Multivitamin supplementation improves haematologic status in children born to HIV-positive women in Tanzania

Introduction: Anaemia is prevalent among children born to HIV-positive women, and it is associated with adverse effects on cognitive and motor development, growth, and increased risks of morbidity and mortality. Objective: To examine the effect of daily multivitamin supplementation on haematologic status and mother-to-child transmission (MTCT) of HIV through breastfeeding. Methods: A total of 2387 infants born to HIV-positive women from Dar es Salaam, Tanzania were enrolled in a randomized, double-blind, placebo-controlled trial, and provided a daily oral supplement of multivitamins (vitamin B complex, C and E) or placebo at age 6 weeks for 24 months. Among them, 2008 infants provided blood samples and had haemoglobin concentrations measured at baseline and during a follow-up period. Anaemia was defined as haemoglobin concentrations <11 g/dL and severe anaemia <8.5 g/dL. Results: Haemoglobin concentrations among children in the treatment group were significantly higher than those in the placebo group at 12 (9.77 vs. 9.64 g/dL, p = 0.03), 18 (9.76 vs. 9.57 g/dL, p = 0.004), and 24 months (9.93 vs. 9.75 g/dL, p = 0.02) of follow-up. Compared to those in the placebo group, children in the treatment group had a 12% lower risk of anaemia (hazard ratio (HR): 0.88; 95% CI: 0.79-0.99; p = 0.03). The treatment was associated with a 28% reduced risk of severe anaemia among children born to women without anaemia (HR: 0.72; 95% CI: 0.56-0.92; p = 0.008), but not among those born to women with anaemia (HR: 1.10; 95% CI: 0.79-1.54; p = 0.57; p for interaction = 0.007). One thousand seven hundred fifty three infants who tested HIV-negative at baseline and had HIV testing during follow-up were included in the analysis for MTCT of HIV. No association was found between multivitamin supplements and MTCT of HIV. Conclusions: Multivitamin supplements improve haematologic status among children born to HIV-positive women. Further trials focusing on anaemia among HIV-exposed children are warranted in the context of antiretroviral therapy. © 2013 Liu E et al;. From February 2004 to June 2007, pregnant women aged 18 years or older presenting for prenatal care at the 32nd week of gestation or earlier in one of the eight clinics in Dar es Salaam, were offered HIV screening with pre- and post-test counselling. HIV-positive women were followed up to enrol their infants into the randomized clinical trial. Infant eligibility criteria included born to HIV-positive women, singleton birth, age 5–7 weeks, and the mother intending to reside in Dar es Salaam for at least two years after delivery. We excluded from the study twin infants and infants with multiple or serious congenital anomalies (such as cyanotic congenital heart disease, spinal bifida) or other medical conditions that would interfere with their ability to comply with the study procedures. Eligible singleton birth infants were randomly assigned to receive a daily oral dose of multivitamins or placebo at age of six weeks (baseline) for 24 months. A randomization list from 1 to 2400 was prepared according to a computer-generated sequence in a block of 20. The randomization list was provided to the pharmacy department, with each a number corresponding to a code denoting one of the two treatments. Infants enrolled at the study clinic were provided the next consecutive number in the series when they received the supply of daily regimen from the pharmacist. From age six weeks to six months, children in the multivitamin treatment group received one capsule containing 60-mg vitamin C, 8-mg vitamin E, 0.5-mg thiamine, 0.6-mg riboflavin, 4-mg niacin, 0.6 mg vitamin B6, 130-µg folate and 1-mg vitamin B12. For children older than six months, two capsules were administered daily. These doses represent 150–600% and 200–400% of US adequate intake for children aged 0–6 and 7–12 months, respectively, and 133–800% of the US Recommended Dietary Allowance (RDA) for children aged 1–3 years. For these nutrients, tolerable upper intake levels have not been defined for children<12 months; the tolerable upper intake levels (if defined) were not exceeded for children≥12 months [13–15]. We did not include vitamin A in our treatment regimen because vitamin A supplementation was associated with increased risk of MTCT through breastfeeding [11, 16]. In addition, all children in this study received periodic large doses of vitamin A supplements (100,000 IU at nine months and 200,000 IU at 15 and 21 months) as per Tanzania Ministry of Health guidelines. The multivitamin supplement used was powder encapsulated in an opaque gelatinous capsule, which was manufactured by Nutriset (Malaunay, France). Mothers were instructed on how to push the capsule through the back of the blister pack, open it, and decant the powder into a small plastic cup. Sterile water (5 mL) supplied with the supplement was added to the powder, and the dose was given to the infants orally. A pilot phase of open-label vitamin use in 12 infants and mothers confirmed that this supplement preparation and use was well-accepted by the mothers and infants. Both the placebo and active capsules contained an orange tasting powder that allowed for identical taste and appearance. All study personnel and participants were blinded to treatment assignment for the duration of the study. Only the study statisticians and the data-monitoring committee saw unblinded data, but none had any contact with study participants. Mothers and their children were asked to return to the clinic every month for research visits and standard clinical care. Mothers were counselled on the risks and benefits of exclusive breastfeeding in keeping with WHO recommendations in place during the study period. At each clinic visit, women were asked about infant feeding in the past seven days. Information on the infant feeding during the first week of life was requested at the first post-partum visit. During monthly follow-up, infant-feeding practices (breastfeeding status and frequency, introduction of other liquid or foods including water, tea, juice, cow's milk, infant formula, porridge, mashed vegetable, meats, rice) was collected. We defined exclusive breastfeeding as feeding a child with breast milk only without additional foods. The duration of exclusive breastfeeding was calculated as the mean of infant ages at which the last time the mother reported that the child was still exclusively breastfeeding and the