-study urine analysis was as before; and 24 h samples underwent analysis for 5 days following the first dose and following dose 10. Symptom forms were completed 24 h following each dose. No treatment-related signs of toxicity were reported in the volunteers. As shown in Table 3.2 mean erythrocyte cholinesterase activity decreased during dosing to 73 % of pre-dose values by day 15. Subsequently activity reached a minimum of 69 % of control values by day 22, before recovering to 91 % of control values by day 54. No urinalysis data was reported. Overall, the study shows that oral exposure of humans to 0.3 mg kg-1 d-1 dichlorvos for 12 - 15 days leads to decreased levels of erythrocyte cholinesterase activity (> 20 % of pre-dosing values), which recover once exposure ceases. A NOAEL cannot be derived from this study (Unpublished, 1997a). Table 3.2 Individual erythrocyte acetylcholinesterase values as a percentage of pre-

dosing values in volunteers administered 0.3 mg kg-1 d-1 dichlorvos for either 12 days (volunteers 5 & 6) or 15 days (volunteers 1 - 4)

Days

after first Volunteers

59

dose 1 2 3 4 5 6 3 90 88 86 91 76* 112 5 84 92 87 86 83 80 8 81 86 77* 73* 73* 81 10 73* 85 79* 77* 73* 77* 12 - 84 74* 80 70* 73* 15 71* 79* 71* 75* 70* 76* 17 70* 72* 66* 74* 72* 73* 19 71* 76* 69* 73* 76* 79* 22 69* 74* 65* 61* 75* 72* 24 74* 78* 61* 75* 72* 72* 26 77* 63* 80 71* 72* 29 79* 73* 72* 72* 72* 33 86 76* 80 77* 82 40 - 81 - 83 87 47 - 81 - 84 86 54 100 85 88 89 94

Volunteers (4 per treated group, 2 per control group), after informed consent, received 0, 1, 2 or 4 mg kg-1 d-1 dichlorvos formulated as PVC capsules for 7 d or 8, 16 or 32 mg kg-1 d-1 for 1- 3 d (Slomka & Hine, 1981). Plasma cholinesterase and erythrocyte acetylcholinesterase activities were assessed 24 and 48 h after the last dose and the lowest result was reported as a percentage of the pre-treatment values. A dose-related decrease in plasma cholinesterase activity, to 10 - 20 % of pre-treatment values at 4 mg kg-1 d-1, was seen with similar reductions in volunteers given higher doses for shorter periods. The range of values reported for erythrocyte acetylcholinesterase activity indicated that inhibition was > 20 % of pre-treatment values at each dose level; 90 - 100 % in concurrent controls, 70 - 95 % at 1 mg kg-1 d-1, 55 - 75 % at 2 mg kg-1 d-1, 40 - 55 % at 4 mg kg-1 d-1, 40 - 55 % at 8 mg kg-1 d-1, 35 - 45 % at 316 mg kg-1 d-1 and 20 - 50 % at 32 mg kg-1 d-1. A second group of volunteers (2 per dosing regime) received increasing doses of dichlorvos (1, 2, 4, 8 or 16 mg kg-1 d-1), with each being maintained for 7 d. No individual was given more than two dose increments. Activity of the cholinesterases was assessed at the end of each dose period. The results were similar to those from the fixed dose study. The in vivo release rate was 27 % over 24 h. A NOAEL cannot be derived from this study due to the significant inhibition of erythrocyte acetylcholinesterase activity (> 20 % of pre-treatment controls) at all dose levels. Healthy male prisoner volunteers (5 per dose and 2 per matched control group, weight unknown) received gelatin capsules containing 1.0, 1.5, 2.0 or 2.5 mg d-1 dichlorvos for 28 d (Unpublished, 1967a). No indication of ethical consent for the study was presented in the report. Dosing started after a control period (of between 14 - 28 days) when subjects received corn oil capsules. Plasma and erythrocyte cholinesterase activities (Michel method) were determined during this period to give pre-exposure levels. Below 2 mg d-1 no effect on plasma or erythrocyte cholinesterase activity was seen. At 2 mg d-1 plasma cholinesterase activity was inhibited from day 7 onwards, falling to 71 % of pre-exposure levels on day 28. Dosing at 2.5 mg d-1 ceased when plasma cholinesterase activity fell to 70 % of pre-exposure

60

on day 20. However activity had returned to pre-exposure levels 15 d later. There was no effect on erythrocyte acetylcholinesterase activity. In the second phase of the study male volunteers (10 exposed and 2 control) received 0 or 1.5 mg d-1 dichlorvos in gelatin capsules for 60 d. Plasma cholinesterase activity was 76 - 87 % of control pre-exposure levels from days 16 to 60, but had recovered within 14 d of exposure ceasing. No effect on erythrocyte acetylcholinesterase activity or blood count was observed at 1.5 mg d-1 for 60 d, which is (assuming a mean bodyweight of 60 kg) approximately equivalent to 0.025 mg kg-1 d-1. Male prisoner volunteers (6 per group, weight unknown) received a total dose of 0 or 2.7 mg d-1 dichlorvos, either divided into three daily pre-meal capsules, three daily doses in a gelatin mix with each meal, three placebo capsules or three gelatin mixes, for 21 d (Boyer et al., 1977). No indication of ethical consent for the study was presented in the report. Cholinesterase activity (by titration) was expressed as a percentage of the pre-treatment average taken over 3 weeks. It was stated that only plasma cholinesterase was inhibited to a significant, activity showing a time related inhibition, reaching a minimum activity of 70 % of pre-treatment levels when dichlorvos was given with meals or 60 % when given as pre-meal capsules. However, no data was presented to indicate the extent of the erythrocyte acetylcholinesterase inhibition and therefore no meaningful conclusions can be drawn from the data.

3.2.4.2.2 Inhalation

Adult male volunteers (3, weight 64 - 82 kg) were exposed to dichlorvos vapour (0.90 - 1.22 mg m-3) for between 5 - 7.5 h d-1 for 4 d (Unpublished, 1970d). No indication of ethical consent for the study was presented in the report. Exposures were carried out in a large inhalation chamber. No change in plasma cholinesterase activity was seen in 2 volunteers. In the third, activity gradually declined to 63 % of pre-exposure values by day 4. Exposure had no effect on erythrocyte acetylcholinesterase activity. A study into the exposure of healthy newborn babies to dichlorvos is available (Cavagna et al., 1970). Two groups of 22 babies were exposed for about 18 h d-1 for 5 d to, either an environment of 1 strip containing 18.8 % dichlorvos per 40 m3 with good ventilation (recommended rate) or 1 strip per 30 m3 in poor ventilation, 20 babies served as unexposed controls. The actual air concentrations of dichlorvos were unknown. Plasma and erythrocyte cholinesterase activity was determined at birth using blood from umbilical cord, and at 5 d. No dose related effect on either plasma or erythrocyte cholinesterase activity was seen.

3.2.4.2.3 Mixed Routes Of Exposure

Studies in Workers During Manufacture A study of blood cholinesterase levels in a group of 30 workers in a production plant in China is available (Jian & Zhiying, 1990). The plant produced dichlorvos to a purity of 80

61

%. Age and sex matched controls (20) were also examined. Blood samples were taken at 0900, 1300, 1700 and 2100 h on a single day. The report stated the air level in the plant was 7.6 mg m-3. The average blood cholinesterase activity (erythrocyte acetylcholinesterase was not measured) was always significantly depressed in exposed workers to 67 - 75 % of the control value. A small diurnal change (highest level at 0900 h) was observed in the controls. This was not seen in the exposed group. A study reported the monitoring of 567 industrial operators working on a mixed OP and carbamate production line between 1970 and 1979 (Wilhelm & Bradamante, 1980). The control group consisted of apparently healthy individuals (257 males, 151 females) with no history of exposure to anticholinesterase compounds. The average working day was 8 h, but could rise to 12 h. Standard PPE, consisting of overalls, rubber gloves and boots and face protection was occasionally used. The majority of workers, when tested, had whole blood cholinesterase activities between 75 - 100 % of normal. Symptoms, such as tiredness, sleepiness, headache and vomiting were noted in all operators having cholinesterase activities below 25 %. A study investigated a group of workers involved with impregnating absorbent pads with dichlorvos and their subsequent incorporation into vaporisation units (Mason, 2000). During work individuals had begun reporting flu-like symptoms, tiredness and respiratory symptoms of wheezing and tightness of the chest. At the same time symptoms were reported, the production equipment was causing unstated problems which meant greater potential dermal exposure, also shift patterns had changed leading to longer, consecutive daily shifts for some workers. The mean measured atmospheric dichlorvos level was 1.15 mg m-3. It is not clear if this value represents an 8-hour TWA (Time Weighted Average) value. It was not noted whether the air sampling was conducted with personal or static sampling and dermal exposures were not measured or estimated. The workers were under a regular health surveillance programme involving blood cholinesterase monitoring. Index blood samples from 20 subjects involved in the production line were measured for erythrocyte acetylcholinesterase and plasma cholinesterase activities by the automated discrete kinetic method of Lewis et al. (1981). Mean erythrocyte acetylcholinesterase activity was 55 % (range 24 - 89 %) of baseline levels. Mean plasma cholinesterase activity was 20 % (range 2 - 75 %) of baseline levels. The eight subjects with lowest acetylcholinesterase activity levels (24 - 43 % of baseline, measured within 1 day of last exposure), who had been removed from exposure, were subsequently monitored for up to 61 days for the recovery of enzyme activity levels. Erythrocyte acetylcholinesterase activity showed a linear recovery with activity recovering at 0.8 % of baseline activity per day. In contrast, plasma cholinesterase activity showed an exponential recovery with a half-life of recovery of approximately 12 days. A study investigated a group of Swiss workers manufacturing dichlorvos-releasing vaporisers (Menz et al., 1974). The group consisted of 5 production workers (4 males and 1 female) and 8 processing workers (7 males and 1 female) who were exposed for 8 months of the year for up to 216 h per month. Medical and laboratory investigations were performed at the beginning and end of the 8 month production run but blood cholinesterase activity was

62

measured weekly. Workers used PPE consisting of rubber apron, boots and disposable gloves. Dichlorvos air levels within the factory showed considerable fluctuation with an average concentration of 0.7 mg m-3 (highest observed value 3 mg m-3); no personal exposures were recorded. Separate values for production and processing areas were not available. In 11/13 workers plasma cholinesterase levels were significantly depressed to about 60 % of pre-exposure levels. The average erythrocyte acetylcholinesterase activity was approximately 65 % of pre-exposure levels. It was not possible to correlate an individual's exposure with plasma or erythrocyte cholinesterase activities. The other haematological and biochemical factors measured in blood and urine were within normal limits, as were the medical examinations. Studies in Pest Control Operators Using Dichlorvos Products A field study of exposure of operators and consumers in 20 single-family houses in Nebraska USA is available (Gold et al., 1984). Dichlorvos was applied as a 0.5 % aqueous emulsion in a band treatment, prepared from a 24.7 % concentrate, to control Blattella germanica. Operators wore PPE consisting of polyester jump suit (with long sleeves), hard hat, respirator and rubber gloves. Pads were attached to outer clothing and to skin beneath clothing at head, forearm just above the wrist, leg just above the ankle, chest and back. Hands were washed in ethanol/water. The mean time to treat each property was 25.5 min. Total dermal exposure, calculated by summing the values determined from hands, interior and exterior skin pads, was 2.35 mg h-1. The respiratory exposure was 0.037 mg h-1, calculated by multiplying the air concentration of dichlorvos by the average ventilation rate for a man doing light work (1740 l h-1). The total exposure was therefore 2.39 mg h-1 or 0.028 mg kg-1 h-1. The authors calculated that 20 % of the dichlorvos penetrated the outer garments. The most heavily contaminated areas were the legs (including thighs and feet) and then the front and back trunk. Two applicators were also monitored for changes to serum and erythrocyte cholinesterase activity. The first treated 7 homes with the last exposure at 7 h. No decrease in erythrocyte acetylcholinesterase activity was observed. Plasma cholinesterase activity varied considerably (31 - 82 % of pre-exposure). The second applicator treated 13 homes with the last exposure to dichlorvos occurring at 56 h, no further details of work pattern were available. At 8 h the serum cholinesterase activity had fallen to 79 % but activity recovered by 24 h and no further changes were seen. Erythrocyte acetylcholinesterase activity decreased from 98.7 % at 24 h to 87.1 % at 30 h, reaching 85 % of pre-exposure values by 56 h. Urine samples taken at the same time as blood samples failed to show any evidence of dichlorvos or dichloroacetic acid (a metabolite); the assay sensitivity was reported as 1 ppb. The air concentration in the treated buildings averaged 0.21 mg m-3over the first 24 h. This was not associated with any change in erythrocyte acetylcholinesterase activity in the 20 residents examined 24 h after the homes were treated, nor was there any evidence of dichlorvos in their urine. Pads were placed in rooms before spraying at sites away from treatment areas. Over the first two hours post-treatment a mean of 0.32 µg cm-2 h-1 of dichlorvos was detected. Thus in this study, exposure of one operator to an estimated dose of 0.15 mg/kg intermittently over a 56 h period resulted in a decrease in erythrocyte acetylcholinesterase activity to 85 % of pre-exposure levels. Exposure of a second operator

63

(estimated dose 0.08 mg kg-1) resulted in no change in erythrocyte acetylcholinesterase activity. This study indicates that the exposure of 20 residents to 0.21 mg m-3 dichlorvos did not result in a decrease in erythrocyte acetylcholinesterase activity, however it is unclear how long and to what extent the exposure occurred. Therefore no meaningful conclusions can be drawn from this study. A study reported exposure of 13 volunteer applicators using aerosol cans (containing 7 % dichlorvos) and a 0.5 % emulsion spray for pest control in the home (Das et al., 1983). The applicators each carried out a full day's work using 10-14 aerosol cans (230-330 g of dichlorvos) and 10.2-11.9 l of spray treating 4 homes. The following PPE was worn; goggles, cap, respirator, coat, gloves and shoes. Exposure was monitored using 10 x 10 cm pads (6 on chest and 6 on back) and a sterile filter placed in the respirator. The mean dichlorvos residues detected on the back pads was 7.1 µg m-², on the chest 3.8 µg m-² and 45.3 µg m-² on the respirator filter (size unspecified). Urine from 3 applicators was analysed for the presence of DMP (a metabolite of dichlorvos) pre-shift, 3, 6 and 18 h post-shift. Blood was taken pre- and post-shift and the following morning. Analysis of blood included haematology, blood chemistry and serum cholinesterase measurement. No changes were seen in any of these parameters. A total of between 0.32-1.39 µg of DMP was recovered in urine sample taken at 3 h. By 18 h the concentration was below the limit of detection (2 ng per 80 ml of urine). A study reported exposure of workers during a malaria eradication programme in Haiti (Stein et al., 1966). They either visited homes to replace old plastic cages containing dichlorvos strips with new cages (12 men) or removed the old wax strips, replacing them with new strips at the field office (4 men). These latter workers had rubber gloves but no further protection was provided. Determinations of cholinesterase activity were made before the programme and weekly during it. The maximum air concentration recorded when workers were replacing old strips was 2.13 mg m-3. There was no significant change in erythrocyte acetylcholinesterase activity. Plasma cholinesterase was depressed in both types of worker (to between 40-60 % of pre-exposure levels) and correlated with their degree of exposure. No illness that was considered to be relevant to dichlorvos exposure was observed. Two reports of ocular disturbances in pesticide workers are available. In the first study 2 of 4 operators, who sprayed fenitrothion and dichlorvos 3 to 5 times per week, showed a decrease in amplitude of some waves on electroretinography (Yoshikawa et al., 1991). All operators had low plasma and/or erythrocyte (sic) cholinesterase activity and detectable alkyl phosphate levels in the urine. The second study reported the case of a 53 year old male presenting with retinitis pigmentosa 10 years after a 15 year's exposure to dichlorvos and other OPs (Utsumi & Miyata, 1977). Blood tests taken on diagnosis showed a reduction of erythrocyte cholinesterase compared to population controls, no change in plasma cholinesterase activity and a plasma dichlorvos concentration of 0.023 ppm. Studies In Consumers Exposed To Dichlorvos Products

64

Three trials were set up in Arizona to investigate exposure to dichlorvos containing strips (Leary et al., 1974). In the third trial, strips were installed in 4 homes at the standard rate of 1/28 m3, with minimal ventilation; 2 control homes contained placebo strips. One month after installation, new strips were placed in the kitchen and dining room at average of 1 strip per 1/14 m3 room space. All strips were removed two weeks later. Air and food samples were taken pre-exposure, during standard exposure, exaggerated exposure and post exposure. Regular cholinesterase assays (Michel method) were performed on 7 adults and 11 children (9 - 20 years) from treated homes and 4 adults and 2 children (3 - 5 years) from control homes. No significant difference was observed in plasma or erythrocyte cholinesterase activities. Air concentration, for strips used at a standard rate, peaked at 0.13 mg m-3 but reached a plateau of 0.09 mg m-3 on days 13-28. After increasing the number of strips, on day 28, the peak air level was 0.16 mg m-3. When strips were removed on day 42 the level was 0.11 mg m-3. The maximum concentration of dichlorvos detected in food was 0.12 ppm. Increasing the application rate did not result in a substantial increase in the amount detected in food. No firm conclusions can be drawn from this study as it is unclear what the individual exposures were and over what length of time. In the two earlier trials no attempt was made to measure air concentrations and so no meaningful conclusions can be drawn from these studies. For the first trial, 5 homes were treated with dichlorvos strips and 2 homes acted as controls using PVC strips containing no dichlorvos. Two types of strip, both containing 20 % Vapona (equivalent to 18.5 % dichlorvos) were used, the dimensions were 10" x 2.5" x 0.25" or 2" x 2.5" x 0.25". Strips were used according to instructions (1 per 28 m3 and replaced every 3 months). Two families also used the small strips in all clothing cupboards. Home occupants kept daily records of: indoor temperature, humidity, time spent in the home and illness. Families were not aware of the identity of test or control homes. Blood cholinesterase activity (from adults and children) was determined at regular intervals. There was no appreciable difference in the amount of time spent in the house (between 55 - 70 % over a 12 month period). The assays of cholinesterase activity did not show any treatment related variation. In the second trial, 16 families took part, 4 acted as controls using placebo strips. Strips were placed in homes at the rate of 1 strip per 28 m3 room space and replaced monthly for six months. Records were kept as in the first study. No dichlorvos related symptoms were observed. Blood samples were taken from adults and children in both control and test homes. From month 2 onwards occupants of treated houses had a slight but significant depression of plasma cholinesterase activity (to 40 - 85 % of control levels). Erythrocyte cholinesterase was not depressed in a treatment-related fashion. A study investigated plasma and erythrocyte cholinesterase activity in patients exposed to dichlorvos in hospital wards (Cavagna et al., 1969). Dichlorvos strips (20 % Vapona) were used at the recommended rate of 1 per 30 m3. A group of 66 men (aged 19 - 73) were admitted to hospital for a variety of medical conditions excluding liver disease. Of these men, 22 were exposed to dichlorvos aerial concentrations of 0.1 - 0.28 mg m-3 for periods of up to 16 d. Only 5 men were confined to bed and therefore exposed 24 h d-1. The remaining 44 subjects were exposed to concentrations < 0.1 mg m-3. Only patients exposed for 24 h d-1 to concentrations > 0.1 mg m-3 showed reductions in plasma cholinesterase activity (to 27 - 65 % of pre-exposure values). No change in erythrocyte acetylcholinesterase activity was seen in any group.

65

A small group (6 men, aged 38 - 64) of patients suffering from a variety of liver diseases were exposed to dichlorvos vapour. Plasma cholinesterase activity was reduced in all patients (to between 33 - 75 % of pre-exposure levels). Reductions were seen in patients exposed to < 0.1 mg m-3 for 16 h d-1 as well as higher concentrations. Erythrocyte acetylcholinesterase activity was unaffected. Babies were exposed to dichlorvos 0.1 - 0.21 mg m-3 (11 subjects) or 0.04 - 0.1 mg m-3 (5 subjects) for 9 - 33 d. All babies exposed to > 0.1 mg m-3 showed a reduction in plasma cholinesterase activity (65 - 91 % of pre-exposure values). Erythrocyte acetylcholinesterase activity was unaffected in both exposure groups. No changes in either plasma or erythrocyte cholinesterase activities were seen in children (6, aged 2-7 years) exposed to 0.04 - 0.21 mg m-3 for about 16 h d-1. Cotton clothes were stored for 20 d in a cupboard containing a mini-strip (weight 20 g, dichlorvos concentration unknown). When removed from cupboard the average dichlorvos content was 0.26 - 0.4 µg g-1 cotton. No changes in plasma or erythrocyte cholinesterase activities were seen in children wearing these clothes. No firm conclusions can be drawn from this study as it is unclear what concentrations of dichlorvos each individual is exposed to (no personal sampling). A report was submitted on consumer exposure to dichlorvos by use of resin vaporiser strips consisting of two studies in the home over 6 months and 2 months respectively and in a dermal toxicity study (Zavon & Kindel, 1966). The formulation consisted of 20 % dichlorvos in a resin, designed to release small quantities of dichlorvos over a prolonged period of time. In all 3 studies cholinesterase activity was measured using the Michel method. Two families, consisting of 2 members each, are reported to have volunteered to have the resin vaporiser strips in their homes for 6 months; one strip being installed per 1000 ft3 (28.32 m3) of air. Erythrocyte acetylcholinesterase activity was determined both before and after the strips were placed in the homes. During the first 4 months new strips were installed monthly, with the strip installed at the beginning of the fourth month remaining in place to the end of the sixth month. The subjects kept daily accounts of the amount of time spent in the house and a record of the temperature and humidity was made. Air levels of dichlorvos were measured only once during the study, approximately one month after the final strip was installed. Temperatures inside the 2 houses ranged between 60 - 82 oF; while humidity ranged from between 30 - 90 %, the lower values being in the one house with air conditioning. The only measurement of dichlorvos revealed levels of 0.87 and 0.97 mg m-3 in the 2 houses, respectively. Erythrocyte acetylcholinesterase activity was only measured once in each individual before the strips were installed, consequently only a single value can be used to calculate changes in activity following instillation of the strips (this analysis was not performed by the authors of the study). Measurements in all 4 individuals showed biologically significant reductions in activity at least once during the study (Table 3.3). It is also noted that 2 of the 4 subjects showed biologically significant reductions in activity at the end of the study. There was no reporting of the times that the subjects had spent in their

66

houses other than subjects 1 and 2, spent more time at home than subjects 3 and 4. However, the authors report that there was no link between variations in cholinesterase activity and exposure times. The absence of any form of personal sampling means there is no indication of what the true exposures were to each individual. Overall, a NOAEL cannot be derived from this study due to all 4 subjects showing a biologically significant reduction in erythrocyte acetylcholinesterase activity.

Table 3.3 Erythrocyte cholinesterase activity as a percentage of baseline values in subjects exposed to vapona resin vaporiser strip for up to 6 months.

Erythrocyte cholinesterase activity (%)

House 1 House 2 Days after first strips installed

Subject 1 Subject 2 Subject 3 Subject 4 1 117 126 115 116 7 92 98 - - 14 106 135 96 116 21 87 95 90 93

New strips 28 79* 90 - 105

42 80 73* 112 75* 56 96 100 96 104 63 New strips 70 94 100 93 100 84 89 97 88 88 86 New strips 112 59* 69* - 75* 154 92 82 96 82 184 85 89 78* 75*

The second study was similar to the 6 month study except it lasted only 2 months and involved an additional 6 families and a total of 16 subjects, including 2 placebo subjects. Where possible 3 baseline measurements were made, though in 7 of the 16 only 1 or 2 measurements were made. Where possible a mean baseline value has been used to calculate changes in erythrocyte cholinesterase activity. However, the information about the timings of both the instillation of the strips is limited and it is impossible to estimate the timings at which the cholinesterase measurements were made. Nevertheless, 7 of the 14 subjects exposed to the strips showed a biologically significant reduction in erythrocyte acetylcholinesterase activity, although 1 of the 2 placebo subjects also showed this effect, which calls into question the reliability of the measurement technique. There were no measurements of air levels of exposure made in this study. Overall, a NOAEL cannot be derived from this study due to biologically significant reductions in erythrocyte acetylcholinesterase activity. The dermal study was conducted in 10 volunteers (6m, 4f). There is no indication in the report that ethical consent was sought for the study. After an initial sighting study on 2 of the volunteers, they were split into 2 groups. The first group handled the strip for 30 minutes each day for 5 days; while the second group had a portion of a strip fixed to their arms with adhesive tape for 30 minutes a day for 5 days. Erythrocyte acetylcholinesterase activity was

67

determined before the first period of contact and after the second and the fifth. No biologically significant reduction in erythrocyte acetylcholinesterase activity was reported in any of the volunteers. However, there is no indication of how much dichlorvos the volunteers were potentially exposed to and therefore it is not possible to have any indication of how much dichlorvos was available to pass through the skin. Therefore no conclusions about human dermal exposure to dichlorvos can be drawn from this study. A study reported on 14 volunteers exposed to dichlorvos slow release strips in either homes or offices (Ottevanger, 1975). No air level data were presented. No changes in either blood cholinesterase or EMG were observed when strips were used according to manufacturer's instructions. Two studies are available that report plasma and erythrocyte cholinesterase activity in residents whose homes contained dichlorvos dispensers (Gratz et al., 1969; Funckes et al., 1963). Several types of dispenser were used but these did not include a modern PVC type. No difference in plasma or erythrocyte cholinesterase activity was seen in occupants exposed to air concentrations of up to 0.84 mg m-3 dichlorvos for periods of up to 10 weeks.

3.2.4.3 CHRONIC TOXICITY

3.2.4.3.1 Oral

Rat A two year combined chronic toxicity and carcinogenicity feeding study in CD rats is available (Unpublished, 1967b). Dichlorvos (93 % pure) was administered to weanling rats (40 per sex per dose) at nominal concentrations of 0, 0.1, 1, 10, 100 or 500 ppm (actual concentrations averaged 0.05, 0.5, 4.7, 47 and 234 ppm). Within each group 25 rats per sex were scheduled for 104 weeks treatment and groups of 5 per sex were sacrificed after 26, 52 and 78 weeks. Necropsies, brain acetylcholinesterase activity assays and limited clinical chemistry (albumin/globulin ratios) were performed on interim kills. Periodic blood samples, weekly for the first month and then at 3 month intervals, were taken from 10 rats per group, chosen at random. A wide range of haematological parameters, including plasma and erythrocyte cholinesterase activity, were determined. Urinalysis, limited clinical chemistry (protein, albumin/ globulin ratio), gross pathology and histopathology were performed at termination. There were no significant differences in mortality between treated and control rats. No clinical signs of toxicity, or any significant differences in body weight, body weight gain, food consumption, absolute or relative organ weights, were observed during the study. No significant changes in urinalyses were reported, although numerical results were not presented. Toxicologically significant reductions in erythrocyte acetylcholinesterase activity were reported throughout the study at 47 ppm and above. Evidence of a reduction in brain acetylcholinesterase activity (to between approximately 50 - 75 % of controls) was seen in both sexes at 234 ppm up to week 78 of the study. At the end of the study, brain

68

cholinesterase activity was reduced in both the control and 234 ppm groups, relative to other treatment groups. Histopathological results were not presented in full but were reported to show diffuse fine vacuoles in hepatic cells in all animals fed 234 ppm for 78 weeks or longer. Similar changes were seen in 80 % of females and 62 % of males receiving 47 ppm. No changes were seen at 4.7 ppm. Verification of the significance of the hepatic cytoplasmic vacuolation was reportedly obtained from a second assessment of the control and 234 ppm sections by another pathologist. The changes seen were considered to be caused by 'natural disease processes'. The significance of the histopathological changes in the liver is uncertain, particularly in the absence of adequate clinical chemistry data, but the effects are considered to be of functional rather than toxicological concern. The NOAEL is 4.7 ppm (equivalent to 0.6 mg kg-1 d-1) based on erythrocyte acetylcholinesterase inhibition at 47 ppm. In another study, groups of 50 per sex per dose Osborne-Mendel rats were administered 150 or 300 ppm dichlorvos (94 % pure) in the diet for 80 weeks and retained untreated for a further 30 weeks (NCI, 1977). The 300 ppm dose was originally set at 1000 ppm, but this was reduced after 3 weeks because of serious signs of toxicity. A matched control group (5 per sex per dose) were housed with the treatment group. A pooled control, drawn from other concurrent chronic studies, was also available (60 per sex). Histopathology, but not haematology or clinical chemistry, was performed on all animals. There were no statistical differences in survival between treated and control animals. Body weight gain was reduced throughout the study in males administered 300 ppm (approximately 10 %). Adverse clinical signs (rough hair, epitaxis, haematuria, alopecia, dark urine, palpable masses and bloating or abdominal distension) were seen in both treated and control animals; although increased in frequency in the former. On histological examination a number of lesions were noted to occur more frequently in treated animals. These consisted of; aggregates of alveolar macrophages (in males 10 % in matched controls, 30 % at 150 ppm and 14 % at 300 ppm and in females 0 % in control, 45 % at 150 ppm and 42 % at 300 ppm), interstitial fibrosis of the myocardium (incidence in both sexes was 10 % in matched controls, 34 - 40 % at 150 ppm and 24 - 30 % at 300 ppm) and focal follicular-cell hyperplasia of the thyroid in males (0 % in both pooled and matched controls, 7 % at 150 ppm and 10 % at 300 ppm). No histopathological findings were observed in the liver. A NOAEL could not be set because of the histopathology findings at 150 ppm (estimated to be 7.5 mg kg-1 d-1). Dog A 2-year feeding study in dogs is available (Unpublished, 1967c). Dichlorvos (93 % pure) was administered to Beagles (3 per sex per dose) at nominal concentrations of 0, 0.1, 1, 10, 100 and 500 ppm of diet. The actual average concentrations were 0.0, 0.01, 0.3, 3.2, 32 and 256 ppm. Plasma cholinesterase and erythrocyte acetylcholinesterase activities were monitored throughout the study. Brain acetylcholinesterase activity was determined at the end of the study. A limited range of tissues was examined histologically. Dichloroacetaldehyde, a breakdown product, was found in some of the diets. One male receiving 256 ppm was sacrificed at 68 weeks in a moribund condition. Necropsy revealed significant pneumonia, which was not considered to be treatment related. No

69

clinical signs of toxicity were observed during the study. Reductions in body weight gain were seen in one control male and one female at 256 ppm. No other changes in body weight were observed. A slight increase (reported as marginally significant, statistics not presented) in relative liver weight was seen in both sexes at 256 ppm (24 - 32 % compared to control). No changes in haematology, urinalysis or clinical chemistry (AST, ALT and albumin/globulin ratio) were reported. A dose-related reduction in plasma cholinesterase activity was observed in all dogs at 3.2, 32 and 256 ppm (to approximately 80 % of control values). A dose-related toxicologically significant reduction in erythrocyte cholinesterase activities was seen in dogs receiving 3.2 (minimum 49 % of controls after 1 month), 32 (minimum 25 % of controls after 1 month) and 256 ppm (minimum 3 % of controls after 3 months). Brain acetylcholinesterase activity was not reduced. Dose related histopathological changes in the liver of treated dogs were seen, consisting of rarefaction of cytoplasmic substance, enlargement of cells and prominence of the cell membranes. These changes were seen to a mild degree in 1/3 females and 0/3 males at 3.2 ppm, increasing in frequency at 32 ppm to 3/3 female and 1/3 males. In the 256 ppm group moderately severe changes were present in 3/3 females and 3/3 males. The report stated that verification of the significance of the hepatic cytoplasmic vacuolation was obtained by consultation with two pathologists who re-examined sections. They reported no significant changes between the control and treated groups. No changes in liver enzyme activities were reported and it is therefore considered that the histopathological findings do not constitute an adverse effect. The NOAEL is 0.3 ppm (equivalent to 0.008 mg kg-1 d-1) based on toxicologically significant reductions in erythrocyte cholinesterase activity at higher dose levels.

3.2.4.3.2 Inhalation

Rat Groups of CFE rats, 50 per sex per dose, were continuously exposed (whole body) to nominal concentrations of 0, 0.05, 0.5 and 5.0 mg m-3 of dichlorvos vapour (greater than 97% pure) for 23 h d-1 for two years (Unpublished, 1976). The average concentration of dichlorvos was 0, 0.05, 0.48 and 4.70 mg m-3. Dichloroacetaldehyde and trimethyl phosphate were detected in the atmosphere during the first year (0.028 and 0.04 mg m-3 respectively at 5.0 mg m-3 dichlorvos). Rats were housed in individual cages within a 10 m3 inhalation chamber. Necropsy and histopathological examination were carried out on all animals. Haematology and clinical chemistry parameters were measured in all animals sacrificed at the end of the study. No dose related change in mortality was observed. Due to high mortality among control males, all male rats were withdrawn from treatment at 100 weeks. No clinical signs of toxicity were observed. Body weight gain was reduced throughout the study for males and females at 4.7 mg m-3 (approximately 10 %) and up to week 76 in males at 0.48 mg m-3 (approximately 6 %). No consistent differences in food consumption among the groups were found. A significant reduction in mean heart, spleen and kidney weight was seen in males at 4.8 mg m-3 but only the spleen showed a reduced relative weight (approximately 15 %). No dose related changes were seen in females. Erythrocyte and plasma cholinesterase showed a dose-related decrease in activity at 0.48 mg m-3 and above; erythrocyte acetylcholinesterase

70

activity falling to approximately 70 % and 5 % of control values at 0.48 and 4.7 mg m-3, respectively. In both sexes brain acetylcholinesterase activity was reduced at 0.48 and 4.7 mg m-3 (to 90 and 20 % of control respectively). In the small number of brains examined (3 female rats per dose) there were no compound related changes in acetylcholine or choline concentrations. Microscopic examination of lungs of rats revealed minor changes in all groups that were not correlated with dichlorvos treatment. Following a short-term investigation comparing head-only and whole body exposure the authors concluded that only 50 % of the actual dose would be received via the lungs. The daily intake following repeated whole body exposure to 4.7 mg m-3 was considered to be approximately 20 times greater than from inhalation exposure alone (accounted for mostly by contamination of food and grooming). Figures provided in the study report were; 6.2 mg from ingestion of food, 3.6 mg from grooming, < 0.1 mg from drinking water and 0.5 mg from inhalation. The NOAEC was 0.05 mg m-3 based on erythrocyte acetylcholinesterase inhibition.

3.2.4.4 SUMMARY

Reliable human volunteer studies are available following oral exposure while less robust data is available following inhalation exposure and no data is available following dermal exposure. Well-conducted, modern ethical human volunteer studies revealed toxicologically significant reductions in erythrocyte acetylcholinesterase activity in 2 of 6 volunteers following oral exposure to 0.1 mg kg-1 d-1 dichlorvos for up to 21 d. An older study involving volunteer male prisoners indicated a NOAEL of 0.025 mg kg-1 d-1 following 28-day exposure, while no effect on erythrocyte acetylcholinesterase activity was observed in male prisoners following administration of 0.025 mg kg-1 d-1 for 60 d. An older inhalation study reports no effect on erythrocyte acetylcholinesterase activity following a 4 d exposure to between 0.9 - 1.22 mg m-3 for 5 - 7.5 h d-1 (equivalent to 0.09 - 0.19 mg kg-1 d-1). The toxicity of dichlorvos following repeated oral exposure has been well studied, with two studies available in rats and dogs consistent with current EC guidelines. More limited information is available following repeated dermal and inhalation exposures. Two well conducted 13-week gavage study in rats and a 52-week oral study in dogs report NOAELs of 0.1 mg kg-1 d-1 and 0.05 mg kg-1 d-1, respectively, based on significant (> 20 %) inhibition of erythrocyte acetylcholinesterase activity at higher dose levels. In an older 2-year feeding study in rats the NOAEL was considered to be 4.7 ppm (equivalent to 0.6 mg kg-1 d-1), based on a reduction in erythrocyte cholinesterase activity at higher doses. An older 2-year dietary study in dogs a NOAEL of 0.008 mg kg-1 d-1 was derived based on decreased erythrocyte cholinesterase activity at higher doses. No adequate studies using dermal application were available. However, in a single non-standard study, a reduction in brain acetylcholinesterase activity was seen in the rat at 2.9 mg kg-1 d-1 (40 applications over 117 d), however, it is unclear whether this was > 20 % of control values. In a rat 2-year inhalation, virtually continuous (23 h d-1), exposure study the NOAEC was

71

0.05 mg m-3 (equivalent to 0.016 mg kg-1 d-1) based on erythrocyte acetylcholinesterase inhibition, with no other significant findings. A NOAEC of 0.11 mg m-3 (equivalent to 0.006 mg kg-1 d-1) was observed in rats following exposure for 4 h d-1 for 120 d, based on decreased brain cholinesterase activities at higher dose levels. A NOAEC of 0.14 mg m-3 (equivalent to 0.21 mg kg-1 d-1) was observed in mice following exposure for 23 h d-1 for 5 d, based on decreased brain and erythrocyte cholinesterase activities at higher dose levels. The potential of dichlorvos to cause delayed neuropathy has been investigated in the adult hen. Repeated gavage administration for 28 d at a dose level causing death in some birds failed to demonstrate delayed neuropathy.

3.2.5 GENOTOXICITY

3.2.5.1 IN VITRO STUDIES

3.2.5.1.1 Gene Mutation

Bacteria Dichlorvos has been tested in a number of well-validated Salmonella typhimurium strains (TA 100, TA 98, TA 1535, TA 1537 and TA 1538) and in other strains (TA 1531, TA 1532, TA 1534 and TA 1536). Tests were conducted using spot and plate-incorporation methods both with and without metabolic activation (provided by liver S9 fractions from Aroclor 1254 treated rats, mice and hamsters and Na-phenobarbital treated rats). Dichlorvos gave positive results in strains TA 100 and TA 1535 when tested with or without activation (Moriya et al., 1983; Zeiger et al., 1988, Carere et al., 1978a; 1978b; Hanna & Dyer, 1975; Shirasu et al., 1976, Breau et al., 1985; Kawachi et al., 1980; Braun et al., 1982; Löfroth, 1978). Some studies were well conducted, although they did not always meet current regulatory requirements due to limited strains or lack of duplicate testing (Moriya et al., 1983; Zeiger et al., 1988). Equivocal results in strain TA 1535 have been recorded in poorly reported studies, where data were either only in summary form or not presented (Torracca et al., 1976; Moriya et al., 1983). Dichlorvos gave positive results in TA 1530 and C117 without metabolic activation (Hanna & Dyer, 1975; Dyer & Hanna, 1973). Negative results were reported when dichlorvos was tested to toxic levels in well-validated strains (TA 98, TA 1536, TA 1537 and TA 1538) in adequately conducted studies with and without S9 (Moriya et al., 1983; Zeiger et al., 1988 (TA 98 only)). A number of more limited studies have also reported negative results in these strains (Carere et al., 1978a; 1978b; Shirasu et al., 1976). Negative results have also been claimed in strains TA 98, C117, G46, TA 1531, TA 1532, TA 1534, TA 1535, TA 1536, TA 1537 and TA 1538 but no data were presented in these studies (Torracca et al., 1976; Hanna & Dyer, 1975; Breau et al., 1985; Carere et al., 1976; Kawachi et al., 1980; Braun et al., 1982).

72

In reverse mutation assays conducted using E. coli in spot, fluctuation and pre- and plate-incorporation tests, dichlorvos gave positive results in strains WP2, WP2hcr, WP2uvrA and WP67 in the absence of metabolic activation (Bridges, 1978; Shirasu et al., 1976; Bridges et al., 1973; Green et al., 1976; Moriya et al., 1978; Moriya et al., 1983; Hanna & Dyer, 1975). A positive result was also seen with WP2 and WP2hcr in one plate incorporation assay conducted using metabolic activation (Moriya et al. 1978). Negative results were reported with strain WP2 in a spot test, which did not give details of toxicity (Dean, 1972). Some unusual isogenic strains of WP2; CM561, CM571, CM611 and WP12 also gave negative results (Hanna & Dyer, 1975; Bridges et al., 1973). Dichlorvos has been shown to induce resistance to a number of compounds (streptomycin, ß-2-thioryl- DL-alanine and 5-methyltryptophan) in E. coli and Streptomyces coelicolor in spot and plate incorporation tests in the absence of metabolic activation (Wild, 1973, Löfroth et al., 1969, Alldrick & Rowland, 1985, Mohn, 1973, Carere et al., 1976; Carere et al., 1978a; 1978b; Torracca et al., 1976). A negative result was claimed in a poorly conducted spot test, which did not induce cytotoxicity, in strain SD-4 (Unpublished, 1971b). In a spot test (without metabolic activation) using two auxotrophic strains of Serratia marcescens dichlorvos was positive (Dean, 1972). A fluctuation test for streptomycin resistance in the following organisms is available; Klebsiella pneumoniae, E. coli K12, Citrobacter freundii 425, Enterobacter aerogenes and S. typhimurium (Voogd et al., 1972). Dichlorvos was positive in all strains except C. freundii, when the result was equivocal. Two assays, conducted on formulations of dichlorvos, were positive (Ashwood-Smith et al., 1972, Nagy et al., 1975). In a modified spot test, dichlorvos impregnated strips (approximately 20 mm2) were placed on lawns of E. coli WP2. A 50 % commercial preparation was used in a spot test with two WP2 trp E. coli mutants, one of which was repair deficient. Metabolites of dichlorvos have also been tested (Löfroth, 1978). Dichloroacetaldehyde (of unknown purity) was positive and dichloroethanol was negative in strain TA 100. Fungi A forward mutation assay in an isogenic mutant at the ade6 locus of Schizosaccharomyces pombe has been reported (Gilot-Delhalle et al., 1983). Analytical dichlorvos was positive without S9 and equivocal in its presence (increase not twice control). In a poorly reported reverse mutation assay, in Saccharomcyes cerevisiae using a spot test method, dichlorvos (3 unspecified doses) was negative (Guerzoni et al., 1976). Two poorly reported spot tests for 8-azaguanine resistance in Aspergillus nidulans are available in which dichlorvos was positive (Bignami et al., 1977; Morpurgo et al., 1977). An old, poorly reported study gave details of an assay using Neurospora crassa (Michalek & Brockman, 1969). Colonies were grown in the presence or absence of an unknown amount of dichlorvos released from a resin strip. After 6 d, inhibition of the rate of vegetative growth and conidiation but no mutation was seen. A cytogenetic study by Xiao-ou et al., (1997) reports a positive finding in Saccharomyces cerevisiae D61.M (a strain capable of detecting aneuploidy) following dichlorvos exposure.

73

However, little or no experimental detail is provided, so the robustness of the result is questionable. Mammalian Cells Two assays at the TK locus were conducted in mouse lymphoma L5178Y cells (NTP, 1989; Myhr et al., 1990). The studies were only conducted without exogenous metabolic activation. Dichlorvos was positive when tested to cytotoxic levels (200 nl ml-1) and in a repeat experiment using lower doses (NTP, 1989). Dichlorvos was positive, when the highest dose tested was 100 nl/ml (Myhr et al., 1990). The frequency of large and small mutant colonies was determined at 0, 25 or 100 nl ml-1 dichlorvos. At both doses the frequency of small colonies was significantly increased (up to 15 fold). The increase in large mutant colonies was equivocal. In a study performed to GLP, mouse lymphoma L5178Y cells were exposed to 0.004 - 0.5 µl ml-1 in the presence and absence of S9. Mutation frequency (TK locus) was increased in a dose-related manner (up to 13-fold, compared with controls) in the absence of S9. Although increases in mutation frequency were observed in the presence of S9, they did not occur in a dose-related manner. Positive controls gave appropriate responses. These findings were reproduced in a subsequent experiment (Unpublished, 1986i). Overall, this study shows that dichlorvos has the potential to induce gene mutation in mammalian cells in the absence of metabolic activation. In a briefly reported study technical dichlorvos was tested in CHO cells (at the HPRT locus) (Oshiro et al., 1991). Dichlorvos gave a positive result without S9. In the presence of S9 significant increases were observed, although not at consecutive doses. In a further study CHO V79 lung cells (HPRT locus) were exposed to dichlorvos for 1 h, then allowed to grow for 48 h before selection for ouabain resistance (Aquilina et al., 1984). There was no significant increase in mutant colonies, or mutation frequency at the highest dose at which survival was 50 %.

3.2.5.1.2 Chromosome Aberration

Mammalian Cells Positive results have been obtained in a number of cytogenetic studies both with and without S9. A well-conducted study investigating chromosome aberrations in CHO cells is available (Anderson et al., 1990). Cells were exposed for 10 - 14 h to one of 10 concentrations of dichlorvos with and without S9. Results from the three highest doses that generated sufficient metaphases (100) were analysed. Dichlorvos significantly increased the frequency of CA, both with and without S9. In the absence of S9, the 'lowest effective concentration' was quoted as 160 µg ml-1, in the presence of S9 it was 500 µg ml-1.

74

A further well-conducted study using dichlorvos in V79 lung cells without metabolic activation is available (Tezuka et al.,1980). A significant increase in the frequency of cells containing aberrations (including gaps) was seen after exposure to 0.1 - 0.5 mM for 26.5 h. There was no increase in the frequency of cells with aberrations when gaps were excluded. The highest dose tested, 1.0 mM, was toxic. Dichlorvos also caused a significant dose related increase in the frequency of polyploid cells in this test. A study in CHO cells, conducted to an adequate protocol, is available (NTP, 1989). Cells were exposed to up to 1000 µg ml-1 dichlorvos for 2 h in the presence or 160 µg ml-1 for 8 - 10 h in the absence of S9 fraction. The harvest time was 12 - 12.5 h. The types of aberrations were not identified. Dichlorvos was positive, increasing the mean number of aberrations/cell and the number of cells containing aberrations with and without S9. Dichlorvos was positive in an assay in Chinese Hamster lung fibroblast cells with and without S9 (Ishidate et al., 1981). Cells were exposed for 24 h (with and without S9) or 48 h (without S9 only) before harvest. The authors considered a result positive if the frequency of cells with aberrations exceeded 10 %. Dichlorvos was positive in a CA assay, in hamster lung cells, without metabolic activation; no further information was available (Kawachi et al., 1980). A positive micronucleus assay using CHO cells is available (Oshiro et al., 1991). Cells were exposed to technical dichlorvos (up to 800 µg ml-1 with and 150 µg ml-1 without S9) for 20 - 24 h and then screened for micronuclei. Dichlorvos caused an increase in micronucleus frequency, with and without S9 (to 42/1000 at 150 µg ml-1 and 70/1000 at 800 µg ml-1; control frequency was 12/1000) No conclusions can be drawn from three cytogenetic studies claiming negative results because of deficiencies in the methodology. In one study, in Chinese Hamster Don-6 without exogenous activation, results were only available from one concentration and no historical control data were presented (Sasaki et al., 1980). Two studies, using freshly isolated human lymphocytes, only examined metaphases after 50-70 h (Unpublished, 1971a; Dean, 1972). Lower Eukaryotes A study reported that dichlorvos induced mitotic non-disjunction in Aspergillus nidulans when a stable diploid strain was grown on plates containing dichlorvos (Morpugo et al., 1979). The maximum dose used (0.8 mg ml-1) did not produce lethality but inhibited growth. Dichlorvos was also reported as positive in a spot test for mitotic non-disjunction in A. nidulans (Bignami et al., 1976).

3.2.5.1.3 Micronucleus

75

A positive result in an in vitro micronucleus test is reported (Zuyao et al., 1993). The test was carried out in Chinese hamster lung fibroblasts (4000 intact cells/group); the solvent control was DMSO and the positive control mitomycin C (0.1 µg ml-1). No details of the concentrations of dichlorvos used were given, only that the maximum resulted in 50 % of the cells dying. An abstract reports that dichlorvos can induce micronuclei in the human lymphoblastoid cell lines AHH-1 and MCL-5. No further details provided (Doherty et al., 1996).

3.2.5.1.4 Sister Chromatid Exchange

Six recent studies in a range of cell lines (CHO, V79 and rat tracheal epithelial cells) are available (NTP, 1989; Tezuka et al., 1980; Lin et al., 1988; Nishio & Uyeki, 1981; Wang et al., 1988; Anderson et al., 1990). In one of these, conducted to a satisfactory protocol, CHO cells were exposed for 2 h to 500 µg ml-1 dichlorvos (with S9) and 2 h at 50 µg ml-1 (without S9) (NTP, 1989). Dichlorvos gave positive results in the presence and absence of S9. In a second, also to a satisfactory protocol, CHO cells were exposed to 10 concentrations (at half log intervals) of dichlorvos, with and without S9, and harvested 25 - 29 h after treatment (Anderson et al., 1990). No delay in cell cycle was observed. Results from the three highest concentrations with sufficient second-division metaphase cells (50) were analysed (concentrations not given). Dichlorvos gave a positive result with and without S9. The lowest effective doses were 10 µg ml-1 (with S9) and 160 µg ml-1 (without S9). The remaining studies all had some limitation in experimental design, such as exclusive use of toxic concentrations, single dose levels, short exposure times or lack of metabolic activation. However, dichlorvos increased the incidence of SCE in all of these studies. A study using freshly isolated phytohaemaglutinin stimulated human lymphocytes, from at least 3 separate individuals, is available (Nicholas et al., 1978). Technical dichlorvos (2.5 - 10 µg ml-1, set following a sighting study) repeatedly failed to induce an increase in the SCE frequency. No significant increase was observed using the same top dose with human fetal lung fibroblasts. A negative result was also noted in an abstract where lymphocytes were treated with dichlorvos (1 or 10 µg ml-1) for up to 72 h (Dzwonkowska et al., 1989). No conclusions can be drawn because of the limited details available.

3.2.5.1.5 Unscheduled DNA Synthesis (UDS)

Three in vitro UDS assays are available, all using human cell lines in the absence of exogenous metabolic activation. In the first a significant increase in net nuclear grains (assessed by autoradiography) was obtained in 'human epithelial like cells' after exposure to 0 - 144 mg ml-1 dichlorvos for 1 h but no further details on survival were presented (Aquilina et al., 1984). In the second study no increase in net nuclear grains was seen following

76

treatment of a heteroploid cell line derived from human kidney with up to 0.02 mg ml-1 for 1 or 48 h (Unpublished, 1971b). In the third study, human lymphocytes were exposed to up to 500 µg/ml dichlorvos for 4 h in the presence of 3H-thymidine (Perocco & Fini, 1980). A significant dose related increase in 3H-thymidine uptake was observed in this non-standard study using scintillation counting as the detection method.

3.2.5.1.6 DNA Strand Breaks

A study by Yamano (1996) reports that dichlorvos (purity 99 %) induced DNA single strand breaks in rat hepatocytes at a concentration of 500 µM. It is reported that this occurs independently of dichlorvos-induced oxidative stress.

3.2.5.1.7 Other Studies

Dichlorvos is known to be capable of alkylating nucleophiles in vitro (Bedford & Robinson, 1972; Löfroth et al., 1969; Jentzsch and Fischer, 1974). Detailed investigations of DNA alkylation, in isolation, in E. coli and in HeLa cells, are available (Löfroth, 1970; Lawley et al., 1974; Wennerberg & Löfroth, 1974). These studies indicate that dichlorvos can methylate DNA, RNA and protein. For DNA, the alkylation products identified were N7-methylguanine (the major adduct), N3-methylguanine, N1-methyladenine, N3-methyladenine and O6-methylguanine. Two studies claimed significant labelling of proteins, 20 - 30 times greater than DNA. Several DNA repair tests in bacteria are available (Rosenkranz, 1973; Kawachi et al., 1980; Adler et al., 1976; Braun et al., 1982). The assays used technical dichlorvos without metabolic activation and were based on spot test or plate incorporation methods. Dichlorvos was positive in E. coli, Bacillus subtilis and Proteus mirabilis. Liquid pre-incubation tests in pol, uvr and exr mutants of E. coli are available (Bridges et al., 1973). Dichlorvos was positive in pol and exr, but not uvr strains. A number of studies have shown that dichlorvos can alter the physical properties of DNA (Rosenkranz & Rosenkranz, 1972; Griffen & Hill, 1978; Olinska et al., 1980; Bridges et al., 1973; Green et al., 1974a; 1974b; Unpublished, 1971b; Shooter, 1975; Nishio & Uyeki, 1982). Dichlorvos reduced both the specific activity and the thermal denaturation point of isolated DNA. It also induced strand breaks in isolated DNA and bacteriophage. Reductions in molecular weight in the presence of dichlorvos have been demonstrated using isolated DNA, in E. coli, CHO cells and heteroploid cell line from human kidney.

3.2.5.2 SUMMARY OF IN VITRO STUDIES

These data were reviewed by the Department of Health Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM). The following assessment of the mutagenic potential of dichlorvos in vitro was agreed:

77

‘Members agreed that dichlorvos is a weak methylating agent (compared to methyl methanesulphonate; MMS). The COM concurred with the following assessment of the in-vitro mutagenicity studies. i) Dichlorvos is mutagenic, both in the presence and absence of exogenous metabolism, to bacteria, yeast cells and in mammalian cell gene mutation assays, chromosome aberrations assay, the in-vitro micronucleus test and sister chromatid exchange assays. ii) Positive results have been reported in in-vitro UDS assays using human lymphocytes and human epithelial-like cells. iii). Dichlorvos has been shown to methylate nucleophiles and to induce strand breaks in isolated DNA. Members agreed that DNA methylation induced by dichlorvos contributed towards the mutagenicity reported in in-vitro test systems but noted that other mechanisms might also be involved. Members considered that the positive results obtained in in-vitro mutagenicity tests with dichlorvos in the presence of an exogenous metabolising fraction and in the assay for single strand breakage of DNA also suggested that dichlorvos and/or its metabolites were genotoxic. This might include dichloroacetaldehyde although the available evidence was insufficient to identify all potential mutagenic metabolites of dichlorvos. The COM concluded that dichlorvos is an in vitro mutagen.’

3.2.5.3 IN VIVO STUDIES (SOMATIC CELLS)

3.2.5.3.1 Micronucleus

A study performed to GLP investigated the potential of dichlorvos to induce micronuclei in the bone marrow cells of mice following ip injection. Mice (CD-1; 5 per sex per group) were administered 0, 4, 13 or 40 mg kg-1 dichlorvos (purity 98.4 %) on 2 consecutive days, based on the results of a sighting study. Positive control animals received 0.15 mg kg-1 triethylenemelamine. Groups of mice were sacrificed at intervals of 30, 48 and 72 h following the second injection and bone marrow cells examined. In the 40 mg kg-1 group 5 animals (2m, 3f) died and signs of toxicity including tremors and lethargy were observed. No signs of toxicity were observed in the other dose groups. No treatment-related increases in the incidence of micronucleated PCE’s were observed in any dose group. There was no difference between treated and control animals in the ratio of PCE/NCE (including positive controls). Positive and negative control group animals showed appropriate responses (Unpublished, 1985a). Overall, this study indicates that dichlorvos does not induce micronuclei in mouse bone marrow cells. In a micronucleus study, mice (strain 615; 5 per sex per group) were administered 4 daily ip doses of dichlorvos at concentrations of 10, 20, 40 and 80 % of the LD50 for 4 ip injections of

78

dichlorvos (no further details given) (Zuyao et al., 1993). The solvent control group was administered maize oil and the positive control group cyclophoshamide (100 mg kg-1). The animals were sacrificed 24 h after the final injection and 1000 PCE’s examined per mouse for micronucleus induction. A negative result was reported but no further details were given. A study, only available in abstract form, and also reporting positive results in in vivo CA and teratology studies, gave limited details of a micronucleus test in mice (Majeeth et al., 1989). Dichlorvos (purity unspecified) was administered by ip injection to Swiss albino mice (0, 2.5, 5 and 10 mg kg-1). Bone marrow preparations were made 24, 48, 72 and 168 h after dosing. In treated mice the frequency of micronuclei was 3 - 3.5 % and 1.15 % in controls. This increase was reported as significant. It is unclear if this study used technical dichlorvos or a commercially available preparation, as no further details were available. The COM considered that this study could not be interpreted as insufficient information on the methods and results were available. Swiss albino mice (3 per group) received ip doses of 0 or 0.015 mg kg-1 technical dichlorvos (reported as equivalent to 50 % of the lethal dose) as a single dose or 2 - 4 daily doses (Paik & Lee, 1977). Bone marrow smears were prepared 6 h after the last dose. Dichlorvos did not increase the frequency of micronucleated polychromatic erythrocytes after either single or repeated dosing. The value of this study is uncertain because of the apparently low lethal dose. In a micronucleus test, modified to examine topical effects, inbred male HRA/Skh hairless mice were given topical doses of technical dichlorvos on the dorsal skin (Tungal et al., 1991). Mice (2 per dose) received dermal applications of 0, 51, 258, 516 and 1033 nmol dichlorvos (estimated to be 0.6, 2.9, 5.7 or 11.4 mg kg-1) or an equal volume of acetone (negative control) or urethane (positive control). Animals were sacrificed at 48 h and skin samples prepared. Keratinocytes were cultured for up to 72 h before micronuclei were assessed in at least 2000 cells. Dichlorvos significantly increased the numbers of micronuclei in all three replicates, as did urethane. The doses and harvest time were chosen after sighting studies. The COM agreed that the approach used in this study had not been fully validated but agreed the authors had used an appropriate positive control chemical and that the results with dichlorvos were indicative of an in vivo site-of-contact mutagenic effect. Members also noted a positive response in a nuclear anomaly assay in hair follicles of mice following topical application (see Schop et al., 1990, below). Although the latter is not considered to be a definitive genotoxicity assay, the results might be indicative of a biological effect in the skin. A combined bone marrow micronucleus and hair follicle nuclear aberration assay is available (Schop et al., 1990). CD1 mice (12 per group) received dermal applications of dichlorvos at 0, 3.1, 6.3 or 12.5 % of the published dermal LD50 (US NIOSH, 1977). Doses were applied 8 d after all hairs had been removed from 10 x 10 mm2 area by plucking. Mice were killed 24 h after treatment and bone marrow smears and serial skin section preparations were made. The incidence of micronuclei, in 1000 polychromatic erythrocytes, and nuclear aberrations, in a minimum of 1500 hair matrix cells, from each mouse was determined. Nuclear

79

aberrations consisted of micronuclei, pyknotic nuclei, karyorrhectic nuclei, condensed chromatin in apoptotic bodies or vacuolated nuclear fragments. The positive control, cyclophosphamide, significantly increased the frequency of both micronuclei and nuclear aberrations while MNNG only increased the frequency of nuclear aberrations. Dichlorvos did not significantly increase the incidence of micronuclei. However the mean ratio of red blood cells to polychromatic erythrocytes was also unaltered. At 12.5 % LD50 dichlorvos significantly increased the percentage of nuclear aberrations in hair follicle cells. The mitotic index was unaltered. Results were only presented graphically, but the incidence at the top dose appears to be approximately 1 % compared to 0.3 % in vehicle control.

3.2.5.3.2 Chromosomal Aberrations

Rat A poorly reported study by Nehez et al., (1994) investigated the potential for dichlorvos to induce chromosome aberrations in the bone marrow of rats following repeated oral dosing. Male rats (SPF outbred; 10 per group) were administered 0, 0.97, 1.29 and 1.94 mg kg-1 d-1 dichlorvos (purity 98 %) by gavage 5 d wk-1, for 6 weeks. Bone marrow cells were extracted 24 h after the final dose and 20 metaphase cells per animal were examined for structural and 'numerical' chromosome aberrations. No information was given about how the metaphases were prepared or how 'numerical' aberrations were scored. It is unclear whether the term 'numerical' relates to aneuploidy or polyploidy, or both of these endpoints. The bodyweights of the animals was measured weekly. At the end of the study the weights of the brain, liver, heart, kidneys, adrenals and spleen were measured. There was no positive control data in this study, which also included dimethoate and parathion-methyl as test substances. The report did not indicate any signs of toxicity in the animals. No difference in the bodyweights or organ weights of treated and control animals were observed. A statistically significant increase in the number of cells with 'numerical' chromosome aberrations was claimed in treated animals compared with controls (5, 16, 15 & 18 % of cells examined at 0, 0.97, 1.29 and 1.94 mg kg-1, respectively). No statistically significant increase in structural aberrations was observed in treated animals. The COM considered that the methods used were satisfactory and noted that, although the adequacy of reporting was limited, the results indicated a positive effect for the induction of numerical chromosome aberrations. It was noted that a clear dose-response would not be expected in this study, as the dose range selected was relatively narrow. Mice A study performed to GLP investigated the potential of dichlorvos to induce chromosomal aberrations in the bone marrow cells of mice following oral exposure. Male mice (ICR; 10 per group) were administered gavage doses of 0, 12.5, 25 or 50 mg kg-1 d-1 dichlorvos (purity 98.1 %) in deionised water for five days (based on the results of a toxicity sighting study). Positive control animals were administered cyclophosphamide 24 h priors to sacrifice. Animals were sacrificed 24 h after the last dose; and colchicine administered 2 h priors to

80

sacrifice. Bone marrow cells were examined for chromosomal aberrations following sacrifice. A minimum of 50 metaphase cells was scored per animal. No animals died during the study. A single animal of the 50 mg kg-1 d-1 group appeared lethargic 15 minutes post-administration, but all other animals appeared normal throughout the study period. No significant increase in chromosomal aberrations was reported following dichlorvos treatment. The mitotic index remained unaffected by treatment. Positive and negative control group animals showed appropriate responses (Unpublished, 1992a). Overall, this study indicates that at the dose levels administered in this dichlorvos does not induce structural chromosomal aberrations in mouse bone marrow cells. A study, only reported in a summary and including positive results from an in vivo micronucleus and teratology studies, is available (Majeeth et al., 1989). Dichlorvos was administered by ip injection to Swiss albino mice (0, 2.5, 5 and 10 mg kg-1). No details of the number of animals used or toxicity were provided. Bone marrow chromosome preparations were made at 24, 48, 72 and 168 h after dosing. The frequency of cells with aberrations in dichlorvos treated mice was 6-21 %, compared to 1.5-2 % in controls. The maximum frequency was seen at 48 h, after a dose of 5 mg kg-1. The type of aberrations was not specified. It is unclear if this study used technical dichlorvos or a commercially available preparation, as no further details are available. The COM considered that this study could not be interpreted as insufficient information on the methods and results were available. In another study, to an acceptable protocol, chromosome aberrations were examined in Carworth Farm strain (CF1) mice (Dean & Thorpe, 1972). Groups of mice (12 per sex) were exposed by inhalation to dichlorvos vapour, either 72 mg m-3 (males only) or 64 mg m-3 (females only) for 16 h or to 5 mg m-3 for 23 h d-1, for 21 d. At the end of the study, animals were sacrificed 2 h after cessation of exposure and 100 chromosome preparations from bone marrow were examined. Acute exposure of mice to dichlorvos produced clinical signs of OP poisoning in both sexes. The only change observed was a non-significant increase in the percentage of cells showing chromatid aberrations (from 0.58 to 0.83 % in males and 0.67 to 1.25 % in females). The positive control gave an increase in aberrations. The mitotic index was not reported. Exposure to 5 mg m-3 of dichlorvos vapour had no effect on the percentage of cells showing aberrations. Dichlorvos did not induce chromosome aberrations after either an acute or repeated exposure in this study. In further study, male Q strain mice (8 per dose) received dichlorvos (99 % pure) at 0 or 2 ppm in the drinking water 5 d wk-1 for 7 weeks (Degraeve et al., 1984a). At the end of the study mice were sacrificed and metaphases from bone marrow were prepared. There was no difference in the percentage of either breaks or gaps between treated or control mice. No exchanges or translocations were observed in any metaphase. Dichlorvos did not induce chromosome aberrations in mice in this study, although toxic doses were not used. In another study, male mice (5 per dose, 10 control) received 0 or 10 mg kg-1 of dichlorvos by ip injection or 100 mg kg-1 by gavage (Kurinnyi, 1975). Animals were sacrificed at 20 h, bone marrow smears were prepared and a total of 1000 metaphases were examined. No

81

positive control was included. There was no significant increase in the incidence of aberrations. Although signs of toxicity were not reported the doses used were close to the respective LD50 values. Dichlorvos was negative in this study. An inadequately reported study gave the results of a cytogenetic assay in mice (Moutschen-Dahmen et al., 1981). Male Q strain mice received 0 or 20 mg kg-1 dichlorvos (99 % pure) by ip injection. Chromosome preparations from bone marrow were made at 24, 48, 72 and 96 h post-treatment. There was no increase in the number of aberrations seen. Mice also received 2 ppm dichlorvos in the drinking water or feed for 5 d wk-1. The results were not clearly reported but they claimed to show no increase in structural aberrations. Hamster A cytogenetic assay in Chinese Hamsters is available (Dean & Thorpe, 1972). Male hamsters (8 per group) were either given an oral dose of 15 mg kg-1, or exposed to atmospheres containing 32 mg m-3 for 16 h or left untreated. Females (4 per group) received oral doses of 0 or 10 mg kg-1 dichlorvos. Animals were killed 2 h after end of inhalation exposure or 24 h (males) or 8 h (females) after oral dosing and bone marrow preparations made. Following oral administration at 15 mg kg-1, 2/8 males died. There was no significant increase in the number of cells containing aberrations after either oral administration or inhalation exposure in either sex. The positive control gave an increase in the number of cells showing aberrations. However this study is limited by a lack of a 48 h sample time. No conclusions can be drawn from two CA assays using commercial preparations of dichlorvos (Degraeve et al., 1984b; Dzwonkoswka and Hübner, 1986).

3.2.5.3.3 Sister Chromatid Exchanges

In a study performed to GLP mice (B6C3F1; 5/sex/group) were administered 0, 3, 10 or 30 mg kg-1 dichlorvos (purity 98.4 %) in corn oil by ip injection. Positive control animals were administered 1 mg kg-1 cyclophosphamide. Animals were sacrificed 24 h post-administration and the bone marrow cells examined. No mortalities were reported, but lethargy was reported in top dose animals. No treatment-related increase in SCEs per cell was observed. Positive and negative control group animals showed appropriate responses (Unpublished, 1985b). Two well conducted studies in male B6C3F1 mice are available (NTP, 1989). Dichlorvos was given by ip injection, at 6.25, 12.5 or 25 mg kg-1 or 10, 20 or 40 mg kg-1 to 4 mice per dose, 1 h after BrdU administration. In both cases the top dose used was approximately equivalent to the LD50. Bone marrow smears were prepared 17 h after dosing. Negative results were obtained in both studies. Positive controls gave a significant increase in SCE. Two more limited studies, that also gave negative results, are available (Kligerman et al., 1985; McFee and Tice, 1986). The first used male B6C3F1 mice (3 per dose), which received ip doses of 0, 5, 15 or 25 mg kg-1. Lymphocytes were isolated 24 h after dosing and cultured for 24 h before BrdU was added. Metaphase cells were examined after a further 60 h incubation. There was no increase in the number of SCEs/cell. The second study was in abstract form only and no further details were presented.

82

3.2.5.3.4 Unscheduled DNA Synthesis

A published paper reported an in vivo-in vitro hepatocyte UDS assay in which male F344 rats (6 per dose) received 2, 10 and 35 mg kg-1 dichlorvos in corn oil by gavage (Mirsalis et al., 1989). At 2 and 24 h, 3 rats per group were sacrificed and suspensions of hepatocytes were prepared and labelled with 3H-thymidine. The test was considered to be positive if the number of net nuclear grains was positive and the percentage of cells undergoing repair was greater than 10 %. Dichlorvos gave a negative result with a net number of grains/nucleus range of -6.6 to -4.9 and the percentage of cells in repair was 1 - 6 %. The positive control gave the appropriate response. A UDS assay in mouse forestomach, conducted to GLP, is available (Unpublished, 1991c). Fasted B6C3F1 mice (20 per sex per group) received oral doses of 0, 10, 20, 40 or 100 mg kg-1 dichlorvos (99.8 % pure). Animals (5 per sex per group) were killed at 2, 4, 12 and 48 h after dosing, the forestomach was removed and sections fixed for histopathology or incubated with 3H-thymidine for 3 h. The positive controls were MNNG and butylated hydroxyanisole (BHA). UDS was assessed by autoradiography in 50 epithelial cells per mouse in animals killed at 2 and 4 h. Replicative DNA synthesis (RDS) was assessed at 12 and 48 h by determining the proportion of S-phase cells in 500 epithelial cells. Mortalities were observed at 100 mg kg-1 dichlorvos (3/20 males), 200 mg kg-1 MNNG (1/20 males) and 300 mg kg-1 BHA (1/20 males). Dichlorvos and BHA did not increase levels of UDS above the background frequency while MNNG produced a statistically significant increase in males but not in females. No increase in RDS was seen with dichlorvos, MNNG or BHA. Hyperplasia was not seen in mice at 2 or 4 h. At 12 h hyperplasia (usually diffuse in nature, but occasionally focal) was seen in all male and female mice receiving dichlorvos. At 48 h the incidence of hyperplasia was slightly lower. Similar changes were seen following BHA treatment. MNNG induced cell damage but only minimal hyperplasia. In a further study it was shown that dichlorvos induced RDS and hyperplasia at 10 h after administration of up to 100 mg kg-1 to B6C3F1 mice, although no dose-response was evident (Unpublished, 1992b). No RDS or hyperplasia was seen at 8 h in dichlorvos treated mice. BHA induced both RDS and hyperplasia at 8 and 10 h, while MNNG induced severe cell damage but no hyperplasia or RDS (also Benford et al., 1994).

