Repetitive Transcranial Magnetic Stimulation in Pain Management: A Narrative Review of Randomized Controlled Trials.

Introduction

Transcranial magnetic stimulation (TMS) is a noninvasive neuromodulation technique that uses electromagnetic induction to generate electric fields in the brain.1,2 TMS can be delivered using different paradigms, including single-pulse TMS (a single stimulus delivered at a time), paired-pulse TMS (2 stimuli delivered with a defined interstimulus interval), and repetitive TMS (rTMS, in which stimuli are delivered in trains). Among these approaches, rTMS has been widely applied for therapeutic purposes because it can induce protocol-dependent changes in cortical excitability, producing either facilitatory or inhibitory effects.2

rTMS was initially developed and investigated as a therapeutic intervention for depression and subsequently received approval from the US Food and Drug Administration (FDA) in 2008 for the treatment of major depressive disorder (MDD).3 Its clinical applications have since expanded to a range of other conditions, including schizophrenia, tinnitus, and auditory hallucinations.4 Notably, rTMS has also been investigated across diverse pain conditions, suggesting its potential as a noninvasive neuromodulatory strategy for pain management. However, rTMS remains an off-label treatment for pain indications, and the current body of evidence is heterogeneous.

Chronic pain is driven by multiple interacting mechanisms, including complex alterations in nociceptive processing such as peripheral and central sensitization and dysfunction of descending pain-modulatory pathways. Peripheral sensitization is characterized by increased responsiveness and reduced thresholds of peripheral nociceptors, whereas central sensitization involves increased responsiveness of nociceptive neurons within the central nervous system to normal or subthreshold afferent input.5,6 Together, these processes contribute to the amplification and persistence of pain.6–8 In parallel, dysfunction of descending pain-modulatory pathways, particularly reduced inhibitory control, may further impair endogenous pain regulation and therefore play a role in chronic pain maintenance.9 These alterations involve distributed cortical and subcortical pain networks, providing a neurophysiological rationale for targeting these networks through neuromodulatory approaches, including rTMS. Although the analgesic mechanisms of rTMS are not fully understood, proposed mechanisms involve modulation of neural network activity through regulation of neurotransmitter levels, alteration of synaptic plasticity, regulation of the excitability of the pain loop, and modulation of neurotrophic, neuroprotective, and glial processes, ultimately contributing to analgesia.10

The therapeutic efficacy of rTMS depends on multiple interrelated factors, such as the stimulation protocol, stimulation site, frequency, intensity, other specific stimulation parameters of trains, coil type, the overall treatment course, and individual patient characteristics. Based on the stimulation pattern, rTMS can be classified into conventional and patterned protocols.2 The conventional protocol delivers pulses at a constant frequency and is typically classified as high-frequency rTMS (HF-rTMS, ≥5 Hz), which consists of bursts of stimuli typically lasting 5–10 seconds, separated by pauses of 20–50 seconds and is generally associated with facilitatory effects on cortical excitability, or low-frequency rTMS (LF-rTMS, ≤1 Hz), which consists of continuous trains of single pulses and is typically associated with inhibitory effects.11 The patterned protocol includes several protocols such as theta burst stimulation (TBS), quadripulse magnetic stimulation (QPS), and paired associative stimulation (PAS).2 Theta burst stimulation (TBS) is a patterned rTMS protocol that has been widely investigated in clinical research, in which bursts of high-frequency pulses (typically 50 Hz) are delivered at a theta frequency of 5 Hz; it is commonly classified into intermittent TBS (iTBS), consisting of 2-second trains repeated every 10seconds for a total duration of 200 seconds, and continuous TBS (cTBS), delivered uninterrupted for 40 seconds, although variations of these protocols have been reported across studies.11 Although iTBS is generally considered facilitatory and cTBS inhibitory, some evidence suggests that these effects are not absolute.11 Despite its short stimulation duration, TBS may produce larger effect sizes with less interindividual variability compared with conventional rTMS, making it a promising neuromodulatory technique.12 rTMS parameters were applied in accordance with the international safety guidelines for TMS, originally established by Wassermann et al in 1998, subsequently updated by Rossi et al in 2009, and further updated in 2021. With respect to stimulation site, the primary motor cortex (M1) is the most commonly targeted region in pain management, although the dorsolateral prefrontal cortex (DLPFC) and other cortical areas may also be stimulated depending on the specific pain condition.

Coil design determines the focality and depth of stimulation, which may influence therapeutic efficacy, with figure-of-eight coils most commonly used, alongside alternative designs such as circular and double-cone coils. To increase stimulation depth, deep rTMS (dTMS) using H-coils has been developed.2

To date, rTMS is generally considered a safe and well-tolerated intervention.2,11,13 The commonly reported adverse events are headache, stimulation-site discomfort, and dizziness.13,14 rTMS is unlikely to be associated with persistent perceptual or cognitive adverse effects, although transient cognitive changes may occur.13 Mood-related or psychiatric changes, including anxiety, agitation, mania, and hypomania, have also occasionally been reported, particularly in psychiatric populations.13,15 Hearing safety is another important consideration because the acoustic output generated by the coil may pose a risk to auditory function without appropriate hearing protection.2,13 Although these adverse effects are generally mild and transient, they may still reduce treatment tolerability and lead to discontinuation. Seizures are rare but represent the most serious adverse event.2,11 The risk of seizure depends on stimulation parameters, including intensity and frequency, as well as the patient’s underlying brain state.11 Theta burst stimulation (TBS) may theoretically carry a higher seizure risk than conventional rTMS protocols because it delivers high-frequency bursts of stimulation.16 To minimize seizure risk, stimulation should adhere strictly to established safety guidelines, and medications known to lower the seizure threshold should be avoided or discontinued.11

Current evidence suggests that rTMS has been investigated across a range of chronic pain conditions, with systematic reviews and meta-analyses focusing on some disorders such as neuropathic pain (NP), fibromyalgia (FM), musculoskeletal pain, and migraine.17–21 However, the literature remains fragmented, with both primary studies and evidence syntheses concentrated on selected pain conditions, while substantial heterogeneity persists in stimulation targets, parameters, coil types, treatment schedules, and comparators. Given the largely off-label use of rTMS for chronic pain and the absence of standardized treatment protocols, an updated overview of the evidence is warranted. This narrative review focuses on randomized controlled trials (RCTs) identified through February 2026 in PubMed to map the distribution of evidence across pain conditions, summarize key methodological and clinical considerations, and identify priorities for future research.

Methods

A literature search was conducted in PubMed from database inception to February 12, 2026. The search strategy used the following Boolean query: (“repetitive transcranial magnetic stimulation” OR rTMS OR “theta burst stimulation” OR TBS OR cTBS OR iTBS) AND (pain OR analgesia), with truncation (random*).

