Download - Streptococci from primary isolation plates grouped by reverse passive haemagglutination

J Clin Pathol 1991;44:670-675

Streptococci from primary isolation platesgrouped by reverse passive haemagglutination

M Farrington, S Winters, D Rubenstein, J Greatorex, R Whetstone

AbstractReverse passive haemagglutination, anovel microtitre based assay, was com-pared with the Streptex (Wellcome UK)latex slide agglutination kit for strepto-coccal grouping in a diagnostic micro-biology laboratory. Three hundred andfifty two extracts from 349 consecutiveprimary isolation plates were assayed byboth methods. Reverse passive haemag-glutination gave identical groupingresults for 98-0% of the 345 streptococciidentified by Streptex, and the K coefficientof agreement between the methods forall 352 extracts tested was 0-973. Crossreactions with Listeria spp seen withStreptex were not found by reverse pas-sive haemagglutination. In the reversepassive haemagglutination method 11streptococci could be grouped on each96-well plate and most reactions werestable for at least 30 minutes.Reverse passive haemagglutination is

more rapid to perform than latex slideagglutination when many organisms areto be grouped, and the patterns ofhaemagglutination are easily recognised.If the method was taken into routine usein a diagnostic laboratory, the persis-tence of reverse passive haemagglutina-tion reactions would enable groupingresults to be checked for quality controlpurposes.

Department ofClinical Microbiologyand Public HealthLaboratory,Addenbrooke'sHospital, Hills Road,Cambridge CB2 2QQM FarringtonS WintersJ GreatorexR WhetstoneDepartment ofMedicineD Rubenstein

Correspondence to:Dr M FarringtonAccepted for publication25 March 1991

Grouping of haemolytic streptococci, whichtraditionally depends on time consuming andexpensive capillary precipitation methods, isusually performed in diagnostic laboratoriesby one of the commercially available "co-agglutination" (COA) or latex agglutinationkits.' Several publications attest to thespecificity and sensitivity of these methods.2Each test takes only a few minutes of technicaltime, although antigen extraction is usuallyrequired and specificity is reduced whenagglutinations are performed on colonies fromprimary isolation plates. Skill is needed in theinterpretation of the agglutination reactions,however, and none of the reported techniquesgives the opportunity for batch or automatedprocessing if large numbers of tests are to beundertaken. Furthermore, specific agglutina-tion is a transient phenomenon that makesquality control of the results difficult toachieve.

Reverse passive haemagglutination(RPH)'0 " relies on the agglutination inmicrotitre trays of antibody coated red cellsby specific antigen. Agglutinability has been

optimised in some systems by prior treatmentof the red cells with proteolytic enzymes, andlong term stability of reagents can be achievedby glutaraldehyde stabilisation and freeze dry-ing.We have grouped clinical isolates from 349

consecutive primary isolation plates in areverse passive haemagglutination assay withstabilised, freeze dried sheep erythrocytescoupled to Lancefield group specific antiseraagainst streptococcal groups A, B, C, D, Fand G (Wellcome Streptococcal GroupingSera, ZJ series; Wellcome Diagnostic, Dart-ford, England). We compared the results withthose obtained by the Streptex LA kit (Well-come Diagnostics). Discrepant reactions werefurther investigated with the PhadebactStreptococcus slide COA system (Karo BioDiagnostics, Sweden) and the API 20 Strepkit (BioMerieux, France).7

