BACKGROUND: Tramadol, a centrally acting analgesic drug, can be administered via multiple routes and is generally well tolerated. OBJECTIVE: This study was designed to assess the pharmacokinetics of epidural tramadol administered preoperatively in Japanese patients undergoing upper abdominal surgery. METHOD: Japanese patients who were scheduled to undergo upper abdominal surgery in The Kitasato Institute Hospital, Tokyo, Japan, were included. Patients received tramadol 2 mg/kg with 5 mL of 1176 mepivacaine epidurally 10 minutes before incision. The serum concentration of tramadol was determined by high-performance liquid chromatography for 21 hours after administration. Serum concentration was determined before tramadol administration and 10, 20, 30, and 60 minutes after tramadol administration, first postoperative night, and first postoperative day. Pain score and adverse events (AEs) were assessed at 1, 3, 6, 12, 18, 24, 36, and 48 hours after surgery by patient interview. RESULTS: Eleven patients were assessed for enrollment. Seven patients (6 men, 1 woman; mean [SD] age, 61.3 [12.6] years; mean [SD] weight, 59.9 [8.9] kg) provided consent and completed the study. The mean (SD) serum C-max of tramadol was 1385.5 (390-8) ng/mL, T-max was 0.33 (0.22) hour, and terminal elimination half-life (t(1/2 beta)) was 10.5 (2.3) hours. Four patients complained of nausea; however, only 1 patient was administered an antiemetic. No other AEs were reported. CONCLUSION: This pilot study found that epidural tramadol administered before incision induced a C-max within 30 minutes of administration. The drug was detected in serum at similar to 21 hours after surgery.