Background/Objective: The effect of daily prenatal and postnatal vitamin supplementation on concentrations of breast milk nutrients is not well characterized in HIV-infected women. We examined the impact of vitamin supplementation during pregnancy and lactation on breast milk concentrations of retinol, carotenoids and tocopherols during the first year postpartum among 626 HIV-infected Tanzanian women. Subjects/Methods: We conducted a randomized, double-blind, placebo-controlled trial. Women were assigned to one of four daily oral supplements: vitamin A þ b-carotene (VA þ BC); multivitamins (MV; B, C and E); MV þ VA þ BC or placebo. Concentrations of breast milk nutrients were determined by high-performance liquid chromatography at birth and every 3 months thereafter. Results: Supplementation with VA þ BC increased concentrations of retinol, b-carotene and a-carotene at delivery by 4799, 1791 and 84 nmol l À1 , respectively, compared to no VA þ BC (all Po0.0001). MV supplementation did not increase concentrations of a-tocopherol or d-tocopherol at delivery but significantly decreased concentrations of breast milk g-tocopherol and retinol. Although concentrations of all nutrients decreased significantly by 3 months postpartum, retinol, a-carotene and b-carotene concentrations were significantly higher among those receiving VA þ BC at 3, 6 and 12 months compared to no VA þ BC. a-Tocopherol was significantly higher, while g-tocopherol concentrations were significantly lower, among women receiving MV compared to no MV at 3, 6 and 12 months postpartum. Conclusions: Sustained supplementation of HIV-infected breastfeeding mothers with MV could be a safe and effective intervention to improve vitamin E concentrations in breast milk. VA þ BC supplementation increases concentrations of breast milk retinol but it is not recommended in HIV-infected mothers due to the elevated risk of vertical transmission.