first time the mother reported that the child was given other foods in addition to breast milk. Compliance with the daily regimen was measured by pill counting by research nurses of unused regimen. When the study began in 2004, routine medical care proposed for pregnant women with HIV infection included malaria prophylaxis, diagnosis and treatment for sexually transmitted diseases and prophylaxis, diagnosis and treatment of opportunistic infections, and iron and folate supplements during pregnancy. One dose of nevirapine was given to the mother at the onset of labour and another dose given to the infant within 72 hours of birth for the purpose of prevention of MTCT [17]. As the study progressed, beginning in July 2005, the availability of antiretroviral drugs increased substantially through programmes including the President's Plan for AIDS Relief (PEPFAR) and other governmental and non-governmental programmes. Women and children in the study were screened for ART eligibility and treated according to Tanzanian Ministry of Health guidelines. Based on the previous findings on the beneficial effects of vitamin B complex, C and E for pregnancy outcomes and to slow HIV disease progression [12], all women received multivitamin supplementation (vitamin B complex, C and E) during and after pregnancy. Maternal HIV-1 serostatus was determined by two sequential enzyme-linked immunosorbent assay (ELISA) using Murex HIV antigen/antibody (Abbott Murex, UK) followed by the Enzygnost anti-HIV-1/2 Plus (Dade Behring, Marburg, Germany); discordant results were resolved by a Western blot test (Bio-Rad Laboratories, Hertfordshire, UK). All children were tested for HIV infection at baseline using the Amplicor HIV-1 DNA assay version 1.5 (Roche Molecular Systems, Inc., Branchburg, NJ, USA) and then again at 18 months using HIV ELISAs. Samples from children who tested positive at 18 months were then backtested using the Amplicor HIV-1 DNA assay version 1.5 to estimate the time of transmission. Children who were HIV-negative at 18 months and still breastfed were tested again before they were discharged from the study after 24 months of follow-up. Blood specimens were requested from each mother at baseline to measure complete blood counts including haemoglobin concentrations and T-cell subset counts. Complete blood counts, haemoglobin concentrations, and T-cell subset counts were also measured for children at baseline and every six months thereafter, until the end of the follow-up. Haemoglobin concentrations were measured using AcT5 Diff AL haematology analyzer (Beckman Coulter, Jersey City, NJ, USA) and T-cell subset, for example absolute CD4+cell count and percentage were performed with the FACSCalibur system (Becton-Dickinson, San Jose, CA, USA). The primary outcomes of the randomized controlled trial were mortality and morbidity, and the main results have been reported [13]. Assuming a mortality rate of 12.5% in the placebo group, we planned to enrol 2360 infants to detect a 30% reduction in mortality in the treatment group with 80% power at a significant level of 0.05 [13]. Assuming haemoglobin concentrations are measured every six months, the current analysis provided 80% power to detect a difference of 0.06 g/dL between the treatment and placebo groups after 24 months of follow-up. Assuming 10% infants are HIV-positive at baseline, the data provided 80% power to detect a maximum relative risk of 0.6 for MTCT of HIV during the follow-up period. Data were double-entered and validated using Microsoft Access software. The final data sets were converted into SAS software and uploaded to a UNIX-based server in Boston, MA. In this article, we aimed to analyze the secondary outcomes including MTCT of HIV, child haemoglobin concentrations, and development of anaemia. Logistic regression model was used to examine the association between multivitamin supplementation and the risk of MTCT through breastfeeding. Children who tested HIV-negative at baseline and who had HIV testing during follow-up were included in the analysis for MTCT (n=1753). t-Tests were used to evaluate the effect of multivitamin supplements on haemoglobin concentrations. Cox proportional hazard models were used to examine the effect of the multivitamin supplements on the risk of development of anaemia and severe anaemia during follow-up. Children with haemoglobin measured at baseline and at least one follow-up haemoglobin measure were included in the analysis for haemoglobin concentrations (n=2008) (Figure 1). We further excluded children who had anaemia or severe anaemia at baseline from the respective analyses conducted to determine the risk of developing anaemia and severe anaemia during the follow-up period. Baseline measures were those obtained within six weeks after randomization. Anaemia was defined as haemoglobin levels<10.0 g/dL at baseline (six weeks of age) and<11 g/dL during follow-up, and severe anaemia was defined as haemoglobin<8.5 g/dL [18]. We classified anaemia as microcytic, normocytic, or macrocytic using mean corpuscular volume (MCV). The normal range of MCV was defined as 70–86 fL for children at the age of 6–24 months [18]. In the multivariate analyses, we considered potential confounders and independent risk factors for anaemia from a list of candidate variables. Covariates including maternal age (≤28, >28 years), CD4+ counts (<200, 200– <350, ≥350 cells/mm3), WHO HIV disease stage (I/II, III/IV), antiretroviral therapy during pregnancy (yes/no), sex of child (male/female), birth weight (<2.5 kg, ≥2.5 kg), preterm birth (<37, ≥37 weeks), HIV status (negative/positive), were adjusted for in the models. Likelihood ratio tests were used to test the interaction between multivitamin supplements and potential effect modifiers. Sample sizes for analyses of haemoglobin (HB), anaemia (HB<11 g/dL), and severe anaemia (HB<8.5 g/dL). All analyses were performed using the SAS software Version 9.1 (SAS Institute, Cary, NC, USA). The significance tests were two-sided and a p-value less than 0.05 was considered statistically significant. The Harvard School of Public Health Human Subjects Committee, and the Muhimbili University of Health and Allied Sciences Research and Publications Committee granted institutional review board approval. Written informed consent was obtained from women for HIV testing and their infant's participation in the trial. The trial was registered at clinicaltrials.gov (identifier {"type":"clinical-trial","attrs":{"text":"NCT00197730","term_id":"NCT00197730"}}NCT00197730).