A pathological perspective on the natural history of cerebral atherosclerosis.

Background: The natural history of intracranial large artery atherosclerosis has been mainly described from lumen-based imaging studies, and much of what is reported to be known about atherosclerosis is derived from non-cerebral arteries.

Aims: To test the hypothesis that atherosclerosis is only partially represented by stenosis and that advanced atherosclerosis is more common that severe stenosis in noncardioembolic infarcts.

Methods: Cerebral large arteries from 196 autopsy cases were studied. The revised American Heart Association classification for atherosclerosis was used to determine the phenotype in each available artery. Cross-sectional lumen stenosis was obtained as defined by the Glagov's method.

Results: As age of cases increased, there was a progressive increment in the frequency of atherosclerotic lesions, rising from 5% of all arteries at age 20-40, to more than 40% at age 60 or older. Stenosis also increased with age: less than 3% of the arteries in those ≤50 years had >40% stenosis, while one out of five arteries in those >80 years had >40% stenosis. In most cases (80%), atherosclerosis and stenosis were directly related. However, one out of five cases with advanced atherosclerosis had <30% stenosis. In arteries supplying brain areas with noncardioembolic infarcts, the majority of segments exhibiting advanced atherosclerosis had lumen stenosis of <40%.

Conclusion: Although intracranial atherosclerosis is typically associated with stenosis, a substantial minority of cases shows advanced atherosclerosis in the absence of stenosis >40%. Definitions based solely on stenosis may underestimate the extent and role of intracranial large artery atherosclerosis.