In resource-poor settings, the WHO recommends that HIV-infected women exclusively breastfeed for 6 months and continue breastfeeding to 12 months.1 In this population, antiretrovirals (ARVs) are provided to the mother or infant to prevent mother-to-child transmission of HIV (PMTCT) if replacement feedings are not acceptable, feasible, affordable, sustainable, and safe.1 HIV-infected women are at risk of impaired iron status during pregnancy and postpartum due to heightened iron demands in this period coupled with the demands of the HIV infection.2–5 Given the strong influence of maternal iron status on infants' iron endowment at birth and thus subsequent iron status,6 infants born to HIV-infected mothers are at high risk of iron deficiency.7,8
Some prenatally administered ARVs, especially zidovudine, are associated with maternal anemia9 and severe infant anemia postpartum.10 This is in contrast to findings in nonpregnant adult populations, where initiation of highly active antiretroviral therapy (HAART) is associated with increases in hemoglobin (Hb).11–13 Although some studies have shown that single-dose infant nevirapine may have transient effects on infant iron status,14,15 extended infant nevirapine regimens do not seem to influence short- and long-term risk of anemia.16,17 However, data regarding the effects of extended postpartum PMTCT regimens on maternal and infant iron status are limited and no studies to date have reported results among mothers also receiving nutritional supplementation.
The Breastfeeding, Antiretrovirals, and Nutrition (BAN) study was a randomized controlled trial designed to test interventions for PMTCT.17 Mother–infant pairs were randomized with a two-by-three factorial design to one of six 28-week treatment assignments: 3 ARV groups (maternal ARV, infant nevirapine or no extended postnatal ARV) and 2 maternal nutritional intervention groups [lipid-based nutrient supplements (LNS) or no LNS]. Previously, we reported that the proportion of low Hb (grade 3 or 4 adverse events) in both mothers and infants did not differ by ARV group; however, this analysis did not evaluate effects of LNS.17 This secondary analysis explores the effects of the 6 treatments on (1) maternal and infant Hb longitudinally during exclusive breastfeeding and in a subsample with multiple iron indicators and (2) maternal and infant ferritin, transferrin receptors (TfR), and Hb adjusted for the acute-phase response. We hypothesized that ARVs would be associated with worsening maternal and infant iron indicators, and that LNS would be associated with improved maternal iron indicators. We did not expect to observe LNS effects in the infants because previous evidence suggests that maternal iron supplementation during breastfeeding does not influence infant iron status.18
Data are from the BAN Study (clinicaltrials.gov number NCT00164736), conducted from 2004 to 2010, whose design19 and primary intervention findings have been previously reported.17,20–22 Briefly, HIV-1–infected pregnant women (n = 3572) were recruited from antenatal clinics in Lilongwe, Malawi. Primary eligibility criteria for initial enrollment included gestational age ≤30 weeks, CD4 ≥250 cells per microliter (CD4 ≥200 cells per microliter before July 2006), and Hb ≥70 g/L. After delivery, secondary eligibility criteria for randomization included infant birth weight ≥2 kg, and no previous ARV use.19 Infants diagnosed with HIV-1 within 2 weeks of delivery were withdrawn from the study and referred for care.17
The Malawi National Health Science Research Committee and the institutional review boards at the University of North Carolina at Chapel Hill and the U.S. Centers for Disease Control and Prevention provided ethical approval for the study, and the institutional review board at the University of California, Davis approved the laboratory analyses at the Western Human Nutrition Research Center. Written informed consent was obtained from all study mothers.
At delivery, mother–infant dyads (n = 2369) were randomized to one of six 28-week treatment arms: maternal LNS/maternal ARV (mLNS-mARV), maternal LNS/infant ARV (mLNS-iARV), maternal LNS only (mLNS), maternal ARV only (mARV), infant ARV only (iARV), or control.
All mothers received daily iron–folic acid supplementation containing 40 mg elemental iron and 0.25 mg folic acid from the initial screening to 1 week postpartum. All mothers and infants also received an intrapartum single dose of nevirapine and twice-daily zidovudine and lamivudine for 7 days after delivery.14
Mothers assigned to the nutritional intervention were given 140 g of LNS per day, providing 746 kcal, 20.8 g protein and 15 mg of elemental iron (see Table S1, Supplemental Digital Content, http://links.lww.com/QAI/A651).22 To prevent sharing of LNS, all mothers were provided maize for family consumption.22 The maternal ARV regimen initially consisted of zidovudine, lamivudine, and nevirapine. Nelfinavir replaced nevirapine after the first 39 women were randomized, and lopinavir/ritonavir replaced nelfinavir shortly thereafter, with most (>75%) mothers randomized to ARV receiving zidovudine/lamivudine/lopinavir–ritonavir. Infants randomized to ARV received daily oral nevirapine (details previously reported).17 Starting on June 13, 2006, cotrimoxazole prophylaxis (CPT) (240 mg once daily) was provided to all infants aged >6 weeks and initiated for all mothers with a CD4 <500 cells per microliter in pregnancy23; in this report, 68% and 43% of infants and mothers received CPT, respectively. In March 2008, the data safety monitoring board at the National Institute of Allergy and Infectious Diseases reviewed data from the enrolled 1857 mother–infant pairs and observed that HIV transmission through breast milk was higher in the no-drug study arms.24 Enrollment was halted in these groups, and mothers in these arms were allowed to choose to switch to the maternal or infant ARVs, or remain in the control group.24
Mothers were provided additional iron supplementation as clinically indicated or with ≥moderate anemia (Hb ≤84 g/L25). Records of additional iron supplementation were abstracted from medication, severe adverse event, and morbidity forms, linked to concomitant study visit(s). Few participants (<3%) were given supplemental iron.
