Public Health Nutrition

Research Article

Vitamin A and causes of maternal mortality: association and biological plausibility

Hala Faisela1 and Rüdiger Pittrofa2 c1

a1 PO Box 67, Abbasia, 11381 Cairo, Egypt

a2 Maternal and Child Epidemiology Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK


Objective To review the association between major causes of maternal mortality and vitamin A, trying to determine if these associations are causal in nature, and to highlight possible biological pathways that may explain vitamin A effects.

Design Literature review, observational studies and clinical trials. The strength of association was determined by applying Bradford Hill criteria of causality.

Results In a vitamin A deficient population, vitamin A is essential for adequate treatment of anaemia. While vitamin A does not seem to be capable of preventing uterine atony, obstetric or surgical trauma, which are important causes of haemorrhage, it might be capable of preventing or decreasing coagulopathy. Possible effects on the placenta as regards implantation, site and size are not clear. As regards pregnancy-related infections, vitamin A supplementation can improve wound healing by decreasing fibrosis and increasing transforming growth factor-β (TFG-β). It can increase resistance to infection by increasing mucosal integrity, increasing surface immunoglobulin A (sIgA) and enhancing adequate neutrophil function. If infection occurs, vitamin A can act as an immune enhancer, increasing the adequacy of natural killer (NK) cells and increasing antibody production. β-carotene in its provitamin form can act as an antioxidant by decreasing endothelial cell damage (the pathognomonic feature of pre-eclampsia) and promote the vasodilator effect of nitric oxide that might bring about a better outcome of toxaemia in pregnancy. It is unlikely that vitamin A or β-carotene has an effect on obstructed labour.

Conclusions Plausible biomedical pathways can only be constructed for obstetric haemorrhage, anaemia in pregnancy, hypertension in pregnancy and pregnancy-related infections. A 40% reduction in the maternal mortality ratio, as observed in Nepal, is unlikely to be solely explained through the aforementioned pathways.

(Received August 12 1999)

(Accepted January 07 2000)


c1 Corresponding author: Email