Objectives: Cerebrovascular accidents (CVA) after CABG are deleterious complications whose prevention remains poorly defined. The aim of this study was to identify the determinants for CVA after CABG.
Methods: Nine thousand nine hundred and sixteen patients underwent CABG at our institution from January 1992 to June 2002. Data were prospectively collected and univariate/multivariate analyses conducted.
Results: Two hundred and eight patients (2.1%) suffered perioperative CVA. Univariate analysis showed a higher risk profile in the CVA group including advanced age, depressed percent left ventricular ejection fraction (LVEF), unstable angina, diabetes mellitus (DM), chronic renal failure (CRF), redo surgery, peripheral vascular disease (PVD), previous CVA, and higher Parsonnet score (P<0.001). Furthermore, the CVA group had longer myocardial ischemia (CVA 56.2 +/-40.9 vs. Control 50.4+/-20.9 min, P=0.03) and cardiopulmonary bypass (CPB) times (CVA 87.4+/-30.0 min vs. Control 78.9 +/-25.9 min, P<0.0001), and lower off-pump surgery rate (CVA 1.4% vs. Control 4.7%, P=0.01). Multivariable analysis identified seven preoperative and two perioperative determinants for CVA: LVEF<30% (odds ratio (OR)=2.49), previous CVA (OR=2.15), DM (OR=1.78), redo (OR=1.76), PVD (OR=1.66), CRF (OR=1.55), age (OR=1.03), perioperative intra-aortic balloon pump (OR=1.83), and transfusion rate (OR=1.59). Perioperative mortality was higher in the CVA group (CVA 18.6% vs. Control 2.6%, P<0.0001).
Conclusions: Although occurrence of CVA seems mainly related to preoperative comorbidities, perioperative surgical variables, such as off-pump surgery, myocardial ischemia and cardiopulmonary bypass time, do not seem to independently influence CVA rate after CABG. In this regard CVA prevention should be performed before posing an indication to CABG, and closer evaluation of patients' risk profiles and tailored clinical/surgical strategies for those patients at higher risk for CVA occurrence should be included.