The trial was carried out in Dar es Salaam, Tanzania. Though malaria remains endemic in Tanzania, infection prevalence in Dar es Salaam is currently in a low risk category (PfPR2-10 ≤ 5%).20 (link) Recent estimates of malaria incidence and mortality show steady declines since a peak in 2003, where the decrease in Tanzania is among the highest in SSA at 7 – 8%.21 (link) Nevertheless, in the absence of malaria interventions, risk for at least one placental infection during pregnancy in the Dar es Salaam area is 40–50% in primi- and secundigravida.22 (link)Pregnant women presenting at the Amtullabai Karimjee, Sinza and Magomeni antenatal clinics (ANCs) between September 2010 and October 2012 were invited to participate in the trial. Eligible participants were HIV-uninfected primigravidae or secundigravidae women that were at or before 27 weeks of gestational age at the time of screening, not severely anemic (hemoglobin (Hb) >8.5 g/dL) not iron deficient (serum ferritin >12 μg/L), and intended to stay in Dar es Salaam until delivery and for at least six weeks thereafter. Recruitment continued until the target sample size of 1500 participants was reached.
Pregnant women presenting at the ANCs were screened for HIV, anemia and low ferritin. ANC staff provided HIV testing, pre- and post-test counseling, and, in the event of ineligibility for the study, standard prenatal care services including antiretroviral therapy and iron supplementation, as needed. Women who then consented to participation were randomized on the same day of their presentation. Enrolled mothers were administered a background questionnaire, a food frequency questionnaire and a full clinical examination.
Participants were individually randomized in equal numbers to receive a daily oral dose of 60 mg elemental iron (as ferrous sulfate) or placebo from the time of enrollment until delivery. The active and placebo tablets and packaging were indistinguishable from one another. Allocation was performed according to a computer-generated randomization sequence using blocks of size 20 created by a scientist not involved in data collection; study clinics were issued pre-labeled regimen bottles according to this sequence. At enrollment, each participant was assigned to the next numbered bottle of regimen at that site. At each subsequent visit, study supplements were dispensed in identical bottles labeled with the participant’s study identification number prepared by study pharmacists who had no contact with the participants. The dose of 60 mg iron is the WHO-recommended dose for universal supplementation during pregnancy. Participants were instructed by clinical staff to consume the supplement with a meal. Participants were given folic acid daily per Tanzanian standard of care.
Each woman attended one of the three study clinics monthly until delivery and was provided with standard prenatal care, including intermittent preventive treatment in pregnancy malaria prophylaxis using SP (IPTp-SP; 1500 mg sulfadoxine, 75 mg pyrimethamine), given in the second and third trimesters. Participants were tested by peripheral blood smear for malaria parasites as needed and incident malaria cases were managed according to Tanzanian MOH guidelines. Vouchers for bednets were issued through a governmental program at all prenatal clinics. At each study visit, participants were administered a health questionnaire, given an obstetric examination and provided a monthly supply of study regimen. Study staff collected used regimen bottles at each visit and counted remaining pills.
On-call study midwives attended participants at delivery, collected, examined and sampled placentas, obtained blood samples and scheduled post-natal appointments. At the 6-week post-partum clinic visit, study staff ascertained survival status and morbidity and conducted anthropometric measurements and a physical examination of mother and child.
At screening, women were provided with HIV testing using two rapid assays (Alere Determine and Uni-Gold HIV-1/2), with confirmation of discrepant results using ELISA (Enzygnost HIV Integral II, Germany) at the Muhimbili University of Health and Allied Sciences (MUHAS) research laboratory. Screening values were obtained during recruitment for hemoglobin (Hemocue AB, Hb 201, Sweden) and ferritin (colloidal gold rapid assay, Glory Science Co., Ltd and Victory Medicine Inc., NY). Women with Hb ≥11g/dL and ferritin results >20 μg/L were eligible for randomization, whereas those with ferritin results in the 10–20 μg/L range required confirmation by immunoturbidimetric assay using the Cobas Integra 400 plus (Roche Diagnostics, IN) in the research laboratory. A peripheral venous blood sample was taken from all women at enrollment, 20 weeks, 30 weeks, delivery, and 6 weeks post-partum. At enrollment and delivery, blood was tested at the research laboratory for a complete blood count (CBC, AcT5 Diff AL (Beckman Coulter, FL)), serum ferritin and C-reactive protein (CRP, Cobas Integra 400 plus).
