Post-injury use of opioid analgesics in patients with traumatic injury: A registry-based study.

Background: Opioid analgesics are commonly used to treat acute and chronic pain following traumatic injury. Psychiatric comorbidity has been reported to be associated with increased pain and persistent opioid use. Our aims were to determine the extent of post-injury opioid use and assess whether pre-injury antidepressant, benzodiazepine, and z-hypnotic drug use is associated with increased post-injury opioid use.

Methods: Data on trauma patients aged 15 years and older included in the Oslo University Hospital Trauma Registry between 2005 and 2014 was linked to data from the Norwegian Prescription Database. We identified opioid dispensing within the first 90 and 365 days following trauma and determined subsequent persistent use. Multivariable logistic regression was applied to examine associations between pre- and post-injury drug use.

Results: 3912 of 11,057 patients (35.4%) had opioids dispensed within 90 days after trauma, and 4644 (42.0%) within 365 days. Among 9800 previously opioid-naïve, the proportions were 33.0% and 39.0%, respectively. One year after the first post-injury opioid dispensing, 9.6% of all opioid users and 4.5% of new users were defined as persistent users. Pre-injury benzodiazepine use and z-hypnotic use was associated with new persistent opioid use (adjusted odds ratio [aOR] 2.25; 95% confidence interval [CI] 1.47-3.45, and aOR 2.04; 95% CI 1.33-3.13, respectively), whereas pre-injury antidepressant use was not (aOR 1.49; 95% CI 0.97-2.30).

Conclusions: Opioid use after trauma is widespread. Development of persistent use is limited, particularly in previously opioid-naïve patients. Pre-injury benzodiazepine or z-hypnotic use seem to increase odds of new persistent use.

Significance statement: This large registry-based study adds to the body of knowledge on opioid use beyond in-hospital care in patients having sustained traumatic injury, a field which is scarcely investigated and not yet fully understood. It suggests that both previous drug therapy and the nature of opioid treatment initiation may affect outcome. This will help guide clinicians in selecting the appropriate pain management in this patient group.