Incremental Prognostic Value of Fetal Doppler Beyond Biometry for Adverse Perinatal Outcomes in Suspected Fetal Growth Restriction: A Retrospective Cohort Study

Objective: To explore the incremental prognostic information of fetal Doppler measurements beyond maternal and fetal biometric characteristics for adverse perinatal outcomes in pregnancies with suspected fetal smallness. Methods: This retrospective cohort study included 606 singleton pregnancies assessed for suspected or diagnosed fetal growth restriction. The principal analysis was restricted to 429 pregnancies with estimated fetal weight (EFW) or abdominal circumference (AC) below the 10th percentile at the index examination. A composite adverse perinatal outcome comprised five-minute Apgar score < 7, umbilical arterial pH < 7.10, neonatal intensive care unit (NICU) admission, respiratory support, neonatal death, or stillbirth. Baseline clinical–biometric, Doppler-integrated, and delivery-week sensitivity logistic models were assessed using stratified 10-fold out-of-fold predictions. Additional robustness analyses evaluated component frequencies, a hard-outcome sensitivity composite less directly influenced by clinical management, paired uncertainty in out-of-fold performance differences, and model performance across alternative cohort definitions. Results: Among 429 pregnancies in the principal cohort, 111 (25.9%) met the composite adverse perinatal outcome, and 206 (48.0%) had abnormal Doppler findings. Event rates were 85/206 (41.3%) among pregnancies with abnormal Doppler and 26/223 (11.7%) among those with normal Doppler. The baseline and Doppler-integrated out-of-fold areas under the receiver operating characteristic curve (AUCs) were 0.642 and 0.690, respectively; corresponding Brier scores were 0.183 and 0.174. Elevated umbilical artery pulsatility index with forward flow and absent or reversed end-diastolic flow (AEDF/REDF) were independently associated with the composite outcome, with adjusted odds ratios of 2.61 (95% confidence interval [CI], 1.58–4.32) and 6.60 (95% CI, 2.40–18.18), respectively, compared with normal umbilical artery flow. The adjusted estimate for the AEDF/REDF subgroup was imprecise because this subgroup included only 19 pregnancies. The paired out-of-fold AUC increment for the Doppler-integrated model was 0.049 (95% CI, 0.005–0.094); the corresponding Brier-score improvement was 0.009 (95% CI, −0.0004 to 0.0179). Sensitivity analyses supported a modest incremental contribution but did not establish the superiority of the integrated model over alternatives based on the cerebroplacental ratio (CPR) or umbilical artery (UA) Doppler. Conclusions: In pregnancies with suspected fetal smallness defined by EFW or AC below the 10th percentile, fetal Doppler assessment provided modest incremental prognostic information beyond maternal and fetal biometric characteristics. Umbilical artery abnormalities were associated with the composite adverse perinatal outcome, although this association should be interpreted in the context of delivery timing and clinical management. External validation and assessment of clinical utility are required before clinical implementation of the model.