OBJECTIVE: To determine the impact of intratwin birthweight discordance and of non-vertex presentation of the second twin on successful vaginal delivery. STUDY DESIGN: The ESPRIT Trial is a national prospective population study of 642 consecutive twin pregnancies managed and delivered at 8 tertiary referral centers in one country. All pregnancies are followed using a uniform sonographic surveillance protocol. RESULTS: Vaginal or cesarean delivery (CD) was planned for 310 (48%) and 332 (52%) women, respectively. Birthweight discordance of 20% was documented in 12% (37/310) of patients who underwent a trial of labor. The sensitivity of ultrasound within 2 weeks of delivery for detection of this degree of discordance was 86%, specificity 61%. While the overall CD rate was 64% (411/642), intrapartum CD was needed in 29% (90/310) of cases where vaginal delivery was planned. Successful trial of labor did not correlate with advancing gestational age (logistic regression p-value 0.2409), vertex presentation of the 2nd twin (p-value 0.3252), nor with concordant size (regression p-value 0.2988). CONCLUSION: Neither size discordance nor non-vertex presentation of the 2nd twin is predictive of CD. When the larger twin is not presenting, successful vaginal delivery can still be anticipated in most cases.