DESCRIPTION (provided by applicant): Inflammation is increasingly recognized to play a central role in atherosclerosis and coronary artery disease, and peripheral blood markers of inflammation have been associated with cardiac events. The relationship of these potentially modifiable risk markers to ischemic stroke risk is less clear. Additional epidemiological studies have suggested that the burden of past exposure to common infectious diseases may be one of the factors contributing to increased levels of inflammation. The proposed research study will capitalize on the on-going Northern Manhattan Study (N INDS R01 29993, PI RL S acco), a prospective cohort study of stroke risk factors in a multi-ethnic urban population, to test the following hypotheses: (1) high-sensitivity C -reactive protein and other inflammatory markers (serum amyloid A, interleukin 6, tumor necrosis factor receptor levels) are independent risk factors for (a) ischemic stroke and (b) ot her vascular outcomes in whites, blacks and Hispanics; (2) the burden of common infectious serologies (C. pneumoniae, H.pylori, CMV, EBV, HSV 1 and 2) is an independent risk factor for stroke and other vascular outcomes in these populations; (3) these inflammatory and infectious marker s are associated with sub-clinical vascular brain diseases seen on MRI scanning, such as white matter hyperintensities and silent infarcts. Levels of these markers will be measured using nephelometric and immunoassay techniques on stored baseline serum samples from 2000 of 3298 initially stroke-free subjects, of whom 1400 will undergo MRI scanning as part of the parent project. Follow-up serum samples will also be collected on these 1400 participants. Outcomes of ischemic stroke, total stroke, myocardial infarction, vascular mortality and total mortality will be ascertained. Cox proportional hazard model analysis will be used to assess the significance of the main exposure variables after adjustment for other risk factors, and inflammatory and infectious markers will be analyzed as time dependen t covariates when follow up values are available. The cross-sectional associations between inflammatory and infectious markers and quantitative MRI abnormalities will also be assessed.