Embed Size (px)
Citation preview
ARTICLE IN PRESS
Available at www.sciencedirect.com
journal homepage: http://intl.elsevierhealth.com/journals/tube
Tuberculosis (2008) 88, 358–365
1472-9792/$ - see frdoi:10.1016/j.tube.
�Corresponding au
E-mail address: i
Human Mycobacterium bovis infection in ten LatinAmerican countries
Isabel N. de Kantora,�, Marta Ambroggib, Susana Poggib, Nora Morcilloc,Maria A. Da Silva Tellesd, Marta Osorio Ribeiroe, Marıa C. Garzon Torresf,Claudia LLerena Polof, Wellman Ribonf, Vicente Garcıag, Dolores Kuffoh,Luis Asenciosi, Lucy M. Vasquez Camposi, Carlos Rivasj, Jacobus H. de Waardk
aTuberculosis Consultants Group, World Health Organization, Av. Libertador 7504, 16 A, 1429 Buenos Aires, ArgentinabHospital FJ Muniz, Instituto R. Vacarezza, Buenos Aires, ArgentinacHospital Dr. A. Cetrangolo, V. Lopez, Buenos Aires, ArgentinadInstituto Adolfo Lutz, Sao Paulo, BrazileInstituto de Pesquisas Biologicas, Laboratorio Central do Estado RGS, Porto Alegre, Rio Grande do Sul, BrazilfInstituto Nacional de Salud, Grupo de Micobacterias, Bogota, ColombiagLaboratorio Veterinario Central, Santo Domingo, R. DominicanahInstituto Nacional de Higiene L.I.Perez, Guayaquil, EcuadoriInstituto Nacional de Salud, Lima, PerujComision Honoraria Lucha Antituberculosa, Departamento de Laboratorio, Montevideo, UruguaykLaboratorio de Tuberculosis, Instituto de Biomedicina, Caracas, Venezuela
Received 31 July 2007; received in revised form 26 October 2007; accepted 29 November 2007
KEYWORDSMycobacterium bovis;Bovine tuberculosis inhumans;Latin America
ont matter & 20072007.11.007
thor. Tel.: +54 11
SummaryThe aim of this work was to obtain the best possible estimate of the relevance of bovinetuberculosis (BTB) in humans in Argentina, Brazil, Chile, Colombia, Costa Rica, DominicanRepublic, Ecuador, Peru, Uruguay and Venezuela. Sources of information were aquestionnaire filled by the participant laboratories, and a search of published literature(1970–2007). Only four of these countries reported bacteriologically confirmed cases ofBTB in humans. Most of these were diagnosed in Argentina, where the mean percentage ofMycobacterium bovis cases in relation to those due to Mycobacterium tuberculosis(2000–2006) ranged from 0.34% to 1.0%, according to the region. A slowly decreasing trendwas observed in non HIV as well as in HIV/AIDS patients in Buenos Aires. In most of thesecountries, the low coverage of culture methods, especially of those including pyruvate-containing media, appropriate to isolate M. bovis, contributes to an underestimate of theproblem. It was confirmed that BTB in humans exists, even though its relevance seems tobe low. Milk pasteurization, sanitary controls to dairy products, and meat inspection atslaughterhouses contribute to the protection of human health. However, occupational
Elsevier Ltd. All rights reserved.
4701 2019; fax: +54 11 4701 7731.
om.ar (I.N. de Kantor).
ARTICLE IN PRESS
Table 1 Bovine population, tubLatin American countries.1,2,8
Country Bovines
Bovine pop
Argentina 51.0Brazil 189.0Chile 2.8Colombia 25.0Costa Rica 0.8Dominican R 1.9Ecuador 3.6Peru 7.5Uruguay 11.7Venezuela 13.5
Human Mycobacterium bovis infection in ten Latin American countries 359
aerogenous exposure to TB cattle and their carcasses remains a source of infection in theregion.& 2007 Elsevier Ltd. All rights reserved.
Introduction
The prevalence of tuberculosis (BTB) infection in cattle, inmost Latin American countries, can be roughly estimated onthe bases of tuberculin surveys results, usually restricted toareas with a history of proved BTB disease, and from officialreports to international animal health agencies. From thereit can be concluded that most of cattle bred in this regionare in areas where BTB infection is still prevalent, eventhough it achieves different levels of prevalence. Inconsequence, human populations living in these areas,especially where livestock and dairy industries do not haveeffective animal health and sanitary controls, are at risk ofinfection.1–5
In public health, smear examination continues to be themain resource available to diagnose TB patients. The culturemethod is indicated for children and extra-pulmonary (EP)TB suspects, as well as for re-treated and HIV(+) patients,either for diagnosing mycobacterial disease, or for investi-gating drug susceptibility of the isolates.6 For that purpose,solid, egg-based Lowenstein Jensen (LJ) or Ogawa media areusually employed. Both contain glycerol, which inhibitsgrowth of most strains of Mycobacterium bovis.7 This is animportant qualification which limits the evaluation of thetrue prevalence. Reference laboratories perform culture,identification and susceptibility tests. In some of these, theStonebrink pyruvate-containing egg medium, or differentsemi-synthetic media without glycerol have been intro-duced. In other cases, even though media with pyruvate arenot included in routine-culture procedures, they are usedfor specifically oriented studies and surveys.
