he Incidence of Stroke after Myocardial Infarction: A eta-Analysis

URPOSE: While the risk of stroke after myocardial infarction (MI) is increased compared with the risk mong those without MI, the magnitude of this risk remains unclear. Although numerous clinical trials ave reported the incidence of stroke following MI, these are among selected populations. We reviewed ohort studies reporting the incidence of stroke after MI to better define the risk of ischemic stroke in an nselected population. ETHODS: A computerized literature search (MEDLINE and PubMed) and manual review of reference ists of identified articles were conducted. Population-based studies published from 1978-2004 with at least 00 subjects that reported number or percent of ischemic strokes experienced by MI survivors were dentified. Data were extracted using standardized forms, and study quality was assessed by 2 independent eviewers. Ischemic stroke rates were reported as number of events per 1000 MI with 95% confidence ntervals (CI) calculated by Poisson distribution. A combined stroke rate was calculated for in-hospital, 30 ays, and 1-year post-MI using weights of 1/variance. A random-effects model also was created to estimate n-hospital stroke rate. Variability in study designs and outcome definitions limit synthesis of available ata. ESULTS: During hospitalization for the index MI, 11.1 ischemic strokes occurred per 1000 MI ompared with 12.2 at 30 days and 21.4 at 1 year. Using a random-effects model, 14.5 strokes occurred er 1000 MI. Positive predictors of stroke after MI included: advanced age, diabetes, hypertension, history f prior stroke, anterior location of index MI, prior MI, atrial fibrillation, heart failure, and nonwhite race. ONCLUSIONS: The public health implications of stroke among MI survivors, as well as the large umber of MI survivors, underscore the need to be aware of this devastating complication. Further research s needed to determine the optimal stroke prevention strategies for MI survivors. © 2006 Elsevier Inc. All ights reserved.