Laboratorymethods for diagnosing cryptosporidiosis · Modified Ziehl-Neelsen staining was introduced for staining cryptosporidial oocysts byveterinary workers whohadfound that Cryptosporidium

J Clin Pathol 1991;44:445-451

ACP Broadsheet 128

Laboratory methods for diagnosingcryptosporidiosis

D P Casemore

IntroductionThe laboratory diagnosis of cryptosporidiosisin man, for both clinical and epidemiologicalpurposes, requires a knowledge of its naturalhistory and pathogenesis. The biology of theorganism and the clinical and epidemiologicalfeatures of the infection have recently beenreviewed in detail. 1-3

CLINICAL FEATURESCryptosporidium is now widely recognised asa cause of acute gastroenteritis, particularly inchildren. The infection in infants or theelderly is not particularly common, nor moresevere. The infection usually produces a per-sistent, watery, offensive diarrhoea, oftenaccompanied by abdominal pain, nausea,vomiting (especially in children), andanorexia; about a third of cases have othersymptoms including clinically importantweight loss, fever, and cough. Cryptosp-oridium is also a cause of severe and poten-tially life-threatening disease in the immun-ocompromised, especially those with AIDS.In these patients the infection sometimesaffects the biliary tract, pancreas, or the res-piratory tract. Oocyst excretion and symp-toms, including the degree of diarrhoea, canfluctuate during the course of the disease.Many patients stop excreting oocysts withinone to three weeks of the disappearance ofsymptoms, but some excrete for severalweeks, and a few continue to excrete formonths. Asymptomatic infection may be morecommonly found in underdeveloped areaswith poor hygiene, or where there is close andfrequent contact with livestock.2

Public HealthLaboratory andDepartment ofMicrobiology, GlanClwyd Hospital,Bodelwyddan, ClwydLLl8 5UJ

D P CasemoreCorrespondence to:Dr C P CasemoreAccepted for publicationI October 1990

EPIDEMIOLOGYInfection in man is derived, either directly or

indirectly, from animals or from other infec-ted people. Many cases are sporadic, but smallclusters and family and community-wideoutbreaks do occur. Seasonal peaks have alsobeen noted in some areas. At tithes, axid insome localities generally, the prevalence may

be very low and hence the rate of detection in

June 1991

laboratory specimens will also be very low (afraction of 1 %).2 In such circumstances it maybe considered unrewarding to look for theparasite except in selected cases.The distribution of positive findings by age

varies: positivity rates, by age range, of morethan 62 000 patients studied in a two yearPublic Health Laboratory Service surveillancestudy are shown in the table.4

BIOLOGY OF CRYTOSPORIDIUMInfection in man and livestock is usually withC parvum. The biology and life cycle havebeen described in detail elsewhere.2" The lifecycle is complex, with several characteristicstages of development after ingestion of anoocyst. It is important to recognise that onlythe oocyst stage is normally detected in stoolspecimens, although other stages can be foundin other types of specimen such as biopsytissues, and in sputum where they may bepresent in exfoliated cells (Casemore DP,unpublished observation). The oocyst is a welldefined structure in which the componentparts and ultrastructures, mainly typical ofapicomplexan protozoa, can be shown byappropriate methods. The oocyst containsfour naked spindle-shaped sporozoites whichare released through a suture during excyst-ment. Most of the features cannot, however,be visualised in any detail by conventionallight microscopy. Differences in the degree ofsporulation and the routes of sporogenousdevelopment may be responsible for some ofthe variation in staining characteristics andinternal "structures" seen by light microscopyin stained preparations.

Laboratory diagnosisDiagnosis is generally by means of detectionof oocysts in faeces, and occasionally in otherspecimens; methods are described in detailbelow. The-endogenous (tissue) stages can beshown by light and electron microscopicaltechniques in biopsy and necropsy tissues.

445

on May 17, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.44.6.445 on 1 June 1991. Dow

nloaded from

http://jcp.bmj.com/

Casemore

Distribution of cryptosporidium positive stools by age*

Percentage Percentage Accumulative rateCrypto- positive positive of all per cent of all

Age sporidium by age cryptosporidium cryptosporidiumgroup n = positive group positive stools positive stools

< 1 4090 79 1-9 6-1 6-11-4 9880 480 4-9 37 0 43-25-14 4822 210 4-4 16-2 59-415-24 8235 160 1 9 12-4 71-725-34 9048 155 1-7 12-0 83-735-44 6710 83 1-2 6-4 90.145-54 4992 32 0-7 2-5 92-655-64 3846 12 07 09 935> 65 5432 21 0-4 1-6 95-1NK 5866 63 1 1 49 -

*Data derived from Palmer and Biffin.2

STOOL EXAMINATIONVarious methods have been described5-11 forthe detection and identification of oocysts inclinical specimens. Those described in detailhere are commonly used and are generallyreliable, given normal good laboratory prac-

tice, including the use of known positive con-

trols. Measurement of putative oocysts isessential and a method is described below.

