STROKE PATIENTS WITH AND WITHOUT HYPERTENSION: METHODS part 2
MRI-based Infarcts and CT-based Infarcts
MRI and CT scans were reviewed by AAASPS local principal investigators and/or local radiolo gists for evidence, location and size of recent and old cerebral infarctions. MRI and CT results were classified according to TOAST criteria.
Stroke Subtypes and Stroke Syndromes
Stroke subtypes and stroke syndromes were classified by local AAASPS investigators according to TOAST criteria. Major hemi-spheral syndromes were those identified as having aphasia with hemiparesis, hemisensory loss and/or homonymous hemianopia or non-dominant hemispheral syndromes with hemiparesis, hemisensory loss and/or homonymous hemianopia or as anterior cerebral syndromes. Minor hemispheral syndromes were those resulting in Broca’s aphasia without hemiparesis, conduction aphasia without hemiparesis, Wernicke’s aphasia without other signs, aphasia with vanishing hemiparesis or mild motor signs, isolated homonymous hemianopia, homonymous hemianopia with associated behavioral signs or as pure non-dominant behavioral signs. Deep hemispheric syndromes had no cortical involvement and were indexed as pure motor hemiparesis, pure sensory-motor stroke, ataxic-hemiparesis, dysarthria-clumsy hand or as hemichorea/hemiballism. Brainstem and cerebellar syndromes were those with major basilar artery syndrome, upper basilar artery syndrome, lower basilar artery syndrome, basilar branch syndrome, Wallenberg’s syndrome, Wallenberg’s syndrome and cerebellar syndrome, and pure cerebellar syndrome. suhagra
The AAASPS is an ongoing clinical trial. The study database used for this analysis contained 1086 enrollees: 1012 patients with htn and 74 without htn. For univariate analyses, chi-square, Fisher’s Exact, Mantel-Haenszel (M-H) test for trend and Wilcoxon Rank-Sum tests were used. Logistic regression analysis was used to test the association between selected risk factors and htn status. All variables that had a p-value <0.15 in univariate logistic regressions were included in a stepwise multivariate logistic regression analysis to obtain the final model. These variables included counts of recent infarcts and old infarcts, stroke subtype, age, sex, history of past stroke, diabetes, angina pectoris, congestive heart fail ure, hypercholesterolemia, other cardiac disease, lack of exercise, smoking and education.