Prediction of white matter lesions and subcortical atrophy for cognitive impairment in patients with ischemic stroke

Abstract

Objective To investigate the predictive role of cerebral white matter lesions (WML) and subcortical atrophy on cognitive function in patients with acute ischemic stroke (AIS) after 3 months. Methods 233 cases of AIS patients admitted to hospital continuously from September 2016 to March 2018 were enrolled and all of them underwent brain MRI.The degree of WML on FLAIR was evaluated according to the Fazekas grading standard.The linear measurement of subcortical atrophy on T1WI was carried out on the subcortical brain atrophy index, including EVANS ratio (ER), inverse cella media index (iCMI), caudate head index (CHI) and basal cistern index (BCI). Demographic, clinical and imaging data of all patients were also recorded.Mini-mental state examination (MMSE) and Montreal cognitive assessment (MoCA) were conducted at 3 months after AIS.The patients were divided into normal cognitive function (NCI) group and post stroke cognitive impairment (PSCI) group according to evaluation results of MMSE and MoCA.Multivariate logistic regression analysis was used to screen the independent risk factors of cognitive impairment. Results Univariate analysis showed that age (t=-4.233, P=0.000), sex (χ2=7.501, P=0.006), education (H=21.188, P=0.000), NHISS score (H=5.791, P=0.016), history of atrial fibrillation (χ2=6.484, P=0.011), TIA (χ2= 9.015, P=0.003), smoking history (χ2=6.943, P=0.008), Fazekas WML score (χ2=27.885, P=0.000), EVANS ratio (H=31.129, P=0.000), inverse cella media index (H=9.434, P=0.002), caudate head index (H=15.148, P=0.000), basal cistern index (t=-1.979, P=0.049) and baseline cognitive function (χ2=136.994, P=0.000) were related to cognitive impairment in patients with AIS after 3 months (P<0.05). Multivariate logistic regression analysis showed that WML score (OR=3.416, P=0.047, 95%CI: 1.017-11.482), EVANS ratio (OR=1.245, P=0.038, 95%CI: 1.012-1.531) and caudate head index (OR=1.187, P=0.040, 95%CI: 1.008-1.397) were risk factors for cognitive impairment in AIS patients after 3 months adjusting for age, education, disease severity and baseline cognitive function. Conclusion WML, EVANS ratio and caudate head index can predict short-term cognitive function in patients with AIS. Key words: White matter lesions; Subcortical atrophy; Acute ischemic stroke; Cognitive impairment