Objectives: In 2017, 67.8% of drug overdose deaths in the US were related to opioid use. Opioids are effective in short-term treatment of pain; however, no evidence shows their long-term effectiveness. Many patients have their first exposure to opioids after a pelvic injury, but not much is known about persistent use afterwards. This study aims to assess the prevalence and predictors of long-term opioid use following pelvic fractures secondary to trauma. We hypothesized that inpatient opioid use was associated with long-term opioid use.
Methods: A retrospective study enrolled patients ≥18 years with acute pelvic fractures from 2015 to 2020. Exclusion criteria were: transfers, no opioids received, and death before discharge. Daily and total morphine milligram equivalents (MME) were calculated. Primary outcome was long-term opioid use, defined as ongoing opioid use 60-90 days after discharge (60-90OP). Secondary outcome was intermediate-term opioid use, defined as ongoing use 30-60 days after discharge (30-60OP). Pelvic fractures were classified by the Tile system. Univariable and logistic regression analyses were performed with p< 0.05 considered as significant.
Results: After exclusions, 277 patients remained. Median age was 43 (interquartile range, 26-64) years. 61% were males and 5% had penetrating trauma. 231 (84%) had type A fractures, while 31 (11%) and 14 (5%) had types B and C respectively. Median total inpatient opioid MME was 422 (157-1667), with a median daily MME of 69 (26-145). 169 (61%) had an average inpatient daily use of ≥ 50 MME/day. The primary outcome, 60-90OP, occurred in 16% while 30-60OP occurred in 29%. Univariable analysis found that total inpatient opioid use and daily inpatient opioid use were each significantly associated with 60-90OP (median MME, 1241 vs 371; median MMEs, 127.7 vs 59.2, respectively) and 30-60OP (median MME, 1140 vs 326; median MMEs, 111.8 vs 57.9, respectively). Similarly, the use of ≥ 50 MME/day was significantly associated with both 30-60OP and 60-90OP. Logistic regression analysis found that total inpatient MME (OR 1.000132 [CI 1.000034 -1.000231]) and pelvic fracture type (B/C vs. A), (OR 3.97 [CI 1.77-8.89]) were independent predictors of 60-90OP.
Conclusion: Total and daily inpatient opioid use were each significantly associated with 60-90OP and 30-60OP, where there is a higher likelihood of post-discharge opioid use for patients who received ≥50 MME/day. Identifying the prevalence and risk factors for long-term opioid use after pelvic fractures requiring opioids for pain management is vital for informing clinical decision making to prevent adverse outcomes.