3.2.5.3.5 COMET Assay

A study by Sasaki et al., (2000) compared data from comet assays for 208 chemicals with known carcinogenic potential. Both published and unpublished data were summarised as part of the study. The data presented on dichlorvos were unpublished and therefore lacked detail in terms of methodology and reporting of results. Groups of 4 male mice (ICR) were administered a single oral dose of 100 mg kg-1 dichlorvos (purity 98 %) in olive oil. This dose is assumed to be close to the maximum tolerated dose, as the authors report an LD50 value 124 mg kg-1. No details of control groups were provided. Sampling times were 3 and 24 h, and the organs analysed were the stomach, colon, liver, kidney, urinary bladder, lung, brain and bone marrow. Subsequently, isolated nuclei were purified from each tissue and

83

slides for the comet assay prepared; 50 nuclei were examined per organ. Migration (the difference between the length of the whole comet and the diameter of the head) was determined for each comet and the average length of DNA migration for each organ calculated. The authors describe the subsequent statistical analysis of the results as follows: 'the average length of migration of the four animals per treatment group was compared by one-way analysis of variance. When significance was obtained by this analysis, differences between treated and control groups were compared using Dunnett’s test. A p-value less than 0.05 was considered statistically significant.' The comet assay data for all 208 chemicals are reported as an appendix to the study. Numerical data are only reported in terms of net migration when there is a statistically significant difference from solvent controls; where numerical data is not reported this indicates that either the test was not carried out ('not done') or the result did not reach statistical significance (a 'dash') when compared with solvent controls. It is assumed that statistical analysis has been performed on the dichlorvos data. At the 3 h sampling time, positive results were reported in all the tissues examined. However, a wide variation in the degree of migration was apparent between the tissues (net migration values of 51.9, 53.5, 54.4, 15.4, 7.63, 16.8, 2.97 and 14.2 mm reported in the stomach, colon, liver, kidney, urinary bladder, lung, brain and bone marrow, respectively). At the 24 h sampling time, positive results were reported, though of a decreased magnitude in terms of migration, in the stomach, colon, liver and lung (net migration values of 27.3, 23.1, 21.7 and 8.89 mm, respectively); while no results were reported for kidney, urinary bladder and bone marrow. It is unclear why no results were reported for these latter tissues, most likely a negative result was observed given the pattern of results reported in the other tissues. The COM considered that the approach adopted by Sasaki and colleagues (2000) to the mutagenicity testing of several hundreds of chemicals had a number of drawbacks, for example, limited reporting of signs of toxicity seen in animals. Members considered that the appropriateness of the isolated nuclei method used by Sasaki and colleagues had not been established and noted that there was no cellular measure of cytotoxicity or apoptosis in this study. In respect of the study on dichlorvos, Members agreed that the dose level chosen (~ 80 % of the LD50) was too high. Members agreed that in view of these limitations, little weight could be placed on this study. The positive data in all tissues examined was unexpected given all the available mutagenicity data on dichlorvos. Members considered that it was not possible to conclude that dichlorvos had mutagenic effects in a wide range of tissues on the basis of these data. Thus, although the authors suggested that dichlorvos had an in-vivo genotoxic effect, the data were uninterpretable.

3.2.5.3.6 DNA adducts

A study by Pletsa et al., (1999) investigated DNA adduct formation in llacZ transgenic mice. Groups of 2 llacZ transgenic mice (Mutamouse; Hazleton) were given a single ip injection with either 0, 4.4 or 11 mg kg-1 (reported to be half of the LD50) dichlorvos in phosphate-buffered saline. Four hours post-exposure, the animals were sacrificed and DNA extracted from the bone marrow, white blood cells, liver, spleen, lung, brain and sperm cells. The purified DNA was analysed for N7-methylguanine and O6-methylguanine using a competitive repair assay and an immunoslot-blot technique, respectively. No evidence was found for DNA adducts above the detection limits of the assays (8 x 10-8 mol/mol guanine for O6-methylguanine and 5 x 10-6 mol/mol guanine for N7-methylguanine). A model methylating agent, dimethylsulphate, was used as a positive control in this study and gave

84

rise to detectable DNA adducts (N7-methylguanine and O6-methylguanine) following ip administration of a single dose of 30 mg kg-1.

3.2.5.3.7 Transgenic study

Mutation frequency has been investigated by Pletsa et al. (1999) in llacZ transgenic mice following both single and multiple exposures. Groups of 4 llacZ transgenic mice (Mutamouse; Hazleton) were given a single ip injection with either 0, 4.4 or 11 mg kg-1 (reported to be half of the LD50) dichlorvos in phosphate-buffered saline. Fourteen days post-treatment, animals from the solvent control and 11 mg kg-1 groups were sacrificed and mutation frequency was determined in the llacZ transgene in DNA from the liver and bone marrow. This was done by isolating the DNA from the tissues and rescuing the l prophages using a l-phage packaging extract (Giga-Pack II Gold). No statistically significant increase in mutation frequency was observed at 11 mg kg-1 compared with controls (consequently a similar analysis was not carried out on the 4.4 mg kg-1 group of mice). For the multiple dosing study groups of 4 llacZ transgenic mice were first given an ip injection of 0 or 11 mg kg-1 dichlorvos in phosphate-buffered saline. A second dose was given 24 h later, but in the dichlorvos group the animals seemed to tolerate this poorly and the 3 further treatments were given at longer intervals of 7 days. Mutation frequency was determined in the liver and bone marrow 14 days after the last treatment (approximately 35 d after the initial treatment). One of the mice in the dichlorvos group died following the final treatment, and was not included in the results. A threefold increase in mutant frequency was found in the liver of treated mice compared with controls and was statistically significant (mean mutant frequencies 145 and 49.7 per 106 plaque forming units, in treated and control animals, respectively). Although, mutant frequency in the bone marrow was almost double that observed in control animals, this did not reach statistical significance. The COM noted that the dose levels used in this study were high and did induce severe toxicity in the animals. They agreed that although the methods used and standards of reporting used in this study had limitations, the data were indicative of a mutagenic effect of dichlorvos in-vivo at the site-of-contact i.e. the liver. The COM noted that the authors had failed to identify any O6 and N7 methylguanine adducts in tissue DNA from transgenic mice given a single intraperitoneal dose of either 4.4 mg kg-1 or 11 mg kg-1 dichlorvos but agreed that the methods used by the authors were of inadequate sensitivity and it was unlikely that any alkyl adducts could have been detected. In support of this conclusion Members commented that the levels of DNA adducts (O6 and N7 methylguanine) in transgenic mice (Muta™ Mouse) following repeated dosing with dimethyl sulphate (10 x 6 mg kg-1 ip) were only approximately 4-fold higher than the limit of detection. Members considered that evaluation of DNA adducts in dichlorvos treated animals after the repeat dosing regime might have provided valuable information but these analyses had not been undertaken.

3.2.5.3.8 Other Studies

A number of studies investigating DNA binding in vivo are available. In no study was a non-genotoxic methyl donating compound used as a negative control. Dichlorvos is rapidly and extensively metabolised in vivo and the entry of radiolabelled carbon into the general metabolic pool has been shown.

85

Male NMRI mice were exposed to known air concentrations of 14C-[methoxy]-dichlorvos for 2 h (Wennerberg & Löfroth, 1974). Urine was collected for 48 h and analysed for radiolabelled metabolites. Urine was found to contain 14C-N7-methylguanine (0.007 - 0.010 % of total urinary radioactivity). A further study also investigated alkylation of DNA in the rat (Löfroth & Wennerberg, 1974). Two male R strain rats received an ip injection of 900 µCi of 14C[-methoxy]-dichlorvos kg-1. Urine was collected for 4 d. During this period, 59 % of the administered radioactivity was recovered in the urine. Analysis of the urine demonstrated that 1.8 nCi of radioactivity was excreted as 1-methylnicotinamide (0.0008 % of the dose). The adducts N7-methylguanine and N3-methyladenine were also identified but co-eluted with each other in the system used. Over the 4 d the combined total was 14.4 nCi. A poorly reported study claimed no increase in concentration of nucleic acid derivatives (deoxycytidine and thymidine) in the urine of female CB hooded rats receiving 10 mg kg-1 dichlorvos by ip injection (Chu & Lawley, 1975). A study into methylation of DNA in mice is available (Kandeel et al., 1987). Male mice (5 per dose) received a single ip dose of 4 or 8 mg kg-1 [14C-methoxy]-dichlorvos. Low levels of radiolabelled N7-methylguanine were recovered from hepatic nucleic acid. This accounted for 0.25 % (as DNA) and 0.63 % (as RNA) of the applied dose 24 h after dosing. The concentration of radioactivity was 0.53, 0.42 and 0.014 µg 14C mg-1 of RNA, DNA and protein respectively at 4 mg kg-1; at 8 mg kg-1 the values were 0.59, 0.53 and 0.032 µg 14C mg-1 respectively. Urine was collected over 24 h and 5.2 % of the total radioactivity detected was in the purine fraction. After chromatography, 0.83 % of the total was identified as N7-methylguanine. Male NMRI mice received either 3H- or 14C-[methoxy]-dichlorvos by injection (Wennerberg & Löfroth, 1974). Urine was collected for 48 h and analysed for radiolabelled metabolites. The urine was found to contain approximately 0.002 - 0.004 % of the total dose as 14C-N7-methylguanine. A review included limited details of an experiment in which mice were given an ip injection of 420 µg kg-1 14C-[methoxy]-dichlorvos (presented as a personal communication) (Wooder & Wright, 1981). When DNA was isolated from the soft organs, pooled and analysed it was found to contain 8 x 10-13 mol methyl g-1 DNA. In an alkylation assay 12 male CBA mice received an ip injection of 1.9 µmol kg-1 of radiolabelled 14C-[methoxy]-dichlorvos (0.21 mCi) (Segerback, 1981). Mice were killed after 5 h. DNA was extracted from soft tissues and purified. The amount of radioactivity found associated with N-7-methylguanine was 7 dpm per 10 mg DNA or 8 x 10-13 mol methyl g-1 DNA. Following a single ip administration of 10 mg kg-1 dichlorvos in dimethylsulphoxide to Wistar rats, animals were sacrificed at various time points up to 24 h post-exposure. No induction DNA single strand breaks in liver samples was observed (Unpublished, 1979c).

86

3.2.5.4 IN VIVO STUDIES (GERM CELLS)

3.2.5.4.1 Chromosomal Aberrations

A study performed to GLP investigated the potential of dichlorvos to induce chromosomal aberrations in the spermatogonial cells of mice following oral exposure. Male mice (ICR; 10 per group) were administered gavage doses of 0, 12.5, 25 or 50 mg kg-1 d-1 dichlorvos (purity 98.1 %) in deionised water for five days (based on the results of a toxicity sighting study). Animals were sacrificed 24 h after the last dose; and colchicine administered 2 h prior to sacrifice. Positive control animals were administered cyclophosphamide 24 h prior to sacrifice. Spermatogonial cells were extracted following sacrifice. A minimum of 50 metaphase cells was scored. No animals died during the study. A single animal of the 50 mg kg-1 d-1 group appeared lethargic 15 minutes post-administration, but all other animals appeared normal throughout the study period. No significant increase in chromosomal aberrations was reported following dichlorvos treatment. The mitotic index remained unaffected by treatment. Positive control group animals showed appropriate responses. (Unpublished, 1992a) Overall, this study indicates that dichlorvos does not induce chromosomal aberrations in mouse spermatogonial cells. A study reporting negative results in Q strain mice is available (Degraeve et al., 1984c). Mice (20 males per dose) received a single ip injection of 0 or 10 mg kg-1 dichlorvos (99 % pure). Animals were killed (2 mice per d) on days 10-15 post-treatment and metaphases from the testes were made. There was no increase in the number of breaks, exchanges, gaps or the total number of aberrations seen. Mitomycin C gave a clear positive response (day 11). In a poorly reported study, male Q strain mice were given 0 or 20 mg kg-1 of dichlorvos (99 % pure) by ip injection (Moutschen-Dahmen et al., 1981). Chromosome preparations from spermatogonia cells were made at 24, 48, 72 and 96 h post-treatment. Spermatocyte metaphases were examined on days 11 - 16. Positive controls gave a significant increase. There was no increase in the number of aberrations seen with dichlorvos. A study in CF1 mice is available (Dean & Thorpe, 1972). Groups of mice (12 per sex) were exposed to 72 mg m-3 for 16 h or 5 mg m-3 of dichlorvos vapour for 23 h d-1, for 21 d. At the end of the study, animals were sacrificed 2 h after cessation of exposure and chromosome preparations made from the testes. Exposure of mice 72 mg m-3 resulted in clinical signs of OP poisoning. A positive control was included (endoxan, considered to be inappropriate for this test) which failed to give an increase in aberrations. No significant changes were seen after either acute or repeated exposure to dichlorvos. Male Q strain mice (8 per dose) received dichlorvos (99 % pure) at 0 or 2 ppm in the drinking water 5 d wk-1 for 7 weeks (Degraeve et al., 1984a). At the end of the study, mice were sacrificed and metaphases from spermatogonia and primary spermatocytes were prepared. There was no difference in the percentage of either breaks or gaps between treated or control mice. No positive controls were included. Dichlorvos did not induce chromosome aberrations in mice in this study, although toxic doses were not used.

87

An assay in Chinese Hamsters is available (Dean & Thorpe, 1972). Male hamsters (8 per group) were either given an oral dose of 15 mg kg-1, exposed to atmospheres containing 32 mg m-3 for 16 h or left untreated. Animals were killed 2 h after end of inhalation exposure or 24 h after oral dosing and metaphase preparations made from the testes. Following oral dosing at 15 mg kg-1, 2/8 males died. A positive control was included (endoxan, considered to be inappropriate for this test) which failed to give an increase in aberrations. Chromosome preparations showed no significant variation between the treated and control hamsters.

3.2.5.4.2 Dominant Lethal Studies

A dominant lethal study performed to GLP and consistent with current EU guidelines was conducted in mice. Groups of male mice (CD-1; 30/group) were administered 0, 8, 16 or 32 mg kg-1 d-1 dichlorvos (purity 97.5 %) in corn oil by intraperitoneal injection for 5 days. These values were based on the results of a sighting study. Following the last dose each male mouse was mated with 2 virgin female mice for a 7 day period, for each of 8 weeks. The females were sacrificed approximately 2 weeks from the midpoint of the mating period and the uterine contents evaluated for the number of live and dead implants. Positive control animals received a single dose of 0.2 mg kg-1 triethyleneamine on the 5th day of treatment. No signs of toxicity were reported in the treated animals. No statistically significant difference in fertility index was reported between treated and control groups. Total implantations per female were statistically significantly decreased compared with control values in the 32 mg kg-1 group in week 1 but this was within the range of control values reported over the 8 weeks and so is not thought to be biologically significant. Live implants/female were statistically significantly reduced in the 32 mg kg-1 group in weeks 1, 5 and 8, but again these values were within the range of control values reported over the 8 weeks; given this and the sporadic nature of the finding, it is not thought to be of biological significance. A statistically significant increase in dead implants/female was reported after 2weeks in the 8 and 32 mg kg-1 groups, due to the control value being very low; and in the 316 mg kg-1 group during weeks 4 and 5. These findings are sporadic and lack dose-response and therefore are of dubious biological significance. Post-implantation loss (ratio of dead implants/live implants) was similar in control animals (0.02 - 0.06) and in treated groups (0.04 - 0.12, 0.03 - 0.09 and 0.05 - 0.09 at 8, 16 and 32 mg kg-1, respectively) over the 8 week period. Positive control responses were appropriate (Unpublished, 1987a). Overall, dichlorvos did not show the potential to produce dominant lethal mutations under the conditions of this assay. In an earlier dominant lethal study, performed to GLP by the same group as above, groups of male mice (CD-1; 10/group) were administered 0, 1, 3 or 10 mg kg-1 d-1 dichlorvos (purity 98.4 %) in corn oil by intraperitoneal injection for 5 days. These values were based on the results of a sighting study. Positive control animals a single dose of 0.5 mg kg-1 triethyleneamine on the 5th day of the treatment schedule. Following the last dose each male mouse was mated with 2 virgin female mice for a 7 day period, for each of 8 weeks. The females were sacrificed approximately 2 weeks from the midpoint of the mating period and the uterine contents evaluated for the number of live and dead implants.

88

No signs of toxicity were reported in the treated animals. No statistically significant difference in fertility index was reported between treated and control groups. Sporadic statistically significant differences were reported between treated and control groups in numbers of implantation sites (live, dead or total) but these could be attributed to unusually high or low control values, when all control values over the 8 week period were looked at. The changes also lacked consistent dose-response characteristics and therefore, are considered to be of no biological significance. Post-implantation loss was similar in control animals (0.02 - 0.07) and in treated groups (0.03 - 0.12, 0.01 - 0.09 and 0.03 - 0.12 at 1, 3 and 10 mg kg-1, respectively) over the 8 week period. Positive control responses were appropriate (Unpublished, 1985c). Again, dichlorvos did not show the potential to produce dominant lethal mutations under the conditions of this assay. A further dominant lethal study is reported in the literature (Dzwonkowska & Hubner, 1991) in which mice (Balb/c; 20 males) were administered either a single ip dose or 5 consecutive daily ip doses of 6 mg kg-1 dichlorvos. Dichlorvos was administered as a 50 % dichlorvos formulation. Negative and positive control groups contained 25 animals. Each male was then mated with a virgin female every 4 days for 48 days. Females were then killed approximately 24 days after the mating period and examined for live and dead implantations. As in previous studies no treatment-related differences were observed in numbers of implantations (live, dead or total) between control and test animals. This study is consistent with previous findings that indicate does not induce dominant lethal mutations at the dose levels used in this study. In an satisfactory study, male CF1 mice (8 per group, 16 per control) were exposed 0, 30 or 55 mg m-3 of dichlorvos vapour for 16 h or to 0, 2.1 or 5.8 mg m-3 for 23 h d-1 for 28 d (Dean & Thorpe, 1972). Following dosing males were each mated with 3 virgin females per week for 8 weeks. Thirteen days after presumed mating females were killed and the uterus removed for examination. There was no evidence of a dominant lethal effect in either the repeated or the single dose experiments. The positive control gave a satisfactory result. In briefly reported studies, dichlorvos did not increase post-implantation losses in mice at doses of; 2 ppm (drinking water), 10 mg kg-1 (oral) and 10 mg kg-1 (ip) (Degraeve et al., 1984a; 1984c; Moutschen-Dahmen et al., 1981; Buselmaier et al., 1973). However, 10 mg kg-1 increased the number of pre-implantation losses in the first or second week's mating. A negative result was also reported in a study using a commercial preparation of malathion (150 g l-1), dichlorvos (60 g l-1) and carbaryl (75 g l-1) (Degraeve et al., 1984b). A study reported the results of a dominant lethal assay using female mice (Dean & Blair, 1976). Female CF1 mice (48 per dose, 96 per control) received 0, 25 or 50 mg kg-1 dichlorvos by gavage. One quarter of the mice from each group were individually mated with males, of proven fertility, on day 0 following dosing and thirteen days after presumed mating females were killed and examined. This process was repeated with mating

89

commencing on days 5, 10 and 15 following dosing. There was no change in the percentage of pregnancies, number of implants or number of early fetal deaths. In a second experiment, female mice (48 per group) were continuously exposed to atmospheres of 0, 2.3 or 7.7 mg m-3. The same mating pattern was followed. Dichlorvos did not cause any observable effects on reproductive performance.

3.2.5.4.3 Other Studies

Host Mediated Assays Four negative host mediated assays are available, using S. cerevisiae, S. typhimurium, and Serratia marcescens in mice (Buselmaier et al., 1972; Buselmaier et al., 1973; Dean et al., 1972; Voogd et al., 1972). Mice were either given oral doses of dichlorvos, sc injections or exposed to vapour for 5 h. Drosophila Assays for accumulation of recessive and lethal mutants have been conducted. Three studies reported positive results using 1 - 5 ppm (Andreu et al., 1988; Bagriaçik & Ünlü, 1991; Velázquez et al., 1987; Hanna & Dyer, 1975). Five studies were negative, at doses of up to 6.0 x 10-3 mM (Sobels & Todd, 1979; Kramers & Knaap, 1978; Jayasuriya & Ratnayaka, 1973; Gupta, 1974).

3.2.5.5 SUMMARY OF IN VIVO STUDIES

The COM noted that there were a large number of in-vivo studies available. Dichlorvos was negative in most published in-vivo mutagenicity assays where it was administered as a single dose. These included mouse bone-marrow micronucleus (using the ip route) and bone marrow chromosome aberration studies in mice and hamsters using oral and, in two studies (mice/hamster) inhalation exposure. Negative results were also reported in SCE in mice and UDS assays (liver (rats)/forestomach (mice)). A negative result was also reported in an adequately conducted bone-marrow chromosome aberration study where mice were given daily oral doses of dichlorvos by gavage for 5 days. Studies in germ cells have given negative results including dominant lethal assays and a chromosome aberration study in mouse spermatogonial cells. The COM also noted that there were a number of positive studies. The COM concluded that a consistent pattern of mutagenic effects had been documented in the in vivo studies in which dichlorvos induced mutagenic effects at high doses in the skin following topical application, and in the liver following repeated intraperitoneal dosing, suggesting a potential site-of-contact effect (i.e. at initial sites of exposure).

90

3.2.5.6 HUMAN STUDIES

A report of a study on CA and SCE in workers exposed to mixed pesticides, including dichlorvos, in the flower industry in north west Italy is available (De Farrari et al., 1991). It is not possible to draw any conclusions about dichlorvos from this study. CA in peripheral lymphocytes were studied in 31 patients, admitted between April 1971 and November 1972, with OP poisoning (van Bao et al., 1974). Chromosome analysis was performed on lymphocytes immediately after intoxication and at intervals afterwards. One patient (39 year old female) had moderate poisoning due to ingestion of dichlorvos. This patient had an increased number of non-modal cells (10/64 cells) containing less than 46 chromosomes. None of the cells analysed contained breaks or exchanges.

3.2.6 CARCINOGENICITY

3.2.6.1 RAT

Fischer 344 rats, 50 per sex per dose, were administered 0, 4 or 8 mg kg-1 d-1 dichlorvos (99 % pure) by gavage in corn oil 5 d wk-1 for 103 weeks (NTP, 1989). There was no significant difference in survival (48 - 64 % for males and 52 - 62 % for females) or body weight gain between the control and test groups. Mild diarrhoea was stated to be compound related. No other clinical signs were observed. Cholinesterase activity was not measured during this study. However a subsequent study using groups of 10 rats per sex per dose indicated that erythrocyte cholinesterase activity was significantly depressed in males but not females at 4 and 8 mg kg-1 d-1. Detailed necropsies and histopathological examination were performed on all animals. Pancreatic lesions were observed in both males and females as shown in Table 3.4.

Table 3.4 Pancreatic lesions in male and female rats

Vehicle control 4 mg kg-1 d-1 8 mg kg-1 d-1 Males Hyperplasia 9/50 9/49 9/50 Adenoma 16/50 25/49 30/50 Females Hyperplasia 2/50 4/47 0/50 Adenoma 1/50 1/47 4/50 Atrophy 5/50 6/47 15/50

The incidence of single and multiple pancreatic adenomas showed a significant positive trend in males but not in females. In control males the incidence was above the historical control range (0 - 12/50). When supplementary horizontal sections of the pancreas were examined, additional acinar cell hyperplasia and adenomas were observed, particularly in the control group. The combined incidence of single and multiple adenomas in males remained significantly greater than the vehicle controls but for single adenomas alone the difference was no longer significant. Although additional lesions were seen in females when these sections were examined, no significant differences emerged.

91

There was a significant increase and positive trend in the incidence of mononuclear cell leukaemias in males (control 11/50, 4 mg kg-1 d-1 20/50 and 21/50 at 8 mg kg-1 d-1, historical control range up to 22/50). This was not seen in females (control 17/50, 4 mg kg-1 d-1 11/50 and 23/50 at 8 mg kg-1 d-1). Alveolar/bronchiolar adenomas were identified in the lungs of 3/49 high dose males only. The incidence of fibro-adenomas of the mammary gland was significantly increased (9/50 in control, 19/50 at 4 mg kg-1 d-1 and 16/50 at 8 mg kg-1 d-1, historical control range up to 21/50). Adenomas (control 0/50, 4 mg kg-1 d-1 0/50 and 8 mg kg-1 d-1 1/10) and carcinomas (2/50 in controls, 2/50 at 4 mg kg-1 d-1 and 0/50 at 8 mg kg-1 d-1) showed no change in incidence. It is considered that this study was adequate for the assessment of the carcinogenic potential of dichlorvos. Increased incidences of pancreatic adenomas and mononuclear cell leukaemia in males and mammary gland fibro-adenomas in females were identified. Mononuclear cell leukaemias and mammary gland tumours have a high background incidence in F344 rats and in all cases the incidences in the treated groups were within the historical control range. Corn oil has been shown to increase the rate of proliferative lesions of the pancreas in male F344/N rats (Eustis & Boorman, 1985). The pancreatic adenomas (incidence in the control group being above the historical control range) are considered to be associated with the use of corn oil as a vehicle. This is supported by the fact that no pancreatic lesions were reported in dietary or inhalation carcinogenicity studies in other species of rats; no evidence of pancreatic lesions are reported in well-conducted 90-day studies in rats; also there is no evidence of pancreatic lesions in studies involving mice and dogs. Therefore it is considered that no treatment related carcinogenicity was seen in this study. Groups of 50 per sex per dose Osborne-Mendel rats were administered 150 or 300 ppm dichlorvos (94 % pure) in the diet for 80 weeks and observed for a further 30 weeks (NCI, 1977). The high dose was originally set at 1000 ppm, but because of signs of toxicity this was reduced to 300 ppm after 3 weeks. The control group consisted of 5 per sex per dose that were matched to the treatment group. However a pooled control drawn from other concurrent chronic studies was available but these animals were housed separately (60 per sex). No details of dietary intake were provided. Body weight gain was reduced in males only administered 300 ppm over the 80 weeks of treatment (approximately 10 %). Adverse clinical signs (rough hair, epistaxis, haematuria, alopecia, dark urine, palpable masses and bloating or abdominal distension) were evident in both treated and control animals. The incidence of these signs increased in frequency in treated groups, particularly 300 ppm females, during the second year. There were no statistical differences in survival for males and females between matched controls and dosed animals over 105 weeks of the study. Survival at 105 weeks in males was 64 % at 150 ppm and 76 % at 300 ppm and in females 84 % at 150 ppm and 80 % at 300 ppm. The only tumour occurring with a significant positive trend was malignant fibrous histiocytoma of the subcutis in males. The incidences in males from the pooled control, matched control, low and high dose groups were 2/58, 1/10, 4/48 and 8/50 respectively. The incidence in the matched controls exceeded that in the low dose group and was not

92

statistically significantly different from that in treated groups. For female rats the incidences were 1/60, 0/10, 5/48 and 1/50 respectively. Thus in this study there was no evidence of carcinogenicity. However the study is considered unsatisfactory for a proper carcinogenicity assessment, as the number of control animals was insufficient and achieved doses not determined. In an inhalation study, groups of Carworth Farm E (CFE) rats, 50 per sex per dose, were continuously exposed to nominal concentrations of 0, 0.05, 0.5 and 5.0 mg m-3 of dichlorvos vapour (> 97 % pure) for two years (Unpublished, 1976). The average concentrations of dichlorvos were 0, 0.05, 0.48 and 4.70 mg m-3. Dichloroacetaldehyde and trimethyl phosphate were detected in the atmosphere when tested during the first year (0.028 and 0.04 mg m-3 respectively in the 5.0 mg m-3 group). The rats were housed in individual cages within a 10 m3 inhalation chamber. Food and water were available ad libitum. Exposures were interrupted as necessary but this did not exceed 1 h d-1. Necropsy and full histopathological examinations were carried out on all animals. For groups exposed to 0, 0.05, 0.48, 4.7 mg m-3 the survival was 22, 42, 30 and 64 % respectively for males and 50, 60, 58 and 76 % respectively for females. Due to the high mortality in control males, all males were withdrawn from treatment at 100 weeks. No clinical signs of organophosphorus poisoning were observed. The body weight of rats exposed to 4.7 mg m-3 was significantly depressed throughout the study (approximately 10 - 13 %). No consistent differences in food consumption among the groups were found. At termination assays indicated that plasma, erythrocyte and brain cholinesterase activities were significantly depressed. Brain acetylcholinesterase activity was approximately 20 % of controls at 4.7 mg m-3 in both males and females. There were no treatment related effects on tumour incidence. Thus in this study it can be concluded that dichlorvos was not carcinogenic to CFE rats. However the study is limited by the low survival in control males. In a study, for which only limited details are available, groups of F344 rats (50 per sex per dose) were administered 0, 140 and 280 ppm dichlorvos (unknown purity) daily in drinking water for 104 weeks (Unpublished, 1987b). Administration was followed by a 4 week recovery period. The actual achieved doses were not presented and only limited details were available. There was no statistically significant difference in mortality rates for either male or females (14 - 25 %). Male and females in both treatment groups showed reduced body weight compared to the controls. However statistical analysis was not presented. A number of malignant and non-malignant changes were observed but none showed any dose related change and individual data were not presented. No conclusions can be drawn form this study because of the limited information available. In another study, in-bred rats of the strain BDIX/Bln were administered dichlorvos (97 % pure) by gavage for 60 weeks and animals observed for up to 111 weeks (Horn et al., 1988). Groups were; vehicle control (59 males and 60 females) receiving 0.2 ml of water 3 times a week, low dose (70 per sex) receiving 0.1 mg dichlorvos per animal 2

93

times a week and high dose (99 per sex) receiving 0.1 mg dichlorvos per animal 3 times a week. There was no significant difference in survival or body weight between any of the treatment groups. No treatment related increase in the number of tumours was seen. A number of pre-neoplastic lesions (focal hyperplasia and papillomas of the forestomach) were identified in the GI tract. However the increase in incidence in the treated groups was not significant. Treated males had increased rates of proliferation of bile duct cells or oval cells of the liver (12/40, 23/48 and 41/77 in control, low and high dose groups). This was significant only in high dose males. No changes were observed in females. A non-significant dose related increase in adrenal phaeochromocytomas in males was observed, 0/30, 2/37 and 4/60 in control, low and high dose respectively. In females the incidence was 1/29, 3/28 and 5/44 for the three groups respectively. This study is inadequate for an assessment of the carcinogenic potential of dichlorvos because of its short dosing period and the low doses used. An abstract (which did not include full experimental details) gave results of 80 week carcinogenicity studies in Osborne-Mendel rats (Robens, 1978). Dichlorvos was administered in food at the MTD and 50 % MTD. A recovery period of 30 weeks was included. No increase in tumours in dichlorvos treated animals were reported. However no conclusions can be drawn from this study because no data were presented. In a poorly reported drinking water study, rats (F344; 50/sex/group) were exposed to 0, 20 or 40 mg kg-1 d-1 dichlorvos for 104 weeks. Surviving animals were sacrificed 4 weeks after the end of exposure. Animals were observed for mortality and signs of toxicity; and weekly observations of bodyweight, food and water consumption were made. All animals were examined macroscopically and the weights of the brain, pituitary, thymus, heart, lung, liver, kidneys, spleen and testes/ovaries were determined. Microscopic examination of these organs and of the eye, harderian glands, skin, subcutis, mammary gland, skeletal muscle, thyroids, trachea, eosophagus, pancreas, stomach, small and large intestine, urinary bladder, prostate, uterus, lymph nodes, bone, bone marrow, spinal cord and peripheral nerves was performed in 20 animals of the control and 40 mg kg-1 d-1 groups, 10 animals of the 20 mg kg-1 d-1 group (and the liver, lung, kidneys and endocrine organs of a further 10 males and females of this group), and all animals found dead or moribund during the study. All tumours and macroscopic lesions were examined microscopically. No statistically significant difference in mortality was observed between treated and control groups. Bodyweight gain was slightly reduced in animals of the 40 mg kg-1 d-1 group, but no other signs of toxicity were reported. No organ weight data were provided in the study report. No treatment-related increase in tumour incidence was observed between treated and control animals (Unpublished, 1987c). Even though relatively poorly reported, this study indicates that dichlorvos exposure via the drinking water does not lead to increased incidences of tumour formation at doses up to 40 mg kg-1 d-1.

94

3.2.6.2 MICE

Groups of 50/sex/dose B6C3F1 mice received 0, 10 or 20 mg kg-1 d-1 (males) or 0, 20 or 40 mg kg-1 d-1 (females) dichlorvos (99 % pure) by gavage in corn oil 5 d wk-1 for 103 weeks (NTP, 1989). Survival was 70 % male and 52 % females (control group), 56 % males and 58 % females (low dose group) and 58 % males and 68 % females (high dose group). No significant difference in mortality was seen for either males or females. No compound related clinical signs or differences in body weight gain were noted during the study. A subsidiary study showed that groups of 10 per sex per dose 8 week old mice administered 5, 10, 20 or 40 mg kg-1 d-1, 5 d wk-1 by gavage in corn oil for 33 d had depressed plasma, but not erythrocyte, cholinesterase activities. Squamous cell papillomas were increased in the forestomach of males (control 1/50, 10 mg kg-1 d-1 1/50 and 5/50 at 20 mg kg-1 d-1) and females (control 5/49, 20 mg kg-1 d-1 6/49 and at 40 mg kg-1 d-1 18/50). Forestomach carcinomas were increased in females only (control 0/49, 20 mg kg-1 d-1 0/40 and at 40 mg kg-1 d-1 2/50). One high dose female was found to have both papilloma and carcinoma. The incidence of hyperplasia of the forestomach showed no treatment related changes (in males 11/50, 1/50 and 9/50 and in females 6/49, 7/49 and 8/50 in control, low and high dose groups respectively). The historical range of incidence for papillomas in male mice at the laboratory was 0/49 to 3/49. Only in one study had a carcinoma been reported. In female mice the laboratory historical incidence was 0/50 to 2/50 for papillomas with no carcinomas being reported. For males, although there were no significant differences between the control and treated groups for papillomas, there was a significant positive trend. For females the difference in incidence between the control and low dose groups was not significant for either papillomas alone or papillomas and carcinomas but was significant between control and the high dose group. However there was a significant positive trend when the incidence of papillomas or papilloma and carcinomas was considered. Thus it can be concluded that there is some evidence of carcinogenicity in this study, with an increase in the incidence of forestomach papillomas in males and females and carcinomas in females in the absence of an increased incidence of hyperplasia. It should be noted that the incidence of papillomas in the female control group exceeded the historical control range. Groups of B6C3F1 (50 per sex per dose) were administered 300 or 600 ppm dichlorvos in the diet for 78 weeks (NCI, 1977). For the first two weeks the doses were 1000 and 2000 ppm but these were reduced because of severe signs of toxicity. The matched controls consisted of 10 animals per sex. Pooled controls (matched controls from concurrent studies on other chemicals) consisted of 100 males and 80 females. Data on dietary intake were not presented. There was no significant dose related change in survival in any of the groups. Body weights in mice receiving 600 ppm were slightly lower than controls and there were no clinical signs of toxicity when doses were 300 and 600 ppm. Several unusual proliferative lesions were identified in the digestive system of treated animals only. Squamous cell carcinoma of the oesophagus was seen in 1/49 low dose males and 1/41 high dose females. One oesophageal

95

papilloma (high dose female) and one focal hyperplasia of the stomach (high dose female) were also noted. Historical control data were not available for this study. Therefore it was not possible to determine whether the incidence of these tumours was within the historical range. No conclusion can be drawn from this study because of the absence of historical control data and the low numbers of matched controls. Groups of C57B1/6/Bln mice were administered dichlorvos (97 % pure) by gavage in water for 50 weeks (Horn et al., 1987). The animals were observed until the study was terminated at 110 weeks. The test groups received 0.2 mg dichlorvos either 2 or 3 times a week, control received water three times a week. Group sizes were unclear, but dichlorvos treated groups were at least 70/sex and control groups were at least 30/sex. Compared with the untreated control group, mean survival in the vehicle control and treated groups was reduced. When compared with the untreated control group, a significantly increased occurrence of focal bladder hyperplasia was seen in vehicle control and treated groups (both sexes). Bladder papillomas were not increased. This study is inadequate for assessing the carcinogenicity of dichlorvos because of the limited dosing schedule and the short duration. An abstract (which did not include full experimental details) gave results of an 80-week carcinogenicity study B6C3F1 mice (Robens, 1978). Dichlorvos was administered in food at the MTD and 50 % MTD. A recovery period of 10 weeks was included. No increase in tumours in the dichlorvos treated animals were reported. No conclusions can be drawn from this study.

3.2.6.3 MECHANISTIC STUDIES

Using non-standard methodology, Benford et al., (1994) investigated the possible mechanism by which dichlorvos could induce forestomach tumours in mice. Groups of mice (B6C3F1; 5 per sex per dose) were administered single gavage doses of either 200 mg kg-1 1-methyl-3-nitro-1-nitrosoguanidine (MNNG; a known genotoxic forestomach carcinogen), 300 mg kg-1 butylated hydroxyanisole (BHA; a non-genotoxic forestomach carcinogen) or 10, 20 40 or 100 mg kg-1 dichlorvos (purity 99.8 %). After various periods of between 2 - 48 h the animals were sacrificed and the stomachs removed. Stomach tissue was examined for evidence of unscheduled DNA synthesis (UDS), replicative DNA synthesis (RDS) or hyperplastic changes. MNNG induced UDS but not RDS or hyperplasia in mouse forestomach epithelium, consistent with its genotoxic mode of action. Both BHA and dichlorvos did not induce UDS, but did induce RDS and hyperplasia. This suggests the forestomach tumours induced by dichlorvos in mice could occur by a non-genotoxic mechanism, in which local tissue damage plays a key role.

3.2.6.4 HUMAN STUDIES

A survey of leukaemia cases linked to farming in Iowa and Minnesota is available (Morris Brown et al., 1990). No conclusions could be drawn from this study because no population was exposed to dichlorvos only.

96

3.2.6.5 SUMMARY

Of the rat studies available only two (one by gavage, one inhalation) are considered to be adequate for a proper assessment of the carcinogenicity of dichlorvos. No evidence of carcinogenicity was seen in the inhalation study in CFE rats. In the gavage study with F344 rats using corn oil as vehicle the increased incidence of mononuclear cell leukaemias (males) and mammary gland tumours (females) were within the historical control range and not considered to be treatment related. The pancreatic lesions (acinar cell hyperplasia and adenoma) may be related to the use of corn oil as vehicle. An additional adequate drinking water study provided no evidence of treatment-related tumour formation in F344 rats. Four studies in mice are available but only one is considered to be entirely adequate for the assessment of carcinogenicity. In this study, in B6C3F1 mice, there was an increase in squamous cell papillomas of the forestomach in males and females (significant in females only) and carcinomas of the forestomach in females only. In a second study, in which historical control data were not available, squamous cell carcinoma of the oesophagus was seen in one low dose male and one high dose female and an oesophageal papilloma in a high dose female. Relevant members of the Department of Health Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC) were asked by the COM to assess the carcinogenicity studies sumitted for this review. Regarding the studies in rats the COC members concluded that there were limitations in most of the studies (e.g. age of study, numbers of animals used, extent of pathology investigations) and that there was no clear evidence for a carcinogenic effect in rats. Regarding other studies in mice, COC members considered that there were limitations in the conduct of these studies similar to those undertaken in the rat. Members considered that it was not possible to undertake a comparison of the studies in mice where dichlorvos had been administered in corn oil and those where dichlorvos had been administered in the drinking water or as an aqueous solution by gavage. COC members reaffirmed, that when the NCI and NTP bioassays in mice were considered together there was limited evidence for an effect on squamous epithelium of the forestomach and oesophagus in mice. However the latter study should be viewed in terms of its age and small number of oesophageal tumours. On considering the overall package of carcinogenicity bioassays COC members felt that there was no consistent evidence for a genotoxic carcinogenic effect. COC members noted that there was no agreed mechanism for the forestomach tumours.

3.2.7 COM CONCLUSION ON THE GENOTOXICITY OF DICHLORVOS

The COM’s conclusion on the genotoxicity of dichlorvos is reproduced below. The full COM statement on dichlorvos can be found at http://www.advisorybodies.doh.gov.uk/Com/dichlorvos.htm (COM, 2002). ‘The COM agreed that there is clear unequivocal evidence that dichlorvos can induce DNA damage, chromosomal breakage and mutations in mammalian cells from in-vitro studies.

97

The compound has been shown to interact with DNA via methylation, however several other mechanisms are theoretically possible. In-vivo, dichlorvos can be rapidly detoxified by hydrolysis before it reaches the systemic circulation. Members noted from the HSE review [undertaken by HSE on behalf of the ACP] that retention of 14C-vinyl-labelled dichlorvos in skin was recorded in a study where radiolabelled dichlorvos was applied to the skin on the backs of male rats. Several non-standard in-vivo mutagenicity assays have indicated that dichlorvos can induce genetic damage when systemic detoxification mechanisms are bypassed, e.g. following exposure to the skin and exposure to the liver following intraperitoneal dosing. The COM agreed that there was a potential risk of mutagenicity at site of contact tissues, i.e. at the initial sites of exposure. The COM felt there was no evidence for systemic mutagenic effects. The COM agreed that until evidence was provided to the contrary and in the absence of appropriate mechanistic data, a precautionary approach should be adopted and no threshold could be assumed for the mutagenic activity of dichlorvos. Members were aware that there was some limited evidence for a carcinogenic effect in mice from standard bioassays. This related to an increase in squamous cell papillomas of the forestomach in mice and carcinomas of the forestomach in female mice given gavage doses of dichlorvos together with the finding of squamous cell papilloma and carcinoma of the oesophagus in a small number of mice. Members noted there was no evidence for carcinogenicity from a number of other carcinogenicity bioassays including an inhalation bioassay in the rat, although there were limitations with all of these studies. Members noted that negative results had been obtained with dichlorvos in a single dose UDS assay in the forestomach of mice using gavage dosing. An increase in replicative DNA synthesis had been reported in this study. Members noted that there were a number of proposals regarding the mechanism of dichlorvos tumourigenicity in the mouse forestomach including localised irritancy of dichlorvos in corn oil. The COM agreed that this proposal had not been proven and considered that it was not possible to exclude a genotoxic effect from these data given the relative insensitivity of the method used as indicated by the response with the positive control chemical; they felt that repeat dosing would most likely be required to identify any mutagenic effect of dichlorvos in this assay. The COM concluded that dichlorvos should be regarded as an in vivo mutagen at the site-of-contact (i.e. at the initial sites of exposure). The COM considered there was no evidence for systemic mutagenic effects. High doses of dichlorvos induced mutagenic effects in the skin following topical application and in the liver following intraperitoneal dosing. The COM noted the limited evidence for a carcinogenic effect of dichlorvos. This related to tumours of the forestomach in mice after gavage dosing and also the oesophageal tumours seen after dietary administration. There was no satisfactory explanation proven for the mechanisms of these tumours and the COM felt, given the available mutagenicity data on dichlorvos, that it would be prudent to assume a genotoxic mechanism. The COM agreed that in the absence of appropriate mechanistic data, a precautionary approach should be adopted and no threshold could be assumed for the mutagenic and carcinogenic effects of dichlorvos.’

98

3.2.8 REPRODUCTIVE TOXICITY

3.2.8.1 FERTILITY

3.2.8.1.1 Rat

In a study performed to GLP and consistent with current EC guidelines rats of the F0 generation (Sprague-Dawley; 30 per sex per group) were administered 0, 5, 20 or 80 ppm dichlorvos (purity 98 %) in drinking water 7 d wk-1 for 10 weeks. The dose levels in terms of mg kg-1 d-1 varied through the study depending mainly on the status of the females. For males the equivalent doses to 5, 20 and 80 ppm were approximately 0.5, 1.9 and 7.2 mg kg-1 d-1, respectively; and for females 0.6 - 1.2, 2.1 - 4.6 and 7.0 - 17.5 mg kg-1 d-1, respectively. Animals were then randomly mated within treatment groups for a three-week mating period to produce the F1 generation, with exposure continued. F0 males were sacrificed following mating and the testes of all animals examined microscopically; the pituitaries and livers of control and high dose animals were also examined. F1 litters were culled to 8 pups on postnatal day 4 and weaned on postnatal day 21. At weaning, 10 weanlings per sex per dose underwent gross necropsy and 30 per sex per dose were selected as F1 parents of the F2 generation. Following a further 11 week exposure the selected F1 animals underwent a 3 week mating period. Following weaning 10 animals per sex per dose of the F2a generation underwent gross necropsy. Due to an unexpectedly poor breeding performance in all F1 animals, including the controls, more detailed additional studies in F1 animals were performed. F1 males were necropsied in the same manner as the F0 group; in addition 20 control males and 10 from each dose group were subjected to a detailed male reproductive assessment including evaluation of sperm number, motility and morphology, reproductive and sex accessory organ weights and microscopic examination of one testis per male. After weaning of the F2a litters, parental F1 females were evaluated for oestrous cyclicity for three weeks and then rebred to naive (untreated) males, to produce the F2b generation. Mating, gestation and lactation were performed as described previously with 10 F2b weanlings per sex per dose necropsied. After weaning of the F2b litters, parental F1 females were then necropsied with ovaries weighed and histopathology performed as described previously. Cholinesterase activity was determined in the erythrocytes, plasma and brain of all parental animals of the F0 and F1 generations. No deaths that could be attributed to treatment were observed in the F0 generation. During the pre-breed exposure period, there were no treatment-related effects on bodyweight gain or food consumption; however, water consumption was statistically significantly reduced in animals of the 80 ppm group. During gestation bodyweight gain was statistically significantly reduced only in F0 females of the 20 ppm group (85 % of control values), however, this effect was not observed during the lactation period. Food consumption was unaffected by treatment in F0 females, but water consumption was statistically significantly reduced in animals of the 80 ppm when compared with controls. Signs of toxicity included an increased incidence of alopecia in animals of the 20 and 80 ppm groups during gestation and in animals of the 80 ppm group during lactation. Cholinesterase activity was reduced in a dose-related manner in both sexes of the F0 generation in erythrocytes (males: 93, 71 and 43 % of control values at

99

5, 20, and 80 ppm, respectively; females: 77, 61 and 40 % of control values, respectively), plasma (males: 96, 71 and 59 % of control values, respectively; females: 88, 45 and 17 % of control values, respectively) and brain (males: 99, 85 and 47 %, of control values, respectively; females: 94, 74 and 41 % of control values, respectively). No gross or microscopic findings were observed at necropsy of animals of the F0 generation. No treatment-related effects on any parameter of fertility was reported in F0 animals. During the lactation period no treatment-related findings were reported in F1 pups; gross necropsy did not reveal any treatment-related changes in the pups, including decedents. No deaths that could be attributed to treatment were observed in the F1 generation. F1 parental animals generally showed no treatment-related changes in bodyweight gain or food consumption throughout exposure, however, water consumption was statistically significantly decreased in the 80 ppm group. In addition, decreased food consumption was observed in F1 females of the 80 ppm group during lactation of the F2b generation, and decreased water consumption was observed in F1 females of the 20 ppm group during gestation of the F2b generation. No other treatment-related signs of toxicity were reported in the F1 animals. Cholinesterase activity was reduced in a dose-related manner in both sexes of the F1 generation in erythrocytes (males: 86, 68 and 45 % of control values at 5, 20, and 80 ppm, respectively; females: 82, 58 and 42 % of control values, respectively), plasma (males: 85, 74 and 42 % of control values, respectively; females: 91, 46 and 19 % of control values, respectively) and brain (males: 101, 94 and 60 %, of control values, respectively; females: 98, 68 and 40 % of control values, respectively). Necropsy of the F1 animals revealed no gross or microscopic treatment-related changes. No treatment-related statistically significant effects on parameters of fertility were apparent in the F1 animals producing the F2a and F2b generations. Necropsy of the F2a and F2b pups revealed no treatment-related findings. The assessment of the male reproductive system in F1 parent animals revealed no treatment-related differences between the groups apart from statistically significant increases in relative weights of the right epididymis; and empty seminal vesicles at 80 ppm. However, absolute values showed no differences between groups. Evaluation of the oestrous cycle in F1 females revealed a decreased number of females cycling and an increased number of cycling females with an abnormal cycle at 80 ppm (Unpublished, 1992c). This study indicates a NOAEL for effects on fertility parameters of 20 ppm (1.9 - 4.6 mg kg-1 d-1) based on the oestrous cycle abnormalities at 80 ppm, however it is noted that the F1 parents produced the F2 generation without a fertility problem. Although, F0 females of the 5 ppm group showed a 23 % inhibition of erythrocyte cholinesterase activity, this finding was not repeated in F1 females nor in brain acetylcholinesterase measurements of either generation or either sex. Thus, it is thought that this is a chance finding and not biologically significant and an overall NOAEL of 5 ppm (0.5 - 1.2 mg kg-1 d-1) for adult toxicity is indicated based on inhibition of cholinesterase activity at higher dose levels. A study reported by Desi & Nagymajtenyi (1999) was designed to investigate the changes of spontaneous and evoked electrical changes in processes of the central and peripheral nervous systems following exposure to moderate doses of dichlorvos following subchronic exposure (up to 12 weeks), over 3 consecutive generations, or during different stages of development. For the 3-generation study, dichlorvos treatment began in groups of 4-week old rats

100

administered gavage doses of either 0, 0.98, 1.96 or 3.92 mg kg-1 d-1 dichlorvos (purity 98 %) in distilled water. These animals were then divided into 2 groups; 10 males/dose for investigations and a group to produce the next generation (10f, 5m per dose). The animals selected to produce the next generation were treated for 7 weeks before mating. Once pregnant the females were treated 7 d wk-1 until weaning when offspring were 4 weeks old. After weaning the rats were treated with dichlorvos 5 d wk-1 for 8 weeks. This cycle was repeated until the F3 generation. The neurotoxicological investigations previously described (in Section 3.2.4.1.1) were then conducted on 12 week old male rats of each generation (F1, F2, F3). No treatment-related signs of toxicity, effects on bodyweight gain or organ weight were reported. Also parameters of fertility were unaffected by treatment. The authors report that the ECoG investigation showed dose and generation-dependent decreases in mean amplitude; and decreases in slow part and increases in fast part of power spectra. The somatosensory, visual and auditory ECoG mean frequencies and latency times of evoked potentials showed dose and generation-dependent increases which were statistically significant, compared to control animals, in the top dose groups of all 3 generations and in the 1.96 mg kg-1 group of the F2 and F3 generations. Conduction velocity and refractory periods were decreased in a dose and generation-dependent manner, being statistically significantly different from controls in the 1.96 and 3.92 mg kg-1 groups. Although brain cholinesterase activity was reported to be inhibited in the treated groups, the inhibition did not reach statistical significance. Overall, this study indicates no effects on fertility up to the top dose used in the study 3.92 mg kg-1 d-1 (consistent with the previous study). As erythrocyte acetylcholinesterase activity was not measured a NOAEL for systemic toxicity could not be established, although changes in the electrical processes measured were observed at dose levels above 0.98 mg kg-1 d-1. A further study is reported in abstract form only. Rats (Wistar; number unknown) were administered oral doses of 0.97, 1.29 or 1.94 mg kg-1 of dichlorvos throughout three consecutive generations including pregnancy and lactation. General toxicological (body weight gain, the weight of 9 organs), haematological (absolute and differential WBC, RBC, Ht, mean volume of RBCs, bone marrow cellularity), and immune function (PFC number of the spleen, DTH reaction) parameters in 8 weeks old males of each generation were determined. At the same time, neurotoxicological parameters (ECoG, somatosensory, visual, and auditory cortical evoked potentials, conduction velocity and refractory periods of peripheral nerve) were recorded at the age of 9 weeks in separate groups of animals. Brain acetylcholinesterase activity was also measured. Among the general toxicological parameters the relative liver weight increased significantly in the second, and the relative thymus weight in the third generation. A decrease in some haematological parameters Ht, MCV, and bone marrow cellularity) was also observed in the second generation, in the third one these changes became more characteristic. The PFC content of the spleen decreased in the first generation only at the top dose, in the second one at the middle and top doses, in the third one at all the three doses. Among the neurological parameters a significant decrease in the ECoG index, and an increase of the absolute and relative refracter periods were found at all doses in all the three generations. The applied

101

doses had no significant effect on the acetylcholinesterase activity of the brain. On the basis of these findings the ECoG index and the refracter periods of the peripheral nerve proved to be the most sensitive parameters in the applied system (Institoris et al., 1997). A three-generation study in CD rats has been conducted but is available only in summary form with very limited details (Witherup et al., 1971). Rats (30 per group, sex ratio not given) were fed dichlorvos (93 % pure) at nominal concentrations of 0, 0.1, 1, 10, 100 and 500 ppm. No effects on fertility, number and size of litters, bodyweight or viability of pups were found. Gross and histopathological examination did not reveal any abnormalities. The actual concentrations of dichlorvos were not measured but were reportedly equivalent to 0, 0.0025, 0.025, 0.25, 2.5 and 12.5 mg kg-1 d-1. The following studies (in mice, rabbits and other studies) were reported in the 1994 HSE review of dichlorvos but were considered to be inadequate to draw conclusions about the potential for dichlorvos to affect fertility. They are included as additional information.

3.2.8.1.2 Mice

A poorly reported and conducted two generation study in C57 and Swiss mice is available (D'Souza & Batra, 1976). Male and female C57 mice (45 control and 35 exposed) were housed in groups of 3 females/1 male. Following successful mating (method of confirmation not given), females were transferred to dichlorvos treated or non-treated cages. Treatment consisted of spraying 2.1-2.4 g dichlorvos into each cage once a week. Treatment continued throughout gestation but was stopped for 2 weeks to allow weaning of the F1 generation. After weaning, treatment was restarted and F0 females were remated. Once mature, F1 animals were also mated using a ratio of 3 females per male. Mating and treatment continued until 10 months for the F0 and 8 months for the F1 generation. Swiss mice (40 control and 43 exposed) were treated using the same regime. All animals (both strains) were maintained untreated until death or sacrificed moribund. Histology was performed on the testes and ovaries. A significant reduction in survival was observed in treated mice. Mortality, at 14 months, was 60 - 66 % in C57 exposed F0 and F1 animals compared with 17 % in controls and, in Swiss mice, 29 - 37 % in exposed F0 and F1 animals compared with 15 % in controls. Overt signs of toxicity, consisting of ruffled fur, breathing difficulties and diarrhoea, were seen in treated mice (no further details given). In control C57 mice all females produced litters, and over the period of the experiment each female had an average of 6 litters (average size was 7.4). No results for a control F1 generation were given. In the exposed F0 generation 1/25 females failed to produce a litter but the average number of litters/female was markedly lower (1.2, average size was 8.4). In the exposed F1 generation 19/21 females failed to litter. The remaining females only had 1 litter each (average size 5.5). In control Swiss mice no females failed to litter and the average number of litters per female was 6.1 (average size of 6.3). In the treated F0 generation 23/24 females produced litters (average of 1.75/female, average size of 6.4). In the F1 generation 37/61 failed to litter. Of the remainder there was an average of 1.8 litters/female (average size of 6.4).

102

Histology of the testes from exposed mice (incidence not given) revealed the presence of large vacuoles resulting from degeneration of seminiferous tubules, absence of Germ cells and cryptorchism. Testicular tumours were seen in 2 treated mice. Ovaries were hyalinised in nature and follicles and oocytes were absent in a large number of mice. Thus in this study there is evidence of failure to litter and decreased littering frequency in dichlorvos exposed C57 and Swiss mice. Additionally there may have been a decrease in litter size in C57 mice. Whilst there may be an association with exposure to dichlorvos, the exposure level cannot be determined from the dosing regime. The mortality data suggest that exposure was high.

3.2.8.1.3 Rabbit

An inadequately conducted and poorly reported fertility study in the French long-eared rabbit is available (Majewski et al., 1979). Dichlorvos (purity unknown) was added to the diet at a concentration intended to give a daily dose of 5 mg kg-1 for one month before mating for both sexes, and throughout pregnancy and lactation for females only. It was not clear whether this daily dose was achieved. Animals (8-12 females and 5 males per group) were assigned to groups receiving no dichlorvos, females only treated, males only treated or both sexes treated. Females were mated with 2 males from their group. Clinical signs of toxicity (nervous agitation, increased pulse and respiration rate, dyspnoea, muscle tremor and abnormal position) were seen in dosed animals but a high degree of individual variation was observed. Dichlorvos exposed males had significantly reduced bodyweights after 45 d. Bodyweights were not presented for females. Measurements of blood or serum parameters on days 1 - 7 indicated: a decrease in plasma cholinesterase activity to 54 % of control on day 1, rising to approximately 80 % on days 3 and 7 and an increase in alanine aminotransferase. There were no obvious differences between the control and any treated group for fertility rate or litter sizes. No results for pup weight were presented. This study is inadequate to assess the reproductive toxicity of dichlorvos because of the short dosing period in males and the low numbers of animals used.

3.2.8.2 OTHER STUDIES

Evidence of effects on the testes (decrease in testes weight, damaged seminiferous tubules, and a reduction in number of Sertoli and Leydig cells) have been observed in rats administered > 10 mg kg-1 d-1 for 18 d or 40 mg kg-1 as a single dose (Krause & Homola, 1974; Krause et al., 1975; 1976). No effect on male sex hormone levels was observed at doses up to 10 mg kg-1 every other day for 2 - 3 weeks (Krause, 1977). In mice a significant increase in the number of abnormal sperm was seen after ip administration at and above 7.5 mg kg-1 d-1 for 5 d (Wyrobek & Bruce, 1975). The onset of first oestrus was significantly delayed in weanling female rats exposed to 2.4 mg m-3 dichlorvos vapour (Timmons et al., 1975). In female mice exposed to up to 4 mg m-3 dichlorvos vapour, during mating or throughout mating and gestation there was no effect on number of litters, mean litter size or gestation length (Casebolt et al., 1990).

103

3.2.8.3 SUMMARY OF FERTILITY STUDIES

No data in humans are available. A well-conducted 3 generation rat drinking water study is available. Although the standard parameters of fertility remained unaffected by treatment, additional analyses carried out during the study indicated that female animals of the 80 ppm group (7 - 17 mg kg-1 d-1) developed treatment-related abnormalities in oestrous cyclicity. Thus, a NOAEL for effects on fertility of 20 ppm (1.9 - 4.6 mg kg-1 d-1) was apparent. The overall NOAEL for adult toxicity was 5 ppm (0.5 - 1.2 mg kg-1 d-1) based on significant (> 20 %) decreases in acetylcholinesterase activity (brain and erythrocyte). This finding is consistent with the results of a study which primarily investigated electrical process changes in the nervous system in 3 generations of rats following dichlorvos exposure, in which no effects on fertility were observed at the top dose used of 3.92 mg kg-1 d-1. A NOAEL for adult toxicity could not be established as erythrocyte acetylcholinesterase activity was not measured. Few conclusions can be drawn from other less robust studies.

3.2.8.4 DEVELOPMENTAL TOXICITY

3.2.8.4.1 Rat

In a study performed to GLP and consistent with current EC guidelines, presumed pregnant female rats (Sprague-Dawley; 25/group) were administered gavage doses of 0, 0.1, 3 or 21 mg kg-1 d-1 dichlorvos (purity 96.9 %) in water on days 5 - 15 of gestation. Animals were sacrificed on gestation day 20 and examined to confirm pregnancy. The bodyweight and weights of the liver and uterus were recorded; ovarian corpora lutea were counted; and uterine contents recorded. All live fetuses were weighed and examined for abnormalities. At sacrifice, a low pregnancy rate was observed in control animals (76 % compared with approximately 90 % in the treated groups). No animals died during the study. Although bodyweight gain was statistically significantly reduced during the dosing period in animals of the 21 mg kg-1 d-1 group, there was no difference between the groups over the 20-day gestation period as a whole. Signs of toxicity were confined to the 21 mg kg-1 d-1 group, the most common observation being tremors within the first hour of dosing. No significant differences between the groups were reported in relation to numbers of corpora lutea, implantation sites, pre- or post-implantation loss, resorptions, litter size or sex ratio. No treatment-related changes in the incidence of external, visceral, skeletal or total fetal malformations were observed (Unpublished, 1991d). Overall, dichlorvos did not adversely affect the developing rat fetus up to maternally toxic dose level of 21 mg kg-1 d-1. An overall NOAEL of 3 mg kg-1 d-1 is apparent for maternal toxicity. A study reported by Desi & Nagymajtenyi (1999) was designed to investigate the changes of spontaneous and evoked electrical processes of the central and peripheral nervous systems following exposure to moderate doses of dichlorvos following subchronic exposure (up to 12 weeks), over 3 consecutive generations, or during different stages of development. In the final experiment, rats were administered gavage doses of either 0, 0.98, 1.96 or 3.92 mg kg-1 d-1 dichlorvos (purity 98 %) in distilled water, according to one of 3 dosing regimes. Either to pregnant rats from the 5th to the 15th day of pregnancy (group A); to pregnant rats from the

104

5th to the 15th day of pregnancy and then through lactation from the 2nd day of delivery until weaning at 4 weeks (group B); or as for group B but weaned F1 males were treated for a further 8 weeks (group C). The neurotoxicological investigations (described in Section 3.2.4.1.1) were made on anaesthetised F1 males (10 per group) at 11-12 weeks of age (this part of the study appears to have also been reported by Desi et al., 1998). No histopathology was performed. No treatment-related signs of toxicity, effects on bodyweight gain or organ weight were reported. Also parameters of fertility were unaffected by treatment. The mean frequency of somatosensory, auditory or visual ECoGs increased in a dose-dependent manner in group C animals. Somatosensory differences were statistically significant compared with controls in animals of the 1.96 and 3.92 mg kg-1 groups; visual and auditory differences only in top dose animals. Group C animals were also reported to show a dose-dependent increase in latency time of evoked potentials which was statistically significant compared with controls in animals of the 1.96 and 3.92 mg kg-1 groups (no further details given). Conduction velocity and refractory period of the tail nerve were dose dependently increased, but only reached statistical significance in the 1.96 and 3.92 mg kg-1 groups. Although brain cholinesterase activity was reported to be inhibited in the treated groups, the inhibition did not reach statistical significance. In the absence of histopathological data, no meaningful conclusions about the effect of dichlorvos on rat developmental can be drawn. A NOAEL for maternal toxicity cannot be derived as no measurement of erythrocyte acetylcholinesterase activity was made, though it is noted that electrical processes were not affected by treatment up to 0.98 mg kg-1 d-1. A poorly reported study by Desi et al., (1998) was designed to investigate the neurotoxicity to offspring of 3 pesticides, including dichlorvos, following exposure to rats during pregnancy, lactation and the early stages of life, by measuring electrical functions of the CNS and PNS both macroscopically and microscopically. Rats (Specific pathogen free outbred Wistar; numbers not given) were administered gavage doses of 0, 0.98, 1.96 or 3.92 mg kg-1 d-1 dichlorvos (based on a reported LD50 value of 98 mg kg-1; purity 97-98%) in saline according to one of 3 dosing regimes. Either to pregnant rats from the 5th to the 15th day of pregnancy (group A); to pregnant rats from the 5th to the 15th day of pregnancy and then through lactation from the 2nd day of delivery until weaning at 4 weeks (group B); or as for group B but weaned F1 males were treated for a further 8 weeks (group C). The neurotoxicological investigations were made on anaesthetised F1 males (10 per group) at 11-12 weeks of age and included measurement of electrocorticograms (ECoG) and evoked potentials, in response to external stimuli, through electrodes placed on the exposed primary somatosensory, visual and auditory centres. Somatosensory stimulations were made via electrodes inserted into the nasal part of the skin. Conduction velocity of the tail nerve was also measured. Single nerve activity (following stimulation) was investigated through a glass microelectrode inserted in the somatosensory area and measurements were made both before and after administration of an ip dose of 19.6 mg kg-1 dichlorvos. Data was recorded at 5 minute intervals following exposure, and evaluation of activity was based on measurement of the interspike interval (ISI; the time between two successive single unit discharges). Hippocampal population spikes were also measured. No histopathology was performed.

105

Only very limited information was reported on the general health of the rats, namely, there no differences between control and treated groups in terms of salivation, muscle tone and bodyweight gain. Although the paper reports that statistically significantly changes in ECoG indices and increased latency times of the evoked potentials following dichlorvos exposure in group C animals, no data for dichlorvos is presented (only a representative graph of results obtained with one of the other pesticides). Tail nerve conduction velocities were reported to be statistically significantly reduced in group C animals following dichlorvos exposure, but again no data is reported for dichlorvos. However, the paper does show that absolute refractory times of the rat tail nerve increases in a dose-related manner following dichlorvos exposure. Following ip administration of dichlorvos, a slight shortening of the mean ISI was followed by a statistically significant lengthening of the mean ISI (corresponding to a decreased firing frequency). Hippocampal population spikes were increased in amplitude following exposure. Overall, the poor reporting of the results of this study and the lack of histopathology make it difficult to draw any firm conclusions apart from continuous exposure to dams and offspring, to these levels of dichlorvos, can apparently effect the bioelectric measurements made on nerves. However, if exposure is ceased no bioelectric changes are observed. A NOAEL could not be derived from this study. A study by Schulz et al., (1995) investigated the effect of dichlorvos exposure on behaviour variables in rats. Groups of pregnant rats (Wistar; 10 per group) received gavage doses of either 0, 0.97, 1.46, 1.94 or 3.88 mg kg-1 d-1 dichlorvos (based on a reported LD50 value of 97 mg kg-1; purity 98%) in distilled water from day 1 of gestation. Following birth 5 males per litter were selected for further study. Dichlorvos treatment of the dams continued throughout the lactation period, and following weaning at 6 weeks of age the male rats were treated with the same dose levels of dichlorvos until 12 weeks of age. For the behavioural investigation, one male from the litter of every treated animal was assigned to an 'isotoxic investigation group'. The remaining animals from each litter were assigned to four other non-behavioural investigation groups (the results of which are not reported in this paper). Thus, each group at each dose level contains 10 males. T-maze training was started at the age of 9 weeks. Rats had to perform daily for 3 consecutive weeks (apart from weekends), and measurements were made approximately 1 hour before dosing. Performance measures (recorded blind) included the running time in the maze; and the number of incorrect choices (defined as the animal moving in a false direction) to find and to enter the goal box. At 11 weeks of age measures of open field behaviour were made on 3 consecutive days after animals finished T-maze testing. These measures, which were scored blind, included horizontal and vertical exploration activity, grooming activity, defecation rate, and delays in starting the open field behaviours. Following the last open field test on day 3, novelty-induced grooming was assessed by placing the animal in a novel environment (a small empty perspex box of limited size) for 40 minutes. Grooming activity was classified as snout washing, head washing, ventral body lick, side body fur, forepaw licking, hindpaw licking, scratching by the hindpaws and ano-genital grooming. Non-grooming components were scored as horizontal and vertical exploratory activity, inactivity (awake) and inactivity (sleeping). Acetylcholinesterase activity of whole brain and serum samples was measured at the end of the study.

106

The authors report that the treated dams showed no treatment-related changes in 'general conditions including state of delivery or nursing, gestation period, birth rate or at necropsy' (though no data is given). The treated F1 males were reported to show no typical signs of clinical intoxication and no treatment-related changes in bodyweights were observed. T-maze running time was statistically significantly increased in a dose dependent only during week 1 of testing (12 s for controls - 27 s for top dose animals); during weeks 2 and 3 of testing times were not affected by treatment. A statistically significant increase in the number of incorrect choices made by animals was observed in weeks 1 and 2 of testing, but no difference was apparent during week 3. Horizontal exploratory activity was statistically significantly increased in a dose dependent manner on days 1 and 2 of testing, but this was not apparent on day 3. Vertical exploratory activity and defecation rates were statistically significantly decreased in a dose dependent manner on all 3 days of testing. No treatment- related effect on grooming activity or starting open field behaviours were reported. No treatment-related effect was observed in novelty-induced grooming behaviour. Only the number of sleeping scores showed a dose-dependent decrease, which reached statistical significance in the top dose group. Significant decreases, compared with controls, were observed in cholinesterase activity in one hemisphere of the brain (65, 55, 41 and 40 % of control values at 0.97, 1.46, 1.94 or 3.88 mg kg-1 d-1, respectively) and in serum (73, 58, 46 and 41 % of control values, respectively). No histopathology was performed on the treated F1 animals. Overall, this study showed that exposure to these levels of dichlorvos may have subtle effects on learning behaviour of offspring. The rats exposed were not reported to show any of the normal signs associated with acetylcholinesterase inhibition. However, acetylcholinesterase levels were clearly decreased at the end of the study. This may be due to the rats being tested pre-dosing, as the effects are most apparent approximately 15 minutes post-dosing. A NOAEL could not be derived from the study given the inhibition of brain acetylcholinesterase activity at 0.97 mg kg-1 d-1 (> 20 % of control values). A NOAEL for developmental toxicity could not be derived from the study as no histopathology was performed on the offspring. The following studies were reported previously in the 1994 HSE review, but were considered inadequate to derive any conclusions about the potential of dichlorvos to induce developmental toxicity in rats. They are included here as additional information. Pregnant CFE rats (9-16 per dose) were exposed to atmospheres of 0, 0.25, 1.25 and 6.25 mg m-3 dichlorvos (> 97 % pure) for 23 h d-1, from days 1-20 of gestation and killed on day 20 (Unpublished, 1971c; Thorpe et al., 1972). No deaths occurred but lethargy was observed in some rats at 6.25 mg m-3. Maternal bodyweights were not presented. Erythrocyte, plasma and brain cholinesterase activities were depressed at 1.25 and 6.25 mg m-3. Brain acetylcholinesterase activity was 72 and 17 % of control respectively. There were no exposure related changes in the pregnancy rate, mean number of resorptions per litter, late fetal deaths, litter size or fetal weights. At 0.25 mg m-3 one fetus displayed severe skeletal abnormalities. No visceral abnormalities were seen. This study is inadequate to assess the developmental potential of dichlorvos because the number of dams bearing live fetuses was low (8 in all exposed groups and 16 in the control group). In a briefly reported study pregnant Sherman rats (4 dosed and 6 control animals) received 0 or 15 mg kg-1 (highest non-fatal dose) ip on day 11 of gestation only (Kimbrough & Gaines,

107

1968). No signs of toxicity were reported. The only reported abnormality was an increased incidence of omphaloceles (3/41 cases in treated pups, no cases in an unknown number of control pups). The authors concluded that dichlorvos was slightly teratogenic. However the study is of limited value because of the low numbers used, the limited dosing regime and it is not known whether abnormalities were all in one litter. A group of animals (10 - 14), which received dichlorvos, was examined in a poorly reported study, which primarily investigated effects of additional stress on fetal development (Baksi, 1978). Charles River female rats were caged with fertile males (ratio not given) until mating took place. On days 8 - 15 of gestation one half received 25 mg kg-1 d-1 of dichlorvos (purity unknown) by gavage, the other half receiving the vehicle. Dichlorvos had no significant effect on the number of implantations or litter size in any treatment groups. A slight decrease in fetal weight (by up to 20 %), not accompanied by a change in litter size, was seen between dichlorvos and vehicle control litters. The authors claimed no fetal abnormalities although the results were not given. Three studies are available that examined cholinesterase activity, and other parameters, in pups from rats exposed to dichlorvos (up to 5.6 mg kg-1 d-1) for the last 7 d of gestation (Zalewska et al., 1977a; 1977b; Dambska et al., 1978). The results indicated that brain acetylcholinesterase activity remained depressed in pups from dams administered 5.6 mg kg-1 d-1 up to day 56. However no differences in weight gain or mobility development were observed.

