Long-Term Opioid Prescriptions After Spine Surgery: A Meta-Analysis of Prevalence and Risk Factors

BACKGROUND:Opioids are frequently prescribed for back pain, but the prevalence of and risk factors for long-term opioid use after spine surgery were not clearly reported. We conducted a systematic review and meta-analysis to summarize the evidence for long-term opioid use (>90 days) among adults who underwent spine surgery. METHODS:PubMed, EMBASE, and Cochrane indexing databases were searched until November 9, 2018 for studies reporting the prevalence of and risk factors for long-term opioid use after spine surgery. Separate meta-analyses were conducted for commercial claims databases or registries (claims/registries) and nonclaims observational studies using the random-effects model to estimate the pooled odds ratio (OR). Prevalence meta-analysis was performed in a clinically homogeneous subset of these patients who underwent lumbar spine surgery. RESULTS:Eight claims and 5 nonclaims were meta-analyzed to avoid double-counting participants. The meta-analysis showed that preoperative opioid users (OR, 5.59; 95% confidence interval [CI], 3.37-9.27 vs. OR 4.21; 95% CI, 2.72-6.51) and participants with preexisting depression and/or anxiety (OR, 1.86, 95% CI, 1.43-2.42 and OR, 1.20; 95% CI, 0.83-1.74, respectively) had a statistically significantly higher odds of long-term postoperative opioids, compared with their peers. Males showed lower odds of long-term postoperative opioid use in the claims group (OR, 0.85; 95% CI, 0.79-0.92), but not in the nonclaims group (OR, 0.99; 95% CI, 0.71-1.39). The pooled prevalence of post-lumbar spine surgery long-term opioid use was 63% (95% CI, 50%-74%) in claims and 47% (95% CI, 38%-56%) in nonclaims. CONCLUSIONS:Patients undergoing spine surgery represent a high-risk surgical population requiring special attention and targeted interventions, with the strongest evidence for those treated with opioids before surgery and those with psychiatric comorbidities.