STAINING METHODSModified Ziehl-Neelsen staining was

introduced for staining cryptosporidialoocysts by veterinary workers who had foundthat Cryptosporidium was associated withscouring of calves. They also recognised theproblems associated with the Romanowskystaining methods.5 Phenol-auramine stainingis very reliable, either when used as for. stain-ing mycobacteria or in a method developed forstaining oocysts which is also widely used.8 Amethod using saffranin and methylene blue hasbeen described by Baxby et al.9 Care needs tobe used in interpretation, particularly ofpreparations stained with modified Ziehl-Neelsen, as a variety of structures can beconfused with oocysts (so-called crypto-sporidium-like bodies). These include fungalspores which are generally larger (6-10 gm)than oocysts (4-6 jm), including mouldspores (such as some Mucor species), and thespores of the common mushroom: fat globulesand bacterial spores have also been mistakenfor oocysts by some workers, although thesecan be distinguished clearly by size alone.Excessive Ziehl-Neelsen staining can result infalse positive reactions from yeast cells.

IMMUNOLOGICALLY BASED METHODSDetection and identification of oocysts can beachieved using monoclonal antibody immuno-fluorescence (IFAT):'01' this often shows thecharacteristic suture line on the surface of theoocyst. Although the use of IFAT has beendescribed for routine stool examination, this isnot widely practised, partly because it isexpensive. Two monoclones are currentlyavailable commercially in the United King-dom in kit form, both containing monoclonalantibodies specific for oocyst wall epitopes. Adirect IFAT, developed in the United King-dom," is available from NorthumbriaBiologicals Ltd, Co Durham, and an indirectIFAT"0 is available from Meridian Diagnos-

tics Inc, USA, and from Launch Diagnostics,Longfield, Kent. A conjugated monoclonalreagent is also available from the PublicHealth Laboratory Service from the Divisionof Microbiological Reagents." None of thesemonoclones is completely specific for Cparvum. Details of methods are provided inthe manufacturer's instructions.Other immunologically based antigen

detection methods such as ELISA and passiveagglutination, such as latex particle tests, havebeen described but have not, so far, provedsufficiently reliable and are not generallyavailable.2 Commercial ELISA tests arecurrently being evaluated and they may even-tually replace microscopical methods in manylaboratories.

Several authors have reported antibodydetection methods which show that animmune response occurs with production ofall of the main antibody classes.' 212 Suchstudies are useful epidemiologically and forthe study of pathogenesis but are of littlevalue for diagnosis in individual cases and arenot described here.

Definitive identification of objects as oocystscan be made using transmission electronmicroscopy of thin sections of pelleted,embedded faecal samples,8 but this is of littlepractical value for diagnostic purposes.

CONCENTRATION METHODSFaeces from patients with acute crypto-sporidiosis do not usually require concentra-tion to detect oocysts, although the numbersof oocysts excreted can fluctuate during thecourse of the infection. Examination of con-centrated specimens may be merited in familycontacts of index cases, or when this isrequired epidemiologically. Concentration ofspecimens may sometimes be indicated in themanagement of immunocompromised patientswith a previous history of unexplained oruninvestigated diarrhoea. Such patients canoccasionally experience remission and sub-sequent recrudescence of cryptosporidiosis.'Concentration methods described includesucrose flotation'3 14 and formol-ether extrac-tion.8 Use of an unmodified formol-ethermethod can result in loss of oocyst8: themodification described below was developedfor recovery of cryptosporidial oocysts, butthe concentrate can also be used for the detec-tion of other parasites. Objects superficiallyresembling oocysts, except that they areosmotically less robust, have been reported tooccur in some sucrose flotation preparations.'5Although associated with gastrointestinalsymptoms, their nature and clinical impor-tance has yet to be resolved (Baxby D, per-sonal communication).

