Longitudinal Assessment of Third-Trimester Fetal Biometry and Cerebroplacental Ratio for Predicting Adverse Perinatal Outcomes in an Unselected Obstetric Population: A Prospective Cohort Study

Background: The clinical value of the cerebroplacental ratio (CPR) for predicting adverse perinatal outcomes in unselected obstetric populations remains uncertain. We aimed to evaluate the prognostic performance of serial third-trimester fetal biometry and Doppler-derived CPR and to compare their ability to identify adverse perinatal outcomes and neonatal growth abnormalities. Methods: In this prospective longitudinal cohort study, 100 consecutive pregnancies from an unselected obstetric population underwent standardized ultrasonographic examinations at both 28 and 37 weeks of gestation. Fetal biometric measurements, estimated fetal weight (EFW), amniotic fluid index, umbilical and middle cerebral artery Doppler indices, and CPR were recorded. Maternal, obstetric, delivery, and neonatal data were collected prospectively. Receiver operating characteristic (ROC) curve analyses and parsimonious multivariable regression models, accompanied by 1000-sample bootstrap internal validation, were performed to evaluate the independent predictive capacity of fetal biometry and CPR metrics. Results: Gestational age-specific CPR percentiles were associated with expected physiological Doppler changes but showed limited associations with obstetric and neonatal outcomes. In separate parsimonious regression models optimized for event-per-variable ratios, neither 28-week nor 37-week CPR independently predicted composite adverse perinatal outcomes, NICU admission, low Apgar scores, or abnormal umbilical cord blood pH. Rigorous bootstrap internal validation confirmed that optimism-corrected area under the curve (AUC) values for CPR models remained close to chance (range: 0.463–0.507). In contrast, 37-week EFW independently predicted neonatal birth weight, while both 37-week EFW and EFW percentile demonstrated good discriminatory performance for identifying small-for-gestational-age neonates (AUC = 0.812 and 0.833, respectively; both p < 0.001). Maternal body mass index and late-pregnancy fetal biometry also showed moderate discriminatory performance for predicting large-for-gestational-age neonates. Overall, late-pregnancy biometry outperformed CPR for identifying growth abnormalities, whereas neither modality alone accurately predicted composite adverse perinatal outcomes. Conclusions: In an unselected obstetric population, serial third-trimester CPR provided limited additional prognostic information beyond routine fetal biometry. Late-pregnancy fetal biometry, particularly 37-week estimated fetal weight, demonstrated superior performance for identifying neonatal growth abnormalities. These findings support the continued use of conventional fetal biometry as the primary component of routine third-trimester surveillance, with CPR serving as a complementary rather than standalone Doppler parameter.