Parvovirus B19 in pregnancy: the number depends on who is infected

Design

systematic review and random-effects meta-analysis of pooled proportions with meta-regression comparing outbreak waves, six databases searched 2000 to April 2025

Population

44 studies, 3,221 women and 748 fetuses with parvovirus B19 infection in pregnancy

Primary outcome

fetal and perinatal outcomes, separated by maternal versus confirmed fetal infection

Effect

intrauterine demise 38.7% (27.3-51.6) with fetal infection versus 7.3% (4.6-11.3) with maternal; intrauterine transfusion 34.9% versus 15.3%; hydrops 58.7%; abnormal neuroimaging 28.6% of those imaged

Parvovirus B19 outbreaks in 2023-24 pushed the question back into antenatal clinics, and the counselling has often been done with a single figure. This meta-analysis pooled 44 studies published from 2000 to April 2025, covering 3,221 women and 748 fetuses, and separated maternal from confirmed fetal infection.

The separation is the point. Intrauterine demise occurred in 38.7% (95% CI 27.3-51.6) of confirmed fetal infections but 7.3% (4.6-11.3) of maternal infections. Intrauterine transfusion was performed in 34.9% of fetal infections and 15.3% of maternal ones. Among infected fetuses, hydrops was present in 58.7%, cardiomegaly in 39.0%, hepatic involvement in 10.8% and fetal growth restriction in 9.5%. Where fetal MRI was performed, 28.6% showed abnormal neuroimaging. Comparing the recent wave with earlier ones, neither intrauterine demise (OR 0.84, 0.46-1.54) nor ultrasound signs of anaemia differed; transfusions were numerically more frequent but not significantly so.

For clinic, this converts one alarming figure into two usable ones. A woman with confirmed maternal infection and a normal fetal assessment faces a much lower risk than the headline number, and telling her so matters. The high rate of abnormal brain findings on fetal MRI among transfused fetuses is the newer signal, and it argues for considering second-level imaging in that group rather than assuming a corrected anaemia is a corrected outcome.

• Counsel separately for maternal infection and confirmed fetal infection — the risks differ fivefold
• Serial middle cerebral artery peak systolic velocity remains the surveillance that matters after maternal infection
• Consider fetal MRI where intrauterine transfusion has been performed; abnormal neuroimaging was found in 28.6%
• The 2023-24 wave was not more severe than previous ones, whatever the case numbers suggested
• Hydrops was present in nearly six in ten infected fetuses — its absence is reassuring, not conclusive

The statistics, in plain English

The wide intervals — 27.3% to 51.6% around the 38.7% demise figure — reflect both small numbers of fetal infections and real differences between the studies pooled. These are pooled proportions from case series, so they describe fetuses that came to attention, and mild or subclinical fetal infections that resolved without detection are largely absent, which biases the rates upward. The comparison between waves is a meta-regression across studies rather than a direct comparison of patients, and its confidence intervals are wide enough that a moderate difference could have been missed.

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.