Vitamin A Supplementation: Implications for Morbidity and Mortality in Children
- Reprints or correspondence: Dr. Wafaie Fawzi, Dept. of Nutrition, Harvard School of Public Health, 665 Huntington Ave., Boston, MA 02115 (mina{at}hsph.harvard.edu).
Abstract
Vitamin A deficiency impairs epithelial integrity and systemic immunity and increases the incidence and severity of infections during childhood. However, findings from vitamin A supplementation trials are not consistent. Supplementation has resulted in significant reductions in mortality in several (but not all) large community-based trials among apparently healthy children. In hospital-based studies, vitamin A supplements have been consistently found to reduce the severity of measles infection, but no effect on nonmeasles respiratory infections has been observed. In some cases, the supplements were associated with an apparently increased risk of lower respiratory infection. Vitamin A supplements also reduced the severity of diarrhea in most (but not all) trials. Potential explanations for the differences in efficacy across trials are reviewed. While vitamin A supplementation is effective in reducing total mortality and complications from measles infections, it is likely to be more effective in populations suffering from nutritional deficiencies.
More than 100 million preschool-aged children live in areas where they are at risk of vitamin A deficiency (VAD) [1]. It is estimated that about 3.3 million children have clinical signs and symptoms of VAD [2]. The evidence for an association between deficient vitamin A status and the occurrence of adverse health outcomes potentially related to the immune response was compiled by Scrimshaw et al. [3] some 30 years ago. Since then, supporting data from animal models, observational studies in humans, and intervention trials have been published [4].
Vitamin A supplementation was found to decrease mortality among measles patients as early as the 1930s [5]. During the 1960s, it was observed that the increased mortality rate among severely malnourished children was particularly high for those with eye signs of VAD [6–8]. This was confirmed by longitudinal studies in Indonesia, where preschool children with mild xerophthalmia followed for 18 months had 3 to 12 times the risk of death than those without clinical VAD [9]. Positive associations between mild VAD and the risk of respiratory disease and acute diarrhea were also reported [10]. Some of these observational studies, however, have limitations, including small sample sizes and lack of adjustment for confounding variables such as socioeconomic and nutritional status. Here we focus on the efficacy of vitamin A supplements on mortality and morbidity end points reported in controlled trials. We divide our presentation into three parts: First, the results of community-based trials that examined the effect of the supplements on mortality; second, study results of the effect of vitamin A supplements on measles; and third, hospital-based and community-based trials that assessed the effect of the supplements on the incidence and/or severity of acute respiratory and diarrheal infections.
Mortality Trials
In community-based studies conducted among children older than 6 months, the protective effect of vitamin A varied (table 1). Vitamin A administered at doses of 200,000 IU every 6 months to Indonesian preschool children was related to a 34% reduction in mortality [11]. However, in studies in India [14] and the Sudan [18], only modest or no effect was observed. In other trials that used the same dose administered every 4 months, the supplements resulted in a significant reduction in mortality: 19% in Ghana [19] and 30% in Nepal [15]. Trials of weekly supplementation at a level equivalent to recommended dietary intake in India [13] and food fortification in Indonesia [12] showed risk reductions of 54% and 45%, respectively. When results of eight studies [11–18] were pooled, vitamin A supplements significantly reduced total mortality by 30% [24]. The effect varied with the dosage and frequency of administration; small frequent doses were associated with a 42% reduction and large spaced doses with a 19% reduction in mortality. The effect also seemed higher among children ⩽ 1 year old and was more strongly protective against death from diarrheal disease. In a trial completed in Tanzania after the publication of this meta-analysis, a 50% reduction in mortality was noted both among children infected with the human immunodeficiency virus (HIV) and among those not HIV infected who received vitamin A [23].
Table 1
Community-based studies of vitamin A (VA) supplementation and total mortality in children.
A few studies have examined the efficacy of vitamin A supplements among children <6 months old. The administration of 52 µmol of vitamin A (≈50,000 IU) to neonates in Indonesia resulted in a significant 64% reduction in the risk of death during the first year of life [21]; however, no effect was observed in a similar trial in Nepal after the administration of 50,000 IU and 100,000 IU to children <1 month and 1–5 months old, respectively [20]. Morbidity and mortality associated with respiratory infections during infancy were not affected by vitamin A supplementation in a meta-analysis that included nine trials with data on this age group [25]. Similarly, in a recent large multicenter trial with data from Peru, Ghana, and India, vitamin A supplements (25,000 IU) did not affect mortality or morbidity when administered at the time of the three diphtheria-tetanus toxoid-pertussis/poliomyelitis immunizations at ages 6, 10, and 14 weeks [22].