Eligibility criteria were defined according to the PICOS framework. Population (P): patients with diagnosed pain conditions; Intervention (I): rTMS used as a therapeutic intervention; Comparator (C): sham stimulation, other therapeutic interventions, or alternative rTMS strategies, when applicable; Outcomes (O): pain-related and condition-specific clinical outcomes; Study design (S): randomized controlled trials. We also included only studies published in English.

Exclusion criteria were: (1) case reports, study protocols, and conference presentations; (2) animal studies.

All identified records were imported into EndNote X9 and independently screened according to the predefined eligibility criteria. Titles and abstracts were initially screened, after which potentially eligible studies underwent full-text assessment.

The methodological quality of the included RCTs was assessed using the Jadad scale, which assigns up to 5 points: 2 for randomization, 2 for blinding, and 1 for reporting withdrawals and dropouts. One point is awarded for reporting randomization or blinding, with an additional point for an appropriate method; one point is deducted if the described method is inappropriate.22 One point is awarded when the number and reasons for withdrawals or dropouts are reported.

Clinical Applications of Repetitive Transcranial Magnetic Stimulation in Pain Management

Our search yielded 735 records. Following eligibility assessment, 134 studies met the inclusion criteria and were categorized into 8 groups: NP, chronic widespread pain, headache, orofacial pain, musculoskeletal pain, postsurgical pain, cancer-related pain, and other conditions. Because most included studies focused on chronic pain, the grouping was informed by the IASP chronic pain framework but adapted for clinical interpretation, with each study assigned to a single category according to the primary pain condition investigated.23 Jadad scores varied across the included RCTs, mainly reflecting differences in the reporting of randomization and blinding.

The main characteristics of the included studies are summarized in Tables 1–5.

Neuropathic Pain

NP is defined as pain caused by a lesion or disease of the somatosensory nervous system.24 NP management is generally initiated with conservative pharmacologic and complementary therapies; however, treatment response is frequently incomplete and often inadequate.25,26 Consequently, noninvasive neuromodulation has gained increasing attention as an adjunctive therapeutic strategy, particularly for patients with refractory NP. In this context, rTMS has been widely investigated for its potential analgesic effects across diverse NP conditions. In this review, 52 studies were identified, including investigations in patients with NP of various etiologies.

These studies encompassed both peripheral NP and central NP conditions. In addition, several studies examined NP populations without restricting the etiology to a specific underlying condition. The included studies in this section are presented in Table 1.

