Transcranial Magnetic Stimulation (TMS) Safety with Respect to Seizures: A Literature Review

Abstract

Transcranial magnetic stimulation is an increasingly popular FDA-approved treatment for resistant depression, migraines, and OCD. Research is also underway for its use in various other psychiatric and medical disorders. Although rare, seizures are a potential adverse event of TMS treatment. In this article, we discuss TMS-related seizures with the various coils used to deliver TMS, the risk factors associated with seizures, the differential diagnosis of its presentations, the effects of sleep deprivation and alcohol use on seizures, as well as seizure risks with protocols for traditional TMS, theta-burst stimulation, and accelerated TMS. A discussion is presented comparing the potential risk of seizures with various psychotropic medications versus TMS. Included are case reports of TMS seizures in the child/adolescent patient, bipolar disorder patients, patients with a history of a traumatic brain injury, and those with epilepsy. Reports are also shared on TMS use without seizures in patients with a history of head injuries and TMS’s continued use if patients have a seizure during their TMS treatment. Findings generated in this review suggest the following. Seizures, if present, are usually self-limiting. Most treatment recommendations for TMS-related seizures are supportive in nature. The risk of TMS-related seizures is <1% overall. TMS has successfully been used in patients with epilepsy, traumatic brain injuries, and those with a prior TMS-related seizure. The rate of TMS-related seizures is comparable to that of most psychotropic medications. While having a seizure is a rare but serious adverse effect of TMS, the benefits of treating refractory depression with TMS may outweigh the risk of suicidal ideation and other significant complications of depression.

Introduction

Transcranial Magnetic Stimulation (TMS) is FDA approved for depression, migraines, and OCD, with other symptoms and disorders being aggressively studied for benefit. Lefaucheur et al 2020¹ have an extensive review of the evidence-based use of rTMS in other disorders such as stroke, multiple sclerosis, tinnitus, anxiety, schizophrenia, substance abuse, Alzheimer’s, ALS, epilepsy, panic, PTSD, and others. TMS uses an electromagnetic coil placed on varying areas of the scalp. This coil generates magnetic pulses and stimulates different brain areas to create the desired pathophysiologic and clinical outcome. TMS was first approved for depression in 2008;² since that time, various coils have been created, altering the area and depth stimulated. It has been suggested that the different coils may have variations in side effects and the possible frequency of the adverse event of a seizure. There are also different protocols of stimulation used with TMS, with the most recent advances called theta-burst stimulation and accelerated treatment. Theta-burst stimulation uses three magnetic pulses 20 ms apart and repeated every 200 ms of 50 Hz stimulation.³ Accelerated TMS uses multiple TMS stimulation treatments per day.⁴

Methodology

This paper will focus on the adverse event of a seizure during TMS, which is an abnormal electrical activity and may be due to systemic or local causes. We will also review TMS use in patients with known epilepsy, which is usually due to a structural abnormality of the brain and is that of repeated seizures.

After having a patient who had a seizure during her TMS treatment in 2018, we began looking for all suggestions and references we could find on TMS-related seizures. We had difficulty finding specific information on whether to continue the TMS and what steps to take immediately after the patient had been stabilized from the seizure. From that time forward, we have reviewed articles/books found on PubMed, MEDLINE, Medscape, Cochrane Libray, Scopus, and Google Scholar while collecting information with Mendeley. In this article, we have presented data from 1998 to 2020 (See Table 1). We will discuss the general risk factors associated with seizures, the risks associated with the different TMS coils, reports of seizures with differing TMS protocols, the risk of psychotropic medication alteration on having a seizure, TMS seizures in the child/adolescent population, TMS seizures in bipolar disorder, as well as the possible effects of alcohol use and sleep deprivation on seizures. We will review TMS use with epilepsy and traumatic head injury patients. We will report on continued TMS after a seizure. We will also describe specific steps to take should the patient have a seizure.

Table 1.

