A recent multicentre trial in India compared single-dose intravenous iron with oral iron for moderate iron deficiency anaemia in the early second trimester, and found less low birthweight and less stillbirth with intravenous iron. This secondary analysis asked a different and more useful question: does it matter whether the haemoglobin actually rises, whatever route was used. It covered 4,252 women with singleton pregnancies and haemoglobin of 7.0 to 9.9 g/dL at 14 to 17 weeks, with haemoglobin, ferritin and transferrin saturation measured again at 20 to 24 weeks, and Poisson models adjusted for maternal age, BMI, parity, treatment allocation, baseline haemoglobin and site.
The answer is that response predicts outcome. Each unit rise in haemoglobin at 20 to 24 weeks was associated with a lower stillbirth risk, relative risk 0.74 (95% CI 0.56 to 0.98). Fitting a curve rather than a straight line, the risk of stillbirth rose progressively as haemoglobin fell below 10.5 g/dL (p<0.0001), and so did early preterm birth before 34 weeks (p=0.01). Small-for-gestational-age infants showed a significant curved relationship with haemoglobin but the risk estimate did not reach significance.
This reframes what the follow-up visit is for. Anaemia in pregnancy is routinely treated as a diagnosis to be prescribed for, with the prescription itself standing in for the outcome. What this shows is that a woman whose haemoglobin has not moved by the second-trimester recheck is carrying a measurably higher risk of losing her baby — and she is identifiable, at 20 to 24 weeks, with a test that costs almost nothing.
The action is straightforward and, in Indian antenatal practice where moderate anaemia affects a large share of pregnancies, high-yield. Recheck haemoglobin four to six weeks after starting iron. If it is still below 10.5 g/dL, do not simply continue the same oral iron. Escalate — reassess adherence and tolerance, check whether the anaemia is actually iron deficiency, consider intravenous iron — and treat that pregnancy as higher risk for surveillance purposes.
One limit worth stating: this is an association, not a demonstration that raising haemoglobin faster prevents stillbirth. The authors say so, and call for prospective trials of early-pregnancy or preconception correction. A woman whose haemoglobin fails to rise may have a different underlying problem — inflammation, haemoglobinopathy, ongoing loss — that itself drives the risk.
- Recheck haemoglobin 4 to 6 weeks after starting iron in pregnancy; do not assume the prescription worked.
- A haemoglobin still below 10.5 g/dL at 20 to 24 weeks marks a pregnancy at higher risk of stillbirth and early preterm birth.
- Escalate rather than repeat: reassess adherence, confirm iron deficiency, consider intravenous iron.
- Screen for anaemia early — the window that mattered here closed by 20 to 24 weeks.
- Investigate a non-responder for haemoglobinopathy, chronic inflammation or ongoing blood loss.
The statistics, in plain English
Two analyses are reported and they answer slightly different questions. The linear model gives a relative risk of 0.74 per unit rise in haemoglobin, with an interval of 0.56 to 0.98 that only just clears 1.0 — real but not precisely measured. The quadratic model, which allows risk to curve rather than change at a constant rate, is the more informative one: it locates a threshold around 10.5 g/dL below which risk climbs steeply, with a very strong p value. Crucially, this is an observational relationship within a randomised trial. Women were randomised to iron route, not to whether their haemoglobin responded, so failure to respond may be a marker of a different problem rather than the cause of the harm.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for obstetrics & gynaecology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free