Table 1Characteristics of Studies Evaluating Repetitive Magnetic Stimulation (rTMS) for Neuropathic PainNo.Author, YearStudy Design/ Group (n)Stimulation SiteCoil Type of Active rTMSActive rTMS ParametersSham ProcedureTreatment CourseFollow -UpSummary of FindingsSafety, n (Active/Sham)Jadad ScoreNonspecified Neuropathic Pain1Hosomi et al27 2013- Crossover- Active rTMS, sham (n = 64)M1Figure-of-eight coil− 5 Hz; 10 trains/session, 50 pulses/train, ITI 50s, total 500 pulses/session- 90% RMTInactive coil + synchronized scalp electrical stimulation and simulated coil sounds10 daily sessions (weekdays) per condition, ≥17-day washout17 days post-treatmentActive rTMS showed significant short-term pain relief compared with sham, but no significant cumulative benefits- Deterioration of squeezing (2/2), headache (0/2), deterioration of numbness (1/1), hypoglycemia (1/0), anxiety (0/1), appetite loss (0/1), and dizziness (0/1)- No serious adverse events5/52Hosomi et al28 2020- Parallel- Active rTMS (n = 72), sham (n = 70)M1Figure-of-eight coil− 5 Hz; 10 trains/session, 50 pulses/train, ITI 50s, total 500 pulses/session- 90% RMTSham coil + electrical scalp stimulation5 daily sessions (days 1–5), responders entered open-label continuous trial: ≥1 session/week for 4 weeks4 weeks post-treatment5 daily sessions: Active rTMS group showed no significant difference in pain relief compared with sham groupWeekly rTMS in responders: Active rTMS showed significantly greater pain relief than the sham group- Pain in extremity (2/7), headache (2/4), hypoaesthesia (2/2), skin pain (1/3), dizziness (2/1), musculoskeletal pain (1/2), excoriation (1/2)- No serious adverse events; adverse event-related withdrawals (0/2)5/53Lefaucheur et al29 2001- Crossover- Active rTMS, sham (n = 14)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s- 80% RMTDedicated placebo figure-of-eight coilSingle session per condition, ≥3-week washoutDaily, days 1–12Active rTMS significantly reduced VAS pain scores for up to 8 days compared with shamNo adverse effects were observed immediately or in the days following rTMS; no seizures occurred1/54Borckardt et al30 2009- Crossover- Active rTMS, sham (n = 4)Left PFCFigure-of-eight coil− 10 Hz; 10s on/20s off for 20 min, total 4000 pulses/session- 100% RMTShielded sham coil3 sessions per phase within 5 days, phases separated by 3 weeks3 weeks after each treatment phaseActive rTMS was associated with a 19% reduction in daily average pain lasting approximately 2 weeks in all patients and produced significantly greater pain relief than sham stimulation in 3 patientsNot explicitly reported2/55Mori et al31 2024- Parallel- Active rTMS (n = 14), sham (n = 16)M1Figure-of-eight coil− 10 Hz; 40 trains/session, 50 pulses/train, ITI 25s, total 2000 pulses/session- 90% RMT; if RMT was not elicitable or >86% MSO, stimulation intensity was fixed at 77% MSODedicated sham figure-of-eight coil mimicking the appearance and sound of active stimulation, combined with synchronized electrical scalp stimulation- Induction: 5 daily sessions- Maintenance: 1 session per week for 7 weeks- Total: 12 sessionsDaily during treatment; 1, 2, 4 and 8 weeks post-treatmentPain relief did not differ significantly between active rTMS and sham; however, active rTMS tended to provide better results than sham- Fever (1/0), scalp pain (0/1), dizziness (0/1), worsening of pain (0/1)- No serious adverse events5/56Khedr et al32 2005- Parallel- Active rTMS (n = 28), sham (n = 20)M1Figure-of-eight coil− 20 Hz; 200 pulses/train, total 2000 pulses/session- 80% RMTActive figure-of-eight coil tilted and elevated away from the scalp to avoid cortical stimulation5 daily sessions over 5 consecutive daysImmediately after the 1st, 4th, and 5th sessions, and 15 days after the final sessionActive rTMS significantly reduced pain scores compared with sham; this effect persisted up to 2 weeks after the final sessionNo adverse effects reported1/57André-Obadia et al33 2014- Crossover- Active rTMS, sham (n = 20)M1Figure-of-eight coil− 20 Hz; 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session- 90% RMTDedicated placebo figure-of-eight coilSingle session per condition, ≥2-week washoutDaily pain assessment for 2 weeks between rTMS sessionsActive rTMS significantly reduced pain; sham did notNot explicitly reported for rTMS1/58Quesada et al34 2020- Crossover- Active rTMS, sham (n = 42)M1Figure-of-eight coil− 20 Hz; 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session- 80% RMT; if RMT is not obtainable, intensity set to 45% MSODouble-sided active-placebo-figure-of-eight coil, providing identical positioning and acoustic stimulation4 sessions/phase, sessions spaced 3 weeks, 8-week washoutInclusion session, 3 weeks after the last rTMSHigh-frequency rTMS of M1 administered at 3-week intervals produces a sustained analgesic effect.- Active: 7 patients; sham: 8 patients- One patient withdrew due to pain exacerbation during active rTMS- No serious adverse events were reported5/59Lefaucheur et al35 2001- Crossover- 0.5-Hz rTMS, 10-Hz rTMS, sham (n = 18)M1Figure-of-eight coil− 0.5 Hz: stimulation for 20 min- 10 Hz: 20 trains/session, 5s/train, ITI 55s- 80% RMTDedicated figure-of-eight placebo coilSingle session per condition, ≥3-week washoutImmediate post-stimulation assessmentOnly 10-Hz rTMS produced significant pain reduction; 0.5-Hz and sham did notNo adverse effects reported, including no seizures1/510André-Obadia et al36 2006- Crossover- 1-Hz rTMS, 20-Hz rTMS, sham (n = 12)M1Figure-of-eight coil− 1 Hz: stimulation for 26 min, total 1600 pulses/session- 20 Hz: 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session - 90% RMTActive coil positioned perpendicular to and separated from the scalpSingle session per condition, 2-week washoutDaily pain assessment for 6 days after each sessionAll rTMS modalities produced immediate pain reduction; however, over the following week, 1-Hz rTMS was significantly less effective than 20- Hz and sham, and even increased pain in some patientsNo adverse effects observed during or after rTMS4/511Saitoh et al37 2007- Crossover- 1-Hz rTMS, 5-Hz rTMS, 10-Hz rTMS, sham (n = 13)M1Figure-of-eight coil− 1 Hz: stimulation for 500s, total 500 pulses/session- 5 Hz: 10 trains/session, 10s/train, ITI 50s, total 500 pulses/session- 10 Hz: 5 trains/session, 10s/train, ITI 50s, total 500 pulses/session- 90% RMT; If RMT could not be elicited due to severe motor pathway damage: 100 A/µsActive coil tilted 45° from the scalp, combined with synchronized electrical forehead stimulationSingle session per condition, 48-h intervalImmediately and at 15, 30, 60, 90, and 180 minutes after stimulationCompared with sham, 5-Hz and 10 Hz rTMS significantly reduced pain for up to 180 min, with 10 Hz tending to be more effective than 5 Hz; 1 Hz was ineffectiveNo transient or persistent adverse effects, including