Effect of pre-injection opioid use on post-injection patient-reported outcomes following epidural steroid injections for radicular pain

Background Context Chronic opioid therapy is associated with worse patient-reported outcomes (PROs) following spine surgery. However, little literature exists on the relationship between opioid use and PROs following epidural steroid injections for radicular pain. Purpose We evaluated the association between pre-injection opioid use and PROs following spine epidural steroid injection. Study Design This study is a retrospective analysis of a prospective longitudinal registry database. Patient Sample A total of 392 patients within our database who were undergoing epidural steroid injections (ESIs) at our institution for degenerative structural spine diagnoses and met our inclusion criteria were included in this study. Outcome Measures Patient-reported outcomes for disability (Oswestry Disability Index/Neck Disability Index [ODI/NDI)]), quality of life (EuroQol-5D [EQ-5D]), and pain (Numerical Rating Scale scores for back pain, neck pain, leg pain, and arm pain [NRS-BP/NP/LP/AP]) were assessed at baseline and at 3 and 12 months post-injection. Methods Multivariable proportional odds logistic regression models were created to examine the relationship between pre-injection opioid use and post-injection PROs. A logistic regression with Bayesian Markov chain Monte Carlo parameter estimation was used to investigate a possible cutoff value of pre-injection opioid use above which the effectiveness of ESI (as measured by minimum clinically important difference [MCID] for ODI/NDI) decreases. Results A total of 276 patients with complete 12-month follow-up following ESI were analyzed. The mean pre-injection daily morphine equivalent amount (MEA) was 14.7 mg (95% confidence interval [CI] 12.4 mg–19.1 mg) for the cohort. Pre-injection opioid use was associated with slightly higher odds of worse disability (odds ratio [OR] 1.03, p=.03) and leg/arm pain (OR 1.01, p=.04) scores at 3 months post-injection only. No significant association between pre-injection opioid use and MCID for ODI/NDI was found, although a cutoff of 55.5 mg/day might serve as a significant threshold. Conclusion Increased pre-injection opioid use does not impact long-term outcomes after ESIs for degenerative spine diseases. A pre-injection MEA around 50 mg/day may represent a threshold above which the 3-month effectiveness of ESI for back- and neck-related disability decreases. Epidural steroid injection is an effective treatment modality for pain in patients using opioids, and can be part of a multimodal strategy for opioid independence.