T-scoring models for fetal growth restriction in Asian women: a multicentre cohort study

The objective of this study was to develop individual scoring models for fetal growth restriction (FGR) in singleton pregnancies, based on routine early pregnancy data, in Asian women. A multicentre pregnancy and delivery cohort between 2009 and 2020 years was analysed, after randomly dividing the data into two subcohorts of training and validation sets at a 70:30 ratio. Maternal social, reproductive and medical history and family history collected at the first antenatal visit, constitute the predictive variables. Risk factors were evaluated using multivariate logistic regression analysis and risk score for stratification of women at high risk of PAH was calculated in a statistic scoring model. The overall incidence of FGR was 9.78%, among 26957 singleton pregnancies. In multivariate analysis, number of gestation (OR, 1.078, 95% CI 1.024-1.134), body mass index before pregnancy (OR, 1.079, 95% CI 1.049-1.11) and at delivery (OR, 0.883, 95% CI 0.86-0.907), family history of hypertension (OR, 1.151, 95% CI 1.022-1.296), underlying maternal diseases of lupus or antiphospholipid syndrome (OR, 2..377, 95% CI 1.616-3.496), Job of husband (OR, 1.36, 95% CI 1.176-1.574), nulliparity (OR, 2.316, 95% CI 2.031-2.642), history of pregnancy associated hypertension (OR, 1.942, 95% CI 1.3-2.901), FGR (OR, 2.25, 95% CI 1.616-3.133), and both (OR, 5.102, 95% CI 3.313-7.857) in a previous pregnancy, were significantly associated with FGR. A scoring model was developed to calculate risk of FGR. Area under the curves were 0.605 and 0.604 in the train and test sets, respectively. In conclusion, scoring model may help counselling individual weight gain in singleton pregnancies, to decrease risk of FGR, although prediction performance needs to be improved more with integrating biomarkers or sonographic findings. (Korea Health Industry Development Institute, grant number: HI21C1300)