UC Berkeley Press Release


CHICAGO Over the past five years, a series of clinical studies—many of them small, uncontrolled, and nonblinded—have documented the possible benefits of repetitive transcranial magnetic stimulation (rTMS) for patients with major depression. Part of the appeal of rTMS has been its potential role as a kinder, gentler alternative to electroconvulsive therapy (ECT); the newer modality does not impair memory or require the induction of seizures, and it has a substantial public relations advantage in that its name does not conjure up images of Jack Nicholson films. What has been lacking, however, is strong evidence that its antidepressive effects are as substantial as those of ECT—or, many experts say, that it works at all.

Some tentative steps toward filling this void were taken earlier this year, with the publication of the first two randomized clinical trials to directly compare the two treatments. Although ECT did come out ahead on some measures in both studies, the evidence did not consistently show that the more established technique is superior to rTMS. Now, early data from a third randomized trial, still in progress, is providing further suggestions that any differences between the efficacy of ECT and that of rTMS may be relatively minor.

"We are pleasantly surprised by our preliminary results," Philip Janicak, MD, lead investigator of the ongoing study, told Neuropsychiatry Reviews. "I think it's premature to say whether or not rTMS is efficacious for depression, but I think so far the weight of the evidence would fall to that side."

Eric Wassermann, MD, Chief of the Brain Stimulation Unit at the National Institute of Neurological Disorders and Stroke, agrees that at present it would be inappropriate to declare rTMS to be as effective as ECT, which remains the gold standard for treatment of severe depression. The success of some of the early rTMS studies, he speculated, may have been due in part to the inclusion of patients whose symptoms were not as severe as those of the typical ECT patient; ongoing rTMS studies at the National Institute of Mental Health and Columbia University "are looking at refractory populations, and we're getting nothing" of clinical significance.

Moreover, he suggested that the idea of head-to-head studies comparing rTMS and ECT is somewhat suspect to begin with. "I think the only reason people began to compare rTMS to ECT is that both plug into the wall," he said. The two approaches presumably work through "different mechanisms" and are "not really related except in some possibly superficial way." A more appropriate clinical comparison, he believes, would be between rTMS and an antidepressant.


However, other researchers see the rTMS–ECT comparison as more valid, even if they, too, approached the issue of the new modality's efficacy with a degree of skepticism. Dr. Janicak, who is Professor of Psychiatry at the University of Illinois at Chicago , acknowledged that he did not expect rTMS to match ECT's antidepressant prowess in his own study. But an analysis of the first 22 patients treated (44 will eventually participate) has shown no statistically significant differences between the two therapies.

In the study, patients who meet DSM-IV criteria for unipolar or bipolar depression are randomized to receive four weeks of treatment with either bitemporal ECT (three times a week) or rTMS (five times a week). All patients are referred by their treating physician, in most cases because of inadequate response to antidepressants; a few have been referred because they could not tolerate medication or because the severity of their symptoms required a fast-acting approach. Depressive symptoms are assessed weekly using the Hamilton Depression Rating Scale (Ham-D); subjects who have a 50% or greater decrease in Ham-D score and a final score of 8 or less are considered responders. (These criteria, of course, are more stringent than the norm; in most studies, a 50% reduction in Ham-D score is sufficient for responder status.) After four weeks of treatment—12 sessions of ECT or 20 of rTMS—nonresponders are offered the opportunity to cross over to the other treatment.

Thus far, both therapies have reduced mean Ham-D scores by about 65%—from the low 30s to nearly 10. Moreover, about two thirds of subjects in each group meet both treatment response criteria. Although the study lacks sham controls, the fact that most patients are severely ill and have failed one or more courses of pharmacotherapy before entering the study suggests that the positive responses are not due to nonspecific treatment effects, according to Dr. Janicak.

Onset of action has been similar in both groups, with patients typically meeting criteria for treatment response in two to three weeks. "So I don't think our data so far indicate that you get a more rapid response with rTMS," Dr. Janicak said. As with ECT, response time to rTMS varies; some patients in the study have recovered after only a few treatments, whereas others required the full four weeks.