CONTROL OF INFECTIONIt is not currently possible reliably to confirmclearance of the infection given the relativeinsensitivity of microscopical methods ofdetection compared with bacteriological en-richment culture. For those with confirmedcryptosporidiosis, enteric precautions are

446



j.com/

J Clin P


nloaded from

http://jcp.bmj.com/

Laboratory methods for diagnosing cryptosporidiosis

0Vs

,

...

.:

I 'A "'*i '.

Figure 1 Faecal smear stained by modified Ziehl-Neelsen showing well stained oocysts,45-50 jim.

4,.~~~~~~~~~~~~~~~~~~t

ebX;;~~~~~~~~~~%L

Figure 3 Faecal smear stained by modified Ziehl-Neelsen showing mushroom sporeabout 6 x 8pum.

Figure 4 Faecal smear stained by modified Ziehl-Neelsen showing mould spores, about8-10 gm.

+-, wt '41

*t ;-|ait: *4' _sIvefw i+* w * £ ~~4-tt *.Fs ! f f P P /~~~6 .̂rf*<'tt < ~~~~~~i;* }

W_;t . >~~~~* '~J

J

4*.* .0

Fr

Figure 2 Faecal smear stained by modified Ziehl-Neelsen showing oocysts, 4 5-5 0 pm, with variablestaining.

advised while diarrhoea persists. Leave ofabsence from work for clinically recoveredsubjects is not required, provided that goodstandards of personal hygiene are main-tained."6

Laboratory proceduresSAFETY

es, Clinical specimens should be handled with thecare normally used for handling infectiousmaterial. It should also be recognised thatcryptosporidiosis is common in patients withAIDS whose specimens may contain a varietyof other infectious agents, including humanimmunodeficiency virus. Smears from suspec-ted or confirmed AIDS patients can beimmersed for one hour in freshly activatedglutaraldehyde solution before staining.Oocysts are unusually resistant to most dis-infectants but are killed by 10 vol. hydrogenperoxide.2" They are very sensitive to the effectsof heat, freezing, and desiccation.

Staining reagents can be hazardous to healthand should be assessed and handled accordingto the CQSHH regulations and guidelines.

QUALITY ASSURANCE/CONTROLA confirmed positive specimen should alwaysbe used to control staining and for comparativepurposes. Stains need to be batch tested when anew batch is purchased and staining times maysometimes need to be adjusted accordingly.Doubtful or equivocal positive samples shouldalways be checked by an alternative methodand can be sent to a reference laboratory.

447

Jr . -..4-.0 , *,. .,.*.h4

P.4 " "sk..e Jk-

M- --.,I L'

I

I

Flt .''e.

.:" t

-44

:7

04* I".



j.com/

J Clin P


nloaded from

http://jcp.bmj.com/

Casemore

t

*.3>

Figure 5 Sputum smear stained by modified Ziehl-Neelsen showing oocysts, 4 5-5 0 pm, with "erythrocyte" staining.

Figure 6 Faecal smear stained by auramine/carbol-fuchsin showing a group of oocysts with characteristic"erythrocyte" pattern of staining.

JA.

77I..f

.>ff,'

4k4

Figure 7 Purified oocysts of Cparvum stained by monoclonal antibody/FITCconjugate, counterstained with 0 1% Evan's blue, showing the suture on the oocystsurface.

Figure 9 Light micrograph ofgut tissue showing a group of endogenous stages ofCparvum in apparently superficial (pseudo-external) location. Plastic embedded semithinsection stained by toluidine blue (Courtesy ofDr A Curry).

Figure 8 Transmission electron-micrograph of an oocystin a faecal pellet showingfour sporozoites and acytoplasmic residual body.

".4

~~ ~~.4 e

V;' u'^i .a Jj|iL.i. -^*Si33; .rj

v- tt?" 3' {s>e i w

Figure 10 Light micrograph ofgut tissue showing agroup of endogenous stages ofC parvum in apparentlysuperficial (pseudo-external) location. Some cells haveapoptotic bodies (haematoxylin and eosin).

448

.,b,.5'

O..



j.com/

J Clin P


nloaded from

http://jcp.bmj.com/


Staining in troughs, as described below, has notcaused problems during more than seven years

ofexperience in this laboratory. Positive resultsshould be checked by examining a fresh smear

stained by an alternative method before report-ing.

