Malaria in pregnancy in rural Mozambique:: the role of parity, submicroscopic and multiple Plasmodium falciparum infections

BACKGROUND Falciparum malaria affects pregnant women, especially primigravidae, but before malaria control programmes targeted to them can be designed, a description of the frequency and parity pattern of the infection is needed. There is little information on the frequency and effect of submicroscopic malaria infection, as well as on multiplicity of Plasmodium falciparum genotypes in pregnancy. This study aimed to describe the prevalence of malaria parasitaemia and anaemia and their relation to parity and age in pregnant women, during two malaria transmission seasons in a rural area of southern Mozambique. It also tried to assess the frequency and effect on anaemia of submicroscopic and multiple falciparum infections. METHODs A total of 686 pregnant women were enrolled in three cross-sectional community-based surveys during different transmission seasons in rural southern Mozambique. In each survey a questionnaire was administered on previous parity history, the gestational age was assessed, the axillary temperature recorded and both haematocrit and malaria parasitaemia were determined. We used polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) analysis to determine submicroscopic and multiple P. falciparum infections in a subsample of women. FINDINGs A total of 156 women (23%) had microscopic parasitaemia, of which 144 (92%) were asexual forms of P. falciparum. The prevalence of clinical malaria was 18 of 534 (3%), that of anaemia, 382 of 649 (59%). In a multivariate analysis age but not parity was associated with an increased risk of microscopic parasitaemia. Anaemia was associated with microscopic P. falciparum parasitaemia. Both malaria parasitaemia and anaemia were more frequent during the rainy season. Although not statistically significant, submicroscopic infections tended to be more frequent among grand-multiparous pregnant women. Subpatent infections were not associated with increased anaemia. Multiplicity of infection was not associated with either parity, age or anaemia. Likewise, there was no correlation between F. falciparum density and multiplicity of infection. INTERPRETATION We did not observe a clear parity pattern of malaria and anaemia in our study. It is possible although unlikely that selection bias may have influenced these findings; but in which direction is unclear. The importance of locally based research before implementation of public health measures needs to be highlighted. According to our findings, a more cost-effective malaria control approach in this area would be targeting all pregnant women regardless of their parity. This would be also more feasible logistically as it would not rely on accurate ascertainment of parity, something that is not always easy in busy antenatal clinics.