Stultz et al TMS Seizure Review Articles

	Reference	Coil Type	Description
Introduction	Lefaucheur et al 20201	Multiple	Review on the use of TMS for multiple disorders using various coil types such as Figure 8, Circular, Double Cone, Hesed
	McClintock et al 20182	H1, Figure 8	TMS approved for depression in 2008
	Chung et al 20153	Multiple	Review of the application of theta burst TMS using multiple coil types such as Double Cone, Figure 8, H-coil
	Sonmez et al 20194	Not Specified	Review on the use of accelerated TMS protocol
TMS and Coils	Rossi et al 20095	N/A	General seizure risk with TMS <0.003%
	Oberman et al 20116	N/A	Seizure risk with theta burst rTMS is around 0.02%
	Carpenter et al 20127	Figure 8	Seizure risk <1% with TMS
	Janicak et al 20208	Figure 8	Neurostar seizure rates lower than initially reported
	Tendler et al 20189	H1	Reviewed 31 seizure cases, rate of 0.087%.
	Cavinato et al 201210	Figure 8	31 y/o male, Hx of severe TBI, had a partial and secondarily generalized tonic-clonic seizure on 4th of 10 daily sessions
	Koudijis et al 201011	Round, Fig. 8	34 children aged 5 mo. to 19 y/o with intractable epilepsy underwent TMS, temporary increase in seizures in 4 children
	Lerner et al 201912	Multiple	Risk of seizures with TMS is 0.08/1000 for Figure 8, 0.12/1000 for Double Cone, 0.43 for H-coil.
TMS Seizure Risk Factors	Rossi et al 20095	N/A	Higher frequencies and short intervals between trains can increase risk; See description under TMS & Coils
	Wasserman 199813	N/A	<20 second intervals between series leads to increased seizure risk
	McClintock et al 20182	H1, Figure 8	Tonic-clonic seizure during TMS due to direct stimulation of motor cortex or adjacent brain areas; Also cited in Intro
	George & Belmaker 200714	N/A	Several factors related to increased seizure risk
Differential Diagnosis	Sheldon et al 200215	N/A	Distinctions between syncope and seizures
	Kinback 201816	H1	42 y/o female, Tx 21 had partial tonic seizure
Theta Burst TMS	Oberman & Pascual-Leone 200917	Figure 8	33 y/o male, possibly sleep deprived, seizure during final train of session
	Purushotham et al 201818	Figure 8	15 y/o female, Hx of schizophrenia, Tx 1 had seizure
	Lenoir et al 201819	Double Cone	2 cases of seizures (One generalized, one partial complex)
	Allen et al 201720	N/A	Reviewed 3 TBS studies involving healthy children and pediatric patients with CNS Disorders, no reported seizures.
Accelerated TMS	Kallel & Brunelin 202021	Figure 8	18 y/o female, had seizure that was first localized then generalized during 3rd session of 2nd day
TMS and Psychotropics	Rossi et al 20095	N/A	Some meds/substances can potentially increase seizure risk; See description under TMS & Coils and Risk Factors
	Loo et al 200822	N/A	If medications are adjusted during TMS, re-check motor threshold
	Dobek et al 201523	N/A	TMS-induced seizures and antidepressant use; Bupropion is not a contraindication
	Lertxurdi et al 201324	N/A	2nd gen antipsychotics may have a higher risk of seizures, especially clozapine
	Khoury & Ghossoub 201925	N/A	0.5–1.2% risk of seizures with antipsychotics, clozapine increases risk the most
	Thanki et al 202026	Figure 8	60 y/o male, Hx of hyponatremic seizures, completed TMS safely twice (once on sertraline, another time on venlafaxine)
TMS in Children/Adolescents	Allen et al 201720	N/A	Reviewed 23 rTMS studies involving children w/CNS disorders & epilepsy, 3 seizures in CNS. See Theta Burst TMS
	Hu et al 201127	Figure 8	15 y/o female, on sertraline, generalized tonic-clonic seizure during Tx 1, became hypomanic from TMS
	Chiramberro et al 201328	Figure 8	16 y/o female, Tx 12 had seizure, later found high level of blood alcohol concentration
	Cullen et al 201629	H1	17 y/o female, Tx 8 had generalized tonic-clonic seizure
	Purushotham et al 201818	Figure 8	See description under Theta Burst TMS
	Wang et al 201830	Figure 8	16 y/o female, Hx of migraines w/auras, Tx 1 had generalized tonic-clonic seizure
	Muir et al 201931	H1	Reviewed 6 pts., 1 case of seizure in pt. w/Hx of autism, multiple head injuries, and D/C of oxcarbazepine prior to TMS
	Zewdie et al 201932	Fig. 8, D.Cone	Reviewed 384 children who underwent TMS, no seizures occurred
TMS and Bipolar Disorder	Tharayil et al 200533	Not Specified	35 y/o pt. with family Hx of seizures had generalized seizure while on lithium and chlorpromazine.
	Sakkas et al 200734	Figure 8	30 y/o female, Type I Bipolar Disorder, manic, stopped diazepam on her own prior to Tx 9, had Jacksonian seizure
	Harel et all 201135	H1	1 patient out of 19 had a generalized seizure
