- Design
- Secondary analysis of two randomised trials
- Population
- 291 women with pre-eclampsia diagnosed at 26–32 weeks under expectant management, Cape Town
- Primary outcome
- Pregnancy latency; composite adverse maternal and perinatal outcomes
- Effect
- Median latency 15.3 days (<3 g/24 h) vs 10.0 (3–5 g) vs 4.9 (≥5 g); maternal complications 8.8% vs 16.1% vs 27.9%; no perinatal association
A secondary analysis pooled 291 women from two randomised trials at Tygerberg Hospital, Cape Town, all being managed expectantly for pre-eclampsia diagnosed at 26 to 32 weeks. The question was whether the amount of protein in a 24-hour collection tells you anything about how long the pregnancy will last and who will be harmed.
It did, for the mother. Median prolongation was 15.3 days with under 3 g/24 h, 10.0 days at 3 to 5 g, and 4.9 days at 5 g or more. Composite maternal complications rose from 8.8% to 16.1% to 27.9% across the same bands. There was no association with the composite perinatal outcome. For latency the sFlt-1/PlGF ratio did slightly better (AUC 0.78 vs 0.67); for maternal complications the two were similar (0.67 vs 0.64).
This matters because many guidelines dropped proteinuria as a severity criterion, and many units have also stopped repeating it. Where angiogenic markers are not available — most Indian district hospitals — a quantified protein estimate carries prognostic information worth using in counselling and in planning steroids, magnesium and transfer. It is not a delivery trigger on its own: the association is with the chance of an early delivery, not a threshold at which to deliver.
- Quantify proteinuria at diagnosis of preterm pre-eclampsia (24-hour urine or protein:creatinine ratio) rather than recording it only as present.
- At 5 g/24 h or more, expect a shorter wait (median 4.9 days in this study, with wide variation): time corticosteroids and arrange in-utero transfer early.
- Tell the family that heavy proteinuria signals a shorter wait and higher maternal risk; in this study it was not associated with the composite perinatal outcome.
- Where sFlt-1/PlGF testing is unavailable, a quantified protein value gives comparable information about maternal risk.
- Do not deliver on the protein value alone; delivery indications remain maternal or fetal deterioration.
Why it matters
Proteinuria was dropped from the severity criteria, but in expectant management it still predicts how long the pregnancy will last.
Don't overread it
This is prognostic association from a single South African centre; it does not show that acting on proteinuria changes outcomes.
The statistics, in plain English
An AUC (area under the curve) of 0.5 is a coin toss and 1.0 is perfect. Values of 0.64 to 0.78 mean the tests separate higher- and lower-risk women moderately well — useful for counselling, not accurate enough to decide delivery for an individual. The interquartile ranges overlap between bands, so a woman with heavy proteinuria can still gain two weeks.
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