Fetal growth restriction (FGR) is associated with increased perinatal morbidity and mortality and requires targeted surveillance to time delivery appropriately. This prospective study evaluated a novel bilateral, segment-based umbilical artery (UA) Doppler scoring system that incorporates multisite assessment of end-diastolic flow, with the objective of testing its association with adverse perinatal outcome in pregnancies complicated by early- or late-onset FGR.
This was a single-center prospective cohort conducted between May 2025 and December 2025. The study enrolled singleton pregnancies meeting Delphi consensus criteria for FGR. Authors performed bilateral UA Doppler measurements at six predefined anatomical locations: the perivesical segment, the free-loop segment and the placental insertion site in each umbilical artery. End-diastolic flow at each site was categorized as present, absent or reversed. These categorical assessments were combined into a cumulative UA Doppler score ranging from 0 to 18 points.
Participants were stratified by UA Doppler score using a prespecified dichotomy (< 3 vs ≥ 3 points) for some analyses, and also grouped into score categories (0–2, 3–5, 6–8, 9–12 and > 12 points) to examine trends across increasing degrees of Doppler abnormality.
Composite adverse perinatal outcome (CAPO) was defined as the occurrence of at least one of the following neonatal events: neonatal hypoglycemia, need for phototherapy, neonatal sepsis, respiratory distress syndrome, requirement for continuous positive airway pressure (CPAP), need for mechanical ventilation, 5‑minute Apgar score < 7 or admission to the neonatal intensive care unit (NICU).
The authors also evaluated the prevalence of individual markers of fetoplacental hemodynamic deterioration across UA Doppler score categories, including abnormal ductus venosus Doppler, cerebroplacental ratio < 5th percentile, intrauterine fetal demise, fetal distress resulting in delivery and 5‑minute Apgar score < 7.
Analyses included comparison of outcomes between UA Doppler score categories and multivariable logistic regression to assess the association between UA Doppler score (as a continuous variable and dichotomized at ≥ 3) and CAPO. Regression models were adjusted for maternal age, body mass index, parity, mode of conception, estimated fetal weight percentile and gestational age at delivery. Trend analyses across ordered UA Doppler score categories were used to test for linear increases in CAPO and markers of hemodynamic deterioration.
A total of 251 pregnancies with FGR were included; 117 were early-onset FGR (< 32 weeks' gestation) and 134 were late-onset FGR (≥ 32 weeks' gestation). The UA Doppler score was applied to all participants based on six-site bilateral assessment and participants were stratified by score to explore associations with perinatal outcomes.
Higher UA Doppler scores (defined in the abstract as ≥ 3 points) were associated significantly with preterm delivery and higher rates of NICU admission and CAPO in both early- and late-onset FGR. In the early-onset FGR subgroup, higher UA Doppler scores were also associated with lower birth weight.
On multivariable logistic regression adjusted for maternal and pregnancy covariates (including gestational age at delivery and estimated fetal weight percentile), the UA Doppler score remained independently associated with CAPO whether considered as a continuous measure or dichotomized at the prespecified threshold.
When outcomes were examined across finer UA Doppler score categories (0–2, 3–5, 6–8, 9–12 and > 12 points), there was a significant linear trend: increasing score categories corresponded to higher rates of CAPO and of individual markers of fetoplacental hemodynamic deterioration such as abnormal ductus venosus Doppler, low cerebroplacental ratio, intrauterine fetal demise, fetal distress resulting in delivery and low 5‑minute Apgar scores.
The findings indicate that a multisite, segment-based UA Doppler score that accounts for bilateral assessments across three anatomical segments per artery can stratify risk for adverse perinatal outcomes in pregnancies affected by FGR. The observed independent association with CAPO after adjustment for key clinical covariates supports the potential incremental prognostic value of a more granular UA Doppler assessment compared with a single-site measurement.
The linear relationship between higher cumulative scores and both composite and individual adverse outcomes suggests the score captures progressive fetoplacental hemodynamic compromise.
These results support the feasibility of a segment-based approach to UA Doppler assessment in surveillance of pregnancies with early- or late-onset FGR. The UA Doppler score could potentially contribute to risk stratification and timing of delivery decisions if further validated. The authors note that additional studies are required to determine the clinical benefit and the practical feasibility of adoption within current surveillance strategies.
The abstract reports key design elements, sample size and primary outcome associations but does not provide details on several important aspects. Specific items not reported in the abstract include interobserver or intraobserver reproducibility for the multisite UA Doppler measurements, the method for selecting the dichotomous cut point (≥ 3), the distribution of scores across the cohort beyond the categorical trend analyses, and longer-term neonatal outcomes. Details on potential confounders beyond those adjusted for in the multivariable model (for example, maternal comorbidities, antenatal interventions, or timing and indications for delivery) are not presented in the abstract. These elements would be important to review in the full text to assess applicability and implementation.
In this prospective single-center cohort of 251 pregnancies with FGR, a novel bilateral multisite UA Doppler scoring system generated from six anatomical sites correlated independently with a composite adverse perinatal outcome and with individual markers of fetoplacental deterioration in both early- and late-onset FGR. The approach appears feasible and promising for risk stratification, but further validation and assessment of clinical utility and operational feasibility are needed.