3.2.8.4.2 Rabbit

In a study performed to GLP and consistent with current EC guidelines, female rabbits (New Zealand White; 16 per group) were artificially inseminated and administered gavage doses of 0, 0.1, 2.5 or 7.0 mg kg-1 d-1 dichlorvos (purity 96.9 %) in water on days 7 - 19 of gestation. Animals were sacrificed on gestation day 30 and examined to confirm pregnancy. The bodyweight and weights of the liver and uterus were recorded; ovarian corpora lutea were counted; and uterine contents recorded. All live fetuses were weighed and examined for abnormalities. At sacrifice 2 animals from each of the 0 and 0.1 mg kg-1 d-1 groups and 3 animals from each of the 2.5 and 7.0 mg kg-1 d-1 groups were found to be non-pregnant. During the dosing period, 4/13 pregnant females of the 7.0 mg kg-1 d-1 group died (1 on gestation day 17 and 3 on gestation day 19); and 2/13 pregnant females of the 2.5 mg kg-1 d-1 group died (gestation days 12 and 15, respectively). Two animals of the 0.1 mg kg-1 d-1 group were removed from the study due to early delivery. Bodyweight gain was reduced during the dosing period in animals of the 2.5 and 7.0 mg kg-1 d-1 groups (33 and 42 % of control values), however, there was no difference between the groups over the 30-day gestation period as a whole. Treatment-related signs of toxicity in surviving animals were observed in animals of the 7.0 mg kg-1 d-1 group only and included tremors, ataxia, lurching, prone position, cyanosis, salivation, diarrhoea, breathing difficulties and excitation. Necropsy of the dams did not reveal any treatment-related changes.

108

No significant differences between the groups were reported in relation to numbers of corpora lutea, implantation sites, pre- or post-implantation loss, litter size, resorptions, litter size or sex ratio. No treatment-related changes in the incidence of external, visceral, skeletal or total fetal malformations were observed (Unpublished, 1991e). Overall, this study indicates that dichlorvos did not adversely effect the developing rabbit fetus up to severely maternally toxic doses. An overall NOAEL of 0.1 mg kg-1 d-1 is apparent for maternal toxicity (based on deaths and reduced bodyweight gain at 2.5 mg kg-1 d-1), while no developmental toxicity was observed at the top dose of 7.0 mg kg-1 d-1. The following studies were reported previously in the 1994 HSE review, but were considered inadequate to derive any conclusions about the potential of dichlorvos to induce developmental toxicity in rabbits or mice. Dutch rabbits (20 per dose) were exposed to atmospheres containing dichlorvos vapour (97 % pure) at 0, 0.25, 1.25 or 6.25 mg m-3 for 23 h d-1 from days 1 - 28 of gestation and killed on day 28 (Unpublished, 1971c). At 6.25 mg m-3 16/20 females died or were killed in extremis. Symptoms included anorexia, lethargy, tremors, nasal discharge and diarrhoea. No deaths or clinical signs of toxicity were seen at 0.25 or 1.25 mg m-3. Erythrocyte, plasma and brain cholinesterase activities, determined at the termination of the study, were reduced (brain only at 0.25 mg m-3 to 90 % of control, all cholinesterases at 1.25 and 6.25 mg m-3, to between 15 - 50 % of control). Neither individual animal nor litter data were presented in the report. No significant changes in the pregnancy rates, the mean numbers of resorptions/litter, mean numbers of late fetal deaths, mean litter size and mean fetal weight were seen. In a separate study, Dutch rabbits (20 per dose) were exposed to atmospheres containing dichlorvos vapour (97 % pure) at 0, 2 or 4 mg m-3 for 23 h d-1 from days 1 - 28 of gestation and killed on day 28 (Unpublished, 1971c). Mortalities were observed in both exposed groups, 6/20 at 4 mg m-3 (all following accidental exposure to 6.6 mg m-3) and 1/20 at 2 mg m-3. Clinical signs of toxicity were not reported. There were no significant changes in; pregnancy rates, mean numbers of resorptions/litter, late fetal deaths, litter size or litter weight. Although individual data were not presented in either of the studies, no skeletal or visceral malformations were seen. Pregnant NZW rabbits (8 control and 12 exposed dams) were given either 0 or 5 mg kg-1 d-1 dichlorvos (96 % pure) by gavage (Schwetz et al., 1979). Another group (19 exposed and 14 control) was exposed to 0 or 4.1 mg m-3 dichlorvos vapour (96 % pure) for 7 h d-1. Both treatments lasted from days 6 - 18 and the animals were killed on day 29 of gestation. No clinical signs of toxicity or change in maternal weight were observed. There was an increased incidence in the number of resorptions/dam (from 0.5 ± 0.5 to 1.8 ± 2.8) but no increase in the number of litters involved. No significant differences in the mean number of implantations, live fetuses, sex ratio, fetal weight and fetal crown rump length were seen in either study. No malformations were observed.

109

Two teratology studies conducted in accordance with 1966 FDA Guidelines, are available (Unpublished, 1969e). NZW rabbits (15 – 26 per group) were artificially inseminated and received either 0, 12, 36 or 62 mg kg-1 d-1 dichlorvos in PVC capsules on days 6 - 18 of gestation and killed on day 29. At 62 mg kg-1 d-1 6 dams died after 6 doses. This group was therefore sub-divided into dams dosed from days 6 - 8 and days 9 - 11. Deaths, not dose related, were observed in all groups (13 - 60 %). No dose related change in the pregnancy rate was seen. However the number of dams bearing live fetuses at day 29 was small (between 3 – 9 per group). There were no obvious changes in the incidence of skeletal abnormalities and no gross malformations were observed. This study is of limited value because of the small number of dams with live fetuses. In another study, rabbits (14 – 15 per group) were artificially inseminated and received 0, 3, 12, 36 or 60 mg kg-1 d-1 of dichlorvos in PVC capsules from days 6 - 16 of gestation and killed on day 29 (Unpublished, 1969e). Ovaries and fetuses were examined macroscopically and fetuses for skeletal abnormalities. At 36 and 60 mg kg-1 d-1 all dams died. No clinical signs of toxicity or mortalities were reported at 3 or 12 mg kg-1 d-1. There were no dose related effects on pregnancy rate. However the number of dams bearing live pups on day 29 was 11, 9 or 5 (gelatin capsule, water and PVC controls respectively) and 12 and 8 (at 3 and 12 mg kg-1 d-1). At 12 mg kg-1 d-1 and in the PVC controls a slight reduction in the mean number of live pups/dam was seen. No skeletal abnormalities were observed. Studies are available that examined cholinesterase activity and other parameters in pups from rabbits exposed to up to 6 mg kg-1 by gavage for the last 10 d of gestation (Dambska et al., 1979; Maslinska & Zalewska, 1978a; 1978b; Dambska et al., 1978). The results indicate brain acetylcholinesterase activity in pups from dams administered 6 mg kg-1 d-1 was significantly reduced from days 1 - 16. No differences were observed in weight gain of pups (on days 0, 8 or 16) or mobility development. Electron microscopy showed certain areas of immaturity in the second layer of the motor cortex that were more evident in treated animals. Glial cells were young, with large nuclei, and often irregular in shape with scanty cytoplasm. Astrocytic processes were scarce. In the control group, an average of 85 % of the synapses examined were immature compared to 90 % in the treated group. The significance of these findings is unclear. Studies investigating effects of postnatal administration (up to 8 mg kg-1 d-1 on days 5 - 16 of life) on CNS development in rabbits are available (Dambska & Maslinska, 1988; Dambska & Maslinska, 1982; Dambska et al., 1984; Maslinska et al., 1984; Dambska et al., 1981). Clinical signs of toxicity, and reductions in brain acetylcholinesterase activity, consistent with OP poisoning were observed. These studies indicated that there were changes in neurones in the cerebral and cerebella cortex, delayed myelination and minor changes in endothelial cells from brain vessels. The significance of these lesions is uncertain.

3.2.8.4.3 Mice

110

A poorly reported study, in abstract form only (also reporting positive results in in vivo cytogenetic and micronucleus tests) is available (Majeeth et al., 1989). Pregnant Swiss albino mice (unknown number) received ip doses of 5 mg kg-1 d-1 dichlorvos (purity unknown) from day 6 to 14 of gestation. There was an increase in the number of dead fetuses and resorptions in treated mice (6.1 % in control group and 21.7 % in the treated group). The frequency of gross malformations and skeletal abnormalities (not further specified) was reported as being increased to 9.2 % but control levels were not given. No conclusions could be drawn because insufficient details were presented. Pregnant CF1 mice (25 treated dams and 28 controls) received 0 or 60 mg kg-1 dichlorvos by gavage in corn oil (Schwetz et al., 1979). Another group (15 exposed dams, 20 control) was exposed to 0 or 4.1 mg m-3 dichlorvos (96 % pure) for 7 h d-1. Both treatments were from days 6 to 15 of gestation. Dams were killed on day 18 of gestation. Clinical signs of toxicity were not seen, but the body weight of dams (oral study only) was significantly reduced on day 16 only. No differences in the numbers of implantations, live fetuses, resorptions/dam, fetal weight, sex ratio or fetal crown rump length were seen. No dose related change in the number of visceral or skeletal abnormalities was seen. Dichlorvos was not teratogenic in this study. However, only the oral component is considered to be an adequate assessment of the teratogenic potential of dichlorvos as no toxicity was seen in the treated animals in the inhalation component.

3.2.8.4.4 Guinea Pig

A study by Mehl et al. (1994) was designed to investigate the effect of short term exposure to a number of compounds, including dichlorvos, on the brain development in guinea pigs. On days 42, 43 and 44 of gestation groups of guinea pigs were administered 15 mg kg-1 d-1 dichlorvos (purity 99 %; 3 animals) or 2 daily doses of 15 mg kg-1 separated by 12 h (4 animals); a further group were administered 2 daily doses of 15 mg kg-1 on days 44, 45 and 46 of gestation (4 animals). The route of administration is not stated in the report, but is assumed to be oral. Once delivered fetal brains were examined. It is reported that in animals administered 2 daily doses of dichlorvos fetal brain weight was statistically significantly decreased (approximately 87 % of control values), however, this was not associated with any changes in the brain enzymes choline acetyltransferase, glutamate decarboxylase and acetyl cholinesterase. Consequently given also the small numbers of animals in each treated group it is unlikely that this effect on fetal brain tissue is biologically significant.

3.2.8.5 SUMMARY OF DEVELOPMENTAL TOXICITY

No studies are available in humans. Two well-conducted gavage studies are available in rats and rabbits. Both studies show no developmental toxicity even at levels, which induce a significant level of maternal toxicity. No effects on development were reported at a dose level of 21 mg kg-1 d-1 (the top dose level in the study) in the rat study for developmental effects, while the NOAEL for maternal toxicity was 3 mg kg-1 d-1 based on tremors and significantly reduced bodyweight gain reported in top dose animals (consistent with dichlorvos acting as an acetylcholinesterase inhibitor). No effects on development were reported at a top dose level of 7 mg kg-1 d-1 in a

111

rabbit study, while the NOAEL for maternal toxicity was 0.1 mg kg-1 d-1, based on deaths and significantly reduced bodyweight gain at 2.5 mg kg-1 d-1 and above. Few conclusions can be drawn from other less-robust studies in different species and by different routes of exposure suggesting that dichlorvos is not a developmental toxicant.

3.2.8.6 SUMMARY OF REPRODUCTIVE TOXICITY

No data in humans are available. A well-conducted 3-generation rat drinking water study is available. Although the standard parameters of fertility remained unaffected by treatment, additional analyses carried out during the study indicated that female animals of the 80 ppm group (7 - 17 mg kg-1 d-1) developed treatment-related abnormalities in oestrous cyclicity. Thus, a NOAEL for effects on fertility of 20 ppm (1.9 - 4.6 mg kg-1 d-1) was apparent. The overall NOAEL for adult toxicity was 5 ppm (0.5 - 1.2 mg kg-1 d-1) based on significant (> 20 %) decreases in acetylcholinesterase activity (brain and erythrocyte). This finding is consistent with the results of a study, which primarily investigated electrical process changes in 3 generations of rats following dichlorvos exposure, in which no effects on fertility were observed at the top dose used of 3.92 mg kg-1 d-1. A NOAEL for adult toxicity could not be established as erythrocyte acetylcholinesterase activity was not measured. Few conclusions can be drawn from other less robust studies. Two well-conducted gavage studies are available in rats and rabbits. Both studies show no developmental toxicity even at levels, which induce a significant level of maternal toxicity. No effects on development were reported at a dose level of 21 mg kg-1 d-1 (the top dose level in the study) in the rat study for developmental effects, while the NOAEL for maternal toxicity was 3 mg kg-1 d-1 based on tremors and significantly reduced bodyweight gain reported in top dose animals (consistent with dichlorvos acting as an acetylcholinesterase inhibitor). No effects on development were reported at a top dose level of 7 mg kg-1 d-1 in a rabbit study, while the NOAEL for maternal toxicity was 0.1 mg kg-1 d-1, based on deaths and significantly reduced bodyweight gain at 2.5 mg kg-1 d-1 and above. Few conclusions can be drawn from other less-robust studies in different species and by different routes of exposure suggesting that dichlorvos is not a developmental toxicant. Overall, these results indicate that dichlorvos is not a reproductive toxicant.

3.3 DATA REQUIREMENTS

The following data should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months.

i. A developmental neurotoxicity study.

112

4 OPERATOR AND CONSUMER EXPOSURE AND RISK ASSESSMENTS

4.1 NO ADVERSE EFFECT LEVELS FOR USE IN RISK ASSESSMENTS

4.1.1 NOAELS IDENTIFIED IN THE MAMMALIAN TOXICITY ASSESSMENT

Table 4.1 NOAELS for dichlorvos identified in the mammalian toxicity assessment

Study Route Species NOAEL mg kg-1 d-1 (NOAEC mg m-3 d-1)

Basis for NOAEL/NOAEC

Acute neurotoxicity

oral/gavage rat 0.5 mg kg-1 Changes in FOB and motor activity at 35 mg kg-1. ChE levels not measured.

13 week oral/gavage rat 0.1 Erythrocyte AChE inhibition at 1.5 mg kg-1 d-1.

91 day oral/gavage rat 0.1 Salivation, tremors and exophthalmus at 7.5 mg kg-1 d-1.

90 day oral/ dietary dog 0.32 Behavioural effects and brain AChE inhibition at 0.96 mg kg-1 d-1.

52 week oral/dietary dog 0.05 Erythrocyte and brain AChE inhibition at 1.0 mg kg-1 d-1.

2 year oral/dietary rat 0.6 Brain AChE inhibition at 6 mg kg-1 d-1. 2 year oral/dietary dog 0.008 Erythrocyte AChE inhibition at

0.09 mg kg-1 d-1. 5 day (23 h d-1)

inhalation (whole body)

mouse 0.211 (0.14)

Brain AChE inhibition at 1.2 mg m-3 d-1 (1.8 mg kg-1 d-1).

120 day (4 h d-1)

inhalation (whole body)

rat 0.0062 (0.11)

Erythrocyte and brain AChE inhibition at 1.07 mg m-3 d-1 (0.06 mg kg-1 d-1).

2 year (23 h d-1)

inhalation (whole body)

rat 0.0162 (0.05)

Erythrocyte AChE inhibition at 0.5 mg m-3 d-1 (0.16 mg kg-1 d-1).

5 to 12 weeks 3 generation

oral/gavage rat 0.98 Bioelectric changes in CNS/PNS activity at 1.96 mg kg-1 d-1.

3 generation fertility

oral/drinking water

rat 0.5-1.2

1.9-4.6

Erythrocyte and brain AChE inhibition at 1.9 - 4.6 mg kg-1 d-1. Abnormalities in oestrous cyclicity at 7.2 - 17.5 mg kg-1 d-1.

developmental oral/gavage rat 3

>21

Reduced body weight gain, tremors at 21 mg kg-1 d-1 . No developmental effects at top dose

developmental oral/gavage rabbit 0.1

>7

Reduced body weight gain, death at 2.5 mg kg-1 d-1. No developmental effects at top dose

1conversion to mean systemic dose based on an assumed respiratory volume of 1.8 l h-1 & mean body weight 0.03 kg (male), 0.025 kg (female) 2conversion to mean systemic dose based on an assumed respiratory volume of 6 l h-1 & mean body weight 0.5 kg (male), 0.35 kg (female)

113

Reliable data in humans are available following acute and repeat oral exposure although it was not possible to identify clear NOAELs from these studies. Less robust data are available following inhalation exposure. These studies, evaluated in Section 3, are summarised in Tables 4.2 to 4.4.

Table 4.2 Key studies involving acute human volunteer exposure to dichlorvos

No. of volunteers

Dose Effects observed Comment Ethical consent

Reference (see Section 3)

4 - 6 0.5 mg kg-1 oral No toxicity and no effect on erythrocyte AChE activity.

Yes Unpublished, 1997a

6 1 mg kg-1 oral No treatment-related signs of toxicity. Erythrocyte AChE activity showed a slight reduction with time (up to 14 days post dosing) but this did not reach biological significance.

Yes Unpublished, 1997b

4-19/group treated 44 controls

0-32 mg kg-1 oral

At 24 h post-administration the range of values reported for erythrocyte AChE activity indicated that inhibition was > 20 % at each dose level and also in a concurrent control group (40 - 100 % of pre-treatment values). This range in controls is greater than that observed in the majority of the treated groups, including the lowest dose group (0.1 - 1 mg kg-1), where inhibition was reported between 50 - 100 % of pre-treatment values.

The wide variation in the concurrent control activities compared with the pre-treatment control activities calls into question the reliability of the study. Given that a decrease of > 20 % in erythrocyte AChE activity is considered a biologically significant effect; a NOAEL cannot be derived from this study.

Yes Slomka & Hine, 1981

6 7-52 mg m-3 (vapour) for 20-240 min. Head only

No treatment-related clinical symptoms reported. Plasma ChE was inhibited after all exposures. Erythrocyte AChE activity was only biologically significantly depressed following exposure to 22.6 mg m-3 for 105 min.

No dose-response relationships were apparent, and it is strange that exposure to higher dose levels, particularly 44.7 and 52 mg m-3 for 90 and 65 min, respectively, produced no change in erythrocyte AChE activity. This questions the reliability of the measurements made during the study and so no meaningful conclusions can be drawn from this data.

No Unpublished, 1969d; Unpublished, 1970c

31 0.9-1.22 mg m-3

(vapour) for 2-7.5 h Whole body

Plasma ChE levels were depressed following exposure for 6.5 h or more. There was no effect on erythrocyte AChE activity,

Given that this study was performed by the same laboratory as the previous one, concerns over the reliability of the methodology exist and therefore no meaningful conclusions can be drawn from the data.

No Unpublished, 1970d

114

Table 4.3 Key studies involving repeated human volunteer exposure to dichlorvos

No. of volunteers

Dose Effects observed Comment Ethical consent

Reference (see Section

3) 6 treated 3 controls

0.1 mg kg-1 d-1 for 21 d oral

No treatment-related toxicity was reported. Erythrocyte AChE activity was reduced in a time dependent manner following dichlorvos exposure, falling to approximately 77 - 92 % of pre-dose activity, at the end of dosing, with two volunteers showing a decrease in erythrocyte AChE activity of > 20% at the end of the dosing period. Also, it is noted that a further volunteer showed a biologically significant reduction in erythrocyte AChE activity 9 days after the final dose. Control values remained in the range 97 - 104 % of pre-dose values, and were statistically significantly greater than the dichlorvos group values from day 7 onwards.

The study shows that oral exposure of humans to approximately 0.1 mg kg-1 d-1 dichlorvos for 21 days leads to significantly decreased levels of erythrocyte AChE activity (> 20 % of pre-dosing values) in 2 of 6 volunteers.

Yes Unpublished, 1997c

6 0.5 mg kg-1 d-1 for 12 - 15 d oral

No treatment-related signs of toxicity were reported. Mean erythrocyte AChE activity decreased during dosing to 73 % of pre-dose values 15 days after the first dose. Subsequently activity reached a minimum of 69 % of control values 22 days after the first dose, before recovering to 91 % of control values 54 days after the first dose.

The study shows that oral exposure of humans to 0.3 mg kg-1 d-1 dichlorvos for 12 - 15 days leads to decreased levels of erythrocyte AChE activity (> 20 % of pre-dosing values), which recover once exposure ceases.

Yes Unpublished, 1997a

4 treated 2 controls

0, 1, 2 or 4 mg kg-1 d-1 for 7d 8, 16 or 32 mg kg-1 d-1 for 2 -3 d oral

Dose-related decrease in plasma ChE activity, to 10 - 20 % of pre-treatment values at 4 mg kg-1 d-1, was seen with similar reductions in volunteers given higher doses for shorter periods. The range of values reported for erythrocyte AChE activity indicated that inhibition was > 20 % of pre-treatment values at each dose level. The range in controls was 90 - 100 % of pre-treatment values.

Yes Slomka & Hine, 1981

115

No. of volunteers

Dose Effects observed Comment Ethical consent

Reference (see Section

3) 5 treated 2 controls

0, 1, 1.5, 2, 2.5 mg d-1 for 28 d

Below 2 mg d-1 no effect on plasma or erythrocyte (A)ChE activity was seen. At 2 mg d-1 plasma ChE activity was inhibited from day 7 onwards, falling to 71 % of pre-exposure levels on day 28. Dosing at 2.5 mg d-1 ceased when plasma ChE activity fell to 70 % of pre-exposure on day 20. However activity had returned to pre-exposure levels 15 d later. There was no effect on erythrocyte AChE activity.

It is noted that this study was performed in healthy male prisoners without indication of ethical consent.

No Unpublished, 1967a

10 treated 2 controls

0 or 1.5 mg d-1

for 28 d oral

Plasma ChE activity was 76 - 87 % of control pre-exposure levels from days 16 to 60, but had recovered within 14 d of exposure ceasing. No effect on erythrocyte AChE was observed.

This study was reported in the same group of prisoners described above. Assuming a mean bodyweight of 60 kg no effect on erythrocyte AChE activity was observed at 0.025 mg kg-1 d-1.

No Unpublished, 1967a

6/group 0 or 2.7 mg d-1 for 21 d oral

Only plasma ChE was inhibited to a significant, activity showing a time related inhibition, reaching a minimum activity of 70 % of pre-treatment levels when dichlorvos was given with meals or 60 % when given as pre-meal capsules.

No data was presented to indicate the extent of the erythrocyte AChE inhibition and therefore no meaningful conclusions can be drawn from the data.

No Boyer et al., 1977

3 0.9-1.22 mg m-3 (vapour) for 2 - 7.5 h for 4 d whole body

No change in plasma ChE activity was seen in 2 volunteers. In the third, activity gradually declined to 63 % of pre-exposure values by day 4. Exposure had no effect on erythrocyte AChE activity.�

No Unpublished, 1970d

116

Table 4.4 The effect of dichlorvos exposure in consumer studies

Subjects Exposure Effects observed Comment Reference (see Section 3)

Exposed 7 adults; 11 children Controls 4 adults; 2 children

Via strips: 0.09- 0.16 mg m-3

for 42 d

No significant difference was observed in plasma or erythrocyte AChE activities.

No firm conclusions can be drawn from this study as it is unclear what the individual exposures were and over what length of time.�

Leary et al., 1974

5 exposed homes 2 control homes

Strips containing 0 or 18.5 % dichlorvos (replaced every 3 months) for 12 months

No treatment-related findings No attempt was made to measure air concentrations and so no meaningful conclusion can be drawn.

Leary et al., 1974

12 exposed homes 4 control homes

Strips as above replaced every month for 6 months.

From month 2 onwards occupants of treated houses had a slight but significant depression of plasma ChE activity (to 40 - 85 % of control levels). Erythrocyte AChE was not depressed in a treatment-related fashion.

No attempt was made to measure air concentrations and so no meaningful conclusion can be drawn.

Leary et al., 1974

14 exposed homes

Slow release strips No changes in either blood ChE or EMG were observed when strips were used according to manufacturer's instructions.

No information about dichlorvos concentration in air was reported and so no meaningful conclusion can be drawn.

Ottevanger, 1975

4 adults, 2 per home

Slow release strips 1 per 1000 ft3 (equiv 28.3 m3) replaced monthly for 4 months, then once in 2 months

All subjects showed biologically significant decrease in erythrocyte AChE at least once during study.

No information given on exposure periods and no personal sampling, so no information could be derived on the true exposures of each individual.

Zavon & Kindel, 1966

6 families: 14 subjects 2 controls

Slow release strips 7 subjects and 1 control showed biologically significant decrease in erythrocyte AChE.

No information given on exposure periods or air levels so no information could be derived on the true exposures of each individual. Reliability of assay method questionable as effects seen in 1 control.

Zavon & Kindel, 1966

117

4.1.2 IDENTIFICATION OF NOAELS FOR USE IN RISK ASSESSMENTS

Risk assessments are presented in this document for several exposure scenarios, which encompass risks to both users of the pesticide products and to consumers who may come into contact with treated areas. Products containing dichlorvos are approved for professional and amateur users. Many of these products will only be used during the warmer periods of the year (May - September) while others, for example, those for professional use in public hygiene applications, may be used throughout the year. Current approvals exist for aerosol space sprays, aerosol space and surface sprays, slow release strips, slow release controllable cassettes and slow release non-controllable cassettes. The use of three products, Product A, Product B and Product C involves operators in handling dichlorvos-impregnated strips directly while placing them in situ, The first product is used in museum display cases, cabinets and small cupboards and the latter two products are used in pheromone traps. The NOAEL values used in each risk assessment scenario are discussed below.

4.1.2.1 PRIMARY EXPOSURE TO AEROSOL PRODUCTS

A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for insecticide application (primary exposure) from an aerosol and slow release strips. This study was of an acceptable quality, was performed to GLP and was consistent with EC guidelines. This NOAEL is considered appropriate for the assessment of risk to both professionals who may use these products repeatedly throughout the year, and for amateur users who may only use aerosols during the warmer periods of the year. It is also considered appropriate for the assessment of risk to professionals who may use slow release strips repeatedly throughout the year.

4.1.2.2 SECONDARY ACUTE EXPOSURE TO AEROSOL AND SLOW RELEASE PRODUCTS

It is considered that these exposure scenarios would be single, rare events and therefore, a NOAEL taken from an acute or short-term study is considered appropriate for the calculation of a toxicity-exposure ratio (TER). The NOAEL, therefore, is taken from the shortest term, acceptable study where cholinesterase inhibition has been measured. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from 2 acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment.

4.1.2.3 CHRONIC SECONDARY EXPOSURE FROM SLOW RELEASE PRODUCTS

These exposure scenarios would involve chronic exposure to dichlorvos vapour over the summer season and possibly throughout the year. A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for residential use (secondary exposure) from a slow release product. This value is consistent with a notional NOAEL of 0.016 mg kg-1 d-1 derived from the NOAEC of 0.05 mg m-3, taken

118

from the 2-year inhalation study in the rat (whole body exposure for 23 h d-1). It is also considered appropriate for professional users of slow release strips in pheromone traps, who may use these products repeatedly throughout the year.

4.1.3 DERMAL ABSORPTION VALUES USED IN THE RISK ASSESSMENT

No information is available on dermal absorption of dichlorvos in humans. A dermal absorption value of 6 - 11 % was identified in a rat dermal absorption study, however, the similarity of the acute oral (LD50 46-105 mg kg-1) and dermal (LD50 210-456 mg kg-1) toxicity for dichlorvos in rats suggests that dermal absorption may be considerably higher than 11%. Alternatively, extensive first pass metabolism following oral exposure, given the rapid metabolism and excretion profile of dichlorvos, may account for the similarity between oral and dermal toxicity. Where appropriate a value of 50 % for dermal absorption has been used in the risk assessments.

4.2 OPERATOR AND CONSUMER EXPOSURE

4.2.1 PRODUCT DATA

Current approvals exist for aerosol space sprays, aerosol surface sprays, aerosol space and surface sprays, slow release strips, slow release controllable cassettes and slow release non-controllable cassettes. Details of the maximum levels of dichlorvos currently permitted are given in Section 1.3. In order for HSE to refine the risk assessments for individual products, approval holders were requested to submit the following information on the patterns of use of their products:

• confirmation of the use sectors of products, • how users apply products and with what application equipment, • how often a contractor would apply the products, • how much in-use solution would be used in a typical application, • how long a typical application would take

If approval holders did not hold these data, estimates were requested. Usage data were also requested from approval holders to enable HSE to estimate the scale of use of individual products and of all dichlorvos usage by the public. This encompassed the total number of units sold in the UK per annum or, where this was unavailable or inappropriate (i.e. where a product is sold in packaging of variable size), total amounts of dichlorvos used in the manufacture of individual products was requested. Approval holders were also asked for information regarding the points of sale (i.e. retail or wholesale). This information is summarised in Section 1.5.

4.2.2 INFORMATION ON PATTERNS OF USE

A request for information was sent to Approval holders. The following table summarises the responses.

119

Table 4.5 Summary of patterns of use of dichlorvos products

Application Type Use Frequency/duration Season Aerosol space spray Space spray, 2 g sec-1 3 x 10 seconds / day Spring,

Summer Aerosol space spray Space spray, 12 g 3 -5 seconds / spray Aerosol surface spray

surface, 2 g sec-1 1 minute maximum Spring, Summer

Slow release strip (Museum use)

Strips containing 14.6 g dichlorvos cut according to volume of display case. Maximum strip: volume ratio 6 g m-3.

Time restriction of 30 mins cutting time per day. Strips are replaced every six months.

Slow release strip (Pheromone Trap)

1 trap (0.49 g dichlorvos) per 100 m2

up to 6 weeks

Slow release controllable cassette

1 unit (up to 20 g dichlorvos) per 20 - 30 m3

up to 4 months

Slow release non- controllable cassette

1 unit (up to 2 g dichlorvos) per 2 m3 cupboard

3 to 6 months Summer

4.2.3 DATA SUBMITTED BY INDUSTRY ON EXPOSURE TO DICHLORVOS

A number of studies relating to exposure were submitted by Approval Holders. These are summarised below. In the studies submitted for this review, the dichlorvos-impregnated strips typically contain 20 % dichlorvos. Very little reference is made to the product description (i.e. slow release strip, slow release controllable or non-controllable cassette) and no data support any distinction between the types of product in terms of the release of dichlorvos into the air to produce efficacious concentrations.

4.2.3.1 'THE MODESTO HOMES' STUDY.

The 'Modesto homes' study aimed to investigate dichlorvos exposure during normal use patterns in the US in response to previously reported work, where use patterns were extreme (Collins & DeVries, 1973). This study of exposure by inhalation and ingestion involved strip devices in controllable release vinyl copolymer lined cardboard packages. The strips, of unspecified size, had a nominal initial concentration of 20 % w/w dichlorvos. Some vapour diffused from the device set in the ‘closed’ position and some appeared to permeate the device material. The ‘open’ position permitted vapour release through holes with an approximate total area of 15 cm2. Three or four devices were placed in each of 15 homes; one device in the kitchen and the others located around the home at the discretion of the home-owner. The indoor air temperatures ranged between 21 and 35 oC. Samples of house (kitchen) air and food were taken for analysis: most of the data within the study related to food residues. The airborne concentrations ranged between <0.01 and 0.11 mg m-3 (40 l air samples) for the 15 homes. The homes were either centrally air conditioned (5 homes), window or wall-air

120

conditioned (5 homes), or had no air conditioner (5 homes). There did not appear to be any obvious impact on dichlorvos concentrations as a result of the use, or otherwise, of air-conditioning. Average values were close to 0.06 mg m-3 at the start of the study but diminished to 0.04 mg m-3 after 7 days and progressively lower to 0.01 mg m-3 after 91 days. The food concentrations showed dichlorvos detected with a frequency of 38 % for air-conditioned homes and 29 % for non air conditioned homes (assuming 'less than' values equal to zero). For non-zero values, the median food concentration was at 0.01 ppm and the highest concentration was at 0.03 ppm. The highest concentration for a food component was in fat (butter on toast) at 0.14 ppm.

4.2.3.2 THE SITTINGBOURNE STUDY

The main purpose of this study was to determine the rate of loss of dichlorvos from an insecticidal strip (Unpublished, 1979). Laboratory studies were undertaken under conditions of ventilation that do not represent typical UK situations in domestic dwellings. However, the experiment was designed to evaluate storage through use of exaggerated conditions. Typically, experiments were carried out with three air changes per hour in the test rooms. HSE consider that typical UK conditions would produce closer to 0.5 - 1 air change per hour, and in some cases ventilation in domestic dwellings is even lower. The second part of the report describes and presents results from field studies carried out in UK homes during 1973 and 1974. In each case just one prototype dichlorvos strip (approx. 20 % dichlorvos) was placed in a selected room in a home and the atmospheric levels monitored over a 119-day period. Rooms were typically ventilated through leaving doors and windows open. Highest concentrations were found in the early stages of the study (up to 0.06 mg m-3) but there is no description of analytical methods and no discussion of air sampling protocols. Therefore, conclusions on the relevance of the air sampling to risk assessment cannot easily be drawn, although the results may be an indicator of typical exposures in UK dwellings where ventilation arrangements are variable.

4.2.3.3 THE MUSEUM STUDY

In this study, measurements were made in two museum rooms with a view to establishing the effect of ventilation on airborne dichlorvos concentrations (Deer et al., 1993). This poorly reported study does not identify the number of dichlorvos strips deployed, their location, or their age. Few samples were taken and it is not clear where they were taken within the rooms. Dichlorvos was measured with ventilation turned on (0.01 mg m-3) and turned off (0.15 mg m-3). The values are persuasive in indicating the impact of ventilation on measured concentrations.

4.2.3.4 PESTICIDE EXPOSURE IN US HOMES

This well conducted study, of its type, measured 28 of the most common pesticides inside and outside a small selection of US homes (Lewis et al., 1998). The study makes no reference to the usage patterns of products or indeed whether they have been used at all. It is not possible to draw any conclusions that influence the exposure assessment of dichlorvos beyond the substance being qualitatively identified in the air of a few houses.

121

4.2.3.5 SIMULATED USE

In this study, airborne concentrations in house rooms containing a dichlorvos strip, and a test chamber containing a strip, were quoted for a ‘typical strip’ at 26.5 oC, 50 % relative humidity and 1 air change per hour (Batth et al., 1973). The house rooms produced a median dichlorvos concentration of 0.04 mg m-3, maximum 0.10 mg m-3 (n=44) and the test chamber showed a median value at 0.09 mg m-3, maximum 0.13 mg m-3 (n=13). The results appear to complement findings from other studies.

4.2.3.6 EUROPEAN AND AUSTRALIAN TRIALS

In a well reported study, trials were carried out to determine the concentrations of dichlorvos that occur in the air of houses when dichlorvos impregnated strips were placed under conditions of normal domestic use (Elgar & Steer, 1972). Ten trials were conducted in the UK, Australia and France in 1969 and 1970. Of 3000 samples 98 % of results of airborne dichlorvos were below 0.1 mg m-3 and 84 % below 0.05 mg m-3 with values ranging from 0.01 mg m-3 to 0.24 mg m-3. The geometric mean of the maximum levels attained was 0.04 mg m-3. In each trial dichlorvos levels rose rapidly over the first day or two and then fell exponentially as dichlorvos became depleted over the 120 days of the trial. Unsurprisingly, highest concentrations were found in poorly ventilated areas and where temperatures were highest. Evidence seems to suggest that deployment of strips in small rooms (less than 30 m3) has no measurable impact on the concentrations attained and that it is ventilation that is the key to the profile of concentrations found over the period of the study. Kitchens tend to be well ventilated and the rate of decline of concentrations is significantly higher than in other rooms. For trials in the United Kingdom, the concentration ranges found were:

Table 4.6 Dichlorvos concentration ranges found in the UK

Time Point Concentration Range(mg m-3)

End of week 1 0.03 to 0.06 End of week 4 0.02 to 0.04 End of week 12 0.01 to 0.02

4.2.3.7 THE WETERINGS STUDY

Aerial concentrations of dichlorvos were measured in a test room (28 m3) after placement of a dichlorvos-impregnated strip, containing 22% dichlorvos (Unpublished, 1997). Ventilation was applied at 3 air changes per hour and room temperature was maintained at 24 oC. Maximum concentrations of 0.08 mg m-3 were attained, slowly reducing to 0.03 mg m-3 over a period of 12 weeks. HSE consider the air change rate is not typical of UK households where values of less than 1 air change per hour may be more usual in some circumstances. However, the paper summarises earlier published data and concludes that dichlorvos concentrations in residences in moderate climatic zones during use of dichlorvos impregnated

122

strips under normal use conditions is lower than the concentration measured in test rooms. All high values have been determined in houses that were closed during the absence of the residents and were not ventilated, while the lowest values were determined in houses with an external door or a large window open.

4.2.3.8 THE 'ARIZONA (TUCSON)' STUDY

This detailed study in 6 closed homes (winter) had dichlorvos impregnated strips placed at a rate 1 per 1000 ft3 (28.3 m3) in four of the homes, and dummy strips in the other two (Leary et al., 1974). Air, meal and drink samples were taken over a 28-d period, with biological monitoring of participants (plasma and erythrocyte cholinesterase, see Section 3.2.4.2.3). After 28 d, new strips were deployed at a higher rate of one per 500 ft3 (14.2 m3) over a 13-day period, with air, food and biological monitoring. There were 97 samples taken in all, and 76 non-zero results for airborne concentrations of dichlorvos. The median of the non-zero results was 0.11 mg m-3 dichlorvos and the 95th percentile was 0.18 mg m-3; with the highest result 0.22 mg m-3. Results for residues in food (124 samples, excluding beverages which had only low concentrations - mostly below 0.1 ppm dichlorvos) are summarised below:

• more than 50 % of meals contained 0.1 ppm or less of dichlorvos, • more than 35 % of meals contained 0.1 to 0.3 ppm of dichlorvos, • 95th percentile was at 0.34 ppm, and the highest residue was 0.51 ppm dichlorvos.

Biological monitoring showed no significant differences in markers between the pre-exposure and exposure periods. Exposed persons spent about 14 hours per day at home, although it is unclear what individual exposures were.

4.2.3.9 SIMULATED USE DATA

A summary of an efficacy study relating to a slow release controllable cassette that contained 80 % w/w dichlorvos (20 g per unit) was submitted (Unpublished, 1999a). Dichlorvos concentrations were determined for an unventilated 1000 ft3 (28.3 m3) room at 1 week, 4 weeks and 8 weeks into an efficacy trial. Concentrations were found to be at 0.09 mg m-3 up to 4 weeks and at approximately 0.05 mg m-3 after 8 weeks.

4.2.3.10 SPRAYING

This report related to professional spraying of a 0.5 % emulsion of dichlorvos in water as a fine band-spray from a compression sprayer operated at 1.4 bar (9.4 ml spray per m2 sprayed) (Gold et al., 1984). However, this mode of use is not one that is currently approved and the report provides no information of use to the current review.

4.2.3.11 HOSPITAL NURSERY STUDY

In this study, dichlorvos impregnated strips containing 20 % dichlorvos were either hung centrally from the ceiling, or close to the walls in hospital nursery wards and adjoining rooms (Cavagna et al., 1970). The strips were deployed at one strip for each 40 m3 or 30 m3 and resultant air concentrations measured over a period of six days. Dichlorvos concentrations were measured over a number of weeks, samples being taken at bench height. In poorly

123

ventilated areas concentrations reached a peak of 0.28 mg m-3 after about two days, with time-weighted average concentrations close to 0.15 mg m-3. In another area, concentrations of dichlorvos reached 0.13 mg m-3 reducing to 0.027 mg m-3 after the ventilation was turned on. The study gives little indication of the degree of ventilation, which appears to be the major determinant of airborne concentration of dichlorvos. These recorded values help define the profile of concentrations in warm rooms with little ventilation.

4.2.3.12 SUMMARY OF DATA SUBMITTED BY APPROVAL HOLDERS

A number of studies have attempted to establish the profile of airborne concentrations of dichlorvos following use of slow release products. There appears to be little information available to distinguish between the concentrations that develop from dichlorvos-impregnated strips, closed or open controllable cassettes or non-controllable cassettes. However, HSE consider that under normal conditions of use there probably is no difference as the devices are designed to generate an efficacious concentration. The principal factor determining concentration of dichlorvos is the level of ventilation in the room. It is clear that good ventilation, however defined, but generally with open doors and windows, will lead to considerably lower concentrations than closed areas. Areas, which are both poorly ventilated and warm, may reach in excess of 0.1 mg m-3 whereas for well-ventilated areas, levels closer to 0.02 mg m-3 could be expected. The 'Arizona (Tucson)' study suggests highest concentrations in domestic dwellings may reach 0.2 mg m-3, but it is not clear how comparable US conditions are to the UK experience and the study used excessive deployment of dichlorvos-impregnated strips. In the studies, air concentrations range from low (less than 0.02 mg m-3) through to above 0.2 mg m-3 under extreme conditions. For risk assessment it would seem reasonable to use the data generated within the UK where high exposures are of the order of 0.06 mg m-3, typically reducing to 0.02 mg m-3 over time. These values are compatible with those found in the field trials of dichlorvos-impregnated strips described in Section 5.3.1.1. A summary of the findings from relevant selected studies is given in Table 4.6. Food or drinking water can become directly contaminated with insecticide by diffusive uptake. A representative concentration in food and water is proposed from industry data submissions at 0.1 ppm reference concentration.

Table 4.7 Reported dichlorvos air concentrations from use of products containing dichlorvos-impregnated strips

Study Maximum air

concentrations (mg m-3)

Average air concentrations

(mg m-3)

Comment

Modesto homes study (Collins & DeVries, 1973)

0.11 0.06 Typical US in-use conditions

Sittingbourne study (Unpublished, 1979)

0.06 0.01 Average concentrations over the whole of the study outweigh the higher exposures from the early stages.

Museum study (Deer et al., 1993)

0.15 (no ventilation)

- 0.01 mg m-3 when ventilation turned on

124

Study Maximum air concentrations

(mg m-3)

Average air concentrations

(mg m-3)

Comment

Simulated use study (Batth et al., 1973)

0.1 0.04 Use of a 'typical' strip

European and Australian Trials (Elgar & Steer, 1972)

98% below 0.1 84% below 0.05 UK typical values up to 0.06 mg m-3, typically 0.02 - 0.04 mg m-3 after 4 w

The Weterings study (Unpublished, 1997)

0.08 0.03 Test rooms with good ventilation rate

Simulated use data (Unpublished, 1999a)

0.09 Unventilated test room, max. concentration after 1 week - slow release cassette

Hospital nursery study (Cavagna et al., 1970)

0.15 - 0.28 with no ventilation, 0.03 with ventilation

Warm hospital nursery in Italy

4.2.4 HSE DATA ON EXPOSURE TO PUBLIC HYGIENE INSECTICIDES

Since 1992, the Health and Safety Executive has gathered information on human exposure to public hygiene insecticide products in the professional and amateur user sectors, to inform its role of assessing exposure and risk to operators and others. The information takes three forms:

• the pattern of work i.e. the frequency and duration of potential exposure, the amount of product used and seasonal factors,

• exposure surveys i.e. the median and realistic worst case exposures in applying products through site surveys of identified tasks or jobs,

• exposure studies i.e. laboratory-based measures of exposure and residues, principally for consumer products.

The exposure studies informing HSE assessments of public hygiene insecticides in respect of dichlorvos are:

• aerosol spray cans on surfaces (Unpublished, 2001), • aerosol space sprays (Unpublished, 2001),

A data model (Unpublished, 1998) and patterns of use information (Unpublished, 1999b) have been derived from these data. The studies’ exposure data points comprise the potential dermal exposure (the amount of product depositing on the outer surface of the person, other than hands), the exposure of hands, exposure by inhalation, the tasks undertaken, the concentration and amount of product used. For aerosol space sprays and aerosol surface sprays, patterns of use information (Unpublished, 1999b) will be used as the default pattern of use:

125

• aerosol space sprays; 1.6 g s-1 discharged for 6 seconds, 4 times daily • aerosol surface sprays; 375 g per treatment, over 7 minutes, once a week

4.2.5 EXPRESSION OF EXPOSURE DATA

All HSE data are quoted in terms of the insecticide product being applied, and are time- weighted. Data are therefore normalised and presented as ‘mg product min-1’ for dermal exposure and ‘mg product m-3’ for inhalation exposure, respectively (Unpublished, 1998). It is misleading to express ‘exposure’ simply as a single value: exposure is more truly expressed in terms of distributions. However, the data relating to specific modes of application are sparse. The exact nature of exposure distributions cannot be proven and consequently, complex statistical treatments are considered inappropriate. The median value in a distribution, moderated by this frequency, represents a ‘central tendency’ value. For volunteer studies where there are, typically, around 10 data points, a realistic worst case is represented by the highest value found. Where the highest data point is a clear outlier (e.g. many times higher than the next highest point), a decision may be taken to disregard it for the purposes of risk assessment. For aerosol uses, the data sets are small but the values that fall out of them are consistent with other modes of application and have the narrow distribution that would be expected of this specific mode of pesticide delivery.

4.2.6 MITIGATION OF EXPOSURE TO INSECTICIDE PRODUCTS

Clothing, whether or not constituting formal personal protective equipment, does have protective properties and needs to be accommodated within the exposure assessment. For the professional user, impermeable clothing (e.g. Tyvek suit) will stop fluids reaching the skin, but easily becomes contaminated inside and is difficult to clean properly. Furthermore, deposits may concentrate on the equipment surface and are available for dislodging. Amateur users would wear only light clothing indoors and for worst case assessments, may wear only a T-shirt and shorts.

4.2.7 FACTORS AFFECTING PRIMARY EXPOSURE ESTIMATES FOR THE USE OF INSECTICIDE PRODUCTS

4.2.7.1 DERMAL EXPOSURE

HSE has data from laboratory based studies showing the range of dermal exposure from using aerosol space sprays and trigger sprays and considers that these can be used for risk assessment, although the data have not yet been presented for consideration by the Committees (Unpublished, 1999b). What is available for uptake following transfer through clothing and onto the skin is another matter. Consequently, the estimation of systemic dose (via a dermal absorption value) from a quantity of in-use product on the outside of clothing requires a degree of professional judgement.

126

4.2.7.2 EXTENT OF PENETRATION THROUGH CLOTHING

HSE data indicate that 20 % is a realistic figure to adopt for the penetration through a single layer of typical clothing (Unpublished, 1999b). Amateur users may wear a long sleeved shirt and long trousers (and household gloves), or they may wear no more than T-shirt, shorts and open footwear. Defaults are assumed; for the central tendency, 20 % clothing penetration with no gloves; and for the worst case, 50 % clothing penetration with no gloves.

4.2.7.3 EXTRAPOLATION FROM CLOTHING TO SKIN TO SYSTEMIC DOSE

The modelling process is not very good at estimating how much of the product finds its way to the skin, and how much of the active substance in the product is eventually absorbed. The current HSE estimates are precautionary, based on the supposition that all of the product predicted to penetrate the layer of clothing reaches the skin, and a proportion of the active substance within the product immediately penetrates through the skin to become a systemic dose. For dichlorvos, a dermal absorption value of 50 % has been used to calculate exposure.

4.2.7.4 PATTERNS OF EXPOSURE

Account needs to be taken of the pattern of work for specific treatments and the pattern of use for any particular product. Professionals spend only a fraction of their time applying insecticides: paperwork, other jobs (e.g. investigation, vermin control) and travel all take time. The most realistic worst case professional exposure scenario means that the use of a specific product would be very unlikely to exceed a few times a day, some days a week. HSE considers that professional operators would be unlikely to use respiratory protective equipment when applying aerosol based dichlorvos products. Amateurs using a product containing dichlorvos would be expected to follow the use pattern suggested in Section 4.2.4 above, and could use the same product throughout the season.

4.2.8 FACTORS AFFECTING SECONDARY EXPOSURE ESTIMATES FOR THE USE OF INSECTICIDE PRODUCTS

Secondary exposure occurs during or following an insecticide application, and the duration may extend for prolonged periods. There are no clear rules to estimate secondary exposure, which needs consideration on a structured case-by-case basis and in the light of the potential persistence of dichlorvos and its mode of use. Post-application exposure occurs after the product has been used, and the exposed population is anyone in the residential environment that may:

• inhale residual aerosols; sprays only, during or immediately after use • inhale vaporised dichlorvos from deposits; any application • dermally contact deposits; both recently applied and dry or on cassettes • ingest dichlorvos; inadvertently through diffusive uptake by food; ingestion of

dislodged deposits by infants

127

The means of secondary exposure (inhaled, dermal contact, ingestion) is unlikely to be linked to the mode of treatment. ‘Realistic worst case’ exposure estimates are presented for these scenarios based on default value calculations and assumptions.

4.2.8.1 SECONDARY DERMAL EXPOSURE

Treatments in places that are inaccessible (e.g. cracks and crevices, out of reach) should not result in any secondary dermal exposure. Contamination dislodged from spot treatments on to hands or on to clothing depends on the dryness of the surface (and hands), the nature of the treated surface, and the nature of the deposit. From preliminary data for dried residues on a flat glazed tile, it is estimated that 50 % of the total surface contamination could transfer to the skin (Unpublished, 2002). Multiple contacts are unlikely to lead to linear progressive accumulation on the skin, since some skin deposit will dislodge back to the surface. There are, as yet, no data relating to rough surfaces (e.g. wood, non-slip flooring, fabrics, carpets etc.), although it is anticipated that the exposure potential will be lower than for smooth flat surfaces.

4.2.8.2 SECONDARY INHALATION EXPOSURE

Aerosol Spraying generates aerosols in the location of application. Coarse aerosol particles clear quickly from air by settling out; fine respirable particles (below 10 µm aerodynamic diameter) are more likely to clear through natural ventilation of the treated space. It is considered that significant secondary exposure by aerosol inhalation after 30 minutes is unlikely to occur. It is proposed that 10 % of the concentration experienced by the user at the realistic worst case will be taken as a default value for exposure over the intervening 30-minute period. Vapour The saturated vapour concentration (SVC) of free dichlorvos is approximately 136 mg m-3 at 20 ºC. HSE surveys of treated residential properties in support of enforcement have never shown airborne levels of any pesticide above a very small fraction of the SVC. Modelling defaults assume that airborne residential concentrations of pesticide vapours are unlikely to exceed 10 % of the saturated vapour concentration. Published studies and some unpublished reports inform on typical air levels of dichlorvos that may be expected under normal use conditions (described in Section 4.2.3). In well-ventilated areas exposure levels are generally around 0.02 mg m-3 - 0.03 mg m-3. In UK studies and trials, exposures of the order of 0.06 mg m-3 have been observed, particularly when dichlorvos-impregnated strips are newly deployed. Under slightly more extreme conditions of warm rooms with no ventilation, dichlorvos levels may rise to slightly above 0.1 mg m-3.

4.2.8.3 SECONDARY EXPOSURE BY INGESTION

Adults may ingest dichlorvos through smoking or eating when they have not washed their hands after undertaking a treatment.

128

An important concern is infants contaminating their hands or feet with freshly applied surface spray and then sucking these; or mouthing articles that have been in contact with treated surfaces (e.g. toys). Scenarios have been developed for situations immediately following treatment and post-application to dried residues. In the case where the potential for dermal absorption is very low, it is likely that ingestion will provide the main route to systemic dose. Food or drinking water can become directly contaminated with insecticide by diffusive uptake. A representative concentration in food and water is proposed from industry data submissions at 0.1 ppm reference concentration.

4.2.9 EXPOSURE ASSESSMENT: DICHLORVOS

4.2.9.1 INDUSTRY DATA

A number of studies relating to exposure were submitted by Approval Holders. These are summarised in Section 4.2.3. All the studies submitted to the dichlorvos review related to exposure to vapour diffused from dichlorvos-impregnated strips, and/or ingested from diffusion into food. The data submitted give no information on exposure to aerosol products, or on the availability of residues on the exterior of cassettes.

4.2.9.2 HSE AND OTHER AUTHORITIES’ DATA

Exposure data are taken from the research referred to in Section 4.2.4. The relevant models and associated exposure calculations are set out in full in Appendix 4, and the estimates are brought forward to the paragraphs below.

4.2.9.3 PATTERNS OF USE

Table 4.8 Default pattern of use data for dichlorvos

Application type Frequency Job duration Aerosol space spray

4 per day most days (summer)

6 seconds (1.6 g s-1)

Aerosol surface spray 1 per week (spring / summer)

7 min (0.89 g s-1)

Slow release strip (museum use)

1Once per 6 months 110 min

Slow release strip (pheromone traps)

15 per day once per 6 weeks

110 min

Slow release controllable cassette 1up to 4 months no information Slow release non-controllable cassette 13 to 6 months no information

1 information supplied by Approval Holders The default values for penetration of clothing are proposed as 20 % at the central tendency and 50 % at the worst case. Occasionally amateur users may wear household gloves but for the purposes of risk assessment it cannot be assumed they do.

129

4.2.10 SYSTEMIC EXPOSURES DURING APPLICATION-PRIMARY EXPOSURE TO AEROSOLS

4.2.10.1 CALCULATION OF EXPOSURE LEVELS

The data sets, assumptions and exposure calculations are set out in Appendix 4, using the pattern of use defaults established in Table 4.8. There is an assumption of 60 kg for adult bodyweight. The pattern of use for professional aerosol space sprays is an unknown but it could be expected that professionals will be asked to treat a greater level of infestation than usually experienced by the amateur user. When professionals use aerosol space sprays, HSE considers the duration of exposure might not be the same as for amateurs - they are probably likely to use them for longer or in greater quantity. In the absence of anything better, increase in exposure through increased use could be balanced by any decrease expected through the use of gloves. A similar risk assessment should be applied. Tables 4.9 - 4.11 show exposure estimates for aerosol products.

Table 4.9 Contact and systemic exposure to dichlorvos-aerosol space spraying

Exposure item Central

Tendency Worst Case

Amount of product in contact with skin (mg d-1) 45 168 Percentage of active substance in product (%) 0.8 % 0.8 % Amount of dichlorvos in contact with skin (mg d-1) 0.36 1.35 Dermal absorption value (%) 50 50 Systemic exposure to dichlorvos via dermal route (mg d-1) 0.18 0.68 Intake of product by inhalation (mg d-1) 1.670 3.740 Amount of dichlorvos inhaled (mg d-1) 0.01 0.03 Total systemic exposure for 60 kg user (mg kg-1 d-1) 0.003 0.012

Table4.10 Contact and systemic exposure to dichlorvos-aerosol surface spraying

Exposure item Central

Tendency Worst Case

Amount of product in contact with skin (mg d-1) 269 770 Percentage of active substance in product (%) 0.5 % 0.5 % Amount of dichlorvos in contact with skin (mg d-1) 1.34 3.85 Dermal absorption value (%) 50 50 Systemic exposure to dichlorvos via dermal route (mg d-1) 0.67 1.93 Intake of product by inhalation (mg d-1) 0.92 4.11 Amount of dichlorvos inhaled (mg d-1) 0.005 0.02 Total systemic exposure for 60 kg user (mg kg-1 d-1) 0.011 0.035

The estimates for exposure of professionals and amateurs using aerosol surface and space spray products (Table 4.11) assume that the exposure while using this type of product would be equivalent to the sum of the exposures from individual surface spray and space spray applications.

130

Comparisons of the NOAEL identified for dichlorvos (0.05 mg-1 kg-1 d-1) with the estimates of a single primary systemic exposure (as detailed above) result in the Toxicity: Exposure Ratios (TER) shown in Table 4.11.

Table 4.11 Contact and systemic exposure to dichlorvos-aerosol space and surface spraying

Product Central Tendency Exposure

(mg kg-1 d-1)

TER: Worst Case Exposure

(mg kg-1 d-1)

TER: Worst Case

Space Spray 0.003 17 0.012 4 Space and Surface Spray 0.014 4 0.047 1

The TER values for professional and amateur use of aerosols containing dichlorvos are very low and are considered unacceptable. The 50 % dermal absorption value may be considered to be conservative. However, it should be noted that TERs would still be low if the value of 11 % from the dermal absorption study had been used.

4.2.11 SYSTEMIC EXPOSURES DURING APPLICATION-PRIMARY EXPOSURE TO SLOW RELEASE CASSETTES AND STRIPS

Slow Release Controllable And Non-Controllable Cassettes It is not possible to estimate whether exposure via the skin occurs in handling slow release cassettes containing dichlorvos-impregnated strips. Studies submitted by an interested party identified a potentially large use of slow release controllable and non-controllable cassettes by museums (Unpublished, 1980; 1994). It should be noted that this use involves the placement of conventional slow release controllable and non-controllable cassettes of an appropriate size in display cases and storage cupboards. One of the studies recently submitted suggested that personal exposures during deployment of slow release cassettes in display cases and storage cupboards resulted in 8-hour time weighted average air concentrations between 0.017 and 0.005 mg m-3. Workers deployed cassettes over a 140-minute period (Unpublished, 1994). The report states that 2-300 cassettes would be deployed in a session and that operators wore nitrile gloves. The results are reported on the basis of a normal work pattern of 3 hours deploying cassettes followed by 5 hours involved in other unspecified tasks. Individual cassettes would be replaced on an annual basis but it is not possible to state whether the task occurs regularly throughout the year or whether there is a campaign to replenish with new cassettes over a brief period. Workers could be exposed on consecutive days or for a number of days regularly throughout the year. Assumptions: 60 kg adult inhalation rate of 1.25 m3 h-1 (Unpublished, 2001)

131

typical 8-hour time weighted average air concentration is 0.005 mg m-3 worst case 8-hour time weighted average air concentration is 0.017 mg m-3

Table 4.12 Acute inhalation exposure to dichlorvos while deploying slow release cassettes in museum display cases and storage cupboards

Typical Worst Case Air concentration 0.005 mg m-3 0.017 mg m-3 Volume inhaled 10 m3 10 m3 Dichlorvos inhaled 0.05 mg 0.17 mg Systemic dose 0.0008 mg kg-1 d-1 0.003 mg kg-1 d-1

A NOAEL taken from an acute or short term inhalation study is considered most appropriate for the calculation of a TER. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from 2 acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment.

Table 4.13 Toxicity to exposure ratios for acute exposure to dichlorvos while deploying cassettes in museum display cases and storage cupboards

Exposure

(mg kg-1 d-1) NOAEL

(mg kg-1 d-1) TER

Typical 0.0008 0.5 625 Worst case 0.003 0.5 167

The TER values for the acute inhalation exposure to professional users while deploying cassettes in museum display cases and storage cupboards are considered acceptable. The exposure assessment for museum staff does not take account of the handling of cassettes. The routine use of regularly changed nitrile gloves should act to minimise exposure via the hands. Given that all slow release cassettes are sold as sealed units, dermal exposure of consumers during normal handling e.g. in residential settings is considered low. Slow Release Strips The use of the three slow release strip products involves users handling dichlorvos-impregnated strips while placing them in situ. Product A consists of a dichlorvos-impregnated strip, cut to an appropriate size for use in museum display cases, cabinets and small cupboards (max strip: volume ratio: 6 g m-3). Users must wear gloves and use forceps, ensure adequate ventilation, preferably under an extraction hood and limit cutting to 30 minutes per day. Cut up strips must be stored in sealed containers prior to use, which should only be opened in well ventilated areas. The strips should be changed every 6 months. Handling of the strips may result in dermal absorption of dichlorvos. However, no information is available to allow estimation of the degree of dermal exposure that occurs during handling of the strips, however, operators wear gloves which should reduce dermal exposure to some degree.

132

The air concentration of dichlorvos is considered to be an important factor in the degree of acute exposure of workers cutting up slow release strips for museum uses. This exposure scenario is based on an assumed rate of output of 10.3 mg h-1. This value is derived from the mean daily weight loss over 48 hours from a fresh 73.5 g dichlorvos strip placed in a 28 m3 test room with approx. 3 air changes per hour (Unpublished, undated). Exposures for a 60 kg user derived using these assumptions are shown in Table 4.14. Assumptions: 60 kg adult inhalation rate of 18.5 m3 d-1 (Unpublished, 2001) typical exposure is 10 minutes worst case exposure is 30 minutes user exposure to dichlorvos is 10.3 mg h-1 work space 8 m3

Table 4.14 Acute inhalation exposure to dichlorvos while

cutting up strips for museum uses Typical Worst Case Release rate 10.3 mg h-1 10.3 mg h-1 Air concentration 0.21 mg m-3 0.64 mg m-3 Volume inhaled 0.13 m3 0.39 m3 Dichlorvos inhaled 0.03 mg 0.25 mg Systemic dose 0.0004 mg kg-1 d-1 0.004 mg kg-1 d-1

A NOAEL taken from an acute or short term inhalation study is considered appropriate for the calculation of a TER. The NOAEL is, therefore, taken from the shortest term acceptable study where cholinesterase inhibition has been measured. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from 2 acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment. The proposed scenarios for acute inhalation exposure to dichlorvos while cutting up slow release strips produces the Toxicity to Exposure Ratios shown in Table 4.15.

Table 4.15 Toxicity to exposure ratios for acute exposure to dichlorvos while cutting up strips for museum uses

Exposure

(mg kg-1 d-1) NOAEL

(mg kg-1 d-1) TER

Typical 0.0004 0.5 1250 Worst case 0.004 0.5 125

The TER values for the acute inhalation exposure to professional users while cutting up dichlorvos-impregnated strips are considered acceptable. However, it should be noted that the exposure estimates take no account of any additional dermal exposure.

133

Two slow release strips; Product B and Product C are used in indoor pheromone traps that monitor levels of, rather than control moth pests of stored products in public hygiene areas including museums, galleries, warehouses, factories and ship holds. Information provided by Approval Holders suggests that handling of strips is infrequent, occurring at most, once every 6 weeks. The maximum recommended application rate is 1 strip per 100 m2. Handling of the strips may result in dermal absorption of dichlorvos. However, strips are supplied foil wrapped and users will be wearing gloves, which may reduce dermal exposure. No information is available to allow estimation of the degree of dermal exposure that occurs during handling of the strips, however exposures are considered to be low. HSE considers that professional operators would be unlikely to use respiratory protective equipment when replacing strips.

4.2.12 SYSTEMIC EXPOSURES POST-APPLICATION-SECONDARY EXPOSURES

Secondary exposures to pesticide products resulting from adventitious contact with treated areas pose a particular problem in risk assessment. There are an almost limitless number of possible scenarios that could be modelled. For every ‘worst case’ there will be several ‘realistic worst cases’ with lower exposure, and multitudes of cases with low (though prolonged) exposure. A selected reference scenario is used to describe a realistic worst case exposure and put other potential exposures into context. A risk assessment based on this reasonably plausible scenario would indicate the acceptability of risk through secondary exposure. However, the probability of the scenario occurring cannot be predicted with any accuracy. Lower exposure events will occur more frequently, and exposures will occur through scenarios that have not been considered. Higher exposures are conceivable depending on the degree of misuse of the product and the inventiveness of the user. For example, abuse would include breaking open a cassette and chewing a dichlorvos-impregnated strip. Where products have a low dermal penetration value, ingestion can become an important route for exposure. As a result of the ready volatilisation of dichlorvos, exposure through food may be significant when a product is used in the presence of food, for example, use of a slow release product in a kitchen (see Section 4.2.3). This review does not address consumer exposure from food residues associated with the use of slow release products in areas where food may be stored, prepared or consumed.

4.2.12.1 ACUTE SECONDARY EXPOSURE

Selected reference scenario for aerosols and residential uses of slow release controllable and non-controllable cassettes: dermal absorption and ingestion

134

For amateur uses of products containing dichlorvos, it is considered the use of aerosol products as a surface spray has the potential to generate the highest systemic exposure. There has been no considered evaluation of behavioural research to support risk assessment in these difficult areas. The selected reference scenario assumes that:

• wasps feeding on spilt food on a kitchen table • a 6 second burst of aerosol at the target (total 7.2 g of product) • solvent evaporates rapidly • 10 kg infant contacts sprayed area • 10 kg infant removes 5 % of total sprayed residue • the product is not unpalatable • 10 kg infant sucks fingers removing 50 % of contamination.

This infant reference scenario produces a single ingestion of 0.18 g spray fluid, equivalent to 1.4 mg dichlorvos (0.8 % in product, assumed density 1 g ml-1), equivalent to 0.14 mg kg-1. The adult reference scenario is an assumed 5 % of the spray (0.4 g) falling on food about to be eaten. This produces a single dose of 400 mg of spray fluid, equivalent to 3.2 mg dichlorvos (0.8 % in product, assumed density 1 g ml-1), equivalent to 0.05 mg kg-1. An non-exhaustive list of other scenarios which could be considered also involve the inadvertent spraying of foodstuffs, contamination of cups and plates prior to use and contamination arising from a leaking trigger resulting in ingestion from contaminated hands. These scenarios are equally relevant to adult exposure. Consideration of these scenarios could produce estimates of ingestion close to the selected reference scenario. Such use of amateur products is foreseeable and research has shown that label advice often goes unread or is ignored. Therefore foreseeable use has to be included within the risk assessment for such products. It is considered that this reference scenario would be a single, rare event and therefore, a NOAEL taken from an acute or short term study is considered more appropriate for the calculation of a TER. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from 2 acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment. The selected reference scenario produces the Toxicity to Exposure Ratio as shown in Table 4.16.

Table 4.16 Toxicity to exposure ratios for acute secondary exposure by dermal absorption and ingestion

Exposure

(mg kg-1 d-1) NOAEL

(mg kg-1 d-1) TER

135

Adult 0.05 0.5 10 Infant 0.14 0.5 4

The TERs for systemic secondary ingestion for infant consumers are unacceptably low and support the recommendation that current approvals for all aerosol products containing dichlorvos be revoked. It should be noted that the above selected reference scenario does not consider dermal absorption, and exposure would be greater if the 50 % dermal absorption value had been applied. Selected reference scenario for aerosols and residential uses of slow release controllable and non-controllable cassettes: inhalation For amateur and professional uses of aerosols and slow release controllable and non-controllable cassettes containing dichlorvos, it is considered that the use of an aerosol product as a space spray has the potential to generate the highest systemic exposure. The selected reference scenario assumes that:

• user exposure to product (0.8 % dichlorvos) is one-tenth the highest point (569 mg m-3) from the indicative exposure model over a 30 minute period (56.9 mg m-3) (Unpublished, 1998; 2001)

• 60 kg adult inhalation rate of 18.5 m3 d-1 (Unpublished, 2001) • 10 kg infant inhalation rate of 4 m3 d-1 (Unpublished, 2001)

Exposures calculated from this scenario are shown in Table 4.17. It is considered that this reference scenario would be a single, rare event and therefore, a NOAEL taken from an acute or short term study is considered more appropriate for the calculation of a TER. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from two acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment.

Table 4.17 Acute secondary exposure by inhalation

Adult Infant Concentration in-use aerosol 56.9 mg m-3 56.9 mg m-3 Inhaled volume 0.39 m3 0.08 m3 Inhaled dichlorvos 0.18 mg d-1 0.038 mg d-1 Systemic dose 0.003 mg kg-1 d-1 0.004 mg kg-1 d-1

The selected reference scenario for systemic secondary inhalation produces the Toxicity to Exposure Ratio as shown in Table 4.18.

Table 4.18 Toxicity to exposure ratios for acute secondary exposure by inhalation

Exposure (mg kg-1 d-1)

NOAEL (mg kg-1 d-1)

TER

Adult 0.003 0.5 167

136

Child 0.004 0.5 125 The TERs for acute secondary exposure by inhalation are considered acceptable. Selected reference scenario for museum uses of slow release controllable and non-controllable cassettes: inhalation A study submitted by an interested party describes the use of dichlorvos-impregnated cellulose pads in sealed specimen cabinets and display cases within a museum (Unpublished, 1980). It suggests that concentrations of dichlorvos inside specimen cabinets can exceed 1 mg m-3 for newly placed cellulose pads. On opening storage cupboards and display cases, workers would be only transiently exposed to the higher concentration of 1 mg m-3 with concentration reducing to below 0.05 mg m-3 within an hour directly in front of the open cabinet. Several thousand specimen cabinets are stored on three floors of an air-conditioned building. It would be anticipated that workers would only open cabinets during the latter phase of the life of the strip when concentrations of dichlorvos within the cabinet would be close to 0.01 mg m-3 and average exposure of the worker during the process of strip replacement would be a much lower than this value. It should also be noted that there is also the possibility of researchers opening display cases in order to examine specimens, although it is considered that exposure would be low in such cases. It is considered that users would be exposed to dichlorvos vapour when opening display cases, specimen cabinets and storage cupboards. As noted previously, the acute secondary exposure of workers opening museum display cases and storage cupboards containing slow-release products is dependent on the air concentrations predicted to develop within display cases and storage cupboards. This risk assessment has been refined to take into account the recently provided information on air levels during museum uses of slow release cassettes. Assumptions: 60 kg adult inhalation rate of 1.25 m3 h-1 (Unpublished, 2001) operator opens and closes one specimen cabinet dichlorvos is inhaled for 1 minute typical air concentration of dichlorvos 0.05 mg m3 worst case air concentration of dichlorvos is 1 mg m3

Table 4.19 Acute inhalation exposure to dichlorvos while opening museum display cases, specimen cabinets and storage cupboards

Typical Worst Case Air concentration 0.05 mg m-3 1 mg m-3

137

Volume inhaled 0.02 m3 0.02 m3 Dichlorvos inhaled 0.001 mg 0.02 mg Systemic dose 0.00001 mg kg-1 d-1 0.0003 mg kg-1 d-1

A NOAEL taken from an acute or short term inhalation study is considered most appropriate for the calculation of a TER. A NOAEL of 0.5 mg kg-1 d-1, derived from combining the information from 2 acute, well conducted, ethical studies in humans, where no signs of toxicity or inhibition of erythrocyte cholinesterase activity were noted at doses of 0.5 mg kg-1 (n = 6) or 1 mg kg-1 (n = 6) has been identified as suitable for use in the risk assessment.

