Abstract
OBJECTIVE: Fetal cleft lip and palate (CLP) is among the most common congenital craniofacial anomalies, posing significant physical and psychological burdens on affected families and contributing to increased societal and healthcare costs. We aimed to develop an individualized nomogram model incorporating fetal nuchal translucency thickness (NT), crown-rump length (CRL), and facial profile markers for predicting CLP in the first trimester of pregnancy. METHODS: Singleton pregnancies undergoing first-trimester screening (11+0 to 13+6 weeks) were enrolled. Fetal NT, CRL, and ultrasonographic facial markers, including the frontomaxillary facial angle (FMFA) and inferior facial angle (IFA), were measured. Feature selection was performed using the least absolute shrinkage and selection operator algorithm, followed by multivariable analysis to establish a nomogram model. The model's diagnostic performance was assessed through receiver operating characteristic (ROC) and precision-recall (PR) curves, calculating the area under the ROC curve (AUC) and area under the PR curve (AUPRC). Calibration curve analysis and decision curve analysis (DCA) were used to evaluate model calibration and clinical utility. Performance was compared with individual ultrasonographic facial markers. RESULTS: A total of 764 fetuses were included, comprising 732 normal and 32 CLP cases. The nomogram demonstrated excellent predictive performance with an AUC of 0.847 (95% CI: 0.751-0.944) and AUPRC of 0.689. The model achieved an accuracy of 0.977, F1-score of 0.640, sensitivity of 0.482, specificity of 0.999, positive predictive value of 0.955, and negative predictive value of 0.978. Compared to FMFA (AUC = 0.757), IFA (AUC = 0.708), NT (AUC = 0.714), and CRL (AUC = 0.592), the nomogram demonstrated significantly improved diagnostic performance (both P < 0.05). DCA confirmed the net clinical benefit of the model, and the calibration curve demonstrated good agreement between predicted and observed outcomes. CONCLUSION: The nomogram model integrating NT ≥ 2.5 mm, FMFA, CRL, and IFA demonstrates superior diagnostic performance for the first-trimester predicting of CLP compared with individual markers. Given its reliance on routinely measured parameters and ease of use, this model offers a practical, non-invasive, and efficient tool that may facilitate early detection of CLP in the first trimester of pregnancy.