BackgroundMitochondrial encephalomyopathy with lactic acidosis and stroke‐like episodes (MELAS) is a rare maternally inherited genetic disease; however, little is known about its underlying brain basis. Furthermore, the dynamic functional connectivity (dFC) of brain networks in MELAS has not been explored.PurposeTo investigate the abnormalities of dFC in patients with MELAS at the acute and chronic stages, and to determine the possible relations between dynamic connectivity alterations and volumes of stroke‐like lesions (SLLs).Study TypeProspective.SubjectsTwenty‐two MELAS patients at the acute stage, 23 MELAS patients at the chronic stage, and 22 healthy controls.Field Strength/SequenceSingle‐shot gradient‐recalled echo planar imaging (EPI) sequence at 3T.AssessmentDynamic FC states were estimated using the sliding window approach and k‐means clustering analyses. Combined with graph theory, the topological properties of the dFC network were also accessed.Statistical TestsPermutation test, Pearson correlation coefficient, and false discovery rate correction.ResultsWe identified four dFC states and found that MELAS patients (especially at the acute stage) spent more time in a state with weaker connectivity (state 1) and less time in states with stronger connectivity. In addition, volumes of acute SLLs were positively correlated with mean dwell time in state 1 (r = 0.539, P < 0.05) and negatively correlated with the number of transitions (r = −0.520, P < 0.05). Furthermore, MELAS patients at the acute stage exhibited significantly increased global efficiency (P < 0.01) and decreased local efficiency (P < 0.001) compared to the controls and the patients at the chronic stage. Patients at the chronic stage only showed significantly (P < 0.001) decreased local efficiency compared to the controls.Data ConclusionOur findings suggest similar and distinct dFC alterations in MELAS patents at the acute and chronic stages, providing novel insights for understanding the neuropathological mechanisms of MELAS.Level of Evidence 2Technical Efficacy Stage Stage 2J. MAGN. RESON. IMAGING 2021;53:427–436.