Table 4.20 Toxicity to exposure ratios for acute exposure to dichlorvos while opening museum display cases, specimen cabinets and storage cupboards

Exposure

(mg kg-1 d-1) NOAEL

(mg kg-1 d-1) TER

Typical 0.003 0.5 50000 Worst case 0.004 0.5 1666

The TER values for the acute inhalation exposure to professional users while opening a single specimen cabinet are considered acceptable. HSE consider it unlikely that a large number of cabinets would be opened at once, other than when cassettes are replaced. At this point, levels of dichlorvos in the specimen cases are likely to be far lower than those used in the risk assessment. Selected reference scenario for slow release strips (museum uses) Product A consists of a dichlorvos-impregnated strip, cut by users to an appropriate size for use in museum display cases, cabinets and small cupboards (max. strip: volume ratio: 6 g m-3). It is considered that users would be exposed to dichlorvos vapour when opening display case, cabinets and small cupboards. HSE consider that the risk from acute inhalation exposure from pieces of slow release strips would be similar, and therefore considered acceptable, to that proposed above for acute secondary exposure from slow release controllable and non-controllable cassettes when opening museum display cases, specimen cabinets and storage cupboards.

4.2.12.2 CHRONIC SECONDARY EXPOSURE

Selected Reference Scenario for museum uses of slow release controllable and non-controllable cassettes Information submitted to this review suggests that it is possible that museum workers may be chronically exposed to dichlorvos through leakage of dichlorvos from display cases, specimen cabinets and storage cupboards. Ambient concentrations around the cabinets rarely exceeded 0.01 mg m-3 (Unpublished, 1980). For risk assessment of chronic exposures it would seem sensible to use a default time-weighted average concentration of around 0.01 mg m-3 for an 8-hour period of work, this taking into account exposure to ambient concentrations measured in adjacent areas.

138

Assumptions:

60 kg adult inhalation rate of 1.25 m3 h-1 (Unpublished, 2001) typical 8-hour time weighted average air concentration is 0.01 mg m-3

A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for chronic secondary exposure to museum workers from the use of slow release cassettes in museum display cases, specimen cabinets and storage cupboards

Table 4.21 Chronic secondary exposure to dichlorvos from museum display cases, specimen cabinets and storage cupboards

Typical Exposure Air concentration 0.01 mg m-3 Volume inhaled 10 m3 Dichlorvos inhaled 0.1 mg Systemic dose 0.002 mg kg-1 d-1 TER 30

HSE consider that the risk from chronic inhalation exposure to museum workers from pieces of slow release strips would be similar to that proposed above for chronic secondary exposure from the use of conventional slow release cassettes in display cases, specimen cabinets and storage cupboards. The TER values for chronic exposure of museum workers to dichlorvos from the use of slow release cassettes and strips in museums display are low and not considered acceptable. However, it should be noted that the ambient air level used in the risk assessment is based on a study conducted in a museum where a large number of cassettes were deployed (300 or more). It is possible that ambient air levels may differ in situations where the number of cassettes deployed and building design differ significantly. Members are asked to consider if use of slow release cassettes and strips in museum display cases, specimen cabinets and storage cupboards may be assessed on a case by case basis on the provision of air level data. HSE consider that the risks to consumers would be acceptable, given that opening museum display cases, specimen cabinets and storage cupboards will rarely occur, museum rooms are generally large and consumers are unlikely to spend long periods of time in a given room should any leakage of dichlorvos from a display case occur. Selected Reference Scenario for use of slow release strips in pheromone traps Two slow release strips are used in indoor pheromone traps used to monitor levels of, rather than control, moth pests of stored products in public hygiene areas including museums, galleries, warehouses, factories and ship holds. Approval Holders have confirmed that that pheromone traps may be sited in areas where food raw materials are stored, food is manufactured and the finished packed product stored. Strips are replaced every 6 to 8 weeks.

139

There is no information available regarding air levels of dichlorvos during the use of these products. In view of this lack of information, air levels have been predicted based on the assumed typical (0.02 mg m3) and worst case (0.06 mg m3) levels for residential uses of slow release controllable cassettes. The products contain approximately 40-fold less dichlorvos (0.49 g compared to 20 g) and are used in an approximately 6-fold greater room volume (minimum 200 m3 compared to 30 m3) than the maximum approved for residential uses of slow release controllable cassettes. Based on this information, predicted air levels for the use of dichlorvos-impregnated strips in pheromone traps fall in the region of 8 x 10-5 mg m3 for a typical case and 2 x 10-4 mg m3 for a worst case. Exposures calculated from this selected reference scenario for a 60 kg operator in vicinity of the pheromone traps for 1 h d-1 for a typical case and 8 h d-1 for a worst case are shown in Table 4.22. Table 4.22 Chronic secondary inhalation exposure to dichlorvos from pheromone traps

Typical Worst Case Air level 8 x 10-5 mg m-3 2 x 10-4 mg m-3 Inhaled volume 0.8 m3 6.1 m3 Inhaled dichlorvos 6 x 10-5 mg d-1 1.4 x 10-3 mg d-1 Systemic dose 9 x 10-7 mg kg-1 d-1 2.3 x 10-5 mg kg-1 d-1

A NOAEL of 0.05 mg kg-1 d-1, taken from the 1 year oral study in the dog, has been identified as suitable for use in the risk assessment for chronic secondary exposure (pheromone traps) from a slow release strip. This value is consistent with a notional NOAEL of 0.016 mg kg-1 d-1 derived from the NOAEC of 0.05 mg m-3, taken from the 2-year inhalation study in the rat (whole body exposure for 23 h d-1). The proposed scenario for secondary inhalation exposure to dichlorvos from pheromone traps produces the Toxicity to Exposure Ratios shown in Table 4.23.

Table 4.23 Toxicity to exposure ratios for secondary chronic inhalation from pheromone traps

Exposure

(mg kg-1 d-1) NOAEL

(mg kg-1 d-1) TER

Typical 9 x 10-7 0.05 55556 Worst case 2.3 x 10-5 0.05 2173

The TER values for the chronic inhalation exposure to professional users from pheromone traps are considered acceptable. Selected Reference Scenario for residential uses of slow release controllable and non-controllable cassettes (for treating areas greater than 6 m3)

140

Slow release units containing up to a maximum of 20 g dichlorvos, are approved, primarily as amateur insecticides for domestic use but also for professional use in public hygiene. Units for treating areas greater than 12 m3 must be controllable i.e. the units are capable of being opened or closed by the user. Slow release products may be replaced at intervals of between three and six months. It is considered that this reference scenario would involve chronic exposure to dichlorvos vapours over the summer season and possibly throughout the year from cassettes placed in rooms; adults and infants inhaling dichlorvos for 18 h d-1 and ingesting dichlorvos residues with 50 % of meals, all taken in the home. A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for chronic secondary exposure (residential) from a slow release product. This value is consistent with a notional NOAEL of 0.016 mg kg-1 d-1 derived from the NOAEC of 0.05 mg m-3, taken from the 2-year inhalation study in the rat (whole body exposure for 23 h d-1). The reference scenario assumes that;

• 60 kg adult inhalation rate 18.5 m3 d-1 (Unpublished, 2001) • 10 kg infant inhalation rate 4 m3 d-1 (Unpublished, 2001) • high air levels 0.06 mg m-3 Section 4.2.3 • typical air levels 0.02 mg m-3 Section 4.2.3 • concentration in food and water 0.1 ppm Section 4.2.3 • adult food intake 1.75 kg d-1 (ECETOC Technical report 58) • infant food intake 0.66 kg d-1 (ECETOC Technical report 58)

Exposures calculated from this scenario are shown in full in Appendix 4.3. The selected reference scenario produces the daily exposures and TERs as shown in Table 4.24.

Table 4.24 TERs resulting from secondary chronic inhalation and ingestion exposure to slow release cassettes (residential uses: areas greater than 6 m3)

Exposure

(mg kg-1 d-1) Typical Exposure

(mg kg-1 d-1) Worst Case

Inhalation 0.005 10 0.014 3.6 Ingestion 0.001 50 0.001 50

Adult

Inhalation + ingestion 0.006 8 0.015 3 Inhalation 0.006 8 0.018 8Ingestion 0.003 17 0.003 17

Infant

Inhalation + ingestion 0.009 6 0.02 2.5 The TER values for residential uses of slow release products for treating areas greater than 6 m3 are low and are considered unacceptable. Selected Reference Scenario for residential uses of slow release controllable and non-controllable cassettes (rooms with limited occupation: areas less than 6 m3)

141

Smaller non-controllable slow release cassettes containing up to a maximum of 2 g dichlorvos are approved for treating small indoor areas, including caravans, boats, bathrooms, wardrobes and cupboards, up to 6 m3. HSE consider that similar air levels and exposure times to those assumed for use of the larger room sized cassettes would be appropriate for situations where these products are used in small rooms, caravans, boats etc. and therefore, based on the risk assessment for the large room sized slow release cassettes, the use of these products is not considered acceptable. For rooms, such as bathrooms, where people would not be expected to spend large amounts of time, HSE consider that exposure times could be reduced to 2 h per day and the risk assessment has been refined accordingly. The selected reference scenario assumes that:

60 kg adult inhalation rate 18.5 m3 d-1 (Unpublished, 2001) 10 kg infant inhalation rate 4 m3 d-1 (Unpublished, 2001) high air levels 0.06 mg m-3 typical air levels 0.02 mg m-3 exposure period 2 h

Table 4.25 Chronic exposure to non-controllable slow release cassettes placed in bathrooms and other rooms where occupation is limited

Adult Infant Typical Worst Case Typical Worst Case Air concentration

0.02 mg m-3 0.06 mg m-3 0.02 mg m-3 0.06 mg m-3

Volume inhaled

1.54 m3 1.54 m3 0.33 m3 0.33 m3

Dichlorvos inhaled

0.03 mg 0.09 mg 0.007 mg 0.02 mg

Systemic dose

0.0005 mg kg-1 d-1 0.0015 mg kg-1 d-1 0.0007 mg kg-1 d-1 0.002 mg kg-1 d-1

A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for chronic secondary exposure (residential) from a slow release cassette. Table 4.26 Toxicity to exposure ratios for chronic exposure to slow release cassettes

placed in bathrooms and other rooms where occupation is limited

142

TER Adult Infant

Typical 100 71 Worst Case 33 25

The worst case TER values for chronic exposure to slow release products in bathrooms and other rooms where occupation is limited are low and are considered unacceptable. Selected Reference Scenario for residential uses of slow release controllable and non-controllable cassettes (wardrobes: areas less than 6 m3) On opening wardrobe doors, dichlorvos vapour will diffuse into the room and produce time-weighted concentrations much lower than those where slow release cassettes are deployed in a room. It is estimated that concentrations may reach 0.006 mg m-3 within the bedroom from freshly deployed slow release cassettes, resulting from intermittent release from wardrobes, where concentrations may reach 0.06 mg m-3 within the enclosed hanging space. Chronic risk assessment is based on assessing exposure to a 12-hour time-weighted average exposure at 0.006 mg m-3 as may be experienced by a sleeping infant or adult. It should be noted that further exposure might occur through wearing clothes hung in treated wardrobes. No information is available on levels of exposure through this route. The selected reference scenario assumes that that wardrobe doors are opened twice daily for a short period and the room is occupied for 12 h per day. Assumptions:

60 kg adult inhalation rate 18.5 m3 d-1 (Unpublished, 2001) 10 kg infant inhalation rate 4 m3 d-1 (Unpublished, 2001) air level in room 0.006 mg m-3 air level in wardrobe 0.06 mg m-3 exposure while doors open 2 mins exposure while doors closed 12 h

Table 4.27 Chronic exposure to slow release cassettes placed in wardrobes

Adult Infant

Exposure when doors are closed Air concentration 0.006 mg m-3 0.006 mg m-3 Volume inhaled 9.25 m3 2 m3 Dichlorvos inhaled 0.06 mg 0.012 mg Systemic dose 0.0009 mg kg-1 d-1 0.0012 mg kg-1 d-1

Exposure when doors are open Systemic dose 0.00002 mg kg-1 d-1 0.00004 mg kg-1 d-1

Total exposure Systemic dose 0.00092 mg kg-1 d-1 0.00124 mg kg-1 d-1

A NOAEL of 0.05 mg kg-1 d-1, taken from the 1-year oral study in the dog, has been identified as suitable for use in the risk assessment for chronic secondary exposure (residential) from a slow release product. This value is consistent with a notional NOAEL of

143

0.016 mg kg-1 d-1 derived from the NOAEC of 0.05 mg m-3, taken from the 2-year inhalation study in the rat (whole body exposure for 23 h d-1).

Table 4.28 Toxicity to exposure ratios for chronic exposure to slow release cassettes placed in wardrobes

TER Adults Infants

Typical 54 40 The TER values for chronic exposure to dichlorvos from the use of slow release cassettes in wardrobes are low and not considered acceptable.

4.2.13 SUMMARY OF TERS

Risk assessments have been presented for both professionals and amateur users and for consumers exposed to aerosols, slow release controllable and non-controllable cassettes and slow release strips containing dichlorvos. The TERs derived from these risk assessments are summarised in Tables 4.29 to 4.34.

Table 4.29 Summary of TERs for primary exposure to aerosols

Product/Use Central Tendency

Worst Case

Aerosol space spray 17 4 Aerosol space and surface spray 4 1

Table 4.30 Summary of TERs for primary exposure to slow release controllable and non-controllable cassettes and strips: museum uses

Product/Use Typical Worst Case

Slow release controllable and non-controllable cassettes-deployment into museum display cases, specimen cabinets and storage cupboards

625 167

Slow release strip-cutting strips for museum use 1250 125 Table 4.31 Summary of TERs for acute secondary exposures to slow release controllable

and non-controllable cassettes and strips: museum uses

Product/Use Typical Worst Case Slow release controllable and non-controllable cassettes-opening of museum display cases, specimen cabinets and storage cupboards

50000 1666

Table 4.32 Summary of TERs for acute secondary exposures to aerosols and residential uses of slow release controllable and non-controllable cassettes (areas greater than 6 m3)

Exposure by Exposure by

144

Ingestion Inhalation Adult 10 167 Infant 4 125

Table 4.33 Summary of TERs for chronic secondary exposures for professional uses of slow release controllable and non-controllable cassettes

and use of slow release strips in pheromone traps Typical Worst

Case Slow release controllable and non-controllable cassettes-diffusion from museum display cases, specimen cabinets and storage cupboards

30 -

Slow release strips-use in pheromone traps 2173 55556

Table 4.34 Summary of TERs for chronic secondary exposures for residential uses of slow release controllable and non-controllable cassettes

Adult Infant Product/Use

Typical Worst

Case

Typical Worst

Case Slow release controllable and non-controllable cassettes-areas greater than 6 m3

Inhalation 10 4 8 3 Ingestion 50 50 17 17 Inhalation + Ingestion 8 3 6 2 Slow release controllable and non-controllable cassettes-areas less than 6 m3 where occupation is limited

100 33 71 25

Slow release controllable and non-controllable cassettes-wardrobes

54 - 40 -

The TER values for primary exposure and acute secondary exposure to aerosols are unacceptable for both professional and amateur users. It is considered unlikely that protective clothing would reduce the systemic exposure of users of these products to an acceptable level. The TER values for primary exposures and acute secondary exposures from museum uses of slow release controllable and non-controllable cassettes and slow release strips are acceptable. However, the selected reference scenarios for chronic exposures from museum uses of these products are low. However, it is possible that ambient air levels may differ in situations where the number of cassettes deployed and building design differ significantly. Therefore it is proposed that the use of slow release cassettes and strips in museum display cases, specimen cabinets and storage cupboards be assessed on a case by case basis on the provision of air level data.

145

TER values for the use of slow release strips in pheromone traps are considered acceptable, however there are concerns about their use in the presence of food. The TER values for chronic secondary exposures from residential uses of slow release controllable and non-controllable cassettes are unacceptable. Further data to refine exposure is not considered appropriate.

4.3 DATA REQUIREMENTS

The following data should be submitted within 1 year. Approval Holders must demonstrate to HSE how they are implementing these requirements within 2 months. Protocols to be agreed with HSE. i. Information on air levels found in a range of museum settings.

The following data should be submitted in regard to the use of slow release strips in pheromone traps, in areas where food may be stored, prepared or consumed, i. Information on consumer exposure from food residues associated with the use of slow

release strips in pheromone traps, in areas where food may be stored, prepared or consumed.

146

5 EFFICACY

5.1 INTRODUCTION

Dichlorvos products are currently approved for use as strips for use in domestic and other premises, including use in museums for the preservation of artefacts. All of the currently approved strips contain dichlorvos as the sole insecticide. Pre-pressurised handheld aerosols are also approved. These products all contain dichlorvos in conjunction with pyrethrins/pyrethroids. Data were previously evaluated in 1994 (ACP published evaluation No. 120). The following evaluation includes these data, and presents further data either submitted in response to the current review or identified as relevant to the review. These data are the results of laboratory, simulated use and field studies, and constitute a data package relating to the use of both strips and pre-pressurised handheld aerosols against a range of target species. The evaluation includes both data held by individual companies and submitted in support of the review and data held in the public domain. Due to the long established use of dichlorvos as an insecticide, a relatively large body of data exists in the public domain. However, many of these data relate to early experiments and sometimes lack details such as the exact formulations, controls, and methods of assessment employed, and, for strip products, the type of product (e.g. controllable or non-controllable cassette). The data evaluation is presented in 6 sections. Section 5.2 presents innate efficacy data on dichlorvos, Section 5.3 presents data on strips, and Section 5.4 presents data on pre-pressurised handheld aerosols. These data are summarised in Section 5.5, and discussed in Section 5.6.

5.2 INNATE EFFICACY - LABORATORY DATA

A study reported an LC50 value for dichlorvos against Musca domestica (common house fly) obtained from a topical application test. A series of doses of the active ingredient formulated in a hydrocarbon solvent were applied to the surfaces of test insects by means of a platinum wire. No control results were reported. The test results showed an LC50 of approximately 0.005 µg l-1. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against M. domestica (Bachmann, 1960). A study investigated the efficacy of a number of insecticides, including dichlorvos, against M. domestica. In the study, which was conducted in the Philippines, the test insects were collected from 15 separate sites, and allowed to lay eggs in breeding media in the bottom of a plastic cup. The eggs were then collected, transferred to a fresh cup, and allowed to pupate. The resultant flies were then reared and raised ready for testing. In addition to the offspring of the wild-caught flies, the Takatsuki strain was used as a standard.

147

The insecticides were diluted with acetone to the required concentration, and applied topically to female insects (21-23 mg body weight). To carry out the application, the insects (20 per test) were anaesthetised using CO2, and 0.5 µl of diluted insecticide applied to the scutum. The insects were then transferred to clean vessels with cotton balls soaked in sugar water to facilitate feeding, and maintained at 25 oC. Mortality counts were then made after 24 h. Each test was replicated 3 times. The authors stated that acetone was used as a negative control. However, these control data were not reported. From the mortality data, LD50 values were determined. The mean LD50 values are presented in Table 5.1

Table 5.1 LD50 values of dichlorvos to 15 colonies of M. domestica from the Philippines

Island Locality Collection site Collection

date Mean LD50

(µg per female fly) Palawan Montible Penal colony 23/01/75 0.0301 Palawan Puerto Princesa Market 24/01/75 0.0315 Palawan Puerto Princesa Airport 26/01/75 0.0832 Mindanao Davao Market 02/02/75 0.1260 Mindanao Davao Market 02/02/75 0.1995 Mindanao Tagum Market 05/02/75 0.0842 Leyte Tacloban Market 20/02/75 0.0539 Leyte Tacloban Rubbish dump 20/02/75 0.1150 Luzon Baguio Slaughter house 01/03/75 0.3540 Luzon Baguio Incinerator 01/03/75 0.1090 Luzon Baguio Rubbish dump 01/03/75 0.1260 Luzon Manila Rubbish dump 12/03/75 0.3305 Luzon Manila Market 12/03/75 0.0710 Luzon Manila Market 12/03/75 0.0910 Luzon Manila Market 12/03/75 0.0820

Takatsuki strain 0.0980 From Table 5.1 the range of LD50 values was 0.0301-0.3540 µg per fly. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against M. domestica (Kano et al., 1977). A study investigated the efficacy of a number of insecticides, including dichlorvos, against M. domestica. In the study, the test insects were collected from food markets in 3 separate districts of Bangkok, and from food markets in 8 separate districts elsewhere in Thailand. A minimum of 100 insects was collected per site. The insects were allowed to lay eggs. The resultant larvae were fed on 1 part corn powder, 1 part fish powder and 3 parts water. Following adult emergence, 2-5 d old adults, fed on 5 % sugar solution, were used in the study. The insecticides were diluted with acetone to the required concentrations ready for topical application. In each test, 25 insects were anaesthetised using ethyl ether, and the diluted insecticide applied to the thorax using a micro syringe at a rate of 0.7 ml per insect. The treated insects were then transferred to a paper cup (50 x 100 mm) and 24 h mortality determined for each concentration. The test temperature and humidity were not reported.

148

Nor were any control data reported. The LD50 values for dichlorvos are presented in Table 5.2. The results in Table 5.2 indicated LD50’s for dichlorvos in the range 0.0021-0.0112 µg per fly. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against M. domestica (Kerdpibule & Hirakoso, 1971).

Table 5.2 LD50 values of dichlorvos to 16 colonies of M. domestica from Thailand

Locality LD50 value

(µg per fly) Bangkok

Phyathai 0.008 Sriyan 0.008 Bang na nai 0.010

Thailand Pakchong 0.0029 Saraburi 0.0042 Lampang 0.0112 Lampoon 0.0021 Chiengmai 0.0027 Surin 0.0060 Srisaket 0.0046 Ubol 0.0070

A study investigated the activity of a range of insecticides, including dichlorvos, against M. domestica. Adult females of the ‘Takatsuki’ strain of M. domestica were used in the test. Larvae were reared on a yeast powder mixture (Tofu), and the adults were fed on sugar and water. On the fourth to fifth day after emergence, the adults were anaesthetised with CO2 and sexed. The knockdown of M. domestica arising from direct contact with the insecticides was assessed by topical application, as was the 24 h mortality. The 24 h mortality was also assessed following film contact and following exposure to dichlorvos vapour. Each test was replicated 3 times. Knockdown The knockdown effect was determined by topical application at a dosage of 1 µg insecticide per insect. A total of 40 insects were used in the 3 replicate tests. The knockdown rate was observed with time (time not reported) and mean KT50 and KT90 values derived. The test temperature and r.h. were not reported. No control data were reported. The mean KT50 and KT90 values for topical application were reported as 0.57 min and 1.18 min, respectively. Although the test conditions were not reported, and no control data

149

were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against M. domestica. Mortality In these tests, anaesthetised females received 1 µl of acetone containing prescribed amounts of insecticide. The application was made to the dorsal side of the thorax using a micrometer - driven syringe. The test insects were kept in a glass vial (dimensions 90 x 70 mm) with a net cover, and a cotton ball impregnated with 5 % sucrose-water solution placed on the cover to facilitate feeding. Mortality counts were taken after 24 h exposure, and LD50 values calculated using probit analysis (Bliss, 1935). Negative (blank) control results were reported. The mean mortalities and resulting LD50 for dichlorvos are presented in Table 5.3.

Table 5.3 Mean mortality and LD50 values of M. domestica following topical application of dichlorvos

LD50 Treatment Dosage

(µg per female) No. of flies Mortality

(%) (µg per female) (µg g-1)* 0.00625 80 0 0.0125 80 2.5 0.025 80 26.3 0.05 80 57.5

Dichlorvos

0.1 80 97.5

0.039 2.051

Control 0 90 0 - - * Average body weight of female M. domestica was 19.0 mg The results in Table 5.3 indicated an LD50 of 0.039 µg per insect (2.051 µg gram-1), and zero mortality in the control. The results therefore provided evidence for the innate activity of dichlorvos when applied topically against M. domestica. Film Contact In these tests filter paper (90 mm diameter) was used as a dry film base. A 1 ml aliquot of the test insecticide dissolved in acetone was pipetted uniformly onto each paper. The papers were placed in Petri dishes (100 mm x 20 mm dimensions); the anaesthetised insects were then transferred into the dishes, and covered with net. Mortality counts were taken after 24 h exposure, and LD50 values calculated using probit analysis (Bliss, 1935).

Table 5.4 Mean mortality and LD50 value of M. domestica

following film contact with dichlorvos

Treatment Dosage (mg m-2)

No. of flies

Mortality (%)

LD50 (mg m-2)

0.1 30 0 0.2 30 3.3 0.4 30 13.3 Dichlorvos

0.8 30 60.0

0.7

Control 0 30 0 -

150

The results in Table 5.4 indicated an LD50 of 0.7 mg dichlorvos m-2, and zero mortality in the control. The results therefore provided evidence for the innate activity of dichlorvos when applied as a contact insecticide against M. domestica. Vapour Toxicity The vapour toxicity of each insecticide was assessed by placing an insecticide-impregnated filter paper in the bottom of a Petri dish. A second Petri dish (same dimensions) enclosing M. domestica in a net was then placed inversely on it to prevent direct contact with the treated filter paper. In this way the flies were exposed to any vapour, but were prevented from coming into contact with the treated filter paper. Vapour toxicity was expressed by the relationship between the dosage of insecticide impregnated in a filter paper (mg m-2) and mortality of M. domestica. The mean mortalities and resulting LD50 for dichlorvos are presented in Table 5.5.

Table 5.5 Mean mortality and LD50 value of M. domestica

following vapour contact with dichlorvos

Treatment Dosage (mg m-2)

No. of flies

Mortality (%)

LD50 (mg m-2)

0.4 30 13.3 0.8 30 26.7 1.6 30 56.7 Dichlorvos

3.2 30 86.7

1.3

Control 0 30 0 - The results in Table 5.5 indicated an LD50 of 1.3 mg dichlorvos m-2, and zero mortality in the control. The results therefore provided evidence for the innate activity of dichlorvos vapour against M. domestica. Residual Effects The residual effect from exposure to dichlorvos was assessed. Each test consisted of 3 replicates. For each test, a series of glass plates and plywood panels (100 mm x 100 mm) were each coated with 1 ml acetone solution containing the prescribed amount of insecticide. Following evaporation of the acetone the test insects were confined on one of the coated plates and on one of the coated panels using a net covered glass ring (90 x 30 mm) for 4 h. The report stated that the test was conducted at 'room temperature', although the precise temperature was not reported. The r.h. was not reported. The treated flies were then transferred to a clean glass vial (90 x 70 mm) and mortality determined after 24 h. The treated plates and panels were maintained under laboratory conditions (conditions not reported), and separate batches of test insects exposed to plates and panels at intervals up to a maximum of 27 d. No control results were reported. The mean mortality rates are presented in Table 5.6.

From Table 5.6 the results for plywood showed that all the application rates except the lowest rate produced 100 % mortality within 24 h. Thereafter, the highest mortality achieved was 66.7 % (2 d after treatment with 50 mg m-2). For glass, the highest application rate produced

151

100 % mortality within 24 h, with the lower rates producing 65-70 % mortality in the same period. Thereafter, the highest mortality was 3.3 % (2 d after treatment with 3.125 mg m-2). Although the test conditions and control results were not reported, the test results provided some evidence that following direct application to plywood or glass, dichlorvos is effective against M. domestica for a maximum of 24 h (Kitagaki et al., 1973).

Table 5.6 Mean percentage mortalities for dichlorvos against M. domestica on plywood and glass substrates over a period of 27 d

Mortality (%)/

Days elapsed after substrates were treated

Substrate Application rate

(mg m-2) 1 2 4 8 16 27

0.78 50.0 0 - - - - 3.12 100.0 0 - - - - 12.5 100.0 50.0 0 - - - Plywood

50.0 100.0 66.7 60.0 40.0 23.3 0 0.78 65.0 0 - - - - 3.12 70.0 3.3 - - - - 12.5 70.0 0 - - - - Glass

50.0 100.0 0 - - - - - Not tested A study investigated the effect of a range of insecticides, including dichlorvos, against M. domestica. A minimum of 4 concentrations of dichlorvos was prepared in corn oil. Three replicates were conducted for each concentration. In each replicate, 10 adult M. domestica (age and sex not reported) were placed in a small cage constructed from a glass cylinder with fine stainless steel screens secured at each end with paraffin film. The 3 cages were suspended with wire inside a 2.78 litre jar and 10-100 µl of dichlorvos in corn oil applied to the inside surface of the jar. The jar was then sealed and continuously stirred for a period of 24 h. The observed mortality of M. domestica after a 24 h exposure to the dichlorvos vapour was recorded, and an LC50 value determined using the trimmed Spearman-Karber method (Hamilton et al., 1977). The test temperature and r.h. were not reported. Both solvent and untreated control tests were conducted. However, the control results were not reported. The results indicated an LC50 of 10 mg m-3. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos vapour against M. domestica (Rice & Coats, 1994). A study investigated the effect of a number of insecticides, including dichlorvos, on larvae of Culex pipiens pallens (northern house mosquito). To determine the susceptibility of the larvae, standard procedures as recommended by the World Health Organisation (WHO) were used (see Appendix 5.1.4). The larvae were collected at several artificial ponds in Seoul, South Korea, and reared to adults under standard insect rearing room conditions. The adults were kept in mosquito cages at 26-28 oC and 70-80 % r.h. The resulting eggs were placed in

152

3 egg rafts, each containing 150 eggs per raft, and each raft placed in 2 litres of deionised water. A series of dichlorvos solutions were prepared in ethyl alcohol. The authors reported that 5 - 6 separate concentrations were used in the study, but the paper did not make clear whether either 5 or 6 concentrations were used in testing dichlorvos. A total of 3 replicate tests were conducted for each concentration. For each replicate a 1 ml aliquot of test solution was added to 249 ml distilled water in a paper cup, and a batch of 25 larvae placed in the cup and exposed to the contents for 24 h. The mortality was recorded after 24 h. The above procedure was repeated in a solvent control test in which 1 ml ethyl alcohol was used instead of dichlorvos. However, the control results were not reported. LC50 values were calculated by probit analysis. The results showed an LC50 of 0.1771 mg dichlorvos l-1 and an LC90 of 0.4167 mg dichlorvos l-1. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied as a contact insecticide against Cx. pipiens pallens larvae (Lee et al., 1997). A study investigated the effect of a number of insecticides, including dichlorvos, on larvae of Aedes aegypti (Linn.) (yellow-fever mosquito), Culex fatigans (Wiedmann) and Anopheles stephensi (Liston) (malaria mosquito). To determine the susceptibility of the larvae, standard procedures as recommended by the WHO were used (Technical Report Series, No. 265, 1963). The test insects were from approximately 2 month old laboratory colonies, established from insects collected in the field at Pondicherry, India. A stock solution of dichlorvos was prepared by dissolution in ethanol, and following a preliminary test with a wide range of dilutions, a number of standard concentrations were prepared. Test concentrations were prepared by adding 1 ml of standard concentration to 249 ml tap water in a 500 ml beaker. The water was then removed from each beaker by means of a strainer, and 25 fourth instar larvae of the target species added. Two replicate tests were conducted for each standard concentration. All of the tests were conducted at a temperature of 25-27 oC, although the r.h. was not reported. A separate test was conducted under the same conditions, in which 1 ml ethanol was added instead of a standard concentration. This acted as a solvent control. Mortality was observed after 24 h and the data corrected for the control results by means of Abbott’s formula (see Appendix 5.1.4). From the mortality data, LC50 values were calculated by probit analysis. The LC50 results are presented in Table 5.7. The results in Table 5.7 indicated an LC50 of 0.045-0.056 mg l-1 (Ae. aegypti), 0.025 - 0.029 mg l-1 (Cx. fatigans) and 0.12-0.15 mg l-1 (An. stephensi). The results therefore provided evidence for the innate activity of dichlorvos when applied as a contact insecticide against mosquito larvae (Das et al., 1979).

Table 5.7 LC50 values for dichlorvos against larvae of Ae. aegypti, Cx. fatigans and An. Stephensi

153

Test species LC50

(mg dichlorvos l-1)

Ae. aegypti 0.045 0.047 0.056

Cx. fatigans 0.025 0.026 0.029

An. stephensi 0.150 0.140 0.120

A study investigated the effect of a number of insecticides, including dichlorvos, on larvae of Culex quinquefasciatus (southern house mosquito). Larvae of 6 different strains of Cx. quinquefasciatus were used in the study. The strains were obtained from 6 different islands in Japan, with one strain obtained from each island. The larvae and egg rafts were collected from artificial containers and ground pools and the mosquitoes maintained separately as laboratory strains. The strains were used for determining insecticide susceptibility within 5 generations. The 4th-instar larvae of the mosquito strains were used for insecticide testing. A series of concentrations of dichlorvos were prepared, and larvae of each strain (number not reported) exposed to each concentration. Neither the number of different concentrations tested, nor the test conditions, was reported. No control data were reported. An LC50 value for each strain was calculated. The LC50’s are presented in Table 5.8.

Table 5.8 LC50’s for 6 strains of Cx. quinquefasciatus

obtained from 6 different islands in Japan

Strain

1 2 3 4 5 6

LC50 (mg l-1)

0.069 0.030 0.050 0.092 0.047 0.054

The results in Table 5.8 showed LC50 ‘s in the range 0.030-0.092 mg l-1. Although the study was poorly reported, the results provided some evidence for the innate activity of dichlorvos when applied as a contact insecticide against Cx. quinquefasciatus larvae (Miyagi et al., 1994). A further study investigated the efficacy of dichlorvos vapour against Anopheles quadrimaculatus (common malaria mosquito). The tests were conducted in a 5.8 m3 chamber with paper lined walls, and were designed to determine the air concentration of dichlorvos required to kill adult An. quadrimaculatus with a 4 h exposure. Dichlorvos vapour was produced by passing dried nitrogen through a glass tube packed with a roll of fibreglass cloth saturated with purified dichlorvos. The rate of vapour production was controlled by varying

154

the flow of nitrogen. To simulate the effects of ventilation, the test chamber was equipped with an exhaust system, which was calibrated to give one air change every 3 min. In the initial series of tests, approximately 50 susceptible An. quadrimaculatus (mixed sexes) were placed in wire cages (approximately 9.0 cm x 10.0 cm) closed with nylon netting. A cage of insects was then suspended at each of 2 positions in the chamber, and air samples obtained within a few inches of the cages. The authors stated that these tests produced a wide variation in the mortalities of female mosquitoes in relation to the dichlorvos air concentration i.e. in some cases 100 % mortality was observed for exposure to 0.009 mg dichlorvos m-3 whilst in other cases 0.022 mg dichlorvos m-3 produced zero mortality. The authors stated that the LC50 was estimated as 0.007-0.020 mg dichlorvos m-3. Given the results obtained in the initial tests, together with further experiments, which indicated that, at lower concentrations, air currents in the exposure chamber produced more distinctive concentration streams in the air than had been previously observed at higher dosage levels, a second series of tests was conducted. In these tests, the arrangement of the test specimens was altered in relation to the location of the air sample intake i.e. 6 cages were arranged around a central cage, and fastened together. Mosquitoes were placed in 3 of the alternating peripheral cages and the related air sample was obtained from inside the central cage. Two such sets of cages were used in each test at the same locations as in the previous experiments. The study made no reference to the use of control tests. The results, expressed as percentage mortality, are presented in Figure 5.1. The results in Figure 5.1 showed that a concentration of 0.0074 mg dichlorvos m-3 produced a mortality rate of < 10 % against A. quadrimaculatus, whereas a concentration of 0.01 mg dichlorvos m-3 produced a mortality rate of between 95 and 100 %. The authors contended that given this large variation in mortality rate over a concentration range of 0.003 mg dichlorvos m-3, then, if it is considered that the air samples represent an average concentration over a period of 4 h, this explains the results in the initial set of tests i.e. that slight variations in concentration, as sampled during the first tests, were frequently not reflected in the biological results. Although the test conditions were not reported, and no control data were reported, the results provided some evidence that a 4 h exposure to an air concentration of 0.01 mg dichlorvos m-3 is effective against A. quadrimaculatus (Maddock & Sedlak, 1961).

155

Dichlorvos concentration (mg m-3)

Figure 5.1 Mortality of female An. quadrimaculatus related to dichlorvos vapour concentration (4 h exposure)

A study investigated the effect of the topical application of a range of insecticides, including dichlorvos, against Periplaneta americana (American cockroach). Prior to the tests, adult female P. americana were collected from manholes and refuse chutes. Doses of each insecticide were then applied to the thorax of batches of 50 insects. No information was given on the test temperature and r.h., or on the amount applied. Knockdown (defined by the authors as 'the inability of cockroaches to walk when gently prodded with a blunt instrument') was recorded after 24 h. Probit analysis was carried out to obtain KD50 values. No control data were reported. A KD50 value of 0.47 µg dichlorvos µl-1 was obtained. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate knockdown activity of dichlorvos when applied topically against P. americana (Ho et al., 1994). A study investigated the effect of a range of synthetic pyrethroids and organophosphate insecticides, including dichlorvos, on field collected strains of Blattella germanica (German cockroach). The test insects were collected in an apartment complex located in Hongje-dong, Seoul from October, 1992 through to April, 1993. The insects were maintained in a rearing room at 27 - 28 °C and 45 - 50 % r.h., and exposed to a photoperiod of 12 h light and 12 h dark. The insects were fed on a mixture of glucose, cheese and lab chow. In each test, adult females (> 98 mg body weight) were randomly selected and anaesthetised with chloroform. A series of 4 or 5 different concentrations of the test insecticide in acetone

0.0050 0.0063 0.0073 0.0074 0.0090 0.0092 0.0093 0.0106 0.01180

20

40

60

80

100

Mor

talit

y (%

)

156

was prepared and applied by the micro-topical application method to the ventral mesothorax. Exactly 1.0 µl of solution was applied to each test insect. Three replicates were performed for each concentration and 10 insects were exposed per replicate. The period of exposure to the insecticide, the test concentrations and the test temperature were not reported. Following exposure the test insects were transferred to 200 mm x 140 mm glass jars and provided with food and water. Mortality was recorded after 24 h and LD50 values calculated using probit analysis. The above procedure was repeated for each test insecticide. The authors reported that the control groups of insects were also treated with acetone and were handled in the same manner as the treated insects. However, no data were reported for these controls. LD50 and LD90 values of 0.405 µg dichlorvos per insect and 1.123 µg dichlorvos per insect, respectively, were reported. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against B. germanica (Shim et al., 1997). A study investigated the effect of a number of insecticides, including dichlorvos, against Cimex lectularius (common bed bug). The test insects were obtained from colonies established about 1 year earlier, and the laboratory colony was maintained in 6 litre containers and provided a bloodmeal twice weekly by placing a restrained 3-week-old chick in the container. Active, partially engorged adults (sex not determined) were removed randomly from the culture 2 d after feeding and used in the tests. Six concentrations were tested, and 4 replicate tests were conducted for each concentration. Solutions of each concentration were prepared by dilution in acetone. In each replicate, 1.2 ml solution was applied to a filter paper contained within a 10 cm Petri dish. Ten insects were then placed on the treated filter paper and maintained at 25 oC (the r.h. was not reported) for 24 h. The mortality rate was determined after 24 h by counting the number of test insects, which did not move when the dish was tapped. The mortality data were then subjected to probit analysis to determine the LC50 and LC90 values for dichlorvos. No control data were reported. The test results showed an LC50 of 2.9 mg dichlorvos l-1, and an LC90 of 5.7 mg l-1 dichlorvos. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied as a contact insecticide against C. lectularius (Fletcher & Axtell, 1993). A study investigated the effect of a range of insecticides, including dichlorvos, against larvae of Cadra cautella (almond moth). In the study, conducted in Australia, three different strains of C. cautella adults were collected from stored food products and grain storage facilities in the south west, central west, and upper north coast, respectively. Cultures of each strain and a laboratory strain, maintained in an insecticide free environment, were reared under natural light at 25 ± 5 °C. The F1, F2 and F3 generations of the test strains were used in the tests. In each test, a series of 5-6 different concentrations was prepared in acetone and applied

157

topically to larvae of the last instar stage, each weighing 15-18 mg. Exactly 1.0 µl of solution was applied to each larva. Four replicates were performed for each concentration and 20 larvae were exposed per replicate. Immediately after treatment, larvae were segregated into groups of 10, placed in clean Petri dishes, supplied with food and held at 27 ± 2 °C and 45-65 % r.h.. Mortality was assessed 120 h after testing. Larvae showing no response to prodding with a needle were considered dead. Probit analysis of the data was used to calculate LD50 values. No control data were reported. The results indicated LD50 values ranging from 0.59-1.47 µg dichlorvos per larvae. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against C. cautella larvae (Attia, 1976). A number of other studies investigated the effect of air concentrations of 0.1 and 0.04 mg dichlorvos m-3 against Tineola bisselliella (common clothes moth), Tinea pellionella (case bearing clothes moth), Attagenus piceus (black carpet beetle) and Anthrenus flavipes (furniture carpet beetle). Although it was reported that the tests were carried out at 18 oC and 50 % r.h., no other information was provided, and nor were any control data reported. The results of the tests were reported as a data summary. This summary is presented in Table 5.9.

Table 5.9 Summary of dichlorvos vapour exposure data on clothes moth and carpet beetle species

Species Age Air Level: 0.1 mg m-3 Air Level: 0.04 mg m-3

Adults 56 % KD and 100 % KD at 2 and 3 h, respectively. All dead after 24 h.

5 % KD and 95 % KD at 2 and 3 h, respectively. All dead after 24 h.

Small larvae Few affected after 5 h. All dead after 48 h.

Few affected after 5 h. All dead after 48 h.

T. bisselliella

Large larvae As for small larvae. 90 % affected after 24 h. 6 % alive after 2 weeks.

Adults 44 % KD and 100 % KD after 1 and 2 h, respectively. All dead after 24 h.

47 % KD and 100 % KD after 1 and 4 h, respectively. All dead after 24 h.

Small larvae 100 % KD after 24 h. All dead after 48 h.

100 % KD after 48 h. All dead after 2 weeks. T. pellionella

Large larvae As for small larvae All affected after 48 h. 34 % alive after 2 weeks (33 % abnormal)

Adults 40 % KD and 100 % KD after 5 and 24 h, respectively. All dead after 48 h.

20 % KD and 100 % KD after 5 and 24 h, respectively. All dead after 48 h.

Small larvae 90 % KD after 24 h. 32 % alive after 4 weeks.

20 % KD after 48 h. 81 % alive after 2 weeks.

A. piceus

Large larvae 12 % KD after 24 h. 77 % alive after 4 weeks.

4 % KD after 48 h. 97 % after 2 weeks.

Adults 50 % KD and 100 % KD after 4 and 24 h, respectively. All dead after 48 h.

50 % KD after 24 h. 100 % dead after 48 h.

Small larvae 40 % KD after 48 h. 13 % alive after 4 weeks.

18 % affected after 48 h. 85 % alive after 2 weeks.

A. flavipes

Large larvae 15 % KD after 48 h. 63 % alive after 4 weeks.

10 % affected after 48 h. 98 % alive after 2 weeks.

KD-Knockdown

158

The results in Table 5.9 indicated that the lower concentration (0.04 mg m-3) produced a knockdown rate of 95 % after 3 h for adult T. bisselliella and 100 % after 4 h for adult T. pellionella. The results also indicated a 24 h mortality rate of 100 % for adults of both species. The results for beetles indicated a knockdown rate of 100 % after 24 h for adult A. piceus and 50 % after 24 h for adult A. flavipes, and indicated a 48 h mortality rate of 100 % for adults of both species. The results also indicated that, for all species, larvae were generally less susceptible than adults. Although Table 5.9 summarised the results of tests conducted to unreported test methodologies, the data provided some evidence that a concentration of 0.04 mg dichlorvos m-3 is effective against adult T. bisselliella, T. pellionella, A. piceus and A. flavipes, but is less effective against larvae of these species (Unpublished, Undated a). A study investigated the activity of a range of insecticides, including dichlorvos, against Tribolium confusum (Duval) (confused flour beetle). A series of 3-6 different concentrations were tested, with 4 replicates conducted per test. The concentrations were prepared by dilution in acetone and 1 ml of each dilution pipetted into a separate Petri dish. The Petri dishes were then allowed to stand at room temperature for 30-40 min, to allow complete solvent vaporisation and the formation of crystals on the glass surface. Fifty adult T. confusum (1-2 weeks, both sexes) were then placed in each Petri dish and held at 30 ± 1 °C and 65 ± 5 % r.h. for 24 h. At the end of this period mortality was assessed, with those insects not moving spontaneously or not responding to gentle pressure, considered to be dead. No control data were reported. From the observed mortality rates, LC50 and LC90 values were calculated by probit analysis. The results showed an LC50 value of 29.58 mg dichlorvos l-1, and an LC90 value of 46.02 mg dichlorvos l-1. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos as a contact insecticide against T. confusum (Aamir, 1983). A study investigated the effect of a range of insecticides, including dichlorvos, against T. confusum (Duval) and T. castaneum (Herbst) (rust-red flour beetle). The test insects were collected from 42 different locations in the USA before being cultured in the laboratory at 27 oC and 60 % r.h.. The tests were conducted using 2-3 week old adults (F3 generation). A number of different dichlorvos concentrations were prepared in acetone and applied topically to the test insects. The mortalities observed after 24 h were recorded and probit analysis used to determine the LC values. No control results were reported. The results indicated an LD50 of 25 µg dichlorvos per g insect for T. castaneum and 106 µg dichlorvos per g insect for T. confusum, and an LD99 of 66 µg dichlorvos per g insect for T. castaneum and 1205 µg dichlorvos per g insect for T. confusum. Although no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically against T. castaneum and T. confusum (Zettler, 1991).

159

A study investigated the effect of a range of insecticides, including dichlorvos, against T. castaneum (Herbst). A minimum of 4 concentrations of dichlorvos was prepared in corn oil. Three replicates were conducted for each concentration. In each replicate, 10 adult T. castaneum (age and sex not reported) were placed in a small cage constructed from a glass cylinder with fine stainless steel screens secured at each end with paraffin film. The 3 cages were suspended with wire inside a 2.78 litre jar and 10-100 µl of dichlorvos in corn oil applied to the inside surface of the jar. The jar was then sealed and continuously stirred for a period of 24 h. The observed mortality after a 24 h exposure to the dichlorvos vapour was recorded, and an LC50 value determined using the Trimmed Spearman-Karber method (Hamilton et al., 1977). The test temperature and r.h. were not reported. Both solvent and untreated control tests were conducted. However, the control results were not reported. The results indicated an LC50 of 11.1 µg dichlorvos cm-3. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos vapour against T. castaneum (Rice & Coats, 1994). A study investigated the efficacy of a range of insecticides, including dichlorvos, against T. castaneum. Two different strains of adult T. castaneum were used, a malathion-resistant strain and a malathion susceptible strain. The susceptible strain was maintained in the laboratory without exposure to insecticide. The test insects were reared on sterilised whole wheat flour at 33 ± 1 °C. For each strain, batches of 300-400 insects were placed inside a glass jar prior to egg laying (oviposition). After every 4 d of oviposition the adults were sieved out and placed in a fresh jar. The jars were left for 1 month, and at the end of this period adult emergence had occurred. In each test, a series of 4-7 different concentrations (concentrations not reported) of dichlorvos diluted in a petroleum ether/acetone mixture was prepared (3 replicates per concentration). For each replicate a 0.5 ml aliquot of each concentration was applied by pipette to the surface of a filter paper (70 mm diameter), using a progressively decreasing spiral to ensure even distribution. The filter papers were allowed to dry for 1 min, transferred to a sheet of plate glass, and left overnight (temperature and r.h. not reported). Immediately before the test the insects were divided into batches of 30 or 40 and held for 1 h without food. Each batch was then randomly assigned to a filter paper and the insects confined for 24 h at 28 ± 1 °C by glass rings (50 mm diameter). At the end of this exposure period mortality was recorded. Each test included 3 control replicates. However, no data were presented for these controls. From the observed mortality rates, an LC50 value for each strain was calculated using probit analysis. The results showed LC50 values of 2123 mg l-1 (malathion resistant strain) and 1750 mg l-1 (malathion susceptible strain). Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied as a contact insecticide against T. castaneum (Pasalu & Bhatia, 1975). A study investigated the efficacy of a number of insecticides, including dichlorvos, against Sitophilus oryzae (rice weevil). The test insects (sex not stated) were maintained in insect

160

rearing jars at 26 + 1 oC and 70 + 5 % r.h. and fed on wheat grains. Four different concentrations of dichlorvos were tested. The diluent was not reported. For each concentration, a deposit of insecticidal film was prepared by adding 1 ml solution to a 5 cm diameter Petri dish. A negative (blank) control test was also conducted. Three replicates were conducted for each concentration and associated control. Where insecticidal solution was added, the solution was allowed to dry for 15 min using an electric fan. Ten insects (7-14 d old) were then introduced into the Petri dish, and mortality counts taken after 3, 6, 9, 12, 18, 24, 48 and 72 h, and thereafter every 24 h until the mortality counts in the control reached 10 %. Moribund insects were treated as dead. The observed percentage mortality was corrected using Abbott’s formula (see Appendix 5.1.4) and LT30 , LT50 and LT90 values determined using probit analysis. The mean LT values are presented in Table 5.10.

Table 5.10 Mean LT values for dichlorvos against S. oryzae

Dichlorvos (mg l-1)

LT30 (h)

LT50 (h)

LT90 (h)

0.64 28.83 44.20 202.14

2.56 20.99 39.50 187.21

5.12 17.40 30.15 116.58

10.24 13.87 23.81 90.04

The results in Table 5.10 indicated that concentrations of 0.64-10.24 mg l-1 produced LT30’s of 13.87 - 28.83 h, LT50’s of 23.81 - 44.20 h and LT90’s of 90.04 - 202.14 h. The results therefore provided evidence that concentrations of 0.64 - 10.24 mg dichlorvos l-1 could produce 90 % mortality against S. oryzae 90 - 202 h after application as a contact insecticide (Karnatak et al., 1991). A study investigated the effect of a number of insecticides, including dichlorvos, against Lasioderma serricorne (F.) (cigarette beetle). The test insects were laboratory-reared on a diet of corn meal-yeast. Females were selected by size (2 mm or larger) 3-6 d after emergence. A number of concentrations were tested (number not stated) in the range 105-230 ng dichlorvos per female. The test solutions were diluted in 80 % acetone and water and were applied topically. Three replicate tests were conducted for each concentration. In each replicate, a minimum of 10 insects was anaesthetised with CO2 and the test solution applied to the dorsum (thorax) of each insect. The insects were then transferred to a glass Petri dish containing a 9 cm filter paper and the dish placed in an incubator at either 18 oC or 32 oC for 24 h. The r.h. in the incubator was not reported. The mortality rate after 24 h was determined, and probit analysis of the data for all concentrations/replicates conducted to determine the concentration required to produce 20, 50 and 95 % mortality at each of the two temperatures. No control results were reported. The results in Table 5.11 indicated that at 18 oC a concentration of 336 ng per female was required to produce 95 % mortality, and 343 ng per female at 32 oC. Although no control

161

data were reported, the results provided some evidence for the innate activity of dichlorvos when applied topically at 18 oC or 32 oC against L. serricorne (Benezet et al., 1988).

Table 5.11 Concentration/mortality data for dichlorvos against L. serricorne at 18 oC and 32 oC

Mean dichlorvos concentration

(ng per female) % Mortality

18 oC 32 oC 20 132 95 50 182 147 95 336 343

Number of test insects 525 517 A study reported LC50 values for various insecticides, including dichlorvos, against Ctenocephalides felis (cat flea). The authors reported that initial tests were conducted in order to establish an approximate range for each LC50. For each active ingredient, a series of dilutions in acetone were then made (0 - 10,000 ppm at 500 ppm intervals) and three 125 µl samples taken from each dilution and pipetted into three 25 ml glass flasks. The flasks were then air dried for 4 h and 15 adult fleas introduced into the bottom of each flask. Mortality was then scored after 24 h. On the basis of the observed mortality, the dilution range was narrowed and the above procedure repeated. The test conditions were not reported. Although the paper reported the use of solutions of acetone as blank controls, no data were reported for these controls. The results for dichlorvos showed an LC50 value of 0.20 mg m-2. Although the test conditions were not reported, and no control data were reported, the results provided some evidence for the innate activity of dichlorvos when applied as a contact insecticide against C. felis (Schwinghammer et al., 1985). A study reported LC50 values for various insecticides, including dichlorvos, against adult C. felis exposed to residues for various periods of time. C. felis adults were reared on domestic cats and the insects then exposed to a range of insecticide active ingredients, including dichlorvos. The paper reported that the number of adult fleas used in these tests varied between 135 - 225, with the majority of the tests using 180 insects. The test insects were exposed to the insecticide for various contact periods. The method of application, the amount/rate applied, and the test conditions were not reported. No control data were reported. The LC50s for dichlorvos are presented in Table 5.12.

Table 5.12 LC50 values for dichlorvos against adult C. felis exposed to residues for various periods of time

Active ingredient Contact time (h) LC50 (mg m-2)

1 11.5 3 7.8 6 5.6 Dichlorvos

24 2.3 The results in Table 5.12 indicated LC50 values ranging from 11.5 mg m-2 (1 h) to 2.3 mg m-2 (24 h). Although the study was poorly reported, the results provided some evidence for the innate activity of dichlorvos against C. felis (Kobayashi et al., 1994).

162

A study investigated the efficacy of various insecticides, including dichlorvos, against C. felis larvae (first instar), pupae and adults. In each of the tests, 19 g sand were mixed with 2 ml acetone solution containing the above active ingredient at one of the following concentrations: 0.1, 1, 10 and 100 ppm. A known weight of Hudson flea medium was then added to the treated sand (to give a mixture of 1 part medium to 19 parts sand) and mixed thoroughly, and 5 g aliquots of this mixture then transferred to each of a series of round tins. Twenty five one-day-old flea eggs were counted onto pieces of black filter paper mounted on a cardboard base, and one filter paper placed on top of the medium in each tin. Two tins (i.e. 50 eggs) were used for each active ingredient concentration. A series of tins were also used as untreated 'blank' controls. The tins were incubated at a temperature of 28 oC and 80 % r.h. After counting the hatching rate (into larvae) for 3 d, the tests were allowed to continue until the remaining adults had hatched. The pupae formed and the number of developed adults was established by sieving of the medium. The results for the inhibition of development of adult fleas by dichlorvos, are presented in Table 5.13, and the effects on eggs-larvae, larvae-pupae and pupae-adults in Table 5.14. The results for inhibition of adult development were corrected using Abbott’s formula (see Appendix 5.3), thus taking into account the control data.

Table 5.13 Results for percentage inhibition of development of C. felis following treatment with dichlorvos

Dichlorvos 0.1 ppm 1.0 ppm 10.0 ppm 100.0 ppm

Inhibition of adult development (%) 0 82 100 100

The results in Table 5.13 show a rate of inhibition of development of 82 % (1 ppm) and 100 % (10 ppm). The results in Table 5.14 show that treatment with 100 ppm dichlorvos prevented the emergence of larvae from eggs. Treatment with 10 ppm dichlorvos resulted in 30 % of the eggs produced larvae, and none of these larvae subsequently developed into pupae. The results for the control tins show high rates of emerged larvae, developed pupae and emerged adults, and little adult mortality. The results provided evidence that dichlorvos at a concentration of 100 ppm inhibited the hatching of C. felis larvae, and at a concentration of 10 ppm prevented the development of pupae and hence the hatching of adults (Unpublished, 1986a).

Table 5.14 Results for the effects of dichlorvos treatment on numbers of hatched larvae, developed pupae and hatched adults: C. felis

163

Active ingredient Conc. (ppm) % larvae % pupae* % adults** 100.0 0 0 0 10.0 30 0 0 1.0 86 16 16 Dichlorvos

0.1 68 94 91 N/A 70 94 89 (11)** N/A 92 87 85 N/A 96 98 94 (10)** N/A 92 93 89 (11)** N/A 92 96 93

Controls

N/A 92 91 91 (8)** *Compared with hatched larvae.**Figures in brackets are percentage mortality Another study by the same authors investigated the activity of dichlorvos against C. felis larvae. This study was conducted according to the same procedure as that followed in the previous study, and the results for the effects on emerged larvae, developed pupae, and emerged adults, are presented in Table 5.15.

Table 5.15 shows that the minimum effective concentrations were: eggs to larvae - 50 ppm; larvae to pupae - 5 ppm; pupae to adults - 5 ppm. The results indicated that a lower concentration of dichlorvos was required to prevent development of larvae and pupae, than that required to prevent development of eggs. The results provided evidence that C. felis larvae and pupae are more susceptible than eggs to dichlorvos (Unpublished, 1986b).

Table 5.15 Results for the effects of dichlorvos treatment on emerged

larvae, developed pupae, and emerged adults: C. felis

Emerged larvae (%)

Developed pupae (%)

Emerged adults (%)

Dichlorvos concentration (ppm)

Trial 1 Trial 2 Trial 1 Trial 2 Trial 1 Trial 2 100.0 0 0 0 0 0 0 50.0 0 0 0 0 0 0 25.0 4 6 0 0 0 0 10.0 35 46 0 0 0 0 5.0 42 45 0 0 0 0 2.5 39 46 3 4 1 4

Control 40 45 35 40 34 39 Minimum Effective Dichlorvos Concentration

(ppm)

50 5 5

5.3 SLOW RELEASE STRIPS

Slow release strips are approved for use as insecticides against flying and crawling insects. The strips are contained within a unit (such as a plastic cassette) and fall into two categories; mini strips for use inside cupboards and wardrobes, limited to a maximum application rate of

164

1 unit per 6 m3, and larger strips for use in rooms at a maximum application rate of 1 unit per 40 m3. Additionally there are approvals for niche market use of dichlorvos strips (which is not applicable for amateur use). An example of such a use involves the placing of small sections of strips in cabinets, cases and small cupboards in museums. Data have been presented on the measured air levels of dichlorvos produced by commercially available strip products when tested in the field. The results of efficacy studies on strips have also been presented. In the evaluation HSE has, wherever possible, considered the measured air levels in relation to the application rate for each product used in these field tests, and for each product tested in the efficacy studies. The application rates have been described in terms of a Strip: Volume ratio (SVR), which relates the amount of available dichlorvos present in a product to the maximum treatment volume, stated on the approval. The SVR’s are expressed as g dichlorvos m-3. The current approved range of SVR’s is 0.49-0.8 g m-3 for general use, and approximately 6 g m-3 for niche market use.

5.3.1 SIMULATED USE DATA

5.3.1.1 DATA ON AIR LEVELS

A study presented air concentration data, and data on dichlorvos release rates, on a commercially available slow release cassette unit under controlled conditions. The unit contained two cellulose pads, each pad impregnated with 8 g dichlorvos. The total area of the pads was 176 cm2. The unit was placed in the dark in a closed room (35 m3, 24 0C and 60 % r.h.) for a period of 16 weeks. Throughout this period, ventilation was maintained at 3 volume air changes per h, which is stated by the author as being the minimum required by UK building regulations (1991) for the design of rooms for habitation. The air concentration of dichlorvos and the dichlorvos release rate were both monitored at intervals during the test, and the results are presented in Figures 5.2 and 5.3. Given that the test strip contained 16 g dichlorvos and was used in a 35 m3 room, the SVR is 0.46 g m-3. The results presented in Figure 5.2 indicated that the dichlorvos air concentration varied between approximately 0.02 and 0.05 mg m-3 during the 16 week test period. The results in Figure 5.3 indicated an identical pattern, with the release rate varying between 2 and 5 mg dichlorvos h-1. These values are compatible with those used in the risk assessment for residential uses of slow release products described in Section 4. Although the room was maintained at typical UK room temperature throughout the test, the level of ventilation was maintained at the minimum level required by UK building regulations. Therefore, although the dichlorvos air levels could, in practice, be lower than those determined in the test, the air levels found in the test are not atypical of those likely to be found in practice (Unpublished, 1995).

165

Figure 5.2 Air concentrations of dichlorvos vapour in a 35 m3 room during use

of a slow release cassette for a period of 16 weeks

Figure 5.3 Release rate of dichlorvos into a 35 m3 room during use

of a slow release cassette for a period of 16 weeks A study presented data on the air concentrations produced by a commercially available product when used in domestic premises in the UK. The test product contained 20.0 % w/w dichlorvos. Each unit contained 14.6 g dichlorvos, and, as the application rate for the product was 1 unit per 30 m3, this was equivalent to an SVR of 0.49 g m-3. In addition to the test product, a slightly smaller version of the product was also tested. The amount of dichlorvos in this second version was not reported. Consequently, the SVR for this version could not be calculated. A total of five test trials were conducted between 1967 and 1970, and the authors stated that the trials were conducted during the period of the year when strips would normally be used for pest control. In each trial, the strips were hung in different types of room e.g. kitchen,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Operation time of unit (weeks)

0

0.01

0.02

0.03

0.04

0.05

0.06

Dic

hlor

vos v

apou

r co

nc. (

mg

m-3

)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Operation time of unit (weeks)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Rat

e of

rel

ease

of d

ichl

orvo

s (m

g h-

1)

166

bedroom, dining room, lounge etc., and, although in the early trials the strips were placed in several rooms in one house, in later trials one room only per house was used. The strips were used at an application rate of 1 unit per 30 + 15 m3 (equivalent to an SVR of 0.32-0.97 g m-3). The air samples were taken each week from each room, and usually twice per week in the first part of the trial. The samples (usually 50 l) were taken from all parts of the room by bubbling the air through 50 ml of distilled water at a rate of 5 l min-1. The samples were taken in the period of the day from 09.00 h to 17.00 h. The dichlorvos content of the air sample extracts was determined by cholinesterase inhibition. In one trial (1970), an investigation was carried out into the variation in the concentration of dichlorvos during the day. The air of four rooms in houses, which were closed during the working day in the absence of the householders, was sampled at regular intervals from the early morning until late at night for a period of 4 weeks. This investigation showed that the air concentrations varied from 0.017 - 0.025 mg dichlorvos m-3. At each sampling time the temperature and r.h. in the room were determined. This showed a mean temperature of 18 - 22 oC, and mean r.h. of 62 - 66 %. An estimate of the ventilation in the room was also made by noting the doors and windows, which were open and also the force of the wind. No attempt was made to regulate the pattern of ventilation and temperature set by the householder. The degree of ventilation found when the houses were visited was maintained during the sampling. The results for the ‘standard’ sized product are presented in Figures 5.4 and 5.5, and those for the ‘small’ version in Figure 5.6.

Figure 5.4 Air concentrations of dichlorvos vapour in rooms in domestic premises

during three separate field trials using a commercially available strip

0 10 20 30 40 50 60 70 80 90 100

110

Time after introduction of strips (days)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Air

con

cent

ratio

n (m

illig

ram

s/cu

bic

met

re)

Trial 1

Trial 2

Trial 3

167

Figure 5.5 Air concentrations of dichlorvos vapour in rooms in domestic premises

during two separate field trials using a commercially available strip

Figure 5.6 Air concentrations of dichlorvos vapour in rooms in domestic premises during two separate field trials using a 'small' version of a commercial strip

The results in Figures 5.4 and 5.5 indicated that in the first 14 d after introduction of the strips the mean air concentration ranged from 0.024 - 0.059 mg dichlorvos m-3. The results also indicated that by the end of the first month the mean concentrations had decreased to

0 10 20 30 40 50 60 70 80 90 100

110

120

Time after introduction of strips (days)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Air

con

cent

ratio

n (m

illig

ram

s/cu

bic

met

re)

Trial 4

Trial 5

0 10 20 30 40 50 60 70 80 90 100

110

120

Time after introduction of strips (days)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Air

con

cent

ratio

n (m

illig

ram

s/cu

bic

met

re)

Trial 1

Trial 2

168

0.019 - 0.044 mg m-3, and, by the end of the 4 month test period, to 0.005 - 0.008 mg m-3. The results for the smaller version of the product (Figure 5.6) indicated a similar trend to the standard sized product (Elgar & Steer, 1972). A study presented data on the air concentrations produced by a commercial product when used in domestic premises in the UK. The test product contained 20.0 % w/w dichlorvos. Each unit contained 14.6 g dichlorvos, and, as the application rate for the product was 1 unit per 30 m3, this was equivalent to an SVR of 0.49 g m-3. The study was conducted in the summer of 1973 in domestic premises in the UK. The test strips were placed in either the kitchens or the living rooms of 30 separate domestic premises, and the dichlorvos air concentration determined, using an enzyme inhibition method, at weekly intervals over a period of 119 d. At each sampling time, the temperature and r.h. were also measured. The mean temperature of the test rooms during the study was 19 - 24 oC, and the mean r.h. was 53 - 75 %. The extent of ventilation of the test rooms was assessed by means of the following scale: 0-calm; 1-gentle breeze; 2-moderate breeze; 3-strong breeze. The criteria for this scale were not reported. The results showed that during the period of the study the ventilation was mainly 0 and 1, although 2 and, occasionally 3, were also recorded. The results for the dichlorvos air levels are presented in Figure 5.7.

Figure 5.7 Air concentrations of dichlorvos vapour in rooms in domestic premises during a 1973 field study using a commercially available strip

The results in Figure 5.7 indicated that during the 119 d period of the study the dichlorvos air concentration decreased from 0.026 mg m-3 (d 7) to 0.004 mg m-3 by the end of the study. Given that the test conditions were not atypical of those to be expected in summer in the UK,

17

1421

2935

4249

5663

7077

8491

98105

112119

Time after introduction of strips (days)

0

0.005

0.01

0.015

0.02

0.025

0.03

Air

con

cent

ratio

n (m

illig

ram

s/cu

bic

met

re)

169

so the air concentrations were not atypical of those to be expected from the product when used at an SVR of 0.49 g m-3 (Unpublished, 1979). A second trial was conducted in domestic premises in the UK in 1974. The trial was conducted on two commercial products similar to the product tested in 1973. Although these products differed from the 1973 product in terms of some of their non-active components, the dichlorvos concentration (20.0 % w/w) and the application rate (1 unit per 30 m3) were identical. The test procedure followed was the same as that in the 1973 trial except that each location was visited at weekly intervals starting with d 1, over the first 10 weeks, and then on d 85, 92 and 106 after which the trial was terminated. In addition, the dichlorvos air concentrations were determined using a packed tube sampling LC technique. The mean temperature of the test rooms during the study was 18.7 - 22.8 oC and the mean r.h. was 56 - 70 %. The results for ventilation showed mainly 1, although 2 were also recorded on a number of occasions. The air concentration results are presented in Figure 5.8. The results in Figure 5.8 indicated that the dichlorvos air concentration produced by Product 1 decreased from 0.023 mg m-3 (d 22) to 0.006 mg m-3 (d 92). The results for Product 2 indicated that the concentration decreased from 0.015 mg m-3 (d 28 & 45) to 0.003 mg m-3 (d 92). As with the 1973 study, the test conditions were not atypical of those to be expected in summer in the UK, and the air concentrations were therefore not atypical of those to be expected from the product when used at an SVR of 0.49 g m-3 (Unpublished, 1979).

Figure 5.8 Air concentrations of dichlorvos vapour in rooms in domestic premises during a 1974 field study using two variants of a commercially available strip

170

5.3.1.2 DIPTERA

A study investigated the effect of a dichlorvos strip and a cassette against M. domestica in 40 locations in domestic homes. The cassette was introduced into 24 of these locations and the strip into the remaining 16. The strip contained 23.6 g dichlorvos and the cassette contained 24.3 g dichlorvos. Two cages, each containing 20 female M. domestica (age unknown), were introduced at each location at weekly intervals for the first 10 weeks and knockdown tests were carried out. Full details were not provided on the test protocol. The temperature, r.h. and the ventilation (the number of open doors or windows) were recorded at all sites on the day the tests were performed. The results are presented in Table 5.16.

The results in Table 5.16 indicated that 45 - 100 % knockdown was achieved after 50 min 1 d after the introduction of the strip, and 75 - 95 % in 120 min 1 day after the introduction of the cassette. The results also indicated that at 15 d 100 % knockdown was achieved within 45 min with the strip and within 90 min with the cassette. Additionally, at 50 d, 40 % knockdown was achieved within 185 min with the strip and 90 % knockdown within 125 min with the cassette (Unpublished, 1979).

Table 5.16 Effect of dichlorvos strips and cassettes on M. domestica

Knockdown (%) Ventilation Generator

type Room

volume (m3)

Day SVR (g m-3) Time

(min) Cage

1 Cage

2

Temp (ºC)

R.h. (%)

Strip 28.1 15 0.84 45 100 100 23 64 1 ID 31.1 1 0.76 <50 100 100 20 66 2 ID Strip 50 0.76 185 40 40 20 58 1 ID

Strip 22.2 1 1.06 50 100 45 19.5 67 2 SW Strip 27.4 1 0.86 60 75 65 20 63 2 ID

41.6 15 0.58 45 0 0 22 60 1 ID Cassette - - 90 100 100 - - - 28.9 1 0.84 120 95 75 21 60 2 ID

50 0.84 65 5 5 23 60 1 ID Cassette - - 125 90 90 - - -

Ventilation: ID: Internal Door Open, SW: Small Window Open - Not tested A study investigated the effects of horizontal distance from a dichlorvos strip on mortality of M. domestica (4 day old females) in a basement room with little natural ventilation. A strip containing 3.6 g dichlorvos was used to treat the 66.8 m3 area (a SVR of 0.05 g m-3 dichlorvos). The results indicated that varying the horizontal distance up to 3 m from the strip had no effect on mortality, and that 97-100 % kill was achieved after 4 h exposure (Mankowska & Goszczynska, 1969). In a further study reported in the same paper, the effect of dichlorvos on M. domestica was assessed by placing 3 containers, each holding 20 flies (4 day old females), into each of 2 treatment areas. The first area had a volume of 115 m3, and was treated with 4 strips (giving a SVR of 0.13 g m-3 dichlorvos) and the second area was of volume 127.5 m3 and was

171

treated with 5 strips (giving a SVR of 0.14 g m-3 dichlorvos). Mortality was recorded after 2, 3, 4, 24 and 48 h exposure, and the test was repeated at intervals up to approximately 8½ months after treatment. The temperature during the study was 18 - 23 oC, with 70 - 80 % r.h. The premises were well ventilated. A summary of the results is shown in Table 5.17. The results in Table 5.17 indicated that the dichlorvos strips produced different mortality results within the two treatment areas. The general trend from both sets of data was that 80 % or greater mortality was achieved after 4 h exposure for up to 119 d (i.e. approximately 4 months) in area 1 and 37 d (approximately 1 month) in area 2. Almost total kill (% mortality) was achieved within 24 h 5 months after initial introduction of the strips. It is likely that the variable results in each 48 h monitoring period reflected changes in the environment of the test areas during the study, hence the variety of exposure/mortality relationships. However, no air level data was available to substantiate this (Mankowska & Goszczynska, 1969).