SELECTION CRITERIA FOR SPECIMENEXAMINATIONIdeally, all diagnostic specimens from patientswith a history of acute or persistent gastro-enteritis should be screened. Otherwise, selec-tion must be based on clinical andepidemiological criteria to yield the greatestnumber of positive results. There is no validityin basing selection on the consistency of stoolspecimens submitted to the laboratory.Examination of the data on positivity by age

shows that about 60% ofpositives will be foundby screening specimens from children, andmore than 90% by extending screening toinclude adults up to the age of 45 years (table).For epidemiological purposes, including out-break control, data from adults can give an

early indication of waterborne outbreaks.

DIRECT EXAMINATION OF FAECES

Macroscopic appearance should be noted buthas little predictive value. Wet film examinationof faeces from patients with acute crypto-sporidiosis does not usually show the presenceof pus cells or red blood cells unless there isconcomitant infection with other organismssuch as campylobacter; mucus is present insome. Oocysts cannot readily be identified insimple wet preparations, even in heavily infec-ted specimens, and are difficult to distinguishreliably from yeast cells or fungal spores,

particularly when the oocysts are present insmall numbers. With experience such bodiescan be distinguished in iodine wet-mounts, theoocysts remaining unstained, the yeasts stain-ing brown. In Sheather's sucrose solution, withsome optical systems, oocysts can have a pink-ish hue compared with yeasts which appear

greenish and in a lower focal plane. Specialisedoptical systems such as Nomarski DifferentialInterference Contrast (DIC) can also showinternal structural details. In general, however,staining methods with fixed smears are to bepreferred.

Staining methodsSLIDE PREPARATIONSmears, about 3 cm x 1 cm, are made on

standard 3" x 1" glass microscope slides. Thebest results are obtained with moderately thickfaecal smears made with the aid of a woodenapplicator stick to give both thick and thinareas. The smears are allowed to air dry and can

be lightly heat-fixed. Oocysts may be attachedto mucus which tends to wash off the slide. Itmay help, with these and with non-fluidspecimens, to emulsify a portion ofthe faeces ina little formol-water containing 0-1% Tween80 (Casemore DP, unpublished observation).1 Modified Ziehl-Neelsen (MZN)Reagents The reagents are unmodified fromthose recommended for the Ziehl-Neelsen

staining of mycobacteria, as described in stan-dard bacteriological texts. Kinyoun's modifica-tion can also be used.Method The following modification has beenfound to be convenient and to give goodresults8: (i) slides are placed in multislidecarriers for fixation and staining in batches introughs. Fix in methanol for three minutes. (ii)Stain with strong carbol fuchsin for about 20minutes (avoid excessive staining), followed bythorough rinsing in tap water. (iii) Give mini-mal decolourisation by agitation in a trough of1% hydrochloric acid in methanol (15-30seconds), followed by rinsing in tap water. (iv)Counterstain for 30-60 seconds in 0 4% mala-chite green (or methylene blue), rinse well, andair dry. (v) Examine by brightfield microscopyusing x 50 and x 100 oil-immersion objectivelenses. Some workers prefer to mount a cover-slip on the smear (using immersion oil, or DPXmountant, or similar) and scan using a lowerpower dry objective lens. Suspicious bodiescan then be further examined using an oil-immersion high power objective lens.Results Oocysts are characteristically roundor slightly ovoid with a consistent modal size,usually of about 4 5 x 5 0 gm (range 4-6 pm).They are acid-fast but oocyst staining, within asmear and between specimens, is very variable,and oocysts vary from unstained to partial redstaining and complete staining: "erythrocyte"stained forms are common, and fullysporulated forms can be found in which redstaining crescentic bodies, the sporozoites, canbe seen within an unstained oocyst wall. (Note:Cryptosporidium in histological sections andempty, excysted, oocyst cases are not acid-fast).2 Phenol-auramine/carbol-fuchsin method8This simple two step method combines fluores-cent staining using auramine with negativestaining by strong carbol-fuchsin to mask back-ground material. Handling time is brief andslides can be stained in multi-carriers asdescribed above. Some workers use phenol-auramine staining, as recommended formycobacteria or incorporate a brief acid-alco-hol decolourisation step into the method des-cribed here.Reagents Suitable reagents are those de-scribed in standard texts for the staining ofmycobacteria: commercially produced solu-tions are generally satisfactory, but see note onquality control. The phenol-auramine formu-lation used in the author's laboratory is com-mercially prepared Lempert's solution and thecarbol-fuchsin is that used for Ziehl-Neelsenstaining, also commercially prepared.Method (i) Slides are placed in multislidecarriers for staining in batches in troughs,without fixation or decolourisation. Stain withphenol-auramine for 10-15 minutes, rinse wellin tap water. (ii) Counterstain by immersing instrong carbol-fuchsin briefly (a few seconds),rinse well, and air dry. (iii) Examine by incidentlight fluorescence microscopy, usingappropriate filters, with low and high power drylens objectives.Results Oocysts appear as bright yellow discs,often with an "erythrocyte" pattern of stainingagainst a dark red background. Yeasts, fumgal