Study visits were conducted at the BAN Study clinic in Lilongwe, Malawi, at birth, 1, 2, 4, 6, 8, 12, 18, 21, and 24 weeks of postpartum. This analysis includes data from 0 to 24 weeks, when mothers were provided with intensive counseling to exclusively breastfeed.26,27 Maternal Hb was measured at prenatal screening and maternal and infant Hb (g/L) was measured at birth, 2, 6, 12, 18, and 24 weeks of postpartum from venous blood samples with a Beckman Coulter AcT or AcT 5-part Differential Analyzer (Beckman Coulter, Fullerton, CA). Infant HIV status was tested with Amplicor 1.5 DNA polymerase chain reaction. Measurements included maternal and infant weights with regularly calibrated Tanita digital scales, infant length with a recumbent length board, and maternal height with a wall-mounted stadiometer. At screening, maternal report of parity, marital status, and education was obtained. Exclusive breastfeeding was obtained by maternal report.
Plasma for additional biomarker assays was separated from red blood cells, aliquoted to 1-mL polypropylene storage tubes, and stored at −70°C. Briefly, subsample laboratory analyses used plasma obtained at either 2 or 6 weeks, as many infants had insufficient plasma at 2 weeks, and 24 weeks postpartum. Transferrin receptors (TfR) and inflammatory marker [C-reactive protein (CRP) and α-1-acid glycoprotein (AGP)] concentrations were measured using Cobas Integra 400 (Roche Diagnostics, Indianapolis, IN).28 Ferritin concentrations were measured with IRMA Ferritin Coat a Count radioimmunoassay (Siemens Health Care Diagnostics Inc., Plainfield, IN).28
The primary outcomes for this analysis are maternal and infant Hb, ferritin, and TfR. The primary independent variables are the study intervention arms. For the maternal outcomes, 4 intervention groups were evaluated; mothers of infants receiving ARVs (mLNS-iARV and iARV) were included in the mLNS and control groups, respectively (see Figure S1, Supplemental Digital Content, http://links.lww.com/QAI/A651). Mother–infant pairs were excluded from both analysis samples if the infant was a multiple birth (n = 49), which may influence fetal iron accretion, or HIV-infected from 2 to 24 weeks (n = 57), which may influence iron mobilization and interpretation of iron indicators.29 Observations after breastfeeding cessation (n = 277) were dropped in longitudinal analyses, as cessation would influence maternal nutritional status and the infant would no longer be exposed to maternal ARVs through breast milk. Of the 2369 randomized dyads, 1765 and 1927 were included in the longitudinal maternal and infant analytic samples, respectively (see Figure S2, Supplemental Digital Content, http://links.lww.com/QAI/A651). Subsample dyads (n = 537) were selected with equal representation from the LNS and no-LNS groups, prioritizing those with anthropometry and dietary data and excluding multiple births and HIV-positive infants (see Figure S3, Supplemental Digital Content, http://links.lww.com/QAI/A651).28 Characteristics of included and excluded dyads in each analytic sample were compared for similarity using t tests and nonparametric tests for normally distributed and skewed continuous variables, respectively. Similar tests compared characteristics of those only in the longitudinal sample versus the subsample. Because a small number of mothers in this analysis received additional iron supplementation, we evaluated whether supplementation varied by treatment arm and observed no differences across treatment arms at each visit (all P > 0.19; data not shown). STATA 12.0 (College Station, TX) was used for all statistical analyses. An α of 0.05 denoted statistical significance.
Infants were included in the longitudinal analyses if they had birth weight and Hb at birth, and at least one other Hb measurement. Mothers were included in the longitudinal analysis if they had Hb at 2-week postpartum and at least one other measurement. Because of hematologic dilution at delivery and inflammation of parturition,30 which would distort initial measurements, maternal models used the 2-week Hb value as baseline. Longitudinal random-effects models with first-order autoregressive disturbance terms were used to evaluate the effect of the study interventions on (1) maternal Hb from 6- to 24-week postpartum and (2) infant Hb from 2 to 24 weeks, adjusting for week and initial Hb.31 Infant models were further adjusted for birth weight, growth rate, and sex. In infant models, age was modeled in weeks and included a knot at 9 weeks of age to capture the shape of the infant Hb curve over time. To determine whether intervention effects varied with infant age, interactions of week with covariates were evaluated using Wald tests for joint significances (α = 0.132) and significant interaction terms were retained.
