Cerebral microbleeds predict first-ever symptomatic cerebrovascular events

Clin Neurol Neurosurg. 2009 Dec;111(10):825-8. doi: 10.1016/j.clineuro.2009.08.011. Epub 2009 Sep 17.

Abstract

Objective: Cerebral microbleeds (CMB) on gradient-echo T2*-weighted magnetic resonance image (MRI) are frequently seen in patients with cerebral diseases. In this observational study we assessed whether CMB are a predictive factor for first-ever cerebrovascular events.

Patients and methods: This study consisted of 698 subjects without a history of symptomatic cerebrovascular events, who received gradient-echo T2*-weighted MRI for 3 months between November 2003 and January 2004 in Kishiwada City Hospital, Osaka, Japan. These subjects were then observed as outpatients for over 3.5 years.

Results: The prevalence of CMB at baseline was 17.0% (119/698) in this population, and the follow-up rate was 51%. A total of 36 first-ever symptomatic cerebrovascular events were observed during the 3 and a half-year follow-up period. First-ever symptomatic cerebrovascular events occurred significantly more frequently in subjects with CMB (15 cases) than those without CMB (21 cases) (p=0.001). Even after adjusting for age, sex and hypertension, it was revealed that the presence of CMB was an independent predictor for the first-ever symptomatic cerebrovascular event by using the Cox proportional hazards model (hazard ratio, 2.87; 95% CI, 1.27-6.48; p=0.01).

Conclusion: The presence of CMB is an independent predictor of first-ever symptomatic cerebrovascular diseases.

MeSH terms

  • Adult
  • Aged
  • Aging / physiology
  • Cerebral Hemorrhage / epidemiology*
  • Cerebral Hemorrhage / pathology
  • Cerebral Infarction / epidemiology
  • Cerebral Infarction / pathology
  • Cerebrovascular Disorders / epidemiology*
  • Cerebrovascular Disorders / pathology
  • Female
  • Follow-Up Studies
  • Forecasting
  • Humans
  • Kaplan-Meier Estimate
  • Magnetic Resonance Imaging
  • Male
  • Middle Aged
  • Prognosis
  • Proportional Hazards Models
  • Prospective Studies
  • Software
  • Stroke / epidemiology