Randomised Controlled Trial of Intravenous Iron (ferric carboxymaltose) for Second Trimester Anaemia in Malawian Pregnant Women (REVAMP)

Introduction

Anemia affects 46% of pregnancies in Africa. WHO estimates 50% of anaemia in pregnancy (44% in Africa) is responsive to iron; other causes of anaemia in Africa include infection (e.g. P. falciparum). Antenatal oral iron supplementation improves maternal anaemia and birthweight, but adherence may be suboptimal. Ferric carboxymaltose (FCM) is a modern intravenous iron formulation which enables provision of 1000mg (or 20mg/kg) in one 15 minute infusion. We reasoned FCM is an option to treat antenatal anaemia in low income primary-care settings, but the feasibility, efficacy and safety of this approach is unproven.

Methods

REVAMP was an open label, individually randomised controlled trial set in Malawi, sub-Saharan Africa, a low income setting where anaemia, malaria and other infections (including HIV) are common. Women were screened at antenatal clinics and were eligible if they met criteria including capillary haemoglobin (Hb) <10.0g/dL, negative Plasmodium rapid diagnostic test, and ultrasound-confirmed singleton pregnancy of 13 to 26 weeks gestation. HIV positivity was not an exclusion. Women were randomly assigned 1:1 to the intervention arm (FCM 20mg/kg, up to 1000mg, once at enrolment) or standard of care (SOC, 60mg elemental iron b.d for 90 days, in alignment with local practice). All women received IPTp with Suphadoxine-Pyrimethamine at enrolment and 28 days. The primary maternal outcome was anaemia at 36 weeks gestation. Secondary maternal outcomes included anaemia at other timepoints (28 days post infusion, during delivery and 28 days post-partum), and Hb, iron deficiency (ID) and iron deficiency anaemia (IDA) at each timepoint. The primary neonatal outcome was birthweight. Subgroup analyses (e.g. by baseline anaemia severity, ID, IDA, HIV status) were prespecified. Safety outcomes included incidence of adverse events (AEs) during infusion, and all AEs from randomisation to 4 weeks postpartum, including serious AEs (SAEs): unplanned hospitalisation or death including pregnancy loss and neonatal death.

Results

We screened 21258 women and randomised 862. Baseline characteristics, including anaemia severity and prevalence of iron deficiency, inflammation, and HIV were similar between arms. Compared with SOC, FCM reduced the prevalence of anaemia at 28 days post infusion (PR 0.91 95% CI 0.85 to 0.97, P=0.007) but did not statistically significantly decrease anaemia prevalence at the primary timepoint of 36 weeks gestation (52.5% vs 56.8%, prevalence ratio (PR) 0.92, 95% CI 0.81 to 1.06, P=0.27), or at delivery or 28 days postpartum (Table 1). Hb concentrations increased from baseline to 28 days post partum in both arms and were greater in women randomised to FCM 28 days post infusion (1.3 vs 1.1 g/dL, difference 0.19 [0.06-0.33]) and at 28 days postpartum (3.1 vs 2.9 g/dL, difference 0.25 [0.06-0.45]). IDA prevalence was reduced at each timepoint, including at 36 weeks gestation (PR 0.29 [0.20 to 0.43], P<0.001). ID prevalence was also reduced at all timepoints. Birthweight did not differ between FCM and SOC. Adverse reactions during FCM administration were reported by 26/430 (6.0%) women; there were no infusion-related SAEs. There was no clinically relevant difference in AEs by any systemic organ class, including infection (including malaria), and SAEs in women and neonates (Table 2). Hypophosphataemia (PO4<0.080mmol/L) at 28 days post infusion was seen in 22/403 (5.5%) women randomised to FCM compared to 8/382 (2.0%) with SOC. There was no difference in pregnancy loss or stillbirth between arms.

Conclusions

In this low-income, malaria endemic setting, treatment of anaemic pregnant women in the second trimester with FCM worked faster than SOC to resolve anaemia but benefit was not evident by 36 weeks. However FCM reduced ID and IDA at every timepoint. Administration of FCM was feasible and safe. Extended follow-up of mothers and babies is underway.