449 on M

ay 17, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.44.6.445 on 1 June 1991. D

ownloaded from

http://jcp.bmj.com/

Casemore

spores, and other cryptosporidium-like bodiescannot readily be confused with oocysts in thismethod once the characteristic appearance ofoocysts has been learned.NB: Presumptive positive smears stained withphenol auramine (standard method or phenolauramine/carbol-fuchsin) can be confirmed, orequivocal fluorescent bodies examined further,by location using a Vernier reading or anEngland Finder slide, overstaining with MZN,and re-examining. Oocyst measurementsshould be made in this way-not on fluorescentobjects.

Measurement of oocystsMeasurements are usually made with an eye-piece graticule calibrated by means of amicrometer scale on a microscope slide(Graticules Ltd, Tonbridge, Kent). Thecalibration procedure needs to be carried outonly once for a given ocular and objective lenscombination on a particular microscope. It isessential, however, to appreciate that thecalibration is for that microscope and lenscombination only and cannot be used for otherlens combinations or with corresponding lenseson other microscopes, which must beseparately calibrated.Procedure (i) Place the eye-piece graticule onthe focal plane diaphragm (between the eye lensand field lens) within the ocular lens. Thisgraticule scale, usually 1 cm divided into 1 mmdivisions, is an arbitrary transfer scale. Returnthe ocular lens to the microscope and focus thegraticule using the eye-piece focusing ring, ifpresent. (ii) Focus on the scale on a slidemicrometer-usually 1 mm divided into 0-1and 0h01 mm-using a low power objective lensto locate the scale. Change to the objective lensto be calibrated and adjust the position of thestage so that the "0" line of the two scales aresuperimposed and clearly focused. (iii) Takingcare not to move the stage, look for other lineswhich are superimposed, as far along the scaleas possible. Record the number of oculardivisions and the distance in mm on the slidebetween the superimposed sets of lines. (iv)Calculate the distance covered by the oculargraticule divisions as follows:1 ocular unit in ,um =

[ distance in mm covered 1x 1000number of ocular divisions 1

Thus for example, with a particular objectivelens, if0-25mm on the micrometer slide scale iscovered by 85 ocular divisions, one oculardivision is equivalent to 2-47 Mm: an objectcovered by two divisions when viewed withthat objective will thus be 4-84 sum. (v) Repeatthe process for each objective to be calibratedand record. The eye-piece graticule can con-veniently be left in place.

Concentration3 Modifiedformol-ether method8Reagents (i) 10% formalin in water (formol-water), (ii) ether (note general comment onsafety, above).Method (i) Emulsify about 0 5 cm3 of stool in a

glass universal in a small volume of formol-water, using an applicator stick to break upformed stools if required, make up to about 3cm3, and mix well using a vortex mixer. (ii) Addabout 3 cm' of ether and shake vigorously for30-40 seconds. (iii) Make up to about 15 cm'with formol-water, remix, and pour through a40 mesh gauze sieve (Endecott's Ltd, London:certified test sieve, 3 inch ASTM, 425 gm) intoa 15 ml glass conical centrifuge tube. (iv) Topup tube with formol-water if required, centri-fuge in a safety bucket by setting centrigue at450 x g (about 1000 rpm) and spin for oneminute. (v) Using a disposable plastic bulbpipette, carefully remove column of fluid frombetween the ether layer and the deposit to asecond tube. Top up the second tube withformol-water, centrifuge at 1000 x g for 10minutes, discard supernatant fluid and usedeposit to prepare smears for staining in theusual way (the primary and secondary depositscan also be examined for ova and cysts of otherparasites if required).

