Dr. O’Reardon, the FDA recently approved TMS for  depression. Can you clarify exactly what this approval was for?

They approved a particular device made by Neuronetics (a TMS device manufacturer) to deliver TMS to a certain patient population. And the approved patient population is adults with major depression who have failed one fully adequate anti depressant trial in this episode.

How did you personally get interested in TMS?

My interest began 10 years ago, when I was doing a fellowship in mood disorders. I was already doing ECT and was interested in treatment-resistant depression. I wanted to find something that would be better tolerated than ECT and my mentor at that time (the late Marty Szuba M.D.) was interested in TMS. We did an initial study that showed that single sessions of TMS, active versus sham, had more than just local effects on the brain. It also had effects on the neuroendocrine system. For example, we showed that TSH (thyroid stimulating hormone) rose in response to active TMS but did not change with sham TMS. This actually replicated a finding that was already established with ECT.

Can you talk a bit about the putative mechanism of TMS?

The TMS device is an electromagnet which sends out magnetic pulses at frequencies that range from 1 hertz (cycle per second) to 20 hertz. When a dynamic magnetic field reaches the surface of the brain, it induces an electric field. Thus, TMS allows us to induce electrical activity in the brain.

Which is ultimately what ECT does as well, right?

Yes, but the difference is that a magnetic field passes unimpeded to the surface of the brain, so that it can be targeted, whereas when we pass electricity in ECT, it immediately dissipates. We don’t have the option with ECT of focal brain stimulation. With TMS we have the option of basically targeting an area on the surface of the prefrontal cortex that is about 2 centimeters squared.

And how do you decide which area of the cortex to target?

We generally target the left dorsolateral prefrontal cortex. This came out of neuroimaging studies of patients with depression that were done in the early 90s, and the single most consistent finding was low metabolic activity in the left dorsolateral prefrontal cortex. The simplistic way of looking at this is that we believe that this area of the cortex is not working as hard as it should or is off-line in major depression, so we are using TMS to reactivate those neurons.

I’ve heard about high vs. low frequency TMS. What is the difference?

This is a fascinating aspect of TMS. Using low frequency TMS, we seem to primarily be stimulating GABAergic interneurons, so this tends to inhibit the prefrontal cortex. Now what would be the relevance of that to depression? It turns out that another way to do TMS successfully in major depression is to give inhibitory TMS to the opposite side of the cortex (the right side) to dampen what appears to be hyperactivity in the right prefrontal cortex (another imaging finding in major depression).

How does one actually do TMS?

You start by locating where the left dorsolateral prefrontal cortex is on an individual patient. To do that, we rest the magnet against the scalp at a point about a third of the way between the vertex of the head and the tip of the left ear. We give single pulses until we get a twitch in the thumb or index finger, and this tells us we are over the motor cortex. Then, we go forwards 5 centimeters from there and we will end up over the dorsolateral prefrontal cortex.

So now you’ve established where you are going to administer the pulses. What do you next?

You determine the motor threshold, which is the energy required to cause that twitch of the thumb. For most patients the motor threshold is somewhere between 50% and 65% of the energy of the device. To actually administer the treatment, instead of giving single pulses, you will be giving trains of TMS at usually 10 pulses per second or 10 hertz. The FDA approval is for 35 minute sessions, though benefits have been found for sessions ranging from 10 to 40 minutes.

So are you giving pulses continuously for 35 minutes?

No. There is what is called an on and off period. There is a train of TMS that lasts 4 seconds, meaning 40 pulses, and then there is an off period of 26 seconds where you are giving no pulses at all. If you stimulated continuously for a very long train you would have an increased risk of seizure.

What is the seizure risk?

In the trials that led to the FDA approval there were 10,000 sessions of TMS administered to 325 patients and the seizure rate was zero. So the risk is generally felt to be certainly less than 1 in 1,000 sessions, possibly as low as 1 in 10,000 sessions which translates to 1 in 500 patients treated with TMS. So you are looking at a risk that is less than what would be the case with Wellbutrin. Worldwide, after 13 years of TMS research, the total number of seizures is on the order of 15, and most of those seizures had been where the duration of stimulus was outside the safety guidelines. At our center at Penn, we have been doing TMS since “1997”, and we have not had any seizures to date, fortunately.

How do patients experience the procedure?

They are resting in a kind of armchair and the coil is placed on the scalp. They get about 5 seconds of stimulus every 30 seconds, and they hear the sound of the magnet pulsing, which is a kind of tapping noise. For the first few days they may feel scalp discomfort because 10 hertz is fairly intense and it will stimulate the muscle and the skin nerves as the pulses go through the scalp. After a few sessions they accommodate to that and it goes away. They can read. They can talk to you. When we had some college students in the study, some of them would be reading their novel or doing their reading homework while on the device.

How many sessions are there in total?

The FDA approval specifies five sessions per week for 20 to 30 sessions, followed by a six-session taper that lasts for three weeks. So an acute course, if you include the taper phase, is anything from seven to nine weeks.

Is TMS being reimbursed by insurance companies?

I would say it is too soon to expect much on that front. The FDA approval was in October 2008, and it will take six to 12 months before insurance companies develop policies to cover it. 

Do you think that the insurance companies will eventually cover it?

I believe it is likely. Certainly the manufacturer is optimistic that TMS will be reimbursable. The cost advantage of TMS is that it is done in an office rather than a hospital. You don’t need the procedure room and the recovery room. Patients have to observed throughout the session, but this can be done by a licensed professional like a nurse or a physician assistant. A psychiatrist needs to supervise the process, but I don’t think it is going to be cost effective for most psychiatrists to be doing the treatment themselves.

You can’t get your secretary to do it?

No. I did hear of one psychiatrist in Germany at one time who did do that, but that would definitely not be recommended. The person administering the treatment must have enough medical training to monitor for unexpected adverse effects, and to make sure that there are no indications that the patient is about to have a seizure, such as focal twitching in their hand.

Thank you for this brief primer on TMS. We’ll check in again in a couple of years and see whether the treatment has gained any traction among psychiatrists.