Zinc supplementation and stunted infants in Ethiopia: a randomised controlled trial.

BACKGROUND

Stunting is highly prevalent in Ethiopia and many other developing countries but the reason for it is poorly understood. Zinc is essential for growth but diets in such countries often do not contain zinc in sufficient quantity or of sufficient bioavailability. Thus zinc deficiency may play a major role in stunting. The aim of the study was to investigate whether the low rate of linear growth of apparently healthy breastfed infants in a rural village in Ethiopia could be improved by zinc supplementation.

METHODS

A randomised, double-blind, placebo-controlled trial was done on apparently healthy breastfed infants aged 6-12 months. 100 non-stunted (length-for-age, Z score < -2) were matched for age and sex with 100 randomly selected stunted (> -2) infants. Infants, both stunted and non stunted, were matched by sex, age (within 2 months) and recumbent length (within 3 cm) for random assignment, to receive a zinc supplement (10 mg zinc per day, as zinc sulphate) or placebo, 6 days a week for 6 months. Anthropometric measurements were taken monthly, data on illness and appetite were collected daily, and samples of serum and hair were taken at the end of the intervention for the analysis of zinc.

FINDINGS

The length of stunted infants increased significantly more (p<0.001) when supplemented with zinc (7.0 cm [SE 1.1]) than with placebo (2.8 cm [0.9]); and the effect was greater (p<0.01) than in non-stunted infants (6.6 [0.9] vs 5.0 [0.8] cm for the zinc and placebo groups respectively, p<0.01). Zinc supplementation also increased the weight of stunted children (1.73 [0.39] vs 0.95 [0.39] kg for the corresponding placebo group, p<0.001) and of non-stunted children (1.19 [0.39] vs 1.02 [0.32] kg for the corresponding placebo group, p<0.05). Zinc supplementation resulted in a markedly lower incidence of anorexia and morbidity from cough, diarrhoea, fever, and vomiting in the stunted children. The total number of these conditions per child was 1.56 and 1.11 in the stunted and non-stunted zinc supplemented children versus 3.38 and 1.64 in the stunted and non-stunted placebo-treated children, respectively. At the end of the intervention period, the concentrations of zinc in serum and hair of stunted infants, who had not been supplemented with zinc, were lower than the respective concentrations of zinc in serum and hair of their non-stunted counterparts.

INTERPRETATION

Combating zinc deficiency can increase the growth rate of stunted children to that of non-stunted infants in rural Ethiopia. This would appear to be due, at least in part, to reduction in morbidity from infection and increased appetite.