Department of Biomedical Engineering, University of Basel, Gewerbestrasse 14, 4123, Allschwil, Switzerland; Medical Image Analysis Center (MIAC) AG, Marktgasse 8, 4051, Basel, Switzerland.
Medical Image Analysis Center (MIAC) AG, Marktgasse 8, 4051, Basel, Switzerland; Department of Neurology, University Hospital of Basel, Petersgraben 4, 4031, Basel, Switzerland.
Neuroimage. 2019 Apr 1;189:727-733. doi: 10.1016/j.neuroimage.2019.01.052. Epub 2019 Jan 24.
Subcortical T2-weighted (T2w) lesions are very common in older adults and have been associated with dementia. However, little is known about the strategic lesion distribution and how lesion patterns relate to vascular risk factors and cognitive impairment.
The aim of this study was to analyze the association between T2w lesion load and location, vascular risk factors, and cognitive impairment in a large cohort of older adults.
1017 patients participating in a large prospective cohort study (INtervention project on cerebroVAscular disease and Dementia in the district of Ebersberg, INVADE II) were analyzed. Cerebral T2w white matter and deep grey matter lesions, the so-called white matter hyperintensities (WMHs), were outlined semi-automatically on fluid attenuated inversion recovery images and normalized to standard stereotaxic space (MNI152) by non-linear registration. Patients were assigned to either a low-risk or a high-risk group. The risk assessment considered ankle brachial index, intima media thickness, carotid artery stenosis, atrial fibrillation, previous cerebro-/cardiovascular events and peripheral artery disease as well as a score based on cholesterol levels, blood pressure and smoking. Separate lesion distributions were obtained for the two risk groups and compared using voxel-based lesion-symptom mapping. Moreover, we assessed the relation between lesion location and cognitive impairment (demographically adjusted z-scores of the Consortium to Establish a Registry for Alzheimer's Disease Neuropsychological Assessment Battery Plus, CERAD-NAB Plus) using voxel-based statistics (α = 0.05).
A total of 878 out of 1017 subjects (86%) had evaluable MRI data and were included in the analyses (mean age: 68.2 ± 7.6 years, female: 515). Patients in the high-risk group were characterized by a significantly higher age, a higher proportion of men, a higher lesion load (p < 0.001), and a worse performance in some of the cognitive subdomain scores (p < 0.05). Voxels with significant associations to the subjects' cerebrovascular risk profiles were mainly found at locations of the corpus callosum, superior corona radiata, superior longitudinal fasciculus, internal and external capsule, and putamen. While several cognitive domains have shown significant associations with the participants' total lesion burden (p < 0.05), no focal WMH locations were found to be associated with cognitive impairment.
Age, gender, several cognitive scores, and WMH lesion load were shown to be significantly associated with vascular risk factors in a population of older, but cognitively preserved adults. Vascular risk factors seem to promote lesion formation most severely at well-defined locations. While lesion load showed weak associations to some cognitive scores, no focal locations causing specific cognitive disturbances were identified in this large cohort of older adults.
皮质下 T2 加权（T2w）病变在老年人中非常常见，并与痴呆有关。然而，对于病变的分布以及病变模式与血管危险因素和认知障碍的关系知之甚少。
本研究旨在分析较大队列中老年患者 T2w 病变负荷与位置、血管危险因素和认知障碍之间的关系。
对参加大型前瞻性队列研究（Ebersberg 区脑血管病和痴呆干预项目，INVADE II）的 1017 例患者进行分析。在液体衰减反转恢复图像上半自动勾画脑 T2w 白质和深部灰质病变，即所谓的脑白质高信号（WMHs），并通过非线性配准将其归一化为标准立体定向空间（MNI152）。患者被分为低危组或高危组。风险评估考虑了踝臂指数、内膜中层厚度、颈动脉狭窄、心房颤动、既往脑/心血管事件和外周动脉疾病，以及基于胆固醇水平、血压和吸烟的评分。为两组分别获得病变分布，并使用基于体素的病变-症状映射进行比较。此外，我们还使用基于体素的统计学方法（α=0.05）评估了病变位置与认知障碍（根据 Consortium to Establish a Registry for Alzheimer's Disease Neuropsychological Assessment Battery Plus，CERAD-NAB Plus 的人口统计学调整后的 z 分数）之间的关系。
共有 1017 例患者中的 878 例（86%）有可评估的 MRI 数据，并纳入分析（平均年龄：68.2±7.6 岁，女性：515 例）。高危组患者的年龄显著更高，男性比例更高，病变负荷更高（p<0.001），一些认知子域评分的表现更差（p<0.05）。与受试者的脑血管风险特征有显著关联的体素主要位于胼胝体、上放射冠、上纵束、内囊和外囊以及壳核。虽然一些认知领域与参与者的总病变负荷有显著相关性（p<0.05），但没有发现特定的 WMH 位置与认知障碍相关。
在一个认知功能正常的老年人群中，年龄、性别、几个认知评分和 WMH 病变负荷与血管危险因素显著相关。血管危险因素似乎在明确的部位最严重地促进病变形成。虽然病变负荷与一些认知评分有较弱的相关性，但在这个较大的老年人群中，没有发现导致特定认知障碍的特定病变位置。
