Skip Navigation

Effects of a Single Large Dose of Vitamin A, Given during the Postpartum Period to HIV-Positive Women and Their Infants, on Child HIV Infection, HIV-Free Survival, and Mortality

  1. Jean H. Humphrey,
  2. Peter J. Iliff,
  3. Edmore T. Marinda,
  4. Kuda Mutasa,
  5. Lawrence H. Moulton,
  6. Henry Chidawanyika,
  7. Brian J. Ward,
  8. Kusum J. Nathoo,
  9. Lucie C. Malaba,
  10. Lynn S. Zijenah,
  11. Partson Zvandasara,
  12. Robert Ntozini,
  13. Faith Mzengeza,
  14. Agnes I. Mahomva,
  15. Andrea J. Ruff,
  16. Michael T. Mbizvo and
  17. Clare D. Zunguza
  1. Reprints or correspondence: Dr. Jean H. Humphrey, #1 Borrowdale Rd., Borrowdale, Harare, Zimbabwe (jhumphrey{at}zvitambo.co.zw)
 
Next Section

Abstract

BackgroundLow maternal serum retinol level is a risk factor for mother-to-child transmission (MTCT) of human immunodeficiency virus (HIV). Multiple-large-dose vitamin A supplementation of HIV-positive children reduces mortality. The World Health Organization recommends single-large-dose vitamin A supplementation for postpartum women in areas of prevalent vitamin A deficiency; neonatal dosing is under consideration. We investigated the effect that single-large-dose maternal/neonatal vitamin A supplementation has on MTCT, HIV-free survival, and mortality in HIV-exposed infants

MethodsA total of 14,110 mother-infant pairs were enrolled ⩽96 h after delivery, and both mother and infant, mother only, infant only, or neither received vitamin A supplementation in a randomized, placebo-controlled trial with a 2×2 factorial design. All but 4 mothers initiated breast-feeding. A total of 4495 infants born to HIV-positive women were included in the present analysis

ResultsNeither maternal nor neonatal vitamin A supplementation significantly affected postnatal MTCT or overall mortality between baseline and 24 months. However, the timing of infant HIV infection modified the effect that supplementation had on mortality. Vitamin A supplementation had no effect in infants who were polymerase chain reaction (PCR) negative for HIV at baseline. In infants who were PCR negative at baseline and PCR positive at 6 weeks, neonatal supplementation reduced mortality by 28% (P=.01), but maternal supplementation had no effect. In infants who were PCR negative at 6 weeks, all 3 vitamin A regimens were associated with ∼2-fold higher mortality (P⩽.05)

ConclusionsTargeted vitamin A supplementation of HIV-positive children prolongs their survival. However, postpartum maternal and neonatal vitamin A supplementation may hasten progression to death in breast-fed children who are PCR negative at 6 weeks. These findings raise concern about universal maternal or neonatal vitamin A supplementation in HIV-endemic areas

In the 1990s, observational studies of HIV-positive women demonstrated that a low serum retinol level during pregnancy is associated with higher infant mortality or vertical transmission [1–3] and a 20-fold greater likelihood of high HIV DNA concentrations in breast milk [4]. On the basis of vitamin A’s ubiquitous role in immune function and the maintenance of epithelial barriers [5], potential mechanisms were articulated [6, 7] and studies were designed [8, 9, 10] to test the hypothesis that providing HIV-positive women with vitamin A supplements would reduce mother-to-child transmission (MTCT)

The results of 3 trials have been disappointing. In South Africa [9], daily supplementation with 5000 IU of vitamin A and 30 mg of β-carotene during the third trimester of pregnancy plus 200,000 IU of vitamin A at delivery and, in Malawi [10], daily antenatal supplementation with 10,000 IU of vitamin A had no overall effect on MTCT. In Tanzania, daily supplementation with 5000 IU of vitamin A and 30 mg of β-carotene throughout pregnancy and lactation plus 200,000 IU vitamin A at delivery increased the risk of MTCT by 38% [11, 12]

Although women in these trials were provided with daily doses of vitamin A, the World Health Organization (WHO) recommends a simpler regimen for women living in areas of prevalent deficiency: 1 dose of 200,000 IU during the 8-week postpartum period [13]. This intervention has been demonstrated to improve maternal and infant vitamin A status [14, 15] and to reduce infant morbidity [16], and it is presently implemented in at least 15 countries. Because a dose of 200,000 IU may be too small to replete vitamin A stores in deficient women [15], doubling this regimen to 2 doses of 200,000 IU is under consideration [17, 18]. Direct supplementation with 50,000 IU of vitamin A in neonates is also under consideration, after 2 trials (neither conducted in HIV-prevalent populations) found that this regimen reduced infant mortality by 20%–60% [19, 20]

The primary objectives of the Zimbabwe Vitamin A for Mothers and Babies (ZVITAMBO) trial were to measure the effect that a single large dose of vitamin A given to postpartum HIV-positive women and/or their infants had on breast-feeding–associated MTCT and HIV-free survival. Because of the results of the Tanzanian study, it also was important to determine whether single-dose postpartum regimens might have adverse effects in HIV-positive women or their infants

A second group of trials conducted during the 1980s and early 1990s demonstrated that vitamin A supplementation in preschool children living in areas of prevalent deficiency reduced mortality by 20%–45% [21]. These findings led to the expansion of vitamin A supplementation programs for children 6 months to 6 years old, and these programs are now implemented in at least 50 countries [13]. Further research in HIV-positive children demonstrated that multiple large doses of vitamin A reduced diarrhea episodes [22, 23], increased CD4 cell counts [24], and reduced all-cause mortality [23, 25]. A second objective of ZVITAMBO was to measure the effect that single-dose postpartum vitamin A supplementation regimens had on mortality in HIV-exposed children. This outcome complements HIV-free survival, because the latter obscures the effects that supplementation has on mortality in HIV-negative infants (when data on their deaths are collapsed into the data on the larger number of infections) and does not detect treatment effects on the duration of survival in HIV-positive children (because their events are counted at the time of infection rather than at death)

