The objective of this study was to conduct a longitudinal assessment of inter-twin growth and Doppler discordance, to identify possible distinct patterns, and to investigate the predictive value of longitudinal discordance patterns for adverse perinatal outcomes in twin pregnancies. This retrospective cohort study included twin pregnancies followed and delivered at a tertiary University Hospital in London (UK), between 2010 and 2023. We included pregnancies with at least three ultrasound assessments after 18 weeks and delivery after 34 weeks' gestation. Monoamniotic twin pregnancies, pregnancies with twin-to-twin transfusion syndrome, genetic or structural abnormalities, or incomplete data were excluded. Data on chorionicity, biometry, Doppler indices, maternal characteristics, and obstetric as well as neonatal outcomes were extracted from electronic records. Doppler assessment included velocimetry of the umbilical artery, middle cerebral artery and cerebroplacental ratio. Inter-twin growth discordance was calculated for each scan. The primary outcome was a composite of perinatal mortality and neonatal morbidity. Statistical analysis involved multilevel mixed-effects regression models and unsupervised machine learning algorithms, specifically k-means clustering, to identify distinct patterns of inter-twin discordance and their predictive value. Predictive models were compared using the area under the receiver operating characteristics curve, calibration intercept, and slope, validated with repeated cross-validation. Analyses were performed using R, with significance set at p<0.05. Data from a total of 823 twin pregnancies (647 dichorionic, 176 monochorionic) were analyzed. Five distinct patterns of inter-twin growth discordance-low-stable (n=204, 24.8%), mild-decreasing (n=171, 20.8%), low-increasing (n=173, 21.0%), mild-increasing (n=189, 23.0%), and high-stable (n=86, 10.4%)-were derived using an unsupervised learning algorithm that clustered twin pairs based on the progression and patterns of discordance over gestation. In the high-stable cluster, the rates of perinatal morbidity (46.5%, 40/86) and mortality (9.3%, 8/86) were significantly higher, compared to the low-stable (reference) cluster (p<0.001). High-stable growth pattern was also associated with a significantly higher risk of composite adverse perinatal outcomes (Odds ratio 70.19, 95% confidence interval 24.18-299.03, p<0.001; adjusted Odds ratio 76.44, 95% confidence interval 25.39-333.02, p<0.001). The model integrating discordance pattern with CPR discordance at the last ultrasound before delivery demonstrated superior predictive accuracy, evidenced by the highest area under the receiver operating characteristics curve of 0.802 (95% confidence interval 0.712 - 0.892 0.046, p<0.001), compared to only discordance patterns (area under the receiver operating characteristics curve 0.785, 95% confidence interval 0.697 -0.873), intertwin weight discordance at the last ultrasound prior to delivery (area under the receiver operating characteristics curve 0.677, 95% confidence interval 0.545 - 0.809), combination of single measurements of estimated fetal weight and CPR discordance at the last ultrasound prior to delivery (area under the receiver operating characteristics curve 0.702, 95% confidence interval 0.586 -0.818) and single measurement of CPR discordance only at the last ultrasound (area under the receiver operating characteristics curve 0.633, 95% confidence interval 0.515 - 0.751). We identified five distinct trajectories of inter-twin fetal growth discordance using an unsupervised machine learning algorithm. Consistent high discordance is associated with increased rates of adverse perinatal outcomes, with a dose-response relationship. Additionally, a predictive model integrating discordance trajectory and CPR discordance at the last visit demonstrated superior predictive accuracy for the prediction of composite adverse perinatal outcomes, compared to either of these measurements alone or a single value of estimated fetal weight discordance at the last ultrasound prior to delivery.
Read full abstract