To the Editor: Ulrike Feske, Ph.D., et al. were too negativistic in their article (1) when they stated, "our findings suggesting that borderline personality disorder patients may not respond adequately to ECT have potentially significant implications for the selection of candidates for ECT" (p. 2079). Although this is unarguable if taken literally, it sounds overly discouraging about the potential use of ECT in this situation.

First of all, there was an acute improvement in over 20% of depressed borderline personality disorder patients, most of whom had been nonresponsive to antidepressant medication. If this were generalizable, then it would amount to a large absolute number of potentially treatable individuals.

Second, in clinical practice, one often sees such patients having been managed with sequential medication trials and polypharmacy lasting over many years. ECT is not, of course, a substitute for the most important factor in treatment, which is a consistent, supportive, and skilled therapist. But if this most powerful biomedical treatment for depression fails, then the failure may help provide critical guidance to future therapy by discouraging the ongoing pursuit of an ultimate biological "magic bullet" that might yet make the patient feel better. To resolve such an issue, in selected cases, may give the therapist and patient more freedom to focus on other problems and may be well worth the effort and expense of ECT.