Objective: The aim of this study is to retrospectively evaluate the influence of diagnosis and three different anesthetic agents on seizure durations, hemodynamic parameters, recovery times, and side effects in electroconvulsive therapy (ECT). Method: We have retrospectively evaluated 1342 ECT sessions conducted on 179 patients under general anesthesia at the Psychiatry Department of Dicle University Hospital between 1 September 2009 and 30 June 2012. The relationship between the indications for ECT and the choice of anesthetic agent, number of ECT sessions, motor and EEG seizure durations, recovery times from anesthesia, hemodynamic changes and complications were analysed. Results: The mean age of the 179 patients included in the study was 36.7 +/- 7.3 years. In terms of sex, 50.8% (n=91) were male and 49.2% (n=88) were female. The most commonly used anesthetic agent was propofol (73.2%). Among the patients, 50.8% was diagnosed with major depression, 28.5% had bipolar disorder and 12.8% was diagnosed with schizophrenia. A total of 179 (13.3%) out of the 1342 ECT sessions were considered as failed. In 92.2% of the failed ECTs, the anesthetic agent was propofol. No statistically significant relationship was observed between the diagnosis forming the basis for the ECT and the motor and EEG seizure times, beginning of spontaneous breathing, eye opening, time until response to verbal stimuli, changes in hemodynamic parameters and complications. The seizure and recovery times were significantly shorter in the propofol group compared to the ketamine and etomidate groups (p < 0.05). The EEG seizure durations were longer in the etomidate group in comparison to the propofol and ketamine groups (p < 0.001 and p < 0.05, respectively). Conclusion: The diagnoses leading to the ECT have no influence on the motor and EEG seizure times, beginning of spontaneous breathing, eye opening, times of response to verbal stimuli, hemodynamic parameters or complications. However, the anesthetic agents used influence all of these parameters.