Submicroscopic malaria in pregnancy and associated adverse pregnancy events: A case-cohort study of 4,352 women on the Thailand-Myanmar border.

BACKGROUND: Malaria in pregnancy detected by microscopy is associated with maternal anaemia, reduced fetal growth, and preterm birth, but the effects of lower density (i.e., submicroscopic) malaria infections are poorly characterised. This analysis was undertaken to investigate associations between submicroscopic malaria at the first antenatal care (ANC) visit and these adverse pregnancy events on the Thailand-Myanmar border. METHODS: Blood samples taken from refugee and migrant pregnant women presenting for their first ANC visit were analysed retrospectively for malaria using ultrasensitive PCR (uPCR, limit of detection 22 parasites/mL). The relationships between submicroscopic malaria and subsequent microscopically detectable malaria, anaemia, birth weight, and preterm birth were evaluated using inverse probability weighting for stratified random sampling. RESULTS: First ANC visit samples from 4,352 asymptomatic women (median gestational age 16.5 weeks) attending between October 1st 2012 and December 31st 2015 were analysed. The weighted proportion of women with submicroscopic malaria infection was 4.6% (95% CI 3.9-5.6), comprising 59.8% (49.5-69.4) Plasmodium vivax, 6.5% (4.0-10.5) Plasmodium falciparum, 1.8% (0.9-3.6) mixed, and 31.9% (22.2-43.5) infections which could not be speciated. Submicroscopic parasitaemia at first ANC visit was associated with subsequent microscopically detected malaria (adjusted hazard ratio [HR] 12.9, 95% CI 8.8-18.8, p < 0.001) and lower birth weight (adjusted predicted mean difference -275 g, 95% CI -510 to -40, p = 0.022). There was no association with preterm birth. Submicroscopic P. falciparum mono-infection (adjusted HR 2.8, 95% CI 1.2-6.6, p = 0.023) and coinfection with P. falciparum and P. vivax (adjusted HR 10.3, 95% CI 2.6-40.4, p = 0.001) was associated with increased risk of maternal anaemia, but submicroscopic P. vivax mono-infection was not. That uPCR was conducted for only a part of the cohort due to cost constraints is a limitation. CONCLUSIONS: In low transmission settings, uPCR identifies substantially more malaria infections at antenatal screening than conventional diagnostic methods. On the Thailand-Myanmar border, submicroscopic malaria at first antenatal consultation was associated with higher risks of microscopically diagnosed malaria later in pregnancy, anaemia, and reduced birth weight.