RANDOMIZED TRIAL TO ASSESS BENEFITS AND SAFETY OF VITAMIN-A SUPPLEMENTATION LINKED TO IMMUNIZATION IN EARLY INFANCY

Background The benefits and safety of vitamin A supplementation linked to immunisation in infancy need to be assessed before it can be widel y recommended. We assessed the safety and benefits of maternal postpar tum and infant vitamin A supplementation administered with each of the three diphtheria-tetanus-pertussis (DPT) and poliomyelitis immunisati ons and with a fourth dose with measles immunisation. Methods From Jan uary, 1995, we enrolled 9424 mother-infant pairs from Ghana, India, an d Peru in this randomised, double-blind, placebo-controlled trial. 471 6 mothers of infants in the vitamin A group received 200 000 IU vitami n A, and their infants were given 25 000 IU vitamin A with each of the first three doses of DPT/poliomyelitis immunisation at 6, 10, and 14 weeks. In the control group, 4708 mothers and their infants received p lacebo at the same times. At 9 months, with measles immunisation, infa nts in the vitamin A group were given a further dose of 25 000 IU and those in the control group received 100 000 IU vitamin A. Infants were followed up to age 12 months. The primary outcome measures were vitam in A status, signs of acute toxic effects, anthropometric indicators, and severe morbidity. Analysis was by intention to treat. Findings 393 3 (93%) of the eligible 4212 infants on vitamin A and 3938 (93%) of th e eligible 4227 controls received all four study doses. At the g-month follow-up, there was a small decrease in vitamin A deficiency in the vitamin A group compared with controls (serum retinol less than or equ al to 0.70 mu mol/L 101 [29.9%] vs 122 [37.1%; 95% CI of the differenc e -14.3% to -0.2%1]). This effect was no longer apparent at 9 and 12 m onths, There were no significant between-group differences in mortalit y throughout the study. The rate ratio to compare all deaths up to age 9 months in the two groups was 0.96 (95% CI 0.73 to 1.27), Fewer than 1% of the infants had bulging fontanelle. The intervention had no eff ect on anthropometric status, or on overall or severe morbidity. Inter pretation The trial confirmed the safety of the intervention, but show s no sustained benefits in terms of vitamin A status beyond age 6 mont hs or infant morbidity.