Incidence, imaging analysis, and early and late outcomes of stroke after cardiac valve operation

Abstract

The aim of this study was to analyze the incidence, topography, and mechanisms of stroke, independent predictors, and late outcome after cardiac valve operations. We retrospectively analyzed prospectively collected data from 2,808 patients (mean age 63 +/- 15 years, n = 1,610, 55% men) who underwent valve surgery with or without concomitant coronary artery bypass grafting from January 1998 to December 2006. Stroke was defined as any new permanent focal neurologic deficit. Overall incidence of stroke was 2.2% (n = 63) and decreased during the study period from 3.3% (1998 to 2002) to 1.3% (2003 to 2006; p = 0.001). The highest stroke rate was observed after double aortic/mitral valve replacement (5.4%) and valve/coronary artery bypass grafting procedures (3.6%). Brain imaging was positive in 74% (n = 43 of 58) and showed ischemic stroke in all patients and hemorrhagic conversion in 28%. Distribution of acute stroke was large territory embolic artery (n = 33, 77%), watershed (n = 7, 16%), and mixed pattern (n = 3, 7%). Multivariate analysis revealed calcified ascending aorta (odds ratio [OR] 2.7), female gender (OR 2.6), ejection fraction <30% (OR 2.3), diabetes (OR 2.2), age >70 years (OR 2.0), and cardiopulmonary bypass time >120 minutes (OR 3.7) as predictors of stroke. Hospital mortality was 24% and 4.6% in patients with and without stroke, respectively. Survival of stroke patients was 78% and 54% at 1 year and 5 years, respectively, and was significantly decreased compared with patients without stroke. Valve pathology including endocarditis did not influence the incidence of stroke. Intraoperative epiaortic scanning may contribute in decreasing the incidence of this complication and may be warranted in all patients undergoing valvular surgery. In conclusion, stroke after valvular surgery is associated with an increased hospital mortality and morbidity and decreased long-term survival.