Background: Cerebral white matter disease (WMD) may result from accumulation of silent embolic brain infarcts in the setting of subclinical, non-stenotic cervical carotid atherosclerosis. The contribution of cervical plaque to the burden of WMD is not well established. Methods: A multicenter, retrospective cohort of consecutive adult patients with stroke due to cervical carotid atherostenosis (>50% luminal stenosis), small vessel disease, or cryptogenic mechanism with unilateral hemispheric stroke was queried. Maximum cervical carotid plaque thickness was used to predict higher grade WMD (Fazekas grade 2-3 versus 0-1) in unadjusted logistic regression, stratified by quartile of inter-side plaque (mean total plaque in axial dimension of the left and right cervical carotid arteries), and adjusted for age, stroke mechanism, atherosclerotic risk factors, and clustering by site. Results: Of the 375 included patients, the median age was 66y (IQR 58-74), 170 (45.3%) were female, and the median inter-side cervical ICA plaque thickness was 1.8mm (IQR 0.2-3.2). Compared to patients in the lowest quartile of inter-side plaque (<0.2mm), those in higher quartiles had higher grade WMD (Q3 adjusted odds ratio aOR 1.44, 95% confidence interval CI, 1.09-1.89; Q4 aOR 1.85, 95% CI, 1.37-2.50). The association with higher grade WMD persisted in a sensitivity analysis considering inter-side plaque thickness as a continuous variable (adjusted incidence rate ratio/1mm plaque 1.09, 95% CI, 1.02-1.17). The effect was preserved across stroke mechanisms, sex, and infarct pattern (cortical versus subcortical); however, younger patients had a stronger association between plaque thickness and WMD while the eldest patients had no association (pinteraction <0.01). Conclusions: In this cohort of non-cardioembolic stroke patients, greater inter-side cervical carotid plaque thickness was strongly associated with greater WMD. This association supports a potential role of subclinical cervical carotid artery atherosclerosis as a contributor to WMD, which may represent accumulation of silent brain infarcts.
Golani et al. (Thu,) studied this question.