Increased Resting-State Gamma-Band Connectivity in First-Episode Schizophrenia

Abstract

Background: Schizophrenia has long been suggested to represent a disorder with prominent neural dysconnectivity. Gamma-band oscillations are highly relevant in this context, due both to their proposed involvement in neuronal synchronization and to their association with neurotransmitter systems relevant for schizophrenia. Several task-related studies have confirmed reduced power and synchronization of gamma-band oscillations in schizophrenia, but it has been suggested that these findings might not apply to the resting state. The present study aimed to investigate resting-state gamma-band connectivity in patients with schizophrenia.

Methods: Sixty-four channel resting-state electroencephalography (eyes closed) was recorded in 22 patients with first-episode schizophrenia and 22 healthy controls matched for age and gender. Orthogonalized power envelope correlation was used as a measure of connectivity across 80 cortical regions at 40 Hz. Mean connectivity at each region was compared across groups using the nonparametric randomization approach. Additionally, the network-based statistic was applied to identify affected networks in patients.

Results: Patients displayed increased mean functional gamma-band connectivity compared to controls in the left rolandic operculum. Network-based analyses indicated increased connectivity in patients within a strongly lateralized network consisting mainly of left inferior frontal/orbitofrontal, lateral and medial temporal, and inferior parietal areas. Within this network, gamma-band connectivity was higher in patients with low positive and disorganization symptom levels.

Conclusions: The present study provides a link between resting-state gamma-band connectivity and the core symptoms of schizophrenia. The observed findings are different than those reported by task-related studies, suggesting that resting-state studies might reveal new aspects in the pathophysiology of schizophrenia.

Keywords: EEG; disorganization; functional connectivity; positive symptoms; power envelope correlation; psychosis.

Fig. 1.

Network of increased resting-state gamma-band connectivity in patients with schizophrenia compared to healthy controls. The size of each node represents its number of connections within the network (degree). Labels are provided for nodes with at least 8 connections. As seen in table 1, these nodes are responsible for about 50% of interactions within the network. Ang, angular gyrus; FrIT, inferior frontal gyrus, pars triangularis; FrMO = medial orbitofrontal gyrus; FSupO, superior frontal gyrus, orbital part; InfT, inferior temporal gyrus; MidT, middle temporal gyrus; RolOp, rolandic operculum; PreC, precentral gyrus; PostC, postcentral gyrus. Color coding (online version only): inferior frontal and orbitofrontal cortex, orange; medial frontal cortex, light blue; lateral frontal cortex, red; remporal cortex and insula, yellow; parietal cortex, dark blue. The figure was created using BrainNet Viewer (http://www.nitrc.org/projects/bnv/, last accessed on Aug 8, 2014).

Fig. 2.

Mean gamma-band connectivity across the 33-node network that displayed significant differences between patients and healthy controls (HC). The patient group is split into 2 depending on the severity of positive (left panel) and disorganized (right panel) symptoms. Significant differences are marked with an asterisk.