MethodsPREPARATION OF STOCK REVERSE PASSIVEHAEMAGGLUTINATION REAGENTSAntibody coupled erythrocytes were preparedas follows: sheep red blood cells that had beenwashed in phosphate buffered saline (PBS)were incubated for 20 minutes at 37'C with a-chymotrypsin in acidified PBS/azide com-plement fixation test buffer (BR16, Oxoid).Stock chromic chloride coupling solution wasmade by dissolving CrCl3 . 6H2O (BDH,England) to a 1% concentration in pyrogen-free physiological saline and adjusting the pHto 5-0 with 5M sodium hydroxide. Immuno-globulin was concentrated from the WellcomeStreptococcal Grouping Sera by precipitationwith 35% ammonium sulphate. Group Aserum was used at 2 mg ml - and groups B,C, D, F and G at 1-5 mg ml-'. The purifiedimmunoglobulins were then coupled to sheepred cells in piperazine buffer." Optimal con-centrations of coupling solution for groups A,B, D, F and G were 1 in 80, and for group C 1in 100. Optimal coupling was confirmed bytitration with anti-rabbit globulin. Samples(1 ml) of coupled cells were stabilised with100 4ul of glutaraldehyde (BDH; 30 Ml in 5 mlPBS) and then freeze dried in PBS with 1-5%sucrose and 1-0% bovine serum albumin(Calbiochem, USA) in an Edward's Modulyofreeze dryer overnight. Before use cells werereconstituted with gentle mixing in PBS totheir original volumes at 4°C for 24 hours;shorter reconstitution times gave weaker dis-tinction between positive and negative results.

ISOLATION OF STREPTOCOCCIIn our diagnostic laboratory primary isolation

670

group.bmj.com on July 15, 2011 - Published by jcp.bmj.comDownloaded from

http://jcp.bmj.com/

http://group.bmj.com/

Streptococcifrom primary isolation plates grouped by reverse passive haemagglutination

plates growing presumptive ,B haemolyticstreptococci or enterococci are passed from alllaboratory areas to a central "streptococcalbench" for confirmation by macroscopicexamination of colonies and sometimes byGram stain, Streptex grouping, and, if neces-sary, API 20 Strep testing. For the purposesof this study, specimens were classified as"skin" (surface swabs sent from general prac-titioners and the hospital dermatology depart-ment), "respiratory" (nose and throat swabs,and sputum), "genital", "urine", and"wound" (surface swabs, tissues, andaspirates from inpatients). Putative strep-tococci from urine were grouped off cysteine-lactose-electrolyte-deficient agar (CM 301;Oxoid UK) and all other # haemolytic isolateswere grouped off bilayered blood agar plates,the top layer comprising 8 ml heart infusionagar (0044-17; Difco, UK) with 10% defibrin-ated horse blood and the bottom layer com-prising 8 ml Blood Agar Base No 2 (CM 271;Oxoid). Nine organisms submitted for group-ing were recognised at this stage as beingviridans streptococci or as belonging to othergenera, but were grouped by both systems asadditional controls.

GROUPING OF STREPTOCOCCIStreptex grouping was performed by one of us(RW) using a modification of the manufac-turer's recommendations on extracts fromconsecutive plates over a five week period. Ifpossible, five or six colonies were picked fromeach primary plate, care being taken tominimise contamination from adjacentcolonies, and emulsified in 0-2 ml of extractionenzyme (ZL 55, Wellcome, diluted in 22 mldistilled water) in Dreyer's tubes. There wereincubated at 56'C for eight minutes, and sixseparate 0-02 ml drops of the extracted super-natant were then transferred by a standardplatinum loop to clean glass slides. Such scal-

Figure Illustrativemicrotitre plate showingreverse passivehaemagglutinationreactions of extractsfrom11 primary isolation platesand controls. Reactions areasfollows: column 1,strong D; 2, strong A; 3,weak G; 4, intermediate Cand weak G; 5, strong C;6, negative; 7,intermediate A; 8,intermediate G; 9, strongB; 10 and 1I strong A;12, control negative.

ing down of the quantities of reagent used hasbeen reported to be reliable and economic.6Latex and control reagents were added to thedrops by the standard 0 02 ml loop, mixed,the slides rocked, and agglutination read byeye usually within 20 seconds. Reactions wereclassified as strong or weak. Each batch ofreagent was controlled against the manufac-turer's positive control antigen before it wasused.The remainder of the extracts were passed

