Abstract: Objective To compare the analgesic efficacy of different formulas of local infiltration analgesia (LIA) after total knee arthroplasty (TKA). Methods 150 patients undergoing TKA under spinal anesthesia were randomized to three groups: group R (ropivacaine 100 mg dissolved to 50 mL with normal saline), group RF (ropivacaine 100 mg + fentanyl 100 μg dissolved to 50 mL with normal saline) and group RM(ropivacaine 100 mg + methylprednisolone 40 mg dissolved to 50 mL with normal saline). Continuous femoral nerve block (CFNB) was established after spinal anesthesia and 20 mL of 0.5% ropivacaine was infused through the catheter placed near femoral nerve. Adjuvant analgesics were IM pethidine. The pain visual analogue scale (VAS) scores were assessed after operation at 4, 8, 12, 16, 24, 36 and 48 h at rest and after operation at 12, 36 and 48 h on passive mobilization by acute pain service blinded to analgesic treatment. Nausea and vomiting, dizziness and other side-effects were assessed postoperatively. Results The pain VAS scores at rest of three groups were not significantly different 4 and 8 h after surgery. The pain VAS scores of group RM were significantly less than group R and group RF at rest 12, 16, 24, 36 and 48 h after surgery (P<0.01), but there was no significant difference between group R and group RF. A significant difference in pain VAS scores was found in group RM vs. group R and group RF on passive mobilization 24, 36, 48 h after operation (P<0.01), but there was no significant difference between group R and group RF. The postoperative pethidine consumptions of group R and group RF were more than that of group RM (P<0.01）at 48 h after operation. Conclusion CFNB combined with LIA was effective in controlling postoperative pain after TKA. The best formula for LIA was ropivacaine combined methylprednisolone and normal saline, but not combined fentanyl and normal saline.

Key words: femoral nerve, local infiltration analgesia, postoperative analgesia, total knee arthroplasty