Table 5.17 Effect of dichlorvos strips on M. domestica

% Mortality Area 1 Area 2

Exposure Time (h) Exposure Time (h)

Time after

application (d) 2 3 4 24 48 2 3 4 24 48 2 0 23 73 100 - 0 12 80 100 - 15 47 72 97 100 - 25 100 - - - 24 2 28 45 100 0 7 58 98 100 - 31 18 55 80 100 - 0 2 33 67 100 34 35 96 100 - - 23 85 98 100 - 37 23 53 88 98 100 32 87 95 100 - 44 48 58 83 95 100 2 22 45 60 100 66 17 50 67 100 - 2 27 37 100 - 94 13 67 78 100 - 0 0 35 100 - 119 7 53 83 100 - 0 0 0 100 - 148 0 23 33 100 - 0 0 0 97 100 183 0 0 0 90 100 0 0 0 7 100 215 0 0 0 100 - 0 0 0 87 100 239 0 0 0 95 100 0 0 0 87 100 259 0 0 0 42 100 0 0 0 43 100

- = Not tested A study investigated the efficacy of a dichlorvos strip against adult M. domestica. The test strip was suspended perpendicularly from the ceiling of a 28.3 m3 test chamber. The chamber was designed to simulate the conditions typically to be expected in a private home i.e. at a temperature maintained at 26.6 + 1.1 oC and 50 + 5 % r.h. The air in the chamber was also continually changed, with the air renewed at rates of 0.33 or 1.0 air changes h-1, rates considered to be consistent with an average home. The test strip was introduced only when the air renewal rate had been standardised to the required rate.

172

The concentration of dichlorvos in the air was monitored throughout the test. For purposes of comparison another dichlorvos strip was suspended from the ceiling of a bedroom in a private house, and the air concentrations monitored in the bedroom and also in the living room, kitchen and basement (dimensions not reported). Air samples were taken each morning and afternoon for 2 weeks, and each morning thereafter until the dichlorvos vapour declined to a non effective level. The results of the air concentration monitoring are presented in Table 5.18.

Table 5.18 Air concentrations of dichlorvos in a 28.3 m3 test chamber and in various rooms of a 684 m3 house

Dichlorvos air concentration (mg m-3) Exposure

period (d) Test chamber

Bedroom Living room

Kitchen Basement

1 0.070 0.050 0.012 0.020 Trace

2 0.080 0.060 0.014 0.040 0.015

3 0.130 0.070 0.032 0.040 0.015

4 N.D. 0.070 0.040 0.051 0.020

5 N.D. 0.100 0.038 0.052 0.014

6 0.090 0.060 0.036 0.046 0.016

7 0.080 0.090 0.036 0.050 0.017

8 0.090 0.072 0.040 0.042 0.020

9 0.120 N.D. N.D. N.D. N.D.

10 0.080 N.D. N.D. N.D. N.D.

11 N.D. 0.083 0.031 0.037 0.016

13 0.110 N.D. N.D. N.D. N.D.

14 0.120 N.D. N.D. N.D. N.D.

15 0.130 0.050 0.035 0.043 0.021

16 N.D. N.D. N.D. N.D. N.D.

17 0.120 N.D. N.D. N.D. N.D.

18 N.D. 0.060 0.018 0.030 0.008

21 0.080 N.D. N.D. N.D. N.D. N.D.: Not determined The results in Table 5.18 indicated that the air concentrations measured in the test chamber were generally comparable with those measured in the bedroom. The biological activity of the strips was evaluated for free flying insects in the test chamber, and for insects introduced into the chamber in cages. For free flying insects, 400-500 active flies (male and female) were released into the chamber and knockdown counts were made at various intervals until only a few flies were seen flying. This point occurred after 65 min. Mortality counts were then made after 24 h. The flies were maintained at the same temperature and r.h. as the test chamber. For caged insects 25 female flies were placed in

173

each of 28 - 40 cages, and the cages placed on the chamber floor. At 5 to 15 min intervals 4 cages were withdrawn from the chamber at a time, and the mortality counted. The maximum exposure time was 90 min. The mortality was then recounted after 24 h. The flies were maintained at the same temperature and r.h. as the test chamber. A control test was undertaken in which the above procedure was repeated for flies introduced into an identical test chamber without a dichlorvos strip. The mortality results for free flying insects are presented in Tables 5.19 and 5.20, and those for caged insects in Tables 5.21 and 5.22.

Table 5.19 Mortality rates for free flying adults of M. domestica at time of withdrawal from a 28.3 m3 test chamber containing a dichlorvos strip

Percentage mortality at indicated min of exposure Air

change rate (h-1)

Dichlorvos air

concentration (mg m-3)

5 10 20 30 40 45 50 60 65

1.0 0.25 17.5 55.3 N.D. N.D. 88.2 N.D. N.D. N.D. N.D. 1.0 0.20 18.4 48.6 N.D. N.D. 84.2 N.D. N.D. N.D. N.D. 1.0 0.10 13.4 28.6 50.9 N.D. N.D. 79.7 N.D. N.D. N.D. 1.0 0.07 N.D. 13.3 44.1 71.6 N.D. N.D. N.D. N.D. 92.6

97.499.3

N.D.: Not determined The results for free flying insects at the time of withdrawal from the test chamber (Table 5.19) indicated that an air concentration of 0.07 mg m-3 produced > 90 % mortality after 65 min exposure. The results also indicated that a concentration of 0.25 mg m-3 produced > 85 % mortality after 40 min. The results for 24 h mortality (Table 5.20) indicated that an air concentration of 0.05 mg m-3 produced > 90 % mortality after 90 min exposure, and 0.22 mg m-3 produced 98 % mortality after 70 min.

Table 5.20 Mortality rates for free flying adults of M. domestica 24 h after withdrawal from a 28.3 m3 test chamber containing a dichlorvos strip

Mortality at indicated min of exposure (%) Air change

rate (h-1) Dichlorvos air concentration

(mg m-3) 5 10 20 30 40 45 50 60 65 70 80 90

0.3 0.22 N N N 17 45 N 77 83 N 98.3 N N

1.0 0.18 N N N 18 55 N 86 97 N 98 N N 1.0 0.12 N N N 16 41 N 69 90 N 95 N N 1.0 0.11 N N N 3 27 N 54 77 N 92 N N 1.0 0.05 N N N 0 5 N 19 21 N 40 77 91

Controls N N N 0 0 N 0 0 N 0 0 0

N: Not determined

Table 5.21 Mortality rates for caged female adults of M. domestica at time of

174

withdrawal from a 28.3 m3 test chamber containing a dichlorvos strip

Mortality at indicated min of exposure (%) Air change

rate (h-1)

Dichlorvos air

concentration (mg m-3)

5 10 20 30 40 45 50 60 65

1.0 0.25 N N N N 99.5* 100**

N N N N

1.0 0.20 N N N N 98.8 99.9**

N N N N

1.0 0.10 N N N N N 95* 100**

N N N

* Females, ** Males, N: Not determined

Table 5.22 Mortality rates for caged female adults of M. domestica 24 h after withdrawal from a 28.3 m3 test chamber containing a dichlorvos strip

Percentage mortality at indicated min of exposure Air change

rate (h-1) Dichlorvos

air concentration

(mg m-3)

5 10 20 30 40 45 50 60 65 70 80 90

0.3 0.22 N N N 52 81 N 97 99 N 99.3 N N 1.0 0.18 N N N 49 81 N 97 99 N 100 N N

1.0 0.12 N N N 32 74 N 91 97 N 99 N N 1.0 0.11 N N N 17 48 N 74 91 N 96 N N 1.0 0.05 N N N 4 22 N 46 48 N 57 91 99

Controls N N N 0 0 N 0 0 N 0 0 0 N: Not determined The results for caged insects at the time of withdrawal from the test chamber (Table 5.21) indicated that an air concentration of 0.10 mg m-3 produced 95-100 % mortality after 45 min exposure. The results also indicated that a concentration of 0.25 mg m-3 produced > 99 % mortality after 40 min. The results for 24 h mortality (Table 5.22) indicated that an air concentration of 0.11 mg m-3 produced > 95 % mortality after 70 min exposure, and 0.22 mg m-3 produced > 99 % mortality after 70 min. The results for the blank controls indicated zero mortality after 24 h for all exposure times. The results indicated that under conditions simulating practical conditions in domestic premises, a dichlorvos strip producing air concentrations of 0.07 - 0.25 mg m-3 could produce > 85 % mortality against free flying M. domestica after 40 - 65 min exposure to the strip (Batth et al., 1973). A study investigated the effect of a slow release strip against M. domestica. Two strips, each containing 15 g of dichlorvos impregnated into a cellulose strip were used. Strip 1 was opened and after 7 weeks the unit was placed in one half of a 28.3 m3 test room, which was divided by a polythene sheet and a wooden frame partition (to give a SVR of 1.02 g m-3

175

dichlorvos). After 1 week, 200 flies (age and sex unknown) were released into this half of the test room and the rate of knockdown recorded. Strip 2 was placed directly into the other half of the test room and after 1 week, the test insects were introduced and the rate of knockdown was recorded. No details of the environmental conditions in the test areas were supplied. The results are shown in Table 5.23.

Table 5.23 Effect of dichlorvos strips on M. domestica

% Knockdown Exposure Time (min) Strip 1 Strip 2

20 1.1 11.4 30 21.7 49.5 40 44.4 74.3 50 59.5 86.9 60 67.8 92 90 91.7 98.8 120 98 100 180 100 -

- Not tested

The results in Table 5.23 indicated that Strip 1 (8 weeks old) achieved around 68 % knockdown after 60 min, over 90 % knockdown after 90 min and 100 % knockdown after 180 min. Strip 2 (1 week old) achieved around 75 % knockdown after 40 min, over 90 % knockdown after 60 min and 100 % knockdown after 120 min (Unpublished, 1981). A study investigated the effect of dichlorvos strips against M. domestica. Two impregnated ceramic strips, identical except for the addition of 0.3 % perfume to one of the formulae, were used in this study. The strips contained 26.4 - 28.8 g dichlorvos per unit. One shutter on each unit was opened one tenth and, at various intervals, placed in a test room of 28 m3 (giving a SVR of 0.94-1.03 g m-3 dichlorvos) in order to carry out knockdown tests. The strips were introduced into the test room, which was maintained at 24 ºC and unventilated (humidity not known), one week before the trial was carried out. In each trial approximately 200 M. domestica (age and sex unknown) were released into the room and the rate of knockdown was recorded. Strip 1 was tested at 12 and 16 weeks and strip 2 at 1, 12 and 16 weeks. The results are presented in Table 5.24. The results in Table 5.24 indicated that after 12 and 16 weeks strip 1 produced 100 % knockdown within 90 min of application. The results also indicated that after 12 weeks strip 2 produced 100 % knockdown within 25 min, and within 90 min after 16 weeks (Unpublished, 1980).

Table 5.24 Effect of dichlorvos strips on M. domestica

176

Knockdown (%)

Strip 1 Strip 2 Time (min)

12 Weeks 16 Weeks 1 Week 12 Weeks 16 Weeks 5 0 0 - 0.3 0 10 3.6 9.3 0 33.3 12.1 15 - - - 95.5 36.3 20 35.4 50.3 25.8 99.5 70.9 25 - - - 100 - 30 67.2 76 52.8 - 91.3 40 84.4 88 75.8 - 96.6 50 92.7 94.5 86 - - 60 95.3 97.2 90.4 - 99.8 90 100 100 97.2 - 100 120 - - 98.9 - -

- Not tested A study investigated the insecticidal activity of a 'lantern' containing 25 g of dichlorvos impregnated into a ceramic tile, against M. domestica. The lantern was suspended in the centre of a closed room of volume 28.3 m3 for 12 weeks (equivalent to a SVR of 0.88 g m-3 dichlorvos). The room was maintained at 24 ºC (r.h. was not specified) and was unventilated. 1, 4 and 8 weeks after the introduction of the strip, 150 M. domestica (3 - 5 d old, sex not specified) were released into the room and the rate of knockdown recorded until 95 % was achieved. The flies were then transferred to clean cages and mortality was recorded after 24 h. At the start of the experiment, one shutter on the lantern was opened one tenth and after 3 weeks this same shutter was opened one third. After 6 weeks the second shutter on the lantern was also one third opened. The results are presented in Table 5.25.

Table 5.25 Effect of a dichlorvos lantern on M. domestica

% Knockdown at indicated weeks after introduction Exposure Time (min)

1 Week 4 Weeks 8 Weeks 10 5.9 12.7 0 20 50.6 57.9 2.6 30 79.8 75.4 35.2 40 95.3 89.3 61.7 60 99.2 98.4 93.4 90 - - 99.1

Mortality after 24 h 100 100 100 - Not tested

The results in Table 5.25 indicated that for up to 4 weeks > 98 % knockdown was achieved 60 min after introduction of the lantern, and 99 % knockdown after 90 min at 8 weeks. The

177

results also indicated that the lantern produced 100 % mortality within 24 h throughout the 8 week test period (Unpublished, 1976). A study investigated the effect of a commercial product containing 66.67 % w/w dichlorvos against M. domestica. The test unit, which had an application rate of 1 unit/20 - 40m3, contained 16 g dichlorvos. The SVR was therefore 0.40-0.80 g m-3. The test was conducted in a 36 m3 test chamber maintained at a temperature of 24 oC and r.h. of 68 + 5 %. The test chamber was subjected to ventilation for 30 min both in the morning and in the evening at a rate of 450 m3 per h. The test strip was installed in the chamber at a height of 1.8 m and the effect of the strip observed 10 min, and 1, 15, 30, 60, 90 and 120 d after installation. At each time point 100 insects (5-8 d old, sex not reported) described as resistant to organophosphate insecticides, were introduced into the chamber immediately after the morning period of ventilation. During the first 15 days the test strip was present in the chamber at all times, with this period then reduced to 12 h per day for the remainder of the test. When not present in the chamber the strip was stored in a room at 22 ± 1 oC under the same ventilation conditions. The report did not state the number of replicate tests conducted. The report also made no reference to negative (blank) control data. However, another commercial product was tested for comparison purposes. At each time point the extent of knockdown was observed 5.5, 30 and 60 min, and 2, 4, 6, 8 and 24 h after introduction of the test insects. A KD50 was then calculated for both the test and reference product. The results are presented in Figure 5.9.

Figure 5.9 KD50 values for test strip and 'competitor' strip against M. domestica

The results in Figure 5.9 indicated that the test strip produced a KD50 of 46 min when the insects were introduced into the chamber 10 min after installation of the strip. This then decreased to 10 min after 60 d before increasing again to 54 min (90 d) and 120 min (120 d).

0 30 60 90 120Time after installation of strip (days)

0

50

100

150

200

250

KD

50 (m

inut

es)

Test productReference product

178

The performance of the strip compared favourably with that of the reference strip, and although no blank control data were reported, the results provided some evidence for the knockdown activity of the test strip for a period of up to 60 d against M. domestica (Unpublished, Undated b). A study investigated the efficacy of a commercial strip product containing 42.2 % w/w dichlorvos against M. domestica. The application rate for the product was 1 unit (containing 18.67 g dichlorvos) per 20-40 m3. This gives an SVR of 0.47-0.93 g m-3. The test was conducted in an unventilated chamber approximately 30 m3 in volume (i.e. at a SVR of 0.62 g m-3), and maintained at a temperature of 22 + 2 oC. The r.h. was not reported. The test strip was installed in the chamber approximately 50 cm from the ceiling. The test report did not state whether the strip was installed in the middle of the room. After a period of 17 h the test insects (5 day old, sex and number not reported) were introduced into the chamber, and the knockdown rate recorded with time. The strips were then stored (conditions not reported), and the above procedure repeated after 4, 8, 12 and 16 weeks. No replicate tests were conducted. A blank control test was conducted at each time point in which approximately 100 M. domestica were released into the test chamber, and the mortality determined every hour for a period of 8 h. The results indicated a maximum mortality of 5 % against M. domestica. In the biological test, the knockdown was observed at 5 min intervals up to 30 min, then at 15 min intervals up to 90 min, and then at 30 min intervals up to 180 min. The knockdown results are presented in Figure 5.10.

Figure 5.10 Knockdown rates produced by a dichlorvos strip against

M. domestica over a period of 16 weeks The results in Figure 5.10 indicated that at the beginning of the test the strip produced a maximum knockdown rate of 92 % after 180 min. The results after 4 and 8 weeks storage indicated a slight increase in the speed of knockdown with time, with the strip producing a maximum knockdown of 98 % after 180 min (4 weeks) and 100 % after 180 min (8 weeks).

5 20 45 90 180 5 20 45 90 180 5 20 45 90 180 5 20 45 90 180 5 20 45 90 180

Time after installation of strip (minutes)

0

20

40

60

80

100

% k

nock

dow

n

Start 8 weeks 16 weeks4 weeks 12 weeks

179

The speed of knockdown then decreased slightly at weeks 12 and 16, with the strip producing maximum knockdown rates of 100 % after 180 min (12 weeks), and 99 % after 180 min (16 weeks). Although the test chamber was unventilated, the results provided some evidence that following 16 weeks storage and 17 h in the test chamber, the product, at a SVR of 0.62 g m-3, was able to produce 99 % knockdown of M. domestica within 180 min (Unpublished, 1993a). A study investigated the effect of a commercial strip product containing 80.0 % w/w dichlorvos against M. domestica. The application rate for this product was 1 unit (containing 20.0 g dichlorvos) per 25-40 m3. This gives an SVR of 0.5-0.8 g m-3. The study comprised two tests. The first test aimed to simulate the intermittent use of the unit over a period of 16 weeks. The second test aimed to simulate continuous use of the unit over the same period. Test 1 The test strip was stored in a room of approximately 140 m3 volume for 4 weeks. This room was ventilated normally and heated as required. The strips were suspended approximately 0.5 m away from doors and windows. Approximately 3 d before the end of the 4-week period the strip was placed in a 28.4 m3 chamber (i.e. at a SVR of 0.7 g m-3) maintained at a temperature of 24 oC and r.h. of 50 + 5 %. The chamber was not ventilated. The position of the strip was not reported. At the end of this period approximately 200 3-4 day old M. domestica of a standard laboratory strain (strain not reported) were released into the room, and the rate of knockdown recorded at intervals. From the knockdown results, KD50 values were calculated. The strip was returned to the larger room after the knockdown test. The above procedure was repeated at 4 week intervals over the 16 week period of the test. The report did not state the number of replicate tests conducted. No negative (blank) control test was conducted. However, another commercial product was tested for comparison purposes. The knockdown results are presented in Figure 5.11 (test product) and Figure 5.12 (positive control). The KD50 and KD95 values are presented in Tables 5.26 and 5.27.

Table 5.26 KD50 values for a dichlorvos strip against M. domestica

KD50 (min) Product 4 weeks 8 weeks 12 weeks 16 weeks

Test 8.5 7.8 9.0 32.0 Reference 11.5 11.6 14.6 36.0

180

Figure 5.11 Knockdown of M. domestica produced by a dichlorvos strip used

intermittently over a period of 16 weeks

Figure 5.12 Knockdown of M. domestica produced by a reference dichlorvos

strip used intermittently over a period of 16 weeks .

Table 5.27 KD95 values for a dichlorvos strip against M. domestica

KD95 (min) Product 4 weeks 8 weeks 12 weeks 16 weeks

Test 15.0 12.0 15.0 69.0 Reference 24.5 21.2 27.0 80.0

6 8 10 12 14 16 20 30 40 50 60Time after introduction of M. domestica (minutes)

0

20

40

60

80

100

% k

nock

dow

n4 weeks

8 weeks

12 weeks

16 weeks

8 10 12 14 16 18 20 24 30 40 50 60Time after introduction of M. domestica (minutes)

0

20

40

60

80

100

% k

nock

dow

n

4 weeks

8 weeks

12 weeks

16 weeks

181

The results in Figure 5.11 indicated that the test strip produced high levels of knockdown throughout the period of the test. The results indicated that the speed of knockdown increased between weeks 4 and 8, and although this had decreased again by week 16, the strip was still able to produce > 95 % knockdown after 60 min. These results were reflected in the KD50 and KD95 results presented in Tables 5.26 and 5.27, respectively. These results compared favourably with those for the reference product in Figure 5.12 and Tables 5.26 and 5.27. Although no negative control results were reported, the test results provided some evidence that after intermittent use over a 16 week period in an unventilated chamber, the strip (at an SVR of 0.7 g m-3) was able to produce > 95 % knockdown of M. domestica within 60 min. Test 2 The test unit was suspended in the centre of the 28.4 m3 chamber (i.e. at a SVR of 0.70 g m-3) at 24 oC continuously for 8 weeks. The chamber was empty, windowless and unventilated throughout the period of the test. The air concentration in the test chamber was monitored after 1, 4 and 8 weeks. Air was drawn from about 1.5 m inside the room without opening it, by inserting a glass tube through a hole in the door. Air was drawn through a sintered glass bubbler in water at a rate of 2 l minute-1. On each occasion 3 x 50 l air samples were taken. The biological test was conducted on the day following air sampling. Approximately 200 test insects (WHO standard strain) were introduced into the chamber and the knockdown recorded until at least 95 % of the insects were affected. All the insects were then collected and held in clean cages (conditions not reported) for 24 h, and the mortality rate determined. Neither replicate tests nor a control test were conducted. The knockdown results are presented in Figure 5.13. The KD50 and KD95 and 24 h mortality results are presented in Table 5.28. The air concentration results are presented in Table 5.29.

Figure 5.13 Knockdown rates produced by a dichlorvos strip against

M. domestica over a period of 8 weeks

10 20 30 40 60 90Time after introduction of M. domestica (minutes)

0

20

40

60

80

100

% k

nock

dow

n 1 week

4 weeks

8 weeks

182

Table 5.28 KD50 and KD95 values, and 24 h mortality, for a dichlorvos strip against M. domestica

1 week 4 weeks 8 weeks

KD50 (min) 20.0 19.0 37.0 KD95 (min) 42.0 47.0 65.0

24 h mortality (%) 100.0 100.0 100.0

Table 5.29 Air concentrations of dichlorvos (mg m-3)

Sample 1 Sample 2 Sample 3

1 week 0.083 0.096 0.089 4 weeks 0.098 0.093 0.095

8 weeks 0.063 0.045 0.054

The results in Figure 5.13 indicated that after 8 weeks the product produced > 95 % knockdown after 90 min. This was reflected in the results in Table 5.28, which indicated a KD95 value of 65 min. The results also indicated 100 % 24 h mortality. Table 5.29 indicated that the test product (at an SVR of 0.7 g m-3) produced an air concentration of 0.083-0.098 mg m-3 after 4 weeks, and 0.045-0.063 mg m-3 after 8 weeks. Although no negative control data were reported, the test results provided some evidence that after 8 weeks continuous use in an unventilated chamber, air concentrations of 0.045 - 0.063 mg m-3 could produce > 95 % knockdown of M. domestica within 90 min. The results also indicated that after 16 weeks intermittent use the strip could produce > 95 % knockdown of M. domestica within 60 min (Unpublished, 1999b). A study investigated the effect of a commercial strip product containing 76.92 % w/w dichlorvos against M. domestica. The application rate for this product was 1 unit (containing 15.0 g dichlorvos) per 18-30 m3. This gives an SVR of 0.50-0.83 g m-3. Two separate strips were tested. Each strip was tested (position not reported) in one half of a 28.4 m3 test chamber, which had been divided by a solid partition. Each strip was therefore tested in a 14.2 m3 chamber (i.e. at an SVR of 1.06 g m-3). Strip 1 was placed in one half of the chamber and left for 1 week. At the end of this period, approximately 200 M. domestica (sex and age not reported) were released into the half chamber, and the rate of knockdown recorded at intervals. Strip 2 was stored in an open topped box (approx. 50 cm x 50 cm x 50 cm) in an unheated store for 7 weeks. At the end of this period, the strip was placed in the other half of the chamber for 1 week, and approximately 200 M. domestica then released into the chamber.

183

The report did not state whether the chamber was ventilated during the test, and no details were reported concerning the test conditions. The report also did not state the number of replicate tests conducted, or report any control data. The knockdown results are presented in Figure 5.14.

Figure 5.14 Knockdown rates produced by two dichlorvos strips against M. domestica over a maximum period of 8 weeks

The results in Figure 5.14 indicated that Strip 1 produced 100 % knockdown within 120 min, and Strip 2 within 180 min. This was reflected in the KD95 values of 65 min (Strip 1) and 102 min (Strip 2). Although the test was conducted in an unventilated chamber, and the test conditions and control data were not reported, the results provided some evidence that the product, at an SVR of 1.06 g m-3, could produce 100 % knockdown of M. domestica within 120 min after 1 week, and within 180 min after 8 weeks (Unpublished, 1999b). A study investigated the effect of a commercially available strip product containing 20.0 % w/w dichlorvos (14.6 g unit-1) against M. domestica. The application rate for the product was 1 unit/18-30 m3 (equivalent to an SVR in the range 0.49-0.81 g m-3). The test insects were derived from the WHO Standard Reference Strain and were tested at 2-5 d old. The test strip was introduced into an unventilated 28 m3 volume test chamber (SVR of 0.52 g m-3) at 25 ± 1 °C and 50 ± 5 % r.h.. Approximately 150 flies were released into the test room 24 h later and knockdown was recorded. This procedure was repeated at 1, 2, 3 and 4 month intervals, and between tests the strip was hung in a large air conditioned store at 19 ± 2 °C and 45-60 % r.h. In each test, after knockdown was complete, or after 4 h for the last two tests, all the flies were collected and held in a clean room for 24 h. At the end of this period the percentage knockdown and mortality was determined. No information was reported on whether any negative controls were carried out. The results are presented in Table 5.30.

20 30 40 50 60 90 120 180Time after introduction of M. domestica (minutes)

0

20

40

60

80

100

% k

nock

dow

n

Strip 1 (1 week)

Strip 2 (8 weeks)

184

Table 5.30 Percentage knockdown and mortality results obtained for M. domestica when exposed to dichlorvos strip over a 4 month period

% knockdown (min) Month

10 15 20 30 40 60 90 120 180 240 % mortality

(24 h) 0 2.1 12 27.5 60.5 73.2 97.2 100 - - - 100 1 0 1 3.1 12.8 37.8 70.4 82.7 95.6 100 - 100 2 - - 0 4.7 14.7 51.8 82.4 95.9 100 - 98.2 3 - 0 1 5.2 22.5 48.7 68.1 83.2 93.2 97.9 96.3 4 - - 0 2.1 - 40.9 72.5 80.3 94.8 98.4 98.4

- Not tested KT50 and KT90 values were estimated by the authors by combining the 1 and 2 month and the 3 and 4 month knockdown results from Table 5.30 and plotting these figures on a logarithmic scale. The KT50 and KT90 values are presented in Table 5.31.

Table 5.31 Time taken to knockdown 50 % and 90 % of the total population

of M. domestica when exposed to dichlorvos strips

Month KT50 (min) KT90 (min) 0 26 51

1 and 2 56 100 3 and 4 78 150

Although control results were not reported, the results in Tables 5.30 and 5.31 provided some evidence that after 4 months storage the product, when tested in an unventilated chamber (SVR of 0.52 g m-3), could produce 90 % knockdown of M. domestica after 150 min and > 98 % mortality after 240 min (Unpublished, 1991a). A study investigated the effect of a dichlorvos strip against M. domestica and Ae. aegypti. No SVR can be calculated for the test strip. The study was conducted according to the method described in Appendix 5.2. In the study, the strip was tested against M. domestica and Ae. aegypti using each of the two test methods (50-100 mixed sex for both species and both tests). The insects were held in 150 x 100 mm diameter gauze cylinders. For the LCT test, there was one test with 1 replicate consisting of 5 exposure times. For the (LT)C test there was 1 test with 5 replicates. The results of the study are presented in Table 5.32.

Table 5.32 LCT, (LT)C and dichlorvos activity results (LT95) following exposure to 0.03 mg m-3 dichlorvos

185

(Concentration of dichlorvos) h Species

LCT50 LCT95 (LT50)C (LT95)C LT95 at 0.03 mg m-3

dichlorvos (h) M. domestica 0.0145 0.0500 0.0550 0.0850 2.8 Ae. aegypti 0.0145 0.0270 0.0425 0.0525 1.8

With reference to the calculation in Appendix 5.2, the LCT results in Table 5.32 indicated that for the test concentration of 0.03 µg l-1, the exposure time required to give 95 % mortality was 1 h 40 min (M. domestica) and 54 min (Ae. aegypti). The (LT95)C results indicated that 95 % mortality was produced after an exposure to 0.03 µg l-1 of 2 h 50 min (M. domestica) and 1 h 45 min (Ae. aegypti). The latter results were supported by the LT95 values of 2 h 48 min (M. domestica) and 1 h 48 min (Ae. aegypti). Although no control results were reported, the results provided some evidence that a dichlorvos vapour concentration of 0.03 µg dichlorvos l-1 was effective against M. domestica and Ae. aegypti after an exposure period of approximately 2-3 h (Unpublished, 1964). A study investigated the effect of a commercially available strip product containing 20.0 % w/w dichlorvos (14.6 g unit-1) against Ae. aegypti. The application rate for the product was 1 unit/18-30 m3 (equivalent to an SVR in the range 0.49-0.81 g m-3). The test insects were a long established non-resistant laboratory strain. One hundred sugar fed adults, 5-8 d old, were used for each of the tests. The test strip was introduced into an unventilated 28 m3 volume test chamber at 25 ± 1 °C and 50 ± 5 % r.h. The insects were released into the room 24 h later and knockdown was recorded. This procedure was repeated at 1, 2, 3, 4 and 6 month intervals, and between tests the strip was hung in a large air conditioned store at 19 ± 2 °C and 45-60 % r.h. In each test, after knockdown was complete all the mosquitoes were collected and held in a clean room for 24 h. At the end of this period the percentage knockdown and mortality was determined. No information was reported on whether any negative controls were carried out. The results are presented in Table 5.33.

Table 5.33 Percentage knockdown and mortality results for Ae. aegypti when

exposed to a dichlorvos strip over a 6 month period

% knockdown (min) Month 5 10 15 20 30 40 60 90 120

% mortality (24 h)

0 0 5.4 37.8 93.2 100 - - - - 100 1 1.2 9 26.9 69.2 97.4 100 - - - 100 2 0 3.9 28.6 77.1 96.2 100 - - - 100 3 0 6.2 32.4 79.5 97.7 100 - - - 100 4 0 3.1 10.9 35.9 75 90.6 98.4 100 - 100 6 - 0 4.8 23.6 54.1 76.2 89.2 95.7 100 100

- Not tested -

KT50 and KT90 values were estimated by the authors by combining the relatively uniform knockdown results months 0 to 3 from Table 5.33 in comparison to month 4 and 6 and plotting these figures on a logarithmic scale. The KT50 and KT90 values are presented in Table 5.34.

186

Table 5.34 Time taken to knockdown 50 % and 90 % of the total population of A. aegypti when exposed to dichlorvos strips

Month KT50 (min) KT90 (min) 0 to 3 17 24

4 23 40 6 28 72

Although control results were not reported, the results in Tables 5.33 and 5.34 provided some evidence that after 6 months storage the product, when tested in an unventilated chamber (SVR of 0.52 g m-3) could produce 90 % knockdown of Ae. aegypti after 72 min and 100 % mortality after 120 min (Unpublished, 1991b). A study investigated the efficacy against Ae. aegypti (L) of a commercial strip product designed for use in small spaces. The product contained 20.0 % w/w dichlorvos (3.5 g dichlorvos/unit), and the application rate was 1 unit per 6.0 m3 (equivalent to an SVR of 0.58 g m-3). In each test, a strip was suspended from the centre of the ceiling of a 0.78 m3 volume test cupboard, each cupboard being one of four placed inside a test chamber of approximately 47 m3 volume. The test chamber and cupboards were maintained at 25 oC and 60 % r.h., and were ventilated twice daily at 1.00 p.m. and 4.30 PM for 15 min each time. Two test cages each containing thirty 3-6 d old adult Ae. aegypti (sex not reported) were introduced into each test cupboard 24 h after introduction of the strips. After an exposure period of 4 h, the insects were transferred to clean cages and provided with food and water. After an observation period of 6 h, the mortality rate was determined. The above procedure was repeated 2, 3, 7, 21, 35, 49, 63, 77, 91 and 119 d after introduction of the strips. The author reported that four cages of insects were placed into untreated cupboards as negative (blank) controls. However, control results were only presented for 91 and 119 d. The test results indicated that the strips produced 100 % mortality throughout the test period. The negative (blank) control results indicated mortality rates of 8.33 % (91 d) and 0 % (119 d). The results provided evidence that the product (at an SVR of 4.49 g m-3) was effective against Ae. aegypti for a period of 17 weeks (Unpublished, 1966).

5.3.1.3 DICTYOPTERA

A study investigated the effect of dichlorvos strips on B. germanica and Blatta orientalis (Oriental cockroach) in premises in Poland using PVC strips containing 3.6 g dichlorvos per strip. The effectiveness of dichlorvos against cockroaches was evaluated by placing containers of 30 B. germanica adults of mixed sex, and 30 B. orientalis into a 66.8 m3 basement room, 3 d after the introduction of a dichlorvos strip into the area (a SVR of 0.05 g m-3 dichlorvos). Mortality was observed for 1 month after treatment. The results are shown in Table 5.35.

The results in Table 5.35 indicated that after 10 d, 100 % mortality of B. germanica and 30 % mortality of B. orientalis was achieved. It took another 19 d to kill the remaining B. orientalis individuals. It is considered by HSE that such mortality rates could result in cockroach control, especially at lower population turnover rates (Mankowska & Goszczynska, 1969).

187

Table 5.35 Effect of dichlorvos strips on B. germanica and B. orientalis

% Mortality Exposure Time (d) B. germanica B. orientalis 3 0 0 6 0 0 8 30 0 10 100 30 13 - 50 19 - 60 23 - 70 26 - 80 29 - 100

- Not tested

In another study against B. germanica, reported in the same paper, separate containers of either 10 adult males or 10 adult females were placed into an area which had been treated with dichlorvos strips, either 3 or 18 d after initial introduction of the strips. Mortality was recorded every few days, up to 60 d exposure. The treatment area for this test was not specifically identified, however it was one of 4 treatment areas described in the introduction to the studies, all of which had strips used at a SVR of 0.13-0.14 g m-3 dichlorvos, and had temperatures of 18-23 oC and 70-80 % r.h. during the trials. The results are presented in Table 5.36.

Table 5.36 Effect of dichlorvos strips on B. germanica

% mortality 3 d after application

% mortality 18 d after application

Exposure time (d)

Female Male Female Male 1 0 0 0 6 2 53 97 0 10 3 100 100 0 13 4 - - 2 13 8 - - 7 46 10 - - 10 53 13 - - 10 70 16 - - 10 86 20 - - 10 96 24 - - 42 100 28 - - 52 - 36 - - 55 - 42 - - 75 - 50 - - 80 - 60 - - 80 -

- Not tested The results in Table 5.36 indicated that the effects of the dichlorvos strips fell dramatically in strips that were 18 d old (time taken to produce 100 % kill rose from 3 d to more than 60 d in female B. germanica and from 3 d to 24 d in males). They also indicated that male

188

B. germanica under the test conditions are more susceptible to dichlorvos than females (Mankowska & Goszczynska, 1969).

A study investigated the effect of dichlorvos vapour on P. americana and B. germanica. Two sets of tests were conducted using either resinous granules impregnated with dichlorvos (20 % w/w), or a wax stick containing 25 % w/w dichlorvos. The first series of tests was conducted in sealed 44 m3 chambers. In each test, 10 or 20 P. americana and 50 or 100 B. germanica adults were placed in screen wire cages and the cages placed in a chamber. The insects were then exposed to either 0.25, 1.04, 1.7 or 7.4 mg dichlorvos m-3 for exposure periods varying from 2-48 h. After exposure the cockroaches were transferred to quart jars containing food, water and harbourages. Knockdown counts were made at the end of the exposure periods and mortalities recorded 24 and 48 h after the exposure periods. The mean temperature of the chambers was 29.7 + 1.7 oC (0.25 and 1.04 mg m-3 concentrations) and 26.6 + 1.7 oC (1.7 and 7.4 mg m-3 concentrations). The r.h. was not reported. The results for percentage knockdown and 24 and 48 h mortalities (test 1) are presented in Table 5.37. No control data were reported. Table 5.37 Knockdown and mortality rates for P. americana and B. germanica exposed

for varying periods of time to varying dichlorvos air concentrations (test 1)

Percentage mortality after indicated post-exposure time

Insect Dichlorvos concentration

(mg m-3)

Exposure period (h)

Percentage knockdown at the end of the

exposure period24 h 48 h

0.25 24 24 36 48

50 90 90 60

50 70 90 60

50 70 90 60

1.04 6

14 16 20

50 0

70 100

50 0

80 100

50 0

80 100

1.70 4 4 5

50 95

100

50 100 100

50 100 100

P. americana

7.40 2 100 100 100

B. germanica 0.25

16 24 30 36 48

10 84 80 74 98

16 80 80 76 98

16 80 82 76 98

189

Percentage mortality after indicated post-exposure time

Insect Dichlorvos concentration

(mg m-3)

Exposure period (h)

Percentage knockdown at the end of the

exposure period24 h 48 h

1.04

8 14 14 16 16 20 24

60 26 26 98 48

100 94

60 32 36 98 54

100 96

60 34 38

100 56

100 96

1.70

4 6 6

24 24

0 82 95

100 100

2 90 97

100 100

2 90 98

100 100

7.40 2 100 100 100

A second series of tests was conducted to determine the effect of dichlorvos vapour on insects hiding in small containers. The containers used were paper bags made of craft paper and boxes consisting of single faced, flexible, corrugated cardboard shaped into cylindrical sleeves slightly larger than the screen-wire cages used to hold the insects. After the insects were placed in the bags, the tops of the bags were twisted to close them. The insects in these containers (10 or 20 P. americana and 50 or 100 B. germanica) were then exposed for different periods of time to both 0.25 and 1.04 mg dichlorvos m-3. The results for test 2 are presented in Table 5.38. No control data were reported.

Table 5.38 Knockdown and mortality rates for contained test insects exposed for

varying periods of time to varying dichlorvos air concentrations (test 2)

Percentage mortality after indicated post-exposure time

Test insect Dichlorvos concentration

(mg m-3)

Exposure container

Exposure period

(h)

Percentage knockdown at the end

of the exposure

period 24 h 48 h

P. americana

0.25

1.04

Bag Box Bag Bag Box

48 48 24 36 24

0 0

60 50 0

0 10 60 60 10

0 10 60 60 10

B. germanica

0.25

1.04

Bag Box Bag Bag Box

48 48 24 36 24

88 2

18 94 8

90 2

26 96 10

90 2

26 92 12

The results in Table 5.37 show that for both P. americana and B. germanica the lowest concentration producing 100 % knockdown and kill was a 20 h exposure to 1.04 mg m-3. Table 5.38 shows that for both cryptic P. americana and B. germanica, the vapour was less effective than when tested in the 44 m3 chamber. Although no control data were reported, the results provided some evidence that in a 44 m3 chamber, an air concentration of 1.04 mg dichlorvos m-3 could produce 100 % knockdown and kill of P. americana and B. germanica after 20 h (Smittle & Burden, 1965).

190

A study investigated the effect of a commercial strip product containing 80.0 % w/w dichlorvos against B. germanica and B. orientalis. The application rate for this product was 1 unit (containing 2.0 g dichlorvos) per 2 m3. This gives an SVR of 1.0 g m-3. In the tests, which were conducted in a 28.4 m3 test chamber, 13 units were pinned to the wall at 1 metre intervals around the chamber just above floor level (giving an SVR of 0.9 g m-3). The temperature of the chamber was maintained at 26.7 oC and 75-80 % r.h.. The room was unventilated. The test insects were exposed in the chamber in open glass dishes with filter paper on the bottom. The dishes (6 for B. germanica and 3 for B. orientalis) were placed in the centre of the chamber, close to a tile placed near the centre of the chamber, and half way between the tile and the centre of the chamber. For each species, 20 two-thirds grown nymphs were used per dish. The tests were conducted 3, 14 and 28 d after the introduction of the tiles. At each time point, the rate of knockdown was recorded at 3, 5 and 24 h. Negative (blank) control tests were conducted. The knockdown results are presented in Tables 5.39 and 5.40.

Table 5.39 Knockdown rates produced by a dichlorvos strip against

B. germanica over a maximum period of 28 d Knockdown rate (%)

3 d 14 d 28 d Position of dish

3 h 5 h 24 h 3 h 5 h 24 h 3 h 5 h 24 hNear tile 15 100 N/T 15 100 N/T 5 40 100 Near tile 30 95 100 35 100 N/T 10 60 100 Half way 25 85 100 20 100 N/T 5 65 100 Half way 20 75 100 20 100 N/T 0 50 100

Centre of chamber 10 65 100 30 90 100 5 80 100 Centre of chamber 10 70 100 20 90 100 0 60 100

Negative control 0 0 0 0 0 0 0 0 0 N/T Not tested

Table 5.40 Knockdown rates produced by a dichlorvos strip against

B. orientalis over a maximum period of 28 d Knockdown rate (%)

3 d 14 d 28 d Position of dish

3 h 5 h 24 h 3 h 5 h 24 h 3 h 5 h 24 hNear tile 40 100 N//T 0 15 100 0 5 100 Half way 25 100 N/T 0 5 100 0 15 100

Centre of chamber 35 100 N/T 0 5 100 0 10 100 Negative control 0 0 0 0 0 0 0 0 0

N/T Not tested The results in Table 5.39 indicated that at 3 d, knockdown of B. germanica tended to be more marked the closer the insects were to the tile. The data in Tables 5.39 and 5.40 indicated that after 28 d, 100 % knockdown of both species was achieved within 24 h. The control results showed zero knockdown.

191

As the product is designed for use in small areas, which may be unventilated, the lack of ventilation in the test chamber reflects this usage pattern. Therefore, the results provided evidence that the product (at an SVR of 1.0 g m-3) produced effective knockdown of Dictyoptera for up to 28 d (Unpublished, 1999c). A study investigated the effect against adults and larvae of B. germanica, P. americana and Leucophaea maderae (grey cockroach), of a commercial strip product designed for use in small spaces. The product contained 20.0 % w/w dichlorvos (3.5 g dichlorvos per unit), and the application rate was 1 unit per 6.0 m3 (equivalent to an SVR of 0.58 g m-3). In each test, a strip was suspended from the centre of the ceiling of each of four 0.78 m3 volume test cupboards placed inside a test chamber of approximately 47 m3 volume. The test chamber and cupboards were maintained at 25 oC and 60 % r.h., and were ventilated twice daily at 1.00 p.m. and 4.30 p.m. for 15 min each time. Two test cages each containing ten 7-14 d old adult B. germanica (sex not reported) were introduced into each cupboard 24 h after introduction of the strips. After an exposure period of 4 h, the insects were transferred to clean cages and provided with food and water. After an observation period of 72 h, mortality was determined. The above procedure was repeated for 3rd nymphal stage larvae, and for both newly emerged P. americana and L. maderae adults and 3rd nymphal stage larvae. The above procedure was repeated 2, 3, 7, 21, 35, 49, 63, 77, 91 and 119 d after introduction of the strips. The author reported that four cages of insects were placed into untreated cupboards as negative (blank) controls. However, control results were only presented for 91 and 119 d. The mortality results are presented in Table 5.41. The results in Table 5.41 show that the test strip produced 100 % mortality of B. germanica and L. maderae adults, and 95 % mortality of P. americana adults for 91 d. The results for nymphs showed rates of 100 % (B. germanica), 85 % (P. americana) and 87.5 % (L. maderae) after 91 d. By the end of the 119 d test period the mortality of both adults and nymphs had declined to levels comparable with the control mortalities. The results provide evidence that the product (at an SVR of 4.49 g m-3) is effective against Dictyoptera for a period of 13 weeks (Unpublished, 1966).

Table 5.41 Percentage mortalities for B. germanica, P. americana and L. maderae adults and nymphs following exposure (4 h) to a

dichlorvos strip over a maximum period of 119 d

Species Percentage mortality/time after introduction of strips (d)

192

1 2 3 7 21 35 49 63 77 91 119 B. germanica (adults)

100 100 100 100 100 100 100 100 100 100 10

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 0 0

B. germanica (nymphs)

N/T 100 100 100 100 100 100 100 100 100 0

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 15 0

P. americana (adults)

100 100 100 100 100 100 N/T N/T N/T 95 2.5

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 0 0

P. americana (nymphs)

N/T 100 100 100 100 100 N/T N/T N/T 85 0

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 0 0

L. maderae (adults)

100 100 100 100 100 100 N/T N/T N/T 100 12.5

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 0 0

L. maderae (nymphs)

N/T 100 100 100 100 100 N/T N/T N/T 87.5 0

Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 0 0

N/T - Not tested. A study investigated the effect of a dichlorvos strip against B. germanica. No SVR could be calculated for the test strip. The study was conducted according to the method described in Appendix 5.1.3. In the study, the test strip was tested against 10 male B. germanica using the (LT)C test method. The insects were held in 150 x 100 mm diameter gauze cylinders. There was 1 test with 5 replicates. The results of the study are presented in Table 5.42.

Table 5.42 (LT)C and dichlorvos activity results (LT95) following exposure to 0.03 mg m-3 dichlorvos

Concentration of

dichlorvos (h) Species

(LT50)C (LT95)C

LT95 at 0.03 mg m-3 dichlorvos (h)

B. germanica 2.19 3.60 120 With reference to the calculation in Appendix 5, the (LT50)C results indicated that an exposure to 0.03 mg m-3 of 73 h was required to produce 50 % mortality of B. germanica. The (LT95)C results indicated that an exposure to 0.03 mg m-3 of 120 h was required to produce 95 % mortality of B. germanica. The latter result was supported by the LT95 value of 120 h.

193

Although no control results were reported, the results provided some evidence that a dichlorvos vapour concentration of 0.03 mg dichlorvos m-3 is effective against B. germanica after an exposure period of 5 d (Unpublished, 1964).

5.3.1.4 LEPIDOPTERA/COLEOPTERA

A study investigated the effectiveness of a commercial product designed for use in small spaces, against adults and larvae of T. bisselliella. The product contained 20.0 % w/w dichlorvos (3.5 g dichlorvos per unit), and the application rate was 1 unit per 6.0 m3 (equivalent to an SVR of 0.58 g m-3). In each test, a strip was suspended from the centre of the ceiling of a 0.78 m3 volume test cupboard, each cupboard being one of four placed inside a test chamber of approximately 47 m3 volume. The test chamber and cupboards were maintained at 25 oC and 60 % r.h., and were ventilated twice daily at 1.00 p.m. and 4.30 p.m. for 15 min each time. Two test cages each containing twenty 21 d old adult T. bisselliella (sex not reported) was introduced into each test cupboard 24 h after introduction of the strips. After an exposure period of 4 h, the insects were transferred to clean cages and provided with food and water. After an observation period of 24 h, the mortality rate was determined. The above procedure was repeated for 2-3 d old larvae. The test procedure was repeated 2, 3, 7, 21, 35, 49, 63, 77, 91 and 119 d after introduction of the strips. The author reported that four cages of insects were placed into untreated cupboards as negative (blank) controls. However, control results were only presented for 91 and 119 d. The mortality results are presented in Table 5.43.

Table 5.43 Percentage mortalities for T. bisselliella adults and larvae following exposure (4 h) to a dichlorvos strip over a maximum period of 119 d

Time after introduction of strips (d) Species

1 2 3 7 21 35 49 63 77 91 119 Adults 100 100 100 100 100 100 100 100 100 100 31.9Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 5 0

Larvae N/T 100 100 100 100 100 100 100 73.1 84.4 1.2 Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 5 0

N/T - Not tested. The results in Table 5.43 indicated that the test strip produced 100 % mortality against adults, and 84.4 % against larvae, after 91 d. By the end of the 119 d test period mortality had decreased markedly, particularly for larvae, which showed mortalities rate comparable with the control mortalities. The results provided evidence that the product (at an SVR of 4.49 g m-3) was effective against T. bisselliella for a period of 13 weeks (Unpublished, 1966). A study investigated the effect of a resin strip containing 20.0 % w/w dichlorvos against Ephestia elutella (Hubner) (tobacco moth). The test was conducted over a period of 115 d in a 28 m3 test chamber, into which a resin strip was introduced. The theoretical concentrations

194

of dichlorvos in the air were calculated using unpublished data provided to the study authors. The method of calculating the theoretical values was not reported, although the unpublished data are described as providing an indication of the relationship between the r.h. and the dichlorvos concentration in the air in an unventilated volume of 28 m3 at 21-26 oC. As well as calculating the theoretical concentrations, the actual air concentration was measured on two separate days during the study. A comparison of the measured and theoretical air concentrations is presented in Table 5.44.

Table 5.44 Theoretical and measured dichlorvos air concentrations on 2 days

Days after installation of

strip

Theoretical dichlorvos concentration

(mg m-3)

Measured dichlorvos concentration

(mg m-3) 27 0.046 0.04 54 0.025 0.05

The results in Table 5.44 show that when the dichlorvos air concentration was measured 27 d after installation of the strips, the measured concentration corresponded closely with the theoretical concentration. However, this was not the case after 54 d, where the measured concentration was twice the theoretical concentration. The authors offered no explanation for this discrepancy. In the biological test, the chamber was maintained at a temperature of 20 oC and mean r.h. of 68 %. The study report made no reference to any ventilation of the chamber during the test. At intervals over a period of 115 d after introduction of the strip, newly emerged E. elutella adults were placed in wire cages and the cages placed in the test chamber. The number of moths used each time varied between 15 and 110 adults. Following introduction of the moths, the rate of knockdown was observed at hourly intervals. At intervals over a period of 32 d after introduction of the strip, larvae in the migratory stage were placed in wire cages and the cages placed in the test chamber. The number of larvae used each time varied between 32 and 123. Following the relevant exposure period, the larvae were maintained at 25 oC and 70 % r.h.. After 1 week, the dead and living larvae were counted. The knockdown rates for adults are presented in Table 5.45, and the mortality rates for larvae in Table 5.46.

Table 5.45 Relationship between percentage knockdown of E. elutella, exposure time and dichlorvos concentration

% knockdown following installation of strips

(h) Days after installation of strip

Theoretical dichlorvos

concentration (mg m-3)

Number of moths

tested 1 2 3 4 5 6 7 8 < 16

195

8 0.099 25 0 0 * 68 95 * * * * 11 0.095 40 0 0 * 75 93 * * * * 21 0.055 33 0 9 61 * 100 * * * * 22 0.052 37 0 * 70 * 94 * * * * 24 0.050 55 0 * 31 * 90 * * * * 24 0.050 37 0 * * * 95 * * * * 26 0.047 46 0 * 40 * 90 * * * * 26 0.047 33 0 0 * 70 100 * * * * 55 0.024 40 0 0 0 25 70 * * * * 58 0.022 110 0 0 1 10 * 73 94 * * 62 0.021 62 0 0 * 15 * * 98 * * 76 0.018 56 0 0 0 4 14 36 66 76 *

114 0.010 25 0 0 0 0 * 0 0 8 100 115 0.010 15 0 0 0 0 * 0 0 13 100

* Not observed

The results for adults (Table 5.45) indicated that during the 115 d test period, the theoretical air concentration steadily decreased from a maximum of 0.099 mg m-3 (d 8) to a minimum of 0.01 mg m-3 (d 115). The results indicated that a theoretical concentration of 0.047-0.099 mg m-3 produced knockdown rates of 90-100 % after 5 h, with 0.01 mg m-3 producing a knockdown rate of 100 % after 8-16 h.

Table 5.46 Percentage mortalities of E. elutella larvae in relation to

exposure time and the theoretical dichlorvos air concentration

Days after installation of strip

Theoretical dichlorvos concentration (mg m-3)

Number of larvae tested

% mortality

4-9 0.095-0.120 101 36 11-18 0.064-0.085 123 59 25-32 0.040-0.049 32 53

The results for larvae (Table 5.46) indicated that a theoretical air concentration of 0.04-0.12 mg m-3 produced a maximum mortality rate of 59 % against larvae. The fact that the calculated theoretical air concentration for d 27 was almost identical to the concentration measured on d 27 (Table 5.44), suggests that the actual air concentrations produced by the strip during the first 27 d were almost identical to the theoretical values i.e. 0.047-0.099 mg m-3 (Table 5.45). This, taken together with the biological test results (Tables 5.45 and 5.46), therefore provides some evidence that under conditions simulating use in the field an actual air concentration of 0.047-0.099 mg m-3 will produce high knockdown rates against E. elutella adults within 5 h for up to 26 d, and that a concentration of 0.04 - 0.12 mg m-3 will produce a maximum of 59 % mortality against larvae. The results as a whole provided some evidence that the product produced these concentrations for 26-32 d.

196

As the calculated theoretical concentration for d 54 was twice the actual concentration measured on the same day, and as the authors presented no explanation for this discrepancy, it is considered by HSE that theoretical air concentrations < 0.04 mg m-3 cannot be considered as a reliable guide to the actual concentration produced by the strip. As a consequence, no conclusion can be reached as to the effectiveness of the strip after approximately 30 d. Given the above, and although no control data were reported, the results provided some evidence that a strip producing an air concentration of 0.047-0.099 mg dichlorvos m-3, was effective for approximately 30 d against E. elutella adults, but not against larvae (Schulten & Kuyken, 1966). A study was conducted to assess the efficacy of a dichlorvos slow release non-controllable cassette against T. bisselliella. The test product, which contained 80.0 % w/w dichlorvos, had an application rate of 1 unit (2.0 g dichlorvos) per 2.5 - 4.0 m3 air. The SVR for the product was therefore 0.5 - 0.8 g m-3. The study was conducted in two stages. In the first stage, ten adult moths and ten 2nd to 4th instar larvae were exposed to the product in each of three experimental wardrobes of 0.525 m3 volume, and containing two towels simulating the presence of clothing. Both adults and larvae were placed into two clear 30 ml plastic tubes (five in each) containing a small amount of larval medium, and with a fine mesh covering each end. One tube was then fastened inside the fold of one towel suspended from a coat hanger at the top of the wardrobe, and the second tube fastened inside the fold of the second towel placed at the bottom of the wardrobe. The towels were left undisturbed in the test wardrobes for a period of 24 h, after which one test cassette was placed approximately two-thirds of the way up the far wall of each wardrobe. Two further wardrobes were set up in a manner identical to that described above, but without the introduction of the product (i.e. negative controls). The wardrobes were ventilated during the test, although no data were provided on temperature and r.h.. The numbers of living and dead adults and larvae at the top and the bottom of each test wardrobe were recorded 48 h and 72 h after introduction of the product, and percentage mortalities were calculated. This was also undertaken for the controls. The results are presented in Figures 5.15 to 5.18.

197

Figure 5.15 Percentage mortalities of T. bisselliella adults/larvae at top of wardrobes 48 h after introduction of slow release dichlorvos strips

Figure 5.16 Percentage mortalities of T. bisselliella adults/larvae at bottom of

wardrobes 48 h after introduction of slow release dichlorvos strips The results for 48 h exposure (Figures 5.15 and 5.16) indicated that the unit performed better when the adults and larvae were at the bottom of the wardrobe. The mortalities were, however, low, with a maximum mortality for adults of 60 % and for larvae of 14.3 %.

1 2 3 4

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

)

Larvae - product (7.1 %)Larvae - control (10.0 %)Adults - product (33.3 %)Adults - control (30.0 %)

1 2 3 4

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

)

Larvae - product (14.3 %)Larvae - control (0 %)Adults - product (60.0 %)Adults - control (10.0 %)

198

Figure 5.17 Percentage mortalities of T. bisselliella adults/larvae at top of

wardrobes 72 h after introduction of slow release dichlorvos strips

Figure 5.18 Percentage mortalities of T. bisselliella adults/larvae at bottom of wardrobes 72 h after introduction of slow release dichlorvos strips

1 2 3 4

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

)

Larvae - product (28.6 %)Larvae - control (10.0 %)Adults - product (46.7 %)Adults - control (40.0 %)

1 2 3 4

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

)

Larvae - product (0 %)Larvae - control (0 %)Adults - product (66.7 %)Adults - control (10.0 %)

199

For 72 h exposure (Figures 5.17 and 5.18) the unit performed better against adults at the bottom of the wardrobe, and better against larvae when the larvae were at the top. However, the mortalities were still low, with a maximum mortality for adults of 66.7 % and for larvae of 28.6 %. The experimental procedure followed in the second stage was identical to that in the first, except that, in addition to testing the product a second time, a second ‘version’ of the product, containing a 50 % ‘overage’ of active ingredient, was also tested. In order to compare the test product with the performance of other, commercially available dichlorvos cassette units, two 'competitor sample' units were tested, with each of three units of each brand introduced into a separate wardrobe. No information was provided on these 'competitor' units. The results are presented in Figures 5.19 to 5.26. For 24 h exposure at the top of the wardrobe (Figures 5.19 and 5.20) the product + 50 % overage performed considerably less effectively than the competitor samples, showing maximum mortalities of 53.3 % and 13.3 % for adults and larvae, respectively. For 24 h exposure at the bottom of the wardrobe (Figures 5.21 and 5.22) the product + 50 % overage produced maximum mortalities of 66.7 % for adults and 13.3 % for larvae. These rates were comparable to those recorded for the two competitor samples, and were higher than the controls.

Figure 5.19 Percentage mortalities for T. bisselliella larvae situated at the top of test wardrobes - 24 h after introduction of cassettes

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (0 %)50 % overage (13.3 %)Competitor sample 1 (0 %)Competitor sample 2 (10.0 %)Control (6.7 %)

200

Figure 5.20 Percentage mortalities for T. bisselliella adults situated

at the top of test wardrobes - 24 h after introduction of cassettes

Figure 5.21 Percentage mortalities for T. bisselliella larvae situated at the bottom of test wardrobes - 24 h after introduction of cassettes

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (46.7 %)50 % overage (53.3 %)Competitor sample 1 (80.0 %)Competitor sample 2 (80.0 %)Control (0 %)

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (0 %)50 % overage (13.3 %)Competitor sample 1 (0 %)Competitor sample 2 (10.0 %)Control (6.7 %)

201

Figure 5.22 Percentage mortalities for T. bisselliella adults situated at the bottom of test wardrobes - 24 h after introduction of cassettes

Figure 5.23 Percentage mortalities for T. bisselliella larvae situated

at the top of test wardrobes - 48 h after introduction of cassettes

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (6.7 %)50 % overage (66.7 %)Competitor sample 1 (60.0 %)Competitor sample 2 (70.0 %)Control (20.0 %)

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (7.7 %)50 % overage (23.1 %)Competitor sample 1 (30.0 %)Competitor sample 2 (60.0 %)Control (13.3 %)

202

Figure 5.24 Percentage mortalities for T. bisselliella adults situated at the top of test wardrobes - 48 h after introduction of cassettes

For 48 h exposure of adults and larvae at the top of the wardrobe (Figures 5.23 and 5.24) the test units again performed better against adults, with the product giving a mortality against adults of 86.7 % (7.7 % for larvae), and the product + 50 % overage giving a mortality of 100 % (23.1 % for larvae). These rates were comparable to those recorded for the two competitor samples, and were higher than the controls.

Figure 5.25 Percentage mortalities for T. bisselliella larvae situated at the bottom of test wardrobes - 48 h after introduction of cassettes

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (86.7 %)50 % overage (100.0 %)Competitor sample 1 (80.0 %)Competitor sample 2 (100.0 %)Control (13.3 %)

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (0 %)50 % overage (35.7 %)Competitor sample 1 (40.0 %)Competitor sample 2 (44.4 %)Control (6.7 %)

203

Figure 5.26 Percentage mortalities for T. bisselliella adults situated at the bottom of test wardrobes - 48 h after introduction of cassettes

For 48 h exposure of adults and larvae at the bottom of the wardrobe (Figures 5.25 and 5.26) the test units again performed better against adults, with the product giving a mortality against adults of 86.7 % (0 % for larvae), and the product + 50 % overage giving a mortality of 100 % (35.7 % for larvae). Again, these rates were comparable to those recorded for the two competitor samples, and were higher than the controls. The results indicated that when tested under conditions simulating practical use at the product SVR of 0.5 - 0.8 g m-3, the product was not consistently effective against T. bisselliella. When the amount of dichlorvos was increased by 50 % to give an SVR of 0.75-1.2 g m-3, the product was effective against adults, but not against larvae (Unpublished, 1997a). A study was conducted to assess the efficacy of a dichlorvos slow release non-controllable cassette product against T. bisselliella. The study was based on test method NL-7200 (see Appendix 5.1.1) in which use of the product in domestic wardrobes was simulated by placing test strips into 0.19 m3 volume wooden boxes, exposing T. bisselliella adults to the strips, and recording knockdown and mortality rates at intervals over a period of 6 months. The test product, which contained 20.0 % w/w dichlorvos, had an application rate of 1 unit (2.1 g dichlorvos) per 2.6 m3 air, giving an SVR of 0.8 g m-3. Therefore, given the volume of the test boxes, the dimensions of each strip were proportionally reduced prior to the test in order to maintain the SVR. The NL-7200 test procedure stipulates the use of one box, as a negative (blank) control. However, in this particular study, two control boxes were used. The results indicated that 100 % knockdown was achieved over a period of 8 h at the beginning of the test period, with this performance maintained throughout the 6-month period of the test. The control results indicated zero mortality throughout the test period. The results also indicated that 100 % mortality was achieved over a period of 3 d at the beginning of the test period, with this performance maintained throughout the 6-month period of the test. The control results for mortality are presented in Figure 5.27.

1 2 3 4 5

0

10

20

30

40

50

60

70

80

90

100

Mor

talit

y (%

) Product (86.7 %)50 % overage (100.0 %)Competitor sample 1 (90.0 %)Competitor sample 2 (100.0 %)Control (26.7 %)

204

Figure 5.27 Mean percentage mortality rates 1-3 d after

introduction of T. bisselliella adults to control boxes The author reported that no eggs or larvae were found after any of the exposure tests, and observed that this could be explained by the 100 % knockdown and mortality obtained for adults. The author also reported that viable eggs were found after the control tests, and stated that as the females die after laying their eggs, the mortalities found in the controls were as expected. The results provided evidence that under conditions simulating use in a small unventilated space, the product (at an SVR of 0.8 g m-3) was effective against T. bisselliella for up to 6 months, and that during this period produced 100 % knockdown within 8 h and 100 % mortality within 3 d (Unpublished, 1997b). A study investigated the effect of slow release strips against one Lepidopterid and two Coleopterid species. This study involved tests to determine the toxic effect, and also included feeding tests reported to conform to test standards SNV 95901 and SNV 95902 (see Appendix 5.1.2), and to the existing protocol of the Swiss Federal Institute (EMPA). In the feeding tests, fifteen T. bisselliella, A. piceus or Anthrenus vorax (carpet beetle) larvae were introduced into two test rooms and two cupboards, to which 3.5 cm2 unbleached flannel cloth was added. The cloth had been treated with a bait consisting of a 15 mm diameter patch of 1:10 aqueous yeast solution. After the strips had been hung in the rooms/cupboards, three feeding tests were carried out at 0, 5, 10 and 14 weeks after the introduction of the strips. The week 10 test with T. bisselliella was not performed due to the unavailability of larvae. Each test lasted for 2 weeks and was conducted in triplicate (moth larvae) or duplicate (beetle

1 2 3

Time following introduction of insects (days)

0

20

40

60

80

100

Mea

n m

orta

lity

rate

(%)

6 months3 months2 months1 monthInitial

205

larvae). A control room/cupboard was employed with no dichlorvos strips. The cloth was weighed before and after the feeding tests to determine loss by feeding. Surviving larvae were counted after each test and mortality rates calculated. The application rates and environmental conditions of the tests are presented in Table 5.47.

Table 5.47 Experimental conditions of the feeding tests conducted

on the test insects exposed to dichlorvos strips

Size of test room (m3)

Application rate (g m-3)

R.H. (%)

Temperature (°C)

Room 1: 23.2 1.53 30 - 47 13 - 18 Room 2 : 9.9 0.76 18 - 38 20 - 22.5 Cupboard 3: 0.668 1.60 30 - 47 13 - 18 Cupboard 4: 0.662 0.80 30 - 47 13 - 18

The results of the feeding tests on T. bisselliella are presented in Table 5.48. The 1.53 and 1.60 g m-3 rates are higher than that in current approvals (0.80 g m-3). Consequently, these results will not be discussed. Table 5.48 indicated that the 0.80 g m-3 rate had 2 surviving larvae after 14 weeks in one of the replicates, but little weight loss of the material. The 0.76 g m-3 application had little effect on the feeding or mortality of the larvae. The maximum weight loss occurred in the controls and all but one of the larvae survived. Table 5.49 presents the results of feeding tests on A. piceus larvae. Table 5.49 indicated that the 0.76 and 0.80 g m-3 application rates did not prevent feeding (51 - 190 % weight loss), and very low mortality of the larvae was recorded. All larvae in the control group survived with 100 % feeding loss of the sample material.

Table 5.48 Results of feeding tests on T. bisselliella larvae

Application

rate (g m-3)

Time of introduction of larvae (weeks)

Number of surviving larvae in each

replicate (N = 15)

Mean weight loss of cloth

(%)

Subjective level of control

0.76 0 5

14

13/15/15/ 13/14/14 14/14/15

59 109 102

‘weak’ ‘weak’ ‘weak’

0.80 0 5

14

0/0/0 0/0/0 2/0/0

10 2.3 11

‘strong’ ‘very strong’

‘strong’ 1.53 0

5 14

15/9/0 0/0/0 0/0/0

33 1.1 2.1

‘middle’ ‘very strong’ ‘very strong’

1.60 0 5

14

1/0/0 0/0/0 0/0/0

16 0.3 2.1

‘fairly strong’ ‘very strong’ ‘very strong’

Control 0 5

14

15/15/15 15/15/14 15/15/15

100 100 100

N/A N/A N/A

Table 5.49 Results of feeding tests on A. piceus larvae

206

Application rate

(g m-3)

Time of introduction of larvae (weeks)

Number of surviving larvae in each

replicate (N = 15)

Mean weight loss (%)

Subjective level of control

0.76 0 5

10 14

15/15 15/15 15/15 15/15

59 190 170 130

‘weak’ ‘weak’ ‘weak’ ‘weak’

0.80 0 5

10 14

14/15 15/15 15/15 15/15

90 51 55 82

‘weak’ ‘weak’ ‘weak’ ‘weak’

1.53 0 5

10 14

14/15 11/10 11/13 13/12

68 21 15 17

‘weak’ ‘strong’ ‘strong’ ‘strong’

1.60 0 5

10 14

15/14 6/10

12/10 13/11

70 16 10 8.5

‘weak’ ‘strong’ ‘strong’ ‘strong’

Control 0 5

10 14

15/15 15/15 15/15 15/15

100 100 100 100

N/A N/A N/A N/A

Table 5.50 presents the results of feeding tests on A. vorax larvae. Table 5.50 indicated that in general the 0.76 g m-3 application rate did not prevent > 100 % weight loss of the material, and mortality was very low. The 0.80 g m-3 application rate generally gave ‘medium strong’ to ‘strong’ control, although over 50 % of the larvae survived in all the tests at this application rate. All larvae in the control group survived with 100 % feeding loss of the sample material. In the feeding tests, the 0.80 g m-3 application rate achieved high mortality up to 14 weeks, with minimal feeding, although two larvae survived in one of the replicates. In addition, both the 0.76 and 0.80 g m-3 application rates allowed 100 % survival and heavy feeding at all the introduction times. Whilst noting that the descriptors stated by the study author for the level of control achieved, and reproduced in Tables 5.48-5.50, are subjective and inconsistent, HSE considers that the data indicated that under the conditions of the study, none of the application rates were sufficient to prevent feeding by A. vorax and A. piceus. Furthermore, the application rates allowed survival of the larvae at all the introduction times.

Table 5.50 Results of feeding tests on A. vorax larvae

207

Application rate

(g m-3)

Time of introduction

of larvae (weeks)

Number of surviving larvae in each replicate

(N = 15)

Mean weight loss (%)

Subjective level of control

0.76 0 5 10 14

15/14 15/14 15/15 15/15

87 330 162 413

‘weak’ ‘weak’ ‘weak’ ‘weak’

0.80 0 5 10 14

14/10 15/14 15/15 12/14

51 17 31 16

‘weak’ ‘strong’

‘medium’ ‘strong’ “ t ” 1.53 0

5 10 14

15/15 10/14 12/14 15/9

49 11 7 11

‘weak’ ‘strong’

‘very strong’ ‘strong’

1.60 0 5 10 14

10/14 0/0 6/3 2/1

58 10 3.1 5.9

‘weak’ ‘strong’

‘very strong’ ‘very strong’

Control 0 5 10 14

15/15 15/15 15/15 15/15

100 100 100 100

N/A N/A N/A N/A

N/A Not applicable Toxicity tests were conducted on 3 - 4 d old adult A. piceus and A. vorax, and 1 - 2 d old T. bisselliella. Tests were also carried out on 21 d old larvae of the above species. Three groups of ten adult insects were placed in plexi glass rings covered with fine wire mesh (8 cm diameter x 2 cm high), and in addition, three groups of ten larvae were placed in Petri dishes covered in fine muslin. The insects were exposed to the same application rates in the same test rooms/cupboards as in the feeding tests. Five sets of tests were conducted, with the insects introduced 0, 4, 8, 12 and 16 weeks after the strips were hung in the rooms. Observations for mortality were made after 15 and 30 min, 1, 2, 4 and 8 h, and 1, 2, 4, 8, 16, and 28 d. KD50 values were calculated by the probit method, and 96.7 % mortality was determined by noting the period after which 29 out of 30 insects were killed. The results of the KD50 tests are presented in Table 5.51.