In contrast to the Chicago study, findings from the first randomized trial to compare ECT and rTMS—published in the February issue of Biological Psychiatry —seemed to favor ECT, at least for patients with psychotic depression. However, a discrepancy in the treatment protocol between the two groups may account for the difference, according to some experts.

The study, led by Leon Grunhaus, MD, Chair of the Department of Psychiatry at Sheba Medical Center in Ramat Gan , Israel , looked at the effects of ECT and rTMS in 40 patients with major depression. ECT was initiated with right unilateral electrode placement, but bilateral therapy was begun if no treatment response was evident after six sessions. Subjects received up to 20 days of the assigned treatment.

By the end of the trial, patients without psychotic symptoms in both treatment groups had experienced a mean decrease of 11 to 12 Ham-D points. In patients with psychotic depression, however, ECT reduced Ham-D scores by an average of 23 points, whereas patients in the rTMS group had only an 8-point decrease, a statistically significant—and, obviously, clinically significant—difference.

The problem, Dr. Janicak noted, is that as per local standard-of-care practices, patients in the ECT group who had psychotic symptoms were also eligible to receive appropriate medication during their course of treatment; thus, four patients received neuroleptics, six received neuroleptics plus antidepressants, and five were given antidepressants alone. However, subjects randomized to rTMS received no medication other than clonazepam (1 to 2 mg/d) to decrease anxiety and reduce the risk of seizures during stimulation. Thus, Dr. Janicak stated, the Israeli researchers should have concluded that for psychotic patients the combination of ECT and medication was more effective than rTMS alone.

The Chicago study, on the other hand, has kept concurrent medication use to a minimum in both groups, typically just benzodiazepines or sedative hypnotics to reduce anxiety and agitation. This may account for the fact that "we're not seeing a difference between rTMS and ECT regardless of whether psychosis is present," Dr. Janicak reported.


A third research team, this one based in the Southern Hemisphere, has also compared the efficacy of rTMS and ECT, and here again the published findings slightly favored the latter modality. In a study reported earlier this year in the International Journal of Neuropsychopharmacology, Saxby Pridmore, MD, Director of Psychological Medicine at the Royal Hobart Hospital in Hobart, Australia, reported that ECT patients had a greater mean reduction in Beck Depression Inventory scores than did rTMS patients (69% vs 46%); this represented, on average, a 7.5 point difference in the magnitude of improvement between the two treatments. ECT recipients also had greater improvement when symptoms were measured via the Ham-D or a visual analogue scale, although neither of these differences reached statistical significance. The rate of remission—when defined as having a final Ham-D score of 8 or below—was identical in both groups (69%).

In addition to its use of blind raters, one of the study's notable features was its use of a flexible regimen: subjects continued receiving rTMS or ECT until the patients achieved a clinical response or plateaued. On average, rTMS remitters received 13.1 treatments, a finding consistent with the Chicago study.

Overall, "this study suggests that rTMS can achieve results approaching ECT, if stimulation is at 100% [of motor threshold] and the length of the course can be flexible, with no upper limit to the number of treatments," Dr. Saxby and colleagues concluded.


Regardless of whether rTMS and ECT prove equally efficacious, the newcomer's superior side effect profile may make it the more desirable approach for some patients. In Dr. Janicak's trial, the most common adverse effect of rTMS has been transient headache, probably due to stimulation of nearby muscles (leading to a tension-type headache); other patients have experienced discomfort from the heat of the magnetic coil. No cases of inadvertent seizure have occurred; Dr. Janicak estimates that the incidence of such seizures should be less than 1% if appropriate safety guidelines are followed. None of the rTMS studies thus far have noted any cases of posttreatment memory impairment, a familiar problem with ECT.

Another advantage that may eventually favor rTMS is its cost. Dr. Janicak estimates the cost of a typical course of ECT as about $10,000, roughly six times the price tag for a comparable course of rTMS. "So if rTMS helps, it would have tremendous cost/benefit value as well," he said.