Taking into account these conditions and possible limitations,the aim of this work was to obtain the best possible estimate ofthe relative importance of BTB in humans in 10 countries of thisregion, with emphasis in the period 2000–2006.
erculosis incidence rates in hum
Human
ulation (million) TB inc
416015451491
1311722842
Materials and methods
Reports on bovine stock, human population, total TB casesand related data were obtained from WHO, PAHO, FAO, FAO/GLIPHA, IUATLD, UNFPA and OIE,1,2,5,8. A systematic searchand review (1970–2006) was made in Scielo and Medlinedata bases for articles on BTB in animals and man, in theLatin American countries.
Other publications, in Spanish or Portuguese, which werenot indexed, were also consulted. The search was extended tocongress and meetings proceedings, to official reports fromanimal health, and public health national departments, aswell as from university and hospital reference laboratories(NRL). In addition, responsible officials from the participant-ing laboratories in Argentina, Brazil, Colombia, Costa Rica,Dominican Republic, Ecuador, Peru, Uruguay and Venezuelafilled in a questionnaire that included the following points:
1.
ans
s
ide
Culture media currently used in the laboratory: eggcontaining and/or semi-synthetic liquid media, with andwithout pyruvate. Description of those media included inroutine work, and of those reserved for special surveys,or for culturing certain type of specimens.
2.
Number of cultures for mycobacteria annually performed(average) from human samples.3.
M. bovis isolations. Number, origin, year of isolation.Percentage of M. bovis isolates in relation to all othermycobacteria.4.
Phenotypic and/or genotypic tests performed to differ-entiate M. bovis from other members of the Mycobacter-ium tuberculosis Complex.Data provided by these laboratories covered periods thatranged from 3 to more than 20 years, and achieved different
, and reported number of cases for 2005, by country, in 10
nce rate per 100,000 No. of cases notified, 2005
15,869111,050
237720,496
6227946
17,33147,976
95711,126
ARTICLE IN PRESS
I.N. de Kantor et al.360
degrees of completion. Information on M. bovis isolationsfrom humans was mainly based on these reports, and wascomplemented with other sources, as indicated.
Results
Global data of bovine population, human TB incidence rateand TB cases notified, for each of these countries, arepresented in Table 1.1,2,5,8 Information on the status of BTBinfection in cattle and in humans for each country ispresented below.
Argentina
Between 1969 and 2005, an average of 10 million bovinecarcasses were annually submitted to the official veterinaryinspection. During that period, the percentage of carcassescondemned for BTB decreased from 6.7% to 1.1%.4 Thenumber of dairy herds officially free from BTB, on the basesof a voluntary program, increased sharply from 44 in 1995 to5663, approximately equivalent to 2 million cattle, in 2005.In this year, dairy herds registered in the three main milkproducer provinces (Buenos Aires, Santa Fe and Entre Rios),which hold some 5 million cattle, were subject to at leastone tuberculin test (P. Torres, SENASA, personal commu-nication). TB in swine underwent a decline parallel to thatobserved in cattle, most probably related to farmingimprovements which include earlier slaughter and discon-
Table 2 Results obtained in studies made in Argentina, at difcases due to M. bovis, in relation to the total number of M. tub
Place Years No. of TBcases
Pac
Muniz Hospital, Buenos Aires11 1971 2157 A
Muniz Hospital, Buenos Aires12 1977–1982 6361 A
Muniz Hospital, Buenos Aires21 1981–1991 10 000 A1981–1991 240 A
Muniz Hospital, Buenos Aires 2000–2006 5551 A2000–2006 1391 A
Posadas Hospital, BuenosAires13
1982 45 E
Nationwide14 1982–1984 7700 P
Santa Fe Province4,15,39 1977–2000 3946 A2001–2006 590 A
Cordoba Province16 2003 723 P48 E
Cetrangolo Hospital, BuenosAires
2001–2005 1470 P
P, pulmonary tuberculosis (TB); EP, extra-pulmonary TB; NR, not reg�All PTB cases.yThree PTB, three EP (lymph-nodular) and two P/EP cases.
tinuance of the practice of feeding hogs with non-pasteurized dairy products.9,10
The Stonebrink medium came into use in public healthlaboratories in the early 70s. This provided some reliabledata on the prevalence of BTB in humans, mainly limited tothe pulmonary cases.11,12 Reports on bacteriologicallyconfirmed childhood and EP TB cases were too exiguous todraw any general conclusion. Nevertheless, the percentageof M. bovis cases diagnosed among EP patients reached 8% inthe 80s.13 A nationwide survey was carried out in the period1982–1984,14 when the highest percentages of cases due toM. bovis were observed in provinces with extensive farmingand cattle breeding activities, like Santa Fe and Cordoba.More specific information came from the National Institutefor Respiratory Disease (INER E. Coni, Santa Fe), and fromCordoba reference laboratories (Table 2).15,16
At the Muniz Hospital (Buenos Aires), in the period1970–1990, the susceptibility tests to isoniazid, streptomy-cin, rifampicin and ethambutol were performed on M. bovisisolates, using the simplified variant of the proportionmethod on a modified LJ.17 Pyrazinamide (PZA) resistancewas determined by the pyrazinamidase test, to confirm thespecies identification. After 1990, the susceptibility of M.bovis strains to antituberculous drugs was tested in theBACTEC 460TB system (BD Argentina).18
Two TB disseminated cases in humans due to M. boviswere reported; one of them, in an HIV/AIDS adult patient,presented drug resistance to rifampicin, in addition toPZA.19,20
ferent times, in terms of number and percentage of humanerculosis Complex isolates obtained and analyzed.