Table 5.51 KD50 values for 3 species of adult insects exposed to dichlorvos

strips in rooms and cupboards

208

KD 50 values (mean of 3 replicates) Application

rate (g m-3)Week of

introduction of insects

T. bisselliella A. piceus A. vorax

0.76 0 4 8 12

2 h 36 min 3 h 6 min

8 h 9 h

10 h 15 min1 d 16 h 5 d 19 h 9 d 4 h

1 d 2 h 3 d 4 h 10 d 9 h 208 d 7 h

0.80 0 4 8 12 16

2 h 36 min 2 h 9 min 2 h 48 min

3 h 6 h 30 min

6 h 42 min 3 h 24 min

4 h 6 h 15 min 17 h 30 min

9 h 48 min 6 h 36 min 6 h 12 min 17 h 42 min

1 d 16 h 1.53 0

4 8 12 16

2 h 48 min 1 h 24 min 1 h 48 min 2 h 24 min 2 h 48 min

10 h 45 min4 h 24 min 4 h 45 min 5 h 6 min 4 h 36 min

16 h 45 min 5 h 6 min 3 h 48 min 5 h 30 min

5 h 1.60 0

4 8 12 16

1 h 36 min 1 h 24 min 1 h 18 min 1 h 48 min 2 h 48 min

5 h 15 min 2 h 48 min 1 h 45 min 2 h 48 min 3 h 30 min

11 h 3 h

2 h 15 min 3 h 30 min 4 h 9 min

Control 0 4 8 12 16

2 d 19 h 3 d 14 h 3 d 7 h

6 d 5 d 14 h

10 d 20 d

30 d 19 h 18 d 7 h

8 d

52 d 2 h 500 d

No data 56 d 16 h No data

Table 5.51 indicated T. bisselliella to have been more susceptible than the two species of beetle, with KD50 values ranging from 2 h 9 min (0.80 g m-3 - insects introduced at week 4) to 9 h (0.76 g m-3 - insects introduced at week 12). Values for A. piceus (the most susceptible beetle species tested) ranged from 3 h 24 min (0.80 g m-3 - insects introduced at week 4) to 9 d 4 h (0.76 g m-3 - insects introduced at week 12). The data for the two lower application rates did show an apparent trend of increasing KD50 values with time, although this was not so pronounced with the 0.80 g m-3 application rate. All the control KD50 values were much higher than the test data. The absence of the week 16 data from the 0.76 g m-3 test was not explained. Table 5.52 presents the time period to achieve KD50 for the larvae of the three species of insect tested.

Table 5.52 Time taken to achieve KD50 for 3 species of insect larvae exposed to

dichlorvos strips in rooms and cupboards

209

KD 50 values (mean of 3 replicates) Application rate (g m-3)

Week of introduction

of insects T. bisselliella A. piceus A. vorax

0.76 0 4 8 12

1 d 4 h 2 d

No data 11d 9 h

22 d 2 h No data

44 d 14 h 51 d 2 h

10 d 14 h 36 d 2 h 21 d 7 h

28 d

0.80 0 4 8 12 16

13 h 45 min 10 h 30 min

No data 2 d 12 h 2 d 19 h

3 d 2 h 5 d 14 h

10 d 13 d 21 h 23 d 18 h

2 d 19 h 5 d 14 h 7 d 16 h

16 d 16 d 21 h

1.53 0 4 8 12 16

13 h 45 min 5 h

No data 13 h 45min

10 h

3 d 19 h 3 d 3 d

3 d 9 h 5 h 4 h

2 d 19 h 3 d

4 d 4 h 5 d 16 h

3 d

1.60 0 4 8 12 16

13 h 45 min 3 h 15 min

No data 13 h 45 min 10 h 30 min

3 d 2 h 2 d 7 h 1 d 12 h 2 d 19 h 2 d 21 h

2 d 16 h 1 d 16 h 2 d 16 h 2 d 12 h 2 d 9 h

Control 0 4 8 12 16

28 d 125 d

No data No data

No data�

No data No data No data No data No data

32 d 12h 208 d

No data 152 d 2 h No data

Table 5.52 indicated T. bisselliella larvae to have been more susceptible than the beetle larvae with KD50 values ranging from 10 h 30 min (0.80 g m-3 - larvae introduced at week 4) to 11 d 9 h (0.76 g m-3 - larvae introduced at week 12). In addition, there was an apparent trend of increasing values with time seen in all the species at the 0.76 and 0.80 g m-3 application rates. The absence of week 16 data from the 0.76 g m-3 test was not explained. There was no apparent dose response relationship between the application rates and the observed knockdown. All the available control values were higher than the test data. However, the gaps in the data were not explained. Table 5.53 presents the time period required to achieve 96.7 % mortality (29 out of 30 individuals) of the three adult insect species used in the test.

Table 5.53 Time taken to achieve 96.7 % mortality for 3 species of adult insects exposed to dichlorvos strips in rooms and cupboards

210

Time to achieve 96.7% mortality (mean of 3 replicates)

Application rate (g m-3)

Week of introduction of

insects T. bisselliella A. piceus A. vorax 0.76 0

4 8 12

1 d 8 d 8 d 8 d

16 d 28 d 28 d 28 d

8 d 28 d > 28 d > 28 d

0.80 0 4 8 12 16

8 h 4 d 8 d 8 d 8 d

4 d 8 d 16 d 16 d 16 d

4 d 4 d 16 d 16 d 16 d

1.53 0 4 8 12 16

8 h 4 d 8 d 8 d 8 d

4 d 4 d 16 d 16 d 16 d

4 d 2 d 2 d 16 d 8 d

1.60 0 4 8 12 16

8 d 4 d 8 d 8 d 8 d

4 d 8 d 16 d 16 d 8 d

4 d 2 d 4 d 4 d 4 d

Control 0 4 8 12 16

8 d 16 d 16 d 16 d 16 d

> 28 d > 28 d > 28 d > 28 d > 28 d

> 28 d > 28 d > 28 d > 28 d > 28 d

Table 5.53 indicated T. bisselliella to have been more susceptible than the beetle species with 96.7 % mortality occurring between 1 and 8 d at both the 0.76 and 0.8 g m-3 application rates. The longest time to obtain 96.7 % mortality (>28 d) was observed with A. vorax exposed to the 0.76 g m-3 application rate, and introduced at weeks 8 and 12. The times obtained for the controls were broadly consistent with the normal life span of the test species. Table 5.54 presents the time taken to achieve 96.7 % mortality for the three species of insect larvae.

Table 5.54 Time taken to achieve 96.7 % mortality for 3 species of insect larvae exposed to dichlorvos strips in rooms and cupboards

211

Time to achieve 96.7% mortality (mean of 3 replicates)

Application rate (g m-3)

Week of introduction of

insects T. bisselliella A. piceus A. vorax 0.76 0

4 8 12

4 d 16 d

No data 28 d

> 28 d > 28 d > 28 d > 28 d

> 28 d > 28 d > 28 d > 28 d

0.80 0 4 8 12 16

2 d 2 d

No data 16 d 8 d

16 d 28 d

> 28 d > 28 d > 28 d

16 d 28 d 28 d

> 28 d > 28 d

1.53 0 4 8 12 16

2 d 2 d

No data 4 d 4 d

16 d 16 d

> 28 d > 28 d 28 d

8 d 16 d 16 d 16 d 16 d

1.60 0 4 8 12 16

2 d 2 d

No data 4 d 4 d

16 d 16 d 28 d

> 28 d 28 d

16 d 16 d 8 d 8 d 16d

Control 0 4 8 12 16

> 28 d > 28 d > 28 d > 28 d > 28 d

> 28 d > 28 d > 28 d > 28 d > 28 d

> 28 d > 28 d > 28 d > 28 d > 28 d

Table 5.54 indicated T. bisselliella to have been the most susceptible larvae with 96.7 % mortality achieved between 2 d and 28 d. For all the species tested the longest times were observed with the 0.76 g m-3 application rate. There was no apparent dose response relationship between the application rates and the observed mortality. All the control values were higher than, or comparable with, the test data. In the toxicity tests, no dose response relationship between the application rates and the observed mortality was discerned. There was limited evidence to suggest a trend of increasing mortality or KD50 values with the time of introduction into the rooms/cupboards. In general, the adult insects were more susceptible than larvae to the dichlorvos strips, and T. bisselliella was more susceptible than the two beetle species. For the 0.76 and 0.80 g m-3 application rates, the values were 28 d or longer for both beetle species. For all of the species tested the time taken to achieve mortality generally increased after 4 weeks. The results from the toxicity tests provided evidence that when tested under conditions simulating practical use, and at SVR’s of 0.76 and 0.8 g m-3, the test strips were effective against T. bisselliella adults within 8 d, but required > 28 d against adult Coleoptera. The

212

results also indicated that the strips were less effective against larvae. The results from the feeding tests provided evidence that the strips, when used at SVR’s of 0.76 and 0.8 g m-3, did not prevent feeding by any of the test species (Unpublished, 1967). A study investigated the effect of a dichlorvos strip against T. confusum, L. serricorne and Sitophilus granarius (granary weevil). No SVR can be calculated for the test strip. The study was conducted according to the method described in Appendix 5.1.3. In the study, the test strip was tested using each of the two test methods (50 to 100 mixed sex for T. confusum and S. granarius, 15 to 25 mixed sex for L. serricorne in both tests). T. confusum and S. granarius were held in open Petri dishes. L. serricorne was held in a gauze covered plate. For the LCT test, there was 1 test with one replicate consisting of 3 exposure times. For the (LT)C test there was one test consisting of 2 replicates. The results of the study are presented in Table 5.55.

Table 5.55 LCT, (LT)C and dichlorvos activity results (LT95)

following exposure to 0.03 mg m-3 dichlorvos

Concentration of dichlorvos (h) Species LCT50 LCT95 (LT50)C (LT95)C

LT95 at 0.03 mg m-3 dichlorvos (h)

T. confusum 0.79 2.02 0.865 2.24 74.0 L. serricorne 0.36* 1.08 0.94 2.52* 84.0 S. granarius 1.01 3.6* 1.59 5.05* 168.0

* By extrapolation With reference to the calculation in Appendix 5.1.3, the LCT results in Table 5.55 indicated that for the test concentration of 0.03 mg m-3, the exposure time required to give 95 % mortality after 24 h was 67 h 20 min (T. confusum), 36 h (L. serricorne) and 120 h (S. granarius). The LT95 results indicated that 95 % mortality was produced after an exposure to 0.03 mg m-3 of 74 h (T. confusum), 84 h (L. serricorne) and 168 h (S. granarius). Although no control results were reported, the results provided some evidence that after an exposure of approximately 70-170 h a dichlorvos vapour concentration of 0.03 mg m-3 dichlorvos was effective against Coleoptera (Unpublished, 1964). A study investigated the effect of dichlorvos vapour against larvae of Anthrenus flavipes (LeConte) (furniture carpet beetle). The vapour was produced from wax sticks (no dimensions reported) impregnated with dichlorvos. The wax sticks were placed in wire baskets suspended 1.83 m above the floor near the rear of a 42.5 m3 closed chamber. The chamber was maintained at 26.7 oC + 1 oC and 75 + 5 % r.h. The concentration of dichlorvos in the air was monitored throughout the duration of the test by the removal of samples from the geometric centre of the chamber. The samples were either 113.3 or 226.6 litres in volume, and were removed with a 1.83 m long by 9.5 mm wide stainless steel tube. The test larvae (2-3 months old) were exposed in 20 wire mesh cages (6.35 cm x 1.9 cm) suspended in the area from which the air samples were taken. Enough caged larvae were introduced to permit the removal of 4 samples of 10 insects each after 4, 24 and 32 h exposure. At the end of each exposure period the larvae were removed and placed in Petri

213

dishes. The extent of mortality was then assessed. The larvae were then kept in the Petri dishes in a room at 26.7 oC + 1 oC and 60 + 5 % r.h. for 28 d, with the mortality assessed at 14 and 28 d. A control test was conducted in which the larvae were placed in an identical test chamber under identical conditions, except without the presence of the impregnated stick. The results are presented in Table 5.56.

Table 5.56 Mortality rates for larvae of A. flavipes immediately following exposure to vapour from a dichlorvos impregnated wax stick and 14 and 28 d post exposure

Percentage mortality after various periods of time post exposure (d)

Exposure period

(h)

Measured dichlorvos air concentration

(mg m-3) 0 14 28

4 0.46 0 3 3

24 0.88 3 85 88

32 0.60 15 100 100

32 Control 0 0 0

The results in Table 5.56 indicated that the air concentration of dichlorvos was maintained at broadly the same level throughout the 32 h exposure period. The results indicated that when the larvae were exposed for 32 h to an air concentration of 0.60 mg dichlorvos m-3 very low mortality (15 %) was observed initially. However, the mortality rate had increased to 100 % 14 d after removal from the chamber. The control results indicated zero mortality for larvae left in the control chamber for 32 h and then left for a further 28 d period. The results provided evidence that an air concentration of 0.60 mg dichlorvos m-3 produced by a dichlorvos impregnated wax stick, was only effective against A. flavipes larvae 14 d after the end of a 32 h exposure period (Boles et al., 1974).

5.3.1.5 HEMIPTERA

A study was conducted on C. lectularius, in which containers of 20 insects of mixed age and sex were exposed to dichlorvos strips. Full details of the treatment area were not given, but will have again been within the general limits described by Mankowska & Goszczynska in the same study against Dictyoptera (see Section 5.3.1.3). Mortality was recorded daily until 100 % kill was achieved, and the study was repeated 13 and 87 d after treatment. The results are shown in Table 5.57.

Table 5.57 Effect of dichlorvos strips on C. lectularius

214

Time after application (d)

Exposure Time (d)

Mortality (%)

1 80 0 2 100 13 1 100

1 60 2 90 87 3 100

The results in Table 5.57 indicated that 100 % mortality of C. lectularius was achieved within 3 d around 3 months after initial application of the strip (Mankowska & Goszczynska, 1969). A study investigated the effect of dichlorvos vapour on C. lectularius. Two sets of tests were conducted using either resinous granules impregnated with dichlorvos (20 % w/w), or a wax stick containing 25 % w/w dichlorvos. The first series of tests was conducted in sealed 44 m3 chambers. In each test, 50 or 100 C. lectularius (equal numbers of each sex) starved for 1-2 d were placed in plastic or glass tubes capped with cloth. A piece of filter paper placed lengthways in each tube provided a resting surface for the specimens. The insects were then exposed to either 0.25, 1.04, 1.7 or 7.4 mg dichlorvos m-3 for exposure periods varying from 2-24 h. After exposure, the insects in the glass tubes were not transferred, but the insects in the plastic tubes were transferred to plastic vials or quart jars containing filter paper and capped with cloth. The mean temperature of the chambers was 29.7 + 1.7 oC (0.25 and 1.04 mg m-3 concentrations) and 26.6 + 1.7 oC (1.7 and 7.4 mg m-3 concentrations). The r.h. was not reported. A second series of tests was conducted to determine the effect of dichlorvos vapour on insects hiding in small containers. The containers used were paper bags made of kraft paper. After the insects were placed in the bags, the tops of the bags were twisted to close them. The insects in the containers (50 or 100-equal numbers of each sex) were then exposed for different periods of time to either 0.25 or 1.04 mg dichlorvos m-3. The results for percentage knockdown and 24 and 48 h mortalities (test 1) are presented in Table 5.58, with the results for test 2 presented in Table 5.59. No control data were reported.

Table 5.58 Knockdown and mortality rates for C. lectularius exposed for varying periods of time to varying dichlorvos air concentrations (test 1)

215

Percentage mortality after indicated post-exposure

time

Dichlorvos concentration

(mg m-3)

Exposure period (h)

Percentage knockdown at the end of the exposure

period 24 h 48 h

0.25 6 8.5

64 100

94 100

94 100

1.04

2 4 6

4 92 100 100

66 100 100 100

66 100 100 100

1.70

2 3 4 6

98 99 99 100 100

100 100 100 100 100

100 100 100 100 100

7.40 2 100 100 100

Table 5.59 Knockdown and mortality rates for contained test insects exposed for varying periods of time to varying dichlorvos air concentrations (test 2)

Percentage mortality after indicated post-

exposure time

Dichlorvos concentration

(mg m-3)

Exposure container

Exposure period

(h)

Percentage knockdown at the end of the exposure

period 24 h 48 h

0.25 Bag 24 100 100 100 1.04 Bag 24 100 100 100

The results in Table 5.58 indicated that an 8.5 h exposure to 0.25 mg m-3 and a 6 h exposure to 1.04 mg m-3, both produced 100 % knockdown and kill. The results in Table 5.59 indicated that a 24 h exposure to 0.25 mg m-3 produced 100 % knockdown and kill. Although no control data were reported, the results provided some evidence that under the conditions of the tests, an air concentration of 0.25 mg dichlorvos m-3 was effective against C. lectularius (Smittle & Burden, 1965).

5.3.1.6 FORMICOID HYMENOPTERA

A study investigated the effect against Monomorium pharaonis L. (pharaoh’s ant), of a commercial strip product designed for use in small spaces. The product contained 20.0 % w/w dichlorvos (3.5 g dichlorvos per unit), and the recommended application rate for the product was 1 unit per 6.0 m3 (equivalent to an SVR of 0.58 g m-3). In each test, a strip was suspended from the centre of the ceiling of a 0.78 m3 volume test cupboard, each cupboard being one of four placed inside a test chamber of approximately 47 m3 volume. The test chamber and cupboards were maintained at 25 oC and 60 % r.h., and

216

were ventilated twice daily at 1.00 p.m. and 4.30 p.m. for 15 min each time. Two test cages each containing thirty 21 d old adult M. pharaonis (sex not reported) were introduced into each test cupboard 24 h after introduction of the strips. After an exposure period of 4 h, the insects were transferred to clean cages and provided with food and water. After an observation period of 4 h, the mortality rate was determined. The test procedure was repeated 2, 3, 7, 21, 35, 49, 63, 77, 91 and 119 d after introduction of the strips. The author reported that four cages of insects were placed into untreated cupboards as negative (blank) controls. However, control results were only presented for 91 and 119 d. The mortality results are presented in Table 5.60.

Table 5.60 Percentage mortalities for M. pharaonis adults following exposure (4 h) to a dichlorvos strip over a maximum period of 119 d

Time after introduction of strips (d)

1 2 3 7 21 35 49 63 77 91 119 Adults 100 100 100 100 100 100 100 100 100 100 87.9Negative control

N/T N/T N/T N/T N/T N/T N/T N/T N/T 6.3 0

N/T - Not tested. The results in Table 5.60 indicated that the test strip produced 100 % mortality after 91 d. By the end of the 119 d test period the mortality had decreased to 87.9 %. The results provided evidence that the test product, when used at an SVR of 4.49 g m-3, was effective against M. pharaonis for a period of 13 weeks (Unpublished, 1966).

5.3.1.7 HYMENOPTERA

A study investigated the effect of a dichlorvos strip against Vespula germanica (German wasp). Neither the amount of dichlorvos on the test strip nor the application rate were reported. Consequently, no SVR can be calculated for the test strip. The study was conducted according to the method described in Appendix 5.1.3. In the study, the test strip was tested against 10 worker V. germanica using the (LT)C test method. The insects were held in 150 x 100 mm diameter gauze cylinders. There was 1 test with 1 replicate. The results of the study are presented in Table 5.61.

Table 5.61 (LT)C and dichlorvos activity results (LT95)

following exposure of V. germanica to 0.03 mg m-3 dichlorvos

Concentration of dichlorvos (h)

Species

(LT50)C (LT95)C

LT95 at 0.03 mg m-3 dichlorvos

(h) V. germanica 0.0825 0.095 3.2

With reference to the calculation in Appendix 5.1.3, the (LT95)C results indicated that an exposure to 0.03 mg m-3 of 3 h 10 min was required to produce 95 % mortality of V. germanica. The latter result was supported by the LT95 value of 3 h 12 min.

217

Although no control results were reported, the results provided some evidence that after an exposure of approximately 3 h a dichlorvos vapour concentration of 0.03 µg dichlorvos l-1 was effective against V. germanica (Unpublished, 1964).

5.3.2 FIELD DATA

5.3.2.1 LEPIDOPTERA

A study investigated the effect of dichlorvos strips against E. elutella. This study involved the monitoring, over a 10 week period (May to August), of adult moth and moth larval populations, and the monitoring (over a 15 week period) of dichlorvos air levels. The monitoring followed the suspension of dichlorvos impregnated PVC strips from the ceilings of six stowages in a multi-storey storage warehouse in London. The warehouse was used to store a variety of infestible commodities such as cocoa, coffee, spices and desiccated coconut. The strips used contained 18.6 % w/w dichlorvos and, at 21-27 oC, had an emission rate of 35 mg dichlorvos h-1 when 3 d old, reducing gradually to 2 mg h-1 at 90 d. The application rate for flying insects was 1 strip per 28 m3. In addition to the dichlorvos treated stowages, a further two were treated sporadically with an oil-based pyrethrin spray. No negative controls could be undertaken due to the impracticability of allowing stowages to go untreated. The stowages chosen for the study were selected because they already contained a large number of hibernating larvae of E. elutella, and a heavy emergence of moths was expected. Each stowage was 2.5 m in height, although the volumes of the stowages varied between 538 and 1700 m3. The stowages were treated in pairs: 1 & 2 - one strip/28 m3, suspended before moth emergence; 3 & 4 - one strip/56 m3, suspended before moth emergence, to determine the effectiveness of treatment at a reduced rate which laboratory trials had indicated might be effective; 5 & 6 - one strip per 28 m3, suspended three weeks after moth emergence started, to determine the amount of time needed to kill an existing moth infestation; 7 & 8 - treated with an oil-based pyrethrin spray, for comparison purposes. The temperature and r.h. in stowages 7 and 8 were monitored in weeks 4 - 10. A temperature variation of approximately 15 - 23 oC was recorded in stowage 8, with a variation of approximately 17 - 19 oC in stowage 7. The temperature variation in the former was considered to be atypical due to the fact that this stowage was on the top floor of the building. The temperature variation in the latter was considered to be more typical of the remainder of the building. The r.h. showed a typical daily fluctuation between 60 - 80 %. The number of live moths in stowage 5, in which strips were introduced three weeks after the beginning of the emergence season, was monitored on a weekly basis. The results are presented in Figure 5.28.

218

Figure 5.28 Effect of dichlorvos strips applied at a rate of 1 strip/28 m3 on emergent

E. elutella population in stowage 5. Strips introduced 3 weeks after beginning of emergence

Note: The break in the graph for the control results between d 3 and 4, is a reflection of the fact that after the treated stowage was monitored at d 3, it was monitored on a further two occasions within 4 h. There are thus two extra data points for the treated stowage. The results in Figure 5.28 indicated a slight increase in the number of live moths in the first two weeks of the emergence season, with a large increase to approximately 8,000 moths at week 3. The results also indicated that after introduction of the strips at week 3 the number of live moths decreased by approximately 90 % within a period of 4 h. The number remained low until the end of the test period. The number of live moths in stowages 1, 2, 3 and 4, those stowages in which strips were introduced at the beginning of the emergence season, was monitored on a weekly basis, as was the number of moths dying in the preceding week. This exercise was also undertaken for stowages 7 and 8, those stowages treated with pyrethroid spray. The results are presented in Figures 5.29 - 5.34. The results for stowage 1 (Figure 5.29) indicated a very low number of live moths (< 55 per 28 m3) throughout the duration of the test. The results also indicated large numbers of dead moths (approximately 3700 per 28 m3 decreasing to 1600 per 28 m3) during the period when the bulk of the moths emerged (i.e. those dying in weeks 2 and 3), with the largest increase found (140 per 28 m3 - 3700 per 28 m3) in week 2.

0.3 1 2 3 3.01 3.02 4 5 6 7 8 9 10

Period following first appearance of emergent moths (weeks)

0

2

4

6

8

10

12

14

16

Est

imat

ed n

o. o

f liv

e m

oths

(tho

usan

ds)

Dichlorvos treatment (stowage 5)

Control (stowage 7)

Green 1966 Figure 1

219

Figure 5.29 Effect of dichlorvos strips applied at a rate of 1 strip/28 m3 on emergent E. elutella population in stowage 1. Strips introduced at beginning of emergence season

Figure 5.30 Effect of dichlorvos strips applied at a rate of 1 strip/28 m3 on emergent E. elutella in stowage 2. Strips introduced at beginning of emergence season

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0

1,000

2,000

3,000

4,000

Num

ber

of m

oths

/28

m3 Alive at time of count

Dying in previous week

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0102030405060708090

100

Num

ber

of m

oths

/28

m3 Alive at time of count

Dying in previous week

220

Figure 5.31 Effect of dichlorvos strips applied at a rate of 1 strip/56 m3 on emergent E. elutella in stowage 3. Strips introduced at beginning of emergence season

Figure 5.32 Effect of dichlorvos strips applied at a rate of 1 strip/56 m3 on emergent E. elutella in stowage 4. Strips introduced at beginning of emergence season

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0

100

200

300

400

500

600

Num

ber

of m

oths

/56

m3 Alive at time of count

Dying in previous week

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0

50

100

150

200

250

300

Num

ber

of m

oths

/56

m3 Alive at time of count

Dying in previous week

221

Although the degree of infestation in stowage 2 was less than in stowage 1, the results for stowage 2 (Figure 5.30) indicated a similar pattern to those for stowage 1. The results for stowages 3 and 4 (Figures 5.31 and 5.32), those stowages treated at a rate of 1 strip/56 m3, indicated a similar trend to those obtained for stowages 1 and 2.

Figure 5.33 Effect of oil-based pyrethroid spray on emergent E. elutella population in stowage 7. Treatment undertaken at beginning of emergence season

Figure 5.34 Effect of oil-based pyrethroid spray on emergent E. elutella population in

stowage 8. Treatment undertaken at beginning of emergence season

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0

100

200

300

400

500

600

700

Num

ber

of m

oths

/28

m3 Alive at time of count

Dying in previous week

0 1 2 3 4 5 6 7 8 9 10

Period of time after treatment (weeks)

0

10

20

30

40

50

Num

ber

of m

oths

/28

m3 Alive at time of count

Dying in previous week

222

The results for stowages 7 and 8 (Figures 5.33 and 5.34), those stowages treated with pyrethroid spray, indicated that the bulk of the emerged moths, having emerged during weeks 1- 4, remained alive until week 6 - 7. The authors estimated that the total number of live moths in stowage 7 increased to approximately 16000 between weeks 2 - 4 before decreasing to approximately 1000 by week 7 and then increasing to approximately 2000 - 3000 during the remaining three weeks of the test. Records were kept of larvae migrating in the autumn from commodities, which had remained in the stowages during the experimental period. However, due to the fact that it was by no means certain that infestible commodities would remain in storage long enough to do this, small bags containing 4.5 kg of clean cocoa were distributed in the stowages to act as 'trap bags'. The number of trap bags was proportional to the size of each stowage. The trap bags were analysed at the end of the 10 week monitoring period. The results are presented in Figure 5.35.

Figure 5.35 Larval infestation of trap bags from stowages treated with

dichlorvos impregnated PVC strips, or with a pyrethroid spray The results in Figure 5.35 indicated that the trap bags placed in the dichlorvos treated stowages, when examined at the end of the test period, contained minimal larval infestation. The trap bags placed in the pyrethroid spray treated stowages contained a higher degree of infestation, particularly so in stowage 7. Dichlorvos air levels were monitored in stowages 1, 2 and 3 over a period of 15 weeks. Air samples were taken before treatment and at intervals during treatment to determine the atmospheric concentrations of dichlorvos at 7 a.m. i.e. when the stowages had been closed for the night, and at 4 p.m., when the doors had usually been open for about 8 h. A total of 50 l

1 2 3 4 5 6 7 8

Stowage number

0

10

20

30

40

50

No.

of l

arva

e m

igra

ting/

trap

bag

223

of air were drawn at 5 l min-1 through a glass filter immersed in 50 ml of distilled water. The apparatus was moved around the stowage and a representative sample taken at a height of 1.5 m. The results are presented in Figures 5.36 - 5.38.

Figure 5.36 Air concentrations of dichlorvos in stowage 1

Figure 5.37 Air concentrations of dichlorvos in stowage 2

1 2 4 6 15

Period of time after treatment (weeks)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08C

once

ntra

tion

(mg/

m3)

7 a.m.

4 p.m

1 2 4 6 15

Period of time after treatment (weeks)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Con

cent

ratio

n (m

g/m

3)

7 a.m.

4 p.m

224

Figure 5.38 Air concentrations of dichlorvos in stowage 3 The results in Figure 5.36 indicated that during the period when the bulk of the moths emerged (weeks 2 - 6) the air concentration varied between approximately 0.02 - 0.06 mg dichlorvos m -3. Figure 5.37 indicated that for the corresponding period in stowage 2 (weeks 4 - 7) the air concentration varied between approximately 0.02 - 0.04 mg dichlorvos m-3 and Figure 5.38 indicated that for stowage 3 (weeks 2 - 6) the air concentration varied between approximately 0.03 - 0.05 mg dichlorvos m-3. The results from this study indicated that strips containing 18.6 % w/w dichlorvos, which were introduced before emergence began, and produced air concentrations of 0.02 - 0.06 mg dichlorvos m-3 during the emergence periods, were effective in killing the bulk of the emerged moths, and were considerably more effective than treatment involving the use of a pyrethroid spray. When strips were introduced during the emergence period i.e. when an infestation had already taken hold, they produced approximately 90 % mortality of the emerged moths within 4 h. The results also indicated that the strips were effective in preventing larval infestation, and were more effective than the pyrethroid spray. Although no negative control test was conducted due to the impracticability of using untreated stowages, the results obtained for the dichlorvos treated stowages were sufficiently marked as to provide clear evidence that dichlorvos strips producing an air concentration of 0.02 - 0.06 mg dichlorvos m -3 were effective against emerged E. elutella (Green et al., 1966). A study investigated the effect of a resin strip containing 20.0 % w/w dichlorvos against E. elutella. The study was conducted in late spring and summer in the Netherlands, in a cocoa storeroom containing a heavy endemic infestation of E. elutella. The storeroom was 30 m x 20-25 m x 5 m in dimensions, with a volume of approximately 2200 m3.

1 2 4 6 15

Period of time after treatment (weeks)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Con

cent

ratio

n (m

g/m

3)

7 a.m.

4 p.m

225

At the beginning of the study 950-1000 m3 of the room were filled with sacks containing cocoa beans. A total of 46 strips (6.25 cm x 25.0 cm x 0.5 cm) were attached to the ceiling in four rows of eight strips. The distance between the strips within a row was approximately 3 m, and the distance between the rows approximately 4-5 m. The mean temperature of the storeroom was 18 oC (range 15-22 oC) and the mean r.h. was 67 % (range 60-72 %). During the study the storeroom was in normal use. About 6 weeks after the beginning of the experiment sacks were removed, and two months later new sacks were stored.

The effect of the strips on newly emerging adults in the existing E. elutella infestation was investigated. The authors reported that although, at the beginning of the emergence period, the number of live moths was so large that living moths could be seen at all times in the storeroom, large numbers died, with the number of dead and dying moths covering the entire floor, particularly during the first month. As a consequence, it became impossible to make an exact count of the number of dead moths on the floor. Therefore, in order to get an impression of the number of moths present, two trays (each 1 m2 area) were placed at random on the floor and the number of moths found in the trays counted at intervals during the study. The results are presented in Figure 5.39.

Figure 5.39 Number of E. elutella adults found on two trays placed on the floor of the store room

The results in Figure 5.39 show that the strips produced a very substantial measure of control of E. elutella adults, with the number of moths present on the floor of the storeroom decreasing from > 150 m-2 after 9-13 d to < 7 m-2 by the end of the test period. The authors also reported that in August and September (the migratory period for E. elutella larvae) no larvae were found in sacks of cocoa beans, which had been stored since April. The authors attributed the absence of migrating larvae to the strips killing adults before they had a chance

7-9

9-13

13-1

4

14-1

6

16-1

9

19-2

1

21-2

7

27-3

1

31-3

6

36-4

2

42-4

5

45-5

3

Time following installation of strips (days)

0

50

100

150

200

Num

ber

of m

oths

/squ

are

met

re

Tray 1

Tray 2

226

to mate or to lay eggs. The authors also reported that from the 1st until the 45th d after installation of the strips approximately 35 newly emerged adults were placed in cages in different positions and heights in the storeroom each week. The authors reported that all of the moths died within 24 h. Although no negative control test was conducted, the results obtained for the dichlorvos treated storeroom were sufficiently marked as to provide clear evidence that when dichlorvos strips were introduced into a well ventilated storeroom under temperature and humidity conditions typical of a temperate climate, the strips could produce effective control of E. elutella (Schulten & Kuyken, 1966).

5.3.3 NICHE-MARKET USE AGAINST COLEOPTERA AND LEPIDOPTERA

The following data are presented in support of the use of dichlorvos against Anobium punctatum (common furniture beetle) in museums, where it is an important pest species. A study investigated the effect of dichlorvos vapour from PVC strips on different stages of the life cycle of A. punctatum in a simulated roof space. Three PVC strips, each containing 20 g dichlorvos, were introduced in to the 41 m3 test area, resulting in a SVR of 1.46 g m-3 dichlorvos. The strips were left in place for a period of 3 months from mid June to mid September, which covered the emergence and egg laying of A. punctatum adults and the incubation of eggs. The temperatures during this period ranged from 14-19 oC. The effect of dichlorvos on the emergence of adults was evaluated by placing plywood boards infested with A. punctatum (which had any existing exit holes plugged so that new holes could be identified) into the treatment room. The boards were introduced in 3 batches, with the final batch of boards being introduced 18 d after the first batch. The number of pupae in the boards was estimated before the test by x-raying the samples. Three months after the introduction of the first batch the new exit holes were counted and categorised. The results are shown in Table 5.62.

Table 5.62 Effect of dichlorvos on emergence of adult A. punctatum from infested plywood

Emergence Holes

Completed Uncompleted Estimated

Number of Pupae Empty Containing

Adult

Treated room 815 0 0 391 Untreated room 448 248 13 6

The results in Table 5.62 indicated that emergence of adults from the wood was completely prevented by the treatment. The effect of dichlorvos on adult A. punctatum was evaluated by placing cages containing 50 adult males and 25 adult females into the treated area and untreated control area.

227

Knockdown and the number of eggs laid were monitored after 18 and 42 h. The trial was conducted twice, and the results are presented in Table 5.63.

Table 5.63 Effect of dichlorvos on adult A. punctatum

Treated room Untreated room % knockdown

after Eggs laid

after % dying after Eggs laid

after

18 h 42 h 18 h 42 h 18 h 42 h 18 h 42 h Trial 1 Male (50) 92 100 - - 2 2 - - Female (50) 68 100 9 13 0 2 48 90 Trial 2 Male (50) 98 100 - - 12 12 - - Female (50) 96 100 4 4 0 0 35 104

- Not tested The results in Table 5.63 indicated that there was 100 % knockdown of adult A. punctatum after 42 h exposure to the strips, and that the number of eggs laid in the treated room was greatly reduced in comparison to the untreated control. In order to investigate the effects of dichlorvos on the hatching of A. punctatum eggs, blocks of Scots pine containing eggs, which were 1-6 d old, were placed into the test room, and were assessed after around 2 months. The results are shown in Table 5.64.

Table 5.64 Effect on dichlorvos on the hatching of A. punctatum eggs

Hatched Unhatched

Treated room 0 436

Untreated room

349 21

The results in Table 5.64 indicated that treatment with dichlorvos completely prevented hatching of A. punctatum eggs (Baker & Harris, 1969). Another set of studies was conducted to follow on from those above, in order to investigate the effects of impregnated PVC strips when used at lower application rates. The simulated roof space was again used together with a domestic sitting room. A single PVC strip, containing 20 g dichlorvos, was introduced into each 41 m3 site, giving an SVR of 0.49 g m-3 dichlorvos. The temperatures in the test areas were recorded on a weekly basis and ranged between 15 and 40 oC. The effect of dichlorvos on the emergence of adult A. punctatum was again evaluated by placing infested plywood boards (which had the existing exit holes plugged) into the treatment areas for a period of 3 months. The number of pupae present in the boards was estimated before treatment by x-raying the boards, and the number and type of exit holes and the number of eggs laid were estimated after treatment. The results are summarised in Table 5.65.

Table 5.65 Effects of dichlorvos on emergence of adult

228

A. punctatum from infested plywood

Exit Holes Egg Laying On Surface In Exit Holes

Situation SVR (g m-3)

Estimd. Pupae

Incomplt. Complt. % Completed Total No

Hatched Total No

HatchedSimulated Roof Space

0.49 548 351 23 7 0 0 0 0

Sitting Room

0.49 200 110 39 26 0 0 2 0

Control - 190 4 122 97 244 242 568 568 The results in Table 5.65 indicated that in the simulated roof space and the sitting room the application of dichlorvos strips led to a marked reduction in the number of adult beetles able to complete their exit holes and emerge. The treatment also prevented egg laying on the surface of the wood and greatly reduced the number of eggs laid in exit holes. The effect of dichlorvos on adult insects was assessed by introducing cages containing 50 male and 25 female A. punctatum adults into the simulated roof space (0.49 g m-3 dichlorvos) and into an untreated control room and recording the number of insects knocked down after 16 and 40 h, as well as the number of eggs laid at these times. The eggs were left in place for a further 8 weeks, and the number hatching was recorded. The results are shown in Table 5.66.

Table 5.66 Effect of dichlorvos on adult A. punctatum

Treated Area Untreated Control % KD Eggs Laid Subsequent

hatching (%)

% KD Eggs Laid

16 h 40 h 16 h 40 h 16 h 40 h 16 h 40 h

Subsequent hatching

(%)

Male (50)

83 100 - - - 4 14 - - -

Female (25)

48 100 36 53 0 4 8 98 195 82

- Not tested The results in Table 5.66 indicated that all of the insects introduced into the treatment area were knocked down within 40 h and that the number of eggs laid was consequently reduced. None of the eggs laid in the treatment area subsequently hatched (Harris et al., 1970). Another study, reported in the same paper, investigated the effect of dichlorvos on the hatching of eggs. Scots pine blocks containing eggs of A. punctatum were placed within the simulated roof space, the domestic sitting room and an untreated control site for a period of between 7-8 weeks, and the number of eggs hatching in this period was recorded. The results are shown in Table 5.67.

Table 5.67 Effect of dichlorvos on the hatching of A. punctatum eggs

229

Situation Strip: Volume Ratio of

Dichlorvos (g/m3)

Total Number of

Eggs

Number Hatched

% of Total Hatched

Simulated Roof Space

0.49 862 0 0

Sitting Room

0.49 810 0 0

Control - 492 444 90 The results in Table 5.67 indicated that hatching was completely prevented in the treated areas (Harris et al., 1970). As the following study on the effects of dichlorvos against T. bisselliella was conducted using dichlorvos strips at a higher SVR than approved for use in domestic situations, this study is presented only in support of the use of dichlorvos strips in museums. A laboratory study was conducted to assess the efficacy of a 1.7 g dichlorvos strip against T. bisselliella in a simulated wardrobe. The 0.5 m3 test wardrobe was maintained at 22-26 oC and the strip was placed in this chamber (resulting in a SVR of 3.4 g m-3 ) for the duration of the test (16 weeks). The insect test cage contained 20 cm2 woollen fabrics, the number of fabrics and size of the test case was not stated. Efficacy was assessed after the initial introduction of the strip into the wardrobe, then after 8 and 16 weeks. The insect free cage was preconditioned in the wardrobe for 4 d, after which T. bisselliella were introduced and mortality was measured over the next 10 d. The number, age and sex of the test insects were not presented in this summary. The test results are summarised in Table 5.68.

Table 5.68 Percentage mortality of T. bisselliella

following exposure to a dichlorvos strip

% mortality of test insects after exposure (d) Time after introduction of the strip (w) 1 d 3 d

0 66 100

8 94 100

16 100 100

The results in Table 5.68 indicated that all test insects were killed within 3 d of introduction into the test cage (Unpublished, 1993b).

230

5.4 PRE-PRESSURISED HANDHELD AEROSOL SPRAYS

5.4.1 DICHLORVOS AS SOLE INSECTICIDE

5.4.1.1 LABORATORY DATA

A study investigated the effect of a dichlorvos aerosol on M. domestica. One hundred insects of mixed sex were released into a chamber of volume 4.4 m3. The chamber was then treated with a dichlorvos aerosol at two different application rates, and the time taken for an effect to be observed, the mortality after 10 min, and the time taken to achieve 100 % kill were recorded. Full details of the treatment, formulation and environmental conditions were not presented. The results are presented in Table 5.69.

Table 5.69 Effects of a dichlorvos aerosol on M. domestica

Calculated Application Rate

(mg dichlorvos m-3 )

Time to start of effect (min)

Mortality after 10 min (%)

Time taken for 100 % kill (min)

1.3 3 - 5 78 - 96 14 - 18 2.3 2 93 11.5

It can be seen from Table 5.69 that both aerosol application rates resulted in 78 % or greater mortality after 10 min, and 100 % mortality was achieved in 11.5 and 14 - 18 min for 2.3 and 1.3 mg m-3, respectively (Bachmann, 1960). A study investigated the effect of a range of insecticides, including dichlorvos, against M. domestica. Adult females of the ‘Takatsuki’ strain of M. domestica were used in the test. Larvae were reared on a yeast powder mixture (Tofu), and the adults were fed on sugar and water. On the fourth to fifth d after emergence, the adults were anaesthetised with CO2 and sexed. The knockdown effect from exposure to dichlorvos spray was assessed. Each test consisted of 3 replicates. To assess the knockdown effect, the settling mist apparatus, as described by Nagasawa (1953) was used (no further details provided). Ten M. domestica were used for each test. A 0.2 ml acetone solution of 0.5 % w/w insecticide concentration was sprayed under a pressure of 1.2 kg cm-2. The knockdown rate was observed with time (time not reported) and mean KT50 and KT90 values derived. Neither the test temperature/humidity nor negative (blank) control data were reported. The mean KT50 and KT90 values were reported as 5.44 min and 7.54 min, respectively. Although the test conditions and control results were not reported, the test results provided some evidence that a 0.5 % w/w dichlorvos spray could produce high knockdown against M. domestica within 8 min (Kitagaki et al., 1973). A study investigated the effect of dichlorvos vapour produced by a residual spray, against adult M. domestica. In these tests, a 5.8 m3 chamber with paper lined walls was sprayed with dichlorvos at a rate of 1.95 g dichlorvos m-2. The dichlorvos air concentration was monitored, although the method of analysis was not reported. Four hundred specimens (200 in each of

231

2 cages) of M. domestica (sex not reported) were exposed to the vapour for 30 min at various intervals during the subsequent decline of the vapour concentration, and the observed mortality after 24 h recorded. No control data were reported. The results indicated that an air concentration of 0.05-0.08 mg dichlorvos m-3 produced low mortality, and that a concentration of 0.15 mg dichlorvos m-3 produced 95-100 % mortality. Although the test conditions and control data were not reported, the results provided some evidence that dichlorvos sprayed at a rate of 1.95 g m-2 and producing an air concentration of 0.15 mg dichlorvos m-3, was effective against M. domestica (Maddock et al., 1961). A study reported the results from spraying trials conducted on the Coleopterids Dermestes lardarius (larder beetle), Tenebrio molitor (yellow mealworm beetle) and T. castaneum, the Lepidopterid Ephestia kuehniella (Mediterranean flour moth) and the Dictyopterid B. orientalis. In each of these trials five insects were sprayed from a distance of 50 cm for 1 s with an aerosol containing 0.4 % w/w dichlorvos, giving a total quantity applied of 6 mg dichlorvos. No control data were reported. The test results are presented in Table 5.70.

Table 5.70 Results for the effects of a formulation containing

0.4 % w/w dichlorvos on various insect species

Findings after Insect species 1 h 1.5 h 16 h

D. lardarius Imago

5 mb 5 mb 3 +, 2 mb

D. lardarius Larva

5 mb 5 mb 5 mb

B. orientalis (Life stage not stated)

5 mb 1 +, 4 mb 3 +, 2 mb

T. castaneum (Life stage not stated)

5 mb 5 + 5 +

E. kuehniella Larva

5 sp 5 mb 5 mb

T. molitor Larva

5 p 5 p 3 sp, 2 mb

p: paralysed, sp: severely paralysed, mb: moribund, +: dead Table 5.70 indicated that 1 h after exposure of D. lardarius larvae and imago, B. orientalis and T. castaneum to dichlorvos, all of the test insects had been knocked down. The results also indicated that after 16 h dichlorvos produced 60 % mortality against B. orientalis and D. lardarius imago, and 100 % mortality against T. castaneum, but failed to produce any mortalities against D. lardarius larvae, E. kuehniella larvae and T. molitor larvae. Although the test conditions and control results were not reported, the test results provided some evidence that when an aerosol formulation containing 0.4 % w/w dichlorvos was sprayed, application was effective against T. castaneum, was less effective against B. orientalis and D. lardarius imago, and was not effective against D. lardarius larvae, E. kuehniella larvae and T. molitor larvae (Bachmann, 1960).

232

A study investigated the effect of various insecticides, including dichlorvos, against B. germanica. The test insecticides were diluted in deodorised kerosene, and applied as a contact spray. The test concentrations were 0.5 % w/w and 2.0 % w/w active ingredient. The author stated that in the contact spray test method used for the test, 20 adult male insects were placed in batches of 10 into screen wire cages, and exposed 'briefly' to 'standard amounts of spray' (test conditions not reported) before being transferred to clean glass dishes. Knockdown was then recorded after 10, 30 and 60 min and mortality after 24 h. No control data were reported. The results for dichlorvos are presented in Table 5.71.

Table 5.71 Results for percentage knockdown and mortality (24 h) following exposure of B. germanica to dichlorvos applied as a kerosene based contact spray

Knockdown (%) Concentration

(% w/w) 10 min. 30 min. 60 min.Mortality after

24 h (%) 0.5 85 100 100 100 2.0 100 100 100 100

The results in Table 5.71 indicated that 0.5 % w/w dichlorvos produced 100 % knockdown within 30 min, and 2.0 % w/w within 10 min. Both concentrations produced a 100 % mortality rate 24 h after exposure. Although the test conditions and control results were not reported, the test results provided some evidence that a solvent based spray containing 0.5 % w/w dichlorvos was effective against B. germanica (Anonymous, 1961). A study reported the results of direct application tests in which various species of Coleoptera and Lepidoptera were exposed to a series of concentrations of dichlorvos, and to a solvent control. In each test, 5 cm3 of each solution was sprayed for 10 s onto 20 test insects, which had previously been anaesthetised by exposure to CO2. The concentrations and amounts of dichlorvos applied, and the application rates, are presented in Table 5.72.

Table 5.72 Concentrations/amounts of dichlorvos applied, and application rates

Concentration applied

(% w/w) 0.001 0.01 0.10 1

Application rate (mg s-1)

0.01 0.1 1 10

Total active ingredient applied (mg)

0.1 1 10 100.

The insects were judged to be dead if they could no longer stand or move, and the mortalities were corrected for the occasional, but negligible, natural mortality observed in control samples, and were corrected using Abbott's formula (see Appendix 5.1.4). The results are presented in Table 5.73. Table 5.73 Mortality rates for various species of Coleoptera and Lepidoptera following

exposure to a series of dichlorvos containing solutions

233

Corrected % mortality (24 h) at different solution concentrations

Insect

0.001 % w/w 0.01 % w/w 0.1 % w/w 1.0 % w/w Sitotroga cerealella (Angoumois grain moth) 100 NT NT NT

T. confusum 0 0 100 100 S. granarius 0 0 93 NT Oryzaephilus surinamensis (saw-toothed grain beetle) 0 0 100 NT

Trogoderma parabile larvae (warehouse beetle) 0* 0* 100* NT

NT-Not tested *Results for 48 h mortality. Table 5.73 indicated that the solution containing 0.001 % w/w dichlorvos, and applied at a rate of 0.01 mg s-1, produced 100 % mortality of S. cerealella after 24 h. The results also indicated that the Coleopterid species tested were less susceptible, with 93-100 % mortality observed 24 h after treatment with 0.1 % w/w, and applied at a rate of 1 mg s-1. T. parabile larvae were the least susceptible, with 100 % mortality observed 48 h after application of 0.1 % w/w dichlorvos. The results provided evidence that a 0.001 % w/w dichlorvos spray applied at a rate of 0.01 mg s-1 was effective against Lepidoptera, and that a 0.1 % w/w dichlorvos spray applied at a rate of 1 mg s-1 was effective against Coleoptera (Strong & Sbur, 1968).

5.4.1.2 SIMULATED-USE DATA

A study investigated the efficacy of dichlorvos vapour produced by a residual spray, against M. domestica. Three tests were conducted in a ‘mock up’ of an aircraft passenger compartment (dimensions not reported) treated with dichlorvos vapour under continuous ventilation. The test chamber was sprayed at an application rate of 1.95 g dichlorvos m-2 to give vapour concentrations ranging from 0.03-0.25 mg dichlorvos m-3. No details were provided on the methodology used for collection and analysis of the air samples. Four replicates were performed for each test. In each replicate, 6 cages, each containing 200 adult female M. domestica, were hung at different locations in the test chamber. The test insects were then exposed to the vapour for a period of 30 min. The cages were then removed and 24 h mortality recorded. No control data were reported. The results are presented in Table 5.74.

Table 5.74 24 h mortality rates for adult female M. domestica

exposed in cages to dichlorvos vapour for 30 min at 6 locations in a simulated aircraft passenger compartment

234

24 h mortality (%) Test Replicate Dichlorvos

concentration (mg m-3)

1 2 3 4 5 6

1 - 100 100 100 100 100 100 2 0.21 100 100 100 100 100 100 3 0.19 100 100 100 100 100 100 1

4 0.18 98 98 99 99 100 100 1 0.20 100 100 100 100 100 100 2 0.18 99 100 100 100 100 100 3 0.03 1 2 0 2 0 4 2

4 0.18 100 97 100 100 100 100 1 0.09 94 96 99 100 99 94 2 0.15 95 98 99 100 99 99 3 0.14 99 100 99 100 100 100 3

4 0.25 100 99 100 100 100 98 - Not measured The results in Table 5.74 show that concentrations in the range 0.14-0.25 mg dichlorvos m-3 produced 95-100 % mortality against M. domestica. The authors reported that in other tests with M. domestica, susceptible An. quadrimaculatus and susceptible and resistant Ae. aegypti were included for comparative purposes. The authors also reported that the results from 50 tests indicated that dichlorvos concentrations giving 44 % mortality of female M. domestica, also gave 64 % and 71 % for the other two species, respectively. The authors reported that in a further series of tests involving 40 replicates, when the females of two strains of susceptible M. domestica, DDT-resistant Ae. aegypti, and susceptible An. quadrimaculatus were exposed simultaneously for 30 min, 100 % mortalities were obtained with concentrations of or above 0.11, 0.16, 0.08 and 0.14 mg dichlorvos m-3, respectively. For all 4 strains, 100 % knockdown of the test insects was observed within 10 min after the 30 min exposure in 17 trials in the concentration range 0.20-0.31 mg dichlorvos m-3. Although no control data were reported, the test results, together with the separate test results reported by the authors, provided some evidence that when dichlorvos was sprayed to an air concentration of approximately 0.1 mg dichlorvos m-3, application was effective against M. domestica (Maddock et al., 1961). A study investigated the effect of a pre-pressurised handheld aerosol spray product against M. domestica. The product contained 0.9 % w/w dichlorvos as the sole active ingredient. In the study, 4-6 replicate tests were conducted on the product, with each test undertaken in a 38 m3 volume test chamber. Each chamber was ventilated for 15 min prior to each test. In each test, approximately 100-150 2-3 d old mixed sex M. domestica adults were introduced into the test room, and the product discharged into the centre of the room for 3.5 s. The application rate for the test product was reported as 110 mg m-3 (equivalent to 0.99 mg dichlorvos m-3).

235

Knockdown counts were made at intervals over a period of 15 min, and paralysed insects removed for assessment of 24 h mortality. Data on percentage knockdown have been presented, together with KD50 values and results for percentage mortality (24 h). No negative (blank) control data were reported. The results for percentage knockdown are presented in Figure 5.40.

Figure 5.40 Percentage knockdown rate for a dichlorvos

containing aerosol product: M. domestica Figure 5.40 indicated a low knockdown rate during the first 4 min, with this increasing markedly at 5 min to reach 97 % at the end of the test period. The test product was reported as having a KD50 of 7 min, and a 24 h mortality of 97 %. Although no control data were reported, the test results provided some evidence that when a pre-pressurised handheld aerosol spray was applied at a rate of 0.99 mg dichlorvos m-3, application was effective against M. domestica within 24 h (Unpublished, 1971).

5.4.2 DICHLORVOS AND PYRETHRINS/PYRETHROIDS

5.4.2.1 LABORATORY DATA

A study investigated the effect of two aerosol formulations against M. domestica. Aerosol 1 contained 0.2 % w/w tetramethrin and 0.5 % w/w dichlorvos. Aerosol 2 contained 0.2 % w/w tetramethrin and 0.9 % w/w dichlorvos (this level is above the currently approved maximum level of dichlorvos (0.8 % w/w)). The 28.4 m3 test room was maintained at

2 3 4 5 7 10 15

Time following discharge (minutes)

0102030405060708090

100

Kno

ckdo

wn

rate

(%)

236

27±1 oC and r.h. 50±5% (lighting conditions and ventilation rates were not reported). After a 3 s aerosol discharge, 300 flies (2-4 d old, mixed sex) were released into the room. Knockdown was recorded at intervals between 1 and 10 min, after which the insects were collected by suction and held in clean cages (under the same environmental conditions) for 24 h, after which percentage mortality was assessed. Each test was repeated 4 times. The results are summarised in Table 5.75.

Table 5.75 Results of two dichlorvos aerosols against M. domestica

Percentage Knockdown Range After Exposure Time in min

Formulation Application Rate

(mg m-3) 1 2 3 5 7 10

24 h Mortality

(%) Aerosol 1 0.38 - 0.42 7.1-

13.4 49.3- 70.3

79.1- 93.4

96.8- 98.8

99.2- 100

100 100

Aerosol 2 0.68 - 0.71 3.3 - 9.0

49.5- 58.9

75.5- 90.0

93.8- 99.5

99.5- 100

100 100

From Table 5.75 the two test formulations gave knockdown of 75 % or greater after 3 min exposure and 100 % knockdown was achieved within the 10 min period. Mortality after 24 h was 100 %. Comparison of the two sets of aerosol data indicates that the higher dichlorvos level in aerosol 2 has little effect on the knockdown action of these formulations (Unpublished, 1987). A study investigated the effect of a commercial aerosol formulation containing 0.5 % w/w dichlorvos and 0.25 % w/w synthetic pyrethroid against P. americana. In the test, adult female P. americana were collected from manholes and refuse chutes. The formulation was then sprayed from a distance of 35 cm onto batches of 10 insects in a perspex cage for 5 s at a rate of 1.7 g formulation s-1. Knockdown was recorded periodically until 100 % were knocked down. The insects were then held in cages and observed daily until end-point mortality was reached. The test was replicated 6-9 times. Probit analysis was carried out to obtain KT50 and KT95 values. No information was given on the test temperature and humidity, and no control results were reported. KD50 and KD95 values of 9.2 min and 29.9 min, respectively, were reported. Although the test conditions and control results were not reported, the test results provided some evidence that when an aerosol formulation containing 0.5 % w/w dichlorvos and 0.25 % w/w synthetic pyrethroid was sprayed onto adult P. americana for 5 s at a rate of 1.7 g formulation s-1, the aerosol could produce high knockdown within 30 min (Ho et al., 1994). A study investigated the efficacy of a commercially available pre-pressurised handheld aerosol contact spray product containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin against B. germanica and P. americana. In this study, in which the product was tested using the standard Chemical Specialities Manufacturers Association direct-spray method, 10 male B. germanica adults were placed in an open ended cylindrical container with a wire gauze base. The container and the aerosol

237

product were then aligned such that the direction of the spray plume would be perpendicular to the base of the container, and the cockroaches then sprayed directly from a distance of 60 cm for a period of 1 s. Immediately following treatment the insects were transferred to clean 15 cm diameter glass crystallising dishes, and knockdown assessed at regular intervals until 95 % of the population were knocked down, or twenty min had elapsed. At the end of the test the insects were transferred to holding containers. Although the author stated that the insects were scored for mortality after 48 h, the mortality results were not reported. The same test protocol was used for P. americana, except that five adult males were treated, and that the insects were sprayed for a period of 2 s. The test was replicated 5 times for each species using a randomised block design. The knockdown results were used to calculate KT50 and KT90 values. No control results were reported. A summary of the test results is presented in Table 5.76.

Table 5.76 Bioassay results for a pre-pressurised handheld aerosol contact spray containing dichlorvos, tetramethrin and d-allethrin

Species Mean

KT50 (min)

Mean KT90 (min)

Discharge rate (mean) (g s-1)

B. germanica 0.65 1.25 2.70 P. americana 3.96 6.10 2.80

The results in Table 5.76 show KT50 and KT90’s of 0.65 min and 1.25 min, respectively, for B. germanica, and 3.96 min and 6.1 min for P. americana. Although control results were not reported, the test results provided some evidence that a pre-pressurised handheld aerosol containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin, could produce high knockdown against B. germanica within 1-2 min (Unpublished, 1989). A study investigated the effect of two pre-pressurised handheld aerosol formulations containing dichlorvos and pyrethrins against various species of Coleoptera. Formulation 1 contained 10.0 % w/w dichlorvos and 0.04 % w/w pyrethrins. Formulation 2 contained 0.4 % w/w dichlorvos and 0.1 % w/w pyrethrins. No application rates were stated. The test species were Attagenus woodroffei (wave-banded fur beetle), adults of Rhyzopertha dominica (lesser grain borer beetle), adults of O. surinamensis, larvae and adults of T. confusum and larvae of Tribolium destructor (dark flour beetle). In the study, which was conducted in Finland, the test insects were reared in glass jars at approximately 22 °C and 30-50 % r.h. in and fed using various food sources. Active adults and last instar larvae were chosen for the tests. The tests were conducted at approximately 22 °C and 30-50 % r.h.. Each formulation was tested separately against the tests species. Each test was replicated 4-8 times, and 30 insects were used in each replicate. Each formulation was applied by holding the aerosol container in a vertical position, and then spraying for 1 s onto filter papers from a distance of 0.5 m. The papers were air-dried overnight before being placed into glass dishes. The test insects were then introduced into the

238

dishes, and the dishes covered with gauze. A small amount of untreated food was placed in the jars to sustain any surviving insects. To assess knockdown and mortality the dishes were examined at 2 h, and at 1, 2, 3, 6, 8, 10 and 13 d after treatment. Insects were judged to have been knocked down if, when stimulated with a bristle brush, movement of the appendages could be seen but the insects were unable to move away. If no movement was seen, the insects were considered to be dead. For each test, a negative (blank) control test was conducted using clean filter papers. The results for each formulation, giving mean 24 h knockdown and mean 6 d mortality, are presented in Tables 5.77 and 5.78.

Table 5.77 Knockdown (24 h) of two dichlorvos- and pyrethrin-containing aerosol formulations against various species of Coleoptera

24 h knockdown

Formulation Species

1* 2**

Control

A. woodroffei (larvae) 100 - 0 R. dominica (adults) - 67 4 O. surinamensis (adults) 100 - 2 T. confusum (larvae) 100 5 1 T. confusum (adults) 100 1 0 T. destructor (larvae) 100 1 1

*10 % w/w dichlorvos and 0.04 % w/w pyrethrins **0.4 % w/w dichlorvos and 0.1 % w/w pyrethrins - Not tested

The results in Table 5.77 indicated that formulation 1 produced 100 % knockdown against all tested species within 24 h. However, formulation 2 produced < 5 % knockdown against T. confusum and T. destructor adults and larvae, and showed a maximum of 67 % (R. dominica adults). The results for the latter formulation did not generally compare favourably with the results for the negative (blank) controls. The results in Table 5.78 indicated that after 6 d, formulation 1 produced 100 % mortality against O. surinamensis adults, and 96 % against T. confusum adults. However, the mortalities against larvae of the tested species were low, particularly for T. destructor (26 %). Formulation 2 tended to produce very low mortality against both adults and larvae, although it 71 % mortality against R. dominica adults. The results for the latter formulation did not generally compare favourably with the results for the negative (blank) controls. The test results provided evidence that an aerosol formulation containing 10 % w/w dichlorvos and 0.04 % w/w pyrethrins could produce 100 % knockdown against stored product Coleoptera within 24 h, and 100 % mortality against adults of stored product Coleoptera within 6 d (Tuovinen & Ekrom, 1982).

Table 5.78 Mortality (6 d) of two dichlorvos- and pyrethrin-containing aerosol formulations against various species of Coleoptera

239

6 d mortality Formulation

Species

1* 2**

Control

A. woodroffei (larvae) 47 - 0 R. dominica (adults) - 71 16 O. surinamensis (adults) 100 - 12 T. confusum (larvae) 63 8 8 T. confusum (adults) 96 0 2 T. destructor (larvae) 26 1 6

* 10.0 % w/w dichlorvos and 0.04 % w/w pyrethrins ** 0.4 % w/w dichlorvos and 0.1 % w/w pyrethrins - Not tested

5.4.2.2 SIMULATED USE DATA

A study investigated the effect of a commercially available pre-pressurised handheld aerosol space spray containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin against M. domestica and Ae. aegypti. In this study, in which the product was tested using a modified version of the British Standard aerosol test method BS 4172 (1967) (see Appendix 5.1.5), approximately fifty 3-4 d old mixed sex M. domestica adults and fifty 4-8 d old sucrose fed mixed sex Ae. aegypti adults were introduced into a 27 m3 volume test room immediately following a 2 s discharge of the test product. Knockdown was then assessed at regular intervals until 95 % of the population were knocked down or fifteen min had elapsed. The test room was maintained at 26 + 2 oC and 45 - 75 % r.h. The test was replicated five times using a randomised block design. Although the author stated that the insects were scored for mortality after 48 h, the mortality results were not reported. The knockdown results were used to calculate KT50 and KT90 values. A summary of these results is presented in Table 5.79.

Table 5.79 Bioassay results for a commercially available pre-pressurised handheld aerosol space spray containing dichlorvos, tetramethrin and

d-allethrin against M. domestica and Ae. aegypti

Species Mean KT50 (min)

Mean KT90 (min)

Mean discharge rate (g s-1)

M. domestica 2.18 6.38 3.10 Ae. aegypti 1.26 3.60 3.10

240

The results in Table 5.79 (Mean KT50 and KT90 of 2.18 min and 6.38 min, respectively, against M. domestica, and 1.26 min and 3.60 min against Ae. aegypti) show that the test product is more effective against Ae. aegypti than against M. domestica. The test results provided evidence that a pre-pressurised handheld aerosol containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin, could produce high rates of knockdown against Diptera within 7 min (Unpublished, 1989). A study investigated the effect of the same pre-pressurised handheld aerosol spray product containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin against M. domestica and Ae. aegypti. In this study, which was conducted using a methodology based on the British Standard aerosol test method BS 4172 (1967) (see Appendix 5.1.5), the knockdown rate was determined under conditions intended to simulate practical use. As well as the test product, a number of other aerosol formulations were also tested. The active ingredient composition of these products is set out in Table 5.80.

Table 5.80 Active ingredient compositions of four competitor products

Active ingredient composition (% w/w) Competitor

Product Permethrin Tetramethrin Cypermethrin

Bioallethrin Propoxur

1 0.075 0.2 - - - 2 - 0.0315 0.108 - 0.753 0.07 - - 0.2 - 4 0.075 0.2 - - -

- Not tested The mean output rate of the test product was determined from 3 discharges from 3 randomly picked cans. Six replicate tests were conducted using M. domestica, and 4 replicates using A. aegypti. In each test, the product was discharged for 3 s into an empty test chamber (volume not reported). The test insects were then introduced into the chamber. The chamber was maintained at 26 + 2 oC and 45-75 % r.h. The percentage knockdown over a period of 15 min was determined and a KT50 value calculated using probit analysis. Insects were introduced into a separate untreated chamber to check for contamination, and observed over a 15 min period. In accordance with BS 4172, the results for a test were considered valid only if the test was conducted on a day in which < 5 % of these control insects were knocked down during the observation period, and < 10 % of the same insects were dead 24 h after release. The mean percentage knockdowns are presented in Figure 5.41 (M. domestica) and Figure 5.42 (A. aegypti), and the mean discharge rate and mean KT50 values are presented in Table 5.81.

241

Figure 5.41 Mean percentage knockdown rates after introduction of M. domestica insects to test room

Figure 5.42 Mean percentage knockdown rates after

introduction of A. aegypti insects to test room

1 2 3 5 7 10 15

Time following release of test insects (minutes)

0

20

40

60

80

100

Mea

n kn

ockd

own

rate

(%)

Competitor product 1Competitor product 2Competitor product 3Competitor product 4Product

1 2 3 5 7 10 15

Time following release of test insects (minutes)

0

20

40

60

80

100

Mea

n kn

ockd

own

rate

(%)

Competitor product 1Competitor product 2Competitor product 3Competitor product 4Product

242

Table 5.81. Mean KT50 values and mean discharge rate for a commercially available pre-pressurised handheld aerosol space spray containing dichlorvos, tetramethrin, d-

allethrin and permethrin against M. domestica and Ae. aegypti

Species Mean KT50 (min)

Mean discharge rate (g s-1)

M. domestica 2.20 3.33 Ae. aegypti 2.30 3.33

Figure 5.41 indicated that the product produced a faster knockdown rate against M. domestica than did the other reference 'competitor' products, producing a knockdown of approximately 95 % within 3-5 min. Figure 5.42 indicated that the product produced a faster knockdown rate against Ae. aegypti than did the reference products 1, 2 and 3, and a rate comparable to that for reference product 4. Both the product and reference product 4 produced a knockdown of approximately 95 % against Ae. aegypti after 3 min. Table 5.81 indicated that the test product produced a mean KD50 of 2.2 min against M. domestica, and 2.3 min against Ae. aegypti. The test results provided evidence that a pre-pressurised handheld aerosol product containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin, could produce high knockdown against the common test species for Diptera i.e. M. domestica and Ae. aegypti within 5 min (Unpublished, 1994). A study investigated the effect of three pre-pressurised handheld aerosol space spray products against M. domestica. The active ingredient composition of the products is set out in Table 5.82. The tests were conducted in a 38 m3 volume test chamber, and 4-6 replicate tests were conducted on each product. The test chamber was ventilated for 15 min prior to each test. Approximately 100-150 2-3 d old mixed sex M. domestica adults were introduced into the chamber and the product discharged into the centre of the chamber for 3.5 s. The application rates for the test products are presented in Table 5.83.

Table 5.82 Active ingredient compositions of three pre-pressurised handheld aerosols

Active ingredient

composition (% w/w)Product

Dichlorvos Pyrethrins 1 0.4 0.1 2 0.4 0.23 0.7 0.1

243

Table 5.83 Application rates for three pre-pressurised handheld aerosol formulations containing dichlorvos: M. domestica

Application rate (mg m-3) Product

Product Dichlorvos Pyrethrins 1 110 0.44 0.11 2 110 0.44 0.22 3 110 0.77 0.11

Knockdown counts were made at intervals over a period of 15 min, and paralysed insects removed for assessment of 24 h mortality. No control data were reported. The percentage knockdown results for product 1 are presented in Figure 5.43. The KD50 values and 24 h percentage mortality results for all three products are presented in Table 5.84.

Figure 5.43 Percentage knockdown rates for a dichlorvos/pyrethrins containing aerosol

product: M. domestica

Table 5.84 KD50 values and mortality rates (24 h) for three different dichlorvos containing aerosol products: M. domestica

Product 1 Product

2Product 3

KD50 (min) 10 9 9 Percentage mortality (24 h) 88 95 95

The results in Figure 5.43 indicated low knockdown during the first 4 min, with this increasing at 5 min to reach 88 % at the end of the test period. The results in Table 5.84

2 3 4 5 7 10 15

Time following discharge (minutes)

0102030405060708090

100

Kno

ckdo

wn

rate

(%)

Product 1

244

indicated KD50’s of 10, 9 and 9 min for Products 1, 2 and 3, respectively, and 24 h mortalities of 88, 95 and 95 %, respectively. Although control results were not reported, the test results provided some evidence that when a pre-pressurised handheld aerosol spray containing 0.4 % w/w dichlorvos and 0.1 % w/w pyrethrins was sprayed against M. domestica, an application rate of 0.44 mg dichlorvos m-3 and 0.11 mg pyrethrins m-3 could produce high knockdown after 15 min and high mortality after 24 h (Unpublished, 1971). A study investigated the effect of a commercially available pre-pressurised handheld aerosol space spray formulation containing dichlorvos and tetramethrin against Vespa crabro (hornet). The knockdown and mortality rates were determined using a method simulating the situation in which the user sprays a short burst directly at individual insects. The test was conducted in a 30 m3 chamber continuously ventilated by means of a fan, and maintained at a temperature of 27 + 1 oC and 50 + 10 % r.h. Three replicate tests were conducted. The test method involved the use of a metal rod to which was attached at one end a sample of the aerosol product, and at the other end a cylindrical cage containing the test insects. The cage was closed at one end by a metal grill, and at the other by a steel plate connected to a trigger mechanism. The distance between the aerosol nozzle and the target cage was adjusted to 80 cm. Twenty insects of mixed sex were then introduced into the cage and the door was closed. The aerosol spray was directed at the centre of the cage door for 1 s, the trigger activated to release the door, and the hornets exposed to a 1 s burst of spray. The amount of formulation discharged in each of the three replicates was 8.2 g, 8.1 g, and 5.9 g, respectively. This gave a mean discharge of 8.07 g. Knockdown counts were made at 30 s intervals for a period of 3 min, and, for each replicate, knockdown expressed as a percentage of the total population. A grading system was applied to the mean percentage knockdown figures obtained from the 3 replicates, to give a broad classification of performance. A negative (blank) control test was conducted using 10 insects. A summary of the test results is presented in Table 5.85.

