Development and Validation of a Small for Gestational Age Screening Model at 21–24 Weeks Based on the Real-World Clinical Data

Abstract

Background: Small for gestational age (SGA) is a condition in which fetal birthweight is below the 10th percentile for the gestational age, which increases the risk of perinatal morbidity and mortality. Therefore, early screening for each pregnant woman is of great interest. We aimed to develop an accurate and widely applicable screening model for SGA at 21–24 gestational weeks of singleton pregnancies. Methods: This retrospective observational study included medical records of 23,783 pregnant women who gave birth to singleton infants at a tertiary hospital in Shanghai between 1 January 2018 and 31 December 2019. The obtained data were nonrandomly classified into training (1 January 2018 to 31 December 2018) and validation (1 January 2019 to 31 December 2019) datasets based on the year of data collection. The study variables, including maternal characteristics, laboratory test results, and sonographic parameters at 21–24 weeks of gestation were compared between the two groups. Further, univariate and multivariate logistic regression analyses were performed to identify independent risk factors for SGA. The reduced model was presented as a nomogram. The performance of the nomogram was assessed in terms of its discrimination, calibration, and clinical usefulness. Moreover, its performance was assessed in the preterm subgroup of SGA. Results: Overall, 11,746 and 12,037 cases were included in the training and validation datasets, respectively. The developed SGA nomogram, comprising 12 selected variables, including age, gravidity, parity, body mass index, gestational age, single umbilical artery, abdominal circumference, humerus length, abdominal anteroposterior trunk diameter, umbilical artery systolic/diastolic ratio, transverse trunk diameter, and fasting plasma glucose, was significantly associated with SGA. The area under the curve value of our SGA nomogram model was 0.7, indicating a good identification ability and favorable calibration. Regarding preterm SGA fetuses, the nomogram achieved a satisfactory performance, with an average prediction rate of 86.3%. Conclusions: Our model is a reliable screening tool for SGA at 21–24 gestational weeks, especially for high-risk preterm fetuses. We believe that it will help clinical healthcare staff to arrange more comprehensive prenatal care examinations and, consequently, provide a timely diagnosis, intervention, and delivery.

Approach and Method for Bayesian Network Modelling: A Case Study in Pregnancy Outcomes for England and Wales

“…Maternal Age: Some works only considered maternal age as a risk factor if the mother was either above or below a particular boundary age [54,55]. Others incorporate age into their models as a continuous variable [39,56].…”