transcranial magnetic stimulation for migraine: a safety review

Review Article

Transcranial Magnetic Stimulation for Migraine:A Safety Reviewhead_1697 1153..1163

David W. Dodick, MD; Carol T. Schembri, MS; Michele Helmuth, RN, MN, CANP; Sheena K. Aurora, MD

Objective.—To review potential and theoretical safety concerns of transcranial magnetic stimulation (TMS), as obtainedfrom studies of single-pulse (sTMS) and repetitive TMS (rTMS) and to discuss safety concerns associated with sTMS in thecontext of its use as a migraine treatment.

Methods.—The published literature was reviewed to identify adverse events that have been reported during the use ofTMS; to assess its potential effects on brain tissue, the cardiovascular system, hormone levels, cognition and psychomotor tests,and hearing; to identify the risk of seizures associated with TMS; and to identify safety issues associated with its use in patientswith attached or implanted electronic equipment or during pregnancy.

Results.—Two decades of clinical experience with sTMS have shown it to be a low risk technique with promise in thediagnosis, monitoring, and treatment of neurological and psychiatric disease in adults. Tens of thousands of subjects haveundergone TMS for diagnostic, investigative, and therapeutic intervention trial purposes with minimal adverse events or sideeffects. No discernable evidence exists to suggest that sTMS causes harm to humans. No changes in neurophysiological functionhave been reported with sTMS use.

Conclusions.—The safety of sTMS in clinical practice, including as an acute migraine headache treatment, is supported bybiological, empirical, and clinical trial evidence. Single-pulse TMS may offer a safe nonpharmacologic, nonbehavioral thera-peutic approach to the currently prescribed drugs for patients who suffer from migraine.

Key words: migraine, safety, repetitive transcranial magnetic stimulation, single-pulse transcranial magnetic stimulation

Abbreviations: CSD cortical spreading depression, rTMS repetitive transcranial magnetic stimulation, sTMS single-pulsetranscranial magnetic stimulation, TMS transcranial magnetic stimulation, VNS vagus nerve stimulator

(Headache 2010;50:1153-1163)

From Mayo Clinic, Phoenix, AZ, USA (D.W. Dodick); Neuralieve, Inc., Sunnyvale, CA, USA (C.T. Schembri and M. Helmuth);Swedish Clinic, Seattle, WA, USA (S.K. Aurora).

Sources of financial support: Not applicable.

Clinical trial registration number: Not applicable.

Address all correspondence to D.W. Dodick, Department of Neurology, Mayo Clinic, 1344 E. Shea Blvd, Scottsdale, AZ 85259USA.

Accepted for publication April 12, 2010.

Conflict of Interest: David W. Dodick, MD: Consulting/Honoraria: Allergan, Addex, Alexza, Almirrall, Coherex, Eli Lilly, Endo,GlaxoSmithKline, HS Lundbeck, Kowa, MAP, Medtronic, Merck, Minster, Neuralieve, Neuraxon, Novartis, NuPathe, Pfizer,Nautilus, Zogenix, Boston Scientific Editorial: Wiley-Blackwell (Cephalalgia, Headache Currents); SAGE (Cephalalgia) AmericanCollege of Physicians (MKSAP Editor); The Neurologist. Research Support: Medtronic, Advanced Neurostimulations Systems, St.Jude, NINDS, Mayo Clinic. Stocks: None. Speakers Bureau: None. Sheena K.Aurora, MD: Grants and Research Support (within last2 years): Advanced Bionics, Alexza, Allergan, Capnia, GlaxoSmithKline, MAP Pharmaceuticals, Merck and Co, OrthoMcNeil,Neuralieve, NuPathe, Takeda. Consulting: Ortho-McNeil Pharmaceutical, Inc, Merck and Co, GlaxoSmithKline, Allergan, Neu-ralieve, NuPathe, MA. Honoraria: Merck and Co., GlaxoSmithKline, Kowa, NuPathe, Ortho-McNeil Pharmaceutical, Inc.

ISSN 0017-8748doi: 10.1111/j.1526-4610.2010.01697.xPublished by Wiley Periodicals, Inc.

Headache© 2010 American Headache Society

1153

Transcranial magnetic stimulation (TMS) is anoninvasive method by which weak electrical cur-rents are induced in the brain by a rapidly changingmagnetic field. When TMS is applied to the head, themagnetic field passes through the skull, inducing mildelectric currents in the brain, which excite and depo-larize neurons in the brain. TMS has been routinelyused for over 2 decades to measure corticomotorexcitation and inhibition of neurophysiologic proper-ties in conscious subjects and has been shown to be alow risk technique with promise in the diagnosis,monitoring, and treatment of neurological and psy-chiatric disease in adults.1-4

