The spectral characteristics of the electroencephalogram (EEG) is the potential biomarkers of efficacy of Transcranial Magnetic Stimulation (TMS) in resistant depression

Abstract

Objectives. The purpose of the study was to find possible biomarkers of TMS effectiveness in resistant depression patients through comparative analysis of the spectral characteristics of EEG in responders (R) and nonresponders (NR) on TMS. Methods. The study included 32 patients with treatment-resistant depression with the severity of symptoms, according to the Hamilton scale (17 items) of at least 20 points (mean 24.7±3.6 points). TMS was performed on a magnetic stimulator “NeuroCMS” on the left dorsolateral prefrontal cortex, 12 sessions of 15 Hz, 100 % of the threshold intensity. To the end of TMS course 20 patients (> 50 % reduction of HAMCD total score) were assigned to the group of R, and the rest (n = 12) consisted of a group of NR. Results. Differences between EEG of R and NR revealed when comparing the value of the asymptotic-symmetry of alpha activity. Pronounced differences in the structure of the main rhythm in R and NR in comparative analysis of background EEG in the quantization to 1 Hz was found in the right occipital region. Theta activity of R not only had a lower representation in all areas of the cortex, but also did not show a significant breach of hemispheric balance both in the frontal areas of the cortex, and in the temporal areas. At NR more pronounced reduction of the alpha rhythm mainly in the amplification of slow-wave activity of delta and theta spectrum, as well as the high activity of the beta range. Discriminant analysis of obtained data show that differences in EEG patterns R and NR no longer differ in the analysis of individual indicators of EEG rhythms, and in their combinations, in particular ratios of the alpha rhythm and theta bands. Moreover, the greatest weight is related to the ratio of spectral power (SP) of alpha and theta rhythms in leads T3 and T5. The higher this ratio, the higher the probability of a hit in the responder group of patients and, therefore, predict favorable effect of TMS. Positive values (Y > 0) of discriminant scores are  efficiency biomarker of TMS treatment in resistant depression. Conclusion. Obtained formula (classification function) allows to predict the effect of TMS with specially-computed index of patient’s EEG. At the same time these data need to be clarified in future researches.