MALARIA DURING THE 1991-1992 NORTH-EAST MONSOON SEASON IN A VILLAGE IN THE INTERMEDIATE RAINFALL ZONE

A longitudinal epidemiological study of malaria was performed at Nikawehera, an established farming village in the intermediate rainfall zone of Sri Lanka. Observations were made over a period of eight months from October 1991 that included the north-east (NE) monsoon season and the earlypart of thebouth-west (SW) monsoon season. The incidence of malaria and clinical indices related to malaria at Nikawehera are reported here. Compared 'with obsemtions made at Nikawehera in the previous year, the present results suggest that malaria transmission is variable and that suitable intervention e.g. through effective insecticide spriying or malaria education, can reduce malaria incidence in such locations.

Malaria is a serious prob1,em affecting the health and economy of the country. Environmental factors, particularly rainfall, determine to a large extent the. occurrence of malaria in different parts of Sri ~anka.' Rainfall is unevenly distributed in the country. The wet zone receives rainfall (>2500 mm per year) throughout the year with two specific monsoonal periods of heavy rainfall viz. the north-east(NE) monsoon in October-January and the south-west (SW) monsoon in May-July. The dry zone gets maximum rainfall (< 2500 mm per year) during the NE monsoon and little or no rainfall during the rest of the year. An intermediate zone with mixed properties may be demarcated between the dry and wet zones (Fig. 1). It is reported that the intermediate zone is more amenable throughout the year to pool formation that is suitable for the breeding of Anopheles mosquitoes.' The ratio of Plasmodium falcipamm to R v i v a infection detected by blood filming ia Sri Lanka has increased in recent times, from 1:33 in 1980 to 1:4 in 1991.2 An understanding of malaria transmission in different regions of Sri Lanka is necessary for developing successful control and eradication measures including the possible use of a malaria ~a c c i n e .~ Recent information on malaria transmission in Sri Lanka shows that there' is significant regional A study was initiated in October 90 at Nikawehera, a farming village located in the intermediate rainfall zone, involving clinical and parasitological examinations as well as measurement of antibody levels at approximately three monthly intervals? The clinical and parasitological data obtained in the period October 91 to May 92 are reported here.
Flgluv 1: Map of Sd h k n with the study site, Nikawebcra, Idenlllted

METHODS AND MATERIALS
Nikawehera, a village in the Matale district located in the intermediate zone but close to the dry zone (Fig.l), was chosen for the study. Nikawehera is predominantly composed of farming families that have been resident for many generations at the site. Records from Dambulla and Galewela district hospitals, located 10 and 5 km respectively from Nikawehera, show a high malaria incidence in the area (Fig.2). A sample of 147 persons composed of 34% children aged 7-15 years and 66% adults were selected for the study in Oct 19907 Of this initial population, 76 (children 36%), 82 (children 41%) and 82 (children 40%) were present in the October 91, January 92 and May 92 surveys respectively. During the period described in this investigation the age range of the same selected group of children is taken as 8-16 years. Informed consent was obtained from each person and from parents in the case of children, for the study. Houses in Nikawehera were sprayed with malathion by the Anti-Malaria Campaign, Ministry of Health during the period of study in October 91, January 92 and March 92, following standard anti-malaria spraying procedures.
Medical data collected at Nikawehera included characteristics of previous fever attacks and accompanying symptoms as well as the time elapsed since the last treatment with antimalarial drugs. Other parameters recorded were anaemia (clinically assessed by conjunctival pallor), hepatomegaly and splenomegaly (measured by conventional palpation with fingers). After the physical examination, blood samples were withdrawn under aseptic conditiori by venepuncture and thin and thick blood films were made and examined for malaria parasites after staining with Giemsa. The number of parasitised cells in fields containing 200 white blood cells (WBC) were determined and parasite densities calculated assuming an average of 8000 WBC perpl of blood. Serum samples were used to detect antibodies against specific epitopes by immunoradiometric assays (submitted for publication).