daily to SW, who performed reverse passivehaemagglutination without examining theprimary culture plates and without knowingthe results of Streptex grouping. Neat strep-tococcal extracts were diluted 1 in 8 with PBScontaining 011% normal rabbit serum.Diluted extract (30 p1) was mixed with 30 ylof reconstituted red cells for each group inAljen "V-well" microtitre plates (Aljen,Burgess Hill, Sussex; product code 100305).Sterlin (Hounslow, Middlesex; code 611 V96),Dynatech (Billingshurst, Kent; code M25ADynatech V), Nunc (Gibco, Paisley, Scotland;code 2-45128A) and Greiner (Dursley,England; code 651161) plates were also inves-tigated for use in the assay, but the angle ofthe Aljen wells gave the best balance ofquality between negative and positive re-actions (figure). Eleven extracts were tested oneach plate. One negative control well for eachgroup was included on each plate, which con-tained buffer and coupled red cells but noorganism extract. Plates were shaken to mix,and the result of haemagglutination was readby the sedimentation pattern after 45 minutesat room temperature. Additional readingswere taken after a further 30 and 60 minutesat room temperature, and after overnightstorage at 4'C.

Extracts of 52 control organisms wereprepared and tested by Streptex and reversepassive haemagglutination. Positive controls

1 2 3 4 5 6 7 8 9 l 1O1.e to

A 0 0 * 0.4

B , . 0 0 0 0 S*

-, ... ..... .

C

D

F

0

9 0 0 : v.

S

S

12

.0

4 4

0 * * I*

t1

0 0 00 0 0 0 ~0 0

0 4 0 A 0 00 0 0 IV'

671


http://jcp.bmj.com/


Farrington, Winters, Rubenstein, Greatorex, Whetstone

Table 1 Site of origin and Lancefield group of 338 streptococci with identical reversepassive haemagglutination and Streptex results

Group

Site A B C D F G t

Skin 15* 5 4* 3* 2 12 41Respiratory 126 3 12 0 1 13 155Genital 7 25 1 16 2 6 57Urine 0 8 0 45 0 1 54Wound 2 4 1 12 3 9 31t 150 45 18 76 8 41 338

*Reactions with more than one antigen in both systems were confirmed in one extract to be dueto a mixed culture of streptococci of groups A and C, and in another to a mixture of groups Aand D.

comprised two group A streptococci (Spyogenes NCTC 8198 and 8322), four group B(S agalactiae NCTC 8180, 8185, 8190, 8194),four group C (C dysgalactiae NCTC 4335, Sequisimilis NCTC 5371 and clinical isolates ofS equi and S zooepidemicus), four group D (Sfaecalis NCTC 775, and clinical isolates of Sfaecium 2 and 3, and S bovis), one group F(NCTC 5389) and one S anginosus (group G;NCTC 10713). Negative controls comprised14 recent clinical isolates of viridans and non-haemolytic streptococci, two S pneumoniae(NCTC 7465, 7978), one S salivarius (NCTC8618), two S aureus (NCTC 6711, 8178), twoListeria monocytogenes (NCTC 11994 and10887), two L innocua (NCTC 11288 and astrain obtained from DMRQC, Colindale),one L seeligeri (DMRQC), one L grayi(NCTC 10815), one L murrayi (NCTC10812), and recent clinical isolates ofEnterobacter cloacae (2), Pseudomonas aerugin-osa (2), Serratia liquifaciens (2), Klebsiellapneumoniae, Citrobacter diversus and Coryn-ebacerium sp.When the results of Streptex and reverse

passive haemagglutination grouping werecontradictory, primary isolation plates werere-examined and subcultured, and coloniesretested by SW using the Phadebact Strepto-coccus Test, which includes reagents forLancefield groups A, B, C, D, F and G. Themanufacturer's instructions for the DirectColony Test were followed. Where possible,subcultured ,B haemolytic colonies were alsoidentified using the API 20 Strep systemaccording to the manufacturer's recommenda-tions.

ResultsThree hundred and fifty two extracts were

tested from the 349 consecutive primary isola-tion plates. Nine ofthese extracts were negativein both systems and were confirmed not to be f,haemolytic streptococci or enterococci, andtwo of the remaining 343 extracts reacted withmore than one grouping reagent in both sys-

tems. Discrepant results between Streptex andRPH were seen with seven extracts; the refer-ence method, Streptex, identified 345 strepto-cocci and reverse passive haemagglutinationgave identical grouping results in 338 of these(98-0%). Table 1 shows the distribution ofthese isolates by specimen of origin andLancefield group.