The variability in the results across trials has a number of potential explanations [26]. The existence of concomitant nutrient deficiencies may impair the bioavailability of the supplements, since some of these nutrients (including fat, protein, and zinc) could be limiting factors for the absorption and utilization of the lipid-soluble vitamin. As discussed before, the effect also varies with the regimen characteristics, such as the dosage and interval of administration: Smaller and more frequent doses of vitamin A seem to be more protective than large periodic doses [12, 13]. The particular epidemiologic profile of the population under study is a relevant factor as well (e.g., the prevalence of infectious diseases at the time of supplementation and their incidence during the follow-up period). Infection at the time of administration may impair the absorption of the supplement while incident infection causes the depletion of vitamin stores and increases losses of vitamin A, thus shortening the protective period of the supplement. The supplements are expected to have greater impact in areas in which VAD is highly prevalent. However, since most trials among children ⩾6 months old did not include a baseline assessment of vitamin A status, it is difficult to examine this potential factor.
Measles
Several hospital-based trials have assessed the efficacy of vitamin A supplements on measles-associated morbidity and mortality (table 2). In 1932, Ellison [5] reported a 60% reduction in the risk of death among children admitted to the Grove Hospital's measles wards in London. Three hundred children were randomized to receive “300 Carr & Price units of vitamin A per day” and 2000 IU of vitamin D; another 300 children were selected as controls. A limitation of the study is that the control group did not receive placebo. In a meta-analysis that included Ellison's and three more recent trials: two placebo-controlled from South Africa [27, 28] and one with no placebo control group from Tanzania [29], large doses of vitamin A given on admission resulted in a significant reduction (∼60%) in the risk of death overall [24]. Mortality in children who were admitted with measles-pneumonia or who developed it during the hospital stay was reduced by about 70% when compared with the control children. The protective effects of the supplements were apparently greater among infants than in older children.
Table 2
Controlled trials on the effect of vitamin A (VA) supplements on measles.
In a later trial of 294 children in Kenya, vitamin A supplements had no effect on mortality; however, the study had limited power to examine this question [30]. Children with diarrhea on admission who received vitamin A recovered significantly faster than those given placebo. Another placebo-controlled trial was conducted in Zambia among children who had measles but whose infection was not severe enough to warrant hospital admission [31]. Children without pneumonia at baseline who were given vitamin A supplements were at lower risk of developing the disease. Those hospitalized with pneumonia and given the supplements had an apparent decreased probability of recovery than the control group. The sample size of this study was also small to allow for precise estimation of the associations of interest. When results of community-based trials were pooled, periodic vitamin A supplementation of children who were apparently healthy at baseline resulted in a 39% reduction in measles-related mortality in addition to the decrease in overall mortality [24].
Diarrheal and Respiratory Infections
In some of the measles trials, vitamin A resulted in a significant reduction in the occurrence and severity of respiratory and diarrheal complications; however, the generalizability of the findings to nonmeasles episodes of these infections was not possible. A number of trials have examined the effect of vitamin A supplementation on diarrhea and respiratory infections in children not infected with measles.
Hospital-based studies
Four vitamin A efficacy trials have been done among children hospitalized with diarrhea (table 3). In a placebo-controlled study among children in Bangladesh with noncholera watery diarrhea, there were no differences in the duration of illness or the stool output between the two treatment arms [32]. Diarrhea was mainly due to rotavirus and enterotoxigenic Escherichia coli. However, in another study in Bangladesh, children with shigellosis who received vitamin A were 32% less likely to be clinically ill by day 5 of the trial than were children given placebo [33]. There was no difference in bacteriologic cure between the treatment groups. The varying efficacy of vitamin A supplements between the two studies in Bangladesh may reflect differences in the pathogenesis of diarrhea attributable to variability in specific etiologic factors within each setting. Shigellosis is a serious intestinal infection associated with mucosal breeches and protein-losing enteropathy that could result in a relatively higher risk of complications than would be observed during episodes of gastroenteritis caused by less pathogenic microorganisms.
Table 3
Hospital-based studies of vitamin A (VA) supplementation and diarrheal and respiratory infections.
