Aphasia in Neurodegenerative Diseases: Can Language Impairment Predict the Underlying Pathology ? (P02.051)

Abstract

Objective: To determine the value of clinical assessment, MRI cortical thickness and [18FDG]-PET metabolism in predicting pathophysiology defined by cerebro-spinal fluid (CSF) biomarkers in patients suffering of predominant language difficulties of a neurodegenerative origin. Background Aphasia is common in neurodegenerative diseases, and may be the main symptom, as in all 3 variants of progressive primary aphasia (PPA). Some studies have tried to predict the underlying pathology. Each variant is associated with a predominant histopathology, but it can only reflect group-wide probabilities. We decided to consider the underlying pathology (Alzheimer9s (AD) or non-AD), and check subsequently if any clinical, structural or metabolic differences could emerge between both defined groups. Design/Methods: 26 aphasic patients underwent comprehensive language and neuropsychological assessments, structural MRI, [18FDG]-PET, and CSF analysis; eleven age-matched healthy controls (HC) underwent the same examinations except for the lumbar puncture. Biomarkers in CSF allowed us to classify patients as AD or non-AD. Results: 11 subjects had biomarkers consistent with AD, leaving 15 in the non-AD group. Compared to the HC, language impairment was the main symptom for all patients. However, no differences between groups emerged in language assessments. AD patients exhibit more praxis and visuo-spatial difficulties compared to the non AD group. The difference was significant but less marked for memory. MRI showed more posterior and symmetric cortical thickness reduction in the AD group. No difference in the medial temporal lobe was observed between the two patients groups. Temporo-parietal hypometabolism was more marked in the AD group but no difference in frontal cortex was found. Conclusions: Language impairment is not a reliable predictor of pathology in our patients, though it is the main symptom. Clinical presentation seems to depend on anatomic damage sites more than the underlying pathology. Disclosure: Dr. Sagot has nothing to disclose. Dr. Saint-Aubert has nothing to disclose. Dr. Bezy has nothing to disclose. Dr. Mirabel has nothing to disclose. Dr. Payoux has nothing to disclose. Dr. Dumas has nothing to disclose. Dr. Vincent has nothing to disclose. Dr. Peran has nothing to disclose. Dr. Barbeau has nothing to disclose. Dr. Puel has nothing to disclose. Dr. Chollet has nothing to disclose. Dr. Pariente has nothing to disclose.