atients characteristicsnd diseaselassification
HIV No. of M. bovis (%)
dults, P NR 38 (1.75)
dults, P and EP NR 57 (0.90)
dults, P and EP Neg. or NI 95� (0.95)dults, P and EP Positive 2� (0.83)
dults, P and EP Neg. or NI 12� (0.22)dults, P and EP Positive 8y (0.58)
P, adults and children NR 4 (8.8)
, EP NR 36 (0.49) (0.00–2.20)
dults, P Neg. or NI 97 (2.46) (0.00–6.60)dults, P Neg. or NI 6 (1.02) (0.00–3.00)
Neg. or NI 9 (1.24)P Neg. or NI 0 (0.00)
Neg. or NI 5 (0.34)
istered; NI, not investigated; Neg., negative.
ARTICLE IN PRESS
Figure 1 DNA spolygotyping pattern of two M. bovis isolates, Cetrangolo Hospital, Vicente Lopez, Buenos Aires, Argentina (2001,2005). Father and daughter, Cases A and B. M: male, F: female.
Human Mycobacterium bovis infection in ten Latin American countries 361
In the 11 years period, 1981–1991, the total numberof bacteriologically confirmed TB cases in the MunizHospital was 10,240. Among these, 0.95% of patients whowere not HIV infected, and 0.83% of the HIV/AIDS cases,were due to M. bovis (Table 2).21 In the same hospital,during a later 7-year period (2000–2006), a total of 6942 TBcases were confirmed by culture. The percentage ofM. bovis cases not associated with HIV infection was0.22%, showing a significant decrease in relation to theperiod 1981–1991 (w2 test, p ¼ 0.00001). The decrease wasless pronounced in the HIV/AIDS group (from 0.83% to 0.58%)(Table 2).
At the Cetrangolo Hospital (Vicente Lopez, Buenos AiresProvince), in the period 2001–2005, PRA (PCR-RestrictionEnzyme Analysis) was systematically applied to everyMycobacterium strain cultured in MGIT960 (BD Argentina)but not on LJ medium, followed by spoligotyping.22 Isolatesfrom five patients were identified as M. bovis (Table 2).These patients presented with pulmonary localization, andwere of rural origin. The strains had an identical geneticpattern.23 Interestingly, two of these cases, were father anddaughter, both from Bahıa Blanca (Buenos Aires Province).The father was a slaughterhouse worker. He had followedirregular therapy, and finally died. His daughter wasdiagnosed 5 years later as having PTB, in the same hospital,and an identical M. bovis isolate was obtained from hersputum specimens (Table 2, Figure 1).
Brazil
A tuberculin survey (2004), was performed on 454,108 cattleselected from a total population of nearly 108 million, in fiveregions of the country (North, North-east, South-east, Southand Central-west). Prevalence of tuberculin reactors rangedbetween 0.37% in the Southern-Eastern, and 3.62% in theNorthern States. Regions where the most important cattlebreeding areas were located, presented the lowest pre-valence (0.37–0.92%).24 A nationwide program of BTBcontrol is operating from 2001.3,4,25
Strains apparently different from M. tuberculosis, iso-lated from patients and examined at the different centralState laboratories, are submitted to the Helio FragaNational Reference Laboratory (NRL) for identification. Inonly one occasion in 20 years (1987–2006) M. bovis wasconfirmed as the etiological agent. The strain was isolatedfrom a blood sample, in an HIV/AIDS female patient. Also atthe NRL, pyruvate egg medium (without glycerol) wasincluded for primary isolation of mycobacteria during 2years (2002–2004), during which no M. bovis was isolated (A.Werneck Barreto, personal communication, 2006, Table 3).
At the Federal University Hospital, Rio de Janeiro, nearly5000 samples are cultured each year for mycobacteria. NoM. bovis isolate was obtained in an 11 months’ especial
survey, using LJ and Stonebrink media. (Leila Fonseca,personal communication, 2006).
In the Sao Paulo laboratory network (RL, Adolfo LutzInstitute), a total of 355,383 cultures were performedin the period 2001–2005. LJ and Ogawa, both glycerolcontaining media, were employed for culturing mycobacter-ia, except in four laboratories, where MB/BacT (BioMerieux) culture vials were used from 2001 onwards, andin another six regional laboratories, which used MGIT960as well. Two M. bovis strains were recovered, in twopatients, one in 2001 from a lymph node biopsy, and otherone in 2002 from a cerebrospinal fluid sample (Table 3).M. bovis was differentiated from BCG on the basisof the susceptibility to TCH (thiophen-2-carboxylic acidhydrazide) and to cycloserine, resistance to PZA and anurease negative test result.22 During a 5-year period(2001–2006), Stonebrink and LJ were used for culturingeach sample analyzed at the Adolfo Lutz Institute, when noM. bovis was isolated.
