Opioid exposure is associated with nonunion risk in a traumatically injured population: An inception cohort study.

INTRODUCTION

Certain common medications are associated with an elevated risk of fracture and recent data suggests that medications can also increase nonunion risk. Medication use is a modifiable nonunion risk factor, but it is unknown whether risk accrues solely to chronic medication use or whether there is also risk inherent to acute use.

METHODS

Multivariate logistic regression was used in an inception cohort to calculate odds ratios (OR) for fracture nonunion associated with medication use, in context with other risk factors demonstrated to influence nonunion. Patient-level health claims for medical and drug expenses were compiled from a payer database. Patients were included if they had a fracture coded in 2011, with continuous enrollment for 1 month prior to and 12 months after fracture. The database contained demographic descriptors, treatment procedures per CPT codes, co-morbidities per ICD-9 codes, and prescriptions per National Drug Codes. Chronic medication use was defined as ≥30 days of prescription prior to fracture with ≥1 day afterward; acute use was any other prescription.

RESULTS

Most non-analgesic medications were safe in acute or chronic use, but risk of nonunion was elevated for a wide range of analgesics. Overall, 45,085 fractures (14.6% of fractures) affected patients using chronic opioids. Nonunion OR was elevated for acute and chronic use of Schedule 2 opioids including acetaminophen/oxycodone, hydromorphone, oxycodone, and acetaminophen/hydrocodone bitartrate, as well as Schedule 3-5 opioids including tramadol (all, p < 0.0001). The highest ORs were associated with chronic administration of Schedule 2 opioids.

DISCUSSION

Most medications do not increase nonunion risk, but acute and chronic use of NSAIDs or opioids was associated with impaired fracture healing. There is particular risk in prescribing opioid analgesics for fracture, though literature suggests that roughly half of opioid-naïve patients receive such a prescription.

CONCLUSIONS

Patients evaluated in this study were not a random sample of Americans; they may approximate a random sample of the Emergency Department population in the United States. Thus, trauma patients may represent a population enriched for nonunion risk factors. Opioids impair recovery from injury; if they also predispose to injury, the ongoing opioid epidemic could presage an increase in nonunion prevalence.