All reverse passive haemagglutination con-

trol wells were negative. Three primary isola-tion plates (from one skin swab, one respiratoryspecimen, and one urine sample) containedmixed cultures of fl haemolytic streptococcibefore Streptex grouping was performed, andthese isolates were submitted separately forStreptex and reverse passive haemagglutina-tion grouping and are counted individually intable 1. Two extracts gave the same mixedreactions by both techniques (one groupsA andC, one groups A and D, both from skin swabs)and were confirmed to contain mixtures ofthesestreptococci on subculture. Of the nine organ-

isms found not be to fl haemolytic streptococcior enterococci, four were viridans streptococci(respiratory), three diphtheroids (one skinswab, two urine samples), one Haemophilusparainfluenzae (respiratory) and one a Gramnegative rod (urine).Table 2 correlates the reactions of each

extract using the Streptex and reverse passivehaemagglutination methods. The K coefficientof agreement between the two methods was

high at 0 973.9 12 Table 3 gives details of theseven extracts from seven primary isolationplates that gave discrepant grouping results.

Five anomalous results were obtained on

testing the 39 control extracts. One clinicalisolate of Ps aeruginosa gave weak positivereactions with all groups by Streptex but was

negative by reverse passive haemagglutination,and one clinical isolate of Klebsiella pneumoniaegave weak reactions against group C, D, and Freagents by reverse passive haemagglutinationbut was negative by Streptex. Several Listeriaspp gave floccular reactions with Streptexreagents, distinct from the usual granularity: Lmonocytogenes NCTC 11994 was positive withgroups D and F, and weakly positive with

Table 2 Reactions by Streptex and reverse passive haemagglutination techniques of 352 extractsfrom presumedstreptococci

Reverse passive haemagglutination grouping reaction:

N A A A A A C AStreptex A B C D F G G + + + + + + Igroup p B C D F G F 1 t

A 148 1 1 150B 45 45C 17 1 18D 1 75 76F 8 1 1 10G 41 1 42NGP 9 9A+C 1 1A±D 1 1t 149 45 17 75 8 41 10 1 1 1 1 1 1 1 352

All = reaction with all six (A, B, C, D, F and G) grouping reagents.NGP = no grouping reaction detected.

672


http://jcp.bmj.com/



Table 3 Details of seven extracts with discrepant Streptex and reverse passive haemagglutination results, and resultsof investigation after subculture

CaseNo Site Streptex RPH Phadebact Comment

I Skin C+ C + /- C + /- Heavily mixed culture, No groups C or FF + /- F+ /- streptococci isolated on subculture

2 Respiratory A + A+ + A + API Strep confirmed S pyogenes,B + /- B + /- only group A streptococci isolated on subculture

3 Respiratory G + G + G + API Strep confirmed group G streptococcus,A+ /- only group G streptococci isolated on subculture

4 Genital A + A + + A+ API Strep confirmed S pyogenes,F + /- F + /- heavily mixed primary culture, only group A

streptococci isolated on subculture5 Urine D+ A+++ ND Primary plate overgrown with Proteus sp6 Wound F + /- All All API strep confirmed S cremoris

7 Wound F + /- - ND Organism died before further testing

Key:Streptex and Phadebact reaction strength: + strong; + /-weak.RPH reaction strength: + + strong; + intermediate; + /-weak;- negative.All: positive reactions with groups A, B, C, D, F and G.ND: not done.

groups A, C and G; NCTC 10887 was positivewith group D and weakly so with groups A andC. L innocua (DMRQC) was positive withgroups D and F and weakly so with group G,and NCTC 11288 was weakly positive withgroup C. L seeligeri gave a weak group Dreaction. L monocytogenes NCTC 11994 alsogave floccular positive reactions with allPhadebact reagents, but no other discrepancieswere seen using this techique. Reactions of Sanginosus (group G) and group F streptococcusNCTC 5389 were strong in both reverse pas-sive haemagglutination and Phadebact tests,but weak by Streptex. Negative controls gavenegative reactions throughout.

