Analgesic effect of perioperative duloxetine in patients after total knee arthroplasty: a prospective, randomized, double-blind, placebo-controlled trial

Abstract Background To investigate the analgesic effect of perioperative use of duloxetine in patients received total knee arthroplasty (TKA). Method This prospective randomized, double-blind, placebo-controlled trial study was registered in the Chinese Clinical Trial Registry (ChiCTR2000033910). 100 patients were finally enrolled. The hospital pharmacy prepared small capsules containing either duloxetine or starch (placebo) which were all identical in appearance and weight (50:50). The 100 enrolled patients were given a capsule (containing either 60 mg duloxetine or 60 mg placebo) every night before sleep since preoperative day 2 till postoperative day 14 (17 days in all) by a nurse who were not involved in this trial. Other perioperative managements were the same in the two groups. The primary outcome was the VAS score, including rVAS (visual analogue scale at rest) and aVAS (visual analogue scale upon ambulation) throughout the perioperative period. The secondary outcomes included opioid consumption, range of motion, including both active range of motion (aROM) and passive range of motion (pROM) and adverse events. The patients were followed up everyday until 7 days after TKA, afterwards, they were followed up at the time of 3 weeks and 3 months after TKA. Result rVAS in duloxetine group were significantly less than placebo group throughout the postoperative period: 4.7 ± 2.3 vs 5.9 ± 2.6 (P = 0.016) at 24 h postoperative; 2.1 ± 1.6 vs 2.8 ± 1.7 (P = 0.037) at 7 days postoperative. In terms of aVAS, similarly, duloxetine group had less aVAS than placebo group throughout the postoperative period: 6.2 ± 2.1 vs 7.1 ± 2.2 (P = 0.039) at 24 h postoperative; 3.3 ± 1.7 vs 4.1 ± 2.0 (P = 0.034) at 7 days postoperative. Patients in duloxetine group consumed significantly less opioids per day than the placebo group: 24.2 ± 10.1 g vs 28.5 ± 8.3 g (P = 0.022) at 24 h postoperative; 2.7 ± 2.5 g vs 4.1 ± 2.6 g (P = 0.007) at 7 days postoperative. aROM in duloxetine group were significantly better than placebo group until postoperative day 6, the aROM became comparable between the two groups: 110.2 ± 9.9° in duloxetine group vs 107.5 ± 11.5° in control group (P = 0.211). In terms of pROM, duloxetine group had significantly better pROM until postoperative day 5, the pROM became comparable between the two groups: 103.8 ± 12.1° in duloxetine group vs 99.5 ± 10.8° in control group (P = 0.064). No significant difference was found between the two groups in the rates of dizziness, bleeding, sweating, fatigue and dryness of mouth. In the placebo group, more patients got nausea/vomiting and constipation (P