The aim of this study was to compare maternal and neonatal outcome of twin births according to mode of delivery and to isolate the factors predicting a successful vaginal delivery and those predicting a failed trial of labor (TOL) leading to an emergent cesarean section. We reviewed all twin deliveries during the years 1995 to 2004. Parameters studied include maternal age, parity, gestational age, maternal antepartum complications and postpartum complications, fetal presentations, birthweight, mode of delivery of each twin, Apgar scores and cord pH. During the study period there were 40,710 deliveries of which 804 (1.9%) were twin deliveries. Of the 804 twins, 398 (49.5%) had planned cesarean sections (PCS) and 406 (50.5%) entered a TOL. Maternal age and parity were similar among the groups. Neonatal outcomes and postpartum complications did not differ between the groups. Of 406 women who had a TOL, 84.9% eventually delivered both twins vaginally. A significantly higher percentage of antepartum complications were noted among those who failed the TOL compared to those with successful TOL (8.2% vs. 1.7%, p =.01). The number of neonates with pH of less than 7.0 did not differ between the groups although more neonates (2.5% vs. 0.4%, p =.05) among the failed TOL had an Apgar score of less than 7.0 at 5 minutes compared to successful TOL. Vaginal delivery of both twins after TOL occurred in 91% of vertex/vertex compared with 71.8% of vertex/nonvertex presenting twins ( p < .01). Neonatal outcomes did not differ between both groups. Our results indicate that both vaginal and PCS are comparable options for vertex presenting first twin regardless of second twin presentation.