Correlation between white matter hyperintensities and stroke etiology classification in patients with acute isolated penetrating artery territory infarction / 国际脑血管病杂志

ABSTRACT

Objective:To investigate the correlation between white matter hyperintensities (WMHs) and stroke etiology classification in patients with acute isolated penetrating artery territory infarction.Methods:Patients with first-ever acute isolated penetrating artery territory infarction admitted to the Department of Neurology, the Affiliated Hospital of Xuzhou Medical University from January 2017 to May 2020 were enrolled retrospectively. According to the Chinese Ischemic Stroke Subclassification (CISS) system, they were divided into large artery atherosclerosis (LAA) and perforating artery disease (PAD). According to the distribution of infarcts, they were divided into lenticulostriate artery (LSA) territory infarction and paramedian pontine artery (PPA) territory infarction. The demographics, vascular risk factors, baseline clinical data, WMHs location, and Fazekas Scale scores were documented. Multivariate logistic regression analysis was used to identify the independent influencing factors of stroke etiology classification. Results:A total of 440 patients with acute isolated penetrating artery territory infarction were enrolled, including 120 (27.3%) in the LAA group, and 320 (72.7%) in the PAD group; 213 (48.4%) with LSA territory infarction, and 227 (51.6%) with PPA territory infarction. The proportion of patients with total Fazekas score 3-6 and periventricular WMHs (PWMHs) score 2-3 in the PAD group was significantly higher than those in the LAA group (all P<0.05). In patients with LSA territory infarction, the proportion of the patients with hypertension, WMHs total Fazekas score 3-6 and PWMHs score 2-3 in PAD subgroup was significantly higher than those in the LAA subgroup, while the proportion of the patients with hyperlipidemia was significantly lower than that in LAA subgroup (all P<0.05). In patients with PPA territory infarction, the levels of low-density lipoprotein cholesterol and homocysteine in the PAD subgroup were significantly lower than those in the LAA subgroup. Multivariate logistic regression analysis showed that PWMHs score 2-3 was an independent correlation factor of PAD (odds ratio [ OR] 2.220, 95% confidence interval [ CI] 1.085-4.541; P=0.029). In patients with LSA territory infarction, hyperlipidemia was independently correlated with LAA ( OR 0.432, 95% CI 0.192-0.972; P=0.042), and PWMHs score 2-3 was independently correlated with PAD ( OR 3.846, 95% CI 1.193-12.397; P=0.024). In patients with PPA territory infarction, higher low-density lipoprotein cholesterol ( OR 0.660, 95% CI 0.494-0.883; P=0.005), homocysteine ( OR 0.958, 95% CI 0.930-0.987; P=0.005) and C-reactive protein ( OR 0.987, 95% CI 0.977-0.997; P=0.008) were independently correlated with LAA. Conclusions:WMHs are common in patients with acute isolated perforating territory infarction caused by LAA and PAD, and more severe PWMHs suggest that PAD is more likely to be the cause of the acute isolated perforating territory infarction, especially in patients with LSA territory infarction.