Department of Psychiatry and Psychotherapy, University Medical Center Hamburg-Eppendorf, Martinistr. 52, W 37, 20246 Hamburg, Germany.
Eur Arch Psychiatry Clin Neurosci. 2013 Jun;263(4):335-44. doi: 10.1007/s00406-012-0350-7. Epub 2012 Sep 1.
Fully automated magnetic resonance imaging (MRI)-based volumetry may serve as biomarker for the diagnosis in patients with mild cognitive impairment (MCI) or dementia. We aimed at investigating the relation between fully automated MRI-based volumetric measures and neuropsychological test performance in amnestic MCI and patients with mild dementia due to Alzheimer's disease (AD) in a cross-sectional and longitudinal study. In order to assess a possible prognostic value of fully automated MRI-based volumetry for future cognitive performance, the rate of change of neuropsychological test performance over time was also tested for its correlation with fully automated MRI-based volumetry at baseline. In 50 subjects, 18 with amnestic MCI, 21 with mild AD, and 11 controls, neuropsychological testing and T1-weighted MRI were performed at baseline and at a mean follow-up interval of 2.1 ± 0.5 years (n = 19). Fully automated MRI volumetry of the grey matter volume (GMV) was performed using a combined stereotactic normalisation and segmentation approach as provided by SPM8 and a set of pre-defined binary lobe masks. Left and right hippocampus masks were derived from probabilistic cytoarchitectonic maps. Volumes of the inner and outer liquor space were also determined automatically from the MRI. Pearson's test was used for the correlation analyses. Left hippocampal GMV was significantly correlated with performance in memory tasks, and left temporal GMV was related to performance in language tasks. Bilateral frontal, parietal and occipital GMVs were correlated to performance in neuropsychological tests comprising multiple domains. Rate of GMV change in the left hippocampus was correlated with decline of performance in the Boston Naming Test (BNT), Mini-Mental Status Examination, and trail making test B (TMT-B). The decrease of BNT and TMT-A performance over time correlated with the loss of grey matter in multiple brain regions. We conclude that fully automated MRI-based volumetry allows detection of regional grey matter volume loss that correlates with neuropsychological performance in patients with amnestic MCI or mild AD. Because of the high level of automation, MRI-based volumetry may easily be integrated into clinical routine to complement the current diagnostic procedure.
基于全自动磁共振成像（MRI）的容积测量可能成为轻度认知障碍（MCI）或痴呆患者诊断的生物标志物。我们旨在通过横断面和纵向研究，调查在遗忘型 MCI 和轻度阿尔茨海默病（AD）痴呆患者中，基于全自动 MRI 的容积测量与神经心理学测试表现之间的关系。为了评估基于全自动 MRI 的容积测量对未来认知表现的可能预后价值，还测试了神经心理学测试表现随时间的变化率与基线时基于全自动 MRI 的容积测量之间的相关性。在 50 名受试者中，18 名患有遗忘型 MCI，21 名患有轻度 AD，11 名对照，在基线时和平均 2.1 ± 0.5 年的随访间隔（n = 19）进行了神经心理学测试和 T1 加权 MRI。使用 SPM8 提供的联合立体定向归一化和分割方法以及一组预定义的二进制叶掩模，对灰质体积（GMV）进行全自动 MRI 容积测量。左、右海马掩模源自概率细胞构筑图。还从 MRI 中自动确定内、外液空间的体积。使用 Pearson 检验进行相关分析。左海马 GMV 与记忆任务的表现显著相关，左颞叶 GMV 与语言任务的表现相关。双侧额、顶和枕叶 GMV 与包含多个领域的神经心理学测试的表现相关。左海马 GMV 的变化率与波士顿命名测试（BNT）、简易精神状态检查和连线测试 B（TMT-B）的表现下降相关。随着时间的推移，BNT 和 TMT-A 表现的下降与多个脑区灰质的丢失相关。我们得出结论，基于全自动 MRI 的容积测量可以检测到与遗忘型 MCI 或轻度 AD 患者神经心理学表现相关的区域性灰质体积损失。由于高度自动化，基于 MRI 的容积测量可以很容易地整合到临床常规中，以补充当前的诊断程序。