Histological diagnosisAlthough the diagnosis of cryptosporidiosis isusually made microbiologically, identificationof the endogenous (tissue) stages of the parasitein biopsy or necropsy tissues is sometimesrequired. The parasite can be seen in suchmaterial when processed and stained by con-ventional histological methods, such ashaematoxylin and eosin staining ofparaffin waxembedded tissue, although it can be incons-picuous and readily overlooked. Better resultscan be obtained by the method described inoutline below (Curry A, personal communica-tion).Method (i) Fix about 1 mm' pieces of tissue in3% 0-1 M cacodylate buffered glutaraldehyde.(ii) Wash in buffer and then postfix in buffered(pH 7 2) 1% osmium tetroxide. (iii) Dehydratein graded alcohols and embed in plastic resinand cut semithin (about 1 gm) sections. (iv)Stain with 1% aqueous toluidine blue, mountpreparation, and examine by light microscopy.Results The parasite can be seen as small (2-8gm), single-celled, round bodies in an appar-ently superficial, pseudo-external, locationalong the brush border, particularly on apicalenterocytes and in the crypts. There may besome change in villous architecture andevidence of inflammation, and some cellularinfiltrate into the mucosa.

Blocks thought to merit further detailedstudy and to confirm the identity of possiblepositives can be examined by transmissionelectron microscopy of ultrathin sections fol-lowing staining-for example, by uranylacetate and lead citrate. The various stages ofthe parasite have been described and elegantlyillustrated by Current et al.37

1 Casemore DP. Human cryptosporidiosis. In: Reeves DS,Geddes AM, eds. Recent advances in infection, No 3.Edinburgh: Churchill Livingstone, 1989:209-36.

2 Casemore DP. Epidemiological aspects of human crypto-

450



j.com/

J Clin P


nloaded from

http://jcp.bmj.com/


sporidiosis. Epidemiol Infect 1990;104:1-28.3 Current WL. Cryptosporidium: Its biology and potential for

environmental transmission. CRC Crit Rev Env Control1986;17:21-51.

4 Palmer SR, Biffin A. Cryptosporidiosis in England andWales: prevalence and clinical and epidemiologicalfeatures. Br Med J 1990;300:774-7.

5 Henriksen SA, Pohlenz JFL. Staining ofcryptosporidia by amodified Ziehl-Neelsen technique. Acta Vet Scand 1981;22:594-6.

6 Ma P, Soave R. Three-step stool examination for crypto-sporidiosis in 10 homosexual men with protracted waterydiarrhea. J Infect Dis 1983;147:824-8.

7 Garcia LS, Bruckner DA, Brewer TC, Shimuzu RY.Techniques for the recovery and identification of Crypto-sporidium oocysts from stool specimens. J Clin Microbiol1983;18:185-90.

8 Casemore DP, Armstrong M, Sands RL. Laboratory diag-nosis ofcryptosporidiosis. J Clin Pathol 1985;38:1337-41.

9 Baxby D, Blundell N, Hart CA. The development andperformance of a simple, sensitive method for the detec-tion of Cryptosporidium oocysts in faeces. J Hyg (Camb)1984;92:317-23.

10 Arrowood MJ, Sterling CR. Comparison of conventionalstaining methods and monoclonal antibody-based

methods for Cryptosporidium oocyst detection. J ClinMicrobiol 1989;27:1490-5.

11 McLauchlin J, Casemore DP, Harrison TG, et al. Identifica-tion of cryptosporidium oocysts by monoclonal antibody.Lancet 1987;i:51.

12 Current WL, Bick PH. Immunobiology of Cryptosporidiumspp. Pathol Immunopathol Res 1989;8:141-60.

13 Sheather AL. The detection of intestinal protozoa andmange parasites by floatation technique. J Comp PatholTher 1923;36:260-75.

14 Current WL, Reese NC, Ernst JV, et al. Human crypto-sporidium in immunocompetent and immunodeficientpersons. N Engl J Med 1983;308:1252-7.

15 Baxby D, Blundell N. Recognition and laboratory character-istics of an atypical oocyst of Cryptosporidium. J InfectDis 1988;158:1038-45.

16 Anonymous. Notes on the control of human sources ofgastrointestinal infections, infestations and bacterialintoxications in the United Kingdom. CommunicableDisease Report Supplement No 1. London: Public HealthLaboratory Service, Communicable DiseasesSurveillance Centre, 1990.

17 Current WL, Reese NC. A comparison of endogenousdevelopment of three isolates of Cryptosporidium insuckling mice. J Protozool 1986;33:98-108.

451



j.com/

J Clin P


nloaded from

http://jcp.bmj.com/

Documents

Laboratorymethods for diagnosing cryptosporidiosis · Modified Ziehl-Neelsen staining was introduced for staining cryptosporidial oocysts byveterinary workers whohadfound that Cryptosporidium