seizures3/512Lefaucheur et al38 2008- Crossover- 1-Hz rTMS, 10-Hz rTMS, sham (n = 46)M1Figure-of-eight coil− 1 Hz: stimulation for 20 min, total 1200 pulses/session- 10 Hz: 20 trains/session, 6s/train, ITI 54s, total 1200 pulses/session- 90% RMTDedicated sham coilSingle session per condition, ≥3-week washout15 minutes before and after each sessionPain scores and thermal thresholds were significantly reduced after 10-Hz rTMS but not after 1-Hz or shamNot explicitly reported3/513Mori et al39 2022- Crossover- 5-Hz 500-pulse rTMS, 10-Hz 500-pulse rTMS, 10-Hz 2000-pulse rTMS, sham (n = 22)M1Figure-of-eight coil− 5 Hz 500 pulses: 10 trains/session, 10s/train, ITI 50s- 10 Hz 500 pulses: 10 trains/session, 5s/train, ITI 25s- 10 Hz 2000 pulses: 40 trains/session, 5s/train, ITI 25s- 90% RMTDedicated sham coilSingle session for each condition, ≥2-week washout between sessionsImmediately after each sessionHigher-intensity protocols, particularly 10-Hz rTMS delivering 2000 pulses, were more effective than lower-dose stimulationTransient scalp pain in 2 patients after active rTMS; no lasting adverse effects or convulsions4/514Hirayama et al40 2006- Crossover- M1 rTMS, S1 rTMS, PMC rTMS, SMA rTMS, sham (n = 20)M1, S1, PMC, SMAFigure-of-eight coil− 5 Hz; 10 trains/session, 10s/train, ITI 50s, total 500 pulses/session (per target)- 90% RMT; If RMT could not be elicited: 100 A/µsActive coil tilted 45° from the scalp, combined with synchronized electrical forehead stimulationSingle session per target, 48-h intervalImmediately and up to 180 minutes after stimulationOnly M1 rTMS significantly reduced pain for up to 3 h; other targets were not effectiveNo transient or persistent adverse effects, including convulsions3/515André-Obadia et al41 2008- Crossover- PA-oriented rTMS, LM-oriented rTMS, sham- PA rTMS, sham: n = 28- PA rTMS, LM rTMS, sham: n = 16M1Figure-of-eight coil− 20 Hz; 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session - 90% RMTActive coil positioned perpendicular to and separated from the scalpSingle-session rTMS per condition, ≥2-week washoutDaily pain assessment after each sessionPA-oriented rTMS was more effective than both LM-oriented rTMS and sham, with analgesic effects lasting approximately 1 weekNo adverse effects reported during or after rTMS4/516André-Obadia et al42 2018- Crossover- Hand M1 rTMS, face M1 rTMS, sham (n = 32)M1(hand area, face area)Figure-of-eight coil− 20 Hz; 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session- 90% RMTDedicated placebo figure-of-eight coilSingle session per condition, ≥2-week washoutDaily pain assessmentStrict somatotopic targeting of rTMS was not required; hand-area M1 was easier to target and yielded better analgesic effects for both upper-limb and facial painNot explicitly reported2/517Galhardoni et al43 2019- Parallel- PSI-dTMS (n = 33), ACC-dTMS (n = 33), sham (n = 32)PSI, ACC- PSI-dTMS: double-cone coil- ACC-dTMS: H-6 coil− 10 Hz; 15 trains/session, 10s/train, ITI 50s, total 1500 pulses/session - 90% RMT- For insular stimulation, an active figure-of-eight coil was placed above an unplugged double-cone coil- For ACC stimulation, the built-in sham system was used- Induction: 5 consecutive days- Maintenance: weekly until week 12- Total: 16 sessionsAssessments during the 12-week treatment periodNeither ACC- nor PSI-dTMS showed significantly greater pain relief than sham; PSI-dTMS showed significant antinociceptive effects, and ACC-dTMS showed significant anxiolytic effects- Pain, predominantly headache, was the most common adverse event across all three groups- No seizures, syncope, falls, persistent headaches, or new neurological deficits were reported4/518Attal et al44 2021- Parallel- M1 rTMS (n = 49), DLPFC rTMS (n = 52), sham (n = 48)Left DLPFC, M1Figure-of-eight coil− 10 Hz; 30 trains/session, 100 pulses/train, ITI 20s, total 3000 pulses/session- 80% RMTDouble-sided active-placebo figure-of-eight coil- Induction: 5 consecutive daily sessions- Maintenance: 1 session weekly for 3 weeks, then 1 session every 2 weeks for 6 weeks, then 1 session every 3 weeks for 12 weeks- Total: 15 sessions15 sessions over 22 weeks; final assessment at week 25 (3 weeks post-treatment)M1 rTMS, but not DLPFC rTMS, was superior to sham in improving pain relief, the sensory dimension of pain, pain intensity, fatigue, and global improvement scores; no significant effects were observed on pain quality, mood, sleep, or quality of life- Headache: M1 17/49 (34.7%), DLPFC 23/52 (44.2%)- Stimulation-site pain: M1 5/49 (10.2%), DLPFC 13/52 (25%)- No seizures or deaths4/519Imperatore et al45 2021- Parallel- Left MC (n = 20), left DLPFC (n = 20)Left MC, left DLPFCFigure-of-eight coil− 10 Hz;60 trains/session, 50 pulses/train, 5s on/15s off, total 3000 pulses/session-MC: 90% RMT; DLPFC: 110% RMTNo sham control10 weekday sessions1 day and 1 week after the final treatment sessionMC stimulation reduced pain interference significantly more than DLPFC stimulation and showed larger effects on pain interference and urge to use opioids; both sites had similar effects on stress, pain, and discomfort- Mild transient headache in 3 patients- No serious adverse events2/520Mori et al46 2021- Crossover- 90% or 110% RMT M1 hand AP, 90% or 110% RMT M1 foot AP, 90% RMT M1 foot ML, sham (n = 11)M1 hand or foot area with AP or ML coil orientationFigure-of-eight coil− 10 Hz;60 trains, 5s/train, ITI 25s, total 3000 pulses/session- 90% or 110% RMTDedicated sham coil3 consecutive days per condition, ≥2-week washoutDaily assessment during 3 treatment days and 7 days post-treatmentOptimal rTMS for neuropathic leg pain may target M1 (foot or hand) at sub-RMT intensityOne patient was unable to complete the 110% RMT foot-M1 condition due to scalp pain; no serious adverse events3/521Kim et al47 2020- Parallel- Active iTBS (n = 15), sham (n = 15)M1Figure-of-eight coil− 3 pulses/burst at 50 Hz, 5 Hz within-train frequency, 10-s intervals for 200s, total 600 pulses/session- 80% AMTActive coil tilted 90° away from the scalp1 session/day for 5 daysImmediately after completing 5 daily iTBS sessionsSignificant improvements in neuropathic pain severity and related symptoms were observed in the active iTBS group but not in the sham groupNo adverse effects reported4/522André-Obadia et al48 2021- Crossover- 20-Hz rTMS, iTBS (n = 39)M1Figure-of-eight coil− 20-Hz rTMS: 20 trains/session, 80 pulses/train, ITI 84s, total 1600 pulses/session- iTBS: 10 bursts of 3 pulses at 50 Hz, delivered at 5 Hz, repeated every 10s to a total of 600 pulses-90% RMTNo sham coil5 consecutive daily sessions per condition, ≥5-week washoutDaily assessments for 3 weeks post-treatment20-Hz rTMS produced greater pain relief and fatigue improvement than iTBS during the month following stimulation; no significant differences in general health or sleep quality were observed between groups- Headache: 3/46 (6.5%), insomnia: 3/46 (6.5%), fatigue: 5/46 (10.8%) during stimulation, not after; all resolved within 2 days without medication- No serious adverse events2/523Shimizu et al49 2017- Crossover- Figure-of-eight coil rTMS, dTMS, sham (n = 18)M1H10-coil (dTMS), figure-of-eight coil− 5 Hz; 10 trains/session, ITI 50s, total 500 pulses/session- 90% RMTSham coil integrated into the BrainsWay helmet, reproducing scalp sensations, acoustic artifacts, and facial muscle activation5 daily sessions per condition, ≥17-day washoutImmediate and 1-hour assessmentsdTMS produced significant short-term pain reduction vs sham, whereas figure-of-eight coil rTMS showed no significant benefit- H-coil: 2/8 patients; figure-8 coil: 4/9 patients.