目的：评估脑小血管病（CSVD）非痴呆患者的认知状态，探讨脑微出血（CMBs）的数量、部位及其与认 知功能损害的相关性。方法：连续收集 CSVD 患者 174 例，应用磁敏感成像（SWI）技术检测 CMBs，记录 CMBs部位、数量。根据有无CMBs将患者分为有CMBs组（62例）和无CMBs组（112例），根据蒙特利尔认知 评估量表（MoCA）对患者认知功能进行评估。比较2组间认知功能评分是否存在差异；采用偏相关分析和 Spearman相关分析，分析CMBs的数量和部位与认知功能评分的关系。结果：皮质-皮质下CMBs出现率占 CMBs 总数量的 40.87%（132/323），深部及幕下为 59.13%（191/323）；CMBs 好发部位是基底节（101 个， 31.27%）、丘脑（39个，12.07%）、额叶（41个，12.69%）、颞叶（35个，10.84%）；有CMBs组较无CMBs组的MoCA 总分、视空间及执行功能、注意力、抽象思维、延迟回忆、定向力认知功能评分均显著下降（P＜0.05），命名和语 言评分无统计学差异（P＞0.05）；CMBs数量与MoCA总分、注意力、延迟记忆呈负相关（r=－0.260，P=0.028； r=－0.242，P=0.039；r=－0.228，P=0.049）；皮质-皮质下CMBs与MoCA总分、视空间与执行功能、注意力、延 迟回忆呈负相关（r=－0.278，P=0.019；r=－0.231，P=0.045；r=－0.213，P=0.049；r=－0.234，P=0.035）；深部及 幕下CMBs与MoCA总分、注意力呈负相关（r=－0.254，P=0.019；r=－0.239，P=0.028）。结论：CMBs可导致 脑小血管病患者多个认知域的功能障碍，尤其是执行功能、注意力、延迟记忆障碍；检出微出血病灶越多，患 者认知功能受损越严重；CMBs可作为CSVD早期诊断及评价血管性认知障碍程度及其转归的重要指标。

To evaluate cognitive function in cerebral small vessel disease (CSVD) patients without dementia and explore the correlation between the number and location of cerebral microbleeds (CMBs) and the impairment of cognitive function． Methods: A total of 174 consecutive patients with CSVD were prospectively recruited. All patients underwent MRI-susceptibility weighted imaging (SWI) to evaluable CMBs, and the location and number of CMBs were recorded. Patients were divided into the CMBs group (62 cases) and non-CMBs group (112 cases) and evaluated by the Montreal Cognitive Assessment (MoCA) scale to compare cognitive function scores between the two groups. The relationship between the location and number of CMBs and cognitive function scores was analyzed by partial correlation analysis and Spearman’s correlation. Results: The frequency of cortical-subcortical and deep and infratentorial CMBs were 40.87% (132/323) and 59.13% (191/323), respectively. There was a predilection for CMBs occurrence in the basal segment (31.27%), thalamus (12.07% ), frontal lobe (12.69% ), and temporal lobe (10.84% ). Compared to the non-CMBs group, the CMBs group showed a decrease in MoCA total score, visuospatial and executive function, attention, abstract thinking, delayed recall, and orientation (P<0.05); there was no significant difference between the two groups in naming and language scores (P>0.05). The number of CMBs was negatively correlated with the MoCA total score, attention, and delayed recall (r=－0.260, P=0.028; r=－0.242, P=0.039; r=0.228, P=0.049). Cortical-subcortical CMBs was negatively correlated with the MoCA total score, visuospatial and executive function, attention, and delayed recall (r=－ 0.278, P=0.019; r=－ 0.231, P=0.045; r=－ 0.213, P=0.049; r=－ 0.234, P=0.035). Deep CMBs was negatively correlated with the MoCA total score and attention (r=－0.254, P= 0.019; r=－0.239, P= 0.028). Conclusion: CMBs may lead to impaired cognitive function in patients with CSVD, particularly affecting executive function, attention, and delayed recall. Patients experience increased impairment when there is an increase in microbleeds detected. CMBs may serve as a biomarker of CSVD with a significant role in the early diagnosis and prognosis of vascular dementia.