RATIONALE: Intravenous iron is increasingly used to treat iron-deficient anaemia (IDA) in pregnancy. A previous network meta-analysis suggested that intravenous irons have a greater effect on haematological parameters than oral irons; however, the impact on serious pregnancy complications such as postpartum haemorrhage (PPH) or the need for blood transfusion was unclear. Since then, several new randomised controlled trials (RCTs) have been conducted.
OBJECTIVES: To evaluate the effect and safety of intravenous versus oral iron preparations for treating IDA in pregnancy.
SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, and two trial registries (ClinicalTrials.gov and the WHO ICTRP) for eligible studies. The latest search was performed on 19 March 2024.
ELIGIBILITY CRITERIA: We included RCTs in pregnant women with confirmed IDA (haemoglobin (Hb) level < 11 g/dL as per World Health Organization (WHO) criteria) comparing intravenous (iron sucrose, ferric carboxymaltose, ferric derisomaltose, ferumoxytol) and oral (ferrous sulfate, ferrous fumarate, ferrous gluconate) iron preparations.
OUTCOMES: Our outcomes were antenatal and postnatal Hb levels, antenatal and postnatal anaemia status, PPH, blood transfusion, maternal satisfaction, maternal well-being, breastfeeding, maternal mortality, maternal morbidity, and adverse events (AEs).
RISK OF BIAS: We used the Cochrane RoB 1 tool to assess risk of bias in the included RCTs.
SYNTHESIS METHODS: We followed standard Cochrane methods. Two review authors independently assessed studies for eligibility and scientific rigour, evaluated the risk of bias of included studies, and extracted data. Where appropriate, we pooled data using a fixed-effect model in the first instance. We reported dichotomous data as risk ratios (RRs) with 95% confidence intervals (CIs) and continuous data as mean differences (MDs) with 95% CIs. We assessed the certainty of the evidence using the GRADE approach.
INCLUDED STUDIES: We included 13 RCTs (3939 participants) mainly conducted in India and Africa (8/13). Gestational age at baseline ranged from 13 to 37 weeks, and Hb levels ranged from 5.0 to just below 11.0 g/dL. The most frequently compared preparations were intravenous iron sucrose versus oral ferrous sulfate (5/13). Most RCTs were at low risk of bias, and the certainty of evidence ranged from moderate to very low, mainly due to concerns over attrition bias, imprecision, and inconsistency.
SYNTHESIS OF RESULTS: Antenatal outcomes Compared with oral iron, intravenous iron likely slightly increases Hb level three to six weeks after treatment start (MD 0.49, 95% CI 0.28 to 0.69; 11 RCTs; 2935 participants; moderate-certainty evidence) and likely reduces anaemia status three to six weeks after treatment start (RR 0.81, 95% CI 0.77 to 0.86; 5 RCTs; 2189 participants; moderate-certainty evidence). Compared with oral iron, intravenous iron likely slightly increases Hb level around birth (MD 0.55, 95% CI 0.33 to 0.77; 6 RCTs; 1574 participants; moderate-certainty evidence) and likely reduces anaemia status around birth (RR 0.85, 95% CI 0.77 to 0.93; 4 RCTs; 1240 participants; moderate-certainty evidence). Postpartum outcomes Compared with oral iron, intravenous iron may slightly increase Hb level postpartum (MD 0.54, 95% CI 0.41 to 0.68; 3 RCTs; 1950 participants; low-certainty evidence). It may also reduce anaemia status (RR 0.66, 95% CI 0.59 to 0.73; 3 RCTs; 1950 participants; low-certainty evidence) and severe anaemia postpartum (RR 0.16, 95% CI 0.03 to 0.84; 2 RCTs; 1581 participants; very low-certainty evidence), although the evidence for the latter outcome is very uncertain. Compared with oral iron, intravenous iron may result in little to no difference in PPH (RR 1.44, 95% CI 0.50 to 4.20; 3 RCTs; 2251 participants; low-certainty evidence) and likely results in little to no difference in the need for blood transfusion (RR 0.97, 95% CI 0.58 to 1.60; 6 RCTs; 2592 participants; moderate-certainty evidence) or rates of breastfeeding (RR 1.04, 95% CI 0.97 to 1.12; 1 RCT; 404 participants; moderate-certainty evidence). No trials reported on maternal satisfaction or maternal well-being. Adverse outcomes Compared with oral iron, intravenous iron may have little to no effect on maternal mortality, but the evidence is very uncertain (RR 0.91, 95% CI 0.13 to 6.39; 4 RCTs; 2152 participants; very low-certainty evidence). Compared with oral iron, intravenous iron likely does not increase maternal morbidity: severe infections (RR 1.01, 95% CI 0.47 to 2.18; 1 RCT; 1881 participants; moderate-certainty evidence) and prolonged hospital stay (RR 0.86, 95% CI 0.62 to 1.21; 1 RCT; 1764 participants; moderate-certainty evidence) and may not increase admissions to the intensive care unit (ICU) (RR 1.99, 95% CI 0.18 to 21.87; 2 RCTs; 2069 participants; low-certainty evidence). Compared with oral iron, intravenous iron likely does not increase AEs (RR 1.05, 95% CI 0.82 to 1.35; 1 RCT; 349 participants; moderate-certainty evidence) and may not increase serious AEs (RR 1.25, 95% CI 0.61 to 2.59; 1 RCT; 1934 participants; low-certainty evidence). However, individual AEs were inconsistently reported across trials.
AUTHORS' CONCLUSIONS: Intravenous iron likely slightly increases Hb levels and likely reduces anaemia in pregnancy compared to oral iron. Hb levels postpartum may be slightly increased with intravenous iron, but the effect on postpartum severe anaemia status is very uncertain. Intravenous iron may result in little to no difference in PPH, and blood transfusion rates are likely unaffected by route of administration. Synthesis of adverse outcomes proved challenging due to their rarity and suboptimal reporting. The effects of intravenous iron on maternal mortality and admissions to the ICU are very uncertain, and there is likely little to no difference between groups in severe infections and prolonged hospital stay. Intravenous iron likely does not increase AEs and may not increase serious AEs; however, the 95% CIs in both cases include potential harm. Furthermore, this finding should be treated cautiously due to the varied adverse event profiles of both types of iron preparations. Data from the ongoing multicentre trials may address some of the identified evidence gaps. However, there is a clear need to strengthen the co-ordination of research efforts around clinically important time points of outcome measure, homogeneity of their definition, and safety reporting.
FUNDING: This Cochrane Review was partially funded by the WHO and was supported by the UK Medical Research Council funding.
REGISTRATION: Registration (2024): PROSPERO, CRD42024523791 via www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42024523791.
