The role of frontal EEG in predicting clinical response of major depressive disorder to intranasal ketamine and esketamine.

Frontal EEG recordings before treatment with ketamine or esketamine in 43 people with major depressive disorder revealed that those who later responded to the medication had increased functional connectivity (measured by phase locking value and phase lag index) and decreased entropy (Renyi and Tsallis) compared to non-responders. Aperiodic spectral parameters were lower in responders but did not predict response. These EEG measures showed moderate predictive accuracy, with area under the ROC curve values of 0.7065 for Renyi entropy, 0.7101 for Tsallis entropy, and 0.7283 for phase lag index, suggesting frontal EEG patterns may serve as biomarkers for identifying individuals likely to benefit from (es)ketamine treatment.

Study at a glance

Abstract

Ketamine and its enantiomer esketamine are NMDA receptor antagonists that have shown consistent antidepressant effects in major depression disorder (MDD). These effects suggest an active role of glutamate in the pathophysiology of depression. It has been hypothesized that acute increase of glutamate neurotransmission in the prefrontal cortex following treatment with (es)ketamine leads to synaptic plasticity and neurotrophic changes with following influence on theta activity of encephalography (EEG). The present retrospective study included a total of 43 patients affected by MDD and treated with either esketamine or ketamine. EEG recordings before treatment were related to clinical outcome as measured by both a self-rating and a rater-based psychometric scale. EEG measures were considered as possible predictive biomarkers for responsivity of MDD to (es)ketamine. In particular, we utilized phase locking value (PLV), phase lag index (PLI), Renyi and Tsallis entropy, as well as aperiodic spectral parameters (offset and decay exponent) by focusing on the frontal region and theta band. We found increased functional connectivity (PLI and PLV) and decreased entropy in responders to (es)ketamine compared to non-responders. Although, we did not find a predictive effect of aperiodic spectral parameters, we observed decreased values in responders compared to non-responders. Our results emphasize the role of frontal EEG as possible biomarker [area under the ROC curve of 0.7065 (Renyi entropy), 0.7101 (Tsallis entropy) and 0.7283 (PLI)] in predicting antidepressant effects of (es)ketamine.