Recent functional neuroimaging studies have examined cognitive inhibitory control, decision-making and stress regulation in heroin addiction using a cue-reactivity paradigm. Few studies have considered impairments in heroin users from an integrated perspective for evaluation of their brain functions. We hypothesized that the brain regions that are dysregulated in the chronic heroin users during cue-reactivity studies may also show dysfunctional connectivity in memory, inhibition and motivation-related dysfunctions during a resting state free of cues. The present study used resting functional magnetic resonance imaging (fMRI) to compare the interaction of brain regions between 12 chronic heroin users and 12 controls by employing a novel graph theory analysis (GTA) method. As a data-driven approach, GTA has the advantage of evaluating the strength as well as the temporal and spatial patterns of interactions among the brain regions. Abnormal topological properties were explored in the brain of chronic heroin users, such as the dysfunctional connectivity in the prefrontal cortex, ACC, SMA, ventral striatum, insula, amygdala and hippocampus. Our results suggest that GTA is a useful tool in defining dysregulated neural networks even during rest. This dysfunctional brain connectivity may contribute to decrease self-control, impaired inhibitory function as well deficits in stress regulation in chronic heroin users.