Reverse passive haemagglutination reactionsare generally easy to read and classify, butbecause of the different characteristic reactionsof each grouping reagent we found it importantto compare reactions horizontally with othersof the same group rather than vertically be-tween different groups (figure). Consideringonly the 345 extracts that did not give dis-crepant reactions with Streptex, the followingpercentages of reverse passive haemagglutina-tion reactions were strong or intermediate at45, 75 and 105 minutes and overnight: 98 6,84 7,734 and 674 for group A; 82-2,71-1,71-1and 64 5 for group B; 100, 88-9, 88-9 and 77.9for group C; 81 6, 68 4, 63 2 and 55 2 for groupD; 87-5, 75-0, 75-0 and 75 0 for group F and75 6, 317, 22-0 and 9-7 for group G.No weakly positive reactions were seen with

group A, B, C and D Streptex reagents, but sixof 10 reactions (60%)with the group F reagentand two of 42 (4-8%) with group G were weak.According to the reverse passive haemag-

glutination methods we used, positive controlsfor each group could readily be included onevery plate, and it would also be possible to testextracts against freeze-dried red cells coupledto rabbit non-specific polyclonal antibody as anadditional negative control. There were suf-ficient wells to have included four more testorganisms, but clarity and ease of comparisonwould be reduced.

After the extracts had been received thetechnical time to set up and read these 11 testswas about 25 minutes (including reconstitutionof reagents, dilution of extracts, inoculation ofwells and reading of the plate). This compares

with about 66 minutes for 11 organisms testedagainst all six antisera by Streptex.

DiscussionThe results of streptococcal grouping shouldbe available quickly, preferably within 24 hoursof submission of the specimen if they are tohave the best chance of influencing clinicalmanagement. This implies that grouping testsmust be carried out on colonies from primaryisolation plates as they were in this study,unless direct antigen detection is performed onclinical material.There was a high degree of agreement be-

tween the grouping reactions of reverse passivehaemagglutination and Streptex. This was des-pite reverse passive haemagglutination testingbeing performed blind of the colonial appear-ances on the primary plate, with some extractscontaining antigens from contaminating bac-teria. Ninety eight per cent of streptococcigrouped by Streptex gave indentical results byreverse passive haemagglutination. The Kcoefficient of agreement of 0 973 between bothmethods was higher than that (0-958) pre-viously reported in a comparison between alatex agglutination kit and a traditional Lance-field grouping method.9 This coefficientmeasures the proportion of times two methodsagree compared with the maximum possiblenumber of times they could agree, corrected forchance agreement.'2 We did not use Lancefieldgrouping as a reference because it is notapplicable to mixed primary isolation plates,and we wished to compare methods currentlyavailable to a clinical diagnostic laboratory.Determinants of the specificity and sen-

sitivity of the available serological groupingmethods include the quantity of streptococcalcells assayed, the adequacy of any antigenextraction method used (which may vary forstreptococci of different groups'3), the inactiva-tion of protein A-like substances,'4 details ofsuspension media5 and the presence of cross-reacting antigens derived from other flora onthe plate,8 15 which may not be epitope specific.These factors may be adversely affected ifgrouping is done from primary isolationmedia,2-4 8 9 16 and the best correlations betweenrapid and reference methods have been ob-

673


http://jcp.bmj.com/


Farrington, Winters, Rubenstein, Greatorex, Whetstone

tained when overnight pure broth cultureshave been used,3 or when the manufacturer'srecommendations for the numbers of coloniesto be extracted have been exceeded.2 Neverthe-less, the resulting cross reactions are usuallyweak and in experienced hands rarely result inincorrect grouping by the Streptex kit.6 Someauthors have applied direct grouping fromprimary plates selectively to those cultures withlarge numbers of easily separable colonies.8