- Reported events included headache, dizziness, upper-limb pain, foot phalanx fracture (due to a slip; unrelated to stimulation5/524Dongyang et al50 2021da Cunha et al51 2022- Crossover- Active dTMS, sham (n = 31)PSIDouble-cone coil− 10 Hz; 30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- 80% RMTOrthogonally positioned active figure-of-eight coil, with the double-cone coil switched off during sham stimulation5 daily sessions per series, ≥21-day washoutAssessments up to 7 days after the fifth sessionActive dTMS produced significant pain reduction during the 5-day induction phase vs sham; treatment response was associated with pain subtype rather than PSI target location- Active: headache (n=2), dizziness (n=1), scalp tenderness (n=1)- Sham: headache (n=1)- All headaches resolved before the next session; one patient required metamizole- No serious adverse events or adverse event–related discontinuations were reported.5/525Bouhassira et al52 2024- Crossover- Figure-of-eight coil rTMS, dTMS, sham (n = 59)M1Figure-of-eight coil,H10-coil (dTMS)− 10 Hz- dTMS coil: 30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- Figure-of-eight coil rTMS: 30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- 80% RMTIntegrated sham F8-coil or H-coil system reproducing scalp sensations and acoustic artifacts without substantial cortical stimulation5 daily sessions per condition, 5–6-week washout1 and 3 weeks after the final session of each treatment periodBoth dTMS and figure-of-eight coil rTMS produced significant analgesia compared with sham, with similar effect magnitudes; figure-of-eight coil rTMS showed earlier effects, whereas H-coil dTMS demonstrated more delayed but longer-lasting benefits- Stimulation-site pain (26% of the patients), 5 patients did not discontinue treatment; headache during or shortly after stimulation (35% of the patients), increased bodily pain and fatigue- One unrelated serious adverse event (rectal cancer)5/526Farnes et al53 2024- Crossover- Active dTMS, sham (n = 17)M1H7-coil− 10 Hz; 30 trains/session, 100 pulses/train, ITI 20s, total 3000 pulses/session- 80% RMTIntegrated sham H7-coil in the same helmet, reproducing acoustic artifacts with negligible intracerebral stimulation5 daily sessions per condition, ≥9-week washout1 and 3 weeks after treatmentActive dTMS significantly reduced pain as well as anxiety and depression compared with sham, with analgesic effects emerging approximately 2 weeks after treatmentStimulation-site pain (82.4%), headache (82.4%); two withdrawals due to adverse events4/527André-Obadia et al54 2023- Crossover- rTMS, tDCS (n = 56)M1Figure-of-eight coil− 10 Hz; 32 trains/session, 50 pulses/train, ITI 25s, total 1600 pulses/session- 90% RMTNo sham control5 daily sessions, ≥4-week washoutAssessments for 4 weeks post-treatmentBoth rTMS and tDCS were effective at 1 month in drug-resistant neuropathic pain, with slightly greater benefit from rTMSNoise intolerance and headache (n=1; withdrawal); immediate tension headache (n=1); delayed headache (n=2)3/5Painful Diabetic Neuropathy28Onesti et al55 2013- Crossover- Active dTMS, sham (n = 23)M1H-coil− 20 Hz; 30 trains/session, 50 pulses/train, ITI 30s- 100% RMTIntegrated sham H-coil reproducing acoustic artifacts, scalp sensations, and facial muscle activation, with negligible intracerebral stimulation5 consecutive days per condition, 5-week washoutImmediately and 3 weeks after each 5-day treatment periodActive dTMS significantly reduced VAS pain scores, and effects disappeared approximately 3 weeks after stimulation; sham showed no significant changeNo minor or major adverse effects reported5/529Yang et al56 2022- Parallel- Active rTMS (n = 10), sham (n = 10)Left M1Figure-of-eight coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s, total 1000 pulses/session- 90% MTActive figure-of-eight coil positioned perpendicular to the scalp during sham stimulation5 consecutive sessions1 day and 1 week after treatmentActive rTMS significantly reduced pain scores at 1 day and 1 week post-treatment and improved physical and mental QOL scores; no significant changes were observed in the sham groupNo adverse effects reported4/530Thakkar et al57 2024- Parallel- M1 pcTBS (n = 21), DLPFC pcTBS (n = 21; sham before active pcTBS)M1, left DLPFCFigure-of-eight coil− 3 pulses at 50 Hz repeated 400 times at intervals of 200 ms, total 1200 pulses in 1 min 44s- 80% RMTDedicated sham coil identical in appearance and sound to the active coil, without active stimulationSingle sessionImmediately and 24 hours after the single treatment session.Pain relief improved significantly after a single session of pcTBS targeting M1 and DLPFC brain regions- Headache: 8, neck pain: 6, discomfort: 5, increased pain: 2, toothache: 1, shoulder pain: 1, nausea: 1- No serious adverse events2/5Postherpetic Neuralgia31Ma et al58 2015- Parallel- Active rTMS (n = 20), sham (n = 20)M1Circular coil− 10 Hz; 300 5-s pulses at 10 Hz, ITI 3s, total 1500 pulses/session- 80% RMTPerpendicular coil placed above an inactive tangential coil, producing minimal or no cortical stimulation10 daily sessions (once daily except weekends)1 and 3 months after treatmentActive rTMS significantly reduced VAS pain scores compared with sham at most points up to 3 months after treatment- Dry mouth (1/0), headache (2/1), neck pain (2/3), dizziness (1/1)- No serious adverse events2/532Pei et al59 2019- Parallel- 5-Hz rTMS (n = 20), 10-Hz rTMS (n = 20), sham (n = 20)M1NS− 5 Hz: 300 1-s pulses at 5 Hz, ITI 2.5s, total 1500 pulses/session- 10 Hz: 300 0.5-s pulses at 10 Hz, ITI 3s, total 1500 pulses/session- 80% MTDedicated sham coil reproducing the sound of active stimulation without generating a magnetic field15 daily sessionsDuring rTMS, 1 and 3 months after treatmentBoth 5-Hz and 10-Hz rTMS significantly reduced pain compared with sham up to 3 months post-treatment; 10-Hz rTMS showed greater analgesic efficacy than 5-Hz rTMS- Active 5 Hz: headache 1, neck pain 1- Active 10 Hz: dry mouth 1, headache 2, neck pain 2, dizziness 13/533Wang et al60 2023- Parallel- M1 rTMS (n = 20), DLPFC rTMS (n = 20), sham (n = 20)M1, left