Table 5.85 Mean percentage knockdown rates for V. crabro after 1 s exposure to dichlorvos and tetramethrin containing pre-pressurised handheld aerosol

space spray formulation

Time following discharge (s) 30 60 90 120 150 180 Mean % knockdown 100 100 100 100 100 100

The results in Table 5.85 indicated that a 1 s exposure produced 100 % knockdown within 30 s. The author stated that 100 % mortality was observed in the exposed insects 24 h after exposure, and that zero mortality was observed in the control insects during the period of the test.

245

The test results provided evidence that a pre-pressurised handheld aerosol spray containing dichlorvos and tetramethrin, could produce 100 % knockdown against V. crabro within 30 s, and 100 % mortality after 24 h (Unpublished, 1996).

5.4.2.3 FIELD DATA

No data have been submitted.

5.5 SUMMARY OF DATA

5.5.1 INNATE EFFICACY

5.5.1.1 DIPTERA

Bachmann (1960) obtained an LC50 of approximately 0.005 µg l-1, and Kitagaki et al., (1973) obtained a KT90 of 1.18 min and an LD50 of 0.039 µg per insect, following topical application against M. domestica. Kano et al., (1977) and Kerdpibule & Hirakoso, (1971) obtained LD50’s for topical application of 0.0301-0.3540 µg per insect and 0.0021 - 0.0112 µg per insect, respectively. These data provided support for those obtained by Kitagaki et al., (1973). Kitagaki et al., (1973) using impregnated filter papers, obtained an LD50 of 0.013 mg 100 cm2, and Rice & Coats (1994) obtained an LC50 of 0.01 µg cm-3, following exposure of M. domestica to dichlorvos vapour. Kitagaki et al., (1973) obtained an LD50 of 0.007 mg 100 cm-2 following exposure of M. domestica to a dichlorvos film, and obtained results providing some evidence that following direct application to plywood and glass, dichlorvos was effective against M. domestica for a maximum of 24 h. Lee et al., (1997) obtained an LC90 of 0.4167 mg l-1 following exposure of Cx. pipiens pallens larvae to dichlorvos vapour. Das et al., (1979) obtained LC50s of 0.045-0.056 mg l-1 for Ae. aegypti larvae, 0.025-0.029 mg l-1 for Cx. fatigans larvae and 0.12-0.15 mg l-1 for An. Stephensi larvae, following exposure to vapour. Miyagi et al., (1994) obtained LC50’s of 0.030-0.092 mg l-1 against Cx. quinquefasciatus larvae, and results by Maddock & Sedlak, (1961) provided some evidence that a 4 h exposure to 0.01 mg dichlorvos m-3 could produce 95-100 % mortality against An. quadrimaculatus.

5.5.1.2 DICTYOPTERA

Ho et al. (1994) obtained a KD50 of 0.47 µg µl-1 following topical application against P. americana, and Shim et al. (1997) obtained an LD90 of 1.123 µg dichlorvos per insect, following topical application against B. germanica.

246

5.5.1.3 HEMIPTERA

Fletcher & Axtell (1993) obtained an LC90 of 5.7 mg dichlorvos l-1, following exposure of C. lectularius to treated filter papers. Although no control data were reported, the test data provided some evidence for the innate activity of dichlorvos against Hemiptera.

5.5.1.4 LEPIDOPTERA

Attia (1976) obtained LD50’s of 0.59-1.47 µg dichlorvos per larva following topical application to C. cautella larvae. The results quoted in the data summary Unpublished (Undated) provided some evidence for the innate activity of 0.04 mg dichlorvos m-3 against adult T. bisselliella and T. pellionella, and to a lesser extent against larvae of these species.

5.5.1.5 COLEOPTERA

The results quoted in the data summary, Unpublished (Undated) provided some evidence for the innate activity of 0.04 mg dichlorvos m-3 against adult A. piceus and A. flavipes, and to a lesser extent against larvae of these species. Aamir (1983) obtained an LC90 of 46.02 mg dichlorvos l-1 following exposure of T. confusum to dichlorvos crystals, and Zettler (1991) obtained LD99s of 1205 µg dichlorvos per g insect and 66 µg dichlorvos per g insect following topical application to T. confusum and T. castaneum, respectively. Rice & Coats (1994) obtained an LC50 of 11.1 µg dichlorvos cm-

3 following exposure of T. castaneum to dichlorvos vapour, and Pasalu & Bhatia (1975) obtained LC50s of 1750-2123 mg l-1 following exposure of T. castaneum to treated filter papers. The results obtained by Karnatak et al. (1991) indicated that concentrations of 0.64 - 10.24 mg dichlorvos l-1 could produce 90% mortality against S. oryzae 90-202 h after application as a contact insecticide. The results obtained by Benezet et al. (1988) provided some evidence that at 18 oC a concentration of 132-336 ng per female was required to produce 20-95 % mortality, and 95-343 ng per female at 32 oC.

5.5.1.6 SIPHONAPTERA

Schwinghammer et al. (1985) obtained an LC50 (24 h) of 0.2 mg m-2, and Kobayashi et al. (1994) an LC50 (24 h) of 2.3 mg m-2, both following application of dichlorvos as a contact insecticide against C. felis. The larval feeding test results (Unpublished, 1986a) indicated that a concentration of 100 ppm dichlorvos could prevent the hatching of C. felis larvae, and that 10 ppm dichlorvos could prevent the development of pupae and hence the hatching of adults. These results were supported by the test results (Unpublished, 1986b), which indicated that 5 ppm dichlorvos was the minimum effective concentration for the development of pupae from larvae, and that 50 ppm dichlorvos was the minimum effective concentration for the development of eggs to larvae.

247

5.5.2 SLOW RELEASE STRIPS

5.5.2.1 AIR MONITORING RESULTS

The results of Unpublished (1995) indicated that when a commercial product with an SVR of 0.46 g m-3 was tested under conditions simulating the minimum level of ventilation required for UK domestic premises, the product produced air concentrations of 0.02 - 0.05 mg dichlorvos m-3 over a period of 16 weeks (112 d). Elgar & Steer (1972) obtained results, which indicated that when a commercial product with an SVR of 0.49 g m-3 was used in UK domestic premises, the product produced a mean air concentration of 0.025 - 0.06 mg dichlorvos m-3 after 14 d, 0.019 - 0.044 mg m-3 after 30 d, and 0.005-0.008 mg m-3 after 120 d. The results from Unpublished (1979) related to two different monitoring studies. The results for the first study indicated that when a commercial product with an SVR of 0.49 g m-3 was used in domestic premises for 4 months in typical UK summer conditions, the product produced a maximum air concentration of 0.026 mg dichlorvos m-3, decreasing to a minimum of 0.004 mg dichlorvos m-3 after 119 d. The second study indicated that when the air concentrations were monitored the following summer, two different versions of the same product produced air concentrations ranging from a maximum of 0.015-0.023 mg dichlorvos m-3, decreasing to a minimum of 0.003-0.006 mg dichlorvos m-3 after 92 d.

5.5.2.2 DIPTERA

The simulated use studies conducted in Unpublished (1976), Unpublished (1980) and Unpublished (1981), and in the field tests conducted by Mankowska & Goszczynska (1969) and in Unpublished (1979), were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support the use of slow release strips at a SVR of 0.49-0.80 g m-3 against Diptera. Further data have now been evaluated in support of strips against Diptera. Batth et al. (1973) indicated that under conditions simulating practical conditions in domestic premises, a dichlorvos strip producing air concentrations of 0.07-0.25 mg m-3 could produce > 85% mortality against free flying M. domestica after 40-65 min exposure to the strip. The simulated use study, Unpublished (Undated a) provided some evidence that a commercial product of unknown SVR could produce effective knockdown of M. domestica for up to 60 d. The simulated use study, Unpublished (1993a) provided some evidence that when a commercial product (SVR of 0.47-0.93 g m-3), previously stored for 16 weeks, was left in an unventilated chamber (at an SVR of 0.62 g m-3) for 17 h, the product produced 99 % knockdown of M. domestica within 180 min. The simulated use study, Unpublished (1999a) conducted on a commercial product (recommended SVR of 0.5-0.8 g m-3) at an SVR of 0.7 g m-3, provided some evidence that after 8 weeks continuous use in an unventilated chamber, air concentrations of

248

0.045 - 0.063 mg m-3 could produce > 95 % knockdown of M. domestica within 90 min. The results also provided some evidence that after 16 weeks intermittent use the strip could produce > 95 % knockdown of M. domestica within 60 min. The simulated use study, Unpublished (1999b) conducted on a commercial product (recommended SVR of 0.5-0.83 g m-3) at an SVR of 1.06 g m-3, provided some evidence that after storage for 8 weeks in an unventilated chamber the product could produce 100 % knockdown of M. domestica within 180 min. The simulated use study Unpublished (1991) provided some evidence that after 4 months storage the product, when tested in an unventilated chamber (SVR of 0.52 g m-3), could produce 90 % knockdown of M. domestica after 150 min and > 98 % mortality after 240 min. The simulated use study, Unpublished (1991b) provided some evidence that after 6 months storage the same product, when tested in an unventilated chamber (SVR of 0.52 g m-3), could produce 90 % knockdown of Ae. aegypti after 72 min and 100 % mortality after 120 min. The simulated use study, Unpublished (1964) provided some evidence that a dichlorvos vapour concentration of 0.03 mg dichlorvos m-3 was effective against M. domestica and Ae. aegypti after an exposure period of approximately 2-3 h. The simulated use study, Unpublished (1966) indicated that when a commercial product designed for use in small spaces (SVR of 0.58 g m-3) was tested at an SVR of 4.49 g m-3, the product could produce 100 % mortality against Ae. aegypti for a period of 17 weeks.

5.5.2.3 DICTYOPTERA

The field studies conducted by Mankowska & Goszczynska (1969) were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support the use of slow release strips at an SVR of 0.49-0.80 g m-3 against Dictyoptera. Further data have now been evaluated in support of strips against Dictyoptera. The simulated use study conducted by Smittle & Burden (1965) provided some evidence that in a 44 m3 chamber, an air concentration of 1.04 mg dichlorvos m-3 could produce 100 % knockdown and kill of P. americana and B. germanica after 20 h. The simulated study, Unpublished (1999c) indicated that when a commercial product designed for use in small spaces (at an SVR of 1.0 g m-3), was tested at an SVR of 0.9 g m-3, the product produced 100 % knockdown of B. germanica and B. orientalis within 24 h for up to 28 d. The simulated use study, Unpublished (1966) indicated that when a commercial product designed for use in small spaces (SVR of 0.58 g m-3) was tested at an SVR of 4.49 g m-3, the product produced 95-100 % mortality against B. germanica, P. americana and L. maderae adults, and 85-100 % mortality against nymphs of the same species, for a period of 13 weeks. The simulated use study, Unpublished (1964) provided some evidence that a dichlorvos vapour concentration of 0.03 mg dichlorvos m-3 was effective against B. germanica after an exposure period of 5 d.

249

5.5.2.4 LEPIDOPTERA

The field study conducted by Green et al. (1966) indicated that the test strips (unknown SVR) produced a dichlorvos vapour concentration of 0.02 - 0.06 mg m-3. These results were in agreement with the air monitoring results obtained in Unpublished (1979), Unpublished (1995) and by Elgar & Steer (1972). The results indicated that, although the test strips introduced into the storage areas at the beginning of the emergence season did not prevent E. elutella emergence, they did produce high mortalities in the emergent populations. The introduction of strips after emergence produced approximately 90 % adult mortality within 4 h. The results therefore indicated that air concentrations of 0.02 - 0.06 mg dichlorvos m-3 were effective against E. elutella. The results by Green et al. (1966) were supported by the results of the field study by Schulten & Kuyken (1966). This indicated that when strips of unknown SVR were tested in a well ventilated storeroom at a temperature and humidity typical of a temperate climate, the strips produced effective control of adult stored product Lepidoptera. The simulated use study conducted by Schulten & Kuyken (1966) reported that a strip containing 20.0 % w/w dichlorvos (unknown SVR), when tested in an unventilated chamber, produced theoretical (i.e. calculated) air concentrations of 0.047 - 0.099 mg dichlorvos m-3. Although no control data were reported, the test results provided some evidence that the strip could produce knockdown rates of 90-100 % against E. elutella adults within 5 h. The results also provided some evidence that the strip was not effective against larvae. The simulated use study, Unpublished (1966) indicated that a commercial product designed for use in small spaces at a recommended SVR of 0.58 g m-3, when tested at a higher SVR (4.49 g m-3), produced 100 % knockdown of T. bisselliella adults (84 % against larvae) for up to 13 weeks. The results of the simulated use study, Unpublished (1997a) indicated that when a commercial product designed for use in small spaces (SVR of 0.5-0.8 g m-3) was tested at an SVR of 0.8 g m-3, the product was not effective against T. bisselliella. However, when the amount of dichlorvos was increased by 50 % to give an SVR of 0.75-1.2 g m-3, the product was effective against adults, but not against larvae. The results of the simulated use study, Unpublished (1997b) indicated that when a commercial product designed for use in small spaces (SVR of 0.8 g m-3) was tested at 0.8 g m-3, the product produced 100 % knockdown of T. bisselliella adults within 8 h, and 100 % mortality within 3 d, for up to 6 months. However, no firm conclusions could be drawn regarding activity against eggs and larvae. The feeding study, Unpublished (1967) conducted in test rooms and wardrobes, indicated that strips producing a concentration of 0.80 g m-3 prevented feeding by T. bisselliella larvae over a period of 14 weeks. The results for adults from toxicity tests indicated that 16 d after introduction of the same strips, the strips produced a minimum mean KD96.7 of 8 d. The simulated use and field data provided evidence that strips with an SVR of 0.46 g m-3 were effective against Lepidoptera. These data were supported by the air monitoring results from Unpublished (1979), Unpublished (1995) and Elgar & Steer (1972), which indicated that when dichlorvos strips with SVR’s of 0.46-0.49 g m-3 were used in domestic premises in

250

the UK, they typically produced air concentrations shown by Green et al. (1966) to be effective against Lepidoptera.

5.5.2.5 COLEOPTERA

The feeding study, Unpublished (1967) conducted in test rooms and wardrobes, indicated that strips producing SVR’s of 0.76. 0.80, 1.53 and 1.60 g m-3 were unable to prevent feeding by A. vorax and A. piceus larvae, and allowed the survival of the larvae. The results from toxicity tests indicated that these strips produced KD96.7 values of 4 - > 28 d. The results indicated that dichlorvos strips within the current approved range of SVR’s (0.49-0.80 g m-3), and at SVR’s up to 1.60 g m-3, are not effective against Coleoptera. The simulated use study, Unpublished (1964) provided some evidence that a dichlorvos vapour concentration of 0.03 mg dichlorvos m-3 was effective against T. confusum, L. serricorne and S. granarius after an exposure period of approximately 70-170 h. The simulated use study by Boles et al. (1974) indicated that when a dichlorvos impregnated wax stick producing 0.60 mg dichlorvos m-3 was tested in an unventilated room, the stick was only able to achieve 100 % mortality against A. flavipes larvae 14 d after the end of a 32 h exposure period.

5.5.2.6 HEMIPTERA

The simulated use results from Mankowska & Goszczynska (1969) were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support the use of slow release strips at a SVR of 0.49-0.80 g m-3 against Hemiptera. Further data have now been evaluated. The simulated use study conducted by Smittle & Burden (1965) provided some evidence that an air concentration of 0.25 mg dichlorvos m-3

was effective against C. lectularius.

5.5.2.7 FORMICOID HYMENOPTERA

The simulated use study, Unpublished (1966) indicated that when a commercial product designed for use in small spaces at an SVR of 0.58 g m-3, was tested at an SVR of 4.49 g m-3, the strip was effective against M. pharaonis for a period of 13 weeks. Given that the maximum approved SVR is 0.80 g m-3, the above do not represent evidence that strips with an SVR of 0.49-0.80 g m-3 are effective against Formicoid Hymenoptera.

5.5.2.8 HYMENOPTERA

The simulated use study, Unpublished (1964) provided some evidence that a dichlorvos vapour concentration of 0.03 mg dichlorvos m-3 will be effective against V. germanica after an exposure period of approximately 3 h.

5.5.2.9 NICHE MARKET USE-LEPIDOPTERA/COLEOPTERA

The simulated use studies conducted by Baker & Harris (1969), Harris et al. (1970) and in Unpublished (1993b), were evaluated during the 1994 review. The ACP considered that these

251

data were sufficient to fully support the niche market use of slow release strips at a SVR of approximately 6.0 g m-3 against Lepidoptera and Coleoptera. No further data have been submitted.

5.5.3 PRE-PRESSURISED HANDHELD AEROSOLS

5.5.3.1 DICHLORVOS AS THE SOLE INSECTICIDE

5.5.3.1.1 Diptera

The study by Bachmann (1960) was evaluated during the 1994 review. The ACP considered these data to be insufficient to support continued approval without the need for a data requirement. Further data have now been evaluated. The laboratory study by Kitagaki et al. (1973) provided some evidence that a 0.5 % w/w dichlorvos spray could produce 90 % knockdown of M. domestica in 7.54 min. The laboratory study conducted by Maddock et al. (1961) provided some evidence that when dichlorvos was sprayed at a rate of 1.95 g m-2 to produce a concentration of 0.15 mg m-3 in a 1.8 m3 chamber, a 30 min exposure produced 95-100 % mortality against M. domestica. The same study also provided some evidence that when dichlorvos was sprayed at a rate of 1.95 mg m-2 under conditions simulating use in an aircraft cabin, a concentration of approximately 0.1 mg dichlorvos m-3 could produce 95-100 % mortality against M. domestica after a 30 min exposure. The simulated use study, Unpublished (1971) provided some evidence that when a pre-pressurised handheld aerosol spray was applied at a rate of 0.99 mg dichlorvos m-3, the spray was effective against M. domestica.

5.5.3.1.2 Dictyoptera

The laboratory study conducted by Bachmann (1960) provided some evidence that an aerosol formulation containing 0.4 % w/w dichlorvos could produce 100 % knockdown of B. orientalis within 1 h, and 60 % mortality within 16 h. The laboratory study, Anonymous, 1961 provided some evidence that a pre-pressurised handheld aerosol contact spray containing 0.5 % w/w dichlorvos could produce 100 % knockdown of B. germanica after 30 min and 100 % mortality after 24 h.

5.5.3.1.3 Coleoptera

The laboratory study conducted by Bachmann (1960) provided some evidence that an aerosol formulation containing 0.4 % w/w dichlorvos could produce 100 % knockdown of D. lardarius imago and T. castaneum adults within 1 h, and 60 - 100 % mortality within 16 h. However, the formulation was not effective against T. molitor larvae.

252

The laboratory study conducted by Strong & Sbur (1968) indicated that a 0.1 % w/w dichlorvos spray applied at a rate of 1 mg s-1, could produce 93-100 % mortality against adults of T. confusum, S. granarius and O. surinamensis after 24 h, and 100 % mortality against T. parabile larvae after 48 h.

5.5.3.1.4 Lepidoptera

The laboratory study conducted by Bachmann (1960) indicated that an aerosol formulation containing 0.4 % w/w dichlorvos was not effective against E. kuehniella. The laboratory study conducted by Strong & Sbur (1968) indicated that a 0.001 % w/w dichlorvos spray, when applied at a rate of 0.01 mg s-1, could produce 100 % mortality of S. cerealella within 24 h.

5.5.3.2 DICHLORVOS AND PYRETHRINS/PYRETHROIDS

5.5.3.2.1 Diptera

The study, Unpublished (1987) was evaluated during the 1994 review. The ACP considered these data to be insufficient to support continued approval without the need for a data requirement. Further data have now been evaluated. The simulated use study, Unpublished (1989) provided some evidence that a commercial pre-pressurised handheld aerosol product containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin, could produce 90 % knockdown of M. domestica within 7 min, and 90 % knockdown of Ae. aegypti within 4 min. The simulated use study, Unpublished (1994), conducted on the same product as Unpublished (1989), provided some evidence that the product could produce approximately 95 % knockdown of M. domestica within 5 min, and 95 % knockdown of Ae. aegypti within 3 min. These results supported the results obtained in the other study. The simulated use study, Unpublished (1971) provided some evidence that a pre-pressurised handheld aerosol spray containing 0.4 % w/w dichlorvos and 0.1 % w/w pyrethrins, when applied at a rate of 0.44 mg dichlorvos m-3 and 0.11 mg pyrethrins m-3, could produce 50 % knockdown of M. domestica within 10 min, and 88 % mortality after 24 h.

5.5.3.2.2 Dictyoptera

The laboratory study conducted by Ho et al. (1994) provided some evidence that when an aerosol formulation containing 0.5 % w/w dichlorvos and 0.25 % w/w synthetic pyrethroid was sprayed for 5 s at a rate of 1.7 g formulation s-1, the spray could produce 95 % knockdown of P. americana within 30 min. The simulated use study, Unpublished (1989) provided some evidence that a commercial pre-pressurised handheld aerosol product containing 0.5 % w/w dichlorvos, 0.1 % w/w tetramethrin, 0.1 % w/w d-allethrin and 0.02 % w/w permethrin, could produce 90 % knockdown of B. germanica within 2 min, and 90 % knockdown of P. americana within approximately 6 min.

253

5.5.3.2.3 Coleoptera

The laboratory study conducted by Tuovinen & Ekrom (1982) indicated that an aerosol formulation containing 10 % w/w dichlorvos and 0.04 % w/w pyrethrins could produce 100 % knockdown against stored product Coleoptera within 24 h, and 100 % mortality against adults of stored product Coleoptera within 6 d.

5.5.3.2.4 Hymenoptera

The simulated use results (Unpublished, 1996) indicated that a pre-pressurised handheld aerosol spray containing dichlorvos and tetramethrin, can produce 100 % knockdown against V. crabro within 30 s, and 100 % mortality after 24 h.

5.6 DISCUSSION

The data in this review includes both supporting data demonstrating the innate insecticidal efficacy of dichlorvos as an active ingredient, and laboratory, simulated use and field data on the efficacy of both strip products and pre-pressurised handheld aerosols and other spray products. The evaluation includes both data held by individual companies and submitted in support of the review and data held in the public domain. Due to the long established use of dichlorvos as an insecticide, a relatively large body of data exists in the public domain. However, many of these data relate to early experiments and sometimes lack details such as the exact formulations, controls, and methods of assessment employed, and, for strip products, the type of product (e.g. controllable or non-controllable cassette).

5.6.1 INNATE DATA

Although, in the main, the results of the simple screening studies were not always well reported or documented, these data demonstrated the insecticidal activity of dichlorvos when applied topically, as a vapour, or by direct contact against a wide spectrum of test insects from the orders Diptera, Dictyoptera, Hemiptera, Lepidoptera, Coleoptera and Siphonaptera.

5.6.2 SLOW RELEASE STRIPS

5.6.2.1 AIR MONITORING

These data demonstrate that standard strip products/cassette units, when used in domestic premises, produced typical maximum air concentrations of approximately 0.02 - 0.06 mg dichlorvos m-3 for approximately 16 weeks, which is the typical lifetime of a product. As the air concentrations are dependent on the environmental conditions and the air sampling/measurement techniques used in the different test studies, the air concentration values are indicators only of the likely concentrations to be found in practice.

254

5.6.2.2 DIPTERA, DICTYOPTERA AND HEMIPTERA

Data on the use of strips against Diptera, Dictyoptera and Hemiptera were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support the use of slow release strips at an SVR of 0.49-0.80 g m-3 against Diptera. Further data have now been evaluated, and these data support the previous recommendation.

5.6.2.3 LEPIDOPTERA

Simulated use and field data were evaluated on the use of strips against Lepidoptera. With the exception of one study, these data provided evidence that strips with an SVR of 0.46 g m-

3 are effective against Lepidoptera. The data were supported by the air monitoring results, which indicated that when dichlorvos strips with SVR’s of 0.46-0.49 g m-3 are used in domestic premises in the UK, they typically produce air concentrations shown in the field study by Green et al. (1966) to be effective against Lepidoptera.

5.6.2.4 COLEOPTERA

Although data have been presented which provide some evidence that dichlorvos strips can be effective against Coleoptera, these data are limited. Consequently, further data are required to fully support a label claim against Coleoptera.

5.6.2.5 FORMICOID HYMENOPTERA

Although data have been presented which provide some evidence that dichlorvos strips can be effective against Formicoid Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Formicoid Hymenoptera.

5.6.2.6 HYMENOPTERA

Although data have been presented which provide some evidence that dichlorvos strips can be effective against Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Hymenoptera.

5.6.2.7 NICHE MARKET USE-LEPIDOPTERA AND COLEOPTERA

Data on the niche market use of strips against Lepidoptera and Coleoptera were evaluated during the 1994 review. The ACP considered that these data were sufficient to fully support this use at an SVR of approximately 6.0 g m-3. No further data have been submitted.

5.6.3 PRE-PRESSURISED HANDHELD AEROSOLS (DICHLORVOS ONLY)

Data were presented to demonstrate that dichlorvos, when formulated as the sole insecticide in aerosol products, has insecticidal activity against a wide spectrum of test insects from the orders Diptera, Dictyoptera, Coleoptera and Lepidoptera. These data are considered by HSE as useful supporting information.

255

5.6.4 PRE-PRESSURISED HANDHELD AEROSOLS (DICHLORVOS AND PYRETHRINS/PYRETHROIDS)

5.6.4.1 DIPTERA

Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Diptera, these data are limited. Consequently, further data are required to fully support a label claim against Diptera.

5.6.4.2 DICTYOPTERA

Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Dictyoptera, these data are limited. Consequently, further data are required to fully support a label claim against Dictyoptera.

5.6.4.3 COLEOPTERA

Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Coleoptera, these data are limited. Consequently, further data are required to fully support a label claim against Coleoptera.

5.6.4.4 HYMENOPTERA

Although data have been presented which provide some evidence that dichlorvos aerosols can be effective against Hymenoptera, these data are limited. Consequently, further data are required to fully support a label claim against Hymenoptera.

5.7 DATA REQUIREMENTS

The following data should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months. Protocols to be agreed by HSE. i. Either simulated use or field data, or a reasoned case based on data, to be submitted in

support of strips against Coleoptera, Formicoid Hymenoptera and Hymenoptera (as appropriate to current label claims).

ii. Either simulated use or field data, or a reasoned case based on data, to be submitted in support of pre-pressurised handheld aerosols containing dichlorvos and pyrethrins/pyrethroids against Diptera, Dictyoptera, Coleoptera and Hymenoptera Either simulated use or field data, or a reasoned case based on data, to be submitted in support of strips against Coleoptera, Formicoid Hymenoptera and Hymenoptera.

256

6 OUTCOME OF THE REVIEW AND DATA REQUIREMENTS

6.1 OUTCOME OF THE REVIEW

At its meetings in April and May 2001, the ACP considered the available physicochemical, toxicological and efficacy data on dichlorvos including reasoned arguments from a number of approval holders and suppliers regarding the interpretation of data. Based on its consideration of anticholinesterase effects, the ACP recommended in May 2001 that Ministers be advised that:

a) approval for all aerosols containing dichlorvos be revoked based on the unacceptable toxicity-exposure ratios (TERs) derived for primary and secondary exposures from professional and amateur use; b) approval for residential uses of slow release controllable and noncontrollable cassettes containing dichlorvos be revoked based on the unacceptable TERs derived for secondary exposures from professional and amateur use; c) approval for the professional use of slow-release strips and controllable and non-controllable cassettes containing dichlorvos in museums be retained subject to the fulfilment of physicochemical, operator exposure and efficacy data requirements; d) approval for the use of slow-release strips in pheromone traps in areas where food may be stored, prepared or consumed be suspended pending the provision of information on food residues; and e) approval for the use of slow release strips in pheromone traps in areas where food is not present be retained subject to the fulfilment of physicochemical and efficacy data requirements.

A review of the agricultural uses of dichlorvos was also considered by the ACP at the same time (ACP evaluation in preparation). The ACP also noted that dichlorvos was under discussion by the Committee on Mutagenicity of Chemicals in Food, Consumer Products and the Environment (COM) and relevant members of the Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC), and recognised that their recommendations would perhaps require modification after consideration of the COM’s findings. At the July 2001 ACP meeting, the Committee considered the implications of the COM’s possible conclusions, and agreed to advise Ministers as follows:

a) If the COM’s final conclusions were that dichlorvos was an in vivo mutagen, and it could not exclude the possibility that the occurrence of tumours in animal tests of carcinogenicity resulted from a genotoxic mechanism, there should be immediate revocation of all uses (both agricultural and non-agricultural).

257

b) Alternatively, if the COM concluded that dichlorvos was an in vivo mutagen, but that the tumours observed in animal tests did not result from a genotoxic mechanism, or if it could not confirm that dichlorvos was an in vivo mutagen, or if it took the view that dichlorvos was not an in vivo mutagen, then the ACP’s previous recommendations would be maintained.

The COM concluded, following a second meeting in July 2001, that dichlorvos should be regarded as an in vivo mutagen (i.e. capable of inducing mutations in living animals) at site of contact and that it could not exclude the possibility of it acting as a genotoxic carcinogen. Consequently the ACP recommended to Ministers that it would be prudent to revoke, with immediate effect, all agricultural and non-agricultural uses of dichlorvos. This advice was given as a precautionary measure, since the possibility of genotoxic carcinogenicity could not be excluded. The ACP considered that any risk of human carcinogenicity was likely to be very small, and would be mainly associated with certain uses in the home and with exposures to some operators in the agricultural sector. Ministers decided that in addition to the revocations and suspensions that were proposed because of concerns about possible anticholinesterase effects of dichlorvos, all uses of dichlorvos should be suspended with immediate effect. Before such regulatory action could be carried out, AMVAC Chemical UK Ltd (an approval holder and manufacturer of dichlorvos) obtained an injunction, which prevented regulatory action. Government agencies were also prohibited from making any announcement to the public about the regulatory action that was proposed. AMVAC also gained permission for a judicial review hearing, which was heard in November 2001. The grounds for the challenge were that AMVAC had not been properly informed of the proposed regulatory action or the basis for it, and had not been given sufficient time to make representations. AMVAC also claimed that Ministers had not given proper regard to the precautionary principle and to the European Convention on Human Rights. The judgement of the Court was issued in December 2001. Mr Justice Crane rejected most of the company’s submissions, including those concerning the precautionary principle and the Convention on Human Rights. However, he ruled that the company had been given insufficient time to respond to the conclusions of the Government’s expert advisers prior to regulatory action being taken. He accepted that the matter was urgent but considered that the claimant had now had full opportunity to present any further material. During the period of the injunction, the ACP was unable to publish the minutes of its meetings. The ACP had concerns that this compromised the openness of the advice given to Ministers, and could thereby have an adverse effect on public confidence in the regulatory process. It was also concerned that speculation about the missing minutes might create unwarranted public anxiety. Notwithstanding these concerns, the ACP agreed that while rapid implementation of regulatory action was desirable once decisions had been made, it was also important that the regulatory process be fair and open to scrutiny. In this case, the ACP’s advice to Ministers had been precautionary (i.e. based on insufficient reassurance that exposures to the compound were acceptable rather than direct evidence that people were being harmed), and the delay caused by the legal action would be acceptable provided that it was not unduly prolonged.

258

59 Following the Court judgement, approval holders for both agricultural and non-agricultural products were asked to provide any further data relating to the potential genotoxic carcinogenicity of dichlorvos. These data were considered by the COM, relevant members of the COC and the ACP in February 2002. The COM considered that there was insufficient evidence to change their previous opinion i.e. that dichlorvos should be regarded as an in vivo mutagen (i.e. capable of inducing mutations in living animals) at site of contact and that it could not exclude the possibility of it acting as a genotoxic carcinogen (The full COM statement regarding the genotoxicity and carcinogenicity of dichlorvos can be found at http://www.advisorybodies.doh.gov.uk/Com/dichlorvos.htm). As the ACP considered that it could not satisfactorily rule out the possibility that dichlorvos is a genotoxic carcinogen (i.e. causes cancer by damaging DNA), it advised Ministers that there could be a small risk of adverse health effects following prolonged exposure to dichlorvos. Ministers decided as a precautionary measure that in addition to the revocations and suspensions that were proposed because of concerns about possible anticholinesterase effects of dichlorvos, the advertisement, sale and supply of all products should be suspended with immediate effect until further studies are conducted to address the mechanisms of tumour formation in mice.

6.2 DATA REQUIREMENTS

In order to address concerns regarding the potential genotoxicity and carcinogenicity of dichlorvos the following data are required. Approval holders must demonstrate to HSE how they are implementing these requirements within 3 months. Protocols are to be agreed by members of the CoM, CoC and ACP as appropriate. All data are to be submitted within 12 months of finalisation of protocols. i. An oral mutation study of the forestomach in the MutaTM Mouse (by gavage/corn oil

vehicle)

ii. An investigation of cell proliferation/inflammation in the mouse forestomach. Should these data be considered sufficient by Ministers to allow the lifting of the suspension, then, the following data requirements shall apply to the few remaining products which have not been revoked due to concerns about possible anticholinesterase effects i.e. pheromone traps to be used in the absence of food and slow release products used for museum specimen preservation. Physical Chemistry The following data should be submitted within 1 year except for 2-year storage stability tests, which should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months. Data Requirements to be addressed by all Approval Holders i. Two year storage stability study on representative strip products if not covered by the

slow release resin strip storage stability studies. Protocols to be agreed with HSE.

259

Approval Holders Supplied by Supplier A i. Confirmation as to whether the impurities mentioned in the 1991 ‘Purity and by-

products of technical active’ data sheet are present in currently produced technical active dichlorvos, as these impurities are not examined in the ‘Method for the determination of by-products and supplementary tests for dichlorvos’ report dated 1993. There is also no reference to the above-mentioned impurities in the ‘Chemical Analysis of 5 Representative batches of Dichlorvos Technical’ report dated 1993 (Unpublished, 1993k). If the impurities are still present, analytical methods for detection and quantification are required.

ii. An analytical method to estimate with adequate reliability dichlorvos in water at 0.2 µg l-1 (chromatograms and full test reports should be submitted).

iii. Freezing point study (a limit test e.g. down to 0 oC, will be acceptable). iv. Evidence (such as test report) for vapour pressure value. v. Evidence (such as test report) for flash point value. vi. Evidence (such as test report or reasoned case) for explosivity statement. viii. The oxidising properties should be addressed: a reasoned case may be sufficient. Approval Holders Supplied by Supplier B i. Mass Spectrum data for the active ingredient (stating purity). This information

should include a copy of the spectrum, with full operating details such as spectrometer, sample inlet system and ionisation mode and assignment of peaks.

ii. Validation for analytical method to determine dichlorvos in technical active ingredient, and chromatograms and typical results for dichlorvos in the technical material.

iii. Validation data for the method of analysis to determine dichlorvos in aerosol products. iv. Confirmation of the stabiliser mentioned in the ‘Analysis of DDVP’ report. v. Method validation for impurities and reasons to support allocation of peaks to column

cyclisation and β-elimination products.

vi. An analytical method to estimate with adequate reliability dichlorvos in water at 0.2 µg l-1 (chromatograms and full test reports should be submitted).

vii. Freezing point study (a limit test e.g. down to 0 oC, will be acceptable).

260

viii. Reasoned case to support applicability of flash point stated in The Pesticide Manual to dichlorvos manufactured by Supplier B.

ix. The explosivity properties should be addressed: a reasoned case may be sufficient. x. The statement regarding the oxidising properties should be supported: a reasoned case

may be sufficient. Approval Holders Supplied by Supplier C i. 1H NMR data on the active ingredient (stating purity). This information should

include a copy of the spectrum, with full operating details such as spectrometer, frequency, solvent and reference material, and assignment of peaks.

ii. Validation for analytical methods to determine dichlorvos in technical active ingredient and to determine impurities in technical dichlorvos.

iii. Chromatograms and typical results for the determination of dichlorvos in the formulation (strip). Validation data for the method of analysis to determine dichlorvos in strips.

v. An analytical method to estimate with adequate reliability dichlorvos in water at 0.1 µg l-1 (chromatograms and full test reports should be submitted).

v. Evidence (such as test report) should be submitted in support of the freezing point. vi. Evidence (such as test report) should be submitted in support of the flash point. vii. The explosivity properties should be addressed: a reasoned case may be sufficient. viii. The oxidising properties should be addressed: a reasoned case may be sufficient. ix. More information on the ‘normal conditions of ambient temperature and humidity’

(i.e. what was the temperature and the humidity or where was the location of the study and the time of year) during the strip storage stability study. More information on the analysis. Confirmation of the composition of the strip. Observations of the physical state and integrity of the strip, and of the colour and weight of the strip. If these details are not available, a new 2-year storage stability study at ambient temperature is required; protocol to be agreed with HSE.

261

Mammalian Toxicity The following data should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months. i. A developmental neurotoxicity study. Operator and Consumer Exposure The following data should be submitted within 1 year. Approval Holders must demonstrate to HSE how they are implementing these requirements within 2 months. Protocols to be agreed with HSE. i. Information on air levels found in a range of museum settings.

The following data should be submitted in regard to the use of slow release strips in pheromone traps, in areas where food may be stored, prepared or consumed, ii. Information on consumer exposure from food residues associated with the use of slow

release strips in pheromone traps, in areas where food may be stored, prepared or consumed.

Efficacy The following data should be submitted within 3 years. Approval Holders must demonstrate to HSE how they are implementing these requirements within 6 months. Protocols to be agreed with HSE. i. Either simulated use or field data, or a reasoned case based on data, to be submitted in

support of strips against Coleoptera, Formicoid Hymenoptera and Hymenoptera (as appropriate to current label claims).

262

7 REFERENCES All identifying product names have been blocked out. PHYSICAL CHEMISTRY Anonymous, (1980). Organophosphorus Pesticides in River and Drinking Water: A Tentative Method. Anonymous a, (1997). The Pesticide Manual, British Crop Protection Council and The Royal Society of Chemistry. Ed. CDS Tomlin Unpublished, (1987a). Determination of Dichlorvos in Nuvan Staykill by Gas Chromatography, GC 208/A Unpublished, (1987b). Water Solubility of DDVP Unpublished, (1987c). Octanol/water Partition Coefficient of DDVP Unpublished, (1988). DDVP: Physical and Chemical Characteristics Unpublished, (1990). Vapour pressure and Henry’s Law Constant of DDVP Unpublished, (1991a). Purity and by-products of technical active ingredient Unpublished, (1991b). Data Sheet, C177 Dichlorvos Physico-chemical properties, PP 2.56 NB/vp Unpublished, (1992a). Method for the Determination of Dichlorvos in Dichlorvos Technical Unpublished, (1992b). Storage Stability Data for Nuvan Staykil (A-5162) Unpublished, (1993a). Chemical Analysis of 5 Representative Batches of Dichlorvos Technical Unpublished, (1993b). Report on Boiling Point/Boiling Range of Dichlorvos Technical Unpublished, (1993c). Report on Density of Dichlorvos Technical Unpublished, (1993d). Report on Water Solubility of Dichlorvos Unpublished, (1993e). Report on Octanol/Water Partition Coefficient of Dichlorvos Unpublished, (1993f). Report on Validation of Analytical Method AW-1/5 for Dichlorvos Technical Unpublished, (1993g). Method for the Determination of By-products and Supplementary Tests for Dichlorvos Technical

263

Unpublished, (1993h). Report on Validation of Analytical Method AK-1/2 for by-products in Dichlorvos Technical Unpublished, (1993i). Analysis of DDVP Unpublished, (1993j). DDVP Chemical Identification Unpublished, (1993k). No-Pest Strip Dichlorvos (DDVP) Analytical Procedure Quantitative Analysis By GC Unpublished, (1994). Insect Strip ZA 9- Storage Stability, 66a/94 Unpublished, (1995a). Agro Base Set- C 177 A Unpublished, (1995b). Report on Surface Tension of Aqueous Solutions Unpublished, (1995c). Product Identity of Technical DDVP Unpublished, (1995d). DDVP: Measurement of Boiling Point Unpublished, (1995e). DDVP: Measurement of Relative Density Unpublished, (1995f). DDVP-Surface Tension Unpublished, (1995g). Quantitation of Technical DDVP Unpublished, (1995h). DDVP Technical: Analytical Method Validation in Freshwater Unpublished, Taylor N, (1997a). Determination Of Vapour Pressure By Balance Method. Unpublished, Heath J, (1997b). Statement of Composition Technical Grade DDVP Unpublished, Brady KT, (1997c). DDVP Technical: Preliminary Analysis, Product identity and Impurity Identification Unpublished, (1999a). Dichlorvos Technical, Volume AA1, Identity and Analysis Unpublished, Krips HJ, (1999b). Determination of the Boiling Temperature of Dichlorvos, Unpublished, Wouters GAJM, (1999c). Determination of solubility, density and partition coefficient of dichlorvos Unpublished, Krips HJ, (1999d). Determination Of The Surface Tension Of An Aqueous Solution of Dichlorvos Unpublished, (undated a). Spectral Data for Dichlorvos Technical Unpublished, (undated b). Denka analytical Method No.31- Gas chromatographic content determination of DDVP in DDVP technical and liquid formulations

264

Unpublished, (undated c). Stability of dichlorvos (DDVP) incorporated into resin strips sealed inside foil pouches MAMMALIAN TOXICOLOGY AND TOXICOKINETICS Adler B, Braun R, Schoeneich J & Böhme H. (1976). Repair-defective mutants of proteus mirabilis as a prescreening system for the detection of potential carcinogens. Biologisches Zentralblatt. 95, 463-469 Alldrick AJ & Rowland IR. (1985). The resistance of e. coli cultivated in low concentrations of dichlorvos to n-methyl-n'-nitro-nitrosoguanidine induced mutagenesis. Mutation Research. 142, 1-4 Anderson BE, Zeiger E, Shelby MD, Resnick MA, Gulati DK, Ivett JL & Loveday KS. (1990). Chromosome aberration and sister chromatid exchange test results with 42 chemicals. Environmental and Molecular Mutagenesis, Supplement 18, 16, 55-137 Andreu H, Velazquez A, Xamena N, Creus A & Marcas R, (1988). Induction of polygenic mutations affecting viability in drosophila melanogaster after dichlorvos malathion and ethyl methanesulfonate treatments. Mutation Research. 203, 230 Aquilina G, Benigni R, Bignami M, Calcagnile C, Dogliotti E, Falcone E & Carere A, (1984). Genotoxic activity of dichlorvos, trichlorfon and dichloroacetaldehyde. Pesticide Science, 15, 439-442 Ashwood-Smith MJ, Trevino R & Ring R, (1972). Mutagenicity of dichlorvos. Nature, 240, 418-420 Astolfi E, Donnewald H, Monserrat J & Calabrese A (1970). The lung in organophosphoric pesticide poisoning. Poumon Coeur, 26, 1067-1073 Bagriacik EU & Unlu H, (1991). Mutagenic effects of dichlorvos (ddvp) in drosophila melanogaster. DOGA - Turkish Journal of Biology, 15, 108-113 Baksi SN. (1978). Effect of dichlorvos on embryonal and fetal development in thyroparathyroidectomized, thyroxine-treated and euthyroid rats. Toxicology Letters, 2, 213-221 Bedford CT & Robinson J, (1972). The alkylating properties of organophosphates. Xenobiotica, 2, 307-337 Benford DL, Price SC, Lawrence JN, Grasso P & Bremmer JN (1994). Investigations of the genotoxicity and cell proliferative activity of dichlorvos in mouse forestomach. Toxicology 92, 203-215 Bidanset JH, (1978). Two unusual suicides involving pesticides as the fatal agent. Clinical Toxicology, 12(5), 615

265

Bignami M, Aulicino F, Velcich A, Carere A & Morpurgo G (1977). Mutagenic and recombinogenic action of pesticides in aspergillus nidulans. Mutation Research. 46, 395-402 Bignami M, Conti L, Morpurgo G & Velcich A, (1976). Comparative analysis of different test systems for somatic recombination with aspergillus nidulanss. Mutation Research, 38, 138-139 Bisby JA & Simpson GR (1975). An unusual presentation of systemic organophosphorus poisoning. Medical Journal of Australia, 2(10), 394-395 Blair D, Hoadley EC & Hutson DH, (1975). The distribution of dichlorvos in the tissues of mammals after its inhalation or intravenous administration. Toxicology and Applied Pharmacology, 31, 243-253 Borkowska J, Kasperek W, Szucki B & Tyburczyk W, (1989). Action of low doses of organophosphoric pesticides in rat organism. ii. Effect of dermal administration of owadofos and nogos. Bromatologia i Chemia Toksykologiczna, 21, 288-229 Boyer AC, Brown LJ, Slomka MB & Hine CH. (1977). Inhibition of human plasma cholinesterase by ingested dichlorvos: effect of formulation vehicle. Toxicology and Applied Pharmacology, 41, 389-394 Bradway DE, Shafik TM & Lores EM. (1977). Comparison of cholinesterase activity, residue levels, and urinary metabolite excretion of rats exposed to organophosphorus pesticides. Journal of Agriculture and Food Chemistry. 25, 1353-1358 Braun R, Schoeneich J, Weissflog L & Dedek W, (1982). Activity of organophosphorus insecticides in bacterial tests for mutagenicity & DNA repair/direct alkylation vs. metabolic activation & breakdown: i. butonate, vinylbutonate, trichlorfon, dichlorvos, demethyl dichlorvos & demethyl vinylbutonate. Chem-Biology Interactions, 39, 339-350 Breau AP, Mitchell WM & Swinson J, (1985). Mutagenic and cell transformation activities of representative phosphorothioateesters in vitro. Journal of Toxicology and Environmental Health, 16, 403-413 Bridges BA (1978). On the detection of volatile liquid mutagens with bacteria: experiments with dichlorvos and epichlorohydrin. Mutation Research, 78, 367-371 Bridges BA, Mottershead RP, Green MHL & Gray WJH, (1973). Mutagenicity of dichlorvos and methyl methane sulphonate for Escherichia coli wp2 and some derivatives deficient in DNA repair. Mutation Research, 19, 295-303 Bryant DH, (1985). Asthma due to insecticide sensitivity. Australia and New-Zealand Journal of Medicine, 15, 66-68 Brzezinski J & Wysocka-Paruszewska B, (1980). Neurochemical alterations in rat brain as a test for studying the neurotoxicity of organophosphorus insecticides. Archives of Toxicology Supplement, 4, 475-478

266

Buselmaier W, Roehrborn G & Propping P, (1972). Mutagenicity investigations with pesticides in the host-mediated assay and the dominant lethal test in mice. Biology Zentralbl, 91, 311-325 Buselmaier W, Roehrborn G & Propping P, (1973). Comparative investigations on the mutagenicity of pesticides in mammalian test systems. Mutation Research, 21(1), 25-26 Carere A, Cardamone G, Ortali V, Bruzzone ML & Giuseppe GD, (1976). Mutational studies with some pesticides in streptomyces coelicolor and Salmonella typhimurium. Mutation Research, 38 (2), 136 Carere A, Ortali VA, Cardamone G & Morpurgo G, (1978a). Mutagenicity of dichlorvos and other structurally related pesticides in salmonella and streptomyces. Chem Biology Interact, 22, 297-308 Carere A, Ortali VA, Cardamone G, Torracca AM & Raschetti K, (1978b). Microbiological mutagenicity studies of pesticides in vitro. Mutation Research, 57, 277-286 Caroldi S & Lotti M, (1981). Delayed neurotoxicity caused by a single massive dose of dichlorvos to adult hens. Toxicology Letters, 9, 157-159 Casebolt DB, Leary SL, & Undeutsch L (1990). Effects of dichlorvos treatment on mouse reproduction Lab Anim Sci 40 (1), 65-7 Casida JE, Mcbride L & Niedermeier RP, (1962). Metabolism of 2,2-dichlorovinyl dimethyl phosphate in relation to residues in milk and mammalian tissues. Journal of Agricultural and Food Chemistry, 10, 370-377 Cavagna G, Locati G & Vigliani EC, (1969). Clinical Effects of Exposure to DDVP (Vapona) Insecticide in Hospital Wards. Archive of Environmental Health, 19, 112-123 Cavagna G, Locati G & Vigliani EC, (1970). Exposure of newborn babies to vapona insecticide. European Journal of Toxicology, 3 (1), 49-57 Chu BCF & Lawley PD, (1975). Increased urinary excretion of nucleic acid and nictotinamide derivatives by rats after treatment with alkylating agents. Chem Biology Interact, 10, 333-338 Cockrell BY, Hall BV & Simon J, (1972). Morphometric analysis of anthelmintic - and prednisolone-induced alterations in the ultrastruture of beagle liver parenchymal cells. American Journal of Veterinary Research, 33, 687-698 Cohen SD & Ehrich M, (1974). Cholinesterase and carboxylesterase inhibition by dichlorvos (ddvp) and interactions with malathion (mal) and thiothocresyl phosphate (tcp). Pharmacologist, 16, 220 COM (2002). COM Statement COM/02/S2: Mutagenicity of dichlorvos URL: http://www.advisorybodies.doh.gov.uk/Com/dichlorvos.htm [17 December 2003]

267

Dambska M, Iwanowski L & Kozlowski P, (1979). The effect of transplacental intoxication with dichlorvos on the development of cerebal cortex in newborn rabbits. Neuropatol Pol, 17, 571-575 Dambska M, Iwanowski L, Maslinska D & Ostenda M, (1984). Blood-brain barrier in young rabbit brain after dichlorvos intoxication. Neuropatol Pol, 22, 129-137 Dambska M & Maslinka D, (1988). Morphological changes after acetylcholinesterase (ache) inhibition by dichlorvos (ddvp) in young rabbit brain. Journal Hirnforsch, 29, 569-571 Dambska M & Maslinka D, (1982). Effect of dichlorvos (DDVP) intoxication of rabbit brain. Neuropatol Pol, 20, 77-84 Dambska M, Maslinska D & Olszewska K, (1981). Effect of prolonged intoxication with dichlorvos (DDVP) on cerebal cortex of young rabbits. Zwierzeta Laboratoryjne, 18, 47-51 Dambska M, Maslinska D, Rakowska I & Rutczynski M, (1978). Effect of dichlorvos poisoning in pregnant rabbits and rats on the condition and growth of progeny. Bromatologia i Chemia Toksykologiczna, 11, 355-357 Das YT, Taskar PK, Brown HD & Chattopadhyay SK, (1983). Exposure of professional pest control operator to dichlorvos (DDVP) and residue on house structures. Toxicology Letters, 17, 95-99 De Ferrari M, Artuso M, Bonassi S, Conatti S, Cavalieri Z, Pescatore D, Marchini E, Pisano V & Abbondandolo A, (1991). Cytogenetic biomonitoring of an italian population exposed to pesticides: chromosome aberration and sister-chromatid exchange analysis in peripheral blood lymphocytes. Mutation Research, 260, 105-113 Dean BJ, (1972). The Mutagenic Effects of OP Pesticides on Micro-organisms. Archiv fur Toxikologie, 30, 67-74 Dean BJ, Doak SMA & Funnell J, (1972). Genetic studies with dichlorvos in the host-mediated assay and in liquid medium using Saccharomyces Cerevisiae. Arch Toxikol, 30, 61-66 Dean BJ & Blair D, (1976). Dominant lethal assay in female mice after oral dosing with dichlorvos or exposure to atmospheres containing dichlorvos. Mutation Research, 40, 67-72 Dean BJ & Thorpe E, (1972). Cytogenetics studies with dichlorvos in mice and chinese hamsters. Archiv fur Toxikologie, 30, 39-49 Degraeve N, Chollet MC & Moutschen J, (1984a). Cytogenetic and genetic effects of subchronic treatments with organophosphorus insecticides. Archives of Toxicology, 56, 66-67 Degraeve N, Chollet MC & Moutschen J, (1984b). Evaluation of the mutagenic potential of four commercial mixtures of insecticides. Food and Chemical Toxicology, 22, 683-687 Degraeve N, Chollet MC & Moutschen J, (1984c). Cytogenetic effects induced by organophosphorus pesticides in mouse spermatocytes. Toxicology Letters, 21, 315-319

268

Dellinger JA, Mckiernan BC, Koroitz GD & Richardson BC, (1987). Latent dichlorvos neurotoxicoty detected by vagal tone monitoring in dogs. Neurotoxicology and Teratology, 9 (3), 197-201 Deschamps D, Questel E, Baud FL, Gervais P & Dally S, (1994). Persistent asthma after acute inhalation of organophosphate insecticide [letter]. Lancet, 344, 1712 Desi I & Nagymajtenyi L (1988). Neurotoxicologic investigations of the pesticide dichlorvos (DDVP). Effects on the central and peripheral nervous system. Toxicology, 49, 141-148 Desi I & Nagymajtenyi L, (1999). Electrophysiological biomarkers of an organophosphorous pesticide, dichlorvos. Toxicol Lett , 107, 55-64 Desi I, Nagymajtenyi L, Papp A & Schulz H, (1998). Experimental model studies of pesticide exposure. Neurotoxicology, 19, 611-616 Desi I, Varga L & Farkas I, (1980). The effect of ddvp, an organophosphorus pesticide on the humoral and cell-mediated immunity of rabbits. Archives of Toxicology Supplement 4, 171-174 Desi I, Varga L & Farkas I, (1978). Studies on the immunosuppressive effect of organochlorine and organophosphoric pesticides in subacute experiments. Journal of Hygiene, Microbiology and Immunology, 22, 115-122 Dicowsky L & Morello A, (1971). Glutathione dependent degradation of 2,2 dichlorovinyl dimethyl phosphate (DDVP) by the rat. Life Science, 10, 1031-1037 Dikshith TSS, Datta KK & Chandra P, (1976). 90-Day dermal toxicity of ddvp in male rats. Bulletin of Environmental Contamination and Toxicology, 15, 574-578 Doherty AT, Ellard S, Parry EM & Parry JM, (1996). Induction of micronuclei and non-disjunction in binucleate human lymphoblastoid cells by trichlorphon and dichlorvos. Mutation Research, 360, 250 D'Souza AV & Batra BK, (1976). A study of the effects of DDVP (propoxure) on laboratory mouse. Current Science. 45 (10), 377 Dyer KF & Hanna PJ, (1973). Comparative mutagenic activity and toxicity of triethylphosphate and dichlorvos in bacteria and drosophila. Mutation Research, 21, 175-177 Dzwonkowska A & Hubner H (1986). Induction of chromosomal aberrations in the Syrian hamster insecticides tested in vivo. Arch Toxicol, 58, 152-156 Dzwonkowska A & Hubner H, (1991). Studies on commercial insecticides with the dominant lethal mutations test. Pol J Occup Med, 4, 43 -53

269

Dzwonkowska A, Hubner H & Porebska I, (1989). In vitro study of the influence of pesticides positive or negative in the ames test on sce and satellite associations. Mutation Research, 216, 310 Ehrich M, Jortner BS & Padilla S (1995). Comparison of the relative inhibition of acetylcholinesterase and neuropathy target esterase in rats and hens given cholinesterase inhibitors. Fundamental and Applied Toxicology, 24, 94-101 Ellinger C, (1985). Enzyme activities after in vitro and in vivo application of dichlorvos. Biomed Biochim Acta, 44, 311-316 Ellman GL, Courtney KD, Andres V & Feather-Stone RM (1961). A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol, 7, 88-95 El-Shimy NM, Mohamed ZA & Ismail SA, (1983). Biochemical studies on the effect of dichlorvos on rats in a short-term feeding study. Egyptian Journal of Food Science. 10, 57-65 Fatranska M, Vargová M, Rosival L, Bátora V, Némoth S & Janeková D, (1978). Circadian susceptibility rhythms to some organophosphate compounds in the rat. Chronobiologia, 5, 39-44 Francis BM, Metcalf RL & Hansen LG, (1985). Toxicity of organophosphorus esters to laying hens after oral and dermal administration. Journal of Environmental Health, B20, 73-95 Fujita K, (1985). Studies on contact dermatitis from pesticides in tea growers. Acta Med Univ Kagoshima, 27 (1), 17-37 Funckes AJ, Miller S & Hayes WJ, (1963). Initial field studies in upper volta with dichlorvos residual fumigant as a malaria eradication technique. no 3 toxicological evaluation. Bulletin of WHO, 29, 243 Fytizas R, (1970). Influence of three solvents on the acute and subacute cutaneous toxicity o dimethyl-2,2-dichlorvos phosphate (DDVP). Ann Inst Phytopathol Benaki, 9 (3), 245-258 Gadamski R & Squmanska G, (1977). Effect of poisoning with organophosphorus pesticide dichlorvos (DDVP) on the acetylcholinesterase (ACHE) activity of rat brain. Neuropath Pol, 15, 537-544 Gajewski D & Katiewicz M, (1981). Activity of certain enzymes and histomorphological changes in subacute intoxication of rats with selected organophosphates. Acta Physiologica Polonica, 32 (5), 507-520 Gilot-Delhalle J, Colizzi A, Moutschen J & Moutschen-Dahmen H, (1983). Mutagenicity of some organophosphorus compounds at the ADE6 locus of Schizoaccharomyces pombe. Mutation Research, 117, 139-148

270

Gold RE, Holcslaw T, Tupy D & Ballard JB, (1984). Dermal and respiratory exposure to applicators and occupants of residences treated with dichlorvos (DDVP). Journal of Economic Entomology, 77, 430-436 Gratz NG, Bracha P & Carmichael A, (1969). A village-sacle trial with dichlorvos as a residual fumigant insecticide in southern Nigeria. Bulletin of WHO, 29, 251 Green MH, Medcalf AS & Stevens SW, (1974). Apparent indirect dna damage by dichlorvos and iodoacetamide in Escherichia coli. Heredity, 33, 446 Green MHL, Medcalf ASC, Arlett CF, Harcourt SA & Lehmann AR, (1974b). DNA strand breakage cause by dichlorvos, methyl methanesulphonate and iodacetamide in Escherichia coli and cultured chinese hamster cells. Mutation Research, 24, 365-378 Green MHL, Muriel WJ & Bridges BA, (1976). Use of a simplified fluctuation test to detect low levels of mutagens. Mutation Research, 38, 33-41 Griffin DE & Hill WE, (1978). In vitro breakage of plasmid DNA by mutagens and pesticides. Mutation Research, 52 (2), 161-169 Guerzoni ME, Del Cupolo L & Ponti I, (1976). Mutagenic activity of pesticides. RIV SCI Tecnol Alimenti Nutr Um. 6, 161-165 Gupta AK & Singh J, (1974). Dichlorvos (DDVP) induced breaks in the salivary gland chromosomes of Drosophila melanogaster. Current Science, 43, 661-662 Haley TJ, Farmer JH, Harmon JR & Dooley KL, (1975). Estimation of the LD1 and extrapolation of the LD(0.1) for five organophosphate pesticides. Archives of Toxicology, 34, 103-109 Hanna PJ & Dyer KF, (1975). Mutagenicity of organophosphorus compounds in bacteria and drosphila. Mutation Research, 28, 405-420 Hass DK, Collins JA & Kodama JK, (1972). Effects of Orally Administered Dichlorvos in Rhesus Monkeys. Journal of American Veterinary Association, 161, 714-719 Hazelwod JC, Stefan GE & Bowen JM. (1978). Motor unit irritability in beagles before and after excposure to cholinesterase inhibitors. American Journal of Veterinary Research, 40, 852-56 Heddle JA, Dean S, Nohmi T, Boerrigter, Casciano D, Douglas GR, Glickman BW, Gorelick NJ, Mirsalis JC, Martus H-J, Skopek TR, Thybaud V, Tindall KR & Yajima N (2000). In vivo transgenic mutation assays. Environmental and Molecular Mutagenesis, 35, 253-259 Hinz V, Blokland A, Van der Staay F, Gebert I, Shcuurman T & Schmidt B, (1996). Receptor interaction profile and CNS general pharmacology of metrifonate and its transformation product dichlorvos in rodents. Drug Dev Res, 38, 31-42 Hodgson E & Casida JE, (1962). Mammalian enzymes involved in the degradation of 2,2-dichlororvinyl dimethyl phosphate. Journal of Agricultural and Food Chemistry, 10, 208-214

271

Horn KH, Teichmann B & Schramm T, (1987). Studies of dichlorvos (DDVP) i. Testing of dichlorovs for carcinogenic activity in mice. Archiv Fur Geschwulstforschung, 57, 353-360 Horn KH, Teichmann B & Schramm T, (1988). Studies in dichlorvos (DDVP). ii. Testing of dichlorvos for carcinogenic activity in rats. Archiv Fur Geschwulstforschung, 58, 1-10 Hussain SS, Rao JV, Aziz A, Jabeen Q & Mustafa M, (1982). Acute inhalation of dichlorvos to albino rats. International Pest Control, 24, 64, 66-74 Hutson DH & Hoadley EC, (1972). The metabolism of [14C-methyl] dichlorvos in the rat and the mouse. Xenobiotica, 2, 107-116 Hutson DH & Hoadley EC, (1972a). The comparative metabolism of [14C-vinyl] dichlorvos in animals and man. Archiv fur Toxikologie, 30, 9-18 Hutson DH, Hoadley EC & Pickering BA, (1971). The metabolic fate of [vinyl-1-14C]dichlorvos in the rat after oral and inhalation exposure. Xenobiotica, 1, 593-611 Hyde KM, Crandall JC, Kortman KE & McMoy WK, (1978). EEG, ECG and respiratory response to acute insecticide exposure. Bulletin of Environmental Contamination and Toxicology, 19 (1), 47-55 Institoris L, Nagymajtenyi L, Siroki O & Desi I, (1997). Comparison of the immuno- and neurotoxicological effects of repeated small doses of an organophosphate pesticide, DDVP, in three generations of rats. Neurotoxicology, 18, 898 Ishidate M, Sofuni T & Yoshikawa K, (1981). Chromosomal aberration tests in vitro as a primary screening tool for environmental mutagens and/or carcinogens. Gann Monogrpahs on Cancer Research, 27, 95-108 Jadhav RK, Sharma VK, Rao GJ, Singh RK & Chandra H, (1989). Distribution of nuvan (DDVP) in autopsy tissues and fluids in a fatal poisoning case. Journal of Indian Academy of Forensic Science, 28, 50-52 Jayasuriya VU & Ratnayake WE, (1973). Screening of some pesticides on Drosophila melanogaster for toxic and genetic effects. Drosophila Information Service, 50, 184-186 Jentzsch R & Fischer GW, (1974). Organophosphorus compounds iv. Kinetic studies on the alkylating properties of insecticidal phosphate esters. Journal fur Praktische Chemie, 316 (2), 249-258 Jian T & Zhiying F, (1990). Chronotoxicological studies on dichlorvos in mice and humans. Journal of Environmental Science, 2, 83-89 Johnson MK, (1978). The anomalous behaviour of dimethyl phosphates in the biochemical test for delayed neurotoxicity. Archives of Toxicology, 41 (1), 107-110 Kandeel KM, Mostafa IY, Zayed SM & Aly MAS, (1987). Effect of carbon-14-labeled dichlorvos on the nucleic acids in mice. Isotope and Radiation Research, 19, 51-59

272

Kawachi T, Yahagi T, Tazima T, Kada T, Ishidate M, Sasaki M & Sugiyama T, (1980). Cooperative programme on short-term assays for carcinogenicity in Japan. IARC Science Publications, 27, 323-330 Kimbrough RD & Gaines TB, (1968). Effect of organic phosphorus compounds and alkylating agents on the rat fetus. Archives of Environmental Health, 16 (6), 805-808 Kirkland VL, Albert JR & van Kampen K, (1974). Studies on dichlorvos induced intestinal enteritis in the dog. Toxicology and Applied Pharmacology, 29, 157 Kligerman AD, Erexson GL & Wilmer JL, (1985). Induction of sister-chromatid exchange (SCE) and cell-cycle inhibition in mouse peripheral blood B lymphocytes exposed to mutagenic carcinogens in vivo. Mutation Research, 157, 181-187 Kobayashi H, Yuyama A & Chiba KI, (1986). Cholinergic system of brain tissue in rats poisoned with the organophosphate, O,O-dimethyl O-(2,2-dichlorovinyl) phosphate. Toxicology and Applied Pharmocology, 82, 29-32 Kramers PGN & Knapp AGAC, (1978). Absence of a mutagenic effect after feeding dichlorvos to larvae of Drosophila melanogaster. Mutation Research, 57, 103-105 Kraus JF, Schröder R & Wulff E, (1970). Unusual Course in a Case of Poisoning by an Insecticide (Demetol-Concentrate VAP). Deutsche Medizinische Wochenschrift, 95, 2615-2617 Krause W, (1977). Influence of DDT, DDVP and malthion on FSH LH and testosterone serum levels and testosterone concentration in testis. Bulletin of Environmental Contamination and Toxicology, 18 (2), 231-240 Krause W, Hamm K & Weissmueller J, (1975). The effect of peroral application of DDVP and malathion on spermatogenesis and leydig cells of the juvenile rat. Andrologia, 7, 109-116 Krause W, Hamm K & Weissmueller J, (1976). Damage to spermatogenesis in juvenile rats treated with DDVP and malathion. Bulletin of Environmental Contamination and Toxicology, 15, 458-460 Krause W & Homola S, (1974). Alterations of the seminiferous epithelium and the leydig cells of the rat testis after the application of dichlorvos (DDVP). Bulletin of Environmental Contamination and Toxicology, 2, 429-443 Kurinniyi AI, (1975). Comparative study of the cytogenetic effect of certain organophosphorus pesticides. Sov Genet, 11, 1534-1538 Kusic R, Jovanovic D, Randjelovic S, Joksovic D, Todorovic V, Boskovic B, Jokanovic M & Vojvodic V, (1991). HI-6 in man: Efficacy of the oxime in poisoning by organophosphorus insecticides. Human and Experimental Toxicology, 10, 113-118 Lewis PJ, Lowing RK & Gompertz D (1981). Automated discrete kinetic method for erythrocyte acetylcholinesterase and plasma cholinesterase. Clin Chem, 27, 926-929

273

Lawley PD, Shah SA & Orr DJ, (1974). Methylation of nucleic acids by 2,2 dichlorovinyl dimethyl phosphate (dichlorvos, DDVP). Chem Biology Interact, 8, 171-182 Leary JS, Keane WT, Fontenot C, Feichtmeir EF, Schultz D, Koos BA, Hirsch L, Lavor EM, Roan CC & Hine CH, (1974). Safety evaluation in the home of polyvinyl chloride resin strip containing dichlorvos (DDVP). Archives of Environmental Health, 29, 308-314 Li C, Miller WT & Jiang J, (1989). Pulmonary edema due to ingestion of organophosphate insecticide. American Journal of Roentgenology, 152, 265-266 Lin SY, Lee TC, Cheng CS & Wang TC, (1988). Cytotoxicity, sister-chromatid exchange, chromosome aberration and transformation induced by 2,2-dichlorovinyl-0,0-dimethyl phosphate. Mutation Research, 206, 439-445 Löfroth G, (1970). Alkylation of DNA by Dichlorvos. Naturwissenschaften. Löfroth G, Kim C & Hussain S, (1969). Alkylating property of 2,2-dichlorovinyl dimethyl phosphate: a disregarded hazard. EMS Newsletter, 2, 21-27 Löfroth G, (1978). The mutagenicity of dichloroacetaldehyde. Znaturforsch, Teil-C-Biosci, 33, 783-785 Löfroth G & Wennerberg R, (1974). Methylation of purines and nicotinamide in the rat by dichlorvos. Z Naturforsch Sectc Bio Sci, 29, 651 Luty S, Latuszyska J, Halliop J, Tochman A, Obuchowska D, Przylepa E, Kprczak E & Bychawski E, (1998). Toxicity of dermally absorbed dichlorvos in rats. Annals of Agriculture and Environmental Medicine, 5, 57-64 Majeeth MA, Marimuthu KM & Gopinath PM, (1989). Cytogenetic and fetotoxic effects of organophosphate pesticides on mice. Environmental and Molecular Mutagenesis, 14 (suppl 15), 122 Majewski D, Podgorski W & Michalowska R, (1979). Retention of dichlorvos (DDVP) in rabbits. PoL Arch. Weter, 21, 249-255 Majewski T, Raczkiewicz J, Podgorski W, Saba L, Bialkowski Z, Polonis A & Rzacynski G, (1979). The influence of DDVP (dichlorvos) on rabbits reproduction. Medycyna Weterynaryjna, 35, 242-224 Markowitz SB, (1992). Poisoning of an urban family due to misapplication of household organophosphate and carbamate pesticides. Clinical Toxicology, 30, 295-303 Maslinska D, Lewandowska I & Prokopczyk J, (1981). Effect of prolonged acetylcholesterase inhibition on postnatal brain development in rabbit. i. level of serotonin in different brain regions. Acta Neurapathol Suppl (Berl), 7, 52-55 Maslinska D, Rutczynski M & Maslinska S, (1980). Histoenzymic study of motoneurons in spinal cord of rat treated with dichlorvos (DDVP). Zwierzeta Laboratoryjne, 17, 27-33

