Field application to malaria studies of the passive haemagglutination (PHA) test with lyophilized cells

MALARIA Field application to malaria studies of the passive haemagglutination (PHA) test with lyophilized cells R. CORNILLE BROGGER 1 & H. M. MATHEWS 2 Lyophilized cells for the passive (formerly indirect). haemagglutination test ior malaria were tested in a smallfield trial. A comparison was made with results obtained with samples from the same persons by another method in another laboratory (Center for Disease Control, Atlanta, GA, USA). The use of lyophilized cells of standardized reactivity clearly adds to the simplicity and rapidity oJ the test, andjustifies its recommendation for use infield studies. Seroimmunological investigations of malaria in the field have been impaired by the difficulty of obtaining standardized, stable antigen that would withstand adverse environmental conditions such as high temperatures. In the WHO Malaria Research Project in Garki district, Kano state, Nigeria (1970-75) the longitudinal seroimmunological studies on approximately 3 000 persons included 6 tests. Five of these tests were performed on plasma specimens in the project laboratory in Kano, namely determinations of serum IgG and IgM, the indirect fluorescent antibody (IFA) test for Plasmodiumfalci- parum and P. malariae, and the precipitin test for P. falciparum. The sixth test, the passive haemagglu- tination (PHA) test, formerly known as the indirect haemagglutination (IHA) test, was performed at the Center for Disease Control (CDC), Atlanta, GA, USA, because of difficulties experienced early in the project in the preparation and preservation of erythrocytes sensitized with malarial antigen. Eluates from blood collected on filter paper, and human 0 erythrocytes sensitized with P. knowlesi (1971) and later with P. falciparum (1971-74), were used in the PHA test, which was performed according to the procedure described in the project immunology baseline report (unpublished). Recently, lyophilized sheep erythrocytes sensitized with P. falciparum have been prepared in the Nether- lands and tested in several laboratories and in a field I Formerly Immunologist, WHO Malaria Research Project IR-0172, Kano, Nigeria. Present address: 17 chemin de Pont-Ceard, 1290 Versoix, Switzerland. 2Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control, Atlanta, GA 30333, USA. trial in Surinam (1). These preparations seemed to open new possibilities for the use of the PHA test in field studies. A small blind study was therefore carried out at Kano in October 1974 to investigate the feasibility of using the PHA test under field conditions, with lyophilized reagents. The results were compared with those obtained with samples of the same blood at the CDC. MATERIALS AND METHODS An a priori selection of plasma specimens was made, based on the results of the test at the CDC. The sample was representative of the complete range of reactivity observed (titres < 16 to >32 768 in a two-fold dilution system), with 20 specimens from each titre level whenever possible (Table 1). The sample comprised 248 specimens from an immuno- logical survey carried out in October 1973, of which 12 were found to be insufficient for retesting; dupli- cate tests were therefore performed with 236 speci- mens. The retesting was done in the order of the serial numbers of the specimens, without knowledge of the CDC results. Lyophilized reagents, including sheep erythrocytes sensitized with P. falciparum antigen prepared from infected Aotus monkey erythrocytes, sheep erythro- cytes sensitized with noninfected Aotus monkey red blood cells for absorption and control, and normal rabbit serum, were provided by Dr J. H. E. T. Meuwissen, transferred to Kano at ambiant temper- ature and stored at 4°C for 2-3 weeks until used. The same conditions were applied to a suspension of stabilized human 0 erythrocytes sensitized with P. falciparum provided by Dr I. Kagan; these were 3319 - 39 BULL. WORLD HEALTH ORGAN., Vol. 52, 1975 R. C. BROGGER & H. M. MATHEWS Table 1. Distribution of the 236 persons studied according to their PHA titres obtained at the two laboratories Kano titre CDC titre <40 40 80 160 320 640 1 280 2560 5120 >5120 >32 768 2 32 768 1 1 1 16384 1 1 2 8 2 4 8192 3 1 2 3 11 4096 1 1 2 2 5 3 6 2048 1 2 2 5 3 2 5 1 024 2 1 5 4 4 4 512 1 4 7 4 1 2 256 2 2 5 4 1 1 2 128 3 1 3 3 7 1 2 64 6 1 1 2 1 1 4 2 1 32 6 5 1 1 4 2 16 13 1 1 2 1 1 1 <16 8 1 5 3 2 used to study possible differences in haemagglutina- tion pattern. The test procedures followed were exactly as described previously (1, 2) and will not be described here in detail. However, the differences between the PHA tests performed at CDC and Kano may be summarized as follows: (1) The P. falciparum antigen extract was pre- pared by different procedures. (2) The red blood cells used for sensitization at CDC were human 0 erythrocytes kept in suspen- sion, whereas the lyophilized reagent was prepared with sheep erythrocytes. (3) The anti-malaria material used for dilution at CDC was an eluate from blood dried on filter paper, whereas that used at Kano was plasma kept at -200C. (4) Absorption of nonspecific agglutinin was car- ried out at Kano, but not at CDC. (5) In both cases the dilution system was two-fold, but started from 1: 16 (eluate) and 1:40 (absorbed plasma), respectively. (6) The final dilution of the reacting mixture was modified by the volume of cell suspension added; this was by a factor of 50% at CDC and by 100% at Kano. (7) The temperature and time of incubation were similar, but the observed sedimentation patterns were different: negative sedimentation pattern ap- peared as a closed button with the sheep erythro- cytes, and a ring (doughnut) of nonagglutinated cells with human 0 cells. RESULTS Table 1 shows the distribution of the 236 results obtained in the two laboratories. Titres are expressed as the reciprocal of the highest dilution that gave a positive reaction. At CDC two-fold dilutions