Prediction of growth abnormalities in monochorionic diamniotic twin pregnancies using estimated fetal weight discordance and clinical factors: a retrospective cohort study

Abstract

Abstract Objectives To explore whether integrating additional conventional clinical or ultrasonic predictors with mid-pregnancy estimated fetal weight discordance (EFWD) can enhance the predictive ability for fetal growth abnormalities in monochorionic diamniotic (MCDA) twin pregnancies. Methods A retrospective cohort study was conducted, including MCDA twin pregnancies delivered over a five-year period. Estimated fetal weight was calculated using the Hadlock formula. The primary growth abnormalities investigated were late-onset selective fetal growth restriction (late-onset sFGR), birth weight discordance (BWD) ≥20 %, and small for gestational age (SGA). Binary logistic regression analysis was used to develop prediction models for growth abnormalities using EFWD alone and in combination with other clinical predictors. The models were compared using receiver operating characteristic (ROC) curves and AIC (Akaike Information Criterion). Results A total of 139 MCDA twin pregnancies were included in the analysis (mean age: 30.40ᵒ±ᵒ3.89 years, 34.5 % conceived through in vitro Fertilization). EFWD (OR 1.25; 95 % CI, 1.14–1.37) and pregnancy complications (OR 2.66; 95 % CI, 1.12–6.32) were significant risk factors for late-onset sFGR. The combined model demonstrated enhanced discriminative ability for late-onset sFGR compared with EFWD alone (area under the ROC curve [AUC]: 0.82 [95 % CI: 0.71–0.94] vs. 0.81 [95 % CI: 0.70–0.92]; p=0.01). EFWD (OR 1.26; 95 % CI, 1.14–1.40) and mean weekly change in EFWD (ΔEFWD/wk) (OR 0.72; 95 % CI, 0.54–0.97) were significant predictors of BWD ≥20 %. The combined model showed superior performance compared with EFWD alone (AUC: 0.84 [95 % CI: 0.76–0.92] vs. 0.79 [95 % CI: 0.70–0.89]; p=0.02). No significant improvement was observed in predicting SGA when combining predictors with EFWD. Conclusions These combined models may offer incremental value for risk stratification in MCDA twin pregnancies without severe complications before 24 weeks of gestation. Prospective validation is required before clinical implementation.

The authors' abstract, as published at the source. Journal of Perinatal Medicine, 2026 · DOI ↗