	Iliceto et al 201836	Figure 8	37 y/o male, Type I Bipolar Disorder, Hx of TBI, completed TMS safely
TMS and TBI	Dhaliwal et al 201537	Fig. 8, Focal	Severe TBI increases seizure risk
	Reti et al 201538	N/A	TBI increases seizure risk, low frequency rTMS may be better option for those with a Hx
	Bernabeu et al 200439	Circular	28 y/o female with Hx of TBI on fluoxetine, had secondarily generalized tonic-clonic seizure using fast rTMS protocol
	Cavinato et al 201210	Figure 8	See description under TMS & Coils
	Pape et al 201440	Not Specified	2 pts. w/Hx of TBI underwent TMS. Pt. #1 completed safely, Pt. #2 had an EEG seizure w/no clinical S/S at Tx 21
	Boes et al 201641	H1	27 y/o male had generalized tonic-clonic seizure at Tx 12, Hx of alcohol use and 4 head injuries
	Muir et al 201931	H1	See description under TMS in Children/Adolescents
TMS and Head Injuries w/o Seizures	Fitzgerald et al 201142	Figure 8	41 y/o female with Hx of mild TBI, completed 20 Txs
	Kreuzer et al 201343	Figure 8	53 y/o male had severe tinnitus after TBI, completed 5 Tx series over 3 years
	Neville et al 201544	Figure 8	Double-blind study involving 36 pts. aged 18–60 y/o with Hx of TBI
	Nielson et al 201545	Figure 8	48 y/o male with Hx of severe TBI, finished 30 Txs; 90% lower risk of seizure if none w/in 2 years of injury
	Englander et al 200347	N/A	CT scan results and neurosurgical procedures useful in determining risk for late posttraumatic seizures
	Koski et al 201548	Figure 8	Treated 12 pts. aged 20–60 y/o, 60% male, 60% had ≥3 concussions
	Rutherford et al 201749	Not Specified	13 pts. with mild TBI, 7 real TMS + 6 sham; rTMS may be effective for some symptoms of post-concussion syndrome
	Paxman et al 201850	Not Specified	61 y/o male had chronic dizziness after a mild TBI, completed 10 Txs
	Lee & Kim 201851	Figure 8	13 pts. aged 19–60 y/o with Hx of TBI, received 10 sessions of real or sham TMS
	Iliceto et al 201836	Figure 8	See description under TMS and Bipolar Disorder
	Saunders & Bermudes 201853	Figure 8	55 y/o female had TBI unrelated to TMS after 10 sessions, resumed TMS 11 days later, 4 different protocols used
	Siddiqi et al 201854	Not Specified	Studied effects of resting-state fMRI-targeted rTMS in a retired NFL player, Hx of repetitive head trauma.
	Stultz et al 201955	H1	23 y/o male with Hx of 4 concussions
TMS and Alcohol Use	Tendler et al 20189	H1	6/31 pts. reviewed who had a seizure involved increased alcohol intake
	Boes et al 201641	H1	See description under TMS and TBI
	Chiramberro et al 201328	Figure 8	See description under TMS in Children/Adolescents
TMS and Sleep Deprivation	Nakken et al 200556	N/A	Reported on 1677 pts. with epilepsy, those with generalized seizures report to be more sensitive to sleep deprivation
	Ferlisi & Shorvon 201457	N/A	Pts. with idiopathic generalized epilepsy more sensitive to seizures while sleep deprived
	Prikryl & Kucerova 200558	Not Specified	45 y/o male who was sleep deprived for 2 nights had a grand mal seizure during Tx 6
	Tendler et al 20189	H1	3/31 pts. reviewed who had a seizure c/o sleep deprivation
	Oberman & Pascual-Leone 200917	Figure 8	See description under Theta Burst TMS
TMS and Epilepsy	Bae et al 200759	N/A	Risk of seizures in pts. with epilepsy <2%
	Pereira et al 201660	N/A	Reviewed 410 epilepsy pts. receiving TMS, 12 of which had a seizure, suggesting a seizure risk of 2.9%
	Vernet et al 201261	Not Specified	22 y/o male, drug-resistant symptomatic focal epilepsy, had seizure clinically similar to typical spontaneous seizures
	Allen et al 201720	N/A	Reviewed 23 rTMS studies involving children with CNS disorders & epilepsy, no seizures reported in epileptic pts.
	Koudijis et al 201011	Round, Fig. 8	See description under TMS and Coils
	Stultz et al 201962	H1	69 y/o female, Hx of complex partial seizures, completed TMS without having a seizure
	Fitzgerald 201463	Not Specified	57 y/o male, tried ECT in the past, had not experienced seizures for 11 years, completed TMS safely at low-frequency
Continued TMS After a Seizure	Bagati et al 201264	Not Specified	44 y/o male, had seizure during Tx 4, continued TMS after being prescribed valproate
	Stultz et al 201965	H1	48 y/o female, developed tonic clonic seizure during Tx 11, continued TMS after evaluation with no reported seizures
	Kallel & Brunelin 202021	Figure 8	See description under Accelerated TMS. Continued with TMS after decreasing to 1 session per day.
Treatment Plan	Fitzgerald & Daskalakis 201266	N/A	rTMS-related seizures likely to terminate quickly and not require additional treatment