The WHO categorizes Zimbabwe as “high-risk” for vitamin A deficiency [13]. At the time when the present study was undertaken, no national vitamin A supplementation programs for women or neonates were underway, although postpartum maternal supplementation was initiated ∼18 months after study recruitment was completed, on the basis of findings from a nationwide micronutrient survey [26]. The prevalence of HIV infection in adult Zimbabweans was 25.8% at the time when the present study was undertaken [27]

Previous SectionNext Section

Subjects, Materials, and Methods

Between November 1997 and January 2000, a total of 14,110 mother-infant pairs were enrolled at 1 of 14 maternity clinics and hospitals in greater Harare. Mother-infant pairs were eligible if neither had an acutely life-threatening condition, the infant was a singleton with birth weight ⩾1500 g, and the mother planned to stay in Harare after delivery. Written, informed consent was obtained from mothers. A 2×2 factorial design was used in the study, and all pairs were randomized, as described elsewhere [28], ⩽96 h after delivery (mean±SD, 20.3±16.6 h after delivery) to 1 of 4 treatment groups. The treatment groups were Aa, Ap, Pa, and Pp, where “A” was maternal vitamin A supplementation (400,000 IU), “P” was maternal placebo, “a” was infant vitamin A supplementation (50,000 IU), and “p” was infant placebo. The single dose of 400,000 IU was chosen for mothers on the basis of data that the existing WHO recommended dose of 200,000 IU was too small to reverse deficiency [15] and a pilot study demonstrating that a dose of 400,000 IU was well tolerated [29]. The regimen now under consideration by the WHO (2 doses of 200,000 IU given at least 1 day apart, to optimize absorption and minimize the risk of adverse effects for the breast-feeding infant) arose from the conclusions of an expert panel convened 2 months after study recruitment was completed [17]

At baseline, mothers were tested for HIV by 2 ELISAs run in parallel (HIV 1.0.2 ICE [Murex Diagnostics]; GeneScreen HIV 1/2 [Sanofi Diagnostics Pasteur]). Duplicate pairs of discordant ELISA results were resolved by Western blot (HIV Blot 2.2; Genelabs Diagnostics). In mothers who were enrolled from 1 October 1998 to the end of the study (∼60% of the total study population), hemoglobin concentrations were measured using a hemoglobinometer (HemoCue) [30]. In mothers who were HIV positive at recruitment, CD4 cells were counted (FACScount; Becton Dickinson), serum retinol levels in a random subsample of 691 mothers were measured using high-performance liquid chromatography [28], and viral load in a random subsample of 444 mothers was measured (Roche Amplicor HIV-1 Monitor test version 1.5; Roche Diagnostics). Baseline characteristics were collected from questionnaires and through transcription of data from hospital records. Gestational age was estimated [31]. Infant birth weight (Seca scale model 727) and maternal mid–upper-arm circumference were measured [32]. HIV testing and antiretroviral prophylaxis for HIV-positive antenatal women were not available in Harare public-sector facilities during ZVITAMBO recruitment, except to 69 women participating in a UNICEF feasibility study (none of whom chose to join ZVITAMBO), who were provided with zidovudine

Follow-up visits were conducted at 6 weeks, 3 months, and then every 3 months for 12–24 months. We initially planned to follow HIV-positive mothers and infants for 24 months. However, in June 2000, economic conditions necessitated shortening the study. Thus, 24%, 48%, and 100% of the pairs were reassigned to 24-month, ⩾18-month, and ⩾12-month follow-up, respectively. Mothers who were HIV positive at baseline were retested at their next visit. Medical care and counseling [28, 33, 34] were offered throughout the trial

Plasma and cell pellets (Amplicor whole-blood PCR sample preparation method; Roche Diagnostics) were prepared from infant blood at all visits and were stored at −70°C. After follow-up was completed, the last available sample from each infant was tested (for samples collected at ⩾18 months of age, plasma was tested using GeneScreen HIV 1/2; for samples collected at <18 months of age, cell pellets were tested using Amplicor HIV-1 DNA test version 1.5 [Roche Diagnostics]). If the last available sample was negative for HIV, the child was classified as being HIV negative; if the last available sample was positive for HIV, samples that had been collected at younger ages were tested to determine the timing of the infection

Statistical methods Continuous and categorical baseline characteristics were compared across the 4 treatment groups by Kruskal-Wallis and χ2 tests, respectively. Breast-feeding rates were estimated for each group by use of Kaplan-Meier methods and were compared by use of pairwise log-rank tests

Data analysis was guided by consensus recommendations [35]. The cumulative proportions of HIV-positive infants were estimated by the Turnbull method [36], and 95% confidence intervals (CIs) were calculated using 2000 bootstraps [37]. Infants were censored at the earliest of 790 days, the last negative HIV test result, or 60 days after the mother’s death or breast-feeding cessation. Analyses were conducted among (1) all infants born to HIV-positive mothers, (2) only infants who were PCR negative for HIV at baseline, and (3) only infants who were PCR negative for HIV at 6 weeks (“6-week-negative infants”). Differences (and 95% CIs) between the cumulative proportions of HIV-positive infants in the Pp group and each of the 3 vitamin A regimens at 6, 12, and 24 months were calculated

