Objective Functional connectivity is essential for brain functions. Focal brain lesions could have important remote effects on the functions of distant brain regions even if outside the damaged area. Network dysfunction may contribute to neurological deficits observed for example after stroke. We explored how functional network characteristics, measured via EEG recordings, correlate with the lesion volume on MRI. Methods Graph characteristics of EEG data of 30 stroke patients in the acute stage (no more than 5 days after the event), were examined. EEG connectivity analysis was performed using eLORETA in both affected and unaffected hemispheres. Lesional volume were evaluated by MRI. Results Considering all subjects as a group, in the stroke hemisphere both Path length and Clustering positively correlate with lesion volume in delta and theta bands and negatively in alpha1 band. Evaluating the contralateral hemisphere, no significant correlation was observed. Conclusions Lesion volume correlates with a progressive reduction of brain interconnection as provided by the proportional increment of connectivity in low frequency and reduction in higher frequency bands. Key message These results support the study of EEG via graph theory parameters for assessing the connectivity characteristics of brain networks in stroke suggesting a possible implications also in their prognosis.