Dynamic brain functional connectivity analysis in  migraine: a resting-state fMRI study

Cláudia Marina Alves Fonseca

Key information

Authors:

Cláudia Marina Alves Fonseca (Cláudia Marina Alves Fonseca)

Supervisors:

Patrícia Margarida Piedade Figueiredo (Patrícia Figueiredo); Raquel Santos Gil-Gouveia

Published in

November 16, 2021

Abstract

Migraine is an incapacitating neurological disorder characterised by recurring, throbbing headache attacks, generally combined with nausea, vomiting, sensory and cognitive disturbances. Migraine has become a major public health concern with a huge impact worldwide, however, its pathophysiology is not entirely understood. In order to investigate the disorder mechanisms, dynamic functional connectivity (dFC) in migraineurs was assessed through the analysis of resting-state fMRI data. Eight female menstrual episodic migraine patients without aura were scanned in the interictal (attack-free) phase of the migraine cycle, five of the eight patients in the ictal (attack) phase, and six healthy controls in the menstrual mid-cycle/post-ovulation phase to control hormonal variation in the interictal phase. To estimate the dFC, the sliding window and phase coherence methods were tested. Then, the leading eigenvector dynamic analysis focused on the dominant patterns of dFC matrices captured by leading eigenvectors, further organised into recurrent dFC states with the k-means clustering algorithm. Finally, between-group differences were statistically assessed with non-parametric permutation-based t-tests. Results revealed a significant increased mean lifetime and probability of occurrence in dorsal attention/frontoparietal, somatomotor and visual networks, and decreased temporal metrics in the fully connected state (global mode) in migraine patients in the ictal phase and controls compared to the interictal phase. This work reinforced the relevance of using dFC to study migraine brain, and suggested that its dysfunction might be associated with an altered dynamic of attentional/cognitive and stimulus processing systems and the global mode, which might potentially constitute neuroimaging biomarkers for disease progression.