BackgroundMalaria antibody responses measured at delivery have been associated with protection from maternal anaemia and low birth weight deliveries. Whether malarial antibodies present in the first half of pregnancy may protect from these or other poor birth outcomes is unclear. To determine whether malaria antibodies in the first half of pregnancy predict pregnancy outcomes, antibodies were measured to a range of merozoite antigens and to antigens expressed on the surface of parasitized red blood cells (pRBCs) in plasma samples collected at 14–20 weeks of gestation from Malawian women. The latter antibodies were measured as total IgG to pRBCs, and antibodies promoting opsonic phagocytosis of pRBCs. Associations between antibodies and maternal haemoglobin in late pregnancy or newborn size were investigated, after adjusting for potential covariates.ResultsAntibodies to pRBC surface antigens were associated with higher haemoglobin concentration at 36 weeks. Total IgG to pRBCs was associated with 0.4 g/l [(95% confidence interval (0.04, 0.8)] increase in haemoglobin, and opsonizing antibody with 0.5 (0.05, 0.9) increase in haemoglobin for each 10% increase in antibody. These antibodies were not associated with birthweight, placental malaria, or newborn anthropometrics. Antibodies to merozoite antigens and non-placental-binding IEs were not associated with decreased risk of any of these outcomes. In some instances, they were negatively associated with outcomes of interest.ConclusionAntibodies to placental-binding infected erythrocytes may be associated with higher haemoglobin levels in pregnancy, whereas antibodies to other malaria antigens may instead be markers of malaria exposure. Trial registration clinicaltrials.gov NCT01239693. Registered Nov 10, 2010.