TMS and Coils

The general risks of a seizure with TMS have been reported to be <1/30,000 (<0.003%) by Rossi et al 2009.⁵ The risk of seizure with theta burst is estimated at 0.02% by Oberman et al 2011.⁶ With the Figure-8 coil, the risk of seizure has been estimated by Carpenter et al 2012⁷ to be 3/1000 or <1% overall. Janicak et al 2020⁸ reported postmarketing seizure rates with the Neurostar coil are even lower than previously reported. Tendler et al 2018⁹ reviewed available data for the Brainsway H1-Coil and reported a seizure rate of 0.087%. They reviewed 31 reported cases of seizures with the H1-Coil and felt most were due to the motor threshold (MT) not being rechecked weekly, as recommended by the company. Six seizures were due to stimulation with intensity >120% of the Motor Threshold. No seizures were reported with the first TMS treatment. Nine seizures were associated with medication issues, increased alcohol intake was reported with six patients, and three seizures were associated with sleep issues. While most seizures occur during or around the time of treatment, Cavinato et al 2012¹⁰ reported on a patient with a traumatic brain injury who had a seizure 3 hours after his TMS treatment. Also, Koudijs et al 2010¹¹ reported an increased frequency of seizures for up to 3 days in four children with known epilepsy who had TMS treatment.

Lerner et al 2019¹² reviewed surveys from 174 providers from 2012 to 2016, revealing 24 seizures out of 318,560 TMS sessions. They reported the risk of seizures with TMS was 0.08/1000 and that “TMS delivered within published guidelines to individuals without risk factors appears to cause fewer than one seizure per 60,000 sessions.” In their review, 62% of seizures occurred on the first exposure to TMS and 75% within the first three treatments. Of the 24 seizures, seven occurred in patients with congenital epilepsy, all of whom were on anti-epileptic medication at the time. Two patients were described as having refractory epilepsy. Lerner et al described the seizure risk for the different coil types as follows: 0.08/1000 for conventional figure-8 coils, double cone coils of 0.12/1000 and 0.43/1000 for H-coils. Only three seizures were reported with the H-coil, but only 7577 (2%) of the 318,560 TMS sessions were completed using the H-coil. The smaller sample size may have made the H-coil numbers appear higher.

TMS Seizure Risk Factors

Rossi et al 2009⁵ conveyed

Seizures are caused by hyper synchronized discharges of groups of neurons in the gray matter, mainly due to an imbalance between inhibitory and excitatory synaptic activity in favor of the latter. Seizures can be induced by rTMS when pulses are applied with relatively high-frequencies and short interval periods between trains of stimulation.

Wasserman 1998¹³ stated that “Intervals between series less than 20 seconds were associated with increased seizures”. (This may have been the cause of the increased seizures reported earlier in the literature using TMS.) McClintock et al 2018² stated

the risk of tonic-clonic seizure, a rare event during rTMS, is related to the direct stimulation of motor cortex or stimulation of the adjacent brain areas with spread of neuronal excitation to the motor cortex. Inspection of the contralateral hand for signs of twitching or movement during stimulation may ensure that stimulation does not spread from prefrontal to primary motor cortex, which can lead to generalized seizure induction with tonic-clonic movement pattern.

McClintock also stated, “seizures can occur within safety guidelines, even in patients who present with no known risk factors.”

The risk factors associated with TMS-related seizures were discussed in the book entitled “Transcranial Magnetic Stimulation in Neuropsychiatry” (George and Belmaker 2007¹⁴). They reported on the increased risk of seizures with focal or generalized encephalopathy, severe head trauma, non-treated epilepsy, family history of epilepsy in first-degree relatives, heavy alcohol use, severe cardiac disease, increased intracranial pressure, medications that lower seizure threshold, cocaine usage, and other epileptogenic drugs. Other risk factors associated with seizures described throughout the years include a personal history of seizure, stroke, epilepsy, concussions, family history of brain tumors, eating disorders, neurologic disease with altered seizure threshold, sleep deprivation, excessive caffeine/stimulant/cocaine use, MDMA (Ecstasy) use, theophylline use, alcohol withdrawal, benzodiazepine withdrawal, electrolyte disturbance (decreased sodium, change in glucose), other medications that lower seizure threshold, depression, demyelinating disorders (such as Multiple Sclerosis), and fever.

Differential Diagnosis of TMS Seizures

Concerning the differential diagnosis, a distinction must be made between vasovagal syncope and seizure during an episode. Vasovagal syncope symptoms include a transient rapid onset, self-limited loss of consciousness associated with lightheadedness, pale skin, yawning, blurred vision, and a feeling of being warm, cold, clammy, or sweaty (Sheldon et al 2002¹⁵). Jerky myoclonic movements have also been reported with both vasovagal syncope and seizures.