In the subsample, we accounted for the acute-phase response and minimized the effects of the infection (such as malaria and HIV) and inflammation [elevated CRP (>5 mg/L) and AGP (>1 g/L)] on TfR, Hb, and ferritin.28 Stage of inflammation [healthy (normal CRP and AGP), incubation (elevated CRP), early convalescence (CRP and AGP elevated), late convalescence (elevated AGP)], and stage-specific correction factors were determined.28,33,34 Plasma ferritin and TfR had non-Gaussian distributions and were log transformed for analyses. Linear regression was used to evaluate the effects of study interventions on maternal and infant iron status (ferritin and TfR) and Hb. Modified Poisson regression with robust variance estimators was used to estimate relative risk of maternal and infant deficiency or anemia at 24 weeks.35 For all models, sensitivity analyses were conducted to evaluate whether inclusion of an indicator variable for maternal iron supplementation or CPT influenced the study intervention effects.
Most baseline maternal characteristics were well balanced by study arm for the longitudinal sample and the subsample (Table 1). A smaller proportion of control mothers were married or educated beyond primary school in the longitudinal sample (Table 1). Compared with randomized mothers not included in this analysis, mothers in both analytic samples were older, more educated, had higher pregnancy Hb, and lower body mass index at delivery. Compared with mothers only in the longitudinal sample, mothers in the subsample had lower body mass index at delivery (data available in online supplemental materials, http://links.lww.com/QAI/A651).
Maternal Hb increased from baseline to 24 weeks (Fig. 1) and thus prevalence of anemia (Hb <120 g/L, unadjusted for inflammation) decreased. In subsample mothers, prevalence of tissue iron depletion (TfR >8.3 mg/L) declined, whereas prevalence of depleted iron stores (ferritin <15 ng/mL) remained somewhat stable (Table 2).
In the longitudinal analysis, mLNS had no effects on maternal Hb (Table 3). There were some transient effects of maternal ARVs on maternal Hb (Table 3). Based on linear combinations of the beta coefficients for the intervention groups and weeks, maternal Hb was significantly lower at 6 and 12 weeks, but not at subsequent observations, in mothers who received mLNS-mARV [6 weeks—β: −3.21 g/L [95% confidence interval (CI): −4.46 to −1.94], P < 0.001; 12 weeks—β: −1.88 g/L (95% CI: −2.91 to −0.85), P < 0.001] or mARV [6 weeks—β: −2.11 g/L (95% CI: −3.35 to −0.87), P = 0.001; 12 weeks—β: −1.47g/L (95% CI: −2.50 to −0.45), P = 0.005] (Table 3).
In subsample mothers, mLNS was not associated with maternal Hb, TfR, or ferritin, compared with control (all P > 0.1) (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651). mARV and mLNS-mARV were associated with lower Hb at 6 weeks and higher TfR at 6 and 24 weeks (indicative of worsening iron status), but were not associated with ferritin (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651). Although the interventions were not associated with risk of maternal anemia (Hb <120 g/L) or depleted iron stores (ferritin <15 ng/mL) at 24 weeks, mARV was associated with increased risk of tissue iron depletion (TfR >8.3 mg/L) but only in ARV-treated mothers who did not receive LNS (Table 4). Although continuous TfR values were higher in mothers in the mLNS-mARV group at 24 weeks (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651), mothers in this group were not at higher risk of tissue iron depletion compared with controls. In sensitivity analyses, adjustment for additional maternal iron supplementation and CPT had no or negligible changes on intervention effects (data not shown).
Baseline infant characteristics were well balanced across the study arms (Table 1). Compared with other randomized dyads, infants in both analytic samples were heavier at birth and more likely to be in the iARV arms (data available in online supplemental materials, http://links.lww.com/QAI/A651).
Mean infant Hb followed the normal pattern of decline from a high of above 170 g/L at birth to a nadir at about 8 weeks, as senescent fetal Hb lyses, with a subsequent slight increase thereafter (Fig. 1).29 In subsample infants, ferritin values declined from initial measurement to 24 weeks, as values normalized after erythrocyte breakdown, whereas TfR increased. Few infants had impaired iron status based on measurements at birth or 2 or 6 weeks, but at 24 weeks, worsening iron status was apparent (Table 2).
In the longitudinal analytic sample, minimal intervention effects were observed on infant Hb (see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A651). Based on linear combinations of the beta coefficients for each intervention arm and age, mLNS-iARV was associated with 2.8 g/L (P = 0.03) lower Hb at 6 weeks; but was not associated with lower Hb values at later visits [12 weeks—β: −2.51 g/L (95% CI: −5.16 to 0.15), P = 0.06; 18 weeks—β: −0.97 g/L (95% CI: −3.44 to 1.51), P = 0.44; 24 weeks—β: 0.57 g/L (95% CI: −2.25 to 3.39), P = 0.7] (see Table S3, Supplemental Digital Content, http://links.lww.com/QAI/A651). In the subsample infants where Hb was corrected for inflammation, however, some intervention effects on continuous Hb values were observed. Although mLNS had no effect on Hb at 2 or 6 weeks, it was associated with lower infant Hb at 24 weeks (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651). Compared with controls, mLNS-iARV and iARV were associated with lower Hb values at 2 and 24 weeks (β range: 4–5 g/L), whereas a similar but nonsignificant trend was observed at 6 weeks (β range: 4.3–5.3 g/L) (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651).
