Annual Incidence of First Silent Stroke in the United States: A Preliminary Estimate

BACKGROUND: Recent estimates of stroke incidence in the US range from 760,000 to 780,000 annually, however these estimates do not reflect the incidence of silent infarcts and hemorrhages. Since population-based studies indicate the prevalence of silent stroke is substantially higher than that of symptomatic stroke, estimates of stroke incidence based solely on symptomatic events may substantially underestimate the annual stroke burden. METHODS: The prevalence of silent infarcts for different age strata were abstracted from two US population-based MRI studies, the Atherosclerosis Risk in Communities Study and the Cardiovascular Health Study. Similarly, first silent cerebral hemorrhage incidence rates were derived from population-based MRI prevalence observations in the Austrian Stroke Prevention Study. Prevalence observations in these studies and death rates from the US Census were inputted to calculate age-specific first silent MRI infarct and hemorrhage incidence. Age- specific incidence rates were projected onto 1998 US population age cohorts to calculate the annual burden of first silent MRI ischemic stroke and first silent MRI cerebral hemorrhage. RESULTS: Estimated age-specific annual incidence rates (per 100,000) of persons experiencing first silent MRI infarct ranged from 1,600 in the age 30-39 stratum to 16,400 at ages 75-79. Estimated incidence rates of first silent MRI cerebral hemorrhage ranged from 180 in the ages 30-39 to 6,900 at age >85. Overall, the projected annual incidence in 1998 of US individuals experiencing first silent MRI infarct was 9,040,000, and first silent MRI hemorrhage 1,940,000. CONCLUSIONS: In 1998, more than 11 million persons experienced stroke in the US, in whom approximately 770,000 were symptomatic and 11 million were first-ever silent MRI infarcts or hemorrhages. These findings demonstrate the incidence of stroke is substantially higher than suggested by estimates based solely on clinically manifest events.