Turnbull estimates for the category “infection or death” were made for baseline-negative and 6-week-negative infants. Infants who died after having a positive HIV test result were retained in the analysis until the date of infection. Infants whose last HIV test result was negative were censored at the earlier of the date of the last negative HIV test result or 60 days after breast-feeding cessation, except for those who died, who were retained until the date of death

To estimate the adjusted effect that vitamin A supplementation had on infection and on infection or death, Cox proportional hazards models were determined via a forward stepwise regression procedure with variable entry at α=0.20 and retention at α=0.10. The effect that vitamin A supplementation had on child mortality was estimated using the Kaplan-Meier method and a proportional hazards model that was constructed using the same procedure described above

Separate analyses were conducted among infants who were infected during the intrauterine period (“IU infants”; defined as those who were PCR positive at baseline), infants who were infected during the late intrauterine/intrapartum/early postnatal period (“IP infants”; defined as those who were PCR negative at baseline and PCR positive at 6 weeks), and infants who were infected during the postnatal period (defined as 6-week-negative infants). These secondary analyses were not based on a priori hypotheses. However, the trend toward harm for the vitamin A–containing study arms in both Kaplan-Meier and Cox analyses, together with the knowledge that the effects of vitamin A on HIV in vitro can vary depending on experimental conditions, compelled us to explore whether any subgroups of infants may have been harmed. In a model testing the interaction between treatment group and the timing of infant HIV infection, the joint test of the 2-way interaction terms between the main effects of the timing of HIV infection and vitamin A treatment was significant (P<.001), which provided us with further support for undertaking these analyses (interaction analyses are presented in table 1)

For infants who died, a study pediatrician who was blinded to treatment and maternal and infant HIV status assigned the cause of death (multiple causes were allowed and were not&amp;ranked [38]) and assigned an HIV infection likelihood score on the basis of clinical evidence of HIV/AIDS reported in medical records or verbal autopsy. Infants whose deaths were associated with signs and symptoms approximating WHO disease stage 1 [39] or less were assigned a score of 0 (meaning no or weak evidence of HIV infection), infants whose deaths were associated with signs and symptoms approximating stage 2 were assigned a score of 1 (meaning some evidence of HIV infection), and infants whose deaths were associated with signs and symptoms approximating stage 3 were assigned a score of 2 (meaning strong evidence of HIV infection). Of the 834 infants with available PCR data and HIV infection likelihood scores who were born to HIV-positive mothers, 193 of 239 infants who received a score of 1 and 194 of 217 infants who received a score of 2 were PCR positive at the time of death (positive predictive value [PPV] of a score of 1 or 2, 81% and 89%, respectively). Of 378 infants who received a score of 0, 174 were not PCR positive at the time of death (negative predictive value [NPV] of a score of 0, 46%)

Ethics approval The Medical Research Council of Zimbabwe, Medicines Control Authority of Zimbabwe, Johns Hopkins Bloomberg School of Public Health Committee on Human Research, and Montreal General Hospital Ethics Committee approved the study protocol

Previous SectionNext Section

Results

A total of 4495 infants were born to HIV-positive mothers and included in the present analysis (figure 1). Follow-up rates to 12 or 24 months did not differ across treatment groups (for the respective χ2 tests [3 df], P=.16 and P=.65). All but 4 HIV-positive mothers initiated breast-feeding, and 97%, 92%, 66%, and 19% were still breast-feeding at 6, 12, 18, and 24 months after delivery, respectively (data not shown). These proportions did not differ across treatment groups (for all χ2 tests [3 df], P>.7). There were statistically significant but small imbalances across treatment groups in maternal education and birth weight (table 2), but other baseline characteristics did not differ. In all treatment groups, ∼60% of women had serum retinol levels <1.05 μmol/L at delivery, but there were no reported cases of maternal night blindness during the just-completed pregnancy (data not shown), as is common in severely deficient populations [40]

Figure 1
View larger version:
Figure 1

Flow of participants through the trial. FU, follow-up; HIV-, HIV negative; HIV+, HIV positive; IND, indeterminate; PCR-, PCR negative; PCR+, PCR positive; wks, weeks; w/o, without

MTCT of HIV For all infants born to HIV-positive mothers, Turnbull estimates of cumulative infection (95% CIs) were 8.6% (7.2%–10.0%), 26.6% (25.1%–27.9%), and 37.5% (35.7%–40.1%) at baseline, 6 weeks, and 24 months, respectively. Thus, proportionately, 22.9%, 48.0%, and 29.1% of all MTCT of HIV occurred during the intrauterine, late intrauterine/intrapartum/early postnatal, and postnatal periods, respectively. Cumulative proportions (95% CIs) of infection or death were 29.1% (26.2%–32.5%) and 43.2% (38.2%–48.8%) at 6 weeks and 24 months, respectively

In all infants born to HIV-positive mothers, infection rates and infection-or-death rates tended to be higher in the Ap and Pa groups, compared with those in the Aa and Pp groups (table 3). This was also the case for baseline-negative infants, and the difference between the infection-or-death rates at 12 months reached statistical significance (tables 3–6 and figure 2A ). In 6-week-negative infants, neither infection rates nor infection-or-death rates differed between the 4 treatment groups (tables 3 and 6 and figure 2B ). The data presented in figure 2A suggest that the infection rates in the Ap and Pa groups diverged from those in the Aa and Pp groups between baseline and 6 weeks and then remained parallel. This observation was supported by a Cox model for infection at 6 months, in which, compared with the Pp treatment, time did not modify the effect on infection by the Aa treatment, but time did modify the effect on infection by the Ap and Pa treatments (when time was entered as a log-transformed continuous variable, P=.030 and P=.028, respectively; when time was entered as a dichotomous variable [<6 weeks vs. ⩾6 weeks], P=.0020 and P=.014, respectively)