At the State Reference Laboratory in Rio Grande do Sul(1997–2005), of approximately 5000 mycobacterial isolatesphenotyped, no M. bovis was confirmed. Genotypic meth-ods, restriction fragment length polymorphism (RFLP) andspoligotyping were used to differentiate 370 M. tuberculosisComplex strains; none of these harbored only one or twoIS6110 copies, or showed the characteristic spacer hybridi-zation of M. bovis (Table 3).
Chile
No case due to M. bovis has been reported by the NRL(National Institute of Health, INS, Santiago) in the last 10years. Meanwhile, BTB infection in cattle affects 56% ofherds in Regions V and the Metropolitan (Santiago) and, inRegions VIII to X, to 5.1% of herds.26 A genotyping assayperformed on nine M. bovis strains isolated from bovinemediastinal lymph nodes taken at the slaughterhouse at postmortem examination, revealed the presence of the se-quence insertion IS1081, specific of the M. tuberculosisComplex, in all the isolates, and a single band pattern withthe IS6110 probe for seven of nine strains analyzed, whatcan indicate M. bovis (M. tuberculosis usually have moreIS6110 copies than M. bovis).27
Colombia
Tuberculin surveys in cattle showed that 2.9% of dairy cattleresponded positively, and that 1.2% of herds had one or morereactors. At slaughterhouses less than 0.01% of the carcassesinspected suffered condemnation for TB in the 90s.1,28 In2005, the finding of 22 infected bovines out of 298 tested infive districts, was reported to OIE. All positive animals wereslaughtered.2
ARTICLE IN PRESS
Table 3 Results obtained in studies made in Brazil, Colombia, Dominican Republic, Ecuador, Peru, Uruguay and Venezuela, atdifferent times, in terms of number and percentage of human cases due to M. bovis, in relation to the total number of M.tuberculosis Complex isolates obtained and analyzed.�
Country Year No. of TB cases orisolates
No. ofM. bovis(%)
Culture media included
BrazilHelio Fraga NRL 1996–2006 2000 isolates 1 (0.05) St. and LJAdolfo Lutz (Sao Paulo) 2001–2002 5617 EP cases 2 (0.04) St. and LJ, MB/BacT,
MGIT960IPB-LACENS/RS (Porto Alegre) 1997–2005 5000 P and EP cases 0 Ogawa and LJ
ColombiaNational Institute Health, (INS,
NRL, Bogota)1989–2005 12,500 isolates 0 St. and LJ
National surveys for drugresistance (INS)
1991–1992;1999–2000;2004–2005
3105 isolates 0 St. and Ogawa
Dominican RepublicLVC and Dermatology Hospital 1980–2005 17,500 isolates 0 St. and LJ
EcuadorIzquieta Perez (NRL) and the
laboratory network1998–2005 75,080 isolates (1044
selected for typing)2 St. and LJ
PeruNational Inst. Health (NRL) and the
laboratory network2004 16,401 isolates 0 LJ and Ogawa, BACTEC 4602005 15,502 isolates 0
UruguayNRL, Montevideo 1979–1983 2000 P and EP cases
1957 selected fortyping
0 St. and LJ
2000–2004 0 St. and LJ, MB-BacT
VenezuelaInstituto Biomedicina, Caracas 1999–2006 1250 cases 1 St. and LJ
St., Stonebrink; LJ, Lowenstein Jensen; NRL, National Reference Laboratory; LVC, Central Veterinary Laboratory; INS, NationalInstitute of Health; IPB-LACENS/RS, Institute for Biological Research, State Reference Laboratory, Rio Grande do Sul.�Details for Argentina are presented in Table 2. No M. bovis isolate was reported in Costa Rica.
I.N. de Kantor et al.362
At the National Institute of Health (INS, NRL, Bogota),since the late 80s every specimen submitted to culture isinoculated on Ogawa and Stonebrink media. In spite of thatno M. bovis strain has been isolated. In the national surveysfor drug resistance, the isolates showing drug resistancewere submitted to phenotypic and genotypic (spoligotypingand RFLP) identification tests. No human M. bovis case wasfound. On the other hand, M. bovis was isolated from bovinespecimens at the INS in 35 occasions between 1971 and 1988(19 from lymphatic node, 8 from lung tissue, and a further 8from milk samples (origin of samples: Instituto ColombianoAgropecuario, ICTA, and the University).
Costa Rica
No BTB infected herd has been detected after the onereported to OIE in 2005, when 33 bovines out of 609
tuberculin tested resulted positive and were, consequently,slaughtered.2 No M. bovis strain has been isolated fromhumans in the last 10 years (1996–2006) at the ReferenceLaboratory for Mycobacteria.
