Silent Strokes Predispose to Dementia

Abstract & Commentary

Source: Vermeer SE, et al. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003;348: 1215-1222.

There is little doubt that cerebrovascular disease is a major contributor to cognitive decline in the elderly. Recurrent symptomatic strokes affecting cortical or subcortical function inevitably result in a stepwise burden of accumulating damage and disability. Less clear, however, is a possible relationship between silent brain infarcts as diagnosed incidentally on MRI and a risk for the subsequent development of dementia.

In the Rotterdam Scan Study, 1015 elderly participants were followed prospectively over a 4-year period to correlate the presence of ischemic brain injury on an initial MRI with their subsequent risk of developing dementia. All participants were free of dementia at baseline, and any patient with a known history of stroke was excluded. The entire cohort was analyzed for the development of dementia, but only 75% were available for subsequent detailed mental status examination and/or MRI primarily due to the death or institutionalization of more than 100 enrollees.

Thirty participants developed dementia; of these, 14 had 1 or more silent infarcts on initial MRI scan, and 7 had multiple infarcts. The presence of silent brain infarcts at baseline increased the risk of dementia more than 2-fold (hazard ratio, 2.26), even when controlling for other MRI findings, such as periventricular or subcortical white matter lesions or brain atrophy. White matter disease also predicted dementia, although less powerfully, with a hazard ratio of 1.59 for periventricular and 1.21 for subcortical disease.

Among those participants who underwent a follow-up MRI scan, strokes appeared in 7/30 (23%) who became demented (3 symptomatic, 4 silent) compared with 79/618 (12%) who did not develop dementia (1 symptomatic, 71 silent). Documented decline in memory as measured by neuropsychiatric testing was restricted to patients with new infarcts, whether they had baseline silent strokes or not. Strokes in the thalamus were more likely to produce cognitive effects than in other locations. Dementia was diagnosed as Alzheimer’s disease (AD) in the majority (26/30), while vascular dementia was only identified in 2/30.

As Vermeer and colleagues note, this study did not focus on subtypes of dementia, but rather an overall association between silent vascular disease and cognitive loss. They postulate that there may be a direct association between ischemia and the development of Alzheimer’s pathology (plaques and neurofibrillary tangles) or alternatively, ischemia may unmask otherwise mild cases of AD.


Dementia nomenclature can be confusing, especially as it relates to cerebrovascular disease. Patients with vascular damage may be diagnosed as suffering from strategic infarct dementia, multi-infarct dementia, or merely vascular dementia. Those with severe hypertension might be labeled as Binswanger’s disease or may be thought to have white matter changes consistent with "leukoariosis." Additional overlap between these vascular diagnoses and AD pathology further muddies these classifications. From a practical therapeutic standpoint, however, acetylcholinesterase inhibitors, such as donepezil, should be considered for all patients regardless of subtype. Given the likely heterogeneous factors at work, it is not surprising that these agents appear to improve memory in patients thought to have AD, vascular-type disease, or some mixture of the 2.

Middle-aged and elderly patients who undergo MRI not for dementia, but often for unrelated reasons, are frequently discovered to have ischemic-type disease, whether lacunar strokes in deep gray matter structures or more nonspecific white matter hyperintensities. While such findings are still not cause for alarm in the majority of patients, the Rotterdam study reminds us that they are not benign and should motivate a comprehensive program of vascular risk factor reduction (including, but not limited to, control of lipid and blood pressure, and smoking cessation) as well as antiplatelet therapy such as daily baby aspirin. — Alan Z. Segal

Dr. Segal is Assistant Professor, Department of Neurology, Weill-Cornell Medical College, Attending Neurologist, New York Presbyterian Hospital