Placental malaria was evaluated using histopathologic examination and by PCR. The fresh placenta was sampled23 (link) and the tissue divided for use in histopathologic examination (microscopic infection) as well as for nucleic acid studies (submicroscopic infection). For placental histopathology, tissue was formalin-fixed, embedded, sectioned, and stained and examined by light microscopy and under polarized light for the presence of malaria pigment and parasitized erythrocytes; infections were classified by a placental histopathologist (DR).24 (link) A subset of 100 slides was submitted for external confirmation of diagnoses. To prepare placental tissue for nucleic acid studies, tissue was stabilized in RNAlater® (Qiagen, Germany) and homogenized, and genomic DNA was extracted using DNeasy® (Qiagen, Germany). Taqman® qRT-PCR was used for amplification using published primer and probe sets (P. falciparum-specific25 (link) and general Plasmodium (18S rRNA genes26 (link))). Tissue was tested for PCR inhibitors, and positive and negative controls were included on each plate for quality assurance.
Of 21,316 women screened for eligibility, 17,891 (84%) were excluded for not meeting the eligibility criteria (including being multigravida (n=7459), intending to deliver out of Dar es Salaam (n=3920), not providing blood sample to assess eligibility (n=2280), iron deficient (n=1762), advanced gestational age (n=1206), age less than 18 years (n=585), HIV positive (n=561), severe anemia (n=69), or high iron stores (n=49)), or for not returning to the study sites for confirmatory laboratory results when needed or declining to participate (n=1925) (Figure 1 ). Of 1500 randomized participants, delivery outcomes were obtained for 1469 (98%). Of these, it was not possible to collect placentas from 342 women who had early fetal loss (n=59), delivered at a non-study hospital (n=167), delivered outside of Dar es Salaam (n=98), or withdrew from the study (n=18). Of the remaining 1127 women, 1003 placenta samples were obtained (88%), and 124 women for whom collection was expected, this was missed.
The primary outcomes were prevalence of placental malaria (by histopathology or PCR), maternal hemoglobin at delivery, and infant birth weight. Secondary outcomes included prevalence of maternal anemia (Hb <11 g/dL) at delivery, LBW (≤2500 g), very LBW (≤2000 g), small for gestational age (≤10th percentile for gestational age, based on the Alexander growth standard27 (link) and the INTERGROWTH standard28 (link)), and placental weight. Maternal hospitalizations during pregnancy, adverse perinatal outcomes such as maternal death, fetal loss, preterm birth (<37 weeks gestational age), and neonatal death, and maternal ferritin and iron deficiency (ferritin <12 μg/L) at delivery were also assessed. Finally, iron deficiency in the presence (ferritin <70 μg/L and CRP >8.2 μg/mL) or absence of inflammation (ferritin <30 μg/L and CRP ≤8.2 μg/mL) at delivery was assessed using published cutoffs.29 (link)The sample size of 1500 was determined to provide at least 80% power at a 5% significance level to detect a 35% or higher effect of the intervention on placental malaria at a background rate of 20%, assuming 10% loss to follow up. Analyses followed the intention-to-treat principle and included all randomized participants. Differences in baseline measures and outcomes between the two treatment arms were assessed with χ2 tests of independence or Fisher’s exact tests for categorical variables and Wilcoxon rank-sum tests for continuous variables. Differences between treatment arms in binary outcomes measured repeatedly on participants were assessed using log-binomial models with an exchangeable correlation structure.30 Twin pregnancies (n=28 pairs) were analyzed as a single outcome, where the final birth weight used was the average of the two twin birth weights. If the delivery outcome was stillbirth for either of the twins, the pregnancy was considered to have been a stillbirth. All P-values were two-sided. SAS version 9.3 (SAS Institute Inc., Cary, NC) was used for all analyses.