DLPFCFigure-of-eight coil− 10 Hz; 5s/train, ITI 25s, total 3000 pulses/session- 100% RMTDedicated sham coil reproducing the sound of active stimulation without generating a magnetic field10 daily sessions over 2 consecutive weeksFirst week of treatment (week 1), post-treatment (week 2), 1-week (week 4), 1-month (week 6), and 3-month (week 14) follow-upM1 rTMS resulted in significantly greater analgesia during and after treatment compared with sham (weeks 2–14) and DLPFC rTMS (weeks 1–14)- Mild headache: M1 5%, DLPFC 15%, sham 10%- Mild scalp discomfort: M1 15%, DLPFC 20%, sham 15%- No serious adverse events4/5Phantom Limb Pain34Ahmed et al61 2011- Parallel- Active rTMS (n = 17), sham (n = 10)M1Figure-of-eight coil− 20 Hz; 200 pulses/train, 10s/train, 1 train/1 min for 10 min- 80% RMTActive figure-of-eight coil elevated and angled away from the scalp, minimizing intracerebral stimulationOnce daily for 5 consecutive daysImmediately after the 1st and 5th sessions, and at 1 and 2 monthsActive rTMS significantly reduced pain compared with sham stimulation, with analgesic effects persisting for up to 2 months after treatment; additionally, β-endorphin concentration increased significantly after active rTMS but not after sham stimulationNot explicitly reported1/535Malavera et al62 2016- Parallel- Active rTMS (n = 27), sham(n = 27)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 6s/train, ITI 54s, total 1200 pulses/session- 90% RMTDedicated sham coil with similar appearance, weight, and acoustic artifacts, but without cortical stimulation or equivalent scalp sensation10 daily sessions15 and 30 days after treatmentActive rTMS significantly reduced pain intensity and showed a higher responder rate (>30% pain reduction) compared with sham at 15 days after treatment; no significant between-group difference was observed at 30 days after treatmentHeadache, neck pain, sleepiness across treatment groups; no serious adverse events4/536Vats et al63 2024- Parallel- Active rTMS (n = 8), sham (n = 6)DLPFCFigure-of-eight coil− 1 Hz; 8 trains/session, 150 pulses/train, ITI 60s, total 1200 pulses/session- 90% RMTActive figure-of-eight coil positioned perpendicular to the scalp, reproducing stimulation sounds without active cortical stimulation10 daily sessionsAfter each treatment session, and at 15, 30 and 60 days after treatmentActive rTMS significantly reduced pain compared with sham; analgesic effects were sustained during follow-up up to 60 days after treatmentNo adverse effects reported during treatment2/5Central Poststroke Pain37de Oliveira et al64 2014- Parallel- Active rTMS (n = 11), sham (n = 10)Left PMC/DLPFCFigure-of-eight coil− 10 Hz; 25 trains/session, 5s/train, ITI 25s, total 1250 pulses/session- 120% RMTDedicated sham coil used with the same stimulation procedure as active rTMS10 daily sessions on weekdays over 2 weeksAfter the 5th and 10th day of rTMS sessions and at 1 week, 1, 2, and 4 weeks after treatmentEarly termination due to lack of efficacy after 21 patients completed the trial and 4-week follow-upMild headache (3/1); no serious adverse events4/538Zhao et al65 2021- Parallel- Active rTMS + PT (n = 19), sham + PT (n = 19)M1Figure-of-eight coil− 10 Hz; 15 trains/session, 1.5s/train, ITI 3s, total 1500 pulses/session- 80% RMT (affected hand); 100% RMT (unaffected hand)Dedicated sham coil identical in shape and color, reproducing acoustic artifacts without magnetic stimulationOnce daily, 6 days/week, 3 weeksAssessments at days 3, 7, 14, and 21 during treatmentActive rTMS significantly reduced pain scores compared with sham with effects lasting up to 3 weeks, increased serum BDNF levels, and decreased RMT; anxiety and depression scores did not change significantlyThree patients experienced transient scalp numbness or facial twitching; no serious adverse effects4/539Ojala et al66 2022- Crossover- M1 rTMS, S2 rTMS, sham (n = 17)M1, S2Figure-of-eight coil− 10 Hz; 101 pulses/train, 10s/train, ITI 50s, 5050 pulses/session- 90% MTActive figure-of-eight coil separated from the scalp using a 75-mm nonconductive plastic spacer to minimize the induced electric field10 sessions/target (daily × 2 weeks), ≥1-month washout1 month after each treatment periodAll conditions produced short-term pain relief with no between-treatment differences; only S2 rTMS significantly reduced pain from baseline to 1 month after treatment- Mostly mild and transient adverse events- Headache (M1: n=1; sham: n=4; S2: n=3), fatigue (2, 2, 3), paresthesia (2, 3, 3), transient pain exacerbation (2, 2, 3), collapse (M1: n=1), increased spasticity (S2: n=2), and dizziness (S2: n=1)4/5Pain Associated with Spinal Cord Injury (SCI)40Defrin et al67 2007- Parallel- Active rTMS (n = 6), sham (n = 5)VertexFigure-of-eight coil− 5 Hz; 500 trains at 5 Hz for 10s, ITI 30s, total 500 pulses/session- 115% RMTDedicated sham coil, identical in shape and producing similar acoustic noise10 daily sessions (2 weeks)During 10 treatment sessions; 2–6 weeks post-treatmentBoth active and sham rTMS produced significant immediate reductions in pain scores, but only real rTMS increased heat-pain thresholds and showed sustained improvements in pain during follow-upNot reported3/541Kang et al68 2009- Crossover- rTMS, sham (n = 11)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s, total 1000 pulses/session- 80% RMTActive figure-of-eight coil elevated and angled away from the scalp5 consecutive daily sessions, 12-week washoutImmediately after the third and fifth stimulation sessions,and 1, 3, 5, and 7 weeks after the end of the 5-day stimulationperiodActive rTMS did not differ from sham in reducing average pain or pain interference; only worst pain showed a significant time effect with active rTMSNot reported3/542Yılmaz et al69 2014- Parallel- Active rTMS (n = 9), sham (n = 7)VertexFigure-of-eight coil− 10 Hz; 30 trains/session, 5s/train, ITI 25s, total 1500 pulses/session- 110% RMTActive figure-of-eight coil angled away from the scalp10 daily sessions10 days (immediately after session), 6 weeks, and 6 months post-treatmentVAS scores significantly decreased in both groups, with no significant between-group differences in pain at any time pointNot reported4/543Nardone et al70 2017- Parallel- Active rTMS (n = 6), sham (n = 6)Left PMC/DLPFCFigure-of-eight coil− 10 Hz; 25 trains/session, 5s/train, ITI 25s, total 1250 pulses/session- 120% RMTDedicated sham coil identical in size, shape, color, and acoustic output10 daily sessions over 2 weeks1 day after the first week of treatment, 1 day, 1 week, and 1-month post-treatmentActive rTMS significantly reduced pain compared with baseline; sham showed no significant changeTransient facial muscle twitching in 2 active-rTMS patients2/544Sun et al71 2019- Parallel- Active rTMS (n = 11), sham (n = 6)Left M1Figure-of-eight coil− 10 Hz; 1.2s/train, ITI 3s, total 1200 pulses/session- 