Studies of TMS in migraineurs have shown excit-ability of the motor cortex,5,6 suggesting that this neu-rophysiological correlate may have a role in migrainemechanisms. Studies of TMS in migraineurs have alsoshown hyperexcitability of the occipital cortex,7,8

which was associated with a propensity to visuallytriggered headaches in some patients.8 Migraine is achronic and often debilitating neurological diseasethat affects nearly 12% of the world’s population.9 Itis characterized by the abnormal modulation ofsensory information processing and increased excit-ability and responsiveness to environmental andendogenous triggers.10,11 Cortical spreading depres-sion (CSD), an intense depolarization of neuronaland glial membranes, has been established as the bio-logical substrate of the migraine aura.11 CSD is thephysiological expression of the underlying brainhyperexcitability, particularly involving the occipitalcortex, which has been demonstrated in subjects withmigraine with aura.12-14 CSD is a noxious event which,in animal models, has been demonstrated to result inneurogenic inflammation and activation of nocicep-tive trigeminal afferents.15 The importance of CSD inhumans with migraine has also been recently high-lighted by the demonstration that drugs establishedas effective for migraine prevention inhibit CSD inanimals.16

Transcranial magnetic stimulation has been dem-onstrated in animal models to inhibit CSD and thushas the potential to terminate aura and reduce theduration or severity of the ensuing migraine head-ache in patients who suffer from migraine with aura.17

Recently, several clinical studies have shown that

single-pulse TMS (sTMS) is an effective and well-tolerated treatment for migraine with or withoutaura,18-20 thus suggesting that sTMS may offer a non-pharmacologic, nonbehavioral therapeutic approachto the currently prescribed drugs for patients whosuffer from migraine.

This article reviews the potential and theoreticalsafety concerns of TMS, as obtained from a review ofpublished data from sTMS studies and related repeti-tive TMS (rTMS) studies, and discusses the safetyconcerns associated with TMS in the context of theuse of sTMS as migraine treatment.

TMS DEVICESTypical commercially available TMS devices have

a microprocessor-controlled power source capable ofstoring electrical energy in 1 or more high voltagecapacitors. The capacitor(s) are connected to an insu-lated, wire coil. In response to a trigger, the storedenergy in the capacitors is rapidly discharged throughthe wire coil. The resulting high current pulse passesthrough the coil creating a transient magnetic pulse,with an approximate rise time of 100-200 microsec-onds and a total pulse time below 1 millisecond.Whenthe coil is located against the head, this transient mag-netic field passes through the scalp and skull, inducingan electrical current in the underlying cortical regionwithout the pain of direct percutaneous stimulation.The peak magnetic field strengths of conventionalTMS devices are 1.5-2 Tesla, comparable to fieldsfrom clinical magnetic resonance imaging scanners.The magnetic field volume is small and the magneticfield decays exponentially with distance. Tissues andobjects several centimeters from the coil are unaf-fected.1,21,22

In s TMS, the magnetic stimuli are delivered notmore than every few seconds (usually longer than5-10 seconds) and are not coupled with another pulse.rTMS involves regularly repeated trains of pulsesdelivered to a single site on the scalp. rTMS causeincreased or decreased excitability of the corticalpathways. The effects depend upon the intensity andfrequency of stimulation and the orientation of thecoils on the head.

The US Food and Drug Administration charac-terizes TMS by frequency. Low-frequency TMS refers

1154 July 2010

to a stimulus delivered less than 1 pulse per secondand does not present a potential for serious risk to thehealth, safety or welfare of the patient. High-frequency TMS refers to a stimulus that is deliveredat 1 pulse per second or more and potentially posesrisk. The division is based on the different physiologi-cal effects and degrees of risk associated with the 2types of stimulation.1,22 From the perspective of char-acterizing electric or magnetic fields within the elec-tromagnetic spectrum (eg, radio waves, light),frequencies below 3000 hertz are generally consid-ered as extremely low frequency.23 The InternationalCommission on Non-Ionizing Radiation Protectionconsiders extremely low frequency to be any fre-quency below 100,000 hertz.24

A PubMed literature search was performed forsafety of single pulse and repetitive TMS and articlesreviewed. The sTMS safety search was extended toadverse events, seizures, cortical excitability, low fre-quency TMS, CSD, electromagnetic fields, pregnancy,

and migraine research. As studies with rTMS provideadditional data with which to evaluate the safety ofTMS, significant findings are discussed to allow forcomparison and a more thorough assessment of thepotential safety concerns associated with the use ofsTMS as migraine treatment.

SUMMARY OF SCIENTIFIC RESEARCHTMS NONCLINICAL SAFETY DATA

Nonclinical safety data are summarized inTable 1.

CLINICAL SAFETY DATAAdverse Events Reported in Clinical Trials of

TMS.—Since its introduction in 1985, tens of thou-sands of subjects have undergone TMS for diagnostic,investigative, and therapeutic intervention purposes.Few adverse events have been reported,2 and noshort- or long-term sequelae have been reported ineither normal subjects or in individuals with neuro-

Table 1.—TMS Safety Conclusions From Nonclinical Data of Direct Electrical or Magnetic Stimulation on Brain and Nerves

Area studied Experiment or guidance TMS result Conclusion

Neural response to directelectrical stimulation andthreshold that yieldshistopathologic damage51,52

Recommendation that chargedensities be kept below40 mC/cm2/phase51

Estimation that TMS induces0.8 to 1.1 mC/cm2/phase53

Induced current by TMS inthe brain is 40-fold lowerthan the recommendedlimit for direct neuralelectrical stimulation46

Direct stimulation of sciaticnerves of cats at 20, 50, 100hertz54

Stimulus pulse amplitudesexceed amount to recruitalpha component of evokedcompound action potential

