OBJECTIVE: Evidence for the role of first-trimester ultrasound in predicting outcome in twin pregnancies is conflicting. The aim of this study was to determine the association between crown-rump length (CRL) discordance and adverse perinatal outcome in twin pregnancies. METHODS: This was a retrospective study of all twin pregnancies of known chorionicity from a large regional cohort over a 10-year period. Terminations of pregnancy, cases with fetal or chromosomal abnormalities and monoamniotic pregnancies were excluded. Receiver-operating characteristics (ROC) curve and logistic regression analyses were performed to evaluate the association between CRL discordance and stillbirth, neonatal mortality, intrauterine growth restriction, preterm birth (PTB) at < 34 weeks' gestation and birth weight (BW) and ultrasound estimated fetal weight (EFW) discordance of ≥ 25%. RESULTS: A total of 2155 twin pregnancies were analyzed, of which 420 were monochorionic (MC) and 1735 dichorionic (DC). There were 42 fetal losses before 24 weeks' gestation and 23 perinatal deaths. CRL discordance was poorly predictive for fetal loss at < 24 weeks (area under the ROC curve (AUC), 0.54 (95% CI, 0.46-0.62)), perinatal loss (AUC, 0.52 (95% CI, 0.41-0.64)), BW discordance (AUC, 0.61 (95% CI, 0.56-0.65)), BW < 5(th) centile (AUC, 0.56 (95% CI, 0.53-0.59)), EFW discordance (AUC, 0.55 (95% CI, 0.51-0.60)) and PTB at < 34 weeks (AUC, 0.50 (95% CI, 0.47-0.54)). Overall mortality was significantly higher in MC (5.0%) than in DC (2.6%) twins (P = 0.016). Logistic regression analysis demonstrated that chorionicity (odds ratio 2.09 (95% CI, 1.06-4.10); P = 0.033) independently contributed to determining mortality, while CRL discordance (P = 0.201) did not. Adjusting for chorionicity did not improve the detection of adverse outcomes using CRL discordance. CONCLUSION: In the absence of aneuploidy or structural fetal abnormality, CRL discordance is of poor predictive value for adverse perinatal outcome in both MC and DC twin pregnancies. CRL discordance should not be used routinely to identify twin pregnancies at high risk of adverse perinatal outcome.
OBJECTIVE: Evidence for the role of first-trimester ultrasound in predicting outcome in twin pregnancies is conflicting. The aim of this study was to determine the association between crown-rump length (CRL) discordance and adverse perinatal outcome in twin pregnancies. METHODS: This was a retrospective study of all twin pregnancies of known chorionicity from a large regional cohort over a 10-year period. Terminations of pregnancy, cases with fetal or chromosomal abnormalities and monoamniotic pregnancies were excluded. Receiver-operating characteristics (ROC) curve and logistic regression analyses were performed to evaluate the association between CRL discordance and stillbirth, neonatal mortality, intrauterine growth restriction, preterm birth (PTB) at < 34 weeks' gestation and birth weight (BW) and ultrasound estimated fetal weight (EFW) discordance of ≥ 25%. RESULTS: A total of 2155 twin pregnancies were analyzed, of which 420 were monochorionic (MC) and 1735 dichorionic (DC). There were 42 fetal losses before 24 weeks' gestation and 23 perinatal deaths. CRL discordance was poorly predictive for fetal loss at < 24 weeks (area under the ROC curve (AUC), 0.54 (95% CI, 0.46-0.62)), perinatal loss (AUC, 0.52 (95% CI, 0.41-0.64)), BW discordance (AUC, 0.61 (95% CI, 0.56-0.65)), BW < 5(th) centile (AUC, 0.56 (95% CI, 0.53-0.59)), EFW discordance (AUC, 0.55 (95% CI, 0.51-0.60)) and PTB at < 34 weeks (AUC, 0.50 (95% CI, 0.47-0.54)). Overall mortality was significantly higher in MC (5.0%) than in DC (2.6%) twins (P = 0.016). Logistic regression analysis demonstrated that chorionicity (odds ratio 2.09 (95% CI, 1.06-4.10); P = 0.033) independently contributed to determining mortality, while CRL discordance (P = 0.201) did not. Adjusting for chorionicity did not improve the detection of adverse outcomes using CRL discordance. CONCLUSION: In the absence of aneuploidy or structural fetal abnormality, CRL discordance is of poor predictive value for adverse perinatal outcome in both MC and DC twin pregnancies. CRL discordance should not be used routinely to identify twin pregnancies at high risk of adverse perinatal outcome.