Application of the passive haemagglutination test for malaria: the problem of false negatives

BRIEF COMMUNICATIONS Application of the passive haemagglutination test for malaria: the problem of false negatives A. VOLLER,1 J. H. E. T. MEUWISSEN,2 & T. GOOSEN 3 Abstract In a study in a population in northern Tanzania, almost all adults and schoolchildren with parasitaemia were positive in the PHA test, whereas only halfof the children under S years of age with parasitaemia were serologically positive. A second study on infants up to the age of 18 months confirmed that many young children who could be expected to have been exposed to malaria did not have PHA-detectable antibody. The malaria passive haemagglutination (PHA) test, formerly known as the indirect haemagglutina- tion (IHA) test, has proved to be a useful addition to the serologic tests used for this disease (2, 4). There have, however, been reports of this test being nega- tive in some young children in malaria endemic areas (3, 5). In addition, under experimental conditions monkeys have had negative PHA tests after recent infection with malaria, whereas they have had posi- tive immunofluorescent titres at the same time (1). It was considered important, from an epidemiolo- gical point of view, to determine the level at which serologically false negatives can occur in the field. Materials and methods Plasma samples, collected in capillary tubes, were transported in the frozen state and were tested by the PHA method of Meuwissen et al. (5) using P. falci- parum sensitized cells. The initial plasma dilution tested was 1: 40, and samples not reacting at this dilution were classified as negative. All samples were preabsorbed with sensitized cells to eliminate non- specific reactions. Thick blood films were made at the time of the plasma collections; these were examined 1 Department of Clinical Tropical Medicine, London School of Hygiene and Tropical Medicine, London, WC1, England, and Nuffield Institute of Comparative Medicine, Zoological Society of London, Regent's Park, London, NWI, England. 2 Institute of Medical Parasitology, University of Nij- megen, The Netherlands. 8 East African Institute for Malaria and Vector Borne Diseases, Amani, Tanzania. for the presence of malaria parasites after conven- tional Giemsa staining. Areas studied Pangani. This is a small township on the coast of northern Tanzania. There is intense malaria trans- mission in the areas surrounding the town where the samples were collected. Voluntary cooperation of the local inhabitants allowed a reasonably represen- tative cross-section of the population to be sampled. Tanga. A large town also on the coast of northern Tanzania, Tanga is situated a few miles north of Pangani. Clinic facilities are provided at Ngombiani and the residents in the area around Tanga, as well as those from the town, attend such clinics. Intense malaria transmission also occurs in the areas around Tanga. Samples were collected from: (a) Babies between 2 days and 1 month old. They were brought to the routine postnatal clinic. (b) Infants from I month to 18 months old. These were accompanying mothers to postnatal clinics. (c) Mothers of children in groups (a) and (b). (d) Pregnant women attending antenatal clinics. Results The results of the blood film examinations and the PHA test on the Pangani residents are shown in Table 1. About one fifth of those aged less than 2 years were serologically positive, and about half of those aged 2-9 years and almost all those over 10 years of age were PHA-positive. There was a high parasite rate in those up to 16 years of age, but thereafter only a fifth of the population had current parasitaemia. Only half of the parasite carriers in children under 5 years of age were serologically PHA-positive. Most of those over 5 years of age who were parasite positive were also serologically posi- tive. The results of the Tanga clinic surveys are shown in Tables 2 and 3. It can be seen that 6000 of the babies under 4 weeks of age had detectable antibody, 3314 662 BULL. WORLD HEALTH ORGAN., Vol. 51, 1974 BRIEF COMMUNICATIONS Table 1. Parasitological (blood film) and serological (PHA test) results of a cross- sectional survey of the population of Pangani, January 1974 Blood films PHA tests Blood film-positiveindividuals Age group number positive/ % number positive/ % number/PHA- % PHA- total examined positive total tested positive number tested pOSitive 0-11 months 9/18 50.0 4/18 22.0 3/9 33.3 12-23 months 2/9 22.3 2/13 15.0 