Vitamin A supplements had no overall effect on the duration of diarrhea among children in India [34] or malnourished children from the Congo [35]. The etiology of diarrhea was not examined in either study. In the Indian trial, children severely deficient in vitamin A (defined by conjunctival impression cytology), who received vitamin A supplements, had diarrhea episodes that were, on average, 32.6 h shorter than in the control group.
The efficacy of vitamin A supplements on the severity of disease among children hospitalized with pneumonia was examined in several placebo-controlled trials (table 3). In Guatemala, the supplements had no effect on the duration of hospital stay or on the number of days of several signs of respiratory disease (e.g., hypoxia, fever, or rapid respiratory rate) [36]. Similarly, in a trial in Tanzania, the vitamin A and placebo groups had comparable mean days of hospitalization and mean days of hypoxia, fever, or rapid respiratory rate [37]. Two other trials from Brazil [38] and Vietnam [39] found no effect of vitamin A on the course of pneumonia. In Brazil, there was an apparent protective effect among a subgroup of children who had pneumonia sufficiently severe to warrant hospitalization. In Vietnam, the duration of hospital stay was, on average, 1.8 days shorter in the vitamin A group than in the placebo group. The effect was limited to children who were moderately malnourished. In the Tanzania trial, in which all subjects enrolled were hospitalized, no differences in the effect of the supplements were observed among children who had a more severe condition at baseline or in categories of age, breastfeeding status, anthropometric status at baseline, or category of dietary vitamin A intake in the 4 months before hospital admission.
Evidence suggests that vitamin A supplements may increase the risk of adverse outcomes when given to children with pneumonia under particular circumstances. In a well-designed placebo-controlled study from Peru, vitamin A supplements resulted in significantly lower blood oxygen saturation and persistence of retractions by day 3 of hospitalization [40]. There was also a nonsignificant greater need for supplemental oxygen in the vitamin A group. In the Tanzania study, mortality was 63% higher in the vitamin A group: Of 346 children given vitamin A, 13 died in hospital compared with 8 deaths among 341 children in the placebo group [37]. The contrast in mortality among groups was not statistically significant; however, the study was not designed with adequate statistical power to examine the effect of the supplements on case fatality.
Three trials specifically examined the efficacy of vitamin A supplements among children hospitalized with pneumonia due to respiratory syncytial virus (RSV). RSV is a paramyxovirus similar to measles and is an important cause of bronchiolitis and pneumonia among infants and children. In a US study, children in the group given supplements needed, on average, 7.3 more days of intensive care and received supplemental oxygen over a longer period than children on placebo, although these differences were not statistically significant [41]. In a larger multicenter US trial, there were no differences in the mean number of days with rapid respiratory rate, need for supplemental oxygen, or intensive care between the vitamin A and placebo groups. However, children who received vitamin A were hospitalized, on average, about half a day longer than those given placebo [42]. In contrast, in a trial in Santiago, Chile, vitamin A supplementation resulted in an apparently more rapid recovery from tachypnea among children with the most severe hypoxemia at baseline (PO2 <90) [43].
Community-based studies
Given the protective effects of vitamin A supplements on mortality, it was presumed that vitamin A would have beneficial effects on morbidity. That was not the case, however, in a number of community trials [44] (table 4). In two trials that found an effect on mortality in Aceh, Indonesia [11, 45], and Tamil Nadu, India [13, 46], there was no effect on the incidence of diarrheal or respiratory infections after the intervention. In Aceh, morbidity was assessed in the week preceding a 6-month visit while in Tamil Nadu, the health status of the children was ascertained weekly. In both trials, however, the primary end point of interest was mortality. Misclassification of morbidity end points is a potential source of bias.
Table 4
Community-based studies of vitamin A (VA) supplementation and diarrheal and respiratory infections.
Several studies specifically examined the effect of the supplements on the incidence and/or severity of infections. A detailed morbidity profile was made on all children by use of passive and active surveillance of the study populations. In several trials, beneficial effects of vitamin A were noted with regard to diarrhea, but no effect was observed on the risk of pneumonia. In a trial in Ghana, where children were followed weekly to ascertain the occurrence of morbidity, there were no significant differences between the two treatment arms with respect to the prevalence of diarrheal or respiratory conditions [19]. However, children who received vitamin A had significantly fewer clinic visits and hospital admissions. In addition, the supplements resulted in a significant reduction in overall and diarrhea-specific mortality, but no effect on respiratory-related deaths was reported. These findings suggest that the protective effect of vitamin A against mortality was mediated by a reduction in the severity, rather than the incidence, of infections.
