Exploration of Overdose Risk Score and Postoperative Complications and Health Care Use After Total Knee Arthroplasty
Ahmed K. Emara,Daniel Santana,Daniel Grits,Alison K. Klika,Viktor E. Krebs,Robert M. Molloy,Nicolas S. Piuzzi
DOI: https://doi.org/10.1001/jamanetworkopen.2021.13977
2021-06-28
JAMA Network Open
Abstract:Importance: The adverse outcomes after total knee arthroplasty (TKA) associated with preoperative prescription drug use (ie, use of narcotics, sedatives, and stimulants) have been established but are not well quantified.Objective: To test the association of preoperative overdose risk score (ORS) with postoperative health care use.Design, Setting, and Participants: This cohort study was conducted using data on a consecutive sample of individuals who underwent primary TKA from November 2018 through March 2020 at a tertiary care health system. Data were collected using the Orthopaedic Minimal Data Set Episode of Care, a validated data-collection system for all elective orthopedic surgical interventions taking place within the health care system. Outcomes were assessed at 90 days postoperatively. Individuals whose preoperative baseline characteristics or ORS were not provided or who declined to participate were excluded. Data were analyzed from September through October 2020.Exposure: Patient-specific preoperative ORS, as measured using NarxCare, associated with patterns of prescription drug use.Main Outcomes and Measures: Associations between patient-specific ORS categories and 90-day postoperative health care use (ie, prolonged hospital length of stay [LOS; ie, >2 days], nonhome discharge, all-cause 90-day readmission, emergency department [ED] visits, and reoperation) were evaluated. Outcomes were also compared between a group of individuals with ORS less than 300 vs those with ORS 300 or greater who were propensity score matched (4:1; caliper, 0.1) using demographic characteristics (ie, age, sex, race, body mass index, and smoking status) and baseline comorbidities.Results: Among 4326 individuals who underwent primary TKA, 2623 (60.63%) were women, 3602 individuals (83.26%) were White, the mean (SD) BMI was 32.8 (6.9), and the mean (SD) age was 66.6 (9.2) years; 90-day follow-up was available for the entire cohort. The predominant preoperative diagnosis was osteoarthritis, occurring among 4170 individuals (96.4%). For individuals with an ORS of 300 to 399, there were significantly higher odds of a prolonged LOS (odds ratio [OR], 2.03; 95% CI, 1.46-2.82; P < .001), nonhome discharge (OR, 2.01; 95% CI, 1.37-2.94; P < .001), all-cause 90-day readmission (OR, 1.56; 95% CI, 1.01-2.42; P < .001), and ED visits (OR, 1.62; 95% CI, 1.11-2.38; P = .01) compared with individuals who were prescription drug naive (ie, ORS = 0). Individuals in the highest ORS category (ie, ORS ≥ 500) had the highest ORs for prolonged LOS (OR, 3.71; 95% CI, 2.00-6.87; P < .001), nonhome discharge (OR, 4.09; 95% CI, 2.02-8.29; P < .001), 90-day readmission (OR, 4.41; 95% CI, 2.23-8.71; P < .001), and 90-day reoperation (OR, 6.09; 95% CI, 1.44-25.80; P = .01). Propensity score matching confirmed the association between an ORS of 300 or greater and the incidence of prolonged LOS (244 individuals [11.6%] vs 130 individuals [23.0%]; P < .001), nonhome discharge (176 individuals [8.4%] vs 93 individuals [16.4%]; P < .001), all-cause 90-day readmission (119 individuals [5.7%] vs 65 individuals [11.5%]; P < .001), and all-cause ED visits (198 individuals [9.4%] vs 76 individuals [13.4%]; P = .006).Conclusions and Relevance: This study found that higher ORS was associated with increased health care use after primary TKA. These findings suggest that an ORS of 300 or greater could be used to designate increased risk and guide the preoperative surgeon-patient discussion to modify prescription drug use patterns.
medicine, general & internal