Figure 2
View larger version:
Figure 2

Turnbull estimates of HIV infection by vitamin A treatment group in infants who were polymerase chain reaction (PCR) negative for HIV at baseline (A) and infants who were PCR negative for HIV at 6 weeks (B)

Table 1
View larger version:
Table 1

Analyses investigating the interaction between vitamin A treatment and timing of infection in the effect on child death at 24 months

Table 2
View larger version:
Table 2

Baseline characteristics of mothers and infants, by vitamin A treatment group

Table 3
View larger version:
Table 3

Cumulative proportions of HIV infection and of HIV infection or death, by vitamin A treatment group

Table 4
View larger version:
Table 4

Probability of HIV infection, according to vitamin A treatment group

Table 5
View larger version:
Table 5

Probability of HIV infection or death, according to vitamin A treatment group

Table 6
View larger version:
Table 6

Proportional hazards models for risk of HIV infection or death by 24 months in infants who were polymerase chain reaction (PCR) negative at baseline and infants who were PCR negative at 6 weeks

Mortality between baseline and 24 months Of the 4495 HIV-exposed infants, 866 died during the 24 months after delivery (155.3/1000 child-years), and 621 (71.7%) of these children who died were PCR positive before death. Vitamin A treatment group was not significantly associated with mortality, after adjustment for other covariates (table 7). These results were similar to the effect that vitamin A supplementation had on mortality of all 14,110 infants enrolled in the trial: the hazard ratios (95% CIs) were 1.06 (0.89–1.26), 1.18 (0.99–1.40), and 1.16 (0.98–1.38) for the Aa, Ap, and Pa groups, respectively, compared with that for the Pp group

Table 7
View larger version:
Table 7

Proportional hazards models for risk of death by 24 months in all infants born to HIV-positive mothers

Mortality between baseline and 24 months according to timing of HIV infection A total of 381, 504, and 2876 infants were classified as IU, IP, and 6-week-negative infants, respectively (figure 1). Of these, 339 (89%), 478 (94.8%), and 2644 (91.9%), respectively, were followed for vital status to ⩾12 months

In IU infants, neither maternal nor neonatal vitamin A supplementation significantly affected mortality (table 8). In IP infants, neonatal vitamin A supplementation significantly reduced mortality at 24 months, by 28% (table 8). An effect of similar magnitude was observed in both groups who received neonatal vitamin A supplementation. Maternal vitamin A supplementation had no significant effect on mortality of these children

Table 8
View larger version:
Table 8

Adjusted child mortality risk, by vitamin A treatment group and infant HIV infection group

In 6-week-negative infants, all 3 vitamin A supplementation regimens were associated with ∼2-fold increases in mortality by 24 months (table 8). Of the 116 6-week-negative children who died, 46 had a PCR-positive test result before death. In a Cox model including postnatal (PN) infection as a time-dependent covariate and interaction terms between each of the treatment groups and PN infection, none of the interaction terms approached significance (P>.4 for Aa×PN, Ap×PN, and Pa×PN, compared with Pp×PN), which suggests that the adverse effect of vitamin A supplementation was similar in children who were and children who were not PCR positive before death. However, because HIV can kill children quickly, some of the 70 children who died with a last available HIV test result that was negative are likely to have been infected. To examine this further, we examined HIV infection likelihood score data, which were available for 69 of 70 infants. Eight infants received a score of 2, 22 infants received a score of 1, and 39 infants received a score of 0. Given that the PPV for a score of 1 or 2 was 81% and 89%, respectively, we estimated that at least 25 of these 70 deaths were associated with HIV infection acquired during breast-feeding. The low NPV of a score of 0 suggested that 25 deaths might be an underestimate. Therefore, we estimated that the majority (61% [46 infants who were PCR positive prior to death and 25 additional infants who were likely to have been HIV infected prior to death, out of 116 infants who died]) of 6-week-negative infants who died were HIV positive, indicating that the adverse effect of vitamin A supplementation possibly reflects hastened HIV disease progression to death

Of the 872 deaths for which the cause of death was available, 71%, 22%, and 18% were associated with acute respiratory tract infection, malnutrition, and gastrointestinal tract disease, respectively; these proportions did not differ across treatment groups. Other causes were associated with <5% of deaths

Previous SectionNext Section

Discussion

Neither maternal nor neonatal vitamin A supplementation, given as single large postpartum doses, significantly affected breast-feeding–associated MTCT. Considered together, evidence from all studies to date suggests that maternal vitamin A supplementation does not reduce postnatal MTCT, regardless of the regimen used; that daily maternal dosing beginning during pregnancy, continued throughout lactation, and including a large dose of β-carotene may increase MTCT [12]; but that supplementation given only during the antenatal period [9, 10] or only as a single large dose at delivery (as in the present study) does not increase breast-feeding–associated MTCT

Neither maternal nor infant vitamin A supplementation significantly affected the overall mortality of HIV-exposed children by 2 years of age. However, the effect of vitamin A supplementation on mortality varied depending on the time when the children were infected with HIV. In IU infants, vitamin A supplementation had no effect, perhaps because the severity of fetal HIV infection overwhelmed any potential effect of postnatal vitamin A supplementation. In IP infants, maternal vitamin A supplementation had no effect, but vitamin A given directly to the infant significantly reduced 2-year mortality, by 28%. However, in HIV-exposed children who did not become infected during pregnancy or delivery (6-week-negative infants), vitamin A supplementation (given to either mother or infant) approximately doubled the risk of death by 2 years of age

