The Benefits and Concerns Related to Vitamin A Supplementation

There has been considerable interest in the role that vitamin A plays in child health. Discovered in 1913 by E. V. McCollum, vitamin A is essential for the production of rhodopsin, the light-sensitive chemical in the retina, and its deficiency results in night blindness and more-severe forms of xerophthalmia, including corneal ulcers and blindness. A role for vitamin A deficiency in the etiology of infections was proposed in the period immediately after its discovery [1], but research to identify whether a causal relationship existed was not pursued. While conducting an observational study to understand the epidemiological aspects of xerophthalmia in Indonesia in the early 1980s, Sommer et al. found that children who had xerophthalmia were more likely to die than were other children [2]. In their first large community trial in Aceh, Indonesia, Sommer et al. found that children >6 months old who received vitamin A every 6 months were significantly less likely to die than were those who did not receive vitamin A [3]. This landmark study led to substantial investments in research and, eventually, field programs related to vitamin A supplementation in developing countries. These findings were confirmed in several other randomized, placebo-controlled trials in diverse populations that were presumably largely free of HIV infection [4], and they have also been extended to children who are HIV infected [5, 6]. Periodic vitamin A supplementation, costing 0.02 US $/dose, to children >6 months old is being implemented in >70 countries and is considered by many international agencies to be one of the most-effective public health interventions ever undertaken Two other main issues pertaining to vitamin A supplementation are important to consider—namely, the effect that postpartum maternal, neonatal, and early infancy supplementation has on child health and the effect that maternal supplementation has on maternal health. Findings about the role that vitamin A supplementation plays during the first 6 months of life have been less consistent than those documented for supplementation later in life. A large dose of vitamin A given only to neonates reduced mortality in infants in Indonesia who were of normal birth weight [7]. A similar trial in India found beneficial effects of supplementation in infants with low birth weight [8]. Results from a large, placebo-controlled trial performed in Ghana, India, and Peru found that providing a large dose of vitamin A to mothers at birth and 3 doses to children at the times of their routine vaccination visits had no effect on infant mortality [9] In this issue of the Journal of Infectious Diseases Humphrey et al. [10] report findings of a placebo-controlled trial in Zimbabwe that shed more light on this question. They examined the efficacy of a single large dose of vitamin A given to women (400,000 IU) early during the postpartum period and/or to infants (50,000 IU). The research team is to be commended for performing such a large and important study under difficult field circumstances. HIV testing was offered to all women after enrollment, and the lack of a significant effect of supplementation on mortality in infants born to HIV-negative women was reported previously [11]. In the article published in this issue of the Journal they present findings on mother-to-child transmission and child mortality in infants of HIV-positive mothers. Given the difficulty in ascertaining the HIV status of all deceased patients, it is customary in this field to examine the effect that interventions have on the combined risk of HIV infection or death (i.e., HIV-free survival), which is an important end point for understanding the public-health implications of the findings. Humphrey et al. found that providing the vitamin A supplement to mothers or infants resulted in an increased risk of infant HIV infection or death, although the effect of providing the supplement to both mother and infant was apparently not different from the effect of providing a placebo. The authors interpret these findings as possibly related to chance, because it would have been more consistent to note a higher risk in infants in all groups in which the vitamin A supplement was given. Little is known about the complex relationships between vitamin A supplementation and the mix of possible adverse and beneficial effects on immunological and virological parameters at the systemic and mucosal levels [12], and it would be difficult to ignore the increased risks that Humphrey et al. observed when vitamin A was given only to either the mothers or the infants. The findings on mortality are equally noteworthy: vitamin A supplementation had no effect on mortality in infants who were positive by polymerase chain reaction (PCR) for HIV at baseline; in infants who were PCR negative at baseline but PCR positive at 6 weeks, neonatal supplementation decreased mortality by 28%, but maternal supplementation had no effect; in the majority of infants—namely, those who were PCR negative at 6 weeks—all 3 vitamin A regimens were significantly associated with ∼2-fold higher mortality Concern related to vitamin A supplementation in patients with HIV infection has been previously noted. Vitamin A has been associated with increased replication of HIV-1 in some, but not all, in vitro studies [13–15]. In HIV-1–negative Kenyan men with genital ulcers, low plasma vitamin A levels were associated with a decreased risk of HIV-1 acquisition [16]. In a prospective study of HIV-infected men in the United States, a U-shaped relationship was observed between dietary vitamin A intake and risks of progression to AIDS [17] and mortality [18], suggesting that men with higher or lower levels of intake were at greater risk than were men who consumed moderate amounts of vitamin A. In a trial from Tanzania, significantly more HIV-1 infections occurred in children of mothers who received daily supplements of preformed vitamin A (5000 IU) and β-carotene (30 mg) during antenatal and breast-feeding periods than in other children [19]. It is plausible that the adverse effects observed in the latter trial may be due to the preformed vitamin A component of the supplement or to β-carotene, which, under highly oxidative conditions, may result in eccentric cleavage