BACKGROUND Cognitive disorders occurring after electroconvulsive therapy (ECT) are regarded as an expression of brain damage, despite computed tomography (CT) and magnetic resonance imaging (MRI) showing no signs of structural brain damage. Serum neuron-specific enolase (NSE) is a sensitive marker of neuronal damage (i.e., after stroke or cardiac arrest). The objective of this study was to investigate whether ECT leads to a rise in the serum NSE level as an expression of neuronal damage. METHODS We investigated seven patients (four women, three men; mean age 6212 years) with major depressive disorder, who were treated with ECT for the first time. ECT was administered every 2 days, three times a week under standard conditions (anaesthesia: thiopental, succinylcholine, 100% oxygen, unilateral ECT, seizure duration more than 20 s). Blood samples were drawn at the following times. For the first ECT: 15 and 1 min before ECT, and 1, 5, 10, 15, 20, 25, 30, 45, 60, 75, 90, 105, 120 min, and 8, 12, 24 h after ECT. For all subsequent ECT: 1 min before and 4 h after every ECT. Serum NSE was measured by means of enzyme immunoassay (Cobas Core NSE, EIA, Hoffmann-La Roche). RESULTS On average, each patient underwent ECT 10 times (range 5-20). In the first ECT there was no difference in serum NSE levels before and at all times following ECT. A comparison of serum NSE levels before and after each subsequent bout of ECT revealed no differences. Moreover, comparing the baseline serum NSE levels (before the first ECT) with the values after final ECT showed no differences either. CONCLUSIONS ECT did not increase serum NSE values, indicating that electroconvulsive therapy does not cause neuronal damage.