The molecular mechanism through which vitamin A works depends on its cytoplasmic metabolite (retinoic acid) binding to specific nuclear receptors [41]. These liganded nuclear retinoid receptors heterodimerize and bind to the regulatory regions of target gene DNA as retinoic acid response elements, to initiate transcription. Through this action, vitamin A has profound and varied effects on the differentiation and function of numerous cells and tissues, including many of the immune system

In vitro and in vivo studies have suggested that the interaction between vitamin A and HIV may vary with the timing of vitamin A supplementation relative to that of infection. Several research groups have shown that vitamin A treatment of HIV-infected cells significantly represses HIV transcription [42–48]. These findings are consistent with the results of clinical trials in which vitamin A supplementation of HIV-positive children halved mortality [23, 25] and reduced morbidity [22, 23]. Our study, in which we found a 28% reduction in mortality after neonatal vitamin A supplementation in the 504 infants infected during the late intrauterine/intrapartum/early postnatal period, is the third study indicating that vitamin A supplementation of HIV-positive children can slow their progression to death

In contrast, at least 3 research groups have reported that pretreatment of human monocytes or monocytelike cells with retinoic acid before infection with HIV results in increased virus production [45, 49, 50]. Our finding of higher mortality in 6-week-negative infants who had received vitamin A supplementation could be explained if priming with vitamin A increased viral load in those who became infected during breast-feeding, thereby hastening their progression to death

An enigma in our findings is the higher transmission rates in the Ap and Pa groups, compared with those in the Aa and Pp groups. These differences manifested between baseline and 6 weeks, resulting in significantly higher infection-or-death rates at 12 months (when statistical power was greater compared with that at 24 months). Given the 2×2 factorial design of the trial, we would expect an adverse or beneficial effect of maternal supplementation to be apparent in both study arms containing the maternal dose and, similarly, any effect of infant supplementation to be apparent in both study arms containing the infant dose. However, the 2 groups in which mother only or infant only received vitamin A (Ap and Pa) had higher MTCT rates, compared with those in the double-placebo group (Pp), whereas the MTCT rate in the Aa group was not different from that in the double-placebo group. This finding, together with the trend toward higher proportions of PCR-positive infants in the Ap and Pa groups at baseline (table 2) and the fact that the differences were apparent by 6 weeks (when the majority of the infants would have acquired HIV infection before dosing), suggests that the higher infection rates in the Ap and Pa groups, compared with those in the Aa and Pp groups, were due to baseline imbalances or chance, rather than the implausible interpretation that giving vitamin A only to the mother or only to the infant increases MTCT, whereas giving vitamin A to both the mother and the infant has no effect

Decisions about whether to implement postpartum maternal and neonatal vitamin A supplementation programs in HIV-endemic areas should be based on the overall balance of benefit, risk, and cost. Our findings support the growing evidence that vitamin A supplementation of HIV-positive infants and children is likely to prolong survival and should be included in their care. However, our findings also provide evidence that the benefits of universal postpartum supplementation in women [14–16] may not justify the costs and risks in HIV-endemic areas. In our study, maternal vitamin A supplementation provided no overall benefit for MTCT or infant survival and apparently increased mortality in HIV-exposed, 6-week-negative children

At least 2 issues require further research to fully inform policy decisions about postpartum maternal supplementation in HIV-endemic areas. First, in severely vitamin A–deficient, presumably HIV-negative women in Nepal, weekly vitamin A supplementation halved maternal mortality [51]. If ongoing studies confirm this finding, the effect that vitamin A supplementation (perhaps given in smaller daily or weekly doses) has on maternal mortality should be reexamined in an HIV-prevalent population that has poorer vitamin A status than did the mothers who participated in ZVITAMBO. Second, in a small subsample of HIV-negative mothers in the ZVITAMBO trial for whom serum retinol levels were measured, those with low levels (<0.7 μmol/L) were 10.4 (95% CI, 3.0–36.3) times more likely to acquire HIV during the postpartum period than were those with higher levels (J.H.H., J. W. Hargrove, L.C.M., P.J.I., L.H.M, K.M., P.Z., K.J.N., F.M., R.N., H.C., L.S.Z., C.D.Z., M.T.M., B.J.W., and the ZVITAMBO Study Group, unpublished data). Supplementation in these apparently vitamin A–deficient mothers tended to be protective against acquisition of HIV infection, although not significantly so (perhaps because of our limited statistical power to detect differences). Should this protection be statistically significantly confirmed, this benefit may override other issues related to postpartum maternal supplementation in populations in which vitamin A deficiency is widespread and HIV incidence is high [52]

Author affiliations ZVITAMBO Study Group (J.H.H., P.J.I., E.T.M., K.M., L.H.M., H.C., B.J.W., K.J.N., L.C.M., L.S.Z., P.Z., R.N., F.M., A.I.M., A.J.R., M.T.M., and C.D.Z.) and College of Health Sciences (J.H.H., P.J.I., K.J.N., L.S.Z., and P.Z.) and Institute of Food Science Nutrition and Family Sciences, University of Zimbabwe (L.C.M. and F.M.), and Ministry of Health and Child Welfare (A.I.M.) and Harare City Health Department (C.D.Z.), Harare, Zimbabwe; Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (J.H.H., L.H.M., and A.J.R.); Research Institute of the McGill University Health Centre, Montreal, Quebec, Canada (B.J.W.); World Health Organization, Geneva, Switzerland (M.T.M.); School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa (present affiliation of E.T.M.)