At 24 weeks, mLNS was associated with lower infant ferritin values (see Table S2, Supplemental Digital Content, http://links.lww.com/QAI/A651). mLNS and mLNS-mARV were associated with higher infant TfR at 2 weeks, but not at other times. The interventions were not associated with risk of infant iron deficiency (inflammation-adjusted ferritin <12 ng/mL; TfR >8.3 mg/L) or anemia (inflammation-adjusted Hb <105 g/L) at 24 weeks (Table 4). In sensitivity analyses, adjustment for additional maternal iron supplementation or infant CPT did not influence intervention effects on infant Hb or iron status (data not shown).
To our knowledge, this is the first study to investigate effects of extended ARVs coupled with maternal nutrition supplementation on maternal and infant iron status during exclusive breastfeeding. Our results suggest that maternal ARV therapy and infant nevirapine prophylaxis were associated with worsening of some maternal and infant iron status indicators. Most notably, maternal ARVs were associated with a 3-fold risk of tissue iron depletion, but this risk was mitigated by mLNS, where adverse effects on tissue iron depletion (TfR levels) were reduced. Although the mLNS contained only 15 mg of iron, this amount was sufficient to mitigate some of the adverse ARV effects on maternal tissue iron status. Interestingly, the interventions did not impact risk of maternal iron store depletion (low ferritin) or anemia. Tissue iron depletion reflects more severe iron deficiency, as TfR levels increase when ferritin values become subnormal (eg, iron stores are depleted); therefore, populations can have depleted iron stores before tissue iron depletion is apparent. Maternal supplementation had few clinically meaningful or sustained effects on infant TfR, Hb, or ferritin, and neither the mLNS nor maternal or infant ARVs were associated with increased risk of infant iron deficiency or anemia at 24 weeks.
Several randomized trials and cohort studies have demonstrated adverse effects of ARVs on maternal or infant iron status, particularly zidovudine-containing regimens9,14,16,36–45; however, none, other than ours,17 evaluated extended postnatal ARV regimens coupled with a nutritional supplement. Furthermore, none of these trials evaluated multiple iron status indicators, nor did they account for the effect of inflammation and infection on iron status. In HIV-affected populations, anemia of inflammation is common, defined as the presence of inflammation without iron deficiency anemia, and initiation of ARV therapy may differentially impact indicators of iron status (increase TfR and Hb); thus, measurement of the acute-phase response to correct iron indicators for concurrent inflammation/infection is recommended for assessment of iron status.46 With inflammation and infection, iron status indicators are distorted; plasma ferritin levels are falsely elevated and do not reflect body iron stores, whereas Hb levels are lowered.46 TfR may be increased with iron deficiency anemia in the presence of inflammation46; but it is believed to be less sensitive to the effects of anemia of inflammation/infection than other indicators. Previously, in BAN, we observed associations between inflammatory markers and TfR, ferritin, and Hb28; therefore, in our subsample analyses, we used methods recommended by Thurnham et al to mitigate the effects of the acute-phase response on iron status indicators.34
This is the first study to use longitudinal data with a true control group to report hematologic effects of maternal ARV exposure on multiple postpartum iron indicators. In a cohort study of pregnant HIV-infected women in Thailand, anemia (Hb <94 g/L) was significantly associated with HAART use compared with ARVs for PMTCT with both regimens containing zidovudine (P = 0.02).9 Another cohort study in Italy compared hematologic outcomes of ARVs in 3 groups: (1) women receiving zidovudine-based HAART from conception, (2) women starting zidovudine-based HAART during pregnancy, and (3) women receiving zidovudine-free HAART from conception.36 Women receiving zidovudine-free HAART had a greater decrease in Hb from baseline to 36-week gestation compared with women who started a zidovudine-containing HAART regimen during pregnancy (−20.3 ± 11.9 g/L vs. −13.6 ± 12.0 g/L, P = 0.04).36 Together, these studies suggest that prenatal maternal Hb may be compromised by some ARV regimens, but there is insufficient evidence that prenatal ARVs always lead to anemia. We did not observe sustained adverse effects of postpartum ARVs on maternal Hb. This may reflect use of different ARV regimens, sample differences (BAN excluded mothers with Hb <70 g/L), or prenatal/postpartum period differences. Although there is evidence that initiation of HAART (with various combinations of drugs) is associated with increased Hb and/or improved anemia in nonpregnant populations,11–13 these results may not be comparable with pregnant/postpartum populations due to heightened iron demands in this period or to relatively healthy HIV-infected pregnant/postpartum women receiving ARVs.
