TMS Safety and Seizure Risk: The Real Numbers

TMS is one of the better-tolerated treatments for depression, and its most serious acute risk, a seizure, is rare. The 2021 international (IFCN) safety guideline concludes that, given the number of people treated and how few seizures have been reported, the risk "is certainly very low." In a large survey of Clinical TMS Society members, 18 seizures occurred across 586,656 sessions in 25,526 patients, or 0.31 per 10,000 sessions and 0.71 per 1,000 patients.

"Very low" is not zero, and the risk is not the same for everyone. It is higher with certain medical conditions, with alcohol withdrawal and stimulant drugs, when safety limits are exceeded, early in a course, and with the deep TMS H-coil. This page sets out the numbers, where they come from, and how a TMS team screens for and manages risk.

Seizure risk: the published numbers

A common error in online copy turns "1 in 30,000 treatments" into "0.03%". The correct figure is about 0.003% per treatment; the per-patient figure is under 0.1%.

Why the figures differ

Each source measures something slightly different. Survey data depend on who answers: only 134 of roughly 500 Clinical TMS Society members responded to the Taylor survey. Per-session rates look smaller than per-patient rates because each patient receives dozens of sessions. And research settings include healthy volunteers and unusual protocols, while clinical surveys mostly reflect standard depression treatment.

The coil difference deserves a plain statement. In the Taylor survey, the relative risk was significantly higher with the H-coil, while the authors still described the absolute risk as low. The survey design cannot separate the coil from differences in the patients treated or the protocols used. The trade-offs between coils are covered in deep TMS vs standard TMS.

What raises seizure risk

Seizure risk rose substantially in people with brain lesions or epilepsy, and seizures were more common when published safety guidelines were exceeded. They were most likely to occur within the first few exposures to TMS.

The IFCN guideline lists medical factors that lower the seizure threshold:

• metabolic problems such as low sodium, calcium or magnesium, low or high blood glucose, and kidney or liver failure
• higher levels of proconvulsant drugs because of reduced clearance
• alcohol withdrawal, and use of stimulants such as cocaine or MDMA
• cyclosporine or tacrolimus, dialysis, systemic infection and fever
• sleep deprivation, stress and anxiety, and increased alcohol use

The single seizure in the pivotal deep TMS depression trial illustrates the point. It occurred near the end of a 26-year-old's ninth session and was attributed to heavy drinking the night before, which she had not disclosed, and possible alcohol withdrawal; she had no lasting effects.

Medications: two sources, two emphases

FDA's 2011 special-controls guidance asks device labeling to call for close monitoring of patients taking "tricyclic antidepressants, neuroleptic medications, or other drugs that are known to lower the seizure threshold." The 2021 IFCN guideline is softer: it says the low observed seizure rate "no longer support[s]" the earlier advice to be cautious with such medications, while still recommending that concurrent drugs be documented. IFCN notes that most rTMS patients take CNS-active medication and that only 3 of the seizures in the Lerner survey occurred in people taking drugs suspected of lowering the threshold.

In practice the two positions are compatible. A medication list is reviewed, the dose of any threshold-lowering drug is noted, and the combination of several risk factors matters more than any single drug. How medication interacts with TMS more broadly is covered in combining TMS with medication.

Screening before the first session

Most of TMS safety happens before the coil is switched on. A careful pre-treatment evaluation covers:

1. Seizure history and neurological illness. FDA guidance asks for close monitoring of people with a history of seizure or epilepsy, stroke, head injury, raised intracranial pressure, severe headaches or other neurologic disease.
2. Metal and implants. Conductive, ferromagnetic or other magnetic-sensitive metal implanted in the head or within 30 cm of the coil is a contraindication in FDA guidance, as are implants such as deep brain stimulators, cochlear implants and vagus nerve stimulators. Pacemakers and implantable defibrillators carry a labeled warning. IFCN is more permissive about some devices, but for patient care the device labeling governs. Practical questions about dental work, tattoos and piercings are answered in the Mind Body Seven guide to metal implants, dental work and tattoos.
3. Alcohol, sedative and stimulant use, including recent changes. Withdrawal is a named risk factor.
4. Sleep, intercurrent illness and electrolyte problems, which FDA guidance and IFCN both flag.
5. Whether a motor threshold can be reliably determined. FDA guidance lists an unreliable threshold among the reasons for close monitoring. Why the threshold matters for dose is explained in motor threshold and TMS dosing.