Dominican Republic
Infected cattle herds are seldom identified (0.63% in2004).1,4 The Veterinary Central Laboratory (LVC) is thereference laboratory for both human and animal TBdiagnosis. At this laboratory, as well as at the Hospital ofDermatology, both LJ and Stonebrink media are used toculture specimens samples. A phenotypic differentiation isperformed on each isolate. No case due to M. bovis has beendetected in the period 1980–2005, neither in the nationwidesurvey for drug resistance, during which 420 pulmonarycases were studied (Table 3).29
ARTICLE IN PRESS
Human Mycobacterium bovis infection in ten Latin American countries 363
Ecuador
Results of a tuberculin survey performed on dairy cattle, inthe Mejia Canton, were recently published. A total of 1012cattle randomly selected from 59 farms were tested; 4.24%were positive. M. bovis was isolated from mesenteric lymphnodes and lung samples in one animal. A 16S ribosomal RNAbased PCR confirmed the culture results. This study confirmsthe presence of BTB in dairy cattle in Ecuador, and reaffirmsthe need for implementing a control program.30
At the NRL Leopoldo Izquieta Perez Institute (INHLIP,Guayaquil), 1.4% of the total mycobacterial isolates made inthe laboratory network, in the period 1998–2005, weresubmitted to species identification, based on their growthand/or morphological characteristics. Additionally, a totalof 997 M. tuberculosis Complex isolates at the nationwidedrug resistance survey (2002) were analyzed. Two M. bovisstrains were identified, one from a lymph node (1 year old,female), and the other from a pulmonary specimen (3 yearsold, male) (Table 3).
Peru
The most recent information available on BTB in cattle is asfollows: in the Province of Huaura, two tuberculin surveyswere made in adult dairy cattle, on 3240 and 3230 animals,respectively, in 2001 and 2002. Prevalence of reactors was0.12% and 0.06%, in the same order. In Ayacucho Province, asample of 461 bovines, selected from a total of 25,502, in 385dairy herds, were tested, and all tested negative. Meanwhile,in another survey, in the Canta Province, near Lima, of 503dairy bovines, 11 (2.20%) animals resulted tuberculin posi-tive.31,32 In 2002, out of 220,370 dairy cattle from severalprovinces which were tested, a total of 426 reactors wereidentified (0.19% prevalence rate). The highest number andpercentage of infected animals were located in the Limadepartment.33 In 2005, the country reported to OIE thedetection of BTB in cattle in only five districts, where 208 outof 3896 tuberculin tested animals (5.3%) resulted positive, andwere subsequently slaughtered.2 According to that, theinfection seems to be almost limited to dairy cattle and tocertain specific country regions.
In 2004, a total of 139,709 specimens were cultured formycobacteria in the public health laboratory network, country-wide, and 11.74% of these were positive. In 2005 the number ofcultures was 149,063, and the positivity rate was 10.40%. Ofthe mycobacterial isolates submitted to the NRL (NationalInstitute of Health, INS, Lima) for typing/identification, no M.bovis strains were identified (Table 3). Media used in thelaboratory network are LJ and Ogawa, both without pyruvate.BACTEC460 is reserved for the examination of specimenscollected from patients at risk of HIV or HIV(+), health workers,and multi-drug resistant (MDR) TB in-contacts. While thenumber of cultures performed from human specimens is veryhigh, most isolates are made on glycerol containing media,which significantly limits the sensitivity for M. bovis.
Uruguay
Voluntary schemes for the attestation of dairy herds startedin the 60s. These activities rapidly gained support from
livestock owners and milk processing plants. Positivetuberculin reactors were moved only to slaughter at anapproved abattoir. No animal classified as a positive reactorwas retested. As a result, only 0.4% of dairy herds presentedBTB infection in 1989.28 In 1996 an official campaign for BTBeradication, with a comprehensive legislation and regula-tions, was launched. In 2005, a total of 10 BTB foci werereported to OIE, a total of 18 cattle were tuberculin testedpositive and were slaughtered; no macrosocpic lesions werefound at necropsy. Thus, TB in cattle seems to be neareradicated.2
The earliest record of any survey aimed to detect M. bovisin humans is a study made in 1962, when 810 strains isolatedfrom TB in-patients were phenotypically analyzed, all whichwere identified as M. tuberculosis, of ‘‘human type’’.34
Stonebrink, in addition to the LJ medium, was introduced in1979 at the NRL (Montevideo), where all the bacteriologicalinvestigations for mycobacteria in the country were carriedout. This procedure was discontinued after 4 years becauseno M. bovis strain had been isolated during that period(Table 3). Since 2000, samples considered to be of specialinterest (EP specimens, HIV positive patients, contacts ofMDR TB cases, among others) are inoculated both in MBBacTand on the egg-containing medium. No M. bovis has beenisolated since then. On the other hand, between 2000 and2003, 3.6 (1.7–4.7%) of the mycobacteria isolates have beenidentified as Mycobacterium avium. These isolates wereobtained from 70 patients (2.2 isolates per case, inaverage). Eleven of these (HIV/AIDS patients) presenteddisseminated disease. Nevertheless, M. avium infection isconfirmed in less than 1% of pulmonary disease cases, andtherefore it is not considered a relevant public healthproblem in Uruguay. On the other hand, the frequency of M.avium disseminated infection associated to HIV/AIDS hasdecreased sharply after the introduction of the highly activeantiretroviral therapy (HAART).35
Venezuela
In 2003, in a countrywide survey, 0.04% of 707,693 bovinesresulted positive to the tuberculin test. PPD positive reactorcattle were detected on 5.2% of the farms investigated.1,2
In the period 1999–2006, a total of 14,620 respiratoryspecimens collected from indigenous rural and urbanpopulations from Caracas and Delta Amacuro, and fromother 717 respiratory symptomatic patients in the CaraboboState, were cultured on LJ and Stonebrink media at theInstituto de Biomedicina, in Caracas. Positivity for M.tuberculosis Complex was 8.55%. All isolates were typedusing spoligotyping and, based on its pattern, one M. boviswas detected, isolated from a pulmonary TB adult patient(Table 3). There were no previous reports of human BTB inthe country.