Several studies have indicated that ARV regimens are associated with adverse effects on infant Hb,10,37–45 whereas 1 study has shown some positive effects.15 Infants who received a short-course regimen, including maternal peripartum nevirapine and/or infant zidovudine and/or nevirapine, had higher Hb from birth to 6 months and lower prevalence of deficiency at 6 months.15 However, no differences in deficiency prevalence were observed beyond 6 months.15 Adverse and often transient hematologic effects of zidovudine-containing ARV regimens during pregnancy have been reported in numerous cohorts,10,38–43 with many of these cohorts reporting severe anemia in HAART-exposed infants.10,41 In addition, postnatal zidovudine exposure has been associated with lower infant Hb and/or increased infant anemia10,37,41,43,44 and may have long-lasting adverse effects on infant Hb up to 18 months postpartum.45 Similar to previous reports, we observed transient effects of maternal and infant ARVs on some infant Hb and TfR; however, we did not observe increased anemia or iron deficiency, which may be attributable to sample differences related to BAN inclusion criteria, type and timing of initiation of ARVs (infants in BAN were not exposed to ARVs in utero), and close follow-up, as well as our report of additional iron indicators and adjustment for inflammation.
While no prior studies reported effects of maternal LNS on maternal or infant iron status, 1 study examined the effects of multivitamin supplementation during pregnancy and postpartum on maternal and infant iron status in the context of HIV.36 In a Tanzanian cohort of ARV-naive and predominately anemic HIV-infected women who received standard iron and folic acid supplementation during pregnancy (containing 120 mg iron and 5 mg folic acid), maternal multivitamin supplementation, containing no iron, starting prenatally and continuing through lactation was associated with 8.8 g/L (P = 0.0002) higher maternal Hb at ≤70 days postpartum compared with placebo.47 The authors speculated that vitamins in the supplement, especially vitamin C, may have contributed to better iron status and enhanced absorption of dietary iron.47 Although we did not observe effects of this magnitude in our mothers, who started receiving supplementation postpartum, mLNS was associated with reduced tissue iron depletion in mother's receiving ARVs. Infant Hb outcomes in this Tanzanian population were not significantly different between supplementation groups at 6 months, infant Hb was 3.1 g/L and 2.9 g/L higher at 2 and 4 years, respectively, for infants whose mothers received supplementation compared with placebo.47 In our study at 24 weeks (approximately 6 months), maternal LNS, with and without infant ARV therapy (mLNS and mLNS-iARV), was associated with approximately 4–5 g/L lower infant Hb at 24 weeks, which contradicts the 6-month findings in the Tanzanian sample.47 These differences may be due to varying time periods of supplementation (LNS in our sample was initiated postpartum) or our ability to account for the effects of inflammation/infection on Hb values.
There are several limitations to this work. Initial subsample measurements were obtained at 2- or 6-week postpartum due to insufficient plasma at 2 weeks. This reflects varying periods of exposure to the interventions; therefore, we evaluated for intervention effects separately at each visit in the subsample analyses. Compared with all randomized infants, included infants were healthier and worse off infants (including those with HIV infection) and mothers were lost to follow-up or excluded from analyses. As such, our ability to detect intervention effects may be reduced. Some mothers received iron supplementation when clinically indicated; however, sensitivity analyses showed that this did not change our interpretation of intervention effects. Finally, our findings may not be generalizable to other populations of HIV-infected women and their HIV-exposed infants, as mothers with low CD4 counts and previous ARV exposure were excluded from BAN and also due to our selection criteria for longitudinal analysis. Dyads were closely followed by study staff and promptly treated for adverse events and other illnesses.
The WHO now recommends that HIV-infected mothers or their uninfected infants receive ARV prophylaxis during 12 months of breastfeeding.1 Given these revised recommendations, the nutritional demands of lactation and HIV,48 and the adoption of lifelong ARV therapy for pregnant and lactating HIV-infected women (option B+), maternal LNS or possibly another iron supplementation method (eg, iron–folic acid) may be valuable to support lactation and prevent impaired iron status and associated consequences such as elevated risk of maternal or infant mortality and maternal HIV-disease progression.5,7,8 However, in settings with endemic malaria, iron supplementation may be associated with increased risk of malaria49,50; thus, supplementation should be coupled with malaria prevention, diagnosis, and treatment efforts. Future randomized controlled trials are needed to examine the effects of ARV regimens beyond 6 months and to establish the optimal composition of supplementation to minimize possible adverse effects of extended ARV regimens. If adverse effects of extended regimens on iron status are observed in clinical trials, it may be possible to provide supplementation along with ARV therapy to support maternal health and micronutrient status through option B+ programming. Further evidence, however, is needed to establish the feasibility and acceptability of providing maternal supplementation in conjunction with ARV therapy and malaria diagnosis, prevention, and treatment efforts.
The authors thank all the women and infants who participated in the study.
1. World Health Organization. Guidelines on HIV and Infant
Feeding: Principles and Recommendations for Infant
Feeding in the Context of HIV and a Summary of Evidence. Geneva, Switzerland: WHO Press; 2010.
