Burden and Outcome of Prevalent Ischemic Brain Disease in a National Acute Stroke Registry

BACKGROUND AND PURPOSE: Previous overt stroke and subclinical stroke are frequent in patients with stroke; yet, their clinical significance and effects on stroke outcome are not clear. We studied the burden and outcome after acute ischemic stroke by prevalent ischemic brain disease in a national registry of hospitalized patients with acute stroke. METHODS: Patients with ischemic stroke in the National Acute Stroke Israeli prospective hospital-based registry (February to March 2004, March to April 2007, and April to May 2010) with information on previous overt stroke and subclinical stroke per computed tomography/MRI (n=3757) were included. Of them, a subsample (n=787) was followed up at 3 months. Logistic regression models were computed for outcomes in patients with prior overt stroke or subclinical stroke, compared with patients with first stroke, adjusting for age, sex, vascular risk factors, stroke severity, and clinical classification. RESULTS: Two-thirds of patients had a prior overt stroke or subclinical stroke. Death rates were similar for patients with and without prior stroke. Adjusted odds ratios (OR; 95% confidence interval [CI]) for disability were increased for patients with prior overt stroke (OR, 1.31; 95% CI, 1.03-1.66) and subclinical stroke (OR, 1.45; 95% CI, 1.16-1.82). Relative odds of Barthel Index≤60 for patients with prior overt stroke (OR, 2.04; 95% CI, 1.14-3.68) and with prior subclinical stroke (OR, 2.04; 95% CI, 1.15-3.64) were twice higher than for patients with a first stroke. ORs for dependency were significantly increased for patients with prior overt stroke (OR, 1.95; 95% CI, 1.19-3.20) but not for those with subclinical stroke (OR, 1.36; 95% CI, 0.84-2.19). CONCLUSIONS: In our national cohort of patients with acute ischemic stroke, nearly two thirds had a prior overt stroke or subclinical stroke. Risk of poor functional outcomes was increased for patients with prior stroke, both overt and subclinical.