Discussion
Information presented here comes from different sourcesthat are heterogeneous in the methodology used, as in theirdegree of completion. Concerning the prevalence of BTBinfection in cattle, the data presented come either from
ARTICLE IN PRESS
I.N. de Kantor et al.364
tuberculin surveys, of variable coverage, or from systematicnational records of condemnations at the slaughterhouse.
It must be emphasized that most of these countries do nothave a consistent program for tuberculin testing and for BTBeradication, in operation countrywide.
Laboratory information on human cases due to M. bovispresents similar variations. Nevertheless, the approachadopted was to include all available reports that could helpto ascertain at national level the presence or absence of BTBin cattle, and the extent of disease due to M. bovis inhumans in these countries, despite the evident limitations ofthe data.
The percentage of reported BTB human cases in relationto those caused by M. tuberculosis ranged between 0% andapproximately 2.5% in these 10 Latin American countries.With the exception of two cases in Ecuador, three in Brazil,and one in Venezuela, all the M. bovis cases reported herewere diagnosed in Argentina. Apparently in Chile, Colombia,Costa Rica, Dominican Republic, Peru and Uruguay, M. bovishas never been isolated from humans. However, thesenegative data most probably underestimate the infection,taking into account the limited use of the egg-mediumcontaining pyruvate, and of semi-synthetic media in whichM. bovis would preferably grow. In effect, a mutation withinthe glycerol kinase gene has been recently identified in theM. bovis genoma, that underlies the requirement forpyruvate when glycerol is the sole carbon source.36,37 It isclear then that the presence of glycerol in culture media, islikely to have been an impediment for the isolation of M.bovis. It would be advisable, especially in the case of Peru, acountry with a relatively high human TB incidence rate(Table 1), and with a well organized TB laboratory network,to introduce pyruvate-containing media, at least in anespecial trial, to investigate the current contribution of M.bovis to human TB.
This failure to culture, due to the use of unsuitableselective media, and to type the organism, and/or limitedavailability of differential tests, have also been emphasizedin the European Region,38 even though laboratory facilitiesand technical standards there are significantly moreadvanced than those available in the Latin American Region.
In the present study, only in Argentina, where tradition-ally the Stonebrink medium has been used, along with LJ, arelatively high yield of M. bovis isolates was obtained. Itcould be argued however, that the prevalence of BTB incattle is very low in most of other countries included here,and that, especially in Uruguay and Costa Rica, a constantpolicy of food protection and safety, have contributed toprotect human health from this zoonotic infection, and sothe risk of M. bovis in humans is very low.
Consequently, low coverage of tuberculin surveys incattle, and technical limitations to isolate M. bovis preventus from attempting to correlate cattle infection rates withfrequency of disease due to M. bovis in man, except inArgentina, where this relation seems quite evident.
Our data suggest that the global relevance of M. bovis as acause of human TB remains low in most of these countries,compared to that of M. tuberculosis, and that it has notshown any increase in recent years.39 Also in Argentina, thetrends seem to be slowly decreasing. Possible explanationsfor this are related to the strengthening of the BTB controland eradication activities, and to the improvement of cattle
sanitary conditions. However, slaughterhouse and dairyindustry workers continue to be at risk, particularly in areasand countries with an extensive livestock industry but inwhich the cattle population is still infected. Currently, OIE,FAO, PAHO and other international regulatory agencies areplacing increased emphasis on the food safety aspects of thedisease caused by M. bovis in humans, and are givingcomprehensive recommendations for the control of meatand milk products, to protect the human health, even insituations of low BTB prevalence in cattle.40 In effect, foodhygiene, along with the pasteurization of milk and othersanitary controls, the epidemiological surveillance and theproper treatment of TB patients continue to be the maintools to control, eliminate, and eventually eradicate humanM. bovis disease.41
Funding: None declared
Competing interests: None declared
Ethical approval: Not required
Acknowledgments
We are especially grateful to: Dr. M. D. Sequeira-Latini, INERE. Coni, Ministry of Public Health, Santa Fe; Dr. AngelaWerner Barreto, National Reference Laboratory Helio Fraga,RJ, Brazil; Prof. Leila S. Fonseca, University Hospital, UFRJ,Brazil; and Dr. M.C. Matamoros, National Reference Labora-tory for Mycobacteria, Costa Rica, who kindly oriented thesearch of data on bovine tuberculosis in humans, and/orparticipated in collecting and reporting information in-cluded in this article. We also wish to thank Dr. Pedro Torres,Tuberculosis Eradication Program, National Service ofAnimal Health, SENASA, Argentina, who provided updatedepidemiological information on the control of tuberculosis incattle and swine in this country, and to Dr. Maria IsabelRomano, CONICET-INTA, Castelar, Argentina, who madeavailable for publication graphic results of spoligotypinganalysis of M. bovis strains.
