Preoperative opioid use disorder has a negative impact on the rate of short-term complications following total and reverse shoulder replacement

Abstract

BackgroundMore than 2 million Americans and 16 million individuals worldwide are estimated to have an opioid use disorder (OUD). It is imperative that we understand the impact OUD has on patients who require surgical procedures. Preoperative OUD has been associated with increased postoperative complications, lengths of stay, and readmissions in elective procedures. The goal of this study was to elucidate the impact that preoperative OUD has on short-term outcomes following total and reverse shoulder arthroplasty. MethodsPatient information was collected using the National Readmissions Database (2016-2018). Persons who underwent anatomic and reverse shoulder arthroplasty were identified using International Classification of Diseases 10th Revision procedural codes (anatomic: 0RRJ0JZ/0RRK0JZ, reverse: 0RRJ00Z, 0RRK00Z). OUD was defined using the following International Classification of Diseases 10th Revision codes: F11.10, F11.11, F11.20, and F11.21. Patient demographics, comorbidity indices (Charlson and Elixhauser), complications, readmission rates, healthcare costs, and lengths of stay were compared between OUD and non-OUD patients. A univariate analysis was first run with OUD as the independent variable followed by a multivariate analysis to identify if OUD was an independent predictor of adverse outcomes. ResultsOne hundred sixty four thousand and five hundred twenty two total patients were identified (OUD: 1098 vs. non-OUD: 163,424). The OUD patients were younger (64.3 ± 10.0 years vs. 69.4 ± 9.5 years, P < .01), had longer lengths of stay (2.5 ± 2.6 days vs. 1.8 ± 2.0 days, P < .01), and had higher cumulative hospital costs ($83,424.88 ± $53,565.26 vs. $73,099.65 ± $47,548.65, P < .01) on average. Univariate analysis revealed that OUD patients had increased rates of respiratory, intraoperative, hardware, and prosthetic complications, dislocations, infection (surgical site and prosthetic), revision surgeries, and 30-day/90-day readmissions (P < .01). Interestingly, through a multivariate analysis, OUD was predictive for less revision surgeries (odds ratio [confidence interval 95%]: 0.69 [0.48-0.99], P = .045) but trended toward being a predictor for increased rates of 90-day readmissions (odds ratio [confidence interval 95%]: 1.28 [0.96-1.72], P = .09). ConclusionOUD patients were younger on average and had increased lengths of stay, healthcare costs, and risks for developing operative, postoperative, and prosthetic complications compared to non-OUD patients undergoing shoulder replacement. However, OUD was not found to be an independent predictor of adverse outcomes. Together, these findings should encourage surgeons to address substance use disorders with all patients preoperatively in an attempt to mitigate potential complications while not identifying OUD as an exclusive operative contraindication.