In this multicenter, randomized, controlled trial, we sought to determine whether patient-controlled epidural analgesia (PCEA) for labor affected the incidence of cesarean delivery when compared with patient-controlled IV opioid analgesia (PCIA). Healthy, term nulliparous patients in 4 Canadian institutions were randomly assigned to receive PCIA with fentanyl (n = 118) or PCEA with 0.08% bupivacaine and fentanyl 1.6 microg/mL (n = 124). There was no difference in the incidence of cesarean delivery-10.2% (12 of 118) versus 9.7% (12 of 124)-or instrumental vaginal delivery-21.2% (25 of 118) versus 29% (36 of 124)-between groups. The duration of the second stage of labor was increased in the PCEA group by a median of 23 min (P = 0.02). Fifty-one patients (43%) in the PCIA group received epidural analgesia: 39 (33%) because of inadequate pain relief and 12 (10%) to facilitate operative delivery. Patients in the PCIA group required more antiemetic therapy (17% versus 6.4%; P = 0.01) and had more sedation (39% versus 5%; P < 0.001). Maternal mean pain and satisfaction with analgesia scores were better in the PCEA group (P < 0.001 and P = 0.02, respectively). More neonates in the PCIA group required active resuscitation (52% versus 31%; P = 0.001) and naloxone (17% versus 3%; P < 0.001). These observations support the hypothesis that PCEA does not result in an increased incidence of obstetrical intervention compared with PCIA. PCEA provides superior analgesia and less maternal and neonatal sedation compared with PCIA.