Wednesday, May 5, 2010

Transcranial Magnetic Stimulation: Does it Live Up to the Hype?

Repetitive Transcranial Magnetic Stimulation (rTMS) is a treatment for depression that was approved by the FDA in October of 2008 (1). Repetitive TMS involves a device (pictured right), which is noninvasive, that excites the neurons in the brain. When this done over the left dorsolateral prefrontal cortex (an area of the brain supposedly less active in depressed patients), brain activity increases. The major selling point is that it has very few side-effects compared to standard antidepressant treatment (most common effects are headache and tingling at the stimulation site).

The FDA approval of this device has been controversial (2). The initial study submitted to the FDA was rejected. The folks at Neurostar (the manufacturers of the device) did a post-hoc analysis of that data. They discovered that patients, who failed to respond to only 1 antidepressant, subsequently responded to rTMS greater than sham (27.3% versus 10.5%). Based on this analysis, the FDA approved rTMS for the treatment of MDD in patients who have failed only 1 antidepressant trial.

In this month's Archives of General Psychiatry, is an article titled "Daily Left Prefrontal Transcranial Magnetic Stimulation Therapy for Major Depressive Disorder" (3). This study was funded by the NIMH and is the first nonindustry funded multisite study of rTMS (though some of the researchers are paid consultants of the TMS manufacturer). It involved 190 people.

What supposedly separates this study from all others, is the sham treatment. One major criticism of the previous TMS research is that the sham treatment was not convincing enough to prevent unblinding (for example, sham did not cause scalp irritation or facial twitching). The researchers went to great lengths to develop a sham treatment which produced the similar physical sensations of rTMS to prevent unblinding.
(Click to Enlarge)

Unfortunately, approximately 50% of the active treatment group correctly guessed which treatment condition they were in. A full 66% of placebo participants correctly guessed their condition. In truth, the level of unblinding is not a whole lot different from standard antidepressant drug trials (since placebos are inactive). The patients, on average, were similar to the patients in the Neurostar post-hoc analysis. The participants failed 1.51 antidepressant trials. 
The primary outcome was remission, defined as a score of 3 or less on the HAM-D or 2 consecutive HAM-D scores less than 10 during phase 1 of the study. Phase 1 was three weeks in duration. Patients received rTMS once a day for 50 minutes (5 days a week).
(Click to Enlarge)
Unfortunately, the results were negative. During the three week period (the right side of the chart) only 6 patients (11%) met criteria for remission. The average drop in HAM-D score for active treatment was only 5 points (26 to 21). The researchers then extended the the length of phase 1 by two weeks. The number of patients who went into remission during this extension phase was 13 (14%). By increasing the length of the phase 1, they obtained statical significance. Sounds fishy to me, but at least they provide all the data.
Similar to the Neurostar analysis, those who did remit were less treatment resistance (i.e., failed only 1 antidepressant trial). The number needed to treat (NNT) was 12. That means, 12 people will need to be treated with rTMS before another person, who otherwise would have not remitted without intervention, finally does remit. That's not very good. However, that number is not far off from standard antidepressant drug trials.
Does rTMS have any practical value as a future treatment for depression? Based on these results, one will need to attend a 50 minutes session everyday (excluding weekends) for 3-5 weeks to see some sort of result. That is in stark contrast to attending psychotherapy 1-2 times a week or visiting a psychiatrist once every 4-6 weeks. As Daniel Carlat points out in his monthly report (1), each treatment session would cost approximately $400. Insurance companies do not currently cover this treatment (and probably never will). Moreover, the group of patients who did remit (i.e., those who failed only 1 antidepressant trial) is not very marketable. Odds are they will try a second antidepressant instead. According to the Star-D results, the odds of improvement are 30% on a second antidepressant compared to 14% of rTMS. Presently, rTMS just does not make economic sense.

George MS, Lisanby SH, Avery D, McDonald WM, Durkalski V, Pavlicova M, Anderson B, Nahas Z, Bulow P, Zarkowski P, Holtzheimer PE 3rd, Schwartz T, & Sackeim HA (2010). Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial. Archives of general psychiatry, 67 (5), 507-16 PMID: 20439832


Dr Shock MD PhD said...

Thanks for this critical discussion of this article. Fully agree with your sepsis about this "novel" treatment and it's efficacy. Especially the cost is another important feature for the effectiveness of this treatment, take care Dr Shock

Bernard Carroll said...

There are additional problems with this report. They reported 18 remitters, of whom 15 came from just 2 of the 4 sites. People are going to be dubious about a treatment that doesn’t seem to work in half the sites. ECT isn’t like that, for instance.

They identified 18 remitters and 19 responders. The ratio of remitters to responders is 0.95. That’s most unusual. The usual ratio of remitters to responders is around 0.65 This discrepancy most likely has to do with the choice of a cutoff score of 10 on the HAMD to define remission. A cutoff score of 7 is more usual. A lot of the patients identified as remitters would have been called just responders by more usual criteria.

Finally, as a matter of pragmatics in therapeutics, the statistics don't come out significant when performed on primary response rates, bypassing the fancy statistical footwork. For remission, 2-sided Fisher exact was marginal at p = 0.046 and Yates chi square was not significant. It doesn’t look to me that rTMS is ready for prime time in this population.

This impression is strengthened by seeing that the outcomes were not statistically significant in the completers or in those who were fully adherent. The reason ITT (Intent to Treat) analyses are required by the FDA is that analyses of completers bias the decision in favor of the treatment. It is paradoxical that the completer analysis came out worse than the ITT analysis.

