Background: Although management of twin deliveries has been a topic of discussion for decades, a consensus on how to deliver twins is lacking. The objective of this study was to examine short-term neonatal outcome of the second twin delivered by cesarean section after vaginal delivery of the first-born twin (combined delivery) and to identify predictors of combined delivery. Methods: This study was a 3-year, population-based, retrospective cohort investigation of 1,254 twin births in Denmark. The twin births were divided into three groups: vaginal deliveries, planned cesarean deliveries, and combined deliveries. Data were extracted from medical records, a fetal medicine software program (Astraia), and the National Birth Registry. Short-term poor neonatal outcome was measured as a 5-minute Apgar score ≤ 7, umbilical cord pH ≤ 7.10, and admission to neonatal intensive care unit for more than 3 days. Results: Vertex-nonvertex fetal presentations were more prevalent in combined deliveries than vaginal deliveries (OR 4.4, 2.5-7.8). Nonvertex second twins born by combined delivery had a higher risk of Apgar score ≤ 7 and umbilical cord pH ≤ 7.10 compared with vaginal delivery, unadjusted OR 6.2 (2.1-18), and unadjusted OR 3.9 (1.6-9.5). Prenatal ultrasound scans were evaluated in combined deliveries, of which 48 percent were vertex-vertex at the last ultrasound scan in pregnancy (mean gestational age 34 + 0) and 37 percent were vertex-vertex at birth. Conclusions: Vertex-nonvertex presenting twins have an increased risk of combined delivery. Combined deliveries are associated with increased neonatal morbidity for the second twin.