2. Papathakis PC, Rollins NC, Chantry CJ, et al.. Micronutrient status during lactation in HIV-infected and HIV-uninfected South African women during the first 6 mo after delivery. Am J Clin Nutr. 2007;85:182–192.
3. Friis H, Gomo E, Koestel P, et al.. HIV and other predictors of serum folate, serum ferritin, and hemoglobin in pregnancy: a cross-sectional study in Zimbabwe. Am J Clin Nutr. 2001;73:1066–1073.
4. Finkelstein JL, Mehta S, Duggan CP, et al.. Predictors of anaemia and iron deficiency in HIV-infected pregnant women in Tanzania: a potential role for vitamin D and parasitic infections. Public Health Nutr. 2012;15:928–937.
5. O'Brien ME, Kupka R, Msamanga GI, et al.. Anemia is an independent predictor of mortality and immunologic progression of disease among women with HIV in Tanzania. J Acquir Immune Defic Syndr. 2005;40:219–225.
6. Chaparro CM. Setting the stage for child health and development: prevention of iron deficiency in early infancy. J Nutr. 2008;138:2529–2533.
7. Isanaka S, Spiegelman D, Aboud S, et al.. Post-natal anaemia and iron deficiency in HIV-infected women and the health and survival of their children. Matern Child Nutr. 2012;8:287–298.
8. Chatterjee A, Bosch RJ, Kupka R, et al.. Predictors and consequences of anaemia among antiretroviral-naive HIV-infected and HIV-uninfected children in Tanzania. Public Health Nutr. 2010;13:289–296.
9. Areechokchai D, Bowonwatanuwong C, Phonrat B, et al.. Pregnancy outcomes among HIV-infected women undergoing antiretroviral therapy. Open AIDS J. 2009;3:8–13.
10. Dryden-Peterson S, Shapiro RL, Hughes MD, et al.. Increased risk of severe infant
anemia after exposure to maternal HAART, Botswana. J Acquir Immune Defic Syndr. 2011;56:428–436.
11. Semba RD, Shah N, Klein RS, et al.. Highly active antiretroviral therapy associated with improved anemia among HIV-infected women. AIDS Patient Care STDS. 2001;15:473–480.
12. Berhane K, Karim R, Cohen MH, et al.. Impact of highly active antiretroviral therapy on anemia and relationship between anemia and survival in a large cohort of HIV-infected women: Women's Interagency HIV Study. J Acquir Immune Defic Syndr. 2004;37:1245–1252.
13. Johannessen A, Naman E, Gundersen SG, et al.. Antiretroviral treatment reverses HIV-associated anemia in rural Tanzania. BMC Infect Dis. 2011;11:190.
14. Taha TE, Kumwenda N, Gibbons A, et al.. Effect of HIV-1 antiretroviral prophylaxis on hepatic and hematological parameters of African infants. AIDS. 2002;16:851–858.
15. Taha TE, Kumwenda N, Kafulafula G, et al.. Haematological changes in African children who received short-term prophylaxis with nevirapine and zidovudine at birth. Ann Trop Paediatr. 2004;24:301–309.
16. Aizire J, Fowler MG, Wang J, et al.. Extended prophylaxis with nevirapine and cotrimoxazole among HIV-exposed uninfected infants is well tolerated. AIDS. 2012;26:325–333.
17. Chasela CS, Hudgens MG, Jamieson DJ, et al.. Maternal or infant
antiretroviral drugs to reduce HIV-1 transmission. N Engl J Med. 2010;362:2271–2281.
18. Baykan A, Yalcin SS, Yurdakok K. Does maternal iron supplementation during the lactation period affect iron status of exclusively breast-fed infants? Turk J Pediatr. 2006;48:301–307.
19. van der Horst C, Chasela C, Ahmed Y, et al.. Modifications of a large HIV prevention clinical trial to fit changing realities: a case study of the Breastfeeding, Antiretroviral, and Nutrition (BAN) protocol in Lilongwe, Malawi. Contemp Clin Trials. 2009;30:24–33.
20. Flax VL, Bentley ME, Chasela CS, et al.. Use of lipid-based nutrient supplements by HIV-infected Malawian women during lactation has no effect on infant
growth from 0 to 24 weeks. J Nutr. 2012;142:1350–1356.
21. Jamieson DJ, Chasela CS, Hudgens MG, et al.. Maternal and infant
antiretroviral regimens to prevent postnatal HIV-1 transmission: 48-week follow-up of the BAN randomised controlled trial. Lancet. 2012;379:2449–2458.
22. Kayira D, Bentley ME, Wiener J, et al.. A lipid-based nutrient supplement mitigates weight loss among HIV-infected women in a factorial randomized trial to prevent mother-to-child transmission during exclusive breastfeeding. Am J Clin Nutr. 2012;95:759–765.
23. Dow A, Kayira D, Hudgens M, et al.. Effects of cotrimoxazole prophylactic treatment on adverse health outcomes among HIV-exposed, uninfected infants. Pediatr Infect Dis J. 2012;31:842–847.