Among children in Brazil who were visited at home 3 times a week as part of a placebo-controlled trial of vitamin A supplementation, there were no significant differences in the incidence of pneumonia or the frequency of hospitalization between the two groups [47]. However, vitamin A resulted in a significant reduction in the mean daily prevalence and in the mean number of episodes of diarrhea, particularly severe episodes, in this study and in another placebo-controlled study in New Delhi [48]. In a trial in South Africa among children born to HIV-infected women, vitamin A supplements resulted in an apparent but nonsignificant reduction of both the incidence of total and severe diarrhea and in lower respiratory infection [49]. Among HIV-infected children, however, vitamin A supplements resulted in a significant 49% reduction of the risk of morbidity associated with diarrhea. In the Tanzania trial of children hospitalized with pneumonia [50], the risk of having ⩾1 episode of severe watery diarrhea during the year after discharge from the clinics was 44% significantly lower in the vitamin A group than among children who received placebo. The risk reduction was apparently stronger among undernourished children. In two trials from Brazil [51] and Indonesia [52], vitamin A—supplemented children had similar risk and severity of respiratory infection compared with children given placebo.
Findings from placebo-controlled trials from Indonesia [53], India [54], and Tanzania [50] suggest that large doses of vitamin A may be harmful when given to well-nourished children. In Indonesia, there was a significant increase (83%) in the risk of acute lower respiratory infection (ALRI) among nonstunted children who received vitamin A compared with the placebo group. For stunted children, supplements resulted in a 29% reduction in the risk of ALRI. In the same study population, although the supplements had no effect on diarrhea, vitamin A resulted in a significantly higher risk of diarrhea among children <30 months old and a significantly lower risk among older children. However, the effect of the supplements on risk of pneumonia was not modified by the occurrence of wasting nor was the effect on diarrhea affected by wasting or stunting. Moreover, there were no differences in the mean duration of ALRI or diarrhea in the vitamin A and placebo arms.
In a randomized trial in India [54], vitamin A supplements had no significant effect on percentage of time ill or number of respiratory infection episodes. Children who received vitamin A had a significantly increased mean duration of diarrheal episodes. Compared with other trials, this population of Indian children had relatively better health care, including high coverage of immunization, awareness among mothers about health and nutrition, and routine deworming. These may have reduced the chances of finding a protective effect of the supplements. In Tanzania, among children with pneumonia who were randomized to receive 2 doses of vitamin A or placebo at the time of hospital admission and 4 monthly doses after discharge, there was a 67% significant increase in the risk of having at least 1 episode of cough and rapid respiratory rate during the year after discharge [50]. This apparently increased risk was limited to HIV-negative children. Treatment was also associated with higher risk of acute diarrhea among well-nourished or stunted children but was protective among wasted subjects. Similar results were reported in a trial of children in Ecuador [55]: Administration of weekly low doses of vitamin A resulted in increased incidence of ALRI in infants with adequate weight for age. In contrast, among underweight children, vitamin A was protective against ALRI.
Results from two trials from Australia and one in Haiti support the hypothesis that vitamin A supplements may increase signs of infection, particularly respiratory signs, such as cough [56]. In a study in Australia, vitamin A—supplemented children experienced a 17% increase in the median number of cough days and a 9% and 43% increase in the median days with runny nose and sore throat, respectively [57]. In Haiti, vitamin A supplementation resulted in a significant increase in the 2-week prevalence of all symptoms of respiratory morbidity, including cough, rapid respiration, and diarrhea [58].
In contrast to all of the above cited studies, in two studies from China [59] and Thailand [60], significant reductions in the incidence of both respiratory infection and diarrhea were noted in vitamin A—supplemented children compared with control children. A main limitation of both studies was that the control group did not receive a placebo. Hence, the investigators were not blinded with respect to the treatment arm, raising the possibility of bias in the ascertainment of outcomes by the research staff.
Comment
Vitamin A supplementation to children decreases the overall risk of mortality by about 30% [24]. In hospitalized children with measles, the mortality reduction attributable to vitamin A supplementation is 60%, on average. Amelioration in the severity of certain episodes of diarrhea is also shown by supplementation trials. Several mechanisms are likely for these effects, including a protective action of the vitamin on the epithelial lining of the gastrointestinal tract, increased mucus secretion, and enhanced local barriers to infection [61]. The correction of VAD through supplementation may also improve humoral and cellular immune functions, including increases in B lymphocyte activation, proliferation and production of IgM and specific IgG, improvements in the T cell—helping response and cytokine synthesis, and enhancements in the function of natural killer cells and the monocyte/macrophage lineage [62].