Barney Carroll.

NeuroPsych said...

Your comments are always appreciated. You're are a veritable font of information.


Anonymous said...

So the upshot is that rTMS is a sham!!! I can believe that. Much like most of the anti-depressant medication effect. What's going on with biological psychiatry, as to the treatment of depression? What have we gained?

Anonymous said...

Anon: It's not snake oil, but this particular regimen doesn't seem to impress. Still, cheaper than the $500m-$1b Merck will spend on its next failed small molecule antidepressant.

Anonymous said...

SSRI's, SNRI's, second generation antipsychotics, rTMS-- I mean WTF? Has any of this advanced the treatment of mood or psychotic disorders? Isn't the most effective treatment for melancholic depression still ECT? And the SSRI's still don't outperform imimpramine? Let alone placebo, in most cases? And isn't the best mood stabilizer still Lithium? 25 years of the biological psychiatry revolution have resulted in nothing more than BULLSHIT.

Anonymous said...

I was severely depressed and I've tried absolutely all anti-depressants available. Nothing worked decently.

Obviously, I also was very skeptical about TMS, but when you have a life disabling disorder, feeling hell-miserable every single day for no apparent reason, why not to try some new miracles the expensive health market is offering?

For my surprise, I discovered it is a very underrated treatment. Nothing else has made me feel so fine. I personally measure Cymbalta (best med I took so far) as a thousand times less intense than TMS as to "happiness degree".

After 24 years in plain misery, giving up everything, even considering well-being as an illusion, I received TMS treatment. 2 sessions later, my life changed, as idiot as it sounds, but I see no other way to describe such great, magnetically forged mood.

All I can think is these proving trials are too massive, there's no individual care.
Most possibly, they only stimulated the dorsolateral prefrontal cortex on everyone.
It's like giving up all anti-depressants after failing one (despite the fact I myself should've done so).
There are other brain areas each depressive subject must get tested before thinking the treatment is useless. Also, other frequencies and magnetic field intensities must be tested on each area, a difficult task when it comes to large trials.

Experiencing, say, the blissful effects of this machine, I believe it will replace anti-depressants soon, unless the psychiatric drug industry, which gained considerable power over the years, helps suppress this fantastic tool

Anonymous said...

TMS has been a Godsend treatment for my 24 year old daughter. She was unsuccesful for a year after several hospitalizations, taking various anti-psychotics, mood stabilizers and antidepressants that did not making any impact in elevating her mood. She continued to be extremely depressed,suicidal and had increased audible hallucinations. I researched TMS vs ECT or vagal stimulation and found that TMS was worthwhile at minimizing possible cognitive/memory side effects or possible permanent throat soreness. My daughter agreed to proceed and she underwent 5 daily treatments for 5 weeks. Yes, this was a sacrifice to get her daily to her treatments and after the 3rd week her depression lifted and her audible hallucinations began to decrease until they fully ceased after week 4. The psychiatirs recommended a total of 5 tapering sessions the following month and this has given my daughter a new desire to live again. She has begun on taking interest resuming her social activities and has signed up to start training for a running marathon for October of this year. TMS has definetely made an improvement in her mood. I would hope that the insurance carrier will consider my appeal in assisting with the financial aspect of this as the meds never helped her get out of the severe depression. TMS was extremely insturmental at saving my daughter's life.

Marc said...

In the open-label phase of treatment, phase II, the overall remission rate was 30%; which compares to STAR-D (assuming remission is similarly defined).

Anonymous said...


Anonymous said...

I realize it has been a long time since this post was written. Curious if you have done anymore recent research. I have had rTMS, and it has saved my life. I have tried 100s of medications and combinations of medications, and nothing worked. I have Bipolar II, with depression being about all I ever felt since I was 12 - with a few hypo-mania moments sprinkled in for good measure. I had suffered this debilitation for over 29 years. rTMS took away the daily suicidal ideation, near the 15 or so treatment. By the end of the fourth week, I could actually feel. I wasn't sad, hopeless, dead inside. I highly recommend rTMS to everyone. I have also, and will continue too fight to get it recognized as a legitimate medical treatment for depression (not experimental), so that government and insurers will cover it. Fine it's not perfect, but it is a viable option that should be explored before ECT. The insignificant side effects(I have had none)are not relevant enough to impair anyone, should it sadly not work for them. I have experienced every side effect possible on almost every medication I have tried. Many medications have a way greater negative impact. I hope you are able to present any updates you find. PS Please note many, if not all, of the early rTMS studies were only conducted for a duration of one to two weeks, it takes a minimum of 4 weeks to see the results. Also it only takes 15 to 20 minutes for a treatment, and you walk out and continue with your day unimpeded. (The first treatment is longer as they have to measure, plot the location, and find your threshold, but subsequent treatments are shorter.) The sacrifice for that 4 weeks is insignificant, especially for the gain. There have also been great advancements in the understanding of positioning and required intensity of the treatment. Albeit I would like to see some continuity in the studies, and clinical trials to enable a better correlation of the results. I would also like to see some standardization, of the treatments presently available from doctors outside of the investigative studies. Well that is my two cents. Not a doctor or professional in any sense of the word, but hopefully my experience will enlighten...

Anonymous said...

That helps a lot to hear. I have been trying to find personal stories not related to the manufacturers or the doctors administering this treatment and haven't found much. My insurance now covers this treatment and I am considering having it done because after 17 years and trials of many drugs, I still feel blah. I have to admit I am worried that it wont work or that it might turn me into a vegetable but I feel at this point that I have not much to lose as far a quality of life.