Predictors of long-term opioid dependence in transforaminal lumbar interbody fusion with a focus on pre-operative opioid usage

Purpose Predictors of long-term opioid usage in TLIF patients have not been previously explored in the literature. We examined the effect of pre-operative narcotic use in addition to other predictors of the pattern and duration of post-operative narcotic usage. Methods We conducted a retrospective cohort study at a single academic institution of patients undergoing a one- or two-level primary TLIF between 2014 and 2017. Total oral morphine milligram equivalents (MMEs) for inpatient use were calculated and used as the common unit of comparison. Results A multivariate binary logistic regression (R-2 = 0.547, specificity 95%, sensitivity 58%) demonstrated that a psychiatric or chronic pain diagnosis (OR 3.95, p = 0.013, 95% CI 1.34-11.6), pre-operative opioid use (OR 8.65, p < 0.001, 95% CI 2.59-29.0), ASA class (OR 2.95, p = 0.025, 95% CI 1.14-7.63), and inpatient total MME (1.002, p < 0.001, 95% CI 1.001-1.003) were positive predictors of prolonged opioid use at 6-month follow-up, while inpatient muscle relaxant use (OR 0.327, p = 0.049, 95% CI 0.108-0.994) decreased the probability of prolonged opioid use. Patients in the pre-operative opioid use group had a significantly higher rate of opioid usage at 6 weeks (79% vs. 46%, p < 0.001), 3 months (51% vs. 14%, p < 0.001), and 6 months (40% vs. 5%, p < 0.001). Conclusions Pre-operative opioid usage is associated with higher total inpatient opioid use and a significantly higher risk of long-term opiate usage at 6 months. Approximately 40% of pre-operative narcotic users will continue to consume narcotics at 6-month follow-up, compared with 5% of narcotic-naive patients. Graphic abstract These slides can be retrieved under Electronic Supplementary Material.