Intertwin umbilical vein diameter-to-abdominal circumference discordance in type 1 selective fetal growth restriction twins: association with risk of perinatal death in the smaller twin

Objective Selective fetal growth restriction (sFGR) is a common complication of monochorionic-diamniotic twin pregnancies, and there is no unified clinical standard for determining the optimal timing of intervention. This study aimed to explore the value of ultrasound parameters for identifying risk of perinatal death (PND) of the smaller twin complicated with type 1 sFGR. Methods This single-center retrospective cohort study included 49 women with type 1 sFGR twin pregnancies who received regular prenatal ultrasound examination and had complete clinical and follow-up data. Cases with type 2/3 sFGR, fetal structural or chromosomal abnormalities, severe maternal complications, intrauterine intervention, or loss to follow-up were excluded. The primary outcome was PND of the smaller twin with type 1 sFGR, defined as either intrauterine death or 28-day neonatal death. The patients were divided into the PND group ( n = 19) and the non-PND group ( n = 30). Baseline ultrasonic indices were compared, with nested logistic regression, reclassification analysis and 500-sample bootstrap internal validation performed to assess incremental value and model stability. Results Gestational age (GA) at diagnosis and the intra-abdominal umbilical vein diameter to abdominal circumference ratio (UV/AC) of the smaller twin in the PND group were significantly lower than those in the non-PND group ( P < 0.05), whereas intertwin UV/AC discordance was significantly higher (P0.05). Multivariate logistic regression identified two factors independently associated with PND in type 1 sFGR smaller twin: earlier GA at diagnosis ( P = 0.013) and elevated intertwin UV/AC discordance ( P = 0.006). The optimal cut-off of intertwin UV/AC discordance was 0.273% (AUC = 0.814, sensitivity 89.5%, specificity 53.3%). Reclassification analyses confirmed the incremental prognostic value of intertwin UV/AC discordance beyond conventional Doppler and clinical variables. The cutoff for GA at diagnosis yielded an AUC < 0.5, indicating reversed predictive performance. Conclusion GA at diagnosis and intertwin UV/AC discordance are independently associated with PND in type 1 sFGR smaller twin. Earlier type 1 sFGR diagnosis and higher intertwin UV/AC discordance are correlated with an increased PND risk. Intertwin UV/AC discordance may serve as a promising ultrasonic marker for antenatal risk stratification in type 1 sFGR.