Low frequency electricalstimulation induces little orno neural injury stimuluseven when the amplitude isvery high54

Effects of chronic rTMS onmetabolite concentrationsin brain tissue

Conscious rats werestimulated 1000 times perday for 5 days at 1 hertzand an intensity of 115% ofthe motor threshold55

No alterations ofN-acetyl-aspartate, creatine,phosphocreatinine,choline-containingcompounds, myo-inositol,glucose or lactate wereobserved. No histologicalchanges in microglial andastrocyte activation

Chronic low-frequency rTMStreatment was not found tohave a delirious effect onthe central nervous system55

Higher cerebral function 3 monkeys received a total of7000 maximum intensitysTMS applications over 30days

No short- or long-termdeficits in higher cerebralfunction or any otheradverse effects wereobserved56

rTMS = repetitive transcranial magnetic stimulation; sTMS = single-pulse transcranial magnetic stimulation; TMS = transcranialmagnetic stimulation.

Headache 1155

logical disorders.1,22 In general, discomfort at the scalpbecause of muscular contraction and associated nervestimulation on the overlying skin and headache,usually with motor or premotor stimulation, havecommonly been reported in rTMS studies;22 however,these symptoms, mild to moderate in severity, haveresponded promptly to simple analgesics. In contrastto rTMS, discomfort is rarely a problem with sTMS.22

Safety data from sTMS use to treat migraine areavailable from a study in which patients with migraineused a Cadwell MES-10 stimulator (Cadwell Labora-tories, Kennewick, WA, USA) to apply 2 brief pulsesof TMS over the area of the perceived pain. In thisrandomized study,18 42 patients with migraine head-aches received treatment with high (50% maximumstimulation) or low (30% maximum stimulation)sTMS treatments. Adverse events included a briefsensation of dizziness (1 patient), drowsy duringtreatment (1 patient), and fatigue (2 patients);however, an increase in energy posttreatment wasreported by some of the patients.

Safety data on sTMS use to treat migraine are alsoavailable from studies in which migraine patients usedan investigational TMS device developed to treatmigraine (Neuralieve, Inc, Sunnyvale, CA, USA) anddeliver 2 single TMS pulses to the occipital area. In thefirst randomized, double-blind study,19 headache epi-sodes over a 3-month period were treated with theNeuralieve investigational TMS device (29 headachesin 23 patients) or with a sham device (21 headaches in19 patients) to evaluate the feasibility and safety ofsTMS as migraine treatment; no adverse events werereported. In the second study, a multicenter, random-ized, double-blind study,20 patients treated up to 3headache episodes in a 3-month period with either theinvestigational TMS device (102 patients) or an iden-tically appearing sham device (99 patients). The inci-dence of adverse events was low and comparablebetween the TMS and sham groups; the primaryadverse events were dizziness, increased nausea, itchyor tingly sensation, and increased headache.20 Noserious adverse events were reported.

Studies evaluating rTMS as treatment for depres-sion, chronic pain, or other neurologic conditionshave likewise reported few adverse events.25,26 In arandomized study in which 301 patients were treated

with rTMS (10 hertz, 3000 pulses per session, 5 daysper week for 3-4 weeks) at an intensity of 120% ofmotor threshold (155 patients) or with sham (146patients), the most commonly reported adverseevents were transient head discomfort (eg, eye pain),toothache, application site discomfort or pain, muscletwitching, skin pain, and headache.25 Of note, the inci-dence of headache was not significantly differentbetween rTMS and sham groups.

In a meta-analysis of 3 studies,26 which includedover 10,000 cumulative treatment sessions of 3000stimulations each at 120% of motor threshold in 268patients, transient headache and scalp irritation werethe most commonly reported adverse events;however, tolerance to these events was observed withrepeated rTMS treatments. No alterations in cogni-tive function or hearing threshold were observed.

Machii et al27 reviewed adverse event data from173 articles published between January 1998 andDecember 2003 and described the use of rTMS tononmotor areas of the brain in 3092 patients. In 7studies, a total of 160 patients applied rTMS to theoccipital area at intensities ranging from 80% to120% phosphene threshold. The patients received600-1500 stimulations at frequencies ranging from 1to 20 hertz in 6 of the 7 studies and 20 stimulations at4 hertz in 1 of the 7 studies. Adverse events in thesepatients included headache (reported incidence of2% to 25% across 5 of the 7 studies); neck ache(reported incidence of 8% to 20% in 3 of the 7studies); and drowsiness (reported in only 1 patient).All adverse events were reported to be mild in sever-ity and self-limited.

In another 5 studies that were reviewed byMachii et al,27 68 patients received from 4 to 1350rTMS applications to the occipital area at intensitiesof 60% to 70% of the maximal output of the stimu-lator. Undefined headache was the only adverse eventreported.

The subjective response to TMS treatments wasevaluated in 40 children, 20 normal boys and girls, and20 boys and girls with attention deficit disorder withhyperactivity who received a total of 200-350 sTMSstimuli (10% at stimulator maximum output; 80%above motor threshold) over a 60- to 90-minuteperiod.21 The children were asked to complete a ques-

1156 July 2010

tionnaire to rank order their TMS experience withordinary life encounters, such as attending a birthdayparty, playing a game, watching television, sitting in along car ride, going to the dentist, throwing up, andreceiving a shot at the doctor’s office. Mean rankorder placed TMS fourth: The children found TMSless enjoyable than watching television but moreenjoyable than a long car ride. These subjective datathus provide further evidence of the tolerability ofTMS.