from 1 : 16 to 1: 32 768 were used. The results are also recorded for titres <16 and >32 768. In Kano dilu- tions of plasma were tested from 1: 40 to 1: 5 120, and results were also recorded for <40 (negative) and >5 120. The correlation coefficient, r = 0.711, was computed by means of the logaritlhimic scales implicit in Table 1, namely for the CDC titres from 0 to 13 and for the Kano titres from 0 to 9. The value of r2 = 0.5 indicates that approxi- mately half of the total variation of one test could be 40 PASSIVE HAEMAGGLUTINATION TEST oo C) C" 0oa u c S I1 5 10 0 5 1092 (CDC titre/8) 1 + 1092 (Kano titre/40) Fig. 1. The mean Kano log titre as a function of the CDC log titre. and the mean CDC log titre as a function of the Kano log titre. predicted by the result of the other test. The relation- ship between the two results is nearly linear, except at the extreme ends of the scale as shown in Fig. 1. Extreme results in both tests tend to revert to the mean in the other test; this is to be expected with open-ended extreme classes. DISCUSSION This investigation aimed to extend studies on the use of the PHA test in epidemiological field surveys in terms of the stability and ease of storage of the reagent, the simplicity of the test procedure, and the interpretation of results by comparison with those obtained previously under carefully standardized conditions. No apparent damage was caused to the reagents by transport and storage, and they proved very satisfactory in use. No information was obtained on the long-term stability of the reagents, nor on the reproducibility of the test. The interpretation of sedimentation patterns presented difficulties, charac- teristic of all investigations using the PHA test, in terms of the definition of positive and negative reactions. The Kano titres were obtained by follow- ing strictly the test procedure recommended by Meuwissen (1) and by keeping to the definition of a negative reaction as that of the dilution showing a closed button of cells similar to that observed for the controls. Deliberate experimentally introduced variations from the recommended test procedure with regard to time and temperature of incubation resulted in widely different results for some but not all of the specimens (not reported in detail here). No other information was obtained on the reproduci- bility of the results. Only one plasma sample tested was found to have nonspecific agglutinating factors that were not removed by the preliminary absorption by non-malaria-sensitized sheep erythrocytes. Patterns of erythrocyte sedimentation similar to those observed in the CDC laboratories for negative reactions, i.e., a ring (doughnut) of nonagglutinated cells, were also observed in Kano in a few tests performed with the CDC human 0 erythrocyte suspension, but they were not seen in the procedure using the lyophilized reagents. The difference in sedimentation pattern could be explained by the differences in the types of cell used, their preserva- tion, and other test differences such as in the concen- tration of the cell suspensions and in the final volume. Other differences, such as those of the antimalaria material used in the tests and the final dilution of the reaction mixture also contributed to the differences in the absolute titre for the same persons. Considering all the above differences, to which could be added individual and local differences (technical staff and equipment), the correlation be- tween the two sets of results is remarkably good, and the performance of the PHA test in Kano was very a IR 11 ax m IC C4 41' 42 R. C. BROGGER & H. M. MATHEWS *encouraging for its future use in seroimmunological studies under field conditions. The use of stan- dardized, stable lyophilized reagents gives the PHA test an incontestable advantage over other malaria tests such as the IFA and precipitin tests. Moreover, the PHA test requires simpler equipment and less technical skill and is less time-consuming than the IFA test. Further evaluation of the respective advan- tages of the different tests should be made from a comparison of the results of these tests with para- sitological findings. Finally, it can be concluded that blood collected and dried on filter paper for use in the PHA test gives results quantitatively comparable with those obtained with plasma specimens. This can be an additional advantage for field epidemiological studies. ACKNOWLEDGEMENTS Thanks are due to Dr J. H. E. T. Meuwissen, University of Nijmegen, Netherlands, who kindly provided the lyophilized cells, to Dr I. Kagan, who provided reagents and valuable comments on the study, and to Dr L. Molineaux for stimulating discussions. RESUME APPLICATION DE L'EPREUVE D HE1MAGGLUTINATION PASSIVE AVEC ERYTHROCYTES LYOPHILISES DANS LES ETUDES PALUDOLOGIQUES SUR LE TERRAIN Parmi des echantillons de plasma recueillis dans le cadre du projet OMS de recherche antipaludique a Kano, Nigeria du Nord, les auteurs ont choisi une serie repre- sentant la gamme complete des r6activit6s d'apres les titres observes dans les epreuves de routine par hemagglu- tination indirecte effectuees au Center for Disease Control d'Atlanta, Etats-Unis d'Amerique. Les auteurs ont alors repete ces 6preuves sur le terrain en utilisant des erythro- cytes lyophilis6s selon la methode recommandee pour leur emploi. L'emploi des erythrocytes Iyophifises a sim- plifie grandement les problemes de transport et de stockage. La correlation entre les deux series de resultats est bonne (r2=O,5), ce qui atteste la valeur de la methode pour application sur le terrain. REFERENCES 1. MEUWISSEN, J. H. E. T. The indirect haemagglutination test for malaria and its application to epidemiological surveillance. Bulletin of the World Health Organization, 50: 277-286 (1974). 2. MATHEWS, H. M. ET AL. The indirect haemagglutination test for malaria. Evaluation of antigens prepared from Plasmodiumfalciparum and Plasmodium vivax. Americanjournalof tropical medicine and hygiene, 24: 417-422(1975).