Background: Late survival of patients having deep sternal wound infection (DSWI) after bilateral internal thoracic artery (BITA) grafting is largely unexplored.
Methods: Outcomes of 3391 consecutive BITA patients were reviewed retrospectively. Patients with DSWI after surgery (n = 142, 4.2%) were compared with those having no sternal complications (n = 3177). Predictors of DSWI and of mortality during the follow-up period were found with negative-binomial and Cox proportional-hazards regression, respectively. One-to-one propensity score-matched analysis, which considered simultaneously baseline patient characteristics, operative data, and postoperative complications was performed. The resulting matched pairs were compared for non-parametric estimates of late survival. The same comparison was performed in matched pairs having no major complications (except DSWI) early after surgery.
Results: In-hospital mortality was higher in DSWI cohort than in patients having no sternal complications (5.6% vs. 1.8%, p = 0.0035). Almost all of postoperative complications were more frequent in DSWI patients. Female sex, obesity, chronic lung disease, renal impairment, extracardiac arteriopathy, congestive heart failure, and urgent/emergency priority were predictors of DSWI common to two DSWI risk models that were developed. DSWI was independent predictor of reduced late survival (multiple covariates-adjusted hazard ratio [HR], 1.91, p < 0.0001). The propensity matching resulted in 135 pairs with same in-hospital mortality (5.2%). Estimates of freedom from all-cause death were lower in DSWI cohort (HR, 1.92, p < 0.0001), even when only pairs (n = 59) having no major postoperative complications (except DSWI) were considered (HR, 1.84, p = 0.026).
Conclusions: DSWI after BITA use seems to reduce late survival even after adjusting for baseline patient characteristics and concomitant postoperative complications.
Keywords: Arterial grafts; Coronary artery bypass grafting; Mortality/survival; Risk factors; Sternal wound infection.
Copyright © 2018 Elsevier B.V. All rights reserved.