Ichimiya A
Department of Neuropsychiatry, Faculty of Medicine, Kyushu University.
Seishin Shinkeigaku Zasshi. 1994;96(1):1-25.
A new rating method was designed to evaluate five symptoms of dementia; disorders of memory, cognition, language, visuospatial functions, and personality, which are defined as recent neuropsychological criteria for dementias by Cummings and Benson. Thirty-two items were made from the items of the Hasegawa dementia scale and neuropsychological tests. Scales for the five symptoms of dementia were then constituted of the items. In semi-structured interview, 113 demented patients and 16 normal elderly volunteers were rated. Interrater reliability of the five symptom scales were tested by ANOVA ICC and all the symptoms had a significant reliability. The scores of the five symptoms were correlated with each other and with the scores of the Hasegawa dementia scale. A Principal Component Analysis was applied to the scores of the five symptoms. The largest principal component had correlations with the scores of the Hasegawa dementia scale and the scores of the all five symptoms and was interpreted as a factor for severity of dementia. This component explains about 70% of total variance of the data. The other 30% of the variance were explained by four principal components which were interpreted as factors of the differences of the scores of the symptom scales. Relations between the scores of the symptoms and regional cerebral blood flow (rCBF) measured using SPECT were investigated in 25 moderately demented patients which met the criteria of probable Alzheimer's disease of NINCDS-ADRDA. rCBF was calculated as a ratio of 123I-IMP uptake of the areas to the mean uptake of right and left cerebellum. Significant correlations were found between the scores of memory symptom and the rCBF of the left temporal lobe (r = 0.59: p < 0.01) and between the scores of personality changes and the rCBF of the both frontal lobes (r = 0.58, 0.54: p < 0.01). Results suggest the possibility that the distribution of brain damage reflects symptoms in dementia and also supports the validity of analytic evaluation of dementia symptoms by this scale.
记忆障碍、认知障碍、语言障碍、视觉空间功能障碍和人格障碍，这些症状被卡明斯和本森定义为痴呆症最新的神经心理学标准。从长谷川痴呆量表和神经心理学测试项目中选取了32个项目。然后由这些项目构成痴呆症五种症状的量表。在半结构化访谈中，对113名痴呆患者和16名正常老年志愿者进行了评分。通过方差分析组内相关系数（ANOVA ICC）对五种症状量表的评分者间信度进行了测试，所有症状均具有显著的信度。五种症状的得分彼此相关，且与长谷川痴呆量表的得分相关。对五种症状的得分进行了主成分分析。最大的主成分与长谷川痴呆量表的得分以及所有五种症状的得分相关，被解释为痴呆严重程度的一个因素。该成分解释了数据总方差的约70%。其余30%的方差由四个主成分解释，这四个主成分被解释为症状量表得分差异的因素。在25名符合NINCDS-ADRDA可能阿尔茨海默病标准的中度痴呆患者中，研究了症状得分与使用单光子发射计算机断层扫描（SPECT）测量的局部脑血流量（rCBF）之间的关系。rCBF计算为各区域123I-异碘安非他明（123I-IMP）摄取量与左右小脑平均摄取量的比值。发现记忆症状得分与左颞叶rCBF之间存在显著相关性（r = 0.59，p < 0.01），人格改变得分与双侧额叶rCBF之间存在显著相关性（r = 0.58，0.5,4：p < 0.01）。结果表明，脑损伤的分布可能反映痴呆症症状，也支持该量表对痴呆症症状进行分析评估的有效性。
