Sarah Bhattacharjee
Pritzker School of Medicine at the University of Chicago
INTRODUCTION: Although studies have reported an association between postoperative non-steroidal anti-inflammatory drugs (NSAIDs) and nonunion rates for various fractures, the effects of postoperative opioids are undetermined. Recent animal models suggest that opioids may contribute to decreased bone fusion integrity, though these findings have yet to be translated in human studies. We sought to assess the relationship between postoperative opioid prescriptions and nonunion in openly treated long bone fractures. METHODS: Patients were identified using a national insurance claims database. All patients who underwent open procedures for fractures of the humerus, femur, radius and/or ulna, and tibia and/or fibula, were stratified by the presence of opioid prescriptions in the 3-month postoperative period. Nonunion rates in the 18-month postoperative period were compared between the opioid cohort and a control group with no opioid prescriptions in the postoperative 18-month period. We performed multiple logistic regression analyses for the independent risk factors of postoperative opioids, NSAIDs, and other comorbidities on nonunion. RESULTS: Of 2,374 fracture patients, 1,532 received opioid prescriptions while 842 were opioid naive. In our univariate analysis, we observed a significant difference between postoperative opioid prescriptions and increased nonunion rates in openly treated long bone fractures (11.81% vs. 4.16%; p < 0.0001). For each individual long bone fracture included in this study, the opioid cohort experienced consistently increased rates of nonunion compared to control. In our pooled regression analysis, we report a significant association between nonunion and postoperative opioids, NSAIDs, tobacco use, and history of diabetes. CONCLUSION: Our findings suggest that postoperative opioids may be associated with higher rates of nonunion in openly treated long bone fractures. This represents an important first step in translating animal models to human outcomes, indicating the necessity of further research into the effects of postoperative analgesics on fracture healing.
Comments