References
1. FAO. Global livestock production and health atlas. AnimalProduction and Health Division, GLIPHA, /http://www.fao.org/ag/aga/glipha/index.jspS; 2006 [accessed January–April2007].
2. OIE. Word Animal Health Information Division, WAHID. DiseaseInformation. Bovine Tuberc 2005 /http://www.oie.int/wahid--prod/public.phpS [accessed October 2007].
3. Ritacco V, Torres P, Sequeira MD, Reniero A, de Kantor IN. Bovinetuberculosis in Latin America and the Caribbean. In: Thoen Ch,Steele JH, Gilsdorf MJ, editors. Mycobacterium bovis infectionin animals and humans. Ames, Iowa: Blackwell Publishing; 2006.p. 149–60 [chapter 16].
4. de Kantor IN, Ritacco V. An update on bovine tuberculosisprogrammes in Latin American and Caribbean countries. VetMicrobiol 2006;112:111–8.
5. Pan American Health Organization (PAHO/WHO). Action plan forthe eradication of bovine tuberculosis in the Americas, Phase 1.HPV/TUB/113/92. Washington DC: PAHO/WHO; 1992.
ARTICLE IN PRESS
Human Mycobacterium bovis infection in ten Latin American countries 365
6. Tuberculosis guide for low income countries. 4th ed. Paris:International Union Against Tuberculosis & Lung Disease; 1996.
7. Grange JM, Yates MD, de Kantor IN. In: Guidelines for speciationwithin the Mycobacterium tuberculosis complex. 2nd ed. WHO/EMC/ZOO/96.4, Geneva: WHO; 1996.
8. World Health Organization. WHO report. /http://www.who.int/tb/publications/global_report/2007/xls/amr.xlsS; 2007[accessed April 2007].
9. Torres P. Situacion de la tuberculosis bovina en la RepublicaArgentina. In: de Zoonosis II T, Cacchione R, Durlach R, Larghi O,editors. Buenos Aires: Asociacion Argentina de Zoonosis; 2004.p. 235–42, chapter 32.
10. Perez A, Debenedetti R, Martınez Vivot M, Bernardelli A, TorresP, Ritacco V. Tendencia de la tuberculosis porcina y validez de lainspeccion bromatologica para su deteccion en areas deproduccion intensiva en la Argentina. Rev Med Vet 2002;83:14–8.
11. Cetrangolo A, Marchesini L, Isola N, Kantor IN, Di Lonardo M.Utilidad del medio de Stonebrink en el diagnostico de latuberculosis. Rev Arg Tub Enf Resp 1971;32:142–4.
12. Di Lonardo M, Fulladosa G, Ambroggi M, Isola N, Poggi S, RybkoA. Tuberculosis humana por Mycobacterium bovis. In: CongresoArgentino de Tisiologıa y Neumonologıa, Buenos Aires. Actas(Proceedings), 1983. p. 11.
13. Peluffo G, de Kantor IN. Bacteriologic diagnosis of extrapul-monary tuberculosis in a general hospital. Rev Argent Microbiol1982;14:91–6 [in Spanish, summary in English].
14. Barrera L, de Kantor IN. Nontuberculous mycobacteria andMycobacterium bovis as a cause of human disease in Argentina.Trop Geogr Med 1987;39:222–7.
15. Ritacco V, Sequeira MD, de Kantor IN. Human tuberculosiscaused by M. bovis in Latin America and the Caribbean. In:Thoen Ch, Steele JH, Gilsdorf MJ, editors. Mycobacterium bovisinfection in animals and humans. Ames, Iowa: BlackwellPublishing; 2006. p. 13–7 [chapter 3].
16. Solda PA, Rojo SC, Cosiansi MC, Barnes AI. Frequency ofpulmonary and extrapulmonary tuberculosis in a referencehospital in Cordoba province, 1991–2003. Rev Argent Microbiol2005;37:89–91 [in Spanish, summary in English].
17. de Kantor IN, Lesslie IW. Isolation of mutants resistant to anti-tuberculosis drugs in wild strains of Mycobacterium bovis.Medicina (Buenos Aires) 1974;34:244–8 [in Spanish, summary inEnglish].
18. Cavirani S, Fanti F, Benecchi M, Calderaro A, Taddei S,Arcangeletti C, et al. Evaluation of susceptibility of Mycobac-terium bovis to antituberculous drugs by radiometric BACTEC460TB system. News Microbiol 2003;26:181–6.
19. Castagnino JP, Musella RM, Rybko AA, Abbate EH. Tuberculosisgeneralizada por M. bovis en un paciente inmunocompetente.Rev AMA 2003;116:16.
20. Valerga M, Viola C, Thwaites A, Bases O, Ambroggi M, Poggi S, etal. Mycobacterium bovis tuberculosis in a female patient withAIDS. Rev Argent Microbiol 2005;37:96–8 [in Spanish, summaryin English].
