The aim of the study was to evaluate post-operative pain and analgesic use after pre-operative or post-incisional i.v. fentanyl plus low dose i.v. ketamine vs. a standard treatment receiving i.v. fentanyl but not ketamine. Men undergoing radical prostatectomy under general anesthesia were randomly assigned in a double-blinded manner to one of three groups. Patients received i.v. fentanyl before incision followed by an i.v. bolus dose (0.2 ml kg(-1)) and an i.v. infusion (0.0025 ml kg(-1)min(-1)) of 1 mg ml(-1) ketamine (group 1) or normal saline (groups 2 and 3). Seventy minutes after incision, patients received i.v. fentanyl followed by an i.v. bolus dose (0.2 ml kg(-1)) and an i.v. infusion (0.0025 ml kg(-1)min(-1)) of saline (groups 1 and 3) or ketamine (group 2). Pain, von Frey pain thresholds, and cumulative morphine consumption using patient-controlled analgesia (PCA) were assessed up to 72 h after surgery. 143 patients completed the study (group 1, n = 47, group 2, n = 50, group 3, n = 46). Cumulative PCA morphine (mean+/-SD) did not differ significantly among groups (group 1, 92.3+/-45.9 mg; group 2, 107.2+/-58.4 mg; group 3, 103.6+/-50.4 mg; P = 0.08 for groups 1 vs. 2, and groups 1 vs. 3). On day 3, the hourly rate (mean+/-SEM) of morphine consumption was significantly lower (p < 0.0009) in group 1 (0.61+/-0.013 mg h(-1)) than group 2 (0.86+/-0.011 mg h(-1)) and group 3 (0.89+/-0.008 mg h(-1)). Pain scores and von Frey pain thresholds did not differ significantly among groups. Two-week and 6-month follow-ups did not reveal significant group differences in pain incidence, intensity, disability or mental health. Pre-operative, low-dose administration of i.v. ketamine did not result in a clinically meaningful reduction in pain or morphine consumption when compared with post-incisional administration of ketamine or a saline control condition.