Abstract
BACKGROUND: Smoking is known to impair bone metabolism, yet the extent to which current, former, and never smokers differ in fracture healing outcomes remains unclear. OBJECTIVE: To evaluate the association between smoking status and radiographic time to union in adults with long-bone fractures. METHODS: This retrospective cohort included adults treated for long-bone fractures at a UK major trauma centre (2022-2025). Smoking status (current, former with at least one year of abstinence, never) was extracted from clinical records. Radiographic union was defined as bridging of at least three of four cortices; the Radiographic Union Score for Tibial (RUST) fractures supported tibial assessments. Normality and variance assumptions were assessed using Shapiro-Wilk and Levene's tests. Differences in healing time were compared using analysis of variance (ANOVA) with appropriate post-hoc testing. Time-to-union was evaluated with Kaplan-Meier curves and multivariable Cox modelling, adjusting for age, sex, diabetes, peripheral vascular disease, open fracture, fracture site, and management. Hazard ratios (HRs) and 95% confidence intervals were reported. RESULTS: Of 234 patients (70 current, 70 former, 94 never smokers), healing time differed significantly among groups (20.1 ± 5.0 vs. 17.2 ± 4.0 vs. 15.3 ± 3.0 weeks; p < 0.001). Current smokers healed significantly slower than former and never smokers. Kaplan-Meier curves showed delayed union in current smokers (log-rank p < 0.001). Adjusted Cox regression demonstrated prolonged time to union among current smokers compared with never smokers (HR < 1). CONCLUSION: Smoking status demonstrates a clear gradient of risk for delayed fracture healing, with current smokers experiencing the longest time to union, former smokers showing partial recovery, and never smokers healing fastest. Smoking cessation counselling should be incorporated into fracture management pathways.