OBJECTIVES: Guidelines recommend antibiotic prophylaxis 60 minutes before skin incision; however, it is unclear whether more precise timing would further reduce sternal wound infection. Our objectives were to examine the relationship between antibiotic timing and infection, test potential efficacy of optimal antibiotic timing in preventing infection, and determine whether patient comorbidity is related to timing and infection. METHODS: From 1/1/1995-1/1/2008, 28,250 patients underwent 28,702 cardiac surgical procedures involving a median sternotomy; 85% received only cefuroxime and 15% received only vancomycin prophylaxis. Multivariable analysis identified factors associated with infection within each phase, and risk-adjusted optimal timing was determined using patient data, risk variables, and hypothetical values of antibiotic timing. RESULTS: Prevalence of sternal wound infection was 2.0% (489 patients) for cefuroxime and 2.3% (101 patients) for vancomycin. Minimum prevalence for infection was 1.8% observed when cefuroxime was administered 15 minutes before incision; risk increased to 2.2% with administration more than 45 minutes before incision and to 2.8% at 60 minutes before incision. Minimum prevalence of infection in patients who received vancomycin was 1.8% observed with initiation 32 minutes before incision; risk increased to 2.2% for administration 45 minutes before incision and 3.2% with administration 60 minutes before incision. Simulation for optimal timing found that it was influenced by phase-specific risk factors. CONCLUSIONS: Refining current antibiotic prophylaxis guidelines may lower sternal wound infections. Antibiotic administration timing resulting in lowest likelihood for infection varied with antibiotic and patient-specific factors. Optimal risk-adjusted timing could potentially reduce infections by 9%-31%.
OBJECTIVES: Guidelines recommend antibiotic prophylaxis 60 minutes before skin incision; however, it is unclear whether more precise timing would further reduce sternal wound infection. Our objectives were to examine the relationship between antibiotic timing and infection, test potential efficacy of optimal antibiotic timing in preventing infection, and determine whether patient comorbidity is related to timing and infection. METHODS: From 1/1/1995-1/1/2008, 28,250 patients underwent 28,702 cardiac surgical procedures involving a median sternotomy; 85% received only cefuroxime and 15% received only vancomycin prophylaxis. Multivariable analysis identified factors associated with infection within each phase, and risk-adjusted optimal timing was determined using patient data, risk variables, and hypothetical values of antibiotic timing. RESULTS: Prevalence of sternal wound infection was 2.0% (489 patients) for cefuroxime and 2.3% (101 patients) for vancomycin. Minimum prevalence for infection was 1.8% observed when cefuroxime was administered 15 minutes before incision; risk increased to 2.2% with administration more than 45 minutes before incision and to 2.8% at 60 minutes before incision. Minimum prevalence of infection in patients who received vancomycin was 1.8% observed with initiation 32 minutes before incision; risk increased to 2.2% for administration 45 minutes before incision and 3.2% with administration 60 minutes before incision. Simulation for optimal timing found that it was influenced by phase-specific risk factors. CONCLUSIONS: Refining current antibiotic prophylaxis guidelines may lower sternal wound infections. Antibiotic administration timing resulting in lowest likelihood for infection varied with antibiotic and patient-specific factors. Optimal risk-adjusted timing could potentially reduce infections by 9%-31%.