Extraction procedures have been modified inseveral ways for increased speed, but coag-glutination methods have the advantage thatextraction may not be required, presumablybecause the Fc portions of group specificantibodies are bound to the carrier staphy-lococci and are not available for cross reactionto protein A-like substances on group A, C, andG streptococci.'4Of the seven extracts that gave discrepant

results (table 3), four were due to additionalreactions to other groups in reverse passivehaemagglutination. In three ofthese Phadebactalso gave the same reaction, and in two theextracts came from heavily mixed cultures.Small amounts of Lancefield antigen fromstreptococci or other species might have beenpresent in these extracts, or non-epitope-specific cross reactions might have occurred toimmunoglobulins or red cells. The strongreaction with group D Streptex of one extractthat gave a strong group A reaction with reversepassive haemagglutination may have been dueto a true difference in reactivity, to splash-overbetween wells on the reverse passive haemag-glutination microtitre plate, or to a transcrip-tion error, but confirmation of this was notpossible because the primary plate had beenheavily overgrown with Proteus sp. There wereextracts with strong group A reactions on thismicrotitre plate, but they were not immediatelyadjacent to the extract in question. The weakreaction with all groups by reverse passivehaemagglutination seen with one S cremorisisolate from an infected abdominal surgicalwound, which normally carries the group Nantigen,'7 would be clinically less confusingthan the group F reaction observed in theStreptex system.

Cross reactions between Streptex reagentsand L monocytogenes have been reported,'5 butwere not noted to be floccular and were mostcommonly against group B reagents. We ob-served floccular reactions with L monocytogenesagainst all groups except B, and also foundstrains of L seeligeri and innocua to give crossreactions. MacGowan et al suggested thatthese problems may be due to rhamnose de-terminants shared between streptococcalpolysaccharides and listerial 0 antigens,"5 butnon-epitope-specific reactions to bound im-munoglobulins or carrier particles were notexcluded. This may explain the lack of crossreactivity in the reverse passive haemagglutina-tion system.

Strong or intermediate positive reactions inreverse passive haemagglutination persistedwell for at least 30 minutes after the conven-tional reading interval of45 minutes. Checkingof results by other staff in a clinical laboratory

would therefore be possible. All extracts testedexcept one (group F by Streptex) were group-able by reverse passive haemagglutination, butgroup G strains gave the weakest reverse pas-sive haemagglutination reactions with onlythree quarters classified as strong or inter-mediate. Such differences are likely to be due tobatch variation in antibody reactivity, or incoupling to the red cell carrier. In commonwith our experience, others have reportedgroup F reactions in the Streptex system to beweak: Keville and Doern found 41% of suchstrains to be ungroupable after enzyme extrac-tion from pure subcultures. 8 In reverse passivehaemagglutination we used only single batchesof each antibody, but we did find some inter-batch variation among successive couplings(data not shown). The reactions of each reversepassive haemagglutination grouping reagentwere consistent, but qualitative differences wereobserved between groups (figure). Part of thisvariation could probably be avoided and thestrength of group G reactions improved byfurther modifications to the coupling or freeze-drying procedures, and for routine laboratoryuse it would be helpful to include positive aswell as negative controls for each group onevery plate.

Batch processing of extracts would be amajor potential advantage of a streptococcalgrouping system based on reverse passivehaemagglutination, and considerable savings oftime would result for laboratories groupinglarge numbers of organisms. Centralisedgrouping in one laboratory area has the dis-advantage of separating the initial recognitionof a streptococcus from its final confirmation,and reorganisation of work would be needed inmany laboratories, but standardisation shouldimprove and quality control be simplified.

We are grateful to RRA Coombs, Emeritus Professor, Univer-sity of Cambridge, for helpful discussion and encouragement,and to the East Anglian Regional Health Authority ResearchCommittee for their support.

I Facklam RR, Carey RB. Streptococci and aerococci. In:Lennette EH, Balows A, Hausler WJ, Shadomy HJ, eds.Manual of clinical microbiology. Washington, DC:American Society for Microbiology, 1985:154-75.

2 Facklam RR, Cooksey RC, Wortham EC. Evaluation ofcommercial latex agglutination reagents for groupingstreptococci. J Clin Microbiol 1979;10:641-6.

3 Slifkin M, Pouchet-Melvin GR. Evaluation of three com-mercially available test products for serogrouping beta-hemolytic streptococci. J Clin Microbiol 1980;11:249-55.

4 Silfkin M, Interval G. Serogrouping single colonies of beta-hemolytic streptococci from primary throat culture plateswith nitrous acid extraction and Phadebact streptococcalreagents. J Clin Microbiol 1980;12:541-5.

