Department of Psychiatry and Psychology, Mayo Clinic Arizona, Scottsdale, AZ, USA.
Arizona Alzheimer's Consortium, Phoenix, AZ, USA.
J Alzheimers Dis. 2018;63(2):603-615. doi: 10.3233/JAD-170852.
Brain imaging measurements can provide evidence of possible preclinical Alzheimer's disease (AD). Their ability to predict individual imminent clinical conversion remains unclear.
To investigate the ability of pre-specified volumetric magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission tomography (FDG-PET) measurements to predict which cognitively unimpaired older participants would subsequently progress to amnestic mild cognitive impairment (aMCI) within 2 years.
From an apolipoprotein E4 (APOE4) enriched prospective cohort study, 18 participants subsequently progressed to the clinical diagnosis of aMCI or probable AD dementia within 1.8±0.8 years (progressors); 20 participants matched for sex, age, education, and APOE allele dose remained cognitively unimpaired for at least 4 years (nonprogressors). A complementary control group not matched for APOE allele dose included 35 nonprogressors. Groups were compared on baseline FDG-PET and MRI measures known to be preferentially affected in the preclinical and clinical stages of AD and by voxel-wise differences in regional gray matter volume and glucose metabolism. Receiver Operating Characteristic, binary logistic regression, and leave-one-out procedures were used to predict clinical outcome for the a priori measures.
Compared to non-progressors and regardless of APOE-matching, progressors had significantly reduced baseline MRI and PET measurements in brain regions preferentially affected by AD and reduced hippocampal volume was the strongest predictor of an individual's imminent progression to clinically significant memory decline (79% sensitivity/78% specificity among APOE-matched cohorts).
Regional MRI and FDG-PET measurements may be useful in predicting imminent progression to clinically significant memory decline.
脑影像学测量可提供可能的临床前阿尔茨海默病（AD）的证据。它们预测个体即将发生临床转化的能力尚不清楚。
研究预先指定的容积磁共振成像（MRI）和氟脱氧葡萄糖正电子发射断层扫描（FDG-PET）测量值预测哪些认知正常的老年人在 2 年内随后会发展为遗忘型轻度认知障碍（aMCI）的能力。
从载脂蛋白 E4（APOE4）丰富的前瞻性队列研究中，18 名参与者随后在 1.8±0.8 年内进展为临床诊断的 aMCI 或可能的 AD 痴呆（进展者）；20 名参与者按性别、年龄、教育程度和 APOE 等位基因剂量匹配，认知能力至少保持 4 年未受损（非进展者）。一个不匹配 APOE 等位基因剂量的补充对照组包括 35 名非进展者。在基线 FDG-PET 和 MRI 测量方面对各组进行比较，这些测量值已知在 AD 的临床前和临床阶段优先受到影响，并且通过区域灰质体积和葡萄糖代谢的体素差异进行比较。使用接受者操作特征、二元逻辑回归和留一法程序对预先测量的临床结果进行预测。
与非进展者相比，无论是否匹配 APOE，进展者的大脑区域的 MRI 和 PET 基线测量值明显降低，这些区域优先受到 AD 的影响，而海马体积的降低是个体即将发生临床显著记忆下降的最强预测因素（在 APOE 匹配队列中，敏感性为 79%/特异性为 78%）。
区域 MRI 和 FDG-PET 测量值可能有助于预测即将发生的临床显著记忆下降。
