Growth velocity adds nothing to Doppler in the small term fetus

Design

retrospective cohort study at a single tertiary perinatal centre

Population

600 singleton pregnancies with estimated fetal weight below the 10th centile after 32 weeks, delivered at term

Primary outcome

composite adverse outcome of neonatal morbidity or preeclampsia

Effect

adding slow growth velocity did not improve prediction (area under the curve 0.64 versus 0.64, P=0.54); composite outcome 22.7% versus 26.6% (P=0.689)

Six hundred singleton pregnancies with an estimated fetal weight below the 10th centile after 32 weeks, all delivered at term at a tertiary centre, were analysed retrospectively to see whether growth velocity between scans added predictive value to the criteria already in use. Velocity was calculated against interval-specific standards in women with at least two scans 14 or more days apart, and slow velocity was set at the threshold giving a 10% false positive rate.

Slow velocity was common — 31.8% of the cohort — and it did track other markers of a genuinely small fetus: an estimated weight below the 3rd centile at the last scan in 52.4% versus 23.2% (P<0.001), and abnormal umbilical artery Doppler in 4.2% versus 1.0% (P=0.023). But the composite adverse outcome, neonatal morbidity or preeclampsia, did not differ: 22.7% versus 26.6% (P=0.689). On multivariable analysis the independent predictors were the ones already in the definition — estimated weight below the 3rd centile, raised uterine artery pulsatility index, and low cerebroplacental ratio. Adding slow velocity left discrimination unchanged, area under the curve 0.64 versus 0.64 (P=0.54).

That is a useful negative. Interval growth velocity is intuitively appealing and increasingly calculated, and it costs an extra scan and a return visit. At term, in a fetus already identified as small, this analysis says the extra measurement does not sharpen the decision — the Doppler and the centile already carry the information.

One caveat on the ceiling: an area under the curve of 0.64 is weak prediction either way. This is not a study showing Doppler is sufficient; it is a study showing velocity does not rescue it.

• Do not schedule an extra interval scan solely to calculate growth velocity in a term fetus already below the 10th centile
• Keep the decision anchored on estimated weight below the 3rd centile, uterine artery pulsatility index and cerebroplacental ratio
• Read slow velocity as a marker of a smaller fetus rather than as an independent risk factor
• Retain interval scanning where it serves another purpose — amniotic fluid, Doppler trend, presentation
• Be explicit with families that no current combination predicts adverse outcome well in this group

The statistics, in plain English

Identical areas under the curve, 0.64 and 0.64 with P=0.54, mean the added variable carried no independent information once the existing criteria were in the model. That is a stronger statement than a non-significant difference in outcome rates, which in a cohort of 600 with about a quarter reaching the composite could easily miss a modest effect. Note also what 0.64 means in absolute terms: prediction that is barely better than a coin weighted slightly in the right direction.

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.