21. Di Lonardo M, Isola NC, Ambroggi M, Rybko A, Poggi S.Mycobacteria in HIV-infected patients in Buenos Aires. TubercLung Dis 1995;76:185–9.
22. Leao SC, Martin A, Mejia GI, Palomino JC, Robledo J, Silva TellesMA, et al. Practical handbook for the phenotypic and genotypicidentification of mycobacteria. Brussels: INCO-DEV-CA; 2004.p. 71.
23. Brudey K, Driscoll JR, Rigouts L, Prodinger WM, Gori A, Al-HajojSA, et al. Mycobacterium tuberculosis Complex geneticdiversity: mining the fourth international spoligotyping data-base (SpolDB4) for classification, population genetics andepidemiology. BMC Microbiol 2006;6:6–23.
24. Roxo E. Situac-ao atual da tuberculose bovina no Brasil.Programa Nacional de Controle e Erradicac-ao de Brucelose eTuberculose Animal, PNCEBT. Secretaria de Defesa Agropecuar-ia. Docum. PNCEBT DDD2005, Sao Paulo: SDA; 2005. p. 1–5.
25. Ministerio da Agricultura, Pecuaria e Abastecimento. ProgramaNacional de Controle e Erradicac- ao de Brucelose e daTuberculose Animal. PNCEBT, /http://www.agricultura.gov.brS[accessed October 2007].
26. Mansilla M, Martınez AH, Palavecino C, Rehren G, Lucero P, LeonG, et al. Variantes geneticas de Mycobacterium tuberculosisaisladas de pacientes de la X Region de Chile. Rev Chil Infect2006;23:220–5.
27. Retamal P, Martınez TMA, Abalos P. Secuencias de insercionIS6110 e IS1081en cepas de Mycobacterium bovis provenientesde bovinos beneficiados en la Region Metropolitana. Rev ChilInfect 2003;20:166–70.
28. de Kantor IN, Ritacco V. Bovine tuberculosis in Latin Americaand the Caribbean: current status, control and eradicationprograms. Vet Microbiol 1994;40:5–14.
29. Espinal MA, Baez J, Soriano G, Garcia V, Laszlo A, Reingold AL,et al. Drug resistant tuberculosis in the Dominican Republic:results of a nationwide survey. Int J Tuberc Lung Dis 1998;2:490–8.
30. Proano-Perez F, Rigouts L, Brandt J, Dorny P, Ron J, Chavez MA,et al. Preliminary observations on Mycobacterium spp. In dairycattle in Ecuador. Am J Trop Med Hyg 2006;75:318–23.
31. Arcelles PM, Delgado C, Alzamora PA, Manchego AS, GavidiaCCh. Prevalencia de tuberculosis bovina en el distrito deVegueta, Huaura. Rev Investig Vet Peru 2005;16:154–7.
32. Sanchez R, Rosario R. Prevalencia de la tuberculosis bovina en laprovincia de Parinacochas, Ayacucho. Rev Investig Vet Peru2002;13:100–2.
33. Flores CF, Delgado CA, Gonzalez ZA, Rivera HG. Determinacionde la presencia de tuberculosis bovina en la provincia de Canta,Lima. Rev Investig Vet Peru 2005;16:65–70.
34. Osorio HS. Prevalencia de los diversos grupos de micobacteriasen un hospital para tuberculosos de Montevideo. Hoja Tisiolo-gica Inst Tisiologıa Prof. J.B. Morelli y Cat de Clin Neumol1963;23:41–6.
35. Coitinho C, Brandes E, Pardinas M, Rivas C. Disseminatedmycobacterial infections in patients with HIV/AIDS. Evaluationof blood cultures. Rev Argent Microbiol 2005;37:196–8 [inSpanish, summary in English].
36. Garnier T, Eiglmeier K, Camus JC, Medina N, Mansoor H, PryorM, et al. The complete genome sequence of Mycobacteriumbovis. Proc Natl Acad Sci USA 2003;100:7877–82.
37. Keating LA, Wheeler PR, Mansoor H, Inwald JK, Dale J,Hewinson RG, et al. The pyruvate requirement of somemembers of the Mycobacterium tuberculosis complex is due toan inactive pyruvate kinase: implications for in vivo growth. MolMicrobiol 2005;56:163–74.
38. de la Rua-Domenech R. Human Mycobacterium bovis infection inthe United Kingdom: incidence, risks, control measures andreview of the zoonotic aspects of bovine tuberculosis. Tubercu-losis (Elsevier) 2006;86:77–109.
39. de Kantor IN. Mycobacterium bovis infection in humans in Southand Central America. In: Symposium on Mycobacterium bovis:contribution to tuberculosis in humans. Abstract book, 37thworld conference on lung health of the international unionagainst tuberculosis and lung disease. Int J Tuberc Lung Dis2006;10(Suppl. 1):S10.
40. Droppers WFGL. OIE philosophy, policy and procedures for thedevelopment of food safety standards. Rev sci tech Off int Epiz2006;25:805–12.
41. World Health Organization. Global tuberculosis control: surveil-lance, planning, financing. WHO report. WHO/HTM/TB/2004.331, Geneva: WHO; 2004. p. 145–6.
