The aim of this study was to compare the effects of epidural analgesia with infiltration analgesia in postoperative pain control for total knee arthroplasty.Thirty patients (28 female, 2 male; mean age: 69.37±5.11 years, range: 61 to 80 years) undergoing total knee arthroplasty between May 2011 and September 2011 were randomly divided into 2 groups. All patients received spinal anesthesia with bupivacaine. Postoperative analgesia of 72 ml 0.9% NaCl + 48 ml bupivacaine (1 ml = 5 mg, total 120 ml) was administered throughout 24 hours to Group 1 (n=15) by epidural catheter and to Group 2 (n=15) by ON-Q infiltration pump. Groups were compared based on the Bromage scores and visual analog scale (VAS), blood pressure, postoperative analgesia requirement and side effects.Demographic data were similar in both groups. Rates of additional analgesia requirement at the postoperative 60th minute and 2nd hour were significantly higher in Group 2 than Group 1 (p<0.05). Rates of nausea-vomiting at the postoperative 60th minute and 2nd hour were significantly higher in Group 1 than Group 2 (p<0.05 and p<0.01, respectively). Bromage scores at 60 minutes and 2 hours was significantly higher in Group 1 than in Group 2 (p<0.01). Mean VAS scores at 60 minutes and 2 hours were significantly higher in Group 2 than Group 1 (p<0.05). While a statistically significant difference was found between systolic arterial pressure measurements at 60 minutes (p<0.05), there was no significant difference in diastolic arterial pressure and peak heart rate.Although the analgesic effect of local infiltration is provided later than by epidural analgesia, the same level of pain control can be achieved with initial additional analgesia. Local infiltration is superior to epidural analgesia in respect of few side effects and early mobilization.