One of the highest exposures to TMS wasreported in a study in healthy men who were exposedto 12,960 magnetic pulses for up to 3 days in 1 week(the equivalent of 38,880 magnetic pulses over 1week) as part of a study to examine the effects ofrTMS on sleep deprivation.28 Despite high exposures,no significant adverse events were reported. It wasconcluded that doses of up to 12,960 pulses per daywere safe and tolerable in healthy men.

Overall, the reported adverse events associatedwith sTMS or rTMS have been mild in severity, and noshort- or long-term sequelae of treatment with TMShave been reported in normal adult subjects.1,22 Thus,sTMS use is regarded as a safe, well-tolerated tech-nique both in normal subjects and in patients withneurological impairment, provided that relevantexclusion criteria (such as presence of metal anywherein the head excluding the mouth, skull defects, intrac-ardiac lines, and cardiac pacemakers) are respected.29

Potential Safety Concerns Associated with TMS.—Effects on brain tissues, cardiovascular system, pro-laction and cortisol levels, cognition and hearing aresummarized in Table 2.

Risk of Seizure.—The most significant safetyconcern with sTMS therapy has been the hypotheticalrisk of inducing an epileptic seizure. This concern hasled to the close monitoring of sTMS and rTMS tech-niques. It has been shown that TMS is able to selec-tively activate the epileptic focus (or foci) in patientswith medically intractable complex partial seizures.Considering these medically refractory epilepticswere in part deprived of their anticonvulsant medica-tion, the article’s authors express surprise that theywere only able to induce a clinical seizure in 1 patient.No drug was necessary to stop undue neuronal hyper-activity and no adverse effects were observed.30 No

adverse effects or sequelae have been identified withsTMS applied over the cerebellum in presumednormal subjects or in individuals with known neuro-logical disorders.13

Based on a literature review, Schrader et al31

reported that the crude risk of TMS-associatedseizure in patients with epilepsy ranged from 0.0% to2.8% for sTMS and from 0.0% to 3.6% for rTMS. Inall cases, the seizures were similar to each patient’stypical seizure and were without long-term adverseoutcomes. In all cases, doubt was expressed in theoriginal reports as to whether the seizures wereinduced by TMS or coincidental. Medically intrac-table epilepsy and lowering anti-epileptic drugs wereassociated with an increased incidence of seizure.

In a meta-analysis of 28 studies, involving over850 healthy infants and children and children with avariety of central nervous system diseases and disor-ders (including epilepsy),21 no seizures were reportedamong the children who were treated with sTMS orrTMS in any of the studies. No published or empiricalevidence was found to suggest that sTMS or rTMSwas associated with more than a minimal risk inchildren.

No reports of seizures in patients with subcorticallesions have been documented.32 Concurrent use ofagents that are known to lower the seizure threshold,such as tricyclic antidepressants or neuroleptics, hasbeen implicated as a potential factor for seizure.32

Additionally, individuals with stroke may have alower threshold for seizure activation after TMS.33

Risks to Attached or Implanted ElectronicEquipment.—The risks of sTMS to implanted orattached electronic devices have been evaluated bothempirically and theoretically by Cadwell.34 Undermost circumstances, the effects of the magnetic fieldon metal objects are of minimal concern. However,the use of magnetic stimulation in persons withcardiac pacemakers and or other implanted metallicobjects with electrically conductive properties is con-sidered a contraindication because of potential thatTMS could dislodge the objects or induce electricalcurrents.

Use of TMS has been shown to be safe in patientswith spinal cord stimulator pulse generatorsimplanted in the lower abdomen, both when the

Headache 1157

implanted generators were turned on and when theywere turned off,35 suggesting that the distance fromthese devices to the application of TMS at the head issufficient to negate the magnetic effects. No damagingstimuli to the brain or damage to the brain stimula-tion device was observed in patients with brain stimu-lators who received TMS; however, it was noted thatinadvertent administration of TMS directly over theimplanted deep brain stimulator could result inmalfunction.36

In a study of the effects of sTMS on vagus nervestimulator (VNS) devices, the TMS coil was set at its

highest intensity and held approximately 5 mm fromthe VNS wires to create an induced charge intensityof 3.3 mC/cm2/phase.37 The function of the pulse gen-erator was unaffected by sTMS; thus, sTMS use inpatients with nerve stimulators would not beexpected to result in nerve stimulation or damage.

