Young Kwang Chae and Jeong Hyun Yun query whether data from our randomised controlled trial of early high-dose vitamin A supplementation (50 000 I given with the three diphtheria-tetanus-pertussis [DPT] vaccinations at 6 10 and 14 weeks) in Gambian infants can contribute to the ongoing controversy about whether there might be harmful interactions between vitamin A supplementation and killed vaccines particularly in girls. Several notes of caution are necessary. The observations by Benn and colleagues that have raised the concerns relate to excess mortality. Our trial is greatly underpowered for this endpoint. There were two postneonatal deaths in the high-dose group (n=99) and none in the standard WHO dose group (n=98). One male infant died of pneumonia and septicaemia at 7 months and one female of unknown causes at 3 months. This overall post-neonatal infant mortality rate equates to 10 in 1000 which is exceptionally low for this region and reflects the high level of monitoring and clinical care offered to the study participants. This point reduces the usefulness of our data in contributing to the debate. In response to Chae and Yuns query as to whether the vitamin A might have exacerbated DTP-related side-effects we have reanalysed our data in an attempt to identify potentially life-threatening illnesses. In our original paper we restricted the analyses to clinic visits that elicited a treatment. Numerous non-critical conditions such as colds and skin rashes remained. We have now excluded these and repeated the analysis of potentially life-threatening events for the time window between the first DPT vaccination and vitamin A dose at 6 weeks and the first live vaccine (measles at 9 months). We analysed four diseases (malaria diarrhoea pneumonia and septicaemia) separately and combined. There were 73 events in the high-dose group and 79 events in the WHO group. Owing to small numbers of events there were wide confidence intervals and no differences were significant. The closest to significance (pneumonia) was if anything lower in the high-dose group. In summary our trial is under powered to contribute to this important debate but we will be happy to contribute the data to future meta-analyses. (full text)
Full Text via ClinicalKey