274

Maslinska D, Strosznajder J, Zalewska T & Orlewski P, (1984). phospholipid-protein ratio in brain of suckling rabbits treated with an organophosphorus compound. International Journal of Tissue Reactions, 6, 317-322 Maslinska D & Zalewska Z, (1978a). Effect of dichlorvos, administered to the pregnant rabbits, on the cholinesterase activity in the progeny. Folia Histochem Cytochem, 16, 335-340 Maslinska D & Zalewska Z, (1978b). Activity of some mitochondrial enzymes in the progeny of rabbits treated with dichlorvos in the gestation period. Folia Histochemica et Cytochemica, 16, 139-140 Maslinska S, Ciurzynska G & Maslinska D, (1979). Dichlorvos intoxication and gastric secretion. Acta Physiologica Polinica, 30, 657-660 Mason HJ (2000). The recovery of plasma cholinesterase and erythrocyte acetylcholinesterase activity in workers after over-exposure to dichlorvos. Occup Med, 50, 343-347 Mathias CGT, (1983). Persistent contact dermatitis from the insecticide dichlorvos. Contact Dermatitis, 9, 217-218 Matsushita T, Aoyama A, Yoshimi K, Fujita Y & Ueda A, (1985). Allergic contact dermatitis from organophosphorus insecticides. Industrial Health, 23, 145-153 McFee AF & Tice R, (1986). Interlaboratory evaluation of a standardized protocol for determination of chemical gernotoxicity in mice. Mamm Chromosomes Newsl, 27, 16 Mehl A, Schanke TM, Johnsen BA & Fonnum F, (1994). The effect of trichlorfon and other organophosphates on prenatal brain development in the guinea pig. Neurochem Res, 19, 569-574 Menz M, Luetkmeier H & Sachsse K, (1974). Long-term exposure of factory workers. Archives of Environmental Health, 28, 72-76 Michalek SM & Brockman HE, (1969). A test for mutagenicity of shell 'NO-PEST strip insecticide' in Neurospora crassa. Neurospora Newsletter, 14, 8 Mirsalis JC, Tyson CK, Steinmetz KL, Loh EK, Hamilton CM, Baleke JP & Spalding JW, (1989). Measurement of unscheduled dna synthesis and s-phase synthesis in rodent hepatocytes following in vivo treatment: testing of 24 compounds. Environmental and Molecular Mutagenesis, 14, 155-164 Modak AT, Weintraub ST & Stavinoha WB, (1974). Dichlorvos and the cholinergic system in the brain. Pharmacologist, 16, 241 Mohn G, (1973). 5-Methyltryptophan resistance mutations in Escherichia coli K-12. mutagenic activity of monofunctional alkylating agents including organophosphorus insecticides. Mutation Research, 20, 7-15

275

Moody SB & Terp DK, (1988). Dystonic reaction possibly induced by cholinesterase inhibitor insecticides. Drug Intelligence and Clinical Pharmacology, 22, 311-312 Moriya M, Kato K & Shirasu Y, (1978). Effects of cysteine and a liver activation system on the activities of mutagenic pesticides. Mutation Research, 57, 259-263 Moriya M, Ohta T, Watanabe K, Miuazawa T, Kato K & Shirasu Y, (1983). Further mutagenicity studies on pesticides in bacterial reversion assay systems. Mutation Research, 116, 185-216 Morpurgo G, Aulicino F, Bignami M, Conti L & Velcich A, (1977). Relationship between structure and mutagenicity of dichlorvos and other pesticides. Atti Accad Naz Lincei Cl Sci Fis Mat Nat Rend, 62, 692-701 Morpurgo G, Bellincampi D, Bualandi F, Baldinelli L & Crescenzi OS, (1979). Analysis of Mitotic Non-Disjunction with Aspergillus nidulans. Environmental Health Perspectives, 31, 81-95. Morris-Brown L, Blair A, Gibson R, Everett GD, Cantor KP, Schuman LM, Burmeister LF, van Lier SF & Dick F, (1990). Pesticide exposures and other agricultural risk factors for leukemia among men in Iowa and Minnesota. Cancer Research, 50, 6585-6591 Moutschen-Dahmen J,Moutschen-Dahmen M, Degraeve N, (1981). Metrifonate and dichlorvos: cytogenetic investigations. Acta Pharmacologica et Toxicologica, 49 (5), 29-39 Myhr B, McGregor D, Bowers L, Riach C, Brown AG, Edwards I, McBride D, Martin R & Caspary WJ, (1990). L5178Y mouse lumphoma cell mutation assay results with 41 compounds. Environmental and Molecular Mutagenesis 16, (supplement 18), 138-167 Nag M & Nandi N, (1991). Effect of three organophosphate on respiration in rat brain and liver tissue. Bioscience Reports, 11, 7-10 Nagy Z, Mile I & Antoni F, (1975). The mutagenic effect of pesticides on Escherichia coli WP2 TRY-. Acta Microbiol Acad Sci Hung, 22 (3), 309-314 Nagymajtenyi L, Desi I & Lorencz R, (1988). Neurophysiological markers as early signs of organophosphate neurotoxicity. Neurotoxicology and Tetraology, 10, 429-434 Naidu KA, Viswanatha S & Krishnakumari MK, (1987). Cardiotoxic effects of dichlorvos (DDVP) in albino rats. Indian Journal of Physiology and Pharmacology, 31, 19-24 Natoff IL, (1970). Influence of the route of exposure on the acute toxicity of cholinesterase inhibitors. Journal of European Toxicology, 3, 363-367 NCI, (1977). Bioassay of Dichlorvos for Possible Carcinogenicity. National Cancer Institute. Nehez M, Toth C & Desi I, (1994). The effect of dimethoate, dichlorvos, and parathionmethyl on bone marrow cell chromosomes of rats in subchronic experiments in vivo. Ecotoxicol Environ Saf, 29, 365-371

276

Nicholas AH, Vienne M, Van den Berghe H, (1978). Sister chromatid exchange frequencies in cultured human cells exposed to an organophosphorus insecticide: dichlorvos. Toxicology Letters. 2, 271-275 Nishio A & Uyeki ED, (1982). Nuclease-sensitivity of methylated dna as a probe for chromatin reconstitution by genotoxicants. Biochemical Biophysical Research Communications, 107, 485-491 Nishio A & Uyeki ED, (1981). Induction of sister chromatid exchanges in chinese hamster ovary cells by organophosphate insecticides and their oxygen analogs. Journal of Toxicology and Environmental Health, 8, 939-946 NTP 342, (1989). Toxicology and Carcinogenesis Studies of Dichlorvos (CAS No. 62-73-7) in F344/N Rats and B6C3F1 Mice (Gavage Studies). Technical Report Series: No 432 Olinski R, Walter Z, Wiaderkiewicz R, Lukásová E & Palecek E, (1980). Changes in DNA properties due to treatment with the pesticides malathion and DDVP. Radiation and Environmental Biophysics, 18, 65-72 Oshiro Y, Piper CE & Balwierz PS (1991). Chinese hamster ovary cell assays for mutation and chromosome damage: data from non-carcinogens. Journal of Applied Toxicology, 11, 167-177 Ottevanger CF, (1975). Neuromuscular function in persons exposed to dichlorvos resin strips value of electromyography. Journal of European Toxicology, 8, 192-195 Pachecka L, Grzadkowska K & Gajewski K, (1980). The activity of plasma phosphatases in rats treated with single doses of dichlorvos and trichlorphon. Acta Physiol Pol, 31, 555-559 Pachecka J, Sulinski A & Ziolkowska G, (1975). The activities of some esterases of the rat brain after intoxication by organophosphate insecticides, dichlorvos and trichlorphon. Neuropatol Pol, 13, 455-462 Paik SG & Lee SY, (1977). Genetic effects of pesticides in the mammalian cells. 1. induction of micronucleus. Korean Journal of Zoology, 20 (1), 19-28 Pauluhn J, Machemer L, Kimmerle G, (1987). Effects of inhaled cholinesterase inhibitors on bronchial tonus and on plasma and erythrocyte acetylcholine esterase activity in rats. Toxicology, 46, 177-190 Perocco P & Fini A, (1980). Damage by dichlorvos of human lymphocyte DNA. Tumori, 66, 425-430 Pletsa V, Steenwinkel MJST, van Delft JHM, Baan RA & Kyrtopoulos SA (1999). Induction of somatic mutations but not methylated DNA adducts in λlacZ transgenic mice by dichlorvos. Cancer Letters, 146, 155-160 Purshottam T & Kaveeshwar U, (1979). Effect of diet on dichlorovinyl dimethyl phosphate toxicity in rats. Aviation Space and Environental Medicine, 50, 581-584

277

Reeves JD, Driggers DA & Kiley VA (1981). Household insecticide associated aplastic anaemia and acute leukaemia in children. Lancet, 2(8241), 300-301 Reiner E & Plestina R, (1979). Regeneration of cholinesterase activities in humans and rats after inhibition by O,O-dimethyl-2,2-dichlorovinyl phosphate. Toxicology and Applied Pharmacology, 49, 451-454 Reiner E, Simeon V & Skrinjaric-Spoljar M, (1980). Hydrolysis of O,O-dimethyl-2,2- dichlorovinyl phosphate (DDVP) by esterases in parasitic helminths, and in vertebrate plasma and erythrocytes. Comp Biochem Physiol C, 66, 149-152 Robens JF, (1978). Tests for possible carcinogenicity of 20 pesticides in osborne-mendel rats and B6C3F1 mice. Toxicology and Applied Pharmacology, 45 (1), 236 Rosenkranz HS, (1973). Preferential effect of dichlorvos (Vapona) on bacteria deficient in DNA polymerase. Cancer Research, 33, 458-459 Rosenkranz HS & Rosenkranz S, (1972). Reaction of dna with phosphoric acid esters, gasoline additive and insecticides. Experientia, 28, 396-397 Ryhäenen R, Honkakoski P, Harri M, Ylitalo P & Hännine O, (1988). Effect of the cold environment on organophosphate toxicity and inhibition of cholinesterase activity. Gen Pharmacol, 19, 741-745 Saito M, Chiba M & Isahai I, (1981). Two cases of pesticide intoxication. Japanese Journal of Legal Medicine, 34 (3), 431 Sarin S & Gill KD (2000). Biochemical characterisation of dichlorvos-induced delayed neurotoxicity in rat. Life, 49, 125-130 Sasaki YF, Sekihashi K, Izumiyama F, Nishidate E, Saga A, Ishida K & Tsuda S, (2000). The comet assay with multiple mouse organs: Comparison of comet assay results and carcinogenicity with 208 chemicals selected from the IARC monographs and US NTP carcinogenicity database. Critical Reviews in Toxicology, 30 (6), 629-799 Sasaki M, Sugimura K, Yoshida MA & Abe S, (1980). Cytogenetic effects of 60 chemicals on culturted human and chinese hamster cells. La Kromosoma, 20, 574-584 Sasinovich LM, (1967). Disorders of some functions of the liver in DDVP intoxication. Vrachebrioe Delo, 10, 93-95 Schmidt H, Schmidt M & Nenner M, (1975). Inhibition of acetylcholinesterase in the bronchial system of rats caused by inhalation of dichlorvos. Nauyn Schmiedeberg's Archives of Pharmacology. 287, R87 Schop RN, Hardy MH & Goldberg MT, (1990). Comparison of the activity of topically applied pesticides and the herbicide 2,4-D in two short-term in vivo assays of genotoxicity in the mouse. Fundamental and Applied Toxicology, 15, 666-675

278

Schulz H, Nagymajtenyi L & Desi I, (1995). Life-time exposure to dichlorvos affects behaviour of mature rats. Hum Exp Toxicol, 14, 721-726 Schwetz BA, Ioset HD, Leong BKJ & Staples RE. (1979). Teratogenic potential of dichlorvos given by inhalation and gavage to mice and rabbits. Teratology, 20, 383-388 Segerback D, (1981). Estimation of genetic risks of alkylating agents. v. methylation of DNA in the mouse by DDVP (2,2-dichlorovinyl dimethyl phosphate). Hereditas, 94, 73-76 Shinoda S, Ito Y, Matsunaga T, Takada T & Nomura R, (1972). Two fatal cases of acute pesticide intoxication. Nippon Noson Igakkai Zasshi, 21 (2), 242-243 Shirasu Y, Moriya M, Kato K, Furuhashi A & Kada T, (1976). Mutagenicity screening of pesticides in the microbial system. Mutation Research, 40, 19-30 Shooter KV, (1975). Assays for phosphotriester formation in the reaction of bacteriophage R17 with a group of alkylating agents. Chem-Biol Interact, 11, 575-588 Sidhu KS & Collisi MB (1989). A case of an accidential exposure to a veterinary insecticide product formulation. Vetinary and Human Toxicology, 31, 63-64 Sim LP, Eng NT, Hock QS. (1980). Insecticide baygon aerosol poisoning - a report of 5 cases. Journal of Singapore Paediatrians Society, 22, 1-4 Sitkiewicz D & Zalewska Z, (1975b). The activity of cytochrome oxidase and succinate dehydrogenase in rat brain mitochondria following trichlorphon and dichlorvos intoxication. Neuropat Pol, 13, 273-280 Slomka MB & Hine CH, (1981). Clinical pharmology of dichlorvos. Acta Pharmacologica et Toxicologica, 49, 105-108 Smith PW, Mertens H, Lewis MF, Funkhouser GE, Higgins EA, Crane CR, Sanders DC, Endecott BR & Flux M, (1972). Toxicology of dichlorvos at operational aircraft cabin altitudes. Aerospace Medicine, 43 (5), 473-478 Snow DH, (1973). The acute toxicity of dichlorvos in the dog. ii. pathology. Australian Veterinary Journal, 49, 120-125 Snow DH & Watson ADJ, (1973). The acute toxicity of dichlorvos in the dog. i. clinical observations and clincial pathology. Australian Veterinary Journal, 49, 113-119 Sobels FH & Todd NK. (1979). Absence of a mutagenic effect of dichlorvos in Drosophila melanogaster. Mutation Research, 67, 89-92 Soininen H, Unni L & Shillcutt S, (1990). Effect of acute and chronic cholinesterase inhibition on biogenic amines in rat brain. Neurochemical Research, 15, 1185-1190 Sprague GL, Sandvick LL, Brookins-Hendricks MJ & Bickford AA (1981). Neurotoxicity of two organophosphate ester flame retardants in hens. Journal of Toxicology and Environmental Health, 8, 507-518

279

Srivastava AK, Raina R, Chaudhary RK & Malik JK, (1989). Acute toxicity and biochemical alterations in rats after single oral exposure to dichlorvos. Pesticide, 23, 35-40 Srivastava AK & Malik JK, (1988). Effect of dichlorvos on tissue esterases in rats. Journal of the Indian Institute of Science, 68, 23-27 Stavinoha WB, Modak AT & Weintraub ST, (1976). Rate of accumulation of acetylcholine in discrete regions of the rat brain after dichlorvos treatment. Journal of Neurochemistry, 27, 1375-1378 Stein WJ, Miller S, Fetzer LE, (1966). Studies with dichlorvos residual as malaria eradication technique in haiti. iii. Toxicological studies. American Journal of Tropical Medicine, 15 (5), 672-675 Stoermer D (1985). [Occupational contact dermatitis from the pesticide dichlorvos.] in German. Dermatol. Monatsschr, 171 (4), 245-249 Takahashi H, Kojima T & Ikeda T, (1991). Differences in the mode of lethality produced through intravenous and oral administration of organophosphorus insecticides in rats. Fundamental and Applied Toxicology, 16, 459-468 Teichert K, Szymczyk T, Consolo S & Ladinsky H, (1976). Effect of acute and chronic treatment with dichlorvos on rat brain cholinergic parameters. Toxicology and Applied Pharmacology, 35, 77-81 Teichert-Kuliszewska K & Szymczyk T, (1979). Changes in rat carbohydrate metabolism after acute and chronic treatment with dichlorvos. Toxicology and Applied Pharmacology, 47, 323-33 Tezuka H, Ando N, Suzuki R, Terahata M, Moriya M & Shirasu Y, (1980). Sister-chromatid exchanges and chromosomal aberrations in cultured chinese hamster cells treated with pesticides positive in microbial reversion assays. Mutation Research, 177-191 Thorpe E, Wilson AB, Dix KM & Blair D, (1972). Teratological studies with dichlorvos vapour in rabbits and rats. Archiv fur Toxikologie, 30, 29-38 Tice RR, Agurell E, Anderson D, Burlinson B, Hartmann A, Kobayashi H, Miyamae Y, Rojas E, Ryu J-C & Sasaki YF (2000). Single cell gel/Comet assay: Guidelines for in vitro and in vivo genetic toxicology testing. Environmental and Molecular Mutagenesis, 35, 206-221 Timmons EH, Chaklos RJ, Bannister TM & Kaplan HM, (1975). Dichlorvos effects on estrous cycles onset in the rat. Laboratory Animal Science, 25, 45-47 Torracca AM, Cardamone GC, Ortali V, Carere A, RAschetti R & Ricciardi G, (1976). Mutagenicity of pesticides as pure compounds and after metabolic activation with rat liver microsomes. Atti Assoc Genet Ital, 21, 28-29 Traverso R, Moretto A & Lotti M, (1989). Human serum 'A'-esterases. Hydrolysis of 0,0-dimethyl-2,2-dichlorovinyl phosphate. Biochemical Pharmacology, 38, 671-676.

280

Tungul A, Bonin AM, He S & Baker RSU. (1991). Micronuclei induction by dichlorvos in the mouse skin. Mutagenesis, 6, 405-408 United States National Institute for Occupational Safety and Health (US NIOSH) (1977). Agricultural chemicals and pesticides: A subfile of the Registry of Toxic Effects of Chemical Substances. Utsumi T & Miyata M, (1977). Biochemical studies on a case of degenerative pigmentosa retinae suspected to be due to chronic intoxication by organophosphorus pesticides. Japanese Review of Clinical Ophthalmology, 1 (5), 550-555 Van Bao T, Szabo I, Ruzicska P & Czeizel A, (1974). Chromosome aberrations in patients suffering acute organic phosphate insecticide intoxication. Humangenetik, 24, 33-57 Velazquez A, Andreu H, Xamena N, Creus A & Marcos R, (1987). Accumulation of drastic mutants in selection lines for resistance to the insecticides dichlorvos and malathion in Drosphila melanogaster. Experimentia, 43, 1122-1123 Voogd CE, Jacobs J, Van Der Stel JJ, (1972). Mutagenic action of dichlorvos. Mutation Research, 16, 413-416 Wadia RS, Chitra S, Amin RB, Kiwalkar RS & Sardesai HV, (1987). Electrophysiological studies in acute organophosphate poisoning. Journal of Neurology, Neurosurgery, and Psychiatry, 50, 1442-1448 Wagner JE & Johnson DR, (1970). Toxicity of dichlorvos for laboratory rats - LD50 and effect on serum cholinesterases. Laboratory of Animal Care, 20, 45-47 Wang T, Wu CT, Lin J & Tarn CY, (1988). Dichlorvos potentiates the insecticide-induced sister chromatid exchanges in chinese hamster ovary cells. Bull Inst Zool Academia Sinica, 27, 111-117 Ward FP & Glicksberg CL, (1971). Effects of dichlorvos on blood cholinesterase activity in dogs. Journal of American Veterinary Medical Association, 158, 457-461 Watanabe H, Shirai O, Ebata N, (1976). A Case Report of Organophosphorus Insecticide (DDVP) Intoxication with Respiratory Intensive Care. Pesticide Abstracts, 9 (2), 94-95 Wennerberg R & Lofroth G, (1974). Formation of 7-methylguanine by dichlorvos in bacteria and mice. Chem Biol Interactions, 8, 339-348 Wild D, (1973). Chemical induction of streptomycin-resistant mutations in Escherichia coli: dose and mutagenic effects of dichlorvos and methyl methanesulphonate. Mutation Research, 19 (1), 33-41 Wilhelm K & Bradamante V, (1980). Blood cholinesterase activity in workers exposed to cholinesterases, a ten-year follow up. Arh Hig Rada Toksikol, 31, 109-124

281

Witherup S, Jolley WJ, Stemmer K & Pfitzer EA, (1971). Chronic toxicity studies with 2,2-dichlorovinyl dimethyl phosphate (DDVP) in dogs and rats including observations on rat reproduction. Toxicology and Applied Pharmacology, 19, 377 Wooder MF & Wright AS, (1981). Alkylation of dna by organophosphorus pesticides. Acta Pharmacologica et Toxicologica, 49 (5), 51-55 Wyrobek AJ & Bruce WR, (1975). Chemical induction of sperm abormalities in mice. Proceedings of the National Academy for Science USA, 72. 114, 425-429 Xiao-ou S, Fan L, Youchun J, Xiaorong L, Peihou Z & Yuqing L (1997). Mutagenicity of 19 organophosphorus pesticides in Saccharomyces cerevisiae D61.M. Chinese Journal of Pharmacology and Toxicology, 11 (4), 291-293 Yamano T, (1996). Dissociation of DDVP-induced DNA strand breaks from oxidative damage in isolated rat hepatocytes. Toxicology, 108, 49-56 Yamashita M, Tanaka J & Ando Y (1997). Human mortality in organophosphate poisonings. Veterinary and Human Toxicology, 39, 84-85 Yamazaki H, Shimazaki K, Kurosawa K & Suzuki A, (1975). Electron microscopic study on the subacute toxicity of organophosphorus pesticide. Nippon Noson Igakkai Zasshi (Journal Japanese Association of Rural Medicine), 24 (1), 1-7 Yoshikawa H, Yoshida M, Ueda T, Yatsugi K & Hara I, (1991). Evaluation of exposure to organophosphorus insecticides and its effect on ocular function in pest control operators. Japanese Journal of Toxicology, 4, 65-71 Yoshikawa H, Yoshida M & Hara I, (1990). Electroretinographic changes induced by organophosphorus pesticides in rats. The Journal of Toxicological Sciences, 15, 87-95 Zalewska Z, Rakowska I, Matraszek G & Sitkiewicz D, (1977a). Effect of dichlorvos on some enzyme activities of the rat brain during postnatal development. i. cholinesterases. Neuropatol Pol, 15, 255-262 Zalewska Z, Wolna B, Sitkiewicz D & Bicz W (1977b). Effect of dichlorvos on some enzyme activities of the rat brain during postnatal development. ii. oxidoreductases. Neuropatol Pol, 15, 367-370 Zavon MR & Kindel EA (1966). Potential hazard in using dichlorvos insecticide resin. Adv Chem Ser, 60, 177-186 Zeiger E, Anderson B & Haworth S, (1988). Salmonella mutagenicity tests. 4.results from the testing of 300 chemicals. Environmental and Molecular Mutagenesis, 11 (12), 1-158 Zuyao N, Shouqi L, Yuqing L, Ying T & Dinguo P, (1993). Induction of micronucleus by organophosphorus pesticides in vivo and in vitro. Journal of the West China University of Medical Sciences, 24 (1), 82-86

282

Unpublished, (1962). 90-Day chronic toxicity studies of vapona insecticide for dogs Unpublished, Witherup S, Stemmer KL & Roell M (1964a). The immediate toxicity of vapona insecticide resin vaporizer Unpublished, (1964b). The effects upon rats, of being fed on diets containing vapona insecticide Unpublished, (1966). The anti-acetylcholinesterase activity of technical DDVP when administered percutaneously to guinea pigs Unpublished, (1967a). Determination of the minimal incipient toxicity of dichlorvos in humans Unpublished, (1967b). The effects exerted upon rats, during a period of two years, by the introduction of vapona insecticide into their daily diets Unpublished (1967c). The effects upon beagle dogs during a period of two years by the introduction of vapona insecticide into their daily diets Unpublished, (1968a). Experimental evaluation in guinea pigs of skin sensitising potential of components of formulated dichlorvos /polyvinyl chloride products Unpublished, (1968b). The toxicity of low concentrations of dichlorvos of inhalation in rodent and avian species Unpublished (1969a). Report on the determination of the oral LD50 to the rat of DDVP, technical Unpublished (1969b). Report on the determination of acute dermal LD50 to the rat of DDVP, technical Unpublished, (1969c). A preliminary report on the inhalation toxicity of high concentrations of dichlorvos Unpublished, (1969d). Report on initial of deliberate exposures to high concentrations of dichlorvos by human subjects Unpublished, (1969e). Teratology studies in rabbits Unpublished (1970a). Report on the Determination of the Acute Oral LD50 to the Rabbit of DDVP, Technical Unpublished (1970b). Report on the determination of the acute intraperitoneal LD50 to the mouse of DDVP, technical Unpublished, (1970c). Dichlorvos; inhalational exposures with human subjects part 2 Unpublished, (1970d). Dichlorvos; inhalational exposures with human subjects

283

Unpublished (1971a). Effects of dichlorvos on human cells in tissue culture Unpublished (1971b). Report on Mutagenic Effect of Technical DDVP Unpublished, (1971c). Teratogenic studies in rats and rabbits given dichlorvos by inhalation Unpublished (1973) Acute oral LD50 of technical dichlorvos in the rat Unpublished (1976). Two-year inhalation exposure of rats to dichlorvos vapour Unpublished, (1977). Cholinesterase activity and neuromuscular function of rhesus monkeys exposed to DDVP vapours Unpublished, (1979a). Determination of the acute oral toxicity of DDVP-technical product sample 1 in rats Unpublished (1979b). Determination of the acute oral toxicity of DDVP-technical product sample 2 in rats Unpublished, (1979c). Studies on the effects of dichlorvos on the integrity of rat liver DNA in vitro Unpublished, (1982a). The acute (4h) inhalation toxicology of dichlorvos vapour in rats and mice Unpublished (1984a), Dichlorvos (L15/20, DDVP) - investigations on acute inhalation toxicity Unpublished, (1984b). L15/20 DDVP-technical material. Investigations on irritation/corrosiveness to skin and eye, rabbit Unpublished, (1985a). A micronucleus test in the mouse using dichlorvos Unpublished, (1985b). An in vivo sister chromatid exchange assay in mice with dichlorvos Unpublished, (1985c). A dominant lethal assay in mice with dichlorvos Unpublished (1986a). Assessment of eye irritation/corrosion by Denkavepon in the rat Unpublished (1986b). Acute oral toxicity in the rat Unpublished (1986c). Evaluation of the acute dermal toxicity of Denkavepon in the rat Unpublished (1986d). Evaluation of the acute inhalation toxicity of Denkavepon in the rat Unpublished (1986e). Assessment of primary skin irriation/corrosion by Denkavepon in the rabbit Unpublished (1986f). Acute dermal irritation/corrosion study in the rabbit

284

Unpublished (1986g). Assessment of eye irritation/corrosion by Denkavepon in the rat Unpublished (1986h). Assessment of the skin sensitisation potential of Denkavepon in the guinea-pig (split adjuvant test Unpublished, (1986i). L5178Y TK+/- mouse lymphoma forward mutation assay with dichlorvos Unpublished, (1987a). A dominant lethal assay in mice with dichlorvos Unpublished (1987b). Personal Communication from data holder Unpublished, (1987c). Studies of carcinogenicity on DDVP (2,2-dichlorovinyl dimethyl phosphate) - (rat) Unpublished, (1988a). DDVP: An acute delayed neurotoxic study in chickens Unpublished, (1988b). C 177 EC 500 (A-4113 B) Acute inhalation toxicity study - LC50 rats (4 hour exposure) Unpublished, (1988c). 13-week gavage toxicity study with DDVP in rats Unpublished, (1989). Metabolism of 14C-DDVP in rats (preliminary and definitive phases) Unpublished, (1990a). Dermal absorption of dichlorvos in rats Unpublished, (1990b). A 52-week chronic toxicity study on DDVP in dogs Unpublished, (1991a). Supplement to: Metabolism of 14C-DDVP in rats (preliminary and definitive phases) Unpublished (1991b). Personal Communication from National Poisons Unit. Unpublished, (1991c). Investigation of the Genotoxic and/or Irritant Effects of Dichlorvos on Mouse Forestomach Unpublished, (1991d). Developmental toxicity evaluation of DDVP administered by gavage to CD (Sprague-Dawley) rats Unpublished, (1991e). Developmental toxicity evaluation of DDVP administered by gavage to New Zealand White rabbits Unpublished, (1992a). In vivo cytogenetics assay: Analysis of chromosomal aberrations in bone marrow and spermatogonial cells following repeated dose administration Unpublished, (1992b). Investigation of the Irritant Effects of Dichlorvos on Mouse Forestomach Unpublished, (1992c). Two-generation reproductive toxicity study of DDVP administered in the drinking water to CD (Sprague-Dawley) rats

285

Unpublished, (1993a). An acute neurotoxicity study of dichlorvos in rats Unpublished, (1993b). A subchronic (13 week) neurotoxicity study of dichlorvos in rats Unpublished, (1994a). DDVP - 28-day neurotoxicity in the domestic hen Unpublished, (1994b). Neuropathological review of studies AVC/1 and RAD/2. Unpublished, (1997a). Dichlorvos: A study to investigate erythrocyte cholinesterase inhibition following oral administration to healthy male volunteers Unpublished, (1997b). Dichlorvos: A study to investigate the effect of a single oral dose on erythrocyte cholinesterase inhibition in healthy male volunteers Unpublished, (1997c). Dichlorvos: A single blind, placebo controlled, randomised study to investigate the effects of multiple oral dosing on erythrocyte cholinesterase inhibition in healthy male volunteers Unpublished, (1998). Pathology working group peer review of DDVP 28-day neurotoxicity study in the domestic hen OPERATOR AND CONSUMER EXPOSURE AND RISK ASSESSMENTS Batth SS, Singh J & Villeneum DC, (1973). Method for evaluation of dichlorvos vaporisers - simulated household use. Journal of Economic Entomology, 66(1), 146-150 Cavagna G, Locali G &Vigliana EC, (1970). Exposure of new-born babies to Vapona insecticide: European Journal of Toxicology, III, 49-57 Collins RD & DeVries DM, (1973). Air concentrations and food residues from use of Shell’s No-Pest insecticide strip. Bulletin of Environmental Contamination and Toxicology vol. 9 (4): 227-233 Deer HM, Beck ED & Roe AH, (1993). Respiratory exposure of museum personnel to dichlorvos insecticide. Vet Hum Toxicol., 35 (3), 226-228 ECETOC (2003). Assessment of Non-Occupational Exposure to Chemicals. Technical Report 58. European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels. Elgar KE & Steer BD, (1972). Dichlorvos concentrations in the air of houses arising from the use of dichlorvos PVC strips. Pesticide Science, 3, 591-600 Gold RE, Holcslaw T, Tupy D & Ballard JB (1984). Dermal and respiratory exposures to applicators and occupants of residences treated with dichlorvos (DDVP). J Econ. Entomol. 77(2), 430-436

286

Health & Safety Executive (2000). Occupational Exposure Limits 2000. Report EH40/2000. ISBN 0 7176 1730 0. HMSO, Norwich. Leary JS, Keane C, Fontenot EF et al, (1974. Safety evaluation in the home of polyvinyl chloride resin strip containing dichlorvos (DDVP, The Arizona (Tucson) study). Arch. Environ. Health 29: 308-314 Lewis RG, Bond AE, Johnson, DE & Hsu, JP (1998). Pilot study-measurement of atmospheric concentrations of common household pesticides. US EPA; Environmental Monitoring and Assessment, 10(1), 59-73 Unpublished, (undated). Summary information on dichlorvos emmision rates from units containing 73 g strips. Unpublished, (1979). The performance of modified dichlorvos/PVC slow-release matrices - a review of the 1973 and 1974 experimental programme Unpublished, (1980). A study of dichlorvos slow-release units used in museum cases for the protection of specimens against insect infestations, Tring Museum 1978-79 Unpublished, (1994). Determination of Airborne Concentration to Dichlorvos at Tring Zoological Museum, Tring Unpublished, (1997). Risk assessment for the private use of dichlorvos-impregnated strips in residences Unpublished, (1998). Operator Exposure To Non-Agricultural Pesticides Unpublished, (1999a). Dichlorvos-Secto Fly Killer Living Room Size Unpublished, (1999b). Patterns Of Use For Some Non-Agricultural Pesticide Products Unpublished, (2001). Consumer Exposure To Non-Agricultural Pesticide Products Unpublished, (2002). Dislodgeable Residues From Absorbent Surfaces EFFICACY Aamir MMI, (1983). Laboratory evaluation of some newer insecticides against Tribolium confusum Duv. (Coleoptera Tenebrionidae). Probleme de Protectia Plantelor, 8, (3), 173-182 Anonymous, (1961). Knockdown and Mortality of German Roaches Exposed to Contact Sprays. Technical Release - National Pest Control Association, U.S.A. Attia FI, (1976). Insecticide Resistance in Cadra cautella in New South Wales, Australia. Journal of Economic Entomology, 69, (6), 773-774 Bachmann F, (1960). Insecticidal effect of DDVP (0,0-dimethyl-2,2-dichlorovinyl phosphate. Anz. Schadlingsh, 33, (3), 41-44

287

Baker JM & Harris EC, (1969). Tests with Dichlorvos against the common furniture beetle Anobium punctatum. Deg. International Pest Control. July 1969, 17-23 Batth SS, Singh J & Villeneuve DC, (1973). Dichlorvos vapourizers: Method for the evaluation under simulated household use. Journal of Economic Entomology, 66 (1), 146-150 Benezet HJ, Huffman BB & Helms CW, (1988). Comparative toxicity of selected insecticides to the cigarette beetle at different temperatures. Tob. Sci. 32, 71-73 Boles HP, Bry RE & McDonald LL, (1974). Dichlorvos vapours: Toxicity to Larvae of the Furniture Carpet Beetle. Journal of Economic Entomology, 67, (2), 308-309 Das PK, et al., (1979). Susceptibility of larvae of Culex fatigans (Wiedmann), Anopheles stephensi (Liston) and Aedes aegypti (Linn.) to insecticides in Pondicherry. Indian J. Med. Res., 70, 412 Elgar KE & Steer BD, (1972). Dichlorvos concentrations in the air of houses arising from the use of dichlorvos PVC strips. Pesticide Science, 3, 591-600 Fletcher MG & Axtell RC, (1993). Susceptibility of the bedbug, Cimexlectularius, to selected insecticides and various treated surfaces. Med. Vet. Entomol., 7, (1), 69-72 Green AA, Kane J & Gradidge JMG, (1966). Experiments on the Control of Ephestia elutella (Hb) (Lepidoptera, Phycitidae) Using Dichlorvos Vapour. J. Stored Prod. Res., 2, 147-157 Hamilton, MA, Russo, RC & Thurston, RV, (1977). Trimmed Spearman-Karber method for estimating median lethal concentrations in toxicity bioassays. Environ. Sci. Technol., 11, 714-719 Harris EC, Elgar KE, Baker JM. & Wyles AG, (1970). Dichlorvos strips for furniture beetle control: A field trial. International Pest Control, July 1970, 28-33 Ho SH, Goh PM & Leong ECW, (1994). Toxicity of some organophosphorus and carbamate insecticides to Periplaneta americana (L.) and inhibition of non-specific esterases by the insecticides. International Pest Control, 36 (6), 154-155 Kano R, Cabrera BD, Hayashi A & Shinonaga S, (1977). Resistant levels of houseflies to six kinds of synthetic insecticides in the Philippines. Southeast Asian Journal of Tropical Medicine and Public Health, 8 (4), 515-518 Karnatak AK, Khare BP & Johari RK, (1991). Relative toxicity of some organophosphorus and synthetic pyrethroid insecticides against Sitophilus oryzae Linnaeus. Agric. Biol. Res. 7, (2), 91-100 Kerdpibule V & Hirakoso S, (1971). Susceptabilities to several insecticides in adult houseflies collected from various districts in Thailand. Japanese Journal of Experimental Medicine, 41, (6), 541-545

288

Kitagaki T, Nakayama, I, Sugiyama S & Otokozawa T, (1973). Effectiveness of Dowco 214 on some insects of public health importance-Laboratory evaluation tests for the common house fly. Japanese Journal of Sanitary Zoology, 24, (2), 103-109. Kobayashi Y, Ono Y, Okano T & Buei K, (1994). Insecticide susceptibility of the cat flea Ctenocephalides felis. Japanese Journal of Sanitary Zoology 45 (2), 121-128 Lee DK, Shin EH. & Shim JC, (1997). Insecticide susceptibility of Culexpipiens pallens (Culicidae, Diptera) larvae in Seoul. Korean Journal of Entomology 27, (1), 9-13 Maddock DR. & Sedlak VA, (1961). Dosage-mortality response of Anophelesquadrimaculatus exposed to DDVP vapour. Bulletin World Health Organisation, 24, 643-644 Maddock, DR, Sedlak VA & Schoof HF, (1961). Dosage-mortality response of Musca domestica exposed to DDVP vapour. Bulletin World Health Organisation, 24, 644-646 Mankowska H & Goszczynska K, (1969). Effectiveness of dichlorvos as a residual fumigant insecticide (Efektywnosc dichlorfosu jako trwalego fumiganta). Roczn. Panstw. Zakl. Hig., 20, 319-328 Miyagi I, Toms T, Zayasu N & Takashita Y (1994). Insecticide susceptibility of Culex quinquefasciatus larvae (Diptera, Culicidae) in Okinawa Prefecture, Japan in 1989. Japanese journal of Sanitary Zoology, 45, (1), 7-11 Pasalu IC & Bhatia SK, (1975). Laboratory evaluation of some insecticides against malathion-resistant and susceptible strains of Triboleum castaneum (Herbst). Bulletin of Grain Technology, 12, (3), 175-179 Rice PJ & Coats JR, (1994). Insecticidal Properties of Several Monoterpenoids to the House Fly (Diptera: Muscidae), Red Flour Beetle (Coleoptera: Tenebrionidae), and Southern Corn Rootworm (Coleoptera: Chrysomelidae). J. Econ. Entomol. 87, (5), 1172-1179 Schulten GGM & Kuyken W, (1966). Dichlorvos resin strips for the control of cocoa moth Ephestia elutella. International Pest Control, Vol: May/June, 18-23 Schwinghammer KA, et al., (1985). Comparative Toxicity of Ten Insecticides Against the Cat Flea, Ctenocephalides felis. J. Med. Entomol., 22, (5), 512-514 Shim JC, Lee DK & Lee KW, (1997). Insecticide Susceptability of German Cockroaches (Blattaria: Blattellidae) in Seoul. Korean Journal of Entomology, 27, (1), 73-77 Smittle BJ & Burden GS, (1965). Dichlorvos as a Vapor Toxicant for Control of Roaches, Bedbugs and Fleas. Pest Control, 33, (10), p 26-32 Strong RG & Sbur DE, (1968). Evaluation of Insecticides for Control of Stored-Product Insects. Journal of Economic Entomology, 61, (4), 1034-1041

289

Tuovinen T & Ekrom P, (1982). Laboratory evaluation of dust, spray and aerosol preparations for control of stored product insects. ANNales AGRICulturae FENNiae (Finland), 21, (4), 177-183 The U.K. Building Regulations-Approved Documents F1&F2 (Means of Ventilation and Condensation. HMSO, 1991. Ref. no. ISBN-0-11-752223-6 Zettler JL, (1991). Pesticide Resistance in Tribolium castaneum and T. confusum (Coleoptera: Tenebrionidae) from Flour Mills in the United States. Journal of Econ. Entom., 84 (3), 763-767 Unpublished, (1964). Vapour Toxicity of Vapona Unpublished, (1966). Vapona Ministrip Performance Unpublished, (1967). Testing for Suitability as Moth Protective of Shell Vapona Strip and Shell Vapona Ministrip Unpublished, (1971). Summary of data abstracted from Shell Company records relating to SHELLTOX-branded flying insect aerosol space sprays containing dichlorvos, developed circa. 1970 Unpublished, (1976). The performance of an experimental Sectovap lantern Unpublished, (1979). The Performance of Modified Dichlorvos/PVC Slow Release Matrices-A Review of the 1973 and 1974 Experimental Programme Unpublished, (1980). Comparison of 2 Sectovap Lanterns and 2 Spacesetter Lanterns for activity against flies Unpublished, (1981). Evaluation of Sectovap Kitchen Size and Fix Up vapourisers against flies Unpublished, (1986a). DDVP, iodophenphos, diazinon and permethrin: Ctenocephalides felis larvicide test Unpublished, (1986b). Ctenocephalides felis larvicide test Unpublished, (1987). A comparison of two insecticidal aerosol formulations for efficacy against houseflies by BS 4172 Unpublished, (1989). Efficacy Testing of Bop Insecticide-Laboratory bioassay on flying insects (Musca domestica and Aedes aegypti), and on crawling insects (Blattella germanica and Periplaneta americana) Unpublished, (1991a). Evaluation of a Vapona Slow Release Strip Against Houseflies. Unpublished, (1991b). Evaluation of a Vapona Slow Release Strip Against Mosquitoes

290

Unpublished, (1993a). Biological Efficacy of Insect Strip ZA 9 Unpublished, (1993b). The biological efficacy of Globol Moth Strip against the webbing clothes moth (Tineola bisselliella Unpublished, (1994). Evaluation of Five Aerosol Flyspray Formulations Against the Housefly M. Domestica and the Mosquito Ae. aegypti. Unpublished, (1995). Dichlorvos Slow Release Cassette Unpublished, (1996). Hunting Spray Test on Hornets - Shelltox Flykiller T1-86021, Sample No.17454 Unpublished, (1997a). Report on the efficacy of Secto Moth Killer Units and two competitor samples against the adults and larvae of the webbing clothes moth Tineola bisselliella. Unpublished, (1997b). Biological Efficacy Study of Vapona Moth Killer DDVP/PVC Strip Unpublished, (1999a). Dichlorvos-Secto Fly Killer Living Room Size Unpublished, (1999b). Dichlorvos-Secto Slow Release Fly Killer Kitchen Size Unpublished, (1999c). Dichlorvos-Secto Mini-Space Insect Unpublished, (Undated a). Results of biological tests carried out to evaluate the activity of dichlorvos-in-air concentrations of 0.1and 0.04mg m-3 against various species of clothes moths and carpet beetles Unpublished, (Undated b). Biological Tests of DDVP Dispensers

291

APPENDIX 1 A1.1 DATA REQUIREMENTS AND CONDITIONS OF APPROVAL RECOMMENDED BY THE ACP WHEN THE REVIEW OF DICHLORVOS IN PUBLIC HYGIENE AMATEUR INSECTICIDES WAS CONSIDERED IN 1994 Physical chemistry Approval holders whose manufacturing source is Supplier A: i. A surface tension study. Approval holders whose manufacturing source is Supplier B: i. Test report supporting boiling point, relative density, vapour pressure and water

solubility values quoted. ii. Spectral data (UV, IR, NMR and Mass Spectrum). iii. A surface tension and partition coefficient study. Mammalian toxicology and toxicokinetics i. No formulation toxicity studies were submitted for the review and consequently all approval holders must now explain the classification of their products for acute toxicity. This may be either by studies conducted using the formulation or by a reasoned case based on the toxicity of the components of the product. ii. Ministers have agreed that all products must carry the phrase ‘MAY CAUSE SENSITISATION BY SKIN CONTACT’ unless the approval holder can justify the removal of the classification by providing adequate data showing that the product is not a sensitiser. iii. A delayed neuropathy study in the adult hen conducted according to OECD Guidelines No. 418. Efficacy i. Data should be provided to support the continued use of slow release strips against Lepidoptera (moths) and Hymenoptera (ants and wasps). ii. Data should be provided to support the continued use of aerosols, (space spray, or

residual surface sprays). This should ideally be data on dichlorvos only products (or laboratory formulations).

These data should support the label claims (including target species and product use patterns) made by individual Approval Holders. Test protocols should be agreed with the Technical Secretariat.

292

A1.2 GENERIC LABELS FOR DICHLORVOS PRODUCTS A1.2.1 Amateur space and surface spray This is a typical example of an aerosol for both space and surface spray. It is for indoor use only. It contains tetramethrin and permethrin as well as dichlorvos. It is for use on hard porous surfaces. It is classified as R12: Extremely flammable and R43: May cause sensitisation by skin contact. FOR INDOOR USE ONLY. DO NOT BREATHE SPRAY MIST. DO NOT CONTAMINATE FOODSTUFFS, EATING UTENSILS OR FOOD CONTACT SURFACES. THIS MATERIAL AND ITS CONTAINER must be disposed of in a safe way. EXTREMELY DANGEROUS TO FISH AND OTHER AQUATIC LIFE. REMOVE OR COVER FISH TANKS AND BOWLS before application. *KEEP OFF SKIN. *WASH HANDS AFTER USE. *KEEP IN A SAFE PLACE, out of reach of children. *Pressurised container: protect from sunlight and do not expose to temperatures exceeding 50 °C. Do not pierce or burn even after use. *Keep away from sources of ignition - No smoking. *Do not spray on a naked flame, electrical equipment or any incandescent material. *UNTIL DRY, TREATED SURFACES CAN POSE A FIRE RISK. *This product contains an anticholinesterase organophosphorus compound. DO NOT USE if under medical advice NOT to work with such compounds. *THESE PHRASES SHOULD NOT APPEAR IN THE STATUTORY BOX. Please note that the phrases: ‘EXTREMELY DANGEROUS TO FISH AND OTHER AQUATIC LIFE’ and ‘REMOVE OR COVER FISH TANKS AND BOWLS before application’ are used due to the presence of the active ingredients, tetramethrin and permethrin and not dichlorvos.

293

A1.2.2 Professional space spray This is a typical example of an aerosol for space spray. It is for indoor use only. It contains tetramethrin as well as dichlorvos. It is classified as R12 : Extremely flammable and R43: May cause sensitisation by skin contact. FOR USE ONLY BY PROFESSIONAL OPERATORS The (COSHH) Control Of Substances Hazardous to Health Regulations 1999 may apply to the use of this product at work. Engineering control of operator exposure must be used where reasonably practicable in addition to the following items of personal protective equipment. However, engineering controls may replace personal protective equipment if a COSHH assessment shows they provide an equal or higher standard of protection. DO NOT CONTAMINATE FOODSTUFFS, EATING UTENSILS OR FOOD CONTACT SURFACES. WEAR SUITABLE PROTECTIVE SYNTHETIC RUBBER/PVC GLOVES when using. FOR INDOOR USE ONLY. THIS MATERIAL AND ITS CONTAINER must be disposed of in a safe way. DO NOT BREATHE SPRAY MIST. EXTREMELY DANGEROUS TO FISH AND OTHER AQUATIC LIFE. REMOVE OR COVER FISH TANKS AND BOWLS before application. *AVOID ALL CONTACT WITH SKIN. *WASH HANDS AND EXPOSED SKIN before meals and after use. * This product contains an anticholinesterase organophosphorus compound. DO NOT USE if under medical advice NOT to work with such compounds. *KEEP IN A SAFE PLACE, out of reach of children. *Pressurised container: protect from sunlight and do not expose to temperatures exceeding 50 °C. Do not pierce or burn even after use. *Keep away from sources of ignition - No smoking. *Do not spray on a naked flame, electrical equipment or any incandescent material. *THESE PHRASES SHOULD NOT APPEAR IN THE STATUTORY BOX. Please note that the phrases: ‘EXTREMELY DANGEROUS TO FISH AND OTHER AQUATIC LIFE’ and ‘REMOVE OR COVER FISH TANKS AND BOWLS before application’ are used due to the presence of the active ingredient tetramethrin and not dichlorvos.

294

A1.2.3 Professional slow release strip for use in museum display cases, cabinets and small cupboards This is a slow release strip. It is an impregnated strip to be cut to required size and is contained within a protective case, which is perforated to allow diffusion of vapour. An extraction cabinet should be used when handling/cutting the strips. Exposure to the dichlorvos strips should be limited to a maximum of 30 minutes per day. It is for use in closable museum display cases/cabinets/small cupboards, not for use in rooms/areas where people could be exposed continuously. It is classified as R43: May cause sensitisation by skin contact. FOR USE ONLY BY PROFESSIONAL OPERATORS. The (COSHH) Control of Substances Hazardous to Health Regulations 1999 may apply to the use of this product at work. DO NOT ALLOW TO COME INTO CONTACT with food or cooking utensils. THIS MATERIAL AND ITS CONTAINER must be disposed of in a safe way. USE ONLY in positions inaccessible to children and animals. DO NOT exceed the use of one unit per.......... cubic metre(s). DO NOT USE in larders or food cupboards. *This product contains an anticholinesterase organophosphorus compound. DO NOT USE if under medical advice NOT to work with such compounds. *WASH HANDS AFTER HANDLING. *KEEP IN A SAFE PLACE AWAY FROM CHILDREN. *THESE PHRASES SHOULD NOT APPEAR IN THE STATUTORY BOX. A1.2.4 Professional slow release strip for use in pheromone traps This is a typical example of a dichlorvos slow release strip. It is for use indoors in a pheromone trapping system/device. It is for use as a public hygiene insecticide for use in museums, galleries, warehouses and grain stores. It is not for use in domestic situations. It is classified as R43: May cause sensitisation by skin contact. It has the same safety phrases as slow release strips for museum uses

295

A1.3.5 Amateur slow release controllable and non-controllable cassettes This is a typical example of a controllable cassette. It is for indoor use only. It is used in domestic situations, but should not be used in rooms/areas where people could be exposed continuously, this is part of the pest and usage area statement and would usually be included in the 'Directions for use' section on the label. It is classified as R43 : May cause sensitisation by skin contact. DO NOT ALLOW TO COME INTO CONTACT with food or cooking utensils. THIS MATERIAL AND ITS CONTAINER must be disposed of in a safe way. USE ONLY in positions inaccessible to children and animals. DO NOT exceed the use of one unit per.......... cubic metre(s). DO NOT USE in larders or food cupboards. *This product contains an anticholinesterase organophosphorus compound. DO NOT USE if under medical advice NOT to work with such compounds. *WASH HANDS AFTER HANDLING. *KEEP IN A SAFE PLACE AWAY FROM CHILDREN. *THESE PHRASES SHOULD NOT APPEAR IN THE STATUTORY BOX. The above labelling would also be present on non-controllable cassettes.

296

APPENDIX 2 A2.1 STORAGE STABILITY Storage Stability Study Conducted On A Pre-Pressurised Aerosol Formulation (Unpublished, 1992a) A batch of aerosols was tested at intervals over a 3 year storage period at ambient temperature. The cans were weighed and cooled to -20 oC for 2 hours. They were then opened and the propellant was allowed to escape. The contents were then transferred to a volumetric flask and diluted with acetone. An aliquot was removed and transferred to another volumetric flask and an internal standard benzyl benzoate was added. The mixture was diluted and analysed by GC. No evidence of significant losses were observed at the 3 year stage. Details on emission parameters were not provided.

Storage Conditions Months Temperature (oC) Content (% w/w)

0 0.48 2 35 0.51 8 35 0.46 24 RT 0.54 36 RT 0.52

Storage Stability Study For A Resin Strip Formulation (Unpublished, undated c) Two separate batches of resin strips sealed inside foil pouches were stored at ambient temperature and humidity for 5 years. No evidence of significant losses was observed at either the 3 year stage or at the 5 year period.

Batch Number Date of analysis 187074 187077

Years

10:04:91 21.14 20.87 0.0 23:10:91 21.85 21.81 0.54 07:04:92 22.34 21.98 0.99 14:10:92 22.03 21.46 1.52 05:04:93 21.39 20.52 1.90 26:10:93 21.26 21.25 2.55 29:04:94 21.76 21.13 3.05 23:10:96 20.71 20.43 5.12

Details on the conditions of the study (such as the ambient temperature) and observations on the strip were not provided. Storage Stability Of Resin Strip Formulation (Unpublished, 1994) A six month accelerated storage stability trial was carried out on the strips, which were sealed in their aluminium foil packaging. The strips were analysed by GC following extraction.

297

No evidence of significant losses after 6 months at 30 oC. The approval holder believes that in their experience this is a guarantee of stability for 2 years at room temperature. This brief single statement is not considered adequate as a reasoned case and observations on the cassette were not provided. The composition of the cassette was not provided.

Storage time (months)

Temperature (oC) Breakdown of decomposed

DDVP (g)

Stabiliser available quantitatively

(% w/w) 1 30 0 100 1 40 0 100 3 30 0 98 3 40 0.1 87 6 30 0 93

Additional to the current data requirements, and therefore not required from other sources, 2 other studies related to the stability of dichlorvos were submitted: a study on the photolytic degradation of dichlorvos in water and methanol (Unpublished, 1970) and a study on the hydrolysis of dichlorvos in aqueous solutions at varying pH, temperature and ionic strength (Unpublished, 1981). The first study revealed that in water after a 6 hour period of illumination (minimum effective wavelength 290 nm) at approximately 20 °C there was 45 % reduction in the level of dichlorvos. The dichlorvos in the methanol solution was not decomposed. The hydrolysis study showed that at 30 °C the half-life decreased with increasing pH as the table below demonstrates. pH Half Life t½ (hours) 1 73.5 5 50.5 7 17.6 9 16.2 13 1.29 x 10-2

298

APPENDIX 4 A4.1 AEROSOL SPACE SPRAYING Task: Ready-for-use aerosol spray can discharged into air (space) Assumptions:

• operator respired volume 1.25 m3 h-1 (Unpublished, 2001) • in-use concentration 0.8 % • task duration 4 bursts of 6 seconds (space sprays) • user wears either light clothing or T-shirt and shorts and vacates room post-spraying • skin penetration value 100 %

Model: Indicative exposure values for contamination rates - aerosol space spray applications (mg min-1 and mg m-3):

Route Likelihood of route Central tendency Worst case Skin contamination 100 % frequency 193 mg min-1 569 mg min-1 Skin, weighted indicative value 193 mg min-1 Inhalation 100 % frequency 167 mg m-3 374 mg m-3 Inhalation, weighted indicative value 167 mg m-3 Models based on laboratory studies using Sr surrogate (Unpublished, 2001). Hand exposure accounted for a median 48 % of skin exposure. Exposure calculation:

Exposure descriptor Central tendency Worst case Potential body exposure rate 193 mg min-1 569 mg min-1 Body exposure rate (52 % of whole) 100 mg min-1 296 mg min-1 Clothing penetration 20 % 50 % Job duration 0.4 min 0.4 min Daily body exposure to product 8.0 mg 59.2 mg Hand exposure rate (48 % of whole) 93 mg min-1 273 mg min-1 Job duration 0.4 min 0.4 min Daily hand exposure to product 37 mg 109 mg Daily skin exposure to dichlorvos (0.8 %) 0.36 mg 1.35 mg Exposure via the skin (50 % penetration) 0.18 mg d-1 0.68 mg d-1 Inhalation exposure rate 167 mg m-3 374 mg m-3 Volume inhaled over job duration 0.01 m3 0.01 m3 Inhalation exposure to dichlorvos (0.8 %) 0.01 mg d-1 0.03 mg d-1 Daily exposure to dichlorvos 0.37 mg d-1 1.37 mg d-1 Daily exposure to dichlorvos, 60 kg operator 0.003 mg kg-1 0.012 mg kg-1

A4.2 AEROSOL SURFACE SPRAYING

299

Task: Ready-for-use aerosol spray can discharged onto surface Assumptions:

• operator respired volume 1.25 m3 h-1 (Unpublished, 2001) • in-use concentration 0.5 % • task duration 7 minutes (surface spray) • user wears either light clothing or T-shirt and shorts and vacates room post-spraying • skin penetration value 100 %

Model: Indicative exposure values for contamination rates - aerosol surface spray applications (mg min-1 and mg m-3):

Route Likelihood of route Central tendency Worst case Skin contamination 100 % frequency 57.2 mg min-1 139 mg min-1 Skin, weighted indicative value 57.2 mg min-1 Inhalation 100 % frequency 6.1 mg m-3 27.4 mg m-3 Inhalation, weighted indicative value 6.1 mg m-3 Models based on laboratory studies using Sr surrogate (Unpublished, 2001). Hand exposure accounted for a median 59 % of skin exposure. Exposure calculation:

Exposure descriptor Central tendency Worst case Potential dermal exposure rate 57.2 mg min-1 139 mg min-1 Body exposure rate (41% of whole) 23.5 mg min-1 57 mg min-1 Clothing penetration 20 % 50% Job duration 7 min 7 min Daily body exposure to product 33 mg 196 mg Hand exposure rate (59% of whole) 33.7 mg min-1 82 mg min-1 Job duration 7 min 7 min Daily hand exposure to product 236 mg 574 mg Daily skin exposure to dichlorvos (0.5 %) 1.34 mg 3.85 mg Exposure via the skin (50 % penetration) 0.67 mg d-1 1.93 mg d-1 Inhalation exposure rate 6.1 mg m-3 27.4 mg m-3 Volume inhaled over job duration 0.15 m3 0.15 m3 Inhalation exposure to dichlorvos (0.5 %) 0.005 mg d-1 0.02 mg d-1 Daily exposure to dichlorvos 1.35 mg d-1 3.87 mg d-1 Daily exposure to dichlorvos, 60 kg operator 0.011 mg kg-1 0.035 mg kg-1

A4.3 SECONDARY EXPOSURE (CHRONIC RESIDENTIAL EXPOSURE)

300

Scenario: Adults and infants inhaling dichlorvos for 18 h d-1 and ingesting dichlorvos residues with 50 % of meals, all taken in the home. Assumptions: 60 kg adult inhalation rate 18.5 m3 d-1 (Unpublished, 2001) 10 kg infant inhalation rate 4 m3 d-1 (Unpublished, 2001) adult food intake 1.75 kg d-1 (ECETOC Technical report 58) infant food intake 0.66 kg d-1 (ECETOC Technical report 58)

Adult

Exposure descriptor Central Tendency Worst Case Airborne concentration dichlorvos 0.02 mg m-3 0.06 mg m-3 Duration 18 h 18 h Inhalation rate 18.5 m3 d-1 18.5 m3 d-1 Respired volume 13.9 m3 13.9 m3 Daily inhalation exposure, dichlorvos 0.28 mg 0.83 mg Food concentration dichlorvos 0.1 ppm 0.1 ppm 50 % daily dietary intake 0.88 kg 0.88 kg Daily ingestion exposure, 0.1 ppm dichlorvos 0.088 mg 0.088 mg Total daily exposure to dichlorvos 0.37 mg 0.92 mg Daily systemic exposure to dichlorvos 0.006 mg kg-1 d-1 0.015 mg kg-1 d-1

Infant

Exposure descriptor Central Tendency Worst Case Airborne concentration dichlorvos 0.02 mg m-3 0.06 mg m-3 Duration 18 h 18 h Inhalation rate 4 m3 d-1 4 m3 d-1 Respired volume 3 m3 3 m3 Daily inhalation exposure, dichlorvos 0.06 mg 0.18 mg Food concentration dichlorvos 0.1 ppm 0.1 ppm 50 % daily dietary intake 0.33 kg 0.33 kg Daily ingestion exposure, 0.1 ppm dichlorvos 0.03 mg 0.03 mg Total daily exposure to dichlorvos 0.09 mg 0.21 mg Daily systemic exposure to dichlorvos 0.009 mg kg-1 d-1 0.02 mg kg-1 d-1

301

APPENDIX 5 A5.1 EFFICACY TEST METHODS A5.1.1 Test method NL-7200: Toxicity test for slow release insecticide strips against moths This method tests for the toxicity of slow release insecticide strips against adult moths, and the effect on egg - laying and newly hatched larvae in scaled-down wardrobes. The test is conducted in the dark (except when making observations) at 24 + 1 oC and 55 + 10 % r.h., inside untreated (plain) wooden boxes measuring 70 x 50 x 55 cm (0.19 m3). The top of each box incorporates a glass panel together with a wooden shutter to allow light to enter when required. The test insects used are adult clothes moths (Tineola bisselliella) from a long-established strain. For the exposure test, two pieces (both 10 x 10 cm folded in half) of dark woollen cloth (to facilitate counting white eggs and larvae) are used. These are treated with a blend of 10 % yeast and 1 % cholesterol and placed at the bottom of each box. The product to be evaluated is placed near the top of the box and 24 h are then allowed to elapse before approximately 250 adults are released into the box. A further box is kept in identical conditions, without the test product, to act as a control. Exposure tests are made initially, and after one, two, three and six months. The number of paralysed insects is counted every h for 8 h, and then at daily intervals for 3 d. The paralysed insects are picked up by a vacuum device and the pieces of wool transferred into jars in the rearing room in order to monitor the viability of the eggs. After the insects are picked up, the boxes are cleaned and ventilated. Between the exposure tests, the test product is stored in wooden boxes similar to the test boxes. The boxes are opened and closed daily (five times per week) to simulate a cupboard in use. The results are expressed as percentage adult mortality, number of eggs laid and number of live larvae found. A5.1.2 Test method SNV 195 901 (1971): Method for determining the resistance of wool and other keratin-containing materials to moths and beetles. This method is a more recent version of the methodology used in the feeding tests described in Section 5.3.1.4. Using the method, the moth and beetle resistance of a material is determined by comparison with a sample of a suitable control material which has been baited and exposed to attack by the test larvae. The standard applies to wool, or mixtures of wool and other fibres, as well as to other keratin-containing materials in all stages of processing, both before and after being put into use. The test insects are: Tineola bisselliella larvae (21-28 d old) raised at 24 oC + 1 oC and 65 + 5 % r.h., and fed on loose wool washed with soap and soda; Attagenus piceus larvae (6-8 week old) and Anthrenus vorax larvae (5-6 week old) raised at 28 oC + 1 oC and 65 + 5 % r.h., with the former fed on dog biscuits, and the latter on loosely woven cloth.

302

For each test, where fabric, knitted materials or felt are to be tested, 6 circles with a diameter of 3-4 cm are used. 4 of the these circles are used as samples for larval attack, and the other 2 circles act as ‘moisture controls’ i.e. blank tests in which untreated, unexposed (to the test species) samples are kept throughout the duration of the test to take into account the weight changes associated with fluctuating moisture levels. In addition, 1 circle of pure wool of diameter 10 cm x 10 cm is used as a negative control i.e. an untreated circle exposed to the test species. Where yarn is to be tested, the yarn is tightly wound in a layer around cardboard strips and glued with gum arabic or some other suitable glue so that the yarn sticks to the edge. For each test, approximately 30-40 g is needed for 18 wrapped pieces of cardboard. 4 pieces are used to test for larval attack, 2 for moisture controls, and 12 for the negative controls (4 for each of the 3 test species). In addition, 4 samples of ‘loose material’ are tested for comparison purposes i.e. as positive controls. The test and moisture control samples are placed in labelled tin cans containing 0.7 mm holes to ensure good ventilation. A felt stamp is used to apply a spot of bait in the middle of each sample. The bait used is an aqueous extract of dried yeast and water. The samples are weighed to + 0.2 g (initial weight). The moisture controls are weighed immediately before and after the associated test samples. A total of 15 test larvae are placed on each test and control sample. The tin cans are closed, including those for the moisture control, and the cans stored for 14 d in the dark at 24 oC + 1 oC and 65 + 5 % r.h. (clothes moth caterpillars) or 28 oC + 1 oC and 65 + 5 % r.h. (fur and carpet beetle larvae). At the end of the test the number of living and dead larvae is counted and the larval skins removed using forceps. In the clothes moth tests the cocoon is also removed. Any anomalous characteristics and visible damage to the samples is noted. With the beetle tests, a crucible or fine sieve is used to separate the excrement from the samples and from loose, bitten off fibres. Any loose, bitten off fibres remaining in the crucible or sieve are added back into the sample, and the sample weighed. The test and control samples are kept for 24 h and are then weighed again. The mean weight loss for each sample is calculated as follows: G = A x C - E B Where, G is the mean weight loss of the samples due to consumption by the test species (in mg); A is the mean weight of the test samples before the test (in mg) E is the mean weight of the test samples after the test (in mg) B is the mean weight of the moisture controls before the test (in mg) C is the mean weight of the moisture controls after the test (in mg) The mean weight loss of the control material should be at least 80 mg/15 larvae for T. bisselliella, A. vorax and A. piceus. If the quantity consumed is less, then the test must be repeated.

303

The determining factor is the ratio of the mean weight loss of the sample material to that of the control material. The sample material is considered to be adequately protected if the relative loss by consumption, expressed in percentage terms, related to the loss by consumption of the control material tested under the same conditions, does not exceed 12 % (T. bisselliella larvae), 15 % (A. vorax larvae) and 18 % (A. piceus larvae). Sample material protected against attack by clothes moths is described as moth resistant, material protected against fur and carpet beetle attack is described as beetle resistant, and material protected against attack by moths and fur and carpet beetles is described as moth and beetle resistant. Borderline cases should be indicated as such in the test report. The test with the sample material without protective treatment should show whether the protective effect established, where applicable, is solely attributable to the protective material or the protective treatment, or whether the sample material is more resistant to attack by moths, carpet beetles or fur beetles than the control material. A5.1.3 Test method for determining the efficacy of dichlorvos vapour against Diptera, Dictyoptera, Coleoptera and Hymenoptera A laboratory study was conducted which examined the concentration/exposure time products of dichlorvos required to kill target insects. The study was carried out using a brick built room with a capacity of approximately 48 m3. Three dichlorvos strips were placed in the room, and the concentration of the active ingredient was determined on a daily basis throughout the study period. Test organisms were exposed on the top of a packing case, which was approximately 600 mm high, situated towards the centre of the room. The test organisms were held in a variety of ways. Two types of test were conducted. In both tests, the mean dichlorvos concentration was 0.03 µg l-1. The conditions were 15 - 18 oC, and 50 - 80 % r.h. No control data were reported. The tests were as follows: To determine LCT values In this test, batches of insects were exposed for known periods, and then held for a 24 h mortality count. This provided data from which LCT50 and LCT95 values could be calculated. The LCT value is defined as “the exposure time (at a given concentration) needed to produce a kill 24 h later”. As an example of the type of result to be expected from this test, the author has stated that, for a test concentration of 0.03 µg l-1, an LCT50 value of 0.09 h means that 50 % of the test insects would be killed 24 h after an exposure time of: 0.09 = 3 h 0.03

304

To determine (LT)C values In this test, batches of insects were held continuously in the test room, and knock down and/or mortality counts were taken at intervals until all individuals were dead. From these data the time for 50 % and 95 % mortality could be calculated and related to concentration to give (LT50)C and (LT95)C values. The (LT)C value is defined as ‘the time to give a certain kill during continuous exposure to a concentration’. As an example of the type of result to be expected from this test, the author has stated that a (LT50)C value of 0.09 h means that 50 % of the test insects would be killed after 0.09/0.03 = 3 h exposure to a test concentration of 0.03 µg l-1 (Unpublished, 1964). A5.1.4 Instructions for determining the susceptibility or resistance of mosquito larvae to insecticides (world health organisation (WHO/VBC/81.807) The purpose of the susceptibility test is to detect the presence of resistant individuals in a mosquito larval population as soon as possible. Establishment of a baseline susceptibility test should include comparison with a ‘normal’ population, one that has never been subject to insecticidal pressure and in which resistant individuals are rare. For a complete test with one insecticide, e.g. dichlorvos, sufficient larvae should be collected from the field in order that about 300 individuals of the same species may be selected; they should be the third or fourth instar and should be retained in the water they were collected in until selected for testing. Lots of 20-25 larvae are distributed in each of 12 small beakers, each containing 25 ml of water. Larvae reared in the laboratory, without pressure to insecticide, should also be separated into lots of 20-25. Into each of 12 glass vessels (75-100 mm diameter) place 225 ml of water. Tap water, rain water or distilled water may be used, providing it is as free as possible of chlorine or organic contaminants. The vessels should be such that the depth of water is between 25 and 75 mm. The average temperature of the water should be recorded and should be approximately 25 °C and it must not be below 20°C or above 30 °C. Prepare the test concentrations by pipetting 1 ml of appropriate standard insecticide solution just above the surface of the water in each glass vessel and stir vigorously for 30 s with a glass rod. In preparing a series of concentrations, the most diluted should be prepared first. In terms of standard solutions fenitrothion is supplied at 31.25, 6.25, 1.25 and 0.25 mg l-1. The standard insecticidal solutions and controls are supplied in 50 ml bottles containing alcohol. The control bottles contain alcohol only. Both the treated solutions and in the controls the alcohol has been denatured by the addition of 2 % butane. The controls are treated by the addition of 1 ml of the alcohol to the water in each container. There should be 2 replicates at each concentration and 2 control replicates. To obtain intermediate concentrations pipette 0.5 ml of the standard solutions instead of 1 ml. Ten to 15 min after the insecticide is placed in to the vessel, add the mosquito larvae by tipping the contents of the small beakers into the vessel. After a period of 24 h exposure mortality counts can be made. In recording the percentage mortality, moribund and dead larvae should be combined. Larvae that cannot be induced to move when they are probed with a needle in the siphon of cervical region are classified as dead. Moribund larvae are

305

those incapable of rising to the surface or of showing the characteristic diving reaction when water is disturbed. Discard larvae that have pupated during the test. If more than 10% of the control larvae pupate in the course of the experiment, then the test should be discarded. Tests with a control mortality of 20 % should be repeated. When 4 replicates have been performed with the same population of mosquito larvae, adequate data should be available for constructing baseline susceptibility. The results obtained should be used to create a dosage-mortality regression line, which will enable LC50 and LC95 values to be obtained. If control mortality is between 5 and 20 %, the percentage mortalities in the test organisms should be corrected using Abbott’s formula. Abbott’s formula is: Corrected % test mortality = ( % test mortality - % control mortality) x 100 100 - % control mortality A5.1.5 Test method BS 4172: hand-held aerosol dispensers against houseflies-method for determination of insecticidal efficacy This describes a method for determining the insecticidal efficacy of hand-held pressurised aerosol dispensers for indoor use as space sprays against houseflies. The method does not apply to metered-valve or total-release dispensers. A dispenser is discharged into a test chamber under controlled conditions, with an internationally recognised susceptible strain of M. domestica (L). Knockdown is assessed at up to 15 min and at 24 h after discharge of the spray. The test chamber should be between 25 and 60 m3 volume, should be maintained at 26 + 2 oC and 45-75 % r.h., and should be ventilated after each test to ensure a minimum of 4 complete air changes before the start of a subsequent test. The test insects are 3-6 d old mixed sex adults. The test insecticide is dispensed into the chamber and at the end of the exposure period, which should be at least 10 min, the M. domestica (both knocked down and flying) are collected into containers by gentle suction or other means. After collection, the M. domestica are supplied with approximately 10 ml of 5 % sugar solution and held for 24 + 6 h at 26 + 2 oC and 45-75 % r.h. The knockdown rate is then assessed and recorded as a percentage of the total population. An untreated control test is conducted. Prior to the test, at least 100 M. domestica are left in the chamber for a minimum of 15 min with the ventilation turned off and the extract vent covered. If there is > 5 % knockdown within 15 min, or more than 10 % knockdown after 18 + 6 h, the chamber should be cleaned, the procedure should be repeated, and all the results from the previous 24 h period should be invalidated. A 'reference' dispenser is also tested. The reference dispenser should be conditioned at 26 ± 2 oC for at least 30 min before use, and then tested using the same procedure as that used for the test dispenser.