Summary of evidence

Publication Details

One systematic review (11) has evaluated the effects and safety of vitamin A supplementation in infants 6 months of age or less in low- and middle-income countries with regard to prevention of morbidity and mortality. It included a subgroup analysis by age at initiation of supplementation (post-neonatal period of 1–6 months of age). The review showed no significant effect of vitamin A supplementation in infants 6 months of age or less on the risk of mortality in the first year of life, but it showed an increase in the risk of developing bulging fontanelles. Analysis of data from three trials in which vitamin A supplementation was initiated between 1 and 6 months of age showed no effect on all-cause mortality as compared with controls (risk ratio (RR) 1.05; 95% confidence interval (CI) 0.84–1.32). The remaining analyses were conducted in all infants 0–6 months of age. There appears to be no significant effect of vitamin A supplementation on mortality or morbidity due to diarrhoea or acute respiratory infections in the first year of life. There is no effect of vitamin A supplementation on all-cause mortality, when given as a cumulative dose of either 50 000 IU or less or more than 50 000 IU, regardless of the status of maternal postpartum vitamin A supplementation. The 10 trials (six of which provided supplements in the post-neonatal period) that provided data on bulging fontanelles following any (first, second or third) dose of vitamin A showed an increased risk of this side-effect (RR 1.55; 95% CI 1.05–2.28). Few trials reported data on other adverse effects, such as vomiting, irritability, diarrhoea and fever, none of which were significant.

WHO performed additional meta-analyses that included only those studies in which infants 1–5 months of age were given supplements (Annex 1). There was no significant effect of vitamin A supplementation on mortality in the first year of life related to diarrhoea (two trials: RR 1.05; 95% CI 0.76–1.46) or respiratory infections (two trials: RR 1.20; 95% CI 0.85–1.68). Additionally, there was no significant effect of vitamin A supplementation on morbidity in the first year of life related to diarrhoea (two trials: RR 0.99; 95% CI 0.94–1.04) or respiratory infections (one trial: RR 1.06; 95% CI 0.96–1.16). There was a significant increase in the occurrence of bulging fontanelles after any dose (first, second or third) of vitamin A (six trials: RR 2.53; 95% CI 1.27–5.03), and one trial reported a significant decrease in vomiting (RR 0.31; 95% CI 0.17–0.58). There was no effect on fever, irritability or diarrhoea as side-effects of the intervention.

The overall quality of the available evidence with regard to mortality during infancy and the side-effect of bulging fontanelles was moderate and for the other outcomes it was low (Annex 2).

The effects of vitamin A supplementation on seroconversion rates to the three poliovirus types (types 1, 2 and 3) was also recently reviewed (12). A meta-analysis of three trials indicated no difference in response to the polio vaccine (specific antibody titres or seroconversion rates) when vitamin A supplements or placebo were given between 1 and 5 months of age concurrently with the oral polio vaccine (OPV). Limited data indicate that vitamin A supplementation does not affect the tetanus or pertussis vaccine response, but may increase the antibody response to diphtheria vaccination. This review also addressed the effect of co-administering vitamin A with vaccines on mortality and other adverse events. A meta-analysis of five trials revealed no significant effect of receiving vitamin A supplements with the DTP vaccine on subsequent mortality (five trials: odds ratio 1.05; 95% CI 0.82–1.36) (13).