Vandenberghe Rik
Department of Neurosciences, Laboratory for Cognitive Neurology, KU Leuven, Leuven, Belgium.
Neurology Department, University Hospitals Leuven, Leuven, Belgium.
Alzheimers Res Ther. 2016 Apr 21;8(1):16. doi: 10.1186/s13195-016-0185-y.
Highly influential recommendations published in 2011 for the classification of the primary progressive aphasias (PPA) distinguished three subtypes: the semantic variant, the nonfluent/agrammatic variant, and the logopenic variant. We review empirical evidence published after 2011 that bears relevance to the validity of the recommended classification scheme. The studies that we review principally rely on monocentric, memory clinic-based consecutive series of PPA patients. We review whether a data-driven analysis of neurolinguistic test scores confirms the subtyping that was based on expert consensus, whether the 2011 subtyping covers the diversity of PPA in a comprehensive manner, and whether the proposed subgroups differ along dimensions that are not explicitly part of the defining criteria, such as diffusion tractography. Data-driven mathematical analyses of neurolinguistic data in PPA broadly confirm the presence of separate clusters corresponding to the subtypes but also leave 15-30% unclassified. A comprehensive description of PPA requires the addition of the mixed variant as a fourth subtype and needs to leave room for cases fulfilling the criteria for a root diagnosis of PPA but not those of any of the three subtypes. Finally, given the limited predictive value of the clinical phenotype for the underlying neuropathology, biomarkers of the underlying pathology are likely of clinical utility in PPA.
2011年发布的关于原发性进行性失语(PPA)分类的极具影响力的建议区分了三种亚型:语义变异型、非流畅/语法缺失变异型和音韵性变异型。我们回顾了2011年之后发表的与推荐分类方案的有效性相关的实证证据。我们所回顾的研究主要依赖于以记忆门诊为基础的单中心连续系列PPA患者。我们探讨了对神经语言学测试分数进行数据驱动分析是否能证实基于专家共识的亚型划分,2011年的亚型划分是否全面涵盖了PPA的多样性,以及所提出的亚组在并非明确作为定义标准一部分的维度上(如弥散张量成像)是否存在差异。对PPA神经语言学数据进行数据驱动的数学分析大致证实了与各亚型相对应的独立聚类的存在,但仍有15% - 30%未分类。对PPA进行全面描述需要增加混合变异型作为第四种亚型,并需要为符合PPA根诊断标准但不符合三种亚型中任何一种标准的病例留出空间。最后,鉴于临床表型对潜在神经病理学的预测价值有限,潜在病理学的生物标志物在PPA中可能具有临床应用价值。