Implementation of a nonopioid multimodal analgesia protocol significantly reduces opioids prescribed after total shoulder arthroplasty: a retrospective study

Abstract

BackgroundThe purpose of this study was to evaluate the impact of a nonopioid multimodal pain protocol on the amount of opioids prescribed, pain level, and complications in patients undergoing total shoulder arthroplasty (TSA) compared to a standard opioid regimen. MethodsWe retrospectively reviewed 2 consecutive cohorts who underwent either primary anatomic or reverse TSA by a single surgeon. The opioid cohort included patients from 2016 to 2020 who received an intraoperative periarticular injection (Ropivacaine, Epinephrine, and Ketorolac cocktail) and were prescribed Oxycodone/Acetaminophen 5/325 mg on discharge. The nonopioid cohort included patients from 2020 to 2022, consisting of preoperative oral analgesics (Celecoxib, Pregabalin, and Tramadol); intraoperative intravenous Dexamethasone and Acetaminophen and periarticular injection; and postoperative oral Dexamethasone and oral analgesics (Pregabalin, Tizanidine, Magnesium, Ibuprofen, and Acetaminophen). The nonopioid cohort had the option to ask for an opioid prescription postoperatively if needed. Patient-reported outcomes collected included visual analog scale scores for pain and Patient-Reported Outcome Measurement Information System up to 1 year after surgery. Opioid consumption using Morphine Milligram Equivalents were compared up to 3 months postoperatively. ResultsThere were 232 patients in the opioid and 112 in the nonopioid cohorts, respectively. The nonopioid protocol had lower mean visual analog scale scores at preoperative (6.4 vs. 7.4, P < .05), 10-day (3.5 vs. 4.2, P < .05), and 6-week postoperative time points (2.1 vs. 2.8, P < .05), but no differences between the groups were noted 3 months postoperatively. No differences in Patient-Reported Outcome Measurement Information System–upper extremity, pain interference, or depression were found preoperatively and up to 1 year after surgery. Opioid consumption was lower in the nonopioid cohort at all time periods (P < .01). Patients in the nonopioid cohort had significantly lower Morphine Milligram Equivalents at discharge, 10-day, 6-week, and 90-day time periods (P < .01) when compared to the opioid cohort. When analyzing risk factors for continued opioid prescriptions for the whole cohort, preoperative opioid use within 90 days before surgery (relative risk 4.69 [95% confidence interval 3.18-6.91; P < .01) and current tobacco use (relative risk 2.61 [95% confidence interval 1.50-4.54]; P = .01) were associated with continued refills >30 days. Complications in the perioperative period and 90-day hospital readmissions were not significantly different between the groups. ConclusionA nonopioid postoperative regimen for patients undergoing TSA significantly reduces opioid prescriptions with similar patient-reported outcomes, subjective pain scores, and without increased complication rates compared to an opioid-only regimen.