ZVITAMBO Study Group Members Additional ZVITAMBO Study Group members are John W. Hargrove, Florence Majo, Mary Ndhlovu, Ellen Piwoz, Lidia Propper, Phillipa Rambanepasi, and Naume Tavengwa

Previous SectionNext Section

Acknowledgments

We thank the Harare City Health Department, Harare Central Hospital, Mashonaland East Provincial Medical Department, and Chitungwiza Hospital, for collaborating and providing access to their facilities

Financial support The ZVITAMBO project was supported by the Canadian International Development Agency (R/C project 690/M3688), the US Agency for International Development (USAID; cooperative agreement HRN-A-00-97-00015-00 between Johns Hopkins University and the Office of Health and Nutrition, USAID), and a grant from the Bill and Melinda Gates Foundation. Additional funding for the study was provided by the Rockefeller Foundation and BASF

Previous SectionNext Section

Footnotes

  • (See the editorial commentary by Fawzi, on pages 756&ndash;9.)

  • ↵Presented in part: 21st International Vitamin A Consultative Group Meeting, Marrakech, Morocco, 3–5 February 2003 (abstract T64)

    Potential conflicts of interest: none reported

    Financial support information appears in the Acknowledgments section

  • Author affiliations and additional ZVITAMBO Study Group members are listed after the text

