Vitamin A supplementation at birth might prime the response to subsequent vitamin A supplements in girls. Three year follow-up of a randomized trial

Within a randomised trial of neonatal vitamin A supplementation (VAS) in Guinea-Bissau, neonatal VAS did not affect overall infant mortality. We conducted a post-hoc analysis to test the hypothesis that neonatal VAS primes the response to subsequent vitamin A. All trial children were offered VAS after follow-up ended at 1 year of age (FU-VAS). We compared mortality between 1 and 3 years of age according to initial randomization to neonatal VAS or placebo in Cox-regression models; we expected that children randomized to neonatal VAS compared with those randomized to placebo would have lower mortality after reception of FU-VAS. Of 4345 infants enrolled in the original trial, 3646 lived in the study area at 1 year of age and 2958 received FU-VAS. Between 1 and 3 years of age, 112 children died. After FU-VAS, neonatal VAS was associated with lower mortality than placebo: Mortality Rate Ratio (MRR) = 0.54 (95%CI: 0.31-0.94). The effect was more pronounced in girls (MRR = 0.37 (0.16-0.89)) than boys (MRR = 0.73 (0.35-1.51)). The beneficial effect of neonatal VAS may have been particularly strong for girls who received both VAS in a campaign and FU-VAS (MRR = 0.15 (0.03-0.67)). Among children who had not received FU-VAS, mortality in the second and third year of life did not differ according to reception of neonatal VAS or placebo. Hence, in the second and third year of life the effect of neonatal VAS versus placebo was different in girls who had or had not received FU-VAS (p for homogeneity = 0.01). The present results suggest that neonatal VAS primes the response in girls such that they get a beneficial effect after a subsequent dose of VAS. Clinicaltrials.gov NCT00168597.