Use in Pregnancy.—Repetitive transcranial mag-netic stimulation use during pregnancy has beendescribed in 2 case reports, one in which a womanwith acute onset of mood disorder received rTMSstimulation (5 hertz; 5 seconds on and 25 seconds offover 20 minutes for 14 days in a 3-week period) at 22

Table 2.—Safety Conclusions From Clinical Data Exploring Potential Side Effects Associated With TMS

Side effect Exposure Conclusion/result

Effects of electricalstimulation on braintissue

2 epilepsy patients received 50 hertz of directsubdural stimulation to anterior temporal lobewith 45 mC maximum charge per phase at57 mC/cm2 before resection of the temporallobe

Stimulation parameters exceeded therecommended limits for direct electricalstimulation; no pathology observed51

Effects of TMS on braintissue

2 medically intractable epilepsy patients received~2000 rTMS stimuli over anterotemporal lobe.Underwent temporal lobe resection 4 weekslater

No pathology attributable to TMS observed45

Effects of TMS on thecardiovascular system

10 healthy subjects received 35 to 50suprathreshold sTMS stimulations per sessionfor 2 sessions

No significant effect on mean blood pressure andpulse rate values47

13 healthy subjects received rTMS at 10 hertz No significant effect on blood pressure and pulserate33

Effects of TMS onprolactin and cortisollevels


Lack of an increase in serum prolactin or cortisollevels following TMS suggests no effect on thehypothalamic-limbic structures47

Effects of TMS oncognition andpsychomotor tests


Significant improvements from pretreatment(practice effect?) observed in the WechslerAdult Intelligence Scale. No significant changein cognitive or psychomotor tests47

50 healthy volunteers received between 20 and60 TMS pulses (mean, 38)

No immediate or delayed effects on verbalmemory, visual memory, language fluency,spatial attention, praxis, coordination or motorstrength48

10 patients with ischemic stroke received TMSstimulations

No effects posttreatment or at 5 to 13 monthslater48

Effects of TMS onhearing

TMS sound is below 140 dB Below impulse noise exposure limit – NationalInstitute for Occupational Health and Safety57

9 healthy volunteers exposed to multiple sessionsof sTMS at 132 dB32

No transient or permanent hearing loss49

68 patients with depression, chronic pain, orother neurological conditions received >10,000cumulative treatment sessions of 3000 pulseseach

No effect of rTMS on hearing threshold26

rTMS = repetitive transcranial magnetic stimulation; sTMS = single-pulse transcranial magnetic stimulation; TMS = transcranialmagnetic stimulation.

1158 July 2010

weeks gestation38 and one in which a woman withrecurrent major depressive disorder was treated with58 sessions of rTMS (25 hertz, 20 trains of 50 pulseseach per session) during the first (35 sessions), second(15 sessions), and third (8 sessions) trimesters of preg-nancy.39 No adverse events were reported for eitherthe mothers or the babies. One of the babies wasfollowed through 22 months of age; both physicaldevelopment and neurological development werenormal.39

The Committee on the Possible Effects of Elec-tromagnetic Fields on Biologic Systems, a committeeof the National Research Council, reviewed expo-sures to electric and magnetic fields (see Table 3) andconcluded that reproduction and development inanimals, particularly mammals, have not been shownto be affected by exposure to extremely low fre-quency electric or magnetic fields.40 Further, the com-mittee concluded that epidemiologic evidence of anassociation between magnetic fields and pregnancyoutcome is not supported.41

The investigational sTMS device that was used totreat migraine in clinical studies19,20 delivers a peakmagnetic field strength of 9000 Gauss when measured0.4 inches (1 cm) from the device surface. The fielddecreases rapidly with distance from the coil. At 18inches (46 cm) below the device, which wouldapproximately correspond to the sternum or thenearest point of a full term pregnant uterus to thedevice, the peak field is less than 1 Gauss. Treatmentconsists of 2 individual pulses. A 30-second delayseparates the 1/1000-second duration pulses, resultingin an exposure time of less than 2/1000 of a second.

The frequency of the sTMS device meets the defini-tion of very low frequency (ie, <100,000 hertz).24

DISCUSSION AND CONCLUSIONSMigraine is a prevalent, often debilitating disease

that typically manifests with moderate or severeintensity headaches and associated symptoms. Themajority of patients with migraine headaches willseek a physician’s care, and most of these patients willbe prescribed some type of drug therapy.42 Currently,the serotonin 5-hydroxytryptamine agonists (ie, “trip-tans”) represent the primary agents used for acutetreatment of migraine headaches. Although the intro-duction of triptans in the 1990s resulted in improvedoutcomes over those obtained with over-the-counteranalgesics, ergot alkaloids, nonsteroidal anti-inflammatory agents, and opioid analgesics, it is esti-mated that as many as 40% of patients do notadequately respond to or tolerate the side effects oftriptans, or have contraindications which precludestheir use.43 Subcutaneous triptans have been shownnot to be effective when administered during the auraphase of a migraine attack, and to date, no treatmentis reliably effective in terminating the aura phase orpreventing the ensuing headache.44 A significant needexists for safe, effective, nonpharmacologic therapeu-tic alternatives for patients with migraine headacheswith and without aura. Recent evidence from a ran-domized, sham-controlled clinical trial demonstratedsTMS to be effective for the acute treatment ofmigraine, administered during the aura phase ofattacks.20

Table 3.—Magnetic Fields in Everyday Life

Type of magnetic field Magnitude of exposure

Background magnetic field from earth Static magnetic field of ~0.5 GaussMagnetic resonance imaging ~20,000 GaussUse of 30 to 60 hertz electricity in homes and offices Time varying magnetic fields of ~0.003 GaussHome appliance use Intermittent exposure of �1 GaussHair dryer 0.7 Gauss at 6 inches from scalpElectric mixer 0.6 Gauss at 6 inchesMicrowave oven 0.3 Gauss at 6 inchesElectric powered subway or street car Peak field of 0.6 Gauss; average field of 0.1 Gauss while seatedAirport screening metal detector Maximum of 3.7 Gauss at a frequency of 100 to 3500 hertz

