Prevalence and Risk Factors of Cerebral Microbleeds in Patients with Nonvalvular Atrial Fibrillation: An Enhanced T2*-Weighted Angiography Imaging Study.

Zhiqiang Li,Lu Cao,Wanqing Dong,Jianping Zhang,Chunyu Wang,Xianjun He,Wenqi Huang,Yuzhou Li,Ming Zhao
DOI: https://doi.org/10.1159/000501123
IF: 2.4
2019-01-01
European Neurology
Abstract:Background/Aims: Cerebral microbleeds (CMBs) may predict the occurrence of intracerebral hemorrhage (ICH). Warfarin use may cause ICH. It is still controversial whether warfarin increases CMBs in atrial fibrillation (AF) patients. The study aimed to investigate the prevalence of CMBs and risk factors in AF patient population. Methods: This single retrospective center study included 113 patients with nonvalvular atrial fibrillation (NVAF) who were on outpatient treatment. CMBs were counted using 3D-enhanced T2*-weighted angiography (ESWAN) imaging. We compared clinical and radiological data between patients who used warfarin and not used warfarin with univariate analysis. The associations between clinical and radiological data and CMBs in NVAF population were analyzed with multivariate linear regression analyses. Results: Among 113 NVAF patients, CMBs were found in 33 (29.2%) patients; there were 53 patients who used warfarin for thromboembolism prevention and 60 patients with similar demographic features who did not use warfarin. CMBs on ESWAN MRI showed no significant difference between the 2 groups (p = 0.061). Patients with CMBs were older than patients without CMBs (p = 0.046), and the frequency of smokers (p = 0.028), hypertension (p = 0.029), previous ICH (p = 0.000), and leukoaraiosis (p = 0.020) in patients with CMBs were significantly higher than patients without CMBs. In multivariate linear regression analyses, previous ICH (beta = 1.438, p = 0.000), age (beta = 0.082, p = 0.000), hypertension (beta = 0.956 p = 0.003), warfarin treatment (beta = 0.849, p = 0.006), and smokers (beta = 0.920, p = 0.016) were positive linear predictors of CMBs number levels in NVAF patients. Conclusion: The present data demonstrated that CMBs were associated with age, hypertension, warfarin treatment, smoking, and a history of ICH. We also found neither CHA2DS2-VASc score nor HAS-BLED score was associated with CMBs in patients with NVAF. (C) 2019 S. Karger AG, Basel
What problem does this paper attempt to address?