TMS seizures can be generalized or partial. Kinback in 2018¹⁶ described a partial tonic seizure in a 42-year-old white female using the H1 coil during treatment 21, where she developed tense jaw and arm muscles, fixed gaze, unresponsiveness, and bilateral upper extremity flexion with clinched wrists. She had slight spine arching but no lower extremity involvement. She stayed fully aware without incontinence or LOC.

Theta-Burst TMS and Seizures

As previously stated, theta-burst TMS has been associated with an estimated seizure risk of 0.02%.⁶ Oberman and Pascual-Leone 2009¹⁷ recounted a seizure in a 33-year-old man with no risk factors (except possible sleep deprivation) using theta burst. Purushotham et al 2018¹⁸ reported on a 15-year-old female with schizophrenia who developed a seizure with theta burst within the first 30 seconds of the first session. Lenoir et al 2018¹⁹ described two cases of seizures with theta burst. One had a generalized seizure, and one a partial complex seizure. Allen et al 2017²⁰ identified no seizures after reviewing three theta-burst studies in 90 healthy children and 40 pediatric patients with CNS disorders.

Accelerated TMS and Seizures

Kallel and Brunelin 2020²¹ described an 18-year-old female with major depression receiving a protocol of 5 sessions per day during four consecutive working days over the left dorsolateral prefrontal cortex using the MagPro X30 figure-eight coil. The patient developed a seizure on the third session of the second day. The seizure underwent secondary generalization and was associated with urinary incontinence and postictal confusion. Twelve days later, the patient agreed to additional TMS given at 1-Hz right DLPFC treatment with one session per day for 30 additional sessions. She also had maintenance TMS scheduled every 2 weeks afterwards and had no seizure activity during her continued or maintenance treatments.

TMS, Psychotropic Medications, and Seizures

Over the years, various medications have been used during which a TMS-related seizure has occurred, but we have found no absolute contraindication with any specific medication to date. Rossi et al⁵ in their safety article of 2009 listed several medications described as having either a strong potential for hazard or a relative hazard for interactions with TMS. They also included medications that if withdrawn could create a strong relative hazard. (Please see their article for a complete list of medications listed.) Since their 2009 article, many of these substances have been used during TMS without incident.

There has been discussion about changing medications during the TMS treatment, and Loo et al 2008²² stated,

If medications are altered during the TMS course, the motor threshold should be re-measured and the stimulus intensity adjusted accordingly. This is based on the premise that medications which alter seizure threshold could also alter motor cortical threshold.

Dobek et al 2015²³ while studying bupropion treatment with TMS and after an extensive literature review reported the following seizure incidence rates in percent for psychotropic medications independent of TMS use: Bupropion SR 0.1%, Bupropion IR 0.4%, Citalopram 0.25%, Duloxetine 0.2%, Fluoxetine 0.2%, Fluvoxamine 0.2%, Mirtazapine 0.04%, Paroxetine 0.1%, Sertraline up to 0.2%, Venlafaxine 0.3%, Tricyclics 0.1–0.4%, Olanzapine 0.9%, Quetiapine 0.8%, Aripiprazole 0.4%, Ziprasidone 0.4%, and Risperidone 0.3%. Lertxurdi et al 2013²⁴ reported that second-generation antipsychotics might have a higher risk of seizures, especially clozapine and possibly olanzapine and quetiapine. Khoury and Ghossoub 2019²⁵ revealed studies with antipsychotics are associated with 0.5% to 1.2% risk of seizures, and also reported clozapine appears to increase the risk the most. The above information reveals that the rates of seizures for psychotropic medications are similar or higher than the general TMS rates of <0.003% to 0.087% mentioned previously. Dobek et al²³ also stated that TMS-induced seizure data are not associated with any particular antidepressant, and bupropion is not a contraindication to TMS. Thanki et al 2020²⁶ even reported on using rTMS in a 60-year-old male with a history of hyponatremic seizures with antidepressants on two different occasions while on sertraline 50 mg and then later on venlafaxine 75 mg, who was later successfully treated with 1 Hz TMS to the right dorsolateral prefrontal cortex at 90% of the resting motor threshold without having a seizure.

TMS Seizures in Child/Adolescent Patients

The overall risk of TMS-related seizures in the child/adolescent population appears to be similar to that of the adult. Allen et al 2017²⁰ reported on a review in the pediatric population of 23 repetitive TMS (rTMS) studies (including 230 CNS and 24 epileptic children) identifying three seizures in the CNS patients, with a risk of 0.14% per session. They reviewed three theta-burst studies in 90 healthy children and 40 pediatric patients with CNS disorders and identified no seizures.

