The Impact of Smoking on Hospital Course and Postoperative Outcomes in Patients With Fracture-Related Infections

OBJECTIVES: To assess the relationship between patient smoking status and fracture-related infection (FRI) characteristics including patient symptoms at FRI presentation, bacterial species of FRI, and rates of fracture union. METHODS: Design: Retrospective cohort study. Setting: Urban level 1 trauma center. Patient Selection Criteria: All patients undergoing reoperation for FRI from January 2013 to April 2021 were identified through manual review of an institutional database. Outcome Measures and Comparisons: Data including patient demographics, fracture characteristics, infection presentation, and hospital course were collected through review of the electronic medical record. Patients were grouped based on current smoker versus nonsmoker status. Hospital course and postoperative outcomes of these groups were then compared. Risk factors of methicillin-resistant Staphylococcus aureus (MRSA) infection, Staphylococcus epidermidis infection, and sinus tract development were evaluated using multivariable logistic regression. RESULTS: A total of 301 patients, comprising 155 smokers (51%) and 146 nonsmokers (49%), undergoing FRI reoperation were included. Compared with nonsmokers, smokers were more likely male (69% vs. 56%, P = 0.024), were younger at the time of FRI reoperation (41.7 vs. 49.5 years, P < 0.001), and had lower mean body mass index (27.2 vs. 32.0, P < 0.001). Smokers also had lower prevalence of diabetes mellitus (13% vs. 25%, P = 0.008) and had higher Charlson Comorbidity Index 10-year estimated survival (93% vs. 81%, P < 0.001). Smokers had a lower proportion of S. epidermidis infections (11% vs. 20%, P = 0.037), higher risk of nonunion after index fracture surgery (74% vs. 61%, P = 0.018), and higher risk of sinus tracts at FRI presentation (38% vs. 23%, P = 0.004). On multivariable analysis, smoking was not found to be associated with increased odds of MRSA infection. CONCLUSIONS: Among patients who develop a FRI, smokers seemed to have better baseline health regarding age, body mass index, diabetes mellitus, and Charlson Comorbidity Index 10-year estimated survival compared with nonsmokers. Smoking status was not significantly associated with odds of MRSA infection. However, smoking status was associated with increased risk of sinus tract development and nonunion and lower rates of S. epidermidis infection at the time of FRI reoperation. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.