An estimated 12% of South African children below five years of age are underweight, and 25% are stunted. Additionally, a high prevalence of vitamin and mineral deficiency exists in infants and young children, and surveys have found a 21% prevalence of anemia [2–4]. Despite this, a recent trial of multiple micronutrients in infants aged six to twelve months was unable to demonstrate an effect on length-for-age Z-scores, with supplemented children continuing to show a decline in Z-scores over time. The effect of prophylactic zinc and/or additional micronutrients on growth in this population remains to be fully defined.
Zinc supplementation is known to improve growth in certain populations of children. A meta-analysis of zinc supplementation trials found significant beneficial effects on height in children, with the largest benefit occurring in children with low height Z-scores at baseline. Notably, none of the three trials from African countries included in the meta-analysis of effect on height found an improvement in height velocity with zinc supplementation. This may be related to lesser degrees of stunting in the African populations compared to Asian populations where the majority of studies finding benefit were conducted. A more recent trial in Burkina Faso, conducted after the meta-analysis was conducted, found no benefit of zinc on height in infants, even in the subgroup of stunted infants. This raises the question of whether concurrent deficiencies of other micronutrients limited the response to zinc. The previous South African study of infants aged six to twelve months, which failed to find an effect on length-for-age Z-scores, did not examine effects in stunted children. Further investigation is therefore needed to assess the benefits of multiple micronutrient supplementation in African children, especially those at highest risk of growth faltering such as those living in rural areas and those at risk of repeated diarrheal illness.
Anemia in infants and children is multifactorial in origin. While iron deficiency is the most common cause, deficiencies in other nutrients that contribute to iron homeostasis - riboflavin, folic acid, vitamins B12, C and A - may also result in anemia. The high phytate content of weaning diets such as maize porridge, the staple food for rural South African children, may further impair absorption of iron and zinc. Infectious disease, especially chronic infection, is known to alter iron metabolism. This multifactorial nature of anemia may partly explain the incomplete response of anemia to iron supplementation in some settings[9, 10]. Some trials of multiple micronutrients, including the combination of iron and zinc, have found them more effective in reducing anemia prevalence than iron alone, [4, 9, 11, 12] while others have not found a beneficial effect of combined supplementation. Because of these discordant results, the utility of zinc or multiple micronutrients in preventing anemia or augmenting its treatment needs further investigation.
In HIV-infected children early viremia and subsequent disease progression are superimposed on the other factors that contribute to growth faltering and anemia in uninfected children. Poor longitudinal growth becomes apparent early in infancy. Unlike weight, height fails to respond to increased feeding. It is conceivable that infected children may show a suboptimal response to supplements due to altered metabolism of these nutrients and reduction in erythropoiesis  or the direct effect of HIV itself.
We analyzed data from a randomized, controlled, double-blinded community-based trial whose primary aims were to assess the benefit of prophylactic daily zinc or multiple micronutrient supplements, both combined with vitamin A, in comparison to vitamin A alone that was used as a control group, on: 1) the primary prevention of respiratory illness and diarrhea, and 2) the effect on growth in rural South African children. Supplements used in this trial had no overall effect on prevalent days or incidence of diarrhea compared to vitamin A that was used as a control group,  but they reduced the incidence of diarrhea in stunted children.
The main purpose of this analysis is to report on the second study aim - effect of supplementation on growth - and to report exploratory analyses of the effect of supplementation on anemia. More specifically, the second aim of this trial was to compare the effect on longitudinal length-for-age Z-scores of adding either zinc alone or in combination with multiple micronutrients to vitamin A supplements, and to describe how HIV, stunting and diarrheal morbidity affect the longitudinal growth response.