A nonopioid multimodal pain protocol achieves equivalent pain control after total shoulder arthroplasty: a randomized controlled trial

Amid efforts to reduce opioid use, identifying effective nonopioid pain protocols for orthopedic procedures is a priority. Shoulder arthroplasty is associated with significant postoperative pain and has traditionally relied on opioids, despite their side effects. This study evaluated the efficacy of a postoperative nonopioid, multimodal pain protocol compared to an opioid protocol for patients undergoing shoulder arthroplasty. We performed a prospective, single blinded, randomized controlled trial including patients undergoing primary anatomic or reverse total shoulder arthroplasty. Patients were excluded if they underwent revision surgery, surgery for fracture, or received opioids within 3 months of surgery. All patients received standardized preoperative analgesics, general anesthesia, and an intraoperative periarticular injection without a regional block. Patients were randomly assigned to a postoperative nonopioid multimodal pain protocol or a standard opioid protocol containing 28 tablets of 5 mg oxycodone in addition to the multimodal regimen. Patients completed visual analog scale and Patient Reported Outcome Measurements Information System Pain Interference surveys and reported opioid use and side effects over 10 postoperative days. An intention-to-treat analysis was performed. Seventy-eight patients were enrolled and included in the analysis with 38 in the multimodal and 40 in the opioid cohort. Demographics (age, sex, race, and body mass index) were comparable between both groups. Pain scores at 10 days postoperatively were comparable between groups with no significant differences in mean visual analog scale pain (2.4 &#xb1; 1.7 in the multimodal group vs. 2.3 &#xb1; 1.6 in the opioid group; P = .52) or median Patient Reported Outcome Measurements Information System-Pain Interference scores (61.5, interquartile range 58.5-67.0 in the multimodal vs. 62.0, interquartile range 56.5-64.0 in the opioid; P = .90. However, the opioid group consumed significantly more morphine milligram equivalents (3.7 &#xb1; 5.07 compared to 0.4 &#xb1; 1.35 for the multimodal group; P < .001). There were no significant differences in side effects between treatment groups. In patients undergoing shoulder arthroplasty, a nonopioid, multimodal pain protocol provided noninferior pain relief compared to an opioid-based regimen. These findings support the shift toward opioid-sparing strategies in orthopedic surgery to minimize postoperative dependence without compromising patient care.