Study design validation for consolidating global with focal neurological events in cardiac surgery stroke risk factor analyses.

ABSTRACT

Neuroimaging shows that both global and focal neurologic deficits after cardiac surgery share an acute, often multifocal, embolic cerebral infarction etiology; yet, analyses of stroke risk factors historically have emphasized the focal deficits. We test if consolidating encephalopathy and coma with focal deficits affects four stroke risk factors and a dummy variable. Overall focal and global events in 575 cardiopulmonary bypass operations identified by retrospective review matched indices reported in large prospective studies. Logistic regression in 189 records selected for completed non-invasive preoperative carotid stenosis screening showed all four conventional stroke risk factors to be independent predictors of overall consolidated global plus focal neurologic risk, specifically age [odds ratio (OR) 1.90 per decade], carotid stenosis >50% (OR 1.91), pump time (OR 1.67 per hour), open chamber (OR 1.95); and successfully eliminated the dummy variable gender (P = 0.6). This analysis indicates that the design of future stroke risk factor studies in the setting of cardiac surgery can and should adopt a neuroimaging evidence-based investigative approach of consolidating global with focal deficits.