Do non-steroidal anti-inflammatory medications or opioids increase the risk of nonunion after surgical and nonsurgical treatment of tibial shaft fractures?

Background: Nonsteroidal anti-inflammatory drug (NSAID) medications have been implicated in fracture nonunion. However, the evidence to date for this association is limited, due in part to the rarity of the outcome. Prior investigations may not sufficiently account for confounding, limiting the ability to draw accurate conclusions regarding the independent association of NSAID exposure with subsequent nonunion rates. Opioids are often used instead of NSAIDs to treat postoperative pain, but the association between opioid use and nonunion risk is unclear. The purpose of this study was to investigate the impact of NSAID use and opioid use on rates of nonunion in in skeletally mature patients with tibial shaft fractures. Patients and Methods: This was a cohort study using the Merative Marketscan Research databases from January 1 2007 to December 31, 2022, which contain inpatient, outpatient, and pharmacy claims data for roughly 273 million individuals covered by employer-sponsored commercial health insurance throughout all US states and territories. Patients were included if they were greater than 16 and less than 65 years of age with an ICD-9 or ICD-10 diagnosis of a fracture of the tibia or tibia and fibula between 2008 and 2022. 297,334 patients met this inclusion criteria. The exposure for the primary aim was defined as having 1 or more fills of an NSAID in the 0-30 days following fracture diagnosis, as compared to no fills of an NSAID in the 0-30 days following fracture diagnosis. The exposure for the secondary aims was defined as having 1 or more fills of an opioid in the 0-30 days following fracture diagnosis, as compared to no fills of an opioid (aim 2) or NSAID (aim 3) in the 0-30 days following fracture diagnosis. The outcome for all analyses was fracture nonunion, defined as a surgical procedure performed to treat tibial nonunion between 90 and 365 days after fracture diagnosis. Propensity score analysis, including injury, treatment, and demographic variables associated with nonunion, was used to compare nonunion rates between the groups. Cox proportional hazard regression was performed to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs), adjusted for the variables included in the propensity match. Results: For patients filling NSAIDs in the 30 days following fracture diagnosis, the adjusted Cox proportional hazard ratio (HR) for nonunion was 1.60 (95% CI 1.20-2.15, p=0.002), indicating a greater risk of nonunion for patients who filled a prescription for NSAIDs within 30 days of the fracture compared to those who did not. Similarly, we observed more treatments for nonunion in the group that filled opioids (1.3% versus 0.9%) and a statistically significant greater rate of nonunion for patients filling opioids compared to those not filling opioids (HR 1.41, 95% CI 1.19-1.66, p<0.001). There was no significant difference in the risk of nonunion for patients who filled NSAID prescriptions compared to those who filled opioid prescriptions (HR 0.89, 95% CI 0.45-1.76, p=0.73) Conclusion: In this propensity matched claims analysis of patients aged 17-64 with tibial shaft fractures, filling NSAIDs and filling opioids postoperatively were separately associated with increased risk of surgical treatment for nonunion compared to patients who did not fill NSAIDs or opioids. Patients who filled NSAIDs only had no difference in the rate of surgical treatment for nonunion compared to patients who filled opioids only. Thus, both NSAIDs and opioids are associated with increased risk of nonunion, but neither appears to be of substantially higher risk than the other. Physicians should consider the side effect profiles of both drugs when deciding the type, quantity, and duration of pain medications to prescribe postoperatively. Level of Evidence: level III Keywords: NSAIDs; non-steroidal anti-inflammatory medications; tibial shaft fractures; tibial nonunion; nonunion Funding: This work was supported by the University of Washington Clinical Learning, Evidence, And Research (CLEAR) Center for Musculoskeletal Disorders, Administrative, Methodologic and Resource Cores and NIAMS/NIH grant P30AR072572. The funding source had no role in the study design, collection, analysis and interpretation of the data, writing of the report, or the decision to submit this article for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.