Screening is repeated informally during a course. A patient who has not slept, has started a new medication, has a fever or has been drinking heavily should say so before that day's session, because those are exactly the circumstances the guidelines describe.

If a seizure happens

A TMS seizure occurs during or immediately after stimulation, in a supervised setting. Stimulation is stopped, the person is protected from injury and monitored until recovered, and medical evaluation follows. Clinics are expected to have a written emergency procedure and staff trained to follow it; the role of physician oversight is discussed in who delivers TMS.

Not every jerking episode is a seizure. IFCN advises distinguishing a true seizure from convulsive syncope, a faint in which brief muscle jerks can occur. After syncope there is usually little confusion, and tongue-biting, incontinence, frothing and vomiting are rare; it is managed by lowering the head, rehydrating, cooling the skin and reassurance. The distinction matters because a faint does not carry the same implications for continuing treatment.

Whether and how TMS can continue after a genuine seizure is an individual medical decision that weighs the cause (for example, undisclosed alcohol withdrawal), the risk factors present and the alternatives.

Hearing

The coil produces a loud click with every pulse. Peak sound levels have been measured near the 140 dB limit for impulsive noise, and IFCN recommends treating the sound as potentially hazardous without hearing protection. FDA guidance requires labeling to state that patients and operators should always wear earplugs or similar protection rated at 30 dB of noise reduction.

With protection in place, studies found no change in hearing sensitivity. One permanent small threshold shift was reported after an earplug slipped out during H-coil stimulation. Anyone who notices hearing loss, ringing or ear fullness after a session should report it and be referred for an auditory evaluation.

Common side effects, cognition and mood switching

The everyday side effects are scalp discomfort or pain at the treatment site and headache. In the NeuroStar program these were the most common adverse events, and 4.5% of patients stopped acute treatment because of adverse events. In the deep TMS pivotal trial, headache occurred in 26.7% with active treatment vs 18.9% with sham. A fuller patient guide is on Mind Body Seven: TMS side effects.

Cognition does not appear to suffer. A systematic review cited by IFCN found no reliable evidence of cognitive side effects across 31 randomized trials of DLPFC rTMS. This is one of the clearest differences from ECT, discussed in TMS vs ECT.

Switching into mania or hypomania is uncommon. A meta-analysis of 25 blinded trials found the risk was not significantly different from sham. In bipolar depression specifically, a 2026 meta-analysis found rates of treatment-emergent mania low and equivalent to sham. FDA guidance still lists bipolar disorder among populations lacking evidence of efficacy or safety, and TMS for bipolar depression is off-label, so it calls for careful mood monitoring; see TMS for bipolar depression.

FDA guidance also asks for a bold labeled warning about worsening depression or suicidality. If you are thinking about suicide or are in crisis, call or text 988 (the 988 Suicide & Crisis Lifeline) or go to the nearest emergency room.

Pregnancy

FDA guidance lists pregnant or nursing patients under a precaution for lack of evidence, not as a contraindication. IFCN's cautious conclusion is that conventional rTMS is minimal risk for mother and child, with the main fetal risk being a maternal seizure; H-coil safety data in pregnancy are not yet available. Reviews disagree: a 2023 review called TMS for peripartum depression safe and well tolerated, while a 2026 review calls it experimental and notes that safety data cover only about 230 pregnancies. In the only small sham-controlled trial, three late preterm births occurred, all in the active group, and the authors could not exclude a link. Treatment options in pregnancy are discussed more broadly in depression and anxiety in pregnancy.

Post-market reports

An analysis of FDA's MAUDE database through April 2025 found 200 unique reports for depression TMS devices; the most common reported symptoms were anxiety, neurocognitive changes, seizures and headache. MAUDE reports are unverified and cannot give incidence rates, so they flag what to watch for rather than how often it happens. No FDA safety communication specific to TMS devices was found in searches run in October 2026.

For the clinical overview of who is a candidate, see the TMS therapy page; the full evidence library is in the TMS guide.

