Effect of early neonatal vitamin A supplementation on mortality during infancy in Ghana (Neovita): a randomised, double-blind, placebo-controlled trial

Background Results of randomised controlled trials of newborn (age 1-3 days) vitamin A supplementation have been inconclusive. The WHO is coordinating three large randomised trials in Ghana, India, and Tanzania (Neovita trials). We present the findings of the Neovita trial in Ghana. Methods This study was a population-based, individually randomised, double-blind, placebo-controlled trial in the Brong Ahafo region of Ghana. The trial participants were infants aged at least 2 h, identified at home or facilities on the day of birth or in the next 2 days, able to feed orally, and likely to stay in the study area for at least 6 months. They were randomly assigned (ratio 1: 1) to receive either one oral dose of vitamin A (50 000 IU) or placebo immediately after recruitment. The research team and parents of the infants were masked to treatment assignment. Follow-up home visits were undertaken every 4 weeks, when data were recorded for deaths, facility use, and care seeking. The primary outcome was post-supplementation mortality to 6 months of age. Analysis was by intention to treat. Potential adverse events were recorded at 1 and 3 days after supplementation. This trial is registered with the Australian New Zealand Clinical Trials Registry (ANZCTR) CTRN12610000582055. Findings We assessed 26 414 livebirths for eligibility between Aug 16, 2010, and Nov 7, 2011. We recruited 22 955 newborn infants, with 11 474 randomly assigned to receive vitamin A and 11 481 to receive placebo. Loss to follow-up was low with vital status at 6 months of age reported for 22 698 (98.9%) infants. We recorded 278 post-supplementation deaths to 6 months of age in the vitamin A group (mortality risk 24.5 in 1000 supplemented infants) and 248 deaths in the placebo group (mortality risk 21.8 per 1000 supplemented infants), relative risk (RR) 1.12 (95% CI 0.95-1.33; p=0.183) and risk difference (RD) 2.66 (95% CI -1.25 to 6.57; p=0.18). Adverse events within 3 days of supplementation did not diff er by trial group. 122 infants died in the first 3 days after supplementation; 70 (0.6%) in the vitamin A and 52 (0.5%) in the placebo group (risk ratio [RR] 1.35, 95% CI 0.94-1.93, p=0.102). 53 infants were reported to have a bulging fontanelle; 32 (0.3%) in the vitamin A group and 21 (0.2%) in the placebo group (RR 1.53, 0.88-2.62, p=0.130). Interpretation The results of this trial do not support inclusion of newborn vitamin A supplementation as a child survival strategy in Ghana.