RESUMO
BACKGROUND: Although presurgical nasal decontamination with mupirocin (NDM) has been advocated as a measure for preventing postsurgical mediastinitis (PSM) due to Staphylococcus aureus, this strategy is not universally recommended due to lack of robust supporting evidence. We aimed to evaluate the role of preoperative NDM in the annual incidence of S. aureus PSM at our institution. METHODS: An interrupted time-series analysis, with an autoregressive error model, was applied to our single-center cohort by comparing preintervention (1990-2003) and postintervention (2005-2018) periods. Logistic regression was performed to analyze risk factors for S. aureus PSM. RESULTS: 12 236 sternotomy procedures were analyzed (6370 [52.1%] and 5866 [47.9%] in the pre- and postintervention periods, respectively). The mean annual percentage adherence to NDM estimated over the postintervention period was 90.2%. Only 4 of 127 total cases of S. aureus PSM occurred during the 14-year postintervention period (0.68/1000 sternotomies vs 19.31/1000 in the preintervention period; Pâ <â .0001). Interrupted time-series analysis demonstrated a statistically significant annual reduction in S. aureus PSM of -9.85 cases per 1000 sternotomies (-13.17 to -6.5; Pâ <â .0001) in 2005, with a decreasing trend maintained over the following 5 years and an estimated relative reduction of 84.8% (95% confidence interval [CI], 89.25-74.09%). Chronic obstructive pulmonary disease was the single independent risk factor for S. aureus PSM (odds ratio, 3.7; 95% CI, 1.72-7.93) and was equally distributed in patients undergoing sternotomy during pre- or postintervention periods. CONCLUSIONS: Our experience suggests the implementation of preoperative NDM significantly reduces the incidence of S. aureus PSM.