The Harvard School of Public Health Human Subjects Committee, the MUHAS Senate Research and Publications Committee, and Tanzania’s National Institute for Medical Research granted institutional review board approval. The Tanzania Food and Drug Authority approved the use of the study regimens. Written informed consent was obtained from all women for their participation in the trial. Study progress was monitored by a Data Safety Monitoring Board (DSMB) annually or more frequently as determined by the DSMB. Interim analyses were conducted using an efficacy stopping rule for unblinding of p<0.001 for primary endpoints and a safety rule for unblinding of p<0.05 and possible further action at the discretion of the DSMB. The trial was registered at clinicaltrials.gov (identifier NCT01119612).
Pregnant women presenting at the ANCs were screened for HIV, anemia and low ferritin. ANC staff provided HIV testing, pre- and post-test counseling, and, in the event of ineligibility for the study, standard prenatal care services including antiretroviral therapy and iron supplementation, as needed. Women who then consented to participation were randomized on the same day of their presentation. Enrolled mothers were administered a background questionnaire, a food frequency questionnaire and a full clinical examination.
Participants were individually randomized in equal numbers to receive a daily oral dose of 60 mg elemental iron (as ferrous sulfate) or placebo from the time of enrollment until delivery. The active and placebo tablets and packaging were indistinguishable from one another. Allocation was performed according to a computer-generated randomization sequence using blocks of size 20 created by a scientist not involved in data collection; study clinics were issued pre-labeled regimen bottles according to this sequence. At enrollment, each participant was assigned to the next numbered bottle of regimen at that site. At each subsequent visit, study supplements were dispensed in identical bottles labeled with the participant’s study identification number prepared by study pharmacists who had no contact with the participants. The dose of 60 mg iron is the WHO-recommended dose for universal supplementation during pregnancy. Participants were instructed by clinical staff to consume the supplement with a meal. Participants were given folic acid daily per Tanzanian standard of care.
Each woman attended one of the three study clinics monthly until delivery and was provided with standard prenatal care, including intermittent preventive treatment in pregnancy malaria prophylaxis using SP (IPTp-SP; 1500 mg sulfadoxine, 75 mg pyrimethamine), given in the second and third trimesters. Participants were tested by peripheral blood smear for malaria parasites as needed and incident malaria cases were managed according to Tanzanian MOH guidelines. Vouchers for bednets were issued through a governmental program at all prenatal clinics. At each study visit, participants were administered a health questionnaire, given an obstetric examination and provided a monthly supply of study regimen. Study staff collected used regimen bottles at each visit and counted remaining pills.
On-call study midwives attended participants at delivery, collected, examined and sampled placentas, obtained blood samples and scheduled post-natal appointments. At the 6-week post-partum clinic visit, study staff ascertained survival status and morbidity and conducted anthropometric measurements and a physical examination of mother and child.
At screening, women were provided with HIV testing using two rapid assays (Alere Determine and Uni-Gold HIV-1/2), with confirmation of discrepant results using ELISA (Enzygnost HIV Integral II, Germany) at the Muhimbili University of Health and Allied Sciences (MUHAS) research laboratory. Screening values were obtained during recruitment for hemoglobin (Hemocue AB, Hb 201, Sweden) and ferritin (colloidal gold rapid assay, Glory Science Co., Ltd and Victory Medicine Inc., NY). Women with Hb ≥11g/dL and ferritin results >20 μg/L were eligible for randomization, whereas those with ferritin results in the 10–20 μg/L range required confirmation by immunoturbidimetric assay using the Cobas Integra 400 plus (Roche Diagnostics, IN) in the research laboratory. A peripheral venous blood sample was taken from all women at enrollment, 20 weeks, 30 weeks, delivery, and 6 weeks post-partum. At enrollment and delivery, blood was tested at the research laboratory for a complete blood count (CBC, AcT5 Diff AL (Beckman Coulter, FL)), serum ferritin and C-reactive protein (CRP, Cobas Integra 400 plus).
