Schulze-Rauschenbach et al (Reference Schulze-Rauschenbach, Harms and Schlaepfer2005) compared repetitive transcranial magnetic stimulation (rTMS) and unilateral electroconvulsive therapy (ECT) and reported a similar treatment response rate. The rTMS methodology produced an impressive improvement with no cognitive side-effects.

However, the reported similar treatment effect with ECT could be misleading, as it is partly due to the rather low success rate of ECT in this study. The Hamilton Rating Scale for Depression (HRSD) score in the ECT group was reduced by a modest 35%. For comparison, the non-psychotic patients in the largest recent ECT study (the CORE study; Reference Petrides, Fink and HusainPetrides et al, 2001) achieved a 74.5% reduction on the HRSD–24 (24-item version).

We started an audit of ECT at our regional psychiatric hospital 1 year ago. So far 23 consecutive patients with treatment-resistant depression, who had an HRSD–17 (17-item version) score of 15 or above (the cut-off used by Schulze-Rauschenbach et al), have completed at least six ECT sessions. We observed a 55% improvement on the HRSD–17: from 24.6 to 11.0 points. The decrease on the self-rated Beck Depression Inventory was 20.1 points (an improvement of 49.9%). This compares with a decrease of only 7.6 points (24%) in the ECT group of Schulze-Rauschenbach et al. Even more importantly, the remission rate in their study was very low. Using the remission criterion of ≤7 points on the HRSD–17 (Reference ThaseThase, 2003), only one of their 13 ECT patients (8%) achieved remission (as shown in Fig. 1). This contrasts with a rate of 43.5% (10 out of 23 patients) in our study and 74.7% (189 out of 253 patients) in the CORE study. Four of our patients scored 0 or 1 point at the end of treatment.

There could be at least two reasons for the low response rate in the ECT group of Schulze-Rauschenbach et al. First, unilateral ECT is less effective than bilateral ECT, and when used at a simulation intensity of 100–150% above seizure threshold, it has produced only a 30% response rate (Reference Sackeim, Prudic and DevanandSackeim et al, 2000). Only four patients in our series and none in the CORE study had unilateral ECT. Second, patients with psychotic depression respond better to ECT (Reference Petrides, Fink and HusainPetrides et al, 2001). None of the patients of Schulze-Rauschenbach et al had psychotic symptoms, but 13 (56.5%) in our group and 77 (30.4%) in the CORE study did. This cannot explain all the difference, as the non-psychotic patients in our group still showed an improvement of 48% on both HRSD–17 and Beck Depression Inventory scores.

Properly administered bilateral ECT still remains by far the most effective treatment for severe depression.