Case reports of seizures in the child/adolescent patient have been reported. Hu et al 2011²⁷ described an adolescent female on Zoloft who had a seizure and became hypomanic with TMS. She subsequently stopped TMS. Chiramberro et al 2013²⁸ reported on a 16-year-old girl who had a seizure during the 12th treatment session with TMS. Of interest, the patient was on sertraline, olanzapine, and hydroxyzine, and she was also found to have used alcohol around the time of treatment. Cullen et al 2016²⁹ recounted a 17-year-old female in a sham-controlled study who developed a generalized tonic-clonic seizure during the 8th treatment session of TMS (which was the first day at 120% MT). Purushotham et al 2018¹⁸ reported a 15-year-old girl with schizophrenia who had a seizure while using theta-burst TMS. Wang et al 2018³⁰ told of a 16-year-old girl previously diagnosed as having migraine with aura who had a seizure 10 seconds after the onset of the third rTMS train.

In the transition years to adulthood ages (18–20), Muir et al 2019³¹ reviewed six patients and presented a case of one seizure using the H 1 coil at 120% MT with the risk factors of autism, multiple head injuries, and the discontinuation of oxcarbazepine during the week before TMS initiation. They felt the spontaneous discontinuation of oxcarbazepine was the most likely cause. Most recently, Zewdie et al 2019³² reviewed the results of 384 children over ten years that underwent TMS or direct current stimulation (tDCS). Neuromodulation TMS occurred in 119 of the patients with a median age of 14. “Despite 221 (58%) of the patients having some sort of brain injury/or epilepsy, no seizures occurred with single, paired, or rTMS or with tCDS.”

TMS Seizures in Bipolar Patients

TMS seizures have been described in diagnoses other than depression and in the bipolar patient have been reported in 2005 by Tharayil et al, by Sakkas et al, and described again by Harel et al in 2011. Tharayil et al 2005³³ reported a generalized seizure in a bipolar patient while on chlorpromazine and lithium who also had a positive family history of seizures. Sakkas et al 2007³⁴ described a Jacksonian seizure in a manic patient treated with rTMS. Harel et al 2011³⁵ studied TMS in 19 Bipolar patients and had only one generalized seizure patient. In 2018, Iliceto et al³⁶ treated a patient with bipolar disorder, generalized anxiety, a TBI, and a history of a suicide gesture without the patient having a seizure. The specific seizure rate with TMS in bipolar disorder is not clear at this time.

TMS in Traumatic Brain Injury Patients

Head injuries are another situation associated with an increased risk of seizures. Dhaliwal et al 2015³⁷ stated, “patients with severe TBI do show an increased risk of unprovoked seizures,” and

there is a strong connection between the severity of a brain injury and the subsequent risk of seizures; individuals with mild to moderate TBI have a substantially lower seizure risk than those with severe TBI.

Reti et al 2015³⁸ indicated,

TBI is associated with an increased risk of both early and late spontaneous seizures, a significant consideration in evaluating rTMS as a potential treatment for TBI depression. Whilst the risk from rTMS is low; underlying neuropathology may somewhat increase that risk.

They suggested that low-frequency rTMS might be less likely to trigger a seizure.

Various reports of patients with a history of a head injury having a TMS-related seizure have been described. Bernabeu et al 2004³⁹ reported a seizure in a TBI patient on fluoxetine but was using a brief interstimulus interval. Cavinato et al 2012¹⁰ described a 31-year-old male with a history of a severe TBI 8 months before TMS initiation. He had a secondary generalized seizure on the 4th of 10 daily sessions. TMS was delivered at 90% MT, 20 Hz, 1s train duration, and 1-minute inter-train interval to the dorsolateral prefrontal cortex (DLPFC). The seizure developed 3 hours after treatment. Pape et al 2014⁴⁰ reported on a 32-year-old male who had a TBI 9 years before TMS initiation who had an EEG identified seizure during the treatment with no clinical expression. Boes et al 2016⁴¹ reported on a 27-year-old male with MDE, GAD, alcohol use, and a history of 4 prior head injuries. He was also sleep deprived. He developed a seizure during the 12th treatment session. As described previously, Muir et al 2019²⁹ presented a patient with a history of oxcarbazepine withdrawal the week before, autism, and multiple head injuries who had a seizure.