24. Chavula C, Long D, Mzembe E, et al.. Stopping the control arm in response to the DSMB: mother's choice of HIV prophylaxis during breastfeeding in the BAN Study. Contemp Clin Trials. 2012;33:55–59.
25. NIAID Division of AIDS. DAIDS/Toxicity Tables for grading severity of adult and pediatric adverse events. 2004. Available at: http://www.niaid.nih.gov/labsandresources/resources/daidsclinrsrch/documents/daidsaegradingtable.pdf
. Accessed on September 2010.
26. Ferguson YO, Eng E, Bentley M, et al.. Evaluating nurses' implementation of an infant
-feeding counseling protocol for HIV-infected mothers: the Ban Study in Lilongwe, Malawi. AIDS Educ Prev. 2009;21:141–155.
27. World Health Organization, United Nations Children Fund. Breastfeeding Counseling: A Training Course. Vol 20121993. WHO/CDR/93.3-5. Available at: http://www.who.int/maternal_child_adolescent/documents/who_cdr_93_3/en/
28. Widen EM, Bentley ME, Kayira D, et al.. Changes in soluble transferrin receptor and hemoglobin concentrations in Malawian mothers are associated with those values in their exclusively breastfed, HIV-exposed infants. J Nutr. 2014;144:367–374.
29. Miller MF, Stoltzfus RJ, Mbuya NV, 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;133:3461–3468.
30. Lee S, Guillet R, Cooper EM, et al.. Maternal inflammation at delivery affects assessment of maternal iron status. J Nutr. 2014;144:1524–32.
31. Rabe-Hesketh SS, Skrondal A. Multilevel and Longitudinal Modeling using Stata. 2nd ed. College Station, TX: Stata Press; 2008.
32. Marshall SW. Power for tests of interaction: effect of raising the Type I error rate. Epidemiol Perspect Innov. 2007;4:4.
33. Thurnham DI, Mburu AS, Mwaniki DL, et al.. Using plasma acute-phase protein concentrations to interpret nutritional biomarkers in apparently healthy HIV-1-seropositive Kenyan adults. Br J Nutr. 2008;100:174–182.
34. Thurnham DI, McCabe LD, Haldar S, et al.. Adjusting plasma ferritin concentrations to remove the effects of subclinical inflammation in the assessment of iron deficiency: a meta-analysis. Am J Clin Nutr. 2010;92:546–555.
35. McNutt LA, Wu C, Xue X, et al.. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157:940–943.
36. Pinnetti C, Baroncelli S, Molinari A, et al.. Common occurrence of anaemia at the end of pregnancy following exposure to zidovudine-free regimens. J Infect. 2011;63:144–150.
37. Bae WH, Wester C, Smeaton LM, et al.. Hematologic and hepatic toxicities associated with antenatal and postnatal exposure to maternal highly active antiretroviral therapy among infants. AIDS. 2008;22:1633–1640.
38. Mussi-Pinhata MM, Rego MA, Freimanis L, et al.. Maternal antiretrovirals and hepatic enzyme, hematologic abnormalities among human immunodeficiency virus type 1-uninfected infants: the NISDI perinatal study. Pediatr Infect Dis J. 2007;26:1032–1037.
39. El Beitune P, Duarte G. Antiretroviral agents during pregnancy: consequences on hematologic parameters in HIV-exposed, uninfected newborn infant
. Eur J Obstet Gynecol Reprod Biol. 2006;128:59–63.
40. Bunders MJ, Bekker V, Scherpbier HJ, et al.. Haematological parameters of HIV-1-uninfected infants born to HIV-1-infected mothers. Acta Paediatr. 2005;94:1571–1577.
41. Feiterna-Sperling C, Weizsaecker K, Buhrer C, et al.. Hematologic effects of maternal antiretroviral therapy and transmission prophylaxis in HIV-1-exposed uninfected newborn infants. J Acquir Immune Defic Syndr. 2007;45:43–51.
42. Pacheco SE, McIntosh K, Lu M, et al.. Effect of perinatal antiretroviral drug exposure on hematologic values in HIV-uninfected children: an analysis of the women and infants transmission study. J Infect Dis. 2006;194:1089–1097.
43. Briand N, Le Coeur S, Jourdain G, et al.. Hematological safety of perinatal exposure to zidovudine in uninfected infants born to HIV type 1-infected women in Thailand. AIDS Res Hum Retroviruses. 2010;26:1163–1166.
44. Mandelbrot L, Landreau-Mascaro A, Rekacewicz C, et al.. Lamivudine-zidovudine combination for prevention of maternal-infant
transmission of HIV-1. JAMA. 2001;285:2083–2093.
45. Le Chenadec J, Mayaux MJ, Guihenneuc-Jouyaux C, et al.; Enquete Perinatale Francaise Study Group. Perinatal antiretroviral treatment and hematopoiesis in HIV-uninfected infants. AIDS. 2003;17:2053–2061.