  • Received March 16, 2005.
  • Accepted September 28, 2005.
Previous Section
 

References

  1. 1.↵
    1. Dushimimana A,
    2. Graham NMH,
    3. Humphrey JH,
    4. et al
    . Program and abstracts of the 8th International Conference on AIDS/3rd STD World Congress (Amsterdam). Amsterdam: International AIDS Society; 1992. Maternal vitamin A levels and HIV‐related birth outcome in Rwanda [abstract Poc4221].
  2. 2.
    1. Semba RD,
    2. Miotti PG,
    3. Chiphangwi JD,
    4. et al
    . Maternal vitamin A deficiency and mother‐to‐child transmission of HIV‐1. Lancet 1994;343:1593-7.
    CrossRefMedlineWeb of Science
  3. 3.↵
    1. Greenberg BL,
    2. Semba RD,
    3. Vink PE,
    4. et al
    . Vitamin A deficiency and maternal‐infant transmission of HIV in two metropolitan areas in the United States. AIDS 1997;11:325-32.
    CrossRefMedlineWeb of Science
  4. 4.↵
    1. Nduati RW,
    2. John GC,
    3. Richardson BA,
    4. et al
    . Human immunodeficiency virus type 1–infected cells in breast milk: association with immunosuppression and vitamin A deficiency. J Infect Dis 1995;172:1461-8.
    Abstract/FREE Full Text
  5. 5.↵
    1. Ross AC
    . Sommer A, West KP Jr. Vitamin A deficiency: health, survival, and vision. New York: Oxford University Press; 1996. The relationship between immunocompetence and vitamin A status; p. 251-73.
  6. 6.↵
    1. Semba RD
    . Overview of the potential role of vitamin A in mother to child transmission of HIV. Acta Paediatr Suppl 1997;421:107-12.
    Medline
  7. 7.↵
    1. Fawzi W
    . Nutritional factors and vertical transmission of HIV‐1: epidemiology and potential mechanisms. Ann N Y Acad Sci 2000;918:99-114.
    CrossRefMedlineWeb of Science
  8. 8.↵
    1. Fawzi WW,
    2. Msamanga GI,
    3. Spiegelman D,
    4. Urassa EJ,
    5. Hunter DJ
    . Rationale and design of the Tanzania vitamin and HIV infection trial. Control Clin Trials 1999;20:75-90.
    CrossRefMedlineWeb of Science
  9. 9.↵
    1. Coutsoudis A,
    2. Pillay K,
    3. Spooner E,
    4. Kuhn L,
    5. Coovadia HM
    . Randomized trial testing the effect of vitamin A supplementation on pregnancy outcomes and early mother‐to‐child HIV‐1 transmission in Durban, South Africa. AIDS 1999;13:1517-24.
    CrossRefMedlineWeb of Science
  10. 10.↵
    1. Kumwenda N,
    2. Miotti PG,
    3. Taha TE,
    4. et al
    . Antenatal vitamin A supplementation increases birth weight and decreases anemia among infants born to human immunodeficiency virus–infected women in Malawi. Clin Infect Dis 2002;35:618-24.
    Abstract/FREE Full Text
  11. 11.↵
    1. Fawzi WW,
    2. Msamanga G,
    3. Hunter D,
    4. et al
    . Randomized trial of vitamin supplements in relation to vertical transmission of HIV‐1 in Tanzania. J Acquir Immune Defic Syndr 2000;23:246-54.
    CrossRefMedlineWeb of Science
  12. 12.↵
    1. Fawzi WW,
    2. Msamanga GI,
    3. Hunter D,
    4. et al
    . Randomized trial of vitamin supplements in relation to transmission of HIV‐1 through breastfeeding and early child mortality. AIDS 2002;16:1935-44.
    CrossRefMedlineWeb of Science
  13. 13.↵
    1. WHO/UNICEF/IVACG Task Force
    . Vitamin A supplements: a guide to their use in the treatment and prevention of vitamin A deficiency and xerophthalmia. Geneva: WHO; 1998.
  14. 14.↵
    1. Stoltzfus RJ,
    2. Hakimi M,
    3. Miller KW,
    4. et al
    . High dose vitamin A supplementation of breast feeding Indonesian mothers: effects on the vitamin A status of mother and infant. J Nutr 1993;123:666-75.
    Abstract/FREE Full Text
  15. 15.↵
    1. Rice AL,
    2. Stoltzfus RJ,
    3. de Francisco A,
    4. Chakraborty J,
    5. Kjolhede CL,
    6. Wahed MA
    . Maternal vitamin A or β‐carotene supplementation in lactating Bangladeshi women benefits mothers and infants but does not prevent subclinical deficiency. J Nutr 1999;129:356-65.
    Abstract/FREE Full Text
  16. 16.↵
    1. Basu S,
    2. Sengupta B,
    3. Roy Paladhi PK
    . Single megadose vitamin A supplementation of Indian mothers and morbidity in breastfed young infants. Postgrad Med J 2003;79:397-402.
    Abstract/FREE Full Text
  17. 17.↵
    1. de Benoist B,
    2. Martines J,
    3. Goodman T
    . Vitamin A supplementation and the control of vitamin A deficiency: conclusions. Food Nutr Bull 2001;22:335-7.
  18. 18.↵
    1. Ross DA
    . Recommendations for vitamin A supplementation. J Nutr 2002;132:2902S-6S.
    Abstract/FREE Full Text
  19. 19.↵
    1. Humphrey JH,
    2. Agoestina T,
    3. Wu L,
    4. et al
    . Impact of neonatal vitamin A supplementation on infant morbidity and mortality. J Pediatr 1996;128:489-96.
    CrossRefMedlineWeb of Science
  20. 20.↵
    1. Rahmathullah L,
    2. Tielsch JM,
    3. Thulasiraj RD,
    4. et al
    . Impact of supplementing newborn infants with vitamin A on early infant mortality: community‐based randomized trial in southern India. BMJ 2003;327:254-7.
    Abstract/FREE Full Text
  21. 21.↵
    1. Beaton GH,
    2. Martorell R,
    3. Aronson KJ,
    4. et al
    . Effectiveness of vitamin A supplementation in the control of young child morbidity and mortality in developing countries. Standing Committee on Nutrition of the United Nations Administrative Committee (ACC/SCN) State of the Art Series Nutrition Policy Discussion Paper no. 13. Geneva: World Health Organization; 1993.
  22. 22.↵
    1. Coutsoudis A,
    2. Bobat RA,
    3. Coovadia HM,
    4. Kuhn L,
    5. Tsai W‐Y,
    6. Stein ZA
    . The effects of vitamin A supplementation on the morbidity of children born to HIV‐infected women. Am J Public Health 1995;85:1076-81.
    Abstract/FREE Full Text
  23. 23.↵
    1. Semba RD,
    2. Ndugwa C,
    3. Perry RT,
    4. et al
    . Effect of periodic vitamin A supplementation on mortality and morbidity of human immunodeficiency virus‐infected children in Uganda: controlled clinical trial. Nutrition 2005;21:25-31.
    CrossRefMedlineWeb of Science
  24. 24.↵
    1. Hussey G,
    2. Hughes J,
    3. Potgieter S,
    4. et al
    . Program and abstracts of the 17th International Vitamin A Consultative Group meeting (Guatemala City, Guatemala). Washington, DC: International Life Sciences Institute; 1996. Vitamin A status and supplementation and its effect on immunity in children with AIDS; p. 81.
  25. 25.↵
    1. Fawzi WW,
    2. Mbise RL,
    3. Hertzmark E,
    4. et al
    . A randomized trial of vitamin A supplements in relation to mortality among human immunodeficiency virus‐infected and uninfected children in Tanzania. Pediatr Infect Dis J 1999;18:127-33.
    CrossRefMedlineWeb of Science
  26. 26.↵
    1. Ministry of Health and Child Welfare
    . Report of the national survey on vitamin A, iron, and iodine. Harare, Zimbabwe: Ministry of Health and Child Welfare, 11–31 October 1999.
  27. 27.↵
