Adding vancomycin to perioperative prophylaxis decreases deep sternal wound infections in high-risk cardiac surgery patients.

OBJECTIVES: Perioperative prophylaxis with cephalosporins reduces sternal wound infections (SWIs) after cardiac surgery. However, more than 50% of coagulase-negative staphylococci, an important pathogen, are cephalosporin resistant. The aim of this study was to determine the impact of adjunctive vancomycin on SWIs in high-risk patients. METHODS: We conducted a pre- and postintervention study in an academic hospital. Preintervention (2010-2011), all patients received prophylaxis with 1.5 g of cefuroxime for 48 h. During the intervention period (2012-2013), high-risk patients additionally received 1 g of vancomycin. High-risk status was defined as body mass index ≤18 or ≥ 30 kg/m2, reoperation, renal failure, diabetes mellitus, chronic obstructive pulmonary disease or immunosuppressive medication. Time series analysis was performed to study SWI trends and logistic regression to determine the effect of adding vancomycin adjusting for high-risk status. RESULTS: A total of 3902 consecutive patients (n = 1915 preintervention and n = 1987 postintervention) were included, of which 1493 (38%) patients were high-risk patients. In the high-risk group, 61 of 711 (8.6%) patients had SWI before and 30 of 782 (3.8%) patients after the intervention. Focusing on deep SWI (DSWI), 33 of 711 (4.6%) patients had DSWI before and 13 of 782 (1.7%) patients afterwards; the absolute risk difference of 2.9% yielded a number-needed-to-treat of 34 to prevent 1 DSWI. Corrected for high-risk status, adding vancomycin significantly reduced the overall SWI rate (odds ratio 0.42, 95% confidence interval 0.26-0.67; P < 0.001) and the subset of DSWI (odds ratio 0.30, 95% confidence interval 0.14-0.62; P = 0.001). The rate of SWI in low-risk patients remained unchanged. CONCLUSIONS: Adding vancomycin to standard antibiotic prophylaxis in high-risk patients significantly reduced DSWI after cardiac surgery.

Risk Factors for Sternal Wound Infection After Open Heart Operations Vary According to Type of Operation.

BACKGROUND: This study evaluated whether risk factors for sternal wound infections vary with the type of surgical procedure in cardiac operations. METHODS: This was a university hospital surveillance study of 3,249 consecutive patients (28% women) from 2006 to 2010 (median age, 69 years [interquartile range, 60 to 76]; median additive European System for Cardiac Operative Risk Evaluation score, 5 [interquartile range, 3 to 8]) after (1) isolated coronary artery bypass grafting (CABG), (2) isolated valve repair or replacement, or (3) combined valve procedures and CABG. All other operations were excluded. Univariate and multivariate binary logistic regression were conducted to identify independent predictors for development of sternal wound infections. RESULTS: We detected 122 sternal wound infections (3.8%) in 3,249 patients: 74 of 1,857 patients (4.0%) after CABG, 19 of 799 (2.4%) after valve operations, and 29 of 593 (4.9%) after combined procedures. In CABG patients, bilateral internal thoracic artery harvest, procedural duration exceeding 300 minutes, diabetes, obesity, chronic obstructive pulmonary disease, and female sex (model 1) were independent predictors for sternal wound infection. A second model (model 2), using the European System for Cardiac Operative Risk Evaluation, revealed bilateral internal thoracic artery harvest, diabetes, obesity, and the second and third quartiles of the European System for Cardiac Operative Risk Evaluation were independent predictors. In valve patients, model 1 showed only revision for bleeding as an independent predictor for sternal infection, and model 2 yielded both revision for bleeding and diabetes. For combined valve and CABG operations, both regression models demonstrated revision for bleeding and duration of operation exceeding 300 minutes were independent predictors for sternal infection. CONCLUSIONS: Risk factors for sternal wound infections after cardiac operations vary with the type of surgical procedure. In patients undergoing valve operations or combined operations, procedure-related risk factors (revision for bleeding, duration of operation) independently predict infection. In patients undergoing CABG, not only procedure-related risk factors but also bilateral internal thoracic artery harvest and patient characteristics (diabetes, chronic obstructive pulmonary disease, obesity, female sex) are predictive of sternal wound infection. Preventive interventions may be justified according to the type of operation.