Subanesthetic ketamine alters EEG signal complexity: Implications for treatment stratification in depression.

Ketamine, a drug that blocks NMDA receptors and is thought to boost neuroplasticity, increased whole-brain EEG signal complexity during infusion in patients with major depressive disorder (MDD), most of whom had treatment-resistant depression. Higher complexity at the end of infusion predicted less symptom improvement the next day, while lower occipital complexity before treatment predicted a favorable response. A logistic model using occipital complexity had moderate predictive accuracy. No changes were seen in a different complexity measure, suggesting it is not useful as a biomarker. Occipital EEG complexity may help predict which patients will benefit from ketamine therapy.

Study at a glance

Abstract

Major depressive disorder, particularly its treatment-resistant form (TRD), poses significant treatment challenges. Ketamine, an N-methyl-d-aspartate receptor antagonist, has shown promise in rapidly alleviating depressive symptoms by influencing neuroplasticity and glutamatergic modulation, which are thought to influence brain activity complexity. In this placebo-controlled study, we examined the effects of subanesthetic doses of intravenous ketamine on EEG signal complexity in 24 MDD patients, 21 of whom had TRD. Treatment response was defined by a ≥ 33 % reduction in Montgomery-Åsberg Depression Rating Scale (MADRS) after ketamine administration. Patients underwent eyes-closed resting state EEG recording pre-, start-, end- and 24 h post-infusion, analyzed for temporospatial and spatiotemporal Lempel-Ziv complexity (LZCT and LZCS). Results indicated that ketamine significantly increased whole-brain LZCT during infusion compared to placebo (sodium chloride 0.9 %) (16.90 % vs. -4.84 %, 95 % CI 4.29 to 39.18, p = 0.017). Elevated LZCT at end-pre was associated with less short-term symptom improvement the following day. Conversely, lower pretreatment occipital LZCT (0.33 vs. 0.46, 95 % CI 0.007 to 0.26, p = 0.040) predicted a favorable response to ketamine, supported by a logistic regression model with an ROC area of 0.75. No significant changes were observed in LZCS, suggesting limited utility as a biomarker. In conclusion, occipital LZCT could serve as an effective predictive biomarker for ketamine's therapeutic effects in MDD, with implications for patients with TRD. This underscores the potential of EEG complexity measures in stratifying treatment and enhancing our understanding of the neural impacts of ketamine in depressive disorders.