Among children who are not undernourished, however, supplementation may be associated with adverse effects, particularly in those with pneumonia [37, 40]. The increased occurrence of signs of respiratory infection associated with vitamin A supplements may indicate an improvement in the inflammatory response, attributable to a pharmacologic effect of the supplement, but the long-term implications of these increased respiratory signs are not clear. Vitamin A supplements should not be given during pneumonia episodes without measles, unless there is evidence of VAD. The conditions under which vitamin A supplements can be harmful need to be examined further. Given that vitamin A supplements may be beneficial in reducing diarrheal disease in the period after discharge from hospital, the supplements could be given after recovery from pneumonia and at the time of hospital discharge.
HIV infection, an increasingly prevalent condition in many developing countries, poses new challenges to the nutritional status and survival of children. Vitamin A supplementation every 3–4 months to HIV-positive children <5 years old is beneficial in reducing diarrheal morbidity [49], total mortality, and AIDS-related deaths [23]. The effect could be mediated through a vitamin A-related enhancement of the immune function. The apparent adverse effects associated with vitamin A supplementation in well-nourished subjects have not been observed among HIV-infected children [50]. Supplementation should be considered as part of the standard of care among children infected with HIV.
Large doses of vitamin A provide an effective solution to the problem of VAD in areas of the world where this is a public health problem. In these communities, vitamin A supplements can reduce total and diarrhea-specific mortality. When periodic dosing is chosen as an intervention strategy, we suggest that the doses be administered every 4 months. Given the varying degree of protection shown in different studies, more research is needed on factors that affect the bioavailability and retention of the large doses.
Even though periodic large doses of vitamin A are beneficial in the short term, their use as the only approach to the problem of VAD has limitations. VAD coexists with other nutrient deficits that are not addressed by the supplementation program. In addition, the effectiveness of this approach is limited to the duration of the program, and children who live in distant places and who probably need the supplement most may be difficult to reach consistently at 1- to 6-month intervals. Furthermore, large programs can put financial and logistical strains on the health care systems in many developing countries. Toxicity due to ingestion of multiple large doses over a short period is also a real possibility that needs to be guarded against.
A more sustainable solution to the problem of VAD is to guarantee that the population has an adequate consumption of vitamin A in the diet. Small frequent doses (in amounts corresponding to those in diet) may be more protective against mortality and morbidity than large periodic doses. Most communities in which VAD is a serious problem have abundant supplies of vegetables and fruits rich in carotenoid with provitamin A activity. Dietary vitamin A intake is associated with a significant reduction in mortality [63], diarrheal and respiratory infections [64], and risks of stunting or wasting [65]. Programs aimed at increasing consumption of dietary vitamin A in these communities should be undertaken in addition to the administration of supplements if the latter strategy is implemented. In areas where vitamin A-containing foods are not so abundant, horticultural approaches should be considered. Food fortification programs can also be useful in improving the vitamin A status of populations, provided that the groups at highest risk of deficiency are reached by these programs.
This review has focused on vitamin A. However, multiple supplementation with other vitamins and minerals is also likely to reduce the burden of adverse health outcomes, since there are physiologic interactions between nutrients and specific micronutrient deficiencies often overlap. Supplementation with iron and zinc, for example, is related to improved vitamin A status among preschool children [66], and combined supplements of vitamin A and iron increase hemoglobin levels above the increments attributable to only one of the nutrients in pregnant women [67] and in anemic preschool children [68]. Multivitamin supplementation during gestation of HIV-infected women has been associated with decreased risk of adverse gestational end points including low birth weight, severe preterm births, and intrauterine growth retardation [69], and supplementation with vitamin A is related to lower risk of preterm delivery [70]. The effect of multiple nutritional supplements on various health outcomes in children is yet to be elucidated.
Vitamin A and other nutritional intervention strategies should be integrated into community programs that are dealing with other health problems rather than being implemented as a vertical program. Additional operational research on the ways of accomplishing this goal is needed.
- © 2000 by the Infectious Diseases Society of America
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.
- 17.
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 6.288
5-Yr impact factor: 6.103
For The Media
For Authors
- Services for authors
- Self-archiving policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