80% RMTActive figure-of-eight coil positioned at 90° to the scalpDaily sessions for 6 weeks with a 1-day interval per weekAfter the first session rTMS, after one week, two weeks, four weeks, and six weeksActive rTMS showed greater posttreatment pain reduction than sham from weeks 2–6; analgesia may be linked to reduced M1 or PMC hypersensitivity- No complaints of discomfort during and after rTMS- No serious adverse events4/545Zhao et al72 2020- Parallel- Active rTMS (n = 24), sham (n = 24)M1Figure-of-eight coil− 10 Hz; 15 pulses/train, ITI 3s, total 1500 pulses/session- 90% RMTDedicated sham coil identical in shape and color, producing sound without magnetic stimulationOnce daily, 6 days/week for 3 weeksDays 3, 7, 14 and 21 during treatmentActive rTMS significantly reduced pain compared with sham at day 3 and week 1 after onset of treatment and significantly increased neurotrophic factors and MEP amplitude- Transient scalp numbness and facial muscle twitching- No serious adverse effects reported4/546Jetté et al73 2013- Crossover- M1 hand rTMS, M1 leg rTMS, sham (n = 16)M1Figure-of-eight coil− 10 Hz;40 trains/session, 5s/train, ITI 25s, total 2000 pulses/session- Hand M1: 90% RMT; Leg M1: 110% RMTDedicated sham coil reproducing stimulation noise and scalp sensationsSingle session per condition, ≥2-week washoutImmediate assessment and daily pain recordings for 1 week after each sessionActive rTMS over the hand or leg area significantly reduced pain up to 48 h postintervention; sham showed no significant change; incomplete lesions showed greater analgesic response; corticospinal excitability increased only after hand-area stimulationNot reported4/547Wu et al74 2026- Parallel- M1 + PMC rTMS (n = 13), M1 rTMS (n = 11), sham (n = 11)Left M1, left PMCFigure-of-eight coil− 10 Hz; 1s/train, ITI 3s, 1000 pulses/site/session- 80% RMTActive coil positioned at 90° to the scalp, reproducing stimulation noise without intended cortical stimulationDaily sessions for 4 weeks, 2-day interval per weekDay 1 and weeks 1, 2, 4 and 6Both M1 and M1 + PMC rTMS significantly reduced NRS scores compared with sham; M1 + PMC rTMS showed the greatest effect, which was associated with reduced activation of left M1 and PMCNot reported4/548Yang et al75 2023- Parallel- iTBS (n = 12), 10-Hz rTMS (n = 13), iTBS + 10-Hz rTMS (n = 12)M1Circular coil- iTBS: triplet 50 Hz bursts, repeated at 5 Hz, 2s on and 8s off, 600 pulses/session, total 3 min 20s- 10-Hz rTMS: 30 trains/session, 5s/train, ITI 25s, total 1500 pulses/session- iTBS + rTMS: iTBS first, followed 2 min later by 10-Hz rTMS, total 2100 pulses-80% RMTNo sham controlOnce daily on weekdays (5 days/week) for 4 weeksAfter week 1 and immediately after the 4-week treatment courseAll 3 modalities significantly reduced pain at 4 weeks but not at 1 week; iTBS priming was more effective than iTBS alone; 10-Hz rTMS was not enhanced by iTBS primingReported to be well tolerated3/549Kim et al76 2025- ParalleliTBS (n = 11), 10-Hz rTMS (n = 10), sham (n = 11)M1Figure-of-eight coil- iTBS: bursts of 3 pulses (50 Hz), 2s/train every 10s, total 600 pulses/session- 10-Hz rTMS: 25 trains/session, 80 pulses/train, ITI 10s, total 2000 pulses/session- iTBS: 80% AMT; 10-Hz rTMS: 80% RMTActive coil tilted 90° to the scalp during sham rTMS1 session/day for 5 consecutive daysImmediately after completing 5 treatment sessionsBoth iTBS and 10-Hz rTMS significantly reduced pain compared with sham; no significant difference was observed between iTBS and 10-Hz rTMS; considering safety, convenience, and adherence, iTBS may be preferable- One patient in the active rTMS group withdrew because of headache- No additional adverse effects reported among those completing treatment4/5Brachial Plexus Injury50de Assis et al77 2022- Parallel- Active stimulation (n = 12; active rTMS or tDCS), sham rTMS or tDCS (n = 8)M1Figure-of-eight coi− 10 Hz;25 trains/session, 10s/train, ITI 17s, total 2500 pulses/session- 90% RMTActive figure-of-eight coil elevated and tilted away from the scalp5 daily consecutive sessions per condition, 30-day washoutAfter the final session and 30 days post-treatmentActive rTMS and tDCS significantly reduced continuous and paroxysmal pain, multidimensional aspects of pain, and anxiety state compared with sham; no significant differences were observed between rTMS and tDCS- Active rTMS: headache (n=1), vertigo (n=1), increased continuous pain (n=1)- Sham rTMS: increased continuous pain (n=2), increased paroxysmal pain (n=3)4/5Lumbosacral Radiculopathy51Attal et al78 2016- Parallel- Active stimulation (n = 21; active rTMS and tDCS), sham rTMS or tDCS (n = 11)M1Figure-of-eight coil− 10 Hz;30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- 80% RMTDedicated sham coil visually indistinguishable from active coil, reproducing acoustic output and similar scalp sensations3 consecutive daily sessions per condition, 3-week washoutAfter each of 3 treatment sessions and 5 days after the final sessionrTMS produced significantly greater pain reduction than tDCS and sham; tDCS was not differ significantly from sham to sham; rTMS lowered cold pain thresholds, which correlated with its analgesic effect, but had no significant impact on individual neuropathic symptomsMild-to-moderate adverse effects; 3 patients withdrew because of adverse effects5/5Pain Associated With mild Traumatic Brain Injury52Choi et al79 2018- Parallel- Active rTMS (n = 6), sham (n = 6)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s, total 1000 pulses/session- 90% MTActive coil positioned perpendicular to the skull during sham stimulation10 sessions (weekdays) over 2 weeksAfter the 5th and 10th sessions; 1, 2 and 4 weeks post-treatmentActive rTMS significantly reduced NRS scores compared with sham up to 4 weeks post-treatment and improved the SF-36 physical component score from post-session 10 to 4-week follow-upNo adverse side effects were reported4/5Abbreviations: ACC, anterior cingulate cortex; AMT, active motor threshold; AP, anteroposterior; BDNF, brain-derived neurotrophic factor; DLPFC, dorsolateral prefrontal cortex; dTMS, deep transcranial magnetic stimulation; ITI, intertrain interval; iTBS, intermittent theta-burst stimulation; LM, lateromedial; M1, primary motor cortex; MC, motor cortex; MEP, motor-evoked potential; ML, mediolateral; MSO, maximum stimulator output; MT, motor threshold; NRS, numeric rating scale; PA, posterior–anterior; pcTBS, prolonged continuous theta-burst stimulation; PFC, prefrontal cortex; PMC, premotor cortex; PSI, posterior superior insula; PT, physical therapy; QOL, quality of life; RMT, resting motor threshold; rTMS, repetitive transcranial magnetic stimulation; S1, primary somatosensory cortex; S2, secondary somatosensory cortex; SCI, spinal cord injury; SMA, supplementary motor area; tDCS, transcranial direct current stimulation; VAS, visual analog scale.