TMS Treatment Without Seizures in Patients with Head Injuries

Transcranial Magnetic Stimulation has also been used successfully without seizures in patients with a history of brain injury. Fitzgerald et al 2011⁴² presented one of the original studies about using TMS without adverse events in a depressed patient with a history of a mild TBI 14 years before the TMS. Kreuzer et al 2013⁴³ treated a TBI patient with TMS for severe tinnitus after a head injury with five treatment series of 1 Hz to the left primary auditory cortex at 110% MT and no reported seizure. Neville et al 2015⁴⁴ completed a randomized controlled trial of 36 patients with TBI divided into two groups and administered ten sessions of 10 Hz TMS over the left DLPFC, which demonstrated improvement in depression and cognitive functioning. Nielson et al 2015⁴⁵ reported on a 48-year-old man with a severe TBI history 5 years before TMS was administered with low-frequency right dorsolateral prefrontal cortex stimulation daily for 6 weeks. The patient demonstrated a 49% improvement in his Hamilton Depression Rating Scale⁴⁶ with treatment. In their article, they set forth that those who have not had a seizure within 2 years of injury have a 90% lower risk of seizure based on work by Englander et al 2003.⁴⁷

Koski et al 2015⁴⁸ treated 12 patients with post-concussive symptoms (headaches, depression, and cognitive deficits) following mild TBI using rTMS to the left DLPFC at 5-sec trains, 10 Hz, and 110% Motor Threshold. Their patients demonstrated a decrease of 14.6 points (p= 0.009) of post-concussive symptoms. Rutherford et al 2017⁴⁹ studied 13 patients with mild TBI delivering 13 treatment sessions of rTMS to the left DLPFC at 20 Hz in 1.5-second trains of 30 pulses and an inter-stimulation separation of 28.5 seconds. Paxman et al 2018⁵⁰ safely used TMS in a mild TBI patient when treating for chronic dizziness with ten sessions of rTMS to the left DLPFC at 70% motor threshold and a frequency of 10 HZ. Lee and Kim 2018⁵¹ studied 13 patients divided into an experimental group and sham group. The patients were given rTMS to the right DLPFC for ten sessions and demonstrated improvements in mood based on the Montgomery-Asberg Depression Rating Scale.⁵² Iliceto et al 2018³⁶ treated a 37-year-old man with a severe TBI secondary to a suicide attempt who had a history of anxiety and bipolar disorder. The patient received 30 treatments of 6 Hz TMS to the left DLPFC at 120% MT and a 26-second inter-stimulation interval. Using the PHQ-9 he had a 70.8% improvement in mood.

Saunders and Bermudes 2018⁵³ described a 55-year female with a TBI with loss of consciousness during her series of TMS treatments. The event was unrelated to the TMS, and the patient was hospitalized for 4 days. TMS was restarted 11 days after the injury without incident. In 2018 Siddiqi et al⁵⁴ used 20 sessions of bilateral TMS in a retired NFL defensive lineman. The patient reported at least 12 previous concussions. Using the Montgomery – Asberg Depression Rating Scale, the patient documented improved scores from 32 to 9. Stultz et al 2019⁵⁵ presented a case of a 23-year-old male with a history of four concussions (two of which required hospitalization) having significant depression, generalized anxiety, panic, and OCD symptoms who received 34 TMS treatments and nine booster sessions without a seizure and with improvement in mood and anxiety. He was treated with the H1 coil at 120% MT to the left DLPFC.

TMS Seizures and Alcohol

Excessive alcohol use and withdrawal can precipitate seizures and has been found to possibly increase the risk of TMS-related seizures. Tendler et al 2018⁹ indicated 6 of the 31 patients reviewed with TMS-related seizures using the H1 coil had a history of increased alcohol intake. Alcohol use was also reported in TMS seizure patients’ history by Boes et al 2016⁴¹ and Chiramberro et al 2013.²⁸ These cases are described elsewhere in this report.

TMS Seizures and Sleep Deprivation

Sleep deprivation has long been identified as a precipitant to having a seizure in epilepsy patients and may increase the risk of a TMS-related seizure. Nakken et al 2005⁵⁶ reported on 1677 patients with epilepsy revealing 53% reported at least one seizure precipitant, with 30% reporting two or more factors had contributed at times to having a seizure. The three most common seizure-precipitating factors were emotional stress, sleep deprivation, and tiredness. Patients having generalized seizures were reported to be more sensitive to sleep deprivation. Ferlisi and Shorvon 2014⁵⁷ studied 104 patients at a tertiary-care adult epilepsy clinic to identify triggering factors to their epilepsy. Ninety-seven percent cited at least one precipitant, with stress, sleep deprivation, and fatigue being the most commonly reported. Those with idiopathic generalized epilepsy were more sensitive to seizures during wake periods and sleep deprivation, with those having extratemporal epilepsy reporting more seizures during sleep.

There are reports of patients having a seizure during TMS in which sleep deprivation may have played a role. Prikryl and Kucerova 2005⁵⁸ described a seizure in a sleep-deprived depressed patient at 110% MT, 15 Hz, and train duration of 10 seconds. Tendler et al 2018⁹ reported that of the 31 patients they reviewed having had a seizure with the H1 coil; three had sleep issues. Oberman and Pascual-Leon 2009¹⁷ reported on a 33-year-old male who had flown from London to Boston and may have been sleep-deprived before the TMS in which he had a seizure.

