Cohort study for a cognitive decline in elderly people living in a Japanese community.

基于体素的阿尔茨海默病特定区域分析(VSRAD)是一款免费提供的软件，使用基于体素的形态计量学检测包括内嗅皮层在内的内侧颞区萎缩，以早期诊断阿尔茨海默病。方法：对日本农村社区非痴呆老年人182例(男53例，女129例)进行头颅MRI检查，年龄65~91岁，平均年龄78岁(SD 6.4)，并于2004~2006年(时间A)进行简明精神状态检查(MMSE)、额头评定量表(FAB)和临床痴呆评定量表(CDR)评定。三年后进行第二次认知功能评估，以评估VSRAD参数与随后的认知缺陷之间的关系。此外，还研究了年龄、性别、教育程度、饮酒、高血压和糖尿病对认知功能下降的影响。结果：代表内侧颞叶萎缩的VSRAD在B时的Z-Score与3年间MMSE的变化呈负相关(r=-.360，P&t；0.001)，而与FAB的变化不相关，与CDR的变化呈正相关(r=.342，P&t；0.001)。与B时刻CDR为0的受试者相比，在B时刻CDR增加到0.5的受试者在A时刻的VSRAD内侧颞区萎缩指数更大。这些数据表明，在社区痴呆患病率研究中，VSRAD可以为预测随后的认知缺陷提供有用的信息。

The Voxel-based Specific Regional Analysis for Alzheimer's Disease (VSRAD) is a freely available software to detect the atrophy of medial temporal areas including the entorholinal cortex using voxel-based morphometry for early diagnosis of Alzheimer's disease. Methods : Cranial MRI with VSRAD was performed on 182 non-demented elderly subjects (male n=53 ; female n=129) aged from 65 to 91 years of age, with a mean age of 78 (SD 6.4) who resided in a Japanese rural community, followed by measurement of Mini-Mental State Examination (MMSE), Frontal Assessment Battery (FAB) and Clinical Dementia Rating (CDR) in 2004 to 2006 (Time A). The second assessment of the cognitive functions was performed 3 years later to assess the relationship between the parameters of VSRAD and consequent cognitive deficits in the subjects. Effects of age, sex, education, alcohol intake, hypertension and diabetes on consequent cognitive decline were also investigated. Results : The Z-score of medial temporal areas of VSRAD at Time B which represents the atrophy of the area was negatively correlated with the changes in the score of MMSE during 3 years (r=-.360, P<0.001), but not in the score of FAB and positively correlated with the changes in the score of CDR (r=.342, P<0.001). The subjects with a consequent cognitive decline assessed by increased CDR to 0.5 at Time B showed a greater atrophy index of medial temporal areas of VSRAD at Time A, compared with the subjects with CDR 0 at Time B. These data indicate that VSRAD could provide an useful information to predict consequent cognitive deficits in a community prevalence study of dementia.