Dale and Deborah Smith Center for Alzheimer's Research and Treatment, Southern Illinois University School ofMedicine, Springfield, IL, USA.
J Alzheimers Dis. 2022;89(3):1119-1129. doi: 10.3233/JAD-220392.
Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) differ in their memory, attention, and visuoconstructional characteristics. The subscales of the well-known Mini-Mental State Examination (MMSE) provide an opportunity to assess these characteristics. Previous research has shown that analysis of the MMSE subscale performance of AD and DLB patients helps to differentiate them.
Study the MMSE scores of AD and DLB patients to see if the ability of previously reported analyses to differentiate them could be improved. Include other dementia patients for perspective.
We studied the MMSEs of all patients seen in our clinics during an 18-month period. Different equations were studied, derived from the subscales of Memory (M, 3 points maximum), Attention (A, 5 points maximum), and Pentagon-copying (P, 1 point maximum).
We obtained 400 MMSEs, 136 from AD patients and 24 from DLB patients, scoring range 1-30. The equation P minus M provided the best discrimination between AD and DLB. Using a P-M score = 1 to identify AD, the positive predictive value was 0.97, negative predictive value 0.22, specificity 0.92, and sensitivity 0.43. As a secondary finding, the P-M = 1 equation was also helpful to differentiate AD from Parkinson's disease dementia.
Considering AD versus DLB in our clinic population, a demented patient who was unable to recall the three memory words on the MMSE but able to copy the intersecting pentagons had a 97% likelihood of having AD. Additional work is needed to improve the sensitivity of the P-M = 1 equation.
阿尔茨海默病（AD）和路易体痴呆（DLB）在记忆、注意力和视空间结构方面存在差异。著名的简易精神状态检查（MMSE）的子量表提供了评估这些特征的机会。先前的研究表明，分析 AD 和 DLB 患者的 MMSE 子量表表现有助于对其进行区分。
研究 AD 和 DLB 患者的 MMSE 评分，以了解之前报告的分析方法是否能够提高区分能力。纳入其他痴呆患者以提供更全面的视角。
我们研究了在 18 个月期间在我们诊所就诊的所有患者的 MMSE。研究了不同的方程，这些方程来自于记忆子量表（M，最高 3 分）、注意力子量表（A，最高 5 分）和五边形复制子量表（P，最高 1 分）。
我们获得了 400 份 MMSE，其中 136 份来自 AD 患者，24 份来自 DLB 患者，评分范围为 1-30。P-M 差值方程提供了 AD 和 DLB 之间最佳的区分能力。使用 P-M 评分=1 来识别 AD，阳性预测值为 0.97，阴性预测值为 0.22，特异性为 0.92，敏感性为 0.43。作为次要发现，P-M=1 方程也有助于区分 AD 与帕金森病痴呆。
在我们的诊所人群中考虑 AD 与 DLB，无法回忆 MMSE 上的三个记忆词但能够复制相交五边形的痴呆患者有 97%的可能性患有 AD。需要进一步研究来提高 P-M=1 方程的敏感性。
