Correlation Between White Matter Lesion Score and Volume and Cerebral Microbleeds in Patients with Cerebral Small Vessel Disease

Chen Xin,Gu Yucheng,Zou Junhui,Jiang Yongcheng,Liu Renyuan,Cheng Yue,Xu Yun

DOI: https://doi.org/10.3760/cma.j.issn.1673-4165.2018.11.004

2018-01-01

Abstract:Objective To investigate the correlation between white matter lesions (WMLs) score and volume and cerebral microbleeds (CMBs) in patients with cerebral small vessel disease (CSVD).Methods Patients with non-acute lacunar infarction from the CSVD Follow-up Cohort Study in the Affiliated Drum Tower Hospital of Nanjing University from January 2017 to December 2017 were enrolled.The relevant clinical data were collected and 3.0 T cranial MRI examinations were performed,including T1-weighted imaging,T2-weighted imaging,fluid attenuated inversion recovery (FLAIR) sequence,susceptibility-weighted imaging,and diffusion-weighted imaging.W2MHS software was used to quantify the volume of WMLs.Fazekas method was used to score periventricular and deep WMLs separately.The number of CMBs was counted visually.Multivariate logistic regression analysis was used to determine the independent influencing factors of CMBs.Multivariate linear regression equation (stepwise method) was used to analyze the independent influencing factors of the number of CMBs.Medcalc 18.6 was used to deseribe receiver operating characteristic (ROC) curve.The predictive value of WMLs Fazekas scores and volumes for CMBs was evaluated.Results A total of 82 patients were enrolled,including 31 (37.8％) in CMBs group and 51 (62.2％) in non-CMBs group.The comparison between the two groups showed that smoking,use of antiplatelet agents,history of transient ischemic attack (TIA) or lacunar infarction,larger WMLs volume,higher levels of triacylglycerol,and lower levels of high-density lipoprotein cholesterol might be the risk factors for CMBs.Multivariate logistic regression analysis showed that higher Fazekas score (odds ratio 1.908,95％ confidence interval 1.210-3.009;P=0.005) and larger WMLs volume (odds ratio 4.620,95％ confidence interval 1.279-16.683;P =0.019) were the independent risk factors for CMBs.Multiple linear regression analysis showed that Fazekas score (r =0.379,P =0.001) and WMLs volume (r =0.260,P =0.023) were independently and positively correlated with the number of CMBs.The ROC curve analysis showed that the area under the curve of Fazekas scores predicting CMBs was 0.768 (95％ confidence interval 0.654-0.881).The optimal cut-off value was 3,the sensitivity was 61.29％,and the specificity was 90.20％.The area under the curve of WMLs volume predicting CMBs was 0.783 (95％ confidence interval 0.677-0.867).The optimal cut-off value was the volume of WMLs reaching 2 137.96 mm3.The sensitivity was 73.33％,and the specificity was 84.00％.Conclusion WMLs and CMBs had significant correlation in patients with CSVD.The WMLs Fazekas score and volume quantification are expected to serve as the alternative indicators for the urgent assessment of CMBs.