Functional network alterations in adult obstructive sleep apnea：A resting-state fNIRS study

Abstract Study Objectives: To evaluate brain network connectivity characteristics and complex network topology properties in adult with obstructive sleep apnea (OSA) based on functional near infrared spectroscopy (fNIRS). Methods Forty-five subjects including 22 with OSA and 23 healthy as the control were recruited for assessment of a 3-minute resting-state prefrontal cortex (PFC) activity with the fNIRS technique. Only the oxygenated hemoglobin(HbO2) signal was used to calculate resting-state functional connectivity (RSFC) and construct brain connection network. To better describe prefrontal connectivity, we also divided it into four types of connectivity (Short-1: intra-hemispheric connection within Regions of interest (ROIs); Short-2: intra-hemispheric connection between ROIs; Long-1: inter-hemispheric connection witnin ROIs; Long-2: inter-hemispheric connection between ROIs). We extracted the relevant indicators of brain network connection and the complex network based on graph theory, and then analyzed the correlation between the indicators and cognitive scale. Results Compared with the healthy control group, patients with OSA showed more chaotic connection patterns, weaker intra- and inter-hemispheric connection intensity, and lower integration efficient. Specifically, patients with OSA exerted significantly lower connection intensity and the total network edge numbers in all four connection types. Furthermore, the average correlation coefficient and global efficiency of brain network were positively correlated with Montreal Cognitive Assessment score. Conclusions The fNIRS-based brain functional connection and complex network topology properties in patients with OSA were significantly different from those in healthy subjects. Such differences based on the fNIRS technique could be considered as potential biomarkers for OSA diagnosis, assessment, and intervention.