Thirteen-year experience with universal Staphylococcus aureus nasal decolonization prior to cardiac surgery: a quasi-experimental study.

INTRODUCTION: Sternal wound infection (SWI) after cardiac surgery is a severe complication. Among preventive measures, pre-operative decolonization of nasal carriage of Staphylococcus aureus has recently been shown to be beneficial. This quasi-experimental study assessed the effect of decolonization on the incidence of S. aureus-associated SWI based on 19 years of prospective surveillance.

METHODS: Segmented negative binomial regression was used to analyse the change over time in the incidence of S. aureus mediastinitis requiring re-operation after cardiac surgery in a French university hospital between 1996 and 2014. Universal nasal decolonization with mupirocin was introduced in December 2001. The association between pre-operative nasal carriage and SWI due to S. aureus was analysed between 2006 and 2012.

RESULTS: Among 17,261 patients who underwent a cardiac surgical procedure, 565 developed SWI (3.3%), which was caused by S. aureus in 181 cases (1%). The incidence of mediastinitis caused by S. aureus decreased significantly over the study period (1.43% in 1996-2001 vs 0.61% and 0.64% in 2002-2005 and 2006-2014, respectively; P<0.001). In segmented analysis, there was a significant break in 2002, corresponding to the introduction of decolonization. Despite this intervention, pre-operative nasal carriage remained a significant risk factor for S. aureus mediastinitis (adjusted odds ratio 2.2; 95% confidence interval 1.2-4.2), as were obesity, critical pre-operative status, coronary artery bypass grafting (CABG), and combined surgery with valve replacement and CABG.

CONCLUSION: Universal nasal decolonization before cardiac surgery was effective in decreasing the incidence of mediastinitis caused by S. aureus. Nasal carriage of S. aureus remained a risk factor for S. aureus-associated SWI.