Construction of SGA prediction model based on multi-dimensional indicators in the second trimester of pregnancy: integrating parturient characteristics, serum markers and ultrasound parameters

TL;DR

This study creates a model to predict small-for-gestational-age infants by combining maternal and fetal data from early and mid-pregnancy.

Contribution

A novel nomogram model integrating first-trimester maternal and serum data with second-trimester ultrasound parameters for SGA prediction.

Findings

The model achieved a C-index of 0.783 in the training set and 0.754 in the validation set.

Key predictors included maternal age, PAPP-A, β-hCG, fetal abdominal circumference, femur length, and umbilical artery PI.

Abstract

To develop and validate a prediction model integrating first-trimester maternal characteristics, serum markers, and second-trimester fetal ultrasound parameters for small-for-gestational-age (SGA) infants.

This retrospective study analyzed 546 pregnant women (training set: n = 382; validation set: n = 164) from February 2022 to December 2024. Maternal baseline data, first-trimester pregnancy-associated plasma protein-A (PAPP-A) and β-human chorionic gonadotropin (β-hCG) levels, and second-trimester ultrasound indicators were collected. Multivariate logistic regression identified independent predictors, and a nomogram was constructed. Model performance was assessed using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA).

The incidence of SGA was 18.85% (72/382) in the training set and 19.51% (32/164) in the validation set. Multivariate logistic regression showed that maternal age, levels of PAPP-A and β-hCG in the first trimester, fetal abdominal circumference, femur length, and umbilical artery PI in the second trimester were independent risk factors for SGA (all P < 0.05). The nomogram model showed good calibration and predictive efficacy in both the training set and the validation set. The C-index reached 0.783 and 0.754 respectively. The areas under the ROC curve (AUC) 0.783 [95% confidence interval (CI): 0.716–0.850] and 0.754 (95% CI: 0.641–0.867) respectively. The optimal thresholds determined based on Youden's index were 0.206 (sensitivity = 0.726, specificity = 0.745) for the training set and 0.227 (sensitivity = 0.747, specificity = 0.714) for the validation set.

The nomogram prediction model constructed with these combined indicators is helpful for evaluating the risk of SGA. However, further verification through large-sample and multi-center studies is still needed to provide a reference for early clinical intervention.

Figures