SUBCORTICAL SILENT BRAIN INFARCTION AS A RISK FACTOR FOR CLINICAL STROKE

Background and Purpose No prospective studies have examined the rate o f symptomatic ischemic or hemorrhagic stroke in patients with subcorti cal silent brain infarction (SSBI) who were otherwise neurologically n ormal at entry into the study. This report investigates SSBI, detected by MRI, as a clinical stroke risk factor. Methods MRI scans were perf ormed in 933 neurologically normal adults (30 to 81 years; mean age, 5 7.5+/-9.2 years) without history of cerebrovascular diseases who recei ved our health screening of the brain 1 to 7 years before investigatio n. We obtained information of their clinical stroke onset through send ing out a questionnaire for subjects. We detected SSBI (focal T2 hyper intensities larger than 3 mm with correlative T1 hypointensity), FWT2H L (focal white matter T2 hyperintensity lesions similar to SSBI but wi thout correlative T2-hypointensity), and PVH (periventricular hyperint ensity) by MRI. Age, sex, family history of stroke, history of hyperte nsion, diabetes mellitus, lipids, hematocrit, blood pressure, fasting blood sugar, smoking, alcohol habits, ischemic changes on electrocardi ogram, and sclerotic changes of retinal arteries were included in the analysis. Results Incidence of SSBI was 10.6% in all subjects. No cort ical infarct was detected in this series. Multiple logistic regression analysis showed that hypertension (odds ratio [OR], 4.07; 95% CI, 2.5 7 to 6.45), diabetes (OR, 2.41; 95% CI, 1.20 to 4.85), alcohol habits greater than or equal to 58 g/day (OR, 2.58; 95% CI, 1.50 to 4.45), re tinal artery sclerosis (OR, 2.14; 95% CI, 1.32 to 2.38), and age (OR, 1.77; 95% CI, 1.32 to 2.38) were significant and independent risk fact ors for SSBI. For FWT2HL, hypertension (OR, 4.49; 95% CI, 2.54 to 7.96 ) and age (OR, 2.08; 95% CI, 1.45 to 3.00) were also independent risk factors. Risk factors for PVH were age (OR, 3.46; 95% CI, 2.23 to 5.36 ), hypertension (OR, 3.06; 95% CI, 1.62 to 5.78), and retinal artery s clerosis (OR, 2.25; 95% CI, 1.02 to 4.96). We found 14 brain infarctio ns, 4 brain hemorrhages, and 1 subarachnoid hemorrhage during observat ion. Annual incidence of clinical stroke was higher in the subjects wi th SSBI than in those without focal lesions (10.1% versus 0.77%). ORs for clinical stroke onset were 10.48 for SSBI (95% CI, 3.63 to 30.21) and 4.81 for FWT2HL (95% CI, 1.13 to 20.58). The PVH did not relate to clinical stroke onset. Conclusions The strong association of SSBI, FW T2HL, and PVH with hypertension suggests a common underlying mechanism (presumably small-vessel vasculopathy). The SSBI showed the most sign ificant association for clinical subcortical stroke. The FWT2HL was al so a risk factor for the stroke but was less significant than SSBI. Th e subjects with SSBI should be considered at high risk for clinical su bcortical brain infarction or brain hemorrhage.