Functional connectivity and network analysis during hypoactive delirium and recovery from anesthesia

OBJECTIVE: To gain insight in the underlying mechanism of reduced levels of consciousness due to hypoactive delirium versus recovery from anesthesia, we studied functional connectivity and network topology using electroencephalography (EEG). METHODS: EEG recordings were performed in age and sex-matched patients with hypoactive delirium (n=18), patients recovering from anesthesia (n=20), and non-delirious control patients (n=20), all after cardiac surgery. Functional and directed connectivity were studied with phase lag index and directed phase transfer entropy. Network topology was characterized using the minimum spanning tree (MST). A random forest classifier was calculated based on all measures to obtain discriminative ability between the three groups. RESULTS: Non-delirious control subjects showed a back-to-front information flow, which was lost during hypoactive delirium (p=0.01) and recovery from anesthesia (p