Nonspecific Neuropathic Pain

Of the 27 studies27–54 investigating rTMS for NP not associated with a specific underlying condition, 8 studies27–34 evaluated the analgesic efficacy of conventional rTMS through comparisons between active rTMS and sham stimulation. Other studies focused on methodological variations, including 5 studies35–39 comparing different stimulation frequencies and 7 studies40–46 comparing different stimulation targets. Additional investigations examined alternative stimulation paradigms, such as 2 studies47,48 investigating iTBS, 4 studies49–53 examining dTMS, and 1 study54 comparing rTMS with anodal transcranial direct current stimulation (tDCS).

Among eight RCTs comparing conventional HF-rTMS (5–20 Hz) with sham stimulation, 727–29,31–34 targeted M1, whereas only 130 targeted the prefrontal cortex (PFC). Analgesic outcomes varied across stimulation frequencies. At 5 Hz, Hosomi et al reported short-term pain relief of active rTMS in their earlier crossover trial27 but no significant benefit over sham in their subsequent parallel RCT, although responders showed improvement during open-label maintenance treatment.28 Findings at 10 Hz were also mixed, with two studies reporting analgesic benefits and one showing no significant difference in analgesic efficacy from sham.29–31 In contrast, all three 20-Hz trials reported significant pain reduction following active stimulation, although differences in patient populations and stimulation protocols limit direct conclusions.32–34

This review revealed 5 crossover studies35–39 that evaluated the analgesic effects of rTMS at different stimulation frequencies in patients with NP. Higher-frequency stimulation (10–20 Hz) generally produced more favorable analgesic outcomes than lower-frequency stimulation. Beyond stimulation frequency, Mori et al reported greater pain reduction with higher stimulation doses, particularly 10-Hz rTMS delivering 2000 pulses per session.39

Seven studies40–46 compared different rTMS stimulation targets for NP. Findings generally favored M1 stimulation.40,44,45 André-Obadia et al reported greater analgesic effects with posterior–anterior than lateral–medial coil orientation, suggesting that coil orientation may influence stimulation efficacy.41 Regarding somatotopic targeting, André-Obadia et al found significantly greater pain reduction with hand-area than face-area M1 stimulation, regardless of pain location, suggesting that strict somatotopic targeting may be unnecessary.42 Similarly, Mori et al also reported that targeting the M1 hand or foot area with sub-RMT intensity may be optimal for neuropathic leg pain.46 Galhardoni et al found no significant analgesic benefit of ACC- or PSI-targeted dTMS over sham, although PSI stimulation showed antinociceptive effects and ACC stimulation showed anxiolytic effects.43

This review identified 2 RCTs comparing iTBS with 20-Hz rTMS or sham stimulation.47,48 Kim et al conducted an RCT in 30 patients with central NP; significant improvements in NP severity and related symptoms were observed in the real iTBS group but not in the sham group.47 In 2021, André-Obadia et al demonstrated that 20-Hz rTMS provided significantly greater pain relief over the month following stimulation compared with iTBS; however, no significant differences in general health or sleep quality were observed between groups.48

In this review, 4 RCTs49–53 compared dTMS using H-coils or double-cone coils with figure-of-eight coil rTMS or sham stimulation. Three studies49,52,53 evaluated M1-targeted dTMS, whereas one50 examined PSI-targeted dTMS. Both M1- and PSI-targeted dTMS demonstrated significant pain reduction compared with sham.

A secondary analysis suggested that PSI-rTMS response may depend on pain phenotype rather than precise stimulation location.51 Findings from two direct comparisons of dTMS with conventional figure-of-eight coil rTMS were mixed. Shimizu et al reported short-term analgesic benefits with dTMS but not conventional rTMS,49 whereas Bouhassira et al observed comparable analgesic effects, with earlier responses to conventional rTMS and more delayed but longer-lasting benefits with dTMS.52

Only one study by André-Obadia et al in 2023, enrolling 56 patients with drug-resistant NP, compared 10-Hz rTMS applied over M1 with anodal tDCS.54 The authors reported that both 10-Hz rTMS and tDCS were effective at 1 month, with a slightly greater benefit observed with rTMS. They also recommended that, because a considerable proportion of patients responded to only one stimulation modality, both approaches should be evaluated before concluding that a patient is nonresponsive to treatment.

Overall, the 27 studies suggest potential analgesic benefits of rTMS for NP, particularly with high-frequency (10–20 Hz) M1 stimulation, although findings remain inconsistent. Stimulation frequency, dose, coil orientation, and cortical targeting may influence analgesic outcomes. Strict somatotopic targeting may not be necessary.

Alternative approaches, including iTBS and dTMS, have also shown promising results. However, substantial methodological heterogeneity and limited comparative evidence preclude firm conclusions regarding the optimal stimulation strategy and long-term analgesic benefits.

Painful Diabetic Neuropathy

Diabetic neuropathy (DN) is a common and debilitating complication of diabetes, with a recent meta-analysis reporting a global pooled prevalence of 36% among individuals with diabetes.80 Among patients with DN, 46.7% experience painful diabetic neuropathy (PDN), which substantially impairs quality of life and contributes to disease burden.81 Oral pharmacological treatments are commonly used to manage PDN; however, their analgesic efficacy is often insufficient and may be limited by adverse effects.82,83 Consequently, rTMS has been investigated as a potential therapeutic option for PDN.

Three studies investigated rTMS for PDN, using different stimulation protocols and cortical targets.55–57 Two studies55,56 evaluated high-frequency (10–20 Hz) stimulation of M1 over five consecutive days, whereas the third investigated a single session of prolonged continuous theta-burst stimulation targeting either M1 or the DLPFC.57 Despite these differences, all three studies reported statistically significant pain reduction following active stimulation. However, the evidence primarily reflects short-term outcomes, with one study55 reporting that the analgesic effect was no longer evident approximately three weeks after treatment.

Overall, these findings suggest potential short-term analgesic benefits of rTMS in PDN, although limited evidence and methodological heterogeneity leave its clinical significance, long-term effects, and optimal stimulation protocol uncertain.

Postherpetic Neuralgia

Postherpetic neuralgia (PHN) is a peripheral NP disorder resulting from reactivation of the varicella–zoster virus in the dorsal root ganglion, affecting the corresponding dermatomes, with a prevalence ranging from 5% to more than 30% among patients with herpes zoster.84,85 PHN is often persistent and refractory to currently available pharmacological treatments, leading to substantial impairment in patients’ quality of life.86 The application of rTMS for the treatment of PHN has been investigated and has shown promising results.

Evidence for rTMS in PHN comes from 3 parallel RCTs.58–60 All three studies reported significant pain reduction with active M1 stimulation compared with sham, with benefits observed for up to three months after treatment. Notably, comparative findings favored 10-Hz over 5-Hz stimulation and M1 over DLPFC targeting, suggesting that both stimulation frequency and cortical target may influence analgesic outcomes.59,60

Collectively, these studies provide relatively consistent evidence supporting the analgesic efficacy of rTMS—particularly 10-Hz stimulation over M1—in PHN. However, larger, well-powered studies are required to define optimal stimulation parameters and to establish treatment effects.

Phantom Limb Pain

Phantom limb pain (PLP) is a common neuropathic pain condition, with a recent meta-analysis reporting a pooled prevalence of 61% among individuals with limb amputations.87 The mechanisms underlying phantom limb pain remain incompletely understood and