The Behavioral and Cognitive Executive Disorders of Stroke: The GREFEX Study

Many studies have highlighted the high prevalence of executive disorders in stroke. However, major uncertainties remain due to use of variable and non-validated methods. The objectives of this study were: 1) to characterize the executive disorder profile in stroke using a standardized battery, validated diagnosis criteria of executive disorders and validated framework for the interpretation of neuropsychological data and 2) examine the sensitivity of the harmonization standards protocol proposed by the National Institute of Neurological Disorders and Stroke and Canadian Stroke Network (NINDS-CSN) for the diagnosis of Vascular Cognitive Impairment.

237 patients (infarct: 57; cerebral hemorrhage: 54; ruptured aneurysm of the anterior communicating artery (ACoA): 80; cerebral venous thrombosis (CVT): 46) were examined by using the GREFEX battery. The patients' test results were interpreted with a validated framework derived from normative data from 780 controls.

Dysexecutive syndrome was observed in 88 (55.7%; 95%CI: 48-63.4) out of the 156 patients with full cognitive and behavioral data: 40 (45.5%) had combined behavioral and cognitive syndromes, 29 (33%) had a behavioral disorder alone and 19 (21.6%) had a cognitive syndrome alone. The dysexecutive profile was characterized by prominent impairments of initiation and generation in the cognitive domain and by hypoactivity with disinterest and anticipation loss in the behavioral domain. Cognitive impairment was more frequent (p = 0.014) in hemorrhage and behavioral disorders were more frequent (p = 0.004) in infarct and hemorrhage. The harmonization standards protocol underestimated (p = 0.007) executive disorders in CVT or ACoA.

This profile of executive disorders implies that the assessment should include both cognitive tests and a validated inventory for behavioral dysexecutive syndrome. Initial assessment may be performed with a short cognitive battery, such as the harmonization standards protocol. However, administration of a full cognitive battery is required in selected patients.