Frequently asked questions

Is TMS safe?

For appropriately screened adults, yes: side effects are usually mild and temporary, and serious events are rare. Seizure is the most serious acute risk, at roughly 0.7 per 1,000 patients in a large survey of TMS clinicians. TMS does carry labeled warnings, so screening and supervision matter.

Can you have TMS if you have had a seizure before?

A seizure history or epilepsy substantially raises risk and is a reason for close monitoring or for choosing a different treatment. Some histories, such as a seizure caused by ECT or an isolated febrile seizure in infancy, are viewed differently. Medicare policy in New York, for example, lists seizure history as a caution but excepts those two situations. This needs an individual assessment.

Does TMS cause brain damage or memory loss?

No evidence shows that it does. A review of 31 randomized trials found no reliable evidence of cognitive side effects from prefrontal rTMS, and the NeuroStar development program found no change in cognitive function.

Is drinking alcohol during a TMS course risky?

Heavy drinking and alcohol withdrawal both lower the seizure threshold, and the one seizure in the pivotal deep TMS trial followed undisclosed heavy drinking. Tell the TMS team about any change in alcohol use before the next session.

Why do I need earplugs?

Each magnetic pulse makes a loud click whose peak can approach the 140 dB limit for impulsive noise. FDA guidance requires hearing protection for every session, and with it in place studies have found no change in hearing.

Sources

1. Safety and recommendations for TMS use in healthy subjects and patient populations, with updates on training, ethical and regulatory issues: Expert Guidelines. Clinical Neurophysiology, 2021. europepmc.org
2. Seizures from transcranial magnetic stimulation 2012-2016: Results of a survey of active laboratories and clinics. Clinical Neurophysiology, 2019. europepmc.org
3. Seizure risk with repetitive TMS: Survey results from over a half-million treatment sessions. Brain Stimulation, 2021. europepmc.org
4. Safety Information. NeuroStar (Neuronetics), 2026. neurostar.com
5. 166 Post Market Rate of Seizures During TMS Treatment with NeuroStar System Appears to Be Lower than Previously Estimated. CNS Spectrums, 2020. cambridge.org
6. Transcranial magnetic stimulation in the treatment of major depressive disorder: a comprehensive summary of safety experience from acute exposure, extended exposure, and during reintroduction treatment. Journal of Clinical Psychiatry, 2008. europepmc.org
7. Efficacy and safety of deep transcranial magnetic stimulation for major depression: a prospective multicenter randomized controlled trial. World Psychiatry, 2015. pmc.ncbi.nlm.nih.gov
8. Class II Special Controls Guidance Document: Repetitive Transcranial Magnetic Stimulation (rTMS) Systems. U.S. FDA, 2011. fda.gov
9. Hypomanic/manic switch after transcranial magnetic stimulation in mood disorders: A systematic review and meta-analysis. World Journal of Psychiatry, 2021. europepmc.org
10. Efficacy, Effectiveness, and Safety of Transcranial Magnetic Stimulation for Bipolar Depression: A Systematic Review and Meta-Analysis. Biological Psychiatry Global Open Science, 2026. europepmc.org
11. Randomized controlled trial of transcranial magnetic stimulation in pregnant women with major depressive disorder. Brain Stimulation, 2019. europepmc.org
12. Digital and repetitive transcranial magnetic stimulation interventions for perinatal depression and anxiety: a systematic review. BMC Psychiatry, 2026. europepmc.org
13. Beyond the efficacy of transcranial magnetic stimulation in peripartum depression: A systematic review exploring perinatal safety for newborns. Psychiatry Research, 2023. europepmc.org
14. Postmarketing Safety of Transcranial Magnetic Stimulation: A 10-Year MAUDE Database Analysis of Adverse Events and Technological Advancements. Journal of Clinical Psychiatry, 2026. pubmed.ncbi.nlm.nih.gov
15. LCD L33398 Transcranial Magnetic Stimulation (version 40). CMS Medicare Coverage Database (Wellpoint Federal), 2026. cms.gov

Educational information, not medical advice. Whether TMS or any treatment fits you is a decision to make with a psychiatrist who knows your history.