Placental malaria was evaluated using histopathologic examination and by PCR. The fresh placenta was sampled23 (link) and the tissue divided for use in histopathologic examination (microscopic infection) as well as for nucleic acid studies (submicroscopic infection). For placental histopathology, tissue was formalin-fixed, embedded, sectioned, and stained and examined by light microscopy and under polarized light for the presence of malaria pigment and parasitized erythrocytes; infections were classified by a placental histopathologist (DR).24 (link) A subset of 100 slides was submitted for external confirmation of diagnoses. To prepare placental tissue for nucleic acid studies, tissue was stabilized in RNAlater® (Qiagen, Germany) and homogenized, and genomic DNA was extracted using DNeasy® (Qiagen, Germany). Taqman® qRT-PCR was used for amplification using published primer and probe sets (P. falciparum-specific25 (link) and general Plasmodium (18S rRNA genes26 (link))). Tissue was tested for PCR inhibitors, and positive and negative controls were included on each plate for quality assurance.
Of 21,316 women screened for eligibility, 17,891 (84%) were excluded for not meeting the eligibility criteria (including being multigravida (n=7459), intending to deliver out of Dar es Salaam (n=3920), not providing blood sample to assess eligibility (n=2280), iron deficient (n=1762), advanced gestational age (n=1206), age less than 18 years (n=585), HIV positive (n=561), severe anemia (n=69), or high iron stores (n=49)), or for not returning to the study sites for confirmatory laboratory results when needed or declining to participate (n=1925) (
The primary outcomes were prevalence of placental malaria (by histopathology or PCR), maternal hemoglobin at delivery, and infant birth weight. Secondary outcomes included prevalence of maternal anemia (Hb <11 g/dL) at delivery, LBW (≤2500 g), very LBW (≤2000 g), small for gestational age (≤10th percentile for gestational age, based on the Alexander growth standard27 (link) and the INTERGROWTH standard28 (link)), and placental weight. Maternal hospitalizations during pregnancy, adverse perinatal outcomes such as maternal death, fetal loss, preterm birth (<37 weeks gestational age), and neonatal death, and maternal ferritin and iron deficiency (ferritin <12 μg/L) at delivery were also assessed. Finally, iron deficiency in the presence (ferritin <70 μg/L and CRP >8.2 μg/mL) or absence of inflammation (ferritin <30 μg/L and CRP ≤8.2 μg/mL) at delivery was assessed using published cutoffs.29 (link)The sample size of 1500 was determined to provide at least 80% power at a 5% significance level to detect a 35% or higher effect of the intervention on placental malaria at a background rate of 20%, assuming 10% loss to follow up. Analyses followed the intention-to-treat principle and included all randomized participants. Differences in baseline measures and outcomes between the two treatment arms were assessed with χ2 tests of independence or Fisher’s exact tests for categorical variables and Wilcoxon rank-sum tests for continuous variables. Differences between treatment arms in binary outcomes measured repeatedly on participants were assessed using log-binomial models with an exchangeable correlation structure.30 Twin pregnancies (n=28 pairs) were analyzed as a single outcome, where the final birth weight used was the average of the two twin birth weights. If the delivery outcome was stillbirth for either of the twins, the pregnancy was considered to have been a stillbirth. All P-values were two-sided. SAS version 9.3 (SAS Institute Inc., Cary, NC) was used for all analyses.
The Harvard School of Public Health Human Subjects Committee, the MUHAS Senate Research and Publications Committee, and Tanzania’s National Institute for Medical Research granted institutional review board approval. The Tanzania Food and Drug Authority approved the use of the study regimens. Written informed consent was obtained from all women for their participation in the trial. Study progress was monitored by a Data Safety Monitoring Board (DSMB) annually or more frequently as determined by the DSMB. Interim analyses were conducted using an efficacy stopping rule for unblinding of p<0.001 for primary endpoints and a safety rule for unblinding of p<0.05 and possible further action at the discretion of the DSMB. The trial was registered at clinicaltrials.gov (identifier NCT01119612).