1/2 50.0 2-4 years 22/53 42.3 30/58 51.7 11/20 55.0 5-9 years 44/116 37.9 63/118 53.5 34/42 80.9 10-16 years 49/125 42.6 100/122 82.0 41/43 95.3 17-30 years 13/85 15.3 83/93 89.2 10/12 83.3 >30 years 15/71 21.1 81/84 96.4 11/12 91.7 Table 2. Results of blood films and PHA tests on people from Tanga regularly attending clinics Blood films PHA test Group number positive/ % oitv number positive/ %pstv total examined positive total tested % positive babies less than 4 weeks of age 1/25 3.6 12/20 60.0 infants 1-18 months of age 4/73 5.5 14/75 18.7 mothers 3/82 3.7 66/81 81.5 pregnant women 7/51 13.7 43/54 79.6 Table 3. Results of PHA tests on very young children from Tanga attending clinics Age group Number positive/total tested positive > 4 weeks 20/32 62.5 1-3 months 5/20 25.0 4-6 months 2/24 8.3 7-9 months 5/18 27.8 10-18 months 2/13 15.4 and fewer (20 %) of the 1-1 8-month-old children had antibody. The mothers were usually positive (80%); the same percentage of pregnant women also had antibody. The detailed serologic positivity rates of the young children are given in Table 3, which shows the results on the children up to 18 months of age seen at the Tanga clinics. There was a high serologic positivity rate for those under 1 month of age, but it then fell and remained low for the subsequent 1 ½/2 years. Discussion These studies, which carry on from those of Meuwissen et al. (5) and of Lobel et al. (3) again show that the malaria PHA test is a convenient means of measuring the serologic profile of a popu- lation in an endemic area. The usual increase in serologic positivity with age can be seen. It should be noted, however, that the large increase in serologic positivity that occurs between the 5-9 and 10-16- year-old age groups precedes the drop in parasite 663 664 BRIEF COMMUNICATIONS rates, which does not occur until the 17-30-year-old group. Although virtually all the Pangani schoolchildren and adults with parasitaemia also had positive PHA tests, only half of those under 5 years of age with parasitaemia were serologically positive. The results from Pangani prompted the Tanga study, which concentrated on the younger age groups. These studies confirmed that many young children who could be expected to have been exposed to malaria did not have PHA-detectable malarial antibody. The high serologic positivity rates of their mothers support the contention that these people lived in highly malarious areas. The pregnant women were usually PHA-positive, as were the babies during the first month of life, after which a rapid decline in seropositivity rates occurred. This shows that the PHA test was measuring passively transferred maternal antibody of the IgG class. The subsequent drop indicates the disappearance of the maternally derived immunoglobulin. It is not known whether the negative phase of the PHA test in some young children is due to the complete absence of antibody or whether it is due to a qualitative differ- ence in the type of antibody produced. Meuwissen et al. (5) have discussed possible reasons for PHA negativity and have attempted to explain why, in some circumstances, the indirect fluorescent anti- body test can be positive when the PHA test is negative. In the present context the mechanism is less important than the fact that young children can frequently have negative malarial serology even when malaria parasites are demonstrable in the blood. This means that surveys based on serology alone could give a misleading impression in that they would underestimate the number of infected young children. This, once again, indicates that serology should be used to complement the parasitological data if an adequate epidemiological picture is to be obtained. ACKNOWLEDGEMENTS We are pleased to acknowledge the financial assistance of the World Health Organization and the Medical Research Council of Great Britain. The work could not have been carried out without the help of the Director, Dr P. Wegesa, and staff of the East African Institute for Malaria and Vector Borne Diseases, Amani, Tanzania. REFERENCES 1. BiDWELL, D. ET AL. Bulletin of the World Health Organization, 49: 313-316 (1973). 2. KAGAN, I. G. American journal of tropical medicine and hygiene, 21: 683-689 (1972). 3. LOBEL, H. 0. ET AL. Bulletin of the World Health Organization, 49: 485-492 (1973). 4. MEUWISEN, J. H. E. T. Er AL. Bulletin of the World Health Organization, 46: 771-782 (1972). 5. MEUWISSEN, J. H. E. T. ET AL. Bulletin of the World Health Organization, 50: 513-519 (1974).