5 Engel HWB, Slithout A. Simplified coagglutination test forserological grouping of beta-haemolytic streptococci.J Clin Microbiol 1981-14:252-5.

6 Castle D, Kessock-Philip S, Easmon CSF. Evaluation ofan improved Streptex kit for the grouping of beta-haemolytic streptococci by agglutination. J Clin Pathol1982;35:719-22.

7 Keville MW, Doern DV. Comparison of the API 20SStreptococcus identification system with an immuno-rheophoresis procedure and two commercial latexagglutination tests for identifying beta-haemolytic strep-tococci. J Glin Microbiol 1982,16-92-5.

8 Hamilton JR. Comparison of Meritec-Strep with Streptexfor direct colony grouping of beta-hemolytic streptococcifrom primary isolation and subculture plates. J ClinMicrobiol 1988;26:692-5.

9 Beck R, Etzion R, Sandler S, Kohn L. Evaluation of therapid Patho Dx Latex Strep Grouping Kit. J ClinMicrobiol 1990;28:398-9.

10 Coombs RRA. Harnessing the red cell for immunoassays.Med Lab Sci 1987;44:66-72.

674


http://jcp.bmj.com/



11 Coombs RRA, Scott ML, Cranage MP. Assays using redcell-labelled antibodies. J Immunol Methods 1987;101:1-14.

12 Cohen J. A coefficient of agreement for nominal scales.Educational and Psychological Measurement 1960;20:37-46.

13 Maxted WR. Preparation of streptococcal extracts forLancefield grouping. Lancet 1948;ii:255-6.

14 Freimer EH, Raeder R, Feinstein A, Herbert J, Gurner BW,Coombs RRA. Detection of protein A-like substances onhaemolytic streptococci prior to use in mixed reversepassive antiglobulin haemagglutination (MRPAH). J

Immunol Methods 1979;31 :219-29.15 MacGowan A, Marshall R, Reeves D. False-positive reac-

tions with Listeria monocytogenes using a commercial kitfor Lancefield grouping of beta-haemolytic streptococci.Eur J Clin Microbiol Infect Dis 1988;7:208-10.

16 Bixler-Forell EW, Martin WJ, Moody MD. Clinical evalua-tion of the improved Streptex method for groupingstreptococci. Diagn Microbiol Infect Dis 1984;2:113-8.

17 Colman G. Streptococcus and Lactobacillus. In: ParkerMT, Duerden BI, eds. Principles of bacteriology, virologyand immunity. Volume 2. London: Edward Arnold, 1990:119-59.

Eponyms in pathology . . .

LEISHMAN, Sir William Boog (1865-1926) was a British microbiologist. Born inGlasgow, he studied medicine, graduating in1886. He joined the RAMC in 1887 and after abrief period in India, he returned to the AMSin Netley, succeeding Sir Almroth Wright asprofessor of pathology in 1900. After the FirstWorld War he became director of pathology atthe war office and eventually director general ofthe army medical services in 1923. In 1903 hedescribed the causal organism of kala azar,Leishmania donovani, as well as the intracellularnon-flagellate forms of the parasite now calledLeishman-Donovan bodies. He was knightedin 1909, elected an FRS in 1910, and rose to therank of lieutenant-general in 1923.

LANGHANS, Theodor (1839-1915) was aGerman pathologist born in Usingen. Hetrained in Wurzburg, central Germany,qualifying in 1864, and served as professor ofpathological anatomy at Giessen (1868) andlater in Berne (1872). He described the charac-teristic multinucleate giant cells of tuberculouslesions in 1867.

675


http://jcp.bmj.com/


doi: 10.1136/jcp.44.8.670 1991 44: 670-675J Clin Pathol

M Farrington, S Winters, D Rubenstein, et al. haemagglutination.grouped by reverse passive Streptococci from primary isolation plates

http://jcp.bmj.com/content/44/8/670Updated information and services can be found at:

These include:

serviceEmail alerting

the box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://jcp.bmj.com/content/44/8/670

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/

http://jcp.bmj.com/


Download - Streptococci from primary isolation plates grouped by reverse passive haemagglutination

Top Related