oxidation breakdown products that may induce adverse effects by interfering with retinoic acid availability at the tissue level. Vitamin A supplements provided only during the antenatal period in 2 other randomized, placebo-controlled trials in Malawi [20] and South Africa [21] had no significant effect on HIV-1 transmission, although the confidence intervals for the estimates were compatible with there being a wide range of effects. Vitamin A supplementation may increase the risk of HIV-1 transmission by enhancing the differentiation of myeloid and lymphoid cells, which is associated with increased expression of CCR5 receptors that increase the susceptibility to HIV-1 infection [16] The findings of the study in Zimbabwe highlight the importance of conducting rigorous research to assess the importance of low-cost interventions that are often presumed to be beneficial. Funding agencies and institutional review boards have a tendency to assume that vitamin supplementation can only be beneficial and, thus, rigorous studies may not be needed—or may even be unethical. In an earlier observational study, low plasma levels of vitamin A were reported to be associated with higher risks of vertical transmission of HIV and infant mortality [22]. Initiation of supplementation programs based on these promising findings would have been problematic, because low vitamin A concentrations as a result of reduced mobilization of liver stores are often observed in patients with infection, and low concentrations may be a marker of an advanced disease stage. Rigorous research to examine claims of the benefits of various interventions that are being marketed to HIV-infected individuals is also crucial. Recent press reports indicate that a company in South Africa has been promoting certain vitamin supplements without evidence that supports their safety or efficacy and have presented these as alternatives to antiretroviral therapy (ART) [23]. Although some nutritional interventions are beneficial in the management of HIV/AIDS, they in no way replace the need for comprehensive treatment and care, including ART, for people living with HIV/AIDS The effect that maternal vitamin A supplementation has on the health of mothers is also important. In a trial in Nepal, weekly maternal supplementation with preformed vitamin A or β-carotene was associated with a significant reduction in maternal mortality [24]. Similar trials are being conducted in Bangladesh and Ghana. Humphrey et al. [10] hypothesize that supplementation in HIV-negative women may reduce the incidence of HIV infection (although no data are presented). As additional data become available, the evidence for and against supplementation programs needs to be reassessed. For now, the evidence available—including these latest findings from Zimbabwe—raise concerns about the safety of maternal vitamin A supplementation programs as recommended by the World Health Organization (WHO) [25], particularly in settings where HIV infection is prevalent. Careful consideration of these data is also warranted before neonatal supplementation programs are introduced. These data are also relevant to ongoing field programs that provide periodic large doses of vitamin A beginning at the age of 6 months. The effect that these supplements have on HIV transmission has not been previously examined, and it is an urgent research priority. Until additional data are available, these programs should continue, particularly in settings where vitamin A deficiency is prevalent and exposure of infants to HIV is minimized through early weaning starting at this age. It is important to note, however, that vitamin A is an essential nutrient and that all individuals (HIV infected or uninfected) should consume the recommended dietary allowance Relatively less information is available on the potential effect that other vitamin deficiencies have on infant and child health and survival in developing countries. Children born to HIV-infected women are particularly at risk for multiple micronutrient deficiencies, because of poor dietary intake, malabsorption, and increased metabolic needs. Children whose mothers decide not to initiate breast-feeding, as well as those who are weaned early, may be at a higher risk of morbidity and mortality that may be counteracted by micronutrient supplements. Several studies have shown that multivitamin supplementation is important for slowing the rate of HIV disease progression and other HIV-related outcomes in adults [26]. In a trial in Tanzania, women who received multivitamins including vitamin B complex, C, and E (but not A) were significantly less likely to progress to WHO stage 4 disease or die, compared with those who received placebo [27]. Providing these supplements to women also resulted in significant improvements in child health outcomes [28], probably through improvements in the quality of breast milk, as well as enhancements in quality of life that would have provided these women with an opportunity to have better infant care practices. The safety and efficacy of direct supplementation with micronutrients other than vitamin A are being examined in children in randomized trials The field of micronutrient research and infectious diseases has developed rapidly during the past decade, although the results from various studies have not always been consistent. Finding consistency in the effects of vitamin supplementation in trials around the world may be less likely than it would be for other interventions, because of the differences in the nutritional backgrounds and the infectious disease burdens in different countries and the different social settings within countries. For instance, evidence is emerging, supported by the findings of Humphrey et al., that the benefits of vitamin A supplementation in pregnant women and their infants may be offset in HIV-infected women by an adverse influence of such supplementation on HIV transmission and disease progression. Albeit still preliminary, these findings reinforce the need for international collaboration to define the baseline nutritional profiles and prevailing burden of infections, to help us interpret the results of interventions, such as those presented in this important study I thank David Hunter and Eduardo Villamor, for reviewing an earlier version of this editorial commentary