Headache 1159

Transcranial magnetic stimulation has been rou-tinely used to measure corticomotor excitation andinhibition of neurophysiologic properties in consciouspatients and has been shown to be a safe techniquewith promise in the diagnosis, monitoring, and treat-ment of neurological and psychiatric disease inadults.1,2

Two decades of clinical experience have shownthat TMS is a harmless technique provided exclusioncriteria (presence of metal anywhere in the headexcluding the mouth, skull defects, intracardiac lines,cardiac pacemakers) are respected.29 The most com-monly reported adverse events associated with TMShave been discomfort at the scalp and headache.18-20,22

Single-pulse TMS has been shown to have no delete-rious effects on brain tissue,22,45 blood pressure33,46,47 orcardiac rhythm,46 prolactin or cortisol levels,47 cogni-tion or psychomotor tests or electroencephalo-grams,47,48 or hearing.26,49 No significant adverse eventswere reported when healthy men were exposed todoses of up to 12,960 TMS pulses per day.28

The authors note that most of the referencedTMS safety studies have been conducted on subjectswith unknown migraine status. Based on clinicalstudies showing that sTMS is a well-tolerated treat-ment for migraine with or without aura,18-20 theauthors argue that migraneurs are no more likely tobe susceptible to adverse events than other groupsstudied. However, more data are desirable.

The most significant safety concern with sTMStherapy has been the hypothetical risk of inducingepileptic seizure, and this concern has led to the closemonitoring of sTMS and rTMS techniques for relatedsafety issues. No seizures secondary to sTMS havebeen reported in healthy subjects or in patients withneurological disorders when applied over the occipi-tal cortex,21 and no reports of seizure after sTMS havebeen documented in patients with subcorticallesions.32 Concurrent use of tricyclic antidepressantsand neuroleptics has been implicated in seizure risk.32

Given these findings, sTMS should be administeredwith caution in patients who are taking tricyclic anti-depressants or neuroleptics. As seizures have beenreported in patients with epilepsy who have beentreated with TMS,31 sTMS should not be used inpatients who have epilepsy and should be used with

caution in patients who have a prevalent familyhistory of epilepsy. Use of the Transmagnetic AdultSafety Screen50 or a similar standard questionnaire toidentify patients with a history of head trauma, sei-zures, neurological, or medical illnesses or a familyhistory of epilepsy should also be considered toensure that patients with such a history do not receivetreatment with TMS.

Under most circumstances, the effect of the sTMSmagnetic field on metal objects is of minimalconcern.34-37 However, to ensure the highest level ofsafety, sTMS should not be used by patients withmetal or conductive materials or implants in theirhead, neck, or upper body, excluding the mouth, or bypatients with cardiac pacemakers, VNS or otherimplanted neurostimulators, implanted cardioverterdefibrillators, or any implanted medical device thatuses a signal from the body.

Experience with TMS during pregnancy islimited. Based on the available information from 2cases, rTMS had no effect on either the mothers orthe babies, and TMS had positive outcomes on thedepression in both mothers.38,39 The NationalResearch Council found no affect of extremely lowfrequency magnetic fields on pregnancy outcome.50

Additional data are needed to fully assess the effectsof sTMS in pregnant women.

The safety of sTMS in clinical practice, includingas a treatment for migraine headaches, is supportedby both biological and empirical evidence. Single-pulse TMS may offer a safe nonpharmacologic, non-behavioral therapeutic approach to the currentlyprescribed drugs for patients who suffer frommigraine.

Acknowledgment: The authors thank Lillian L. Neff

for writing and editorial assistance.

REFERENCES

1. Anand S, Hotson J. Transcranial magnetic stimula-tion: Neurophysiological applications and safety.Brain Cogn. 2002;50:366-386.

2. Rossini PM, Rossi S. Transcranial magnetic stimula-tion: Diagnostic, therapeutic, and research potential.Neurology. 2007;68:484-488.

3. Barker AT, Freeston IL, Jalinous R, Jarratt JA. Mag-netic stimulation of the human brain and peripheral

1160 July 2010

nervous system: An introduction and the results ofan initial clinical evaluation. Neurosurgery. 1987;20:100-109.

4. Barker AT, Jalinous R, Freeston IL. Non-invasivemagnetic stimulation of human motor cortex.Lancet. 1985;1:1106-1107.

5. Maertens de Noordhout AL, Pepin JL, Schoenen J,Delwaide PJ. Percutaneous magnetic stimulation ofthe motor cortex in migraine. ElectroencephalogrClin Neurophysiol. 1992;85:110-115.

6. Bettucci D, Cantello R, Gianelli M, Naldi P, MutaniR. Menstrual migraine without aura: Cortical excit-ability to magnetic stimulation. Headache. 1992;32:345-347.

7. Aggugia M, Zibetti M, Febbraro A, Mutani R. Tran-scranial magnetic stimulation in migraine with aura:Further evidence of occipital cortex hyperexcitabil-ity. Cephalalgia. 1999;19:465.

