Multimodal MRI cerebral correlates of verbal fluency switching and its impairment in women with depression

L. Domain a, M. Guillery a, N. Linz b, A. König c, d, J.M. Batail a, R. David e, I. Corouge f, E. Bannier f, J.C. Ferré f, T. Dondaine g, D. Drapier a, G.H. Robert a, f, *


a Universitary Department of Psychiatry, Centre Hospitalier Guillaume R ́egnier, Rennes, France
b ki:elements, Saarbrücken, Germany
c Stars Team, Institut National de Recherche en Informatique et en Automatique (INRIA), Sophia Antipolis, France d CoBTeK (Cognition-Behaviour-Technology) Lab, FRIS-University Coˆte d’Azur, Nice, France
e Old-age Psychiatry DEPARTMENT, Geriatry Division, University of Nice, France
f U1228 Empenn, UMR 6074, IRISA, University of Rennes 1, France
g Univ. Lille, Inserm, CHU Lille, LilNCog, Lille Neuroscience & Cognition, F-59000 Lille, France

Abstract

Multimodal MRI cerebral correlates of verbal fluency switching and its impairment in women with depression

Background: The search of biomarkers in the field of depression requires easy implementable tests that are biologically rooted. Qualitative analysis of verbal fluency tests (VFT) are good candidates, but its cerebral cor- relates are unknown.
Methods: We collected qualitative semantic and phonemic VFT scores along with grey and white matter anatomical MRI of depressed (n = 26) and healthy controls (HC, n = 25) women. Qualitative VFT variables are the “clustering score” (i.e. the ability to produce words within subcategories) and the “switching score” (i.e. the ability to switch between clusters). The clustering and switching scores were automatically calculated using a data-driven approach. Brain measures were cortical thickness (CT) and fractional anisotropy (FA). We tested for associations between CT, FA and qualitative VFT variables within each group.
Results: Patients had reduced switching VFT scores compared to HC. Thicker cortex was associated with better switching score in semantic VFT bilaterally in the frontal (superior, rostral middle and inferior gyri), parietal (inferior parietal lobule including the supramarginal gyri), temporal (transverse and fusiform gyri) and occipital (lingual gyri) lobes in the depressed group. Positive association between FA and the switching score in semantic VFT was retrieved in depressed patients within the corpus callosum, right inferior fronto-occipital fasciculus, right superior longitudinal fasciculus extending to the anterior thalamic radiation (all p < 0.05, corrected).
Conclusion: Together, these results suggest that automatic qualitative VFT scores are associated with brain anatomy and reinforce its potential use as a surrogate for depression cerebral bases.

Full paper below: