Dhiren S Sheth1, Ngoc Ho2, Jose R Pio2, Peggy Zill3, Stephanie Tovar2, Robert S Namba1. 1. Department of Orthopedic Surgery, Southern California Permanente Medical Group, Irvine Medical Center, Irvine, CA. 2. Department of Research and Evaluation, Southern California Permanente Medical Group, Pasadena CA. 3. Department of Pharmacy, Southern California Permanente Medical Group, Irvine Medical Center, Irvine, CA.
Abstract
BACKGROUND: Forty percent of patients continue to use opioids at 3 months after joint arthroplasty. We sought to identify clinical and psychological risk factors associated with prolonged opioid use. METHODS: In this prospective study, psychological profile data were collected preoperatively. Prolonged use was defined as dispensation of an opioid after 90 days. Logistic regressions were used for univariate and multivariate modeling and to create receiver operating characteristic curves. A backward stepwise regression analysis was used to select significant factors in the multivariable model. RESULTS: The study included 258 patients (163 total knee arthroplasty, 95 total hip arthroplasty). 29.84% of patients were on preoperative opioids and 14% (37 of 258) of patients had prolonged use of opioids. In the univariate analysis, age <65, associated back pain, chronic pain syndrome or fibromyalgia, prior opioid use, drug potency of more than 10 morphine equivalent, and total score on Opioid Risk Tool of more than 7 were associated with prolong use. In the multivariate analysis, age <65, associated back pain, chronic pain, and preoperative use of opioids were significant risk factors for prolonged use (combined area under the curve = 0.83). Preoperative opioid use had the highest area under the curve = 0.72 (P = .0005). Psychological profile tests did not predict prolonged opioid use. CONCLUSION: Preoperative opioid use was the strongest predictor of postoperative prolonged opioid use. Younger age, associated backpain, and chronic pain syndrome were the other identified risk factors. Screening tools to detect aberrant drug-related behavior may be more helpful than those for depression or pain catastrophizing.
BACKGROUND: Forty percent of patients continue to use opioids at 3 months after joint arthroplasty. We sought to identify clinical and psychological risk factors associated with prolonged opioid use. METHODS: In this prospective study, psychological profile data were collected preoperatively. Prolonged use was defined as dispensation of an opioid after 90 days. Logistic regressions were used for univariate and multivariate modeling and to create receiver operating characteristic curves. A backward stepwise regression analysis was used to select significant factors in the multivariable model. RESULTS: The study included 258 patients (163 total knee arthroplasty, 95 total hip arthroplasty). 29.84% of patients were on preoperative opioids and 14% (37 of 258) of patients had prolonged use of opioids. In the univariate analysis, age <65, associated back pain, chronic pain syndrome or fibromyalgia, prior opioid use, drug potency of more than 10 morphine equivalent, and total score on Opioid Risk Tool of more than 7 were associated with prolong use. In the multivariate analysis, age <65, associated back pain, chronic pain, and preoperative use of opioids were significant risk factors for prolonged use (combined area under the curve = 0.83). Preoperative opioid use had the highest area under the curve = 0.72 (P = .0005). Psychological profile tests did not predict prolonged opioid use. CONCLUSION: Preoperative opioid use was the strongest predictor of postoperative prolonged opioid use. Younger age, associated backpain, and chronic pain syndrome were the other identified risk factors. Screening tools to detect aberrant drug-related behavior may be more helpful than those for depression or pain catastrophizing.