Preoperative Corticosteroid Injection Timing and Revision Risk After Midfoot Arthrodesis.

Abstract

Midfoot osteoarthritis (OA) is commonly treated with intraarticular corticosteroid injections before surgical fusion. However, the impact of preoperative injection timing on postoperative outcomes remains unclear. This study evaluates the association between the timing of corticosteroid injections and postoperative 90-day complications and 1-year revision following midfoot arthrodesis. We conducted a retrospective cohort study using the PearlDiver Mariner database (2015-2024), identifying 82 144 patients who underwent midfoot arthrodesis. Patients were stratified into 5 groups based on the timing of corticosteroid injection before surgery: ≤1 month, >1-2 months, >2-3 months, ≥3 months (reference), and no preoperative injection (control). Outcomes included 90-day postoperative complications and revision surgery within 1 year of the procedure. Multivariable logistic regression and Cox proportional hazards models were used to evaluate adjusted odds and time-to-event outcomes, respectively. Unadjusted 90-day complication rates were higher among injection cohorts than controls; however, in adjusted logistic models, odds of complications did not differ significantly vs the ≥3-months reference (unmatched and propensity-matched). In time-to-event analyses, preoperative injections administered within 3 months of surgery were associated with a significantly higher hazard of 1-year revision vs ≥3 months (HRs ~1.25-1.49). Older age, male sex, and greater comorbidity burden were also associated with higher risk. Preoperative corticosteroid injections administered within 3 months of midfoot arthrodesis were associated with increased rates of 1-year revision compared with injections given ≥3 months preoperatively, whereas adjusted odds of 90-day complications were similar across timing groups. These timing-related risks may inform preoperative planning for elective midfoot fusion but should be interpreted cautiously given the limitations of administrative claims data and residual confounding by indication. Level III, retrospective cohort study.