8. Aurora SK, Cao Y, Bowyer SM, Welch KMA. Theoccipital cortex is hyperexcitable in migraine; evi-dence from TMS, fMRI and MEG studies. Head-ache. 1999;39:469-476.

9. Lipton RB, Bigal ME, Diamond M, Freitag F, ReedML, Stewart WF, AMPP Advisory Group. Migraineprevalence, disease burden, and the need for preven-tive therapy. Neurology. 2007;68:343-349.

10. Goadsby PJ, Lipton RB, Ferrari MD. Migraine –current understanding and treatment. N Engl J Med.2002;346:257-270.

11. Afridi SK, Giffin NJ, Kaube H, et al. A positronemission tomographic study in spontaneousmigraine. Arch Neurol. 2005;62:1270-1275.

12. Aurora SK, Ahmad BK, Welch KM, Bhardhwaj P,Ramadan NM. Transcranial magnetic stimulationconfirms hyperexcitability of occipital cortex inmigraine. Neurology. 1998;50:111-114.

13. Aurora SK, Cao Y, Bowyer SM, Welch KM. Theoccipital cortex is hyperexcitable in migraine:Experimental evidence. Headache. 1999;39:469-476.

14. Reilly CE. Cortical spreading depression is linked topain transmission by the trigeminal nerve. J Neurol.2002;249:643-645.

15. Bolay H, Reuter U, Dunn AK, Huang Z, Boas DA,Moskowitz MA. Intrinsic brain activity triggerstrigeminal meningeal afferents in a migraine model.Nat Med. 2002;8:136-142.

16. Ayata C, Jin H, Kudo C, Dalkara T, Moskowitz MA.Suppression of cortical spreading depression inmigraine prophylaxis. Ann Neurol. 2006;59:652-661.

17. Holland PR, Schembri CT, Fredrick JP, Goadsby PJ.Transcranial magnetic stimulation for the treatmentof migraine aura. Proceedings of the 61st AnnualMeeting of the Academy of Neurology; 2009 April25; Seattle, WA.

18. Clarke BM, Upton AR, Kamath MV, Al-Harbi T,Castellanos CM. Transcranial magnetic stimulationfor migraine: Clinical effects. J Headache Pain.2006;7:341-346.

19. Mohammad YM, Kothari R, Hughes G, et al. Tran-scranial magnetic stimulation (TMS) relievesmigraine headache. Proceedings of the 48th AnnualScientific Meeting of the American HeadacheSociety; 2006 June 23; Los Angeles, CA.

20. Lipton RB, Dodick DW, Goadsby PJ, et al. Single-pulse transcranial magnetic stimulation for acutetreatment of migraine with aura: A randomised,double-blind, parallel-group, sham-controlled trial.Lancet. 2010;9:373-380.

21. Gilbert D, Garvey MA, Bansal AS, Lipps T, ZhangJ, Wasserman EM. Should transcranial magneticstimulation research in children be consideredminimal risk? Clin Neurophysiol. 2004;115:1730-1739.

22. Wassermann E. Risk and safety of repetitive transc-ranial magnetic stimulation: Report and suggestedguidelines from the International Workshop on theSafety of Repetitive Transcranial Magnetic Stimula-tion, June 5-7, 1996. Electroencephalogr Clin Neuro-physiol. 1998;108:1-16.

23. Committee on the Possible Effects of Electromag-netic Fields on Biologic Systems, National ResearchCouncil. Possible Health Effects of Exposure to Resi-dential Electric and Magnetic Fields. Washington,DC: National Academies Press; 1997:12.

24. International Commission on Non-IonizingRadiation Protection. Guidelines on limits of expo-sure to static magnetic field. Health Phys. 2009;96:504-514.

25. O’Reardon JP, Solvason HB, Janicak PG, et al. Effi-cacy and safety of transcranial magnetic stimulationin the acute treatment of major depression: A mul-tisite randomized controlled trial. Biol Psychiatry.2007;62:1208-1216.

26. Janicak PG, O’Reardon JP, Sampson SM, et al.Transcranial magnetic stimulation in the treatmentof major depressive disorder: A comprehensivesummary of safety experience from acute exposure,and during reintroduction treatment. J Clin Psychia-try. 2008;69:222-232.

Headache 1161

27. Machii K, Cohen D, Ramos-Estebanez R, Pascual-Leone A. Safety of rTMS to non-motor corticalareas in healthy participants and patients. Clin Neu-rophysiol. 2006;117:455-471.

28. Anderson B, Mishory A, Nahas Z, et al. Tolerabilityand safety of high daily doses of repetitive transcra-nial magnetic stimulation in healthy young men. JECT. 2006;22:49-53.

29. Tassinari A, Cincotta M, Zaccara G, Michelucci R.Transcranial magnetic stimulation and epilepsy. ClinNeurophysiol. 2003;114:777-798.

30. Hufnagel A, Elger CE, Durwen HF, Böker DK,Entzian W. Activation of epileptic focus by transc-ranial magnetic stimulation of the human brain. AnnNeurol. 1990;27:49-60.

