AN ATTEMPT TO LOOK FOR DIFFERENCE BETWEEN DEGENERATIVE AND VASCULAR APHASIA: A DATA-DRIVEN CLUSTER ANALYSIS.

Objectives: Vascular and degenerative damage are so different brain pathologies that the impact on cognition cannot be expected to be the same. The present study aims to identify possible markers able to distinguish vascular and neurodegenerative aphasias. Materials: 18 logopenic variant/mixed primary progressive aphasias (lv-mPPA) and 11 fluent post-stroke aphasics (psA) of comparable severity, and 12 matched healthy controls (HC) were enrolled. All participants underwent a comprehensive neuropsychological battery, and the examination of macro and microlinguistic aspects of language. Methods: A Principal Component Analysis (PCA) was performed (after z-transformation of the scores) to identify factors explaining the overall performance on patient groups only; factor scores were entered into a cluster analysis by fixing 2-groups solution in order to verify if lv-mPPA and psA were assigned to different clusters. Results: The PCA extracted two factors, “global cognition/lexical semantic factor” and “morphosyntactic factor”, accounting for 42.9% of variability, and two factors loading macrolinguistic variables (accounting for 22.7% of variability). A two-step cluster analysis, exclusively based on behavioural data, allowed to some extent the re-emergence of the two etiologies: a first cluster composed of 18 subjects (78% lv-mPPA and 22% PSA), a second cluster composed of 11 subjects (64% PSA and 36%lvPPA patients). Subjects of the first cluster were more impaired both in global cognition/lexical semantic and morphosyntactic factors. Macrolinguistic factors did not differentiate instead, the two clusters. Lv-mPPA that fell in the first cluster showed widespread cortical thinning in both the cerebral hemispheres (with left prevalence) including the temporal poles; atrophy was confined in the left hemisphere, and the temporal pole was relatively spared in lv-mPPA of the second cluster. The ischemic area was distributed in the posterior regions of the left middle cerebral artery territory in all PSA, and the anterior portion of the temporal lobe was always spared. Underestimation of the severity of leukoencephalopathy with modification of the intra and interhemispheric connectivity might have generated inclusion of PSA in the first cluster (with a majority of PPA) in spite of monohemispheric lesion. Discussion: In lv-mPPA the cognitive damage tends to be more widespread than expected on the basis of a focal symptom such as aphasia, and primary language disorders seems intrinsically more severe than vascular deficits. Conclusions: Aphasic syndromes are largely dependent on damage localization and extension, but not on etiology as such. -Marini, A., Andretta, S., Del Tin, S., & Carlomagno, S. A multi-level approach to the analysis of narrative language in aphasia. Aphasiology, 2011, 25(11), 1372-1392. McNeil, M., & Duffy, J. Primary progressive aphasia. In R. Chapey (ed.), Language intervention strategies in aphasia and related neurogenic communication disorders (4 th ed.), 2001, pp. 472-486. Philadelphia: Lippincott Williams & Wilkins. -Mesulam, M. M., Rogalski, E. J., Wieneke, C., Hurley, R. S., Geula, C., Bigio, E.H., Thompson, C.K., & Weintraub, S. Primary progressive aphasia and evolving neurology of the language network. Nature reviews. Neurology, 2014, 10 (10), pp. 554-569