When to consider Deep TMS: how I decide with my patients

Transcranial magnetic stimulation has been FDA-cleared for depression since 2008, and the Deep TMS system we use at Stedfast Psychiatry since 2013, yet most patients who would qualify for it have never been offered it. I oversee the Deep TMS program at our Palm Beach Gardens center, and I want to lay out how I actually decide, with a patient sitting across from me, whether TMS belongs in their plan.

What Deep TMS is

TMS uses a magnetic coil placed against the scalp to induce small electrical currents in the brain. Repeated pulses over the left dorsolateral prefrontal cortex, a region that is underactive in many people with depression, change the activity of that circuit over a course of treatment. No anesthesia, no sedation, no memory effects. Patients drive themselves to and from the appointment.

Deep TMS refers to the BrainsWay H-coil, which the FDA cleared for major depressive disorder in 2013. Compared with the figure-8 coil used in standard TMS, the H-coil stimulates a deeper and broader region of the prefrontal cortex. The same device has since been FDA-cleared for obsessive-compulsive disorder (2018), smoking cessation (2020) and depression with comorbid anxiety (2021). In our practice the depression indication accounts for most of the treatments we deliver.

The evidence I quote to patients comes from the pivotal trial published by Yechiel Levkovitz and colleagues in World Psychiatry in 2015, a multicenter randomized sham-controlled study of 212 patients with medication-resistant depression. After the acute phase, remission rates were roughly double in the active group compared with sham.

Where TMS sits in the sequence

The FDA label is straightforward: adults with major depressive disorder who have not had a satisfactory response to antidepressant medication in the current episode. In practice, insurers usually require documented failure of several adequate medication trials before they will cover it, so the real-world entry point is somewhere around the second or third failed antidepressant.

That happens to be where the evidence says the medication route starts to run out of road. The STAR*D trial, reported by A. John Rush and colleagues in the American Journal of Psychiatry in 2006, showed remission rates falling with each successive medication step, from about a third on the first trial to roughly 13 percent on the third and fourth. When a patient has had two adequate trials and is still depressed, the odds that the next pill will produce remission are modest, and I say so plainly. That is the conversation in which I raise Deep TMS, not as a last resort but as an option with a different mechanism and a different risk profile.

The questions I ask

Before I recommend TMS to anyone I work through a specific set of questions.

• Is the diagnosis right? TMS treats major depressive disorder. If what looks like resistant depression is actually bipolar depression, a substance-related mood disorder, untreated sleep apnea or a medical illness, TMS is the wrong tool. I re-examine the diagnosis before every TMS referral, including my own patients.
• Were the medication trials adequate? Two weeks at a starting dose is not a trial. I want to see at least six to eight weeks at a therapeutic dose, and I want to understand why each one stopped.
• Is there anything that makes TMS unsafe? Metal implants in or near the head, other than dental work, rule it out. Cochlear implants, aneurysm clips and some implanted stimulators do too. A history of seizures does not automatically exclude a patient but changes the risk discussion.
• Can the patient make the schedule? The standard Deep TMS course is 36 sessions: 20 daily sessions, five days a week for four weeks, followed by 16 continuation sessions twice a week over the next 12 weeks. Each session lasts about 20 minutes in the chair. A patient who cannot get to Palm Beach Gardens every weekday for a month will not get the benefit, and I would rather know that before we start.
• What else is in the plan? TMS is usually added to medication, not substituted for it, and I keep patients on their current regimen through the course unless there is a reason not to. Psychotherapy continues as well.

What I tell patients about side effects

The common ones are scalp discomfort at the stimulation site and headache, both of which usually fade over the first week as patients adapt. The serious one is seizure, which in the published literature occurs in well under one tenth of one percent of patients and has been rare in our own program. We screen for it, we follow the FDA-cleared protocol, and a trained technician is in the room for every session. I do not minimize the risk and I do not inflate it.

Patients sometimes ask about cognitive effects because they have heard of electroconvulsive therapy. TMS is not ECT. There is no anesthesia, no induced seizure and no memory loss. People return to work after their session.

How I judge whether it is working

We take a PHQ-9 at baseline and weekly through the acute phase. I look for movement by the third week. Some patients respond early; a meaningful number respond late, in the fifth or sixth week, which is one reason I ask people to commit to the full acute course rather than judge it after ten sessions. Response is a 50 percent drop in the score. Remission is a PHQ-9 of 5 or below plus, in my practice, a return to the functional goals we set at the start.

If there is no response at all by the end of the acute phase, we stop and reconsider. If there is a partial response, the continuation phase often builds on it. For patients who remit and later relapse, a shorter retreatment course often works, and the literature supports that approach.

Who I steer away from TMS

Patients who are acutely suicidal need faster treatment than a six-week TMS course can provide, and for them I talk about Spravato, which the FDA approved for depression with acute suicidal thinking, or hospital-based care. Patients with psychotic depression generally do better with ECT. Patients whose depression is mild, or who have not yet had a proper medication trial, should try medication and therapy first. And patients who tell me honestly that they cannot make 20 weekday visits should not start a course they will not finish.

The decision, in the end

TMS is not a miracle and it is not a gimmick. It is an FDA-cleared treatment with a reasonable evidence base, a favorable side effect profile and a demanding schedule, and it works for a substantial fraction of the patients who have run out of good medication options. My job is to put the numbers in front of a patient, check the practical obstacles, and make sure the diagnosis is right. Having done that, I let the patient decide, and I have found that patients who understand the trade-offs make good decisions.

If you have had two antidepressant trials without remission and no one has mentioned TMS to you, it is a reasonable question to ask.

More from Dr. Husted

• Why I measure depression treatment by function, not just mood2026-06-11
• Spravato in practice: what I tell patients before their first session2026-08-06
• Anxiety that does not respond to the first medication: my next steps2026-08-27

Educational content and not a substitute for evaluation. If you are in crisis, call or text 988.