TMS in Patients with Epilepsy

TMS has been used in patients with known epilepsy. Bae et al 2007⁵⁹ reported the seizure risk with TMS in patients with epilepsy is less than 2% (4 of 280 patients). Pereira et al 2016⁶⁰ recounted 12/410 patients who had a history of epilepsy developed a seizure during TMS, suggesting a crude rate of seizures per subject of 2.9% in epilepsy patients treated with TMS. Vernet et al 2012⁶¹ usually indicated TMS-associated seizures in epilepsy patients have been clinically similar to the patients’ typical spontaneous seizures, and the risk of a seizure in temporal association with TMS is less than 2% in epilepsy patients. Allen et al 2017,²⁰ in their previously mentioned review of 23 rTMS studies using TMS in children, included 24 pediatric patients with epilepsy. Koudijs et al 2010¹¹ reported an increased frequency of seizures for up to 3 days after TMS in 4 children with known epilepsy based upon follow-up phone conversations. Stultz et al 2019⁶² reported on a 69-year-old white female with a history of complex partial seizures, resistant depression, generalized anxiety disorder, and panic disorder who had three generalized seizures in her lifetime. She received 31 TMS treatments with the H1 coil at 120% MT over the left DLPFC without having a seizure. She has returned for booster TMS sessions without having a seizure. Fitzgerald 2014⁶³ described a 57-year-old man diagnosed with epilepsy when he was 26 years old who had been stabilized on sodium valproate and lamotrigine. He was later treated with rTMS of 1 Hz stimulation to the right dorsolateral prefrontal cortex for 20 consecutive weekdays with documented improvement in both mood and anxiety without a seizure.

Continued TMS After a Seizure

If the patient does have a seizure, then the question is that of whether to continue treatment. Bagati et al 2012⁶⁴ described a patient who developed a seizure during treatment session #4, and while using valproate to cover for seizures, he finished TMS. Stultz 2019⁶⁵ reported previously on a patient who had a seizure during treatment session #11 but was then able to continue TMS without additional seizures or additional medications. The patient received 39 treatment sessions in total. In the accelerated TMS article referenced previously by Kalle and Brunelin²¹, the patient continued treatment and later maintenance treatment after having a seizure.

The Treatment Plan Should a Patient Have a Seizure

Fitzgerald and Daskalakis⁶⁶ report,

Seizures are likely to terminate fairly rapidly and not require medication treatment: emergency response medical units should potentially be summoned if the seizure does not terminate in several minutes.

Most treatment recommendations for TMS seizures are supportive, with ensuring patient safety, avoiding aspiration, and monitoring blood pressure if possible. Currently, there are no specific recommendations for oxygen or suctioning during the seizure.

As seizures are usually a rare occurrence in the outpatient psychiatric office, one must also debrief the staff and may use the episode as a teachable moment to review seizure protocols with the staff and emphasize the need to be discussing sleep hygiene continually, alcohol use, medication change, etc., with patients daily before their TMS treatment. Concerning documentation, always include the date of occurrence, TMS treatment number, train, placement, motor threshold, and device used. A detailed description of the seizure, how it first presented and then progressed should be described. Documentation should include any identifiable risk factors and the absence of others. List all medications the patient was taking at the time of the seizure, including non-psychotropic meds. Review with the patient any changes they may have had before the seizure concerning sleep, medications, or alcohol/caffeine use. The risk of not treating severe depression versus the risk of another seizure must be discussed with the patient and the family if TMS is to be continued. While not always necessary, some practitioners have added anti-epileptic medications to the treatment regimen before continuing TMS treatment. The seizure should be reported to the FDA and the manufacturer of the TMS unit. Also, one should consider a letter to the editor of a journal to help keep others informed of potential risks and outcomes.

Conclusion

In summary, while TMS has a risk of seizure that is generally <1% overall, some situations may increase that risk, such as alcohol use, brain injury, sleep deprivation, family history, etc. Various case studies have been described in this article in which these situations may have been contributing factors. Traditional TMS, theta burst, accelerated TMS, and child/adolescent TMS use have been reviewed with similar risks for seizures. Comparable seizure rates are reported with both TMS and various psychotropic medications we prescribe daily. (TMS has a lower risk than some of the medications.) We have described successful continued TMS use in patients having had a seizure during TMS and in those with a history of head injury/concussion. TMS use in patients with known epilepsy has also been reviewed. While seizure is a potentially serious adverse event of TMS and must be discussed with the patient, the benefit of treating refractory depression and possibly suicidal ideation may outweigh the risk.

Disclosure

The authors have no conflict of interests to report.