Recheck haemoglobin by 20–24 weeks after iron: a poor response was linked to stillbirth

Design

Secondary analysis of a multicentre randomised trial (IV vs oral iron)

Population

4,252 pregnant women in India with moderate iron deficiency anaemia at 14–17 weeks

Primary outcome

Stillbirth by haemoglobin at 20–24 weeks

Effect

RR 0.74 per 1 g/dL higher haemoglobin (95% CI 0.56–0.98); risk rose below about 10.5 g/dL

This is a secondary analysis of a multicentre Indian randomised trial that compared a single intravenous iron infusion (ferric derisomaltose or ferric carboxymaltose) with oral iron for moderate iron deficiency anaemia (haemoglobin 7.0–9.9 g/dL) at 14–17 weeks. Of 4,252 women, about a third received each treatment. The analysis asked whether the haemoglobin reached at 20–24 weeks, whatever the treatment, predicted outcome.

Each 1 g/dL higher haemoglobin at 20–24 weeks was associated with lower stillbirth risk (RR 0.74, 95% CI 0.56–0.98). Modelling the curve showed stillbirth and birth before 34 weeks rising progressively as haemoglobin fell below about 10.5 g/dL. Small-for-gestational-age births did not show a significant relationship.

This is association within a trial, not a randomised comparison of targets, so it does not prove that driving haemoglobin higher prevents stillbirth. But it makes a practical point for Indian antenatal care, where moderate anaemia at booking is common: a first course of iron is not the end of the job. A repeat haemoglobin a few weeks after treatment identifies the women whose anaemia persists into the second half of pregnancy, and they are the ones this study flags. It was published in April 2026.

• Recheck haemoglobin 4–8 weeks after starting iron for moderate anaemia, and by 20–24 weeks at the latest.
• Treat a haemoglobin still below about 10.5 g/dL at 20–24 weeks as a pregnancy at higher risk.
• Check adherence and absorption problems when oral iron has not worked, and consider intravenous iron.
• Look for causes other than iron deficiency when haemoglobin does not rise, such as haemoglobinopathy or folate or B12 deficiency.
• Record the response in the antenatal card so later visits see the trend, not just one value.

Why it matters

It moves the question from whether iron was given to whether it worked.

Don't overread it

The link between haemoglobin response and stillbirth is observational; the trial did not randomise women to haemoglobin targets.

The statistics, in plain English

A relative risk of 0.74 per 1 g/dL means about a quarter lower stillbirth risk for each extra gram of haemoglobin reached. The upper confidence limit (0.98) is close to 1, so the size of the effect is uncertain even though it is statistically significant. Women who respond poorly may differ in other ways, such as infection or nutrition, that also affect outcome.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.