Is postoperative ketorolac administration associated with nonunion in adults after proximal humerus open reduction and internal fixation? a propensity-matched retrospective cohort study.

Introduction: Although ketorolac's association with poor bone healing remains debated, no study has examined the impact of ketorolac administration in adults with proximal humerus fractures (PHFs) after open reduction and internal fixation (ORIF), limiting surgeon decision-making. Therefore, the primary aim of this study was to examine the association between short-term ketorolac administration within the first three days after ORIF for PHF and the incidence and risk of nonunion or malunion through one year.

Methods: A pre-registered retrospective propensity-matched cohort study was performed using a large United States health records-based database (TriNetX, LLC). Patients included adults (≥18 years old) who underwent first-time proximal humerus ORIF and received either acute (≤3 days) postoperative ketorolac (ketorolac cohort) or acetaminophen (control cohort). The primary outcome was the risk ratio (RR) of nonunion through one year. Secondary outcomes explored the incidence and risk of reoperation by surgery type, other relevant postoperative adverse events (such as malunion), and RR and mean count of postoperative oral opioid prescription. Over fifteen risk factors associated with bone union were used for propensity matching.

Results: There were 2143 patients per cohort (n = 4286 total) with a mean age of 55 years. Comparing the ketorolac cohort to the control cohort, there was a statistically significant increase in risk of nonunion (p = 0.040; RR: 1.46 [1.02, 2.10]; 3.3% versus 2.2%; 70 patients versus 48 patients). Individual outcomes demonstrated no statistically significant difference in risk of malunion (p = 0.288; RR: 1.28; 1.9% versus 1.5%), revision ORIF (p = 0.493), total shoulder arthroplasty (p = 0.354), or acute kidney injury (p = 0.423). There was a statistically significant decrease in risk (p = 0.015) and mean count (p = 0.033) of oral opioid prescription.

Conclusion: Acute postoperative ketorolac after ORIF for PHF is associated with a modest increase in risk of nonunion and reduction in opioid prescriptions, with no significant differences in malunion, reoperation, or acute kidney injury. These findings support the need for individualized decision-making to weigh risks and benefits in postoperative pain management, with future research needed on dosages.