RESULTS
Questioning during each of the surveys revealed that about 50% of persons presenting for clinical examinations experienced one or more malaria attacks during the period October 91 -May 92. The prevalence of anaemia was 61%, 56% and 28% in October 91, January 92 and May 92 respectively. The incidence of splenomegaly and hepatomegaly in the study population are shown in Figures 3 and 4 respectively. The proportion of children among those with palpable spleens, were 9/16,10/12 and 7/12 in October 91, January 92 and May 92 respectively. Children had higher spleen enlargement rates (p < 0.05 by chi square analysis) than adults in October 91 and January 92 (33% and 29% in children compared with 14% and 4% for the adults in October 91 and January 92 respectively). A similar comparison (21% children vs 10% adults) in May 92 showed that the difference was not statistically significant. 22%, 35% and 9% children had palpable liver in October 91, January 92 and May 92 respectively. Corresponding values for the incidence of anaemia in children were 52%, 50% and 18% respectively. Only the liver index showed a peak in January 92 in the study population. Table 1 gives the parasite prevalence rate and Table 2 the mean parasite densities in the study population. The total parasite prevalence rate was 17.1% in October 91, 12.2% in January 92 and 19.54 in May 92. There were no statistically significant differences between the parasite prevalence rates in adults and children except in May 92 when children had a higher prevalence of I? v i v a (p < 0.05 by chi square analysis). The parasite densities in adults and children were not significantly different at any time.

DISCUSSION
Results from the Nikawehera study obtained over the preceding twelve month period (October 90 -September 91), showed that malaria incidence and the clinical indices were related to entomological transmission rates of malaria? They indicated a tendency towards a peak of clinical symptoms and malaria incidence in January 91 at the end of NE monsoon transmission season. Antibody levels (submitted) and entomological inoculation rates7 were also maximal in January 91. In contrast, the present results show that there was a lower malaria incidence in January 92 than in May 92. It can be speculated that this difference may be related to more effective spraying by the Anti- When compared to the preceeding N E monsoon when the proportions of palpable spleens were 24% in October 90,40% in January 91 and 27% in May 91 respectively and the proportions of ~e m i c individuals were 76%,71% and 14% in October 90, January 91 and May 91 respectively? the present observations show lower spleen rates and overall fewer anaemic persons. This may partly be due to lower malaria transmission rates in the 91 -92 NE monsoon season at Nikawehera and partly to our observations that persons in the study group having been educated on malaria for several months, took more rapid treatment after the onset of symptoms of malaria. Apart from signiticantly higher spleen rates in children in October 91 and January 92 and the higher Rvivax incidence among children in May 92, our results did not show other evidence for differences in clinical immunity to malaria between 8-16 y olds and adults. This may be partly due to the small sample size in our study, since studies in highly malaria endemic areas of Africa show that there is signif~cantly greater immunity in adults? However, we have observed that the population in Sri Lanka, in contrast to Africa, have more rapid access to medical treatment? We have suggested that this difference may result in a slower rate of acquisition of clinical immunity with age in Sri Lanka, which may in turn reduce immunity differences between 8-16 y old children and adults 9. Younger children (< 8y), who may have been significantly more susceptible than adults to malaria, were not included in the study at Nikawehera fof logistical reasons.
During the 91 -92 NE monsoon season, Nikawehera had characteristics of a location mesoendemic for malaria (spleen and parasite rates typically between 11 -50%). In contrast, our studies showed that in Weheragala in the dry zone6 and in Nikawehera during the 90 -91 NE monsoon season? malaria transmission had characteristics of hyperendemicity (spleen rates in children > 50%). The results therefore suggest that malaria transmission in Nikawehera is variable from year to year. This may be due to factors such as rainfall, a better understanding of malaria among the population and the efficacy of spraying insecticides at Nikawehera.