TO THE EDITOR: Benoit H. Mulsant, M.D., and colleagues (1) bemoan the fact that in their study almost half of the older patients with psychotic depression received inadequate or no antidepressant or neuroleptic therapy before being referred for ECT. Many of these same authors already reported in an overlapping study (2) that a substantial percentage of these medication-resistant patients subsequently responded to ECT (but that those who received ECT after not responding to antidepressants were less likely to respond than those who had not first been given antidepressants). Thus, the question arises: why give drug therapy at all to patients with psychotic depression who are willing to accept ECT?

The authors cited a study in which there was a 100% response rate to a tricyclic and higher-dose neuroleptic combination in patients with psychotic depression (3). This evidence is unconvincing, since the study was uncontrolled and retrospective. Indeed, as far as we know, no prospective, controlled comparison of ECT and pharmacotherapy in the treatment of depression exists.

Until the results of such controlled studies are available, there are ample clinical grounds for offering ECT as the initial low-risk, high-efficacy treatment of choice to older patients who suffer from psychotic major depression (4), rather than subjecting them to the often intolerable side effects (including potentially permanent tardive dyskinesia) of many weeks or months of an antidepressant and high-dose neuroleptic combination. I believe that most of the authors of the article privately agree with this point of view. What puzzles me is why they didn't say so publicly.