A systematic investigation of the invariance of resting-state network patterns: is resting-state fMRI ready for pre-surgical planning?

Objectives: Measurements of resting-state networks (RSNs) have been used to investigate a wide range of diseases, such as dementia or epilepsy. This raises the question whether this method could also serve as a pre-surgical planning tool. Generating reliable functional connectivity patterns is of crucial importance, particularly for pre-surgical planning, as these patterns may directly affect the outcome. Methods: This study investigated the reproducibility of four commonly used resting-state conditions: fixation of a black crosshair on a white screen; fixation of the center of a black screen; eyes-closed and fixation of the words “Entspann dich!” (Engl., “relax”). Ten healthy, right-handed male subjects (mean age, 25 years; SD 2) participated in the experiment. The spatial overlap for different RSNs across the four conditions was calculated. Results: The spatial overlap across all four conditions was calculated for each seed region on a single subject and at the group level. Activation maps at the single-subject and group levels were highly stable, especially for the reading network (RNW). The lowest consistency measures were found for the visual network (VIN). At the single-subject level spatial overlap values ranged from 0.31 (VIN) to 0.45 (RNW). Conclusion: These findings suggest that RSN measurements are a reliable tool to assess language-related networks in clinical settings. Generally, resting-state conditions showed comparable activation patterns, therefore no specific conditions appears to be preferable.