The aim of this study was to evaluate the potential of immunonutritional markers, specifically the hemoglobin, albumin, lymphocyte, and platelet (HALP) score and the prognostic nutritional index (PNI), in predicting late-onset fetal growth restriction (LO-FGR) during the first trimester. This retrospective study was conducted at a tertiary care center between October 2022 and August 2023. The study included a total of 213 singleton pregnancies, with 99 women in the LO-FGR group and 114 in the healthy control group, matched by maternal age and gestational age at delivery. All blood samples were collected between 11 and 14 weeks of gestation (during the first-trimester screening test). We analyzed first-trimester laboratory parameters, specifically focusing on hemoglobin levels, white blood cells (WBCs), lymphocytes, platelets, and albumin levels. Afterwards, we calculated the HALP score and PNI, and then compared the values of both groups. Both HALP score (3.58 ± 1.31 vs. 4.19 ± 1.8, p = 0.012) and PNI (36.75 ± 2.9 vs. 39.37 ± 3.96, p < 0.001) were significantly lower in the FGR group than in the control group. The HALP score cut-off value of < 3.43 in predicting FGR had a sensitivity of 62.3% and specificity of 54.5% (AUC = 0.600, 95% CI: 0.528–0.672, p = 0.012). The PNI cut-off value of < 37.9 in predicting FGR had a sensitivity of 65.8% and specificity of 62.9% (AUC = 0.707, 95% CI: 0.632–0.778, p < 0.001). While the HALP score was not a significant predictor of composite adverse neonatal outcomes in the FGR group, PNI showed a cut-off value of < 37.7 with a sensitivity of 60.9% and specificity of 59.7% (AUC = 0.657, 95% CI: 0.581–0.733, p < 0.001). The HALP score and PNI are valuable prognostic tools for predicting the risk of FGR in the first trimester. Low PNI values are also associated with composite adverse neonatal outcomes in pregnancies complicated by FGR.

Objective: To evaluate the performance of sonographic fetal biometry and Doppler parameters assessed at routine third‐trimester ultrasound examination for predicting small‐for‐gestational age (SGA) and fetal growth restriction (FGR). Methods: This was a retrospective cohort study of low‐risk singleton pregnancies undergoing routine ultrasound examination between 35 + 0 and 37 + 6 weeks' gestation and delivered at St George's University Hospital, London, UK, between December 2019 and February 2024. The study outcomes were SGA (birth weight < 5 ^th centile) and FGR (birth weight < 3 ^rd centile or birth weight < 10 ^th centile with composite adverse perinatal outcome). Composite adverse perinatal outcome comprised intrauterine death, neonatal death or admission to the neonatal intensive care unit. Demographic characteristics, estimated fetal weight (EFW) and abdominal circumference centiles, as well as Doppler indices, including pulsatility indices (PI) of the umbilical artery (UA), middle cerebral artery (MCA) and uterine artery (UtA) were evaluated. The cerebroplacental ratio (CPR) was calculated, and all indices were converted to multiples of the median (MoM). Multivariable logistic regression analysis was performed to identify and adjust for confounders. The area under the receiver‐operating‐characteristics curve (AUC) was used to evaluate the model's performance for predicting small neonates. Results: A total of 14 161 pregnancies were included in the study. The prevalence of SGA and FGR neonates was 3.1% and 1.5%, respectively. Independent predictors of SGA and FGR, respectively, were: EFW centile (adjusted odds ratio (aOR) 0.91 (95% CI, 0.90–0.92); P < 0.001 and aOR 0.90 (95% CI, 0.89–0.91); P < 0.001); AC centile (aOR 0.91 (95% CI, 0.90–0.92); P < 0.001 and aOR 0.91 (95% CI, 0.90–0.92); P <0.001); UA‐PI MoM (aOR 4.60 (95% CI, 2.19–9.64); P < 0.001 and aOR 2.53 (95% CI, 1.05–6.10); P = 0.038); MCA‐PI MoM (aOR 0.37 (95% CI, 0.20–0.70); P = 0.002 and aOR 0.26 (95% CI, 0.12–0.59); P = 0.001); CPR MoM (aOR 0.23 (95% CI, 0.13–0.42); P < 0.001 and aOR 0.25 (95% CI, 0.12–0.53); P < 0.001); and UtA‐PI MoM (aOR 2.54 (95% CI, 1.68–3.83); P < 0.001 and aOR 2.16 (95% CI, 1.31–3.58); P = 0.003). The EFW centile alone was associated with an AUC of 0.917 (95% CI, 0.907–0.929) for the prediction of SGA and 0.925 (95% CI, 0.908–0.939) for the prediction of FGR. This was similar to AUCs of around 0.92 for the prediction of SGA and AUCs of around 0.93 for the prediction of FGR when the EFW centile was combined with any Doppler parameters. Conclusions: Sonographic fetal biometry evaluation in the late third trimester can predict delivery of a neonate affected by SGA or FGR, including those at risk for adverse perinatal outcomes. In an unselected population, fetal arterial Doppler parameters were independent predictors of SGA and FGR, but the addition of Doppler parameters to fetal biometry did not improve prediction of the incidence of small neonates. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Objective: Shoulder dystocia (ShD) is a rare but serious obstetric emergency associated with significant neonatal morbidity. This study aimed to evaluate the predictive performance of machine learning (ML) models based on fetal biometric ratios and clinical characteristics for the identification of ShD in pregnancies without clinical suspicion of macrosomia. Methods: We conducted a retrospective case-control study including 284 women (84 ShD cases and 200 controls) who underwent spontaneous vaginal delivery between 37 and 42 weeks of gestation. All participants had an estimated fetal weight (EFW) below the 90th percentile according to Hadlock reference curves. Univariate and multivariate logistic regression analyses were performed on maternal and neonatal parameters, and statistically significant variables (p < 0.05) were used to construct adjusted odds ratio (aOR) models. Supervised ML models—Logistic Regression (LR), Random Forest (RF), and Extreme Gradient Boosting (XGB)—were trained and tested to assess predictive accuracy. Performance metrics included AUC-ROC, sensitivity, specificity, accuracy, and F1-score. Results: The BPD/AC ratio and AC/FL ratio markedly enhanced the prediction of ShD. When added to other features in RF models, the BPD/AC ratio got an AUC of 0.884 (95% CI: 0.802–0.957), a sensitivity of 68%, and a specificity of 83%. On the other hand, the AC/FL ratio, along with other factors, led to an AUC of 0.896 (95% CI: 0.805–0.972), 68% sensitivity, and 90% specificity. Conclusions: In pregnancies without clinical suspicion of macrosomia, ML models integrating fetal biometric ratios with maternal and labor-related factors significantly improved the prediction of ShD. These models may support clinical decision-making in low-risk deliveries where ShD is often unexpected.