Objective: S100B is a glial cell protein with bimodal function. In low concentrations, it exerts neurotrophic effects, but higher levels reflect neuronal distress. Recent research suggests that this molecule may be a biomarker of response to electroconvulsive therapy (ECT). We examined the effect of ECT on serum S100B and its utility as 1) a biomarker of a depressive state and 2) a predictor of ECT response. We also wanted to ensure that ECT does not cause a marked serum S100B-elevation, indicating neural distress. Methods: We measured serum S100B in 22 in-patients treated with ECT due to depression. Depression severity was assessed using 17-item Hamilton Rating Scale for Depression (HAMD-17). The data were collected before an ECT series, within one week after the series (post-ECT), and at a six-month follow-up. Changes in serum S100B and clinical outcomes were tested using a linear mixed model. A relationship between serum S100B and the clinical outcomes was examined using Spearman and partial correlation. Results: Serum S100B did not change significantly immediately after an ECT series or six months later. The post-ECT serum S100B-change was not associated with the clinical effect (rho=.14, n=22, p=.54). The baseline serum S100B did not predict the clinical effect when controlling for age (r =.02, n=22, df=19, p=.92). Conclusion: The study neither supports serum S100B as a state marker of depression nor a predictor of ECT response. No evidence for ECT-related neural distress was found.