BACKGROUNDIntermittent preventive treatment for malaria in pregnancy (IPTp) can improve birth outcomes, but whether it confers benefits to postnatal growth is unclear. We investigated the effect of IPTp on infant growth in Uganda and its pathways of effects using causal mediation analyses.METHODSWe analysed data from 633 infants born to mothers enrolled in a randomised trial of monthly IPTp with dihydroartemisinin-piperaquine (DP) vs. sulfadoxine-pyrimethamine (SP) (NCT02793622). Weight and length were measured from 0 to 12 months of age. Using generalised linear models, we estimated effects of DP vs. SP on gravidity-stratified mean length-for-age (LAZ) and weight-for-length Z-scores (WLZ). We investigated mediation by placental malaria, gestational weight change, maternal anaemia, maternal inflammation-related proteins, preterm birth, birth length, and birth weight. Mediation models adjusted for infant sex, gravidity, gestational age at enrolment, maternal age, maternal parasitaemia at enrolment, education, and wealth.FINDINGSSP increased mean LAZ by 0.18-0.28 Z from birth through age 4 months compared to DP, while DP increased mean WLZ by 0.11-0.28 Z from 2 to 8 months compared to SP among infants of multigravidae; at these ages, confidence intervals for mean differences excluded 0. We did not observe differences among primigravida. Mediators of SP included birth weight, birth length, maternal stem cell factor, and DNER. Mediators of DP included placental malaria and birth length, maternal IL-18, CDCP1, and CD6 at delivery.INTERPRETATIONIn high malaria transmission settings, this exploratory study suggests different IPTp regimens may influence infant growth among multigravidae, potentially through distinct pathways, in the exclusive breastfeeding period, when few other interventions are available.FUNDINGStanford Center for Innovation in Global Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bill & Melinda Gates Foundation.