Deep Sternal Wound Infection after Coronary Artery Bypass Grafting Markedly Increases Hospital Length of Stay: Estimates from Community Hospitals

Abstract

BACKGROUND: Despite improved use of peri-operative antibiotic prophylaxis, surgical site infections following coronary artery bypass grafting (CABG) continue to occur, with high case-fatality rates and costs. Few data are currently available on the impact of deep sternal wound infections (DSWI) on length of hospital stay, and no such data exist for community hospitals (as opposed to teaching hospitals). We assessed the impact of DSWI on length of stay, in-hospital survival, and disposition using a large administrative database. METHODS: We evaluated outcomes in a cohort of individuals with DSWI identified using the Quality Resources and Resource Management System (QRS), a large database that was constructed by Tenet HealthSystem to evaluate quality of care among consecutive individuals undergoing CABG at 38 participating hospitals in Southern California, Southern Florida, and New Orleans. Entry of data values into QRS is performed by case managers as a routine duty and occurs concurrently with patient care. Predictors of in-hospital mortality were evaluated through creation of logistic regression models; rates of discharge were evaluated using Kaplan-Meier methods. Factors associated with length of stay were evaluated using a series of Cox proportional hazards models. RESULTS: A total of 16,383 individuals underwent CABG between January 1999 and February 2003; 125 procedures (0.76%) were complicated by DSWI. DSWI was associated with a marked increase in in-hospital mortality (RR 4.37, 95% CI 2.94–6.50), and in length of hospital stay (median LOS 37.5 days versus 6.5 days, p <0.001 by log-rank test). Individuals who survived DSWI were less likely to be discharged home rather than to a skilled nursing or rehabilitation facility (RR 0.61, 95% CI 0.48–0.77). Mortality after DSWI was predicted by a pre-operative history of ventricular arrhythmia (RR 5.22, 95% CI 2.04–13.36), and by prolonged ventilation after surgery (odds ratio for death with each additional day on ventilator 1.05, 95% CI 1.02–1.09). Pre-operative patient characteristics were poorly predictive of length of stay. Lower rates of hospital discharge among those with DSWI were predicted by admission for non-elective cardiac surgery (hazard ratio 0.39, 95% CI 0.22–0.66). CONCLUSIONS: The impact of deep sternal wound infections on patient mortality and resource consumption is substantial in the community hospital setting. Given the high impact on healthcare resources, even modestly effective interventions that reduce DSWI could be cost-effective or cost-saving in community hospitals.