Of course, despite these favorable attributes, rTMS remains an experimental therapy, awaiting more definitive evidence of its efficacy. Most of the trials to date that have compared rTMS to sham stimulation have found some benefits to treatment, but, with the exception of another Israeli study, none had more than 20 patients. For example, Robert M. Berman, MD, and colleagues at Yale University School of Medicine reported earlier this year that a two-week course of rTMS reduced Ham-D scores by an average of 14 points in 10 patients with treatment-resistant depression; sham "treatment" resulted in no improvement. Only one rTMS patient had an improvement that was both robust (a 40-point decrease in Ham-D score) and long-lasting, however, leading the researchers to term the benefits of rTMS as only "clinically modest."

Some researchers find such studies less than persuasive. One problem, many argue, is that current sham controls are not sufficiently convincing, so that patients may be aware of which treatment they are receiving. Yet another reason for skepticism is that the theoretical foundation for how rTMS works is shaky at best. "The idea that you can produce lasting alterations in cortical activity with stimulation at 10 or 20 Hz is borne out nowhere," Dr. Wassermann said, and the assumption that a predominantly focal treatment like rTMS can cure depression—which has neurobiologic manifestations in many brain regions—is similarly suspect. "I don't think that the driving hypotheses are strong enough that when I see positive data, I think, 'Oh yes, this must be due to what we did in some understandable way.'"


Ultimately, the fate of rTMS lies in the hands of several large trials currently being planned. For example, David Avery, MD, Professor of Psychiatry and Behavioral Sciences at the University of Washington School of Medicine, in Seattle , will be conducting a sham-controlled trial expected to include 86 medication-free depressed outpatients—a sample large enough not only to determine overall efficacy but perhaps to help identify patient subgroups that might be best suited for rTMS. To ensure the reliability of any observed treatment effects, the clinical profiles of the two groups—from medication resistance to episode length—will be carefully matched.

The Chicago group is also preparing for what they hope will be a more definitive rTMS study. It will examine the acute effects of treatment, as prior studies have; but it will also compare the relative efficacy of rTMS versus medication as maintenance therapies for depression.

If these and other studies prove rTMS effective, the therapy could find a niche in treatment algorithms for depression, perhaps as the next stage for patients who have failed one or more antidepressant trials, Dr. Avery suggested. "Going on to ECT is a very big step, so rTMS could be an intermediate step."

Indeed, if only a third of depressed individuals who would normally receive ECT turn out to benefit from rTMS, Dr. Janicak noted, "that would still be a significant advantage in our ability to manage severe depression" and would "avoid many of the complications associated with ECT." Thus, in the event that one of the two modalities does prove to be somewhat more effective than the other, the real winner, ultimately, will be severely depressed patients.

—Peter Doskoch

Suggested Reading
1. Berman RM, Narasimhan M, Sanacora G, et al. A randomized clinical trial of repetitive transcranial magnetic stimulation in the treatment of major depression. Biol Psychiatry. 2000;47:332-337.
2. Grunhaus L, Dannon PN, Schreiber S, et al. Repetitive transcranial magnetic stimulation is as effective as electroconvulsive therapy in the treatment of nondelusional major depressive disorder: an open study. Biol Psychiatry. 2000;47:314-324.
3. Martis B, Janicak PG. Transcranial magnetic stimulation for major depression: therapeutic possibilities. International Drug Therapy Newsletter. July, 2000:1-10.
4. Post RM, Kimbrell TA, McCann UD, et al. Repetitive transcranial magnetic stimulation as a neuropsychiatric tool: present status and future potential. J ECT. 1999;15:39-59.
5. Pridmore S, Bruno R, Turnier-Shea Y, et al. Comparison of unlimited numbers of rapid transcranial magnetic stimulation (rTMS) and ECT treatment sessions in major depressive episode. Int J Neuropsychopharmacol. 2000;3:129-134.
6. Pridmore S, Belmaker R. Transcranial magnetic stimulation in the treatment of psychiatric disorders. Psychiatry Clin Neurosci. 1999;53:541-548.