46. Northrop-Clewes CA. Interpreting indicators of iron status during an acute phase response–lessons from malaria and human immunodeficiency virus. Ann Clin Biochem. 2008;45:18–32.
47. Fawzi WW, Msamanga GI, Kupka R, et al.. Multivitamin supplementation improves hematologic status in HIV-infected women and their children in Tanzania. Am J Clin Nutr. 2007;85:1335–1343.
48. World Health Organization. Nutrient Requirements for People Living With HIV/AIDS: Report of a Technical Consultation. Geneva, Switzerland: WHO Press; 2003.
49. Sangare L, van Eijk AM, Ter Kuile FO, et al.. The association between malaria and iron status or supplementation in pregnancy: a systematic review and meta-analysis. PloS One. 2014;9:e87743.
50. Schumann K, Solomons NW. Can iron supplementation be reconciled with benefits and risks in areas hyperendemic for malaria? Food Nutr Bull. 2013;34:349–356.
We like to thank the members of the BAN study team: Linda Adair, Yusuf Ahmed, Mounir Ait-Khaled, Sandra Albrecht, Shrikant Bangdiwala, Ronald Bayer, Margaret Bentley, Brian Bramson, Emily Bobrow, Nicola Boyle, Sal Butera, Charles Chasela, Charity Chavula, Joseph Chimerang'ambe, Maggie Chigwenembe, Maria Chikasema, Norah Chikhungu, David Chilongozi, Grace Chiudzu, Lenesi Chome, Anne Cole, Amanda Corbett, Amy Corneli, Anna Dow, Ann Duerr, Henry Eliya, Sascha Ellington, Joseph Eron, Sherry Farr, Yvonne Owens Ferguson, Susan Fiscus, Valerie Flax, Ali Fokar, Shannon Galvin, Laura Guay, Chad Heilig, Irving Hoffman, Elizabeth Hooten, Mina Hosseinipour, Michael Hudgens, Stacy Hurst, Lisa Hyde, Denise Jamieson, George Joaki (deceased), David Jones, Elizabeth Jordan-Bell, Zebrone Kacheche, Esmie Kamanga, Gift Kamanga, Coxcilly Kampani, Portia Kamthunzi, Deborah Kamwendo, Cecilia Kanyama, Angela Kashuba, Damson Kathyola, Dumbani Kayira, Peter Kazembe, Caroline C. King, Rodney Knight, Athena P. Kourtis, Robert Krysiak, Jacob Kumwenda, Hana Lee, Edde Loeliger, Dustin Long, Misheck Luhanga, Victor Madhlopa, Maganizo Majawa, Alice Maida, Cheryl Marcus, Francis Martinson, Navdeep Thoofer, Chrissie Matiki (deceased), Douglas Mayers, Isabel Mayuni, Marita McDonough, Joyce Meme, Ceppie Merry, Khama Mita, Chimwemwe Mkomawanthu, Gertrude Mndala, Ibrahim Mndala, Agnes Moses, Albans Msika, Wezi Msungama, Beatrice Mtimuni, Jane Muita, Noel Mumba, Bonface Musis, Charles Mwansambo, Gerald Mwapasa, Jacqueline Nkhoma, Megan Parker, Richard Pendame, Ellen Piwoz, Byron Raines, Zane Ramdas, John Rublein, Mairin Ryan, Ian Sanne, Christopher Sellers, Diane Shugars, Dorothy Sichali, Wendy Snowden, Alice Soko, Allison Spensley, Jean-Marc Steens, Gerald Tegha, Martin Tembo, Roshan Thomas, Hsiao-Chuan Tien, Beth Tohill, Charles van der Horst, Esther Waalberg, Elizabeth Widen, Jeffrey Wiener, Cathy Wilfert, Patricia Wiyo, Innocent Zgambo, and Chifundo Zimba.
The BAN study was supported by grants from the Prevention Research Centers Special Interest Project of the Centers for Disease Control and Prevention (SIP 13-01 U48-CCU409660-09, SIP 26-04 U48-DP000059-01, and SIP 22-09 U48-DP001944-01); the National Institute of Allergy and Infectious Diseases; the University of North Carolina Center for AIDS Research (P30-AI50410); and the NIH Fogarty AIDS International Training and Research Program (DHHS/NIH/FIC 2-D43 TW01039-06 and R24 TW007988; the American Recovery and Reinvestment Act). The Malawi Mothers and Infants (MaMi) study was supported by a grant from the Bill and Melinda Gates Foundation (OPP53107) and the Carolina Population Center (R24 HD050924). The antiretrovirals used were donated by Abbott Laboratories, GlaxoSmithKline, Boehringer Ingelheim, Roche Pharmaceuticals, and Bristol-Myers Squibb. The Call to Action for the Preventing Mother-to-Child Transmission programme was supported by the Elizabeth Glaser Pediatric AIDS Foundation, the United Nations Children's Fund, the World Food Program, the Malawi Ministry of Health and Population, Johnson and Johnson, and the US Agency for International Development.