Silent Brain Infarction and Subcortical White Matter Lesions Increase the Risk of Stroke and Mortality: A Prospective Cohort Study

Bokura, Hirokazu; Kobayashi, Shotai; Yamaguchi, Shuhei; Iijima, Kenichi; Nagai, Atsushi; Toyoda, Genya; Oguro, Hiroaki; Takahashi, Kazuo

doi:10.1016/j.jstrokecerebrovasdis.2005.11.001

« Previous Next »Journal of Stroke & Cerebrovascular Diseases
Volume 15, Issue 2 , Pages 57-63, March 2006

Silent Brain Infarction and Subcortical White Matter Lesions Increase the Risk of Stroke and Mortality: A Prospective Cohort Study

Hirokazu Bokura, MD, PhD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
- Address reprint requests to Hirokazu Bokura, MD, PhD, Shimane University, Faculty of Medicine, 89-1 Enya-cho, Shimane 693-8501, Japan
Shotai Kobayashi, MD, PhD
- Director of University Hospital, Shimane University, Faculty of Medicine, Shimane, Japan
Shuhei Yamaguchi, MD, PhD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
Kenichi Iijima, MD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
Atsushi Nagai, MD, PhD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
Genya Toyoda, MD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
Hiroaki Oguro, MD, PhD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan
Kazuo Takahashi, MD, PhD
- Department of Neurology, Hematology, & Rheumatology, Shimane, Japan

Received 11 July 2005; received in revised form 13 October 2005; accepted 17 October 2005.

Silent brain infarction (SBI) and white matter lesions (periventricular hyperintensity [PVH] and subcortical white matter lesions [SWML] are detected in both stroke patients and normal elderly persons. We prospectively examined the association between these lesions and the risk of subsequent stroke and mortality in neurologically normal adults. Magnetic resonance imaging scans were performed in 2,684 neurologically normal subjects with no history of stroke (mean age, 58 ± 7 years old at entry) who underwent our health screening of the brain. After the brain screening, we obtained information about clinical stroke onset and death using a questionnaire sent annually to all subjects. When a subject suffered from medical events, we confirmed the detailed information in a telephone interview and by asking the attending physician. SBI was defined as a focal T2-hyperintensity and T1-hypointensity lesion > 3 mm. PVH and SWML were graded according to their severity. The average follow-up period was 6.3 years. Stroke occurred in 102 subjects (3.8%), and 93 subjects died during follow-up. The incidence of clinical stroke was significantly higher in the subjects with SBI than in those without SBI. Marked PVH and marked SWML independently increased the risk of stroke (for SBI, stroke risk factor–adjusted odds ratio [OR] = 3.66, 95% confidence interval [CI] = 2.28–5.89; for marked PVH, stroke risk factor–adjusted OR = 2.08, 95% CI = 1.04–4.17; for marked SWML, stroke risk factor–adjusted OR = 2.73, 95% CI = 1.32–5.63). Regarding mortality, SBI and marked PVH increased the risk of death (for SBI, stroke risk factor–adjusted OR = 1.95, 95% CI = 1.16–3.29; for PVH, stroke risk factor–adjusted OR = 4.01, 95% CI = 1.91–8.45). Death attributable to stroke occurred more frequently in those subjects with SBI, marked PVH, and marked SWML. We conclude that SBI, marked PVH, and marked SWML are important risk factors for clinical stroke and that SBI and marked PVH also increase the risk of mortality.

Key Words: Silent brain infarction , periventricular hyperintensity , magnetic resonance imaging , stroke , mortality