Pain catastrophizing and prolonged opioid use following lumbar fusion

Healthcare providers commonly prescribe long-term opioid therapy for patients following lumbar fusion despite a lack of evidence that opioids are a safe or effective intervention for chronic pain. The purpose of this prospective, longitudinal study was to examine the prevalence and predictors of prolonged, prescribed opioid use in a cohort of 57 patients undergoing elective lumbar fusion. Prior to surgery, participants completed a demographic and clinical variables questionnaire and the Pain Catastrophizing Scale (PCS). Sixty-one percent (n = 35) of participants reported preoperative opioid use. The mean preoperative pain rating was 7.65 (SD = 1.87), and the mean pain catastrophizing score was 28.85 (SD = 14.72). Three months following lumbar fusion, participants self-reported their prescribed opioid use and their postoperative pain intensity. Forty-four percent (n = 22) of participants reported continued opioid use. The mean postoperative pain intensity rating was 3.12 (SD = 2.15). Pain catastrophizing was neither significantly correlated with time to opioid cessation (r = .03, p = .86), nor with postoperative pain intensity (r = -.04, p = .82). Multiple regression analysis was conducted to identify the best combination of age, sex, employment status, educational level, preoperative pain intensity, preoperative opioid use, disability status, and pain catastrophizing to predict time to opioid cessation. Bivariate analysis identified a strong correlation between time to opioid cessation and preoperative opioid use (r = .46, p = .000), and a moderate correlation between time to opioid cessation and disability (r = .29, p = .022). Multiple regression analysis indicated that preoperative patient characteristics predicted prolonged, postoperative opioid use, and accounted for 18% of the variance in time to opioid cessation [R2 = .322, R2adj = .179, F(8, 38) = 2.254, p = .044]. Among preoperative patient characteristics, preoperative opioid use was the sole predictor that significantly contributed to the model (β = .466; p = .005). Thus, screening patients for opioid use prior to lumbar fusion may help to identify patients at increased risk of prolonged opioid use following lumbar fusion.