31. Schrader LM, Stern JM, Koski L, Nuwer MR, EngelJ Jr. Seizure incidence during single-and paired-pulse transcranial magnetic stimulation (TMS) inindividuals with epilepsy. Clin Neurophysiol. 2004;115:2728-2737.

32. Rosa MA, Odebrecht M, Rigonatti SP, MarcolinMA. Transcranial magnetic stimulation: Review ofaccidental seizures. Rev Bras Psiquiatr. 2004;26:131-134.

33. Foerster A, Schmitz JM, Nouri S, Claus D. Safety ofrapid rate transcranial magnetic stimulation: Heartrate and blood pressure changes. Electroencepha-logr Clin Neurophysiol. 1997;104:207-212.

34. Cadwell J. Principles of magnetoelectric stimula-tion. In: Chokroverty S, ed. Magnetic Stimulation inClinical Neurophysiology. Boston, MA: ButterworthHeinemann; 1990:26-32.

35. Kofler M, Leis A, Sherwood AM, Delapasse JS,Halter JA. Safety of Transcranial magnetic stimula-tion in patients with abdominal implanted devices.Lancet. 1991;338:1275-1276.

36. Wassermann EM, Lisanby SH. Therapeutic applica-tion of transcranial magnetic stimulation: A review.Clin Neurophysiol. 2001;112:1367-1377.

37. Schrader LM, Stern JM, Fields TA, Nuwer MR,Wilson CL. A lack of effect from transcranial mag-netic stimulation (TMS) on the vagus nerve stimu-lator (VNS). Clin Neurophysiol. 2005;116:2501-2504.

38. Nahas Z, Bohning DE, Molloy MA, Oustz JA, RischSC, George MS. Safety and feasibility of repetitivetranscranial magnetic stimulation in the treatmentof anxious depression in pregnancy: A case report.J Clin Psychiatry. 1999;60:50-52.

39. Tan O, Tarhan N, Coban A, et al. Antidepressanteffect of 58 sessions of rTMS in a pregnant woman

with recurrent major depressive disorder: A casereport. Prim Care Companion. J Clin Psychiatry.2008;10:69-71.



42. Lipton RB, Stewart WF. Prevalence and impact ofmigraine. Neurol Clin. 1997;15:1-13.

43. Ferrari MD, Goadsby PJ, Roon KI, Lipton RB. Trip-tans (serotonin, 5-HT1B/1D agonists) in migraine:Detailed results and methods of a meta-analysis of53 trials. Cephalalgia. 2002;22:633-658.

44. Bates D, Ashford E, Dawson R, et al. Subcutaneoussumatriptan during the migraine aura. SumatriptanAura Study Group. Neurology. 1994;44:1587-1592.

45. Gates JR, Dhuna A, Pasual-Leone A. Lack ofpathologic changes in human temporal lobes aftertranscranial magnetic stimulation. Epilepsia. 1992;3:504-508.

46. Bridgers SL. The safety of transcranial magneticstimulation reconsidered: Evidence regarding cogni-tive and other cerebral effects. ElectroencephalogrClin Neurophysiol Suppl. 1991;43:170-179.

47. Chokroverty S, Hening W, Wright D, et al. Magneticbrain stimulation: Safety studies. ElectroencephalogrClin Neurophysiol. 1995;97:36-42.

48. Bridgers SL, Delaney RC. Transmagnetic stimula-tion: An assessment of cognitive and other cerebraleffects. Neurology. 1989;39:417-419.

49. Pascual-Leone A, Houser CM, Reese K, et al. Safetyof rapid-rate transcranial magnetic stimulation innormal volunteers. Electroencephalogr Clin Neuro-physiol. 1993;89:120-130.

50. Keel JC, Smith MJ, Wassermann EM. A safetyscreening questionnaire for transcranial magneticstimulation. Clin Neurophysiol. 2001;112:720.

51. Agnew WF, McCreery DB. Considerations forsafety in the use of extracranial stimulation formotor evoked potentials. Neurosurgery. 1987;20:143-147.

52. Agnew WF, McCreery DB. Considerations forsafety with chronically implanted nerve electrodes.Epilepsia. 1990;31(Suppl 2):S27-S32.

1162 July 2010

53. Jalinous R. Guide to Magnetic Stimulation. Newton,MA: Jali Medical Inc; 1998.

54. McCreery DB, Agnew WF, Yuen TG, Bullara L.Relationship between stimulus amplitude, stimulusfrequency and neural damage during electricalstimulation of sciatic nerve of cat. Med Biol EngComput. 1995;33:426-429.

55. Liebetanz D, Fauser S, Michaelis T, et al. Safetyaspects of chronic low-frequency transcranial mag-netic stimulation based on localized proton mag-netic resonance. J Psychiatr Res. 2003;37:277-286.

56. Yamada H, Tamaki T, Wakano K, Mikami A,Transfeldt EE. Effect of transcranial magneticstimulation on cerebral function in a monkey model.Electroencephalogr Clin Neurophysiol. 1995;97:140-144.

57. National Institute for Occupational Health andSafety (NIOHS). Criteria for recommended stan-dard: Occupational noise exposure (NIOS Publica-tion 98-126) [online]. Available from URL: http://www.cdc.gov/niosh/docs/98-126/ (accessed May 21,2010).

Headache 1163

transcranial magnetic stimulation for migraine: a safety review

Documents