Electroacupuncture has been considered an effective neurorehabilitative approach to relieve neuropathic pain originating in the central nervous system. However, the neural mechanism underlying the effect of electroacupuncture on pain-relief remains largely unknown. The objective of this study was to investigate the alteration of hub configurations of brain networks caused by the sustained impact of electroacupuncture on a clinically relevant animal model of neuropathic pain. Rats were divided into four groups: normal, model, electroacupuncture, and sham-electroacupuncture. Rats of the last three groups received complete brachial plexus avulsion to evoke neuropathic pain. Electroacupuncture was conducted continuously for three months on the electroacupuncture group, while the sham intervention was performed on the sham-electroacupuncture group. Mechanical withdrawal thresholds were evaluated at the end of the first and third month of intervention. Graph theoretical network analysis compared the regional topological parameters and explored hub configurations of brain networks by longitudinal resting-state fMRI. Three-months electroacupuncture showed a significant pain-relief effect. Not the spatial distribution of hubs, but the hubness distribution showed a significant difference among groups after a three-month intervention. The proportion of more highly connected hub regions was significantly higher in the model rats than the normal rats, while that of the electroacupuncture group was considerably lower than the model group. Additionally, regional parameter changes showed a very similar distribution of hub proportions. It was concluded that long-term electroacupuncture might restore an adaptive equilibrium to a disrupted network and suppress maladaptive plastic changes that follow neuropathic pain. This may provide an important avenue for future strategies appropriate for therapeutic interventions.