    1. Ministry of Health and Child Welfare
    . Zimbabwe National HIV and AIDS Estimates. Harare, Zimbabwe: Ministry of Health and Child Welfare; 2003.
  28. 28.↵
    1. Malaba L,
    2. Iliff PJ,
    3. Nathoo KJ,
    4. et al
    . Effect of postpartum maternal or neonatal vitamin A supplementation on infant mortality among infants born to HIV‐negative mothers in Zimbabwe. Am J Clin Nutr 2005;81:454-60.
    Abstract/FREE Full Text
  29. 29.↵
    1. Iliff PJ,
    2. Humphrey JH,
    3. Mahomva AI,
    4. et al
    . Tolerance of large doses of vitamin A given to mothers and their babies shortly after delivery. Nutr Res 1999;19:1437-46.
    CrossRefWeb of Science
  30. 30.↵
    1. Miller MF,
    2. Stoltzfus RJ,
    3. Mbuya NV,
    4. et al
    . Total body iron in HIV‐positive and HIV‐negative Zimbabwean newborns strongly predicts anemia throughout infancy and is predicted by maternal hemoglobin concentration. J Nutr 2003;133:3461-8.
    Abstract/FREE Full Text
  31. 31.↵
    1. Capurro H,
    2. Konichezky S,
    3. Fonseca D,
    4. Caldeyro‐Barcia R
    . A simplified method for diagnosis of gestational age in the newborn infant. J Pediatr 1978;93:120-2.
    MedlineWeb of Science
  32. 32.↵
    1. Gibson RS
    . Principles of nutritional assessment. New York: Oxford University Press; 1990.
  33. 33.↵
    1. Piwoz EG,
    2. Iliff PJ,
    3. Tavengwa NV,
    4. et al
    . An education and counselling program for preventing breastfeeding‐associated HIV transmission in Zimbabwe: design and impact on maternal knowledge and behaviour. J Nutr 2005;135:950-5.
    Abstract/FREE Full Text
  34. 34.↵
    1. Gavin L,
    2. Tavengwa NV,
    3. Iliff PJ,
    4. Humphrey JH,
    5. Nathoo KJ
    . The development of an intervention to inform women about infant feeding in the context of HIV [poster 269]. Program and abstracts of the 2nd Conference on Global Strategies for the Prevention of HIV Transmission from Mothers to Infants (Montreal) 1999.
  35. 35.↵
    1. Alioum A,
    2. Dabis F,
    3. Dequae‐Merchadou L,
    4. et al
    . Estimating the efficacy of interventions to prevent mother‐to‐child transmission of HIV in breast‐feeding populations: development of a consensus methodology. Stat Med 2001;20:3539-56.
    CrossRefMedlineWeb of Science
  36. 36.↵
    1. Turnbull B
    . The empirical distribution function with arbitrarily grouped, censored, and truncated data. J Royal Statist Soc Ser B 1976;38:290-5.
  37. 37.↵
    1. Hudgens MG,
    2. Satten GA,
    3. Longini IM Jr
    . Nonparametric maximum likelihood estimation for competing risks survival data subject to interval censoring and truncation. Biometrics 2001;57:74-80.
    CrossRefMedlineWeb of Science
  38. 38.↵
    1. WHO
    . The measurement of overall and cause‐specific mortality in infants and children: report of a joint WHO/UNICEF consultation, 15–17 December 1992. Geneva: WHO; 1992.
  39. 39.↵
    1. Hammer ES,
    2. Gibb D,
    3. Havlir D,
    4. Mofenson L,
    5. Beek IV,
    6. Vella S
    . Scaling up antiretroviral therapy in resource‐limited settings: guidelines for a public health approach. Geneva: WHO; 2002.
  40. 40.↵
    1. Christian P,
    2. West KP Jr,
    3. Khatry SK,
    4. et al
    . Night blindness during pregnancy and subsequent mortality among women in Nepal: effects of vitamin A and beta‐carotene supplementation. Am J Epidemiol 2000;152:542-7.
    Abstract/FREE Full Text
  41. 41.↵
    1. Sporn MB,
    2. Roberts AB
    . Interactions of retinoids and transforming growth factor‐beta in regulation of cell differentiation and proliferation. Mol Endocrinol 1991;5:3-7.
    Abstract/FREE Full Text
  42. 42.↵
    1. Hanley TM,
    2. Kiefer HL,
    3. Schnitzler AC,
    4. Marcello JE,
    5. Viglianti GA
    . Retinoid‐dependent restriction of human immunodeficiency virus type 1 replication in monocytes/macrophages. J Virol 2004;78:2819-30.
    Abstract/FREE Full Text
  43. 43.
    1. Kiefer HL,
    2. Hanley TM,
    3. Marcello JE,
    4. Karthik AG,
    5. Viglianti GA
    . Retinoic acid inhibition of chromatin remodeling at the human immunodeficiency virus type 1 promoter. J Biol Chem 2004;279:43604-13.
    Abstract/FREE Full Text
  44. 44.
    1. Brown XQ,
    2. Hanley TM,
    3. Viglianti GA
    . Interleukin 1β and interleukin 6 potentiate retinoic acid‐mediated repression of human immunodeficiency virus type 1 replication in macrophages. AIDS Res Hum Retroviruses 2002;18:649-56.
    CrossRefMedlineWeb of Science
  45. 45.↵
    1. Poli G,
    2. Kinter AL,
    3. Justement JS,
    4. Bressler P,
    5. Kehrl JH,
    6. Fauci AS
    . Retinoic acid mimics transforming growth factor β in the regulation of human immunodeficiency virus expression in monocytic cells. Proc Natl Acad Sci USA 1992;89:2689-93.
    Abstract/FREE Full Text
  46. 46.
    1. Maciaszek JW,
    2. Coniglio SJ,
    3. Talmage DA,
    4. Viglianti GA
    . Retinoid‐induced repression of human immunodeficiency virus type 1 core promoter activity inhibits virus replication. J Virol 1998;72:5862-9.
    Abstract/FREE Full Text
  47. 47.
    1. Towers G,
    2. Harris J,
    3. Lang G,
    4. Collins MK,
    5. Latchman DS
    . Retinoic acid inhibits both the basal activity and phorbol ester‐mediated activation of the HIV long terminal repeat promoter. AIDS 1995;9:129-36.
    MedlineWeb of Science
  48. 48.↵
    1. Yamaguchi K,
    2. Groopman JE,
    3. Byrn RA
    . The regulation of HIV by retinoic acid correlates with cellular expression of the retinoic acid receptors. AIDS 1994;8:1675-82.
    MedlineWeb of Science
  49. 49.↵
    1. Kitano K,
    2. Baldwin GC,
    3. Raines MA,
    4. Golde DW
    . Differentiating agents facilitate infection of myeloid leukemia cell lines by monocytotropic HIV‐1 strains. Blood 1990;76:1980-8.
    Abstract/FREE Full Text
  50. 50.↵
    1. Turpin JA,
    2. Vargo M,
    3. Meltzer MS
    . Enhanced HIV‐1 replication in retinoid‐treated monocytes: retinoid effects mediated through mechanisms related to cell differentiation and to a direct transcriptional action on viral gene expression. J Immunol 1992;148:2539-46.
    Abstract
  51. 51.↵
    1. West KP Jr,
    2. Katz J,
    3. Khatry SK,
    4. et al
    . Double blind, cluster randomised trial of low dose supplementation with vitamin A or β‐carotene on mortality related to pregnancy in Nepal. BMJ 1999;318:570-5.
    Abstract/FREE Full Text
  52. 52.↵
    1. Sommer A
    . Innocenti Micronutrient Research Report no. 1. International Vitamin A Consultative Group; 2005. Available at: http://ivacg.ilsi.org. Accessed 13 September 2005.
| Table of Contents

This Article

  1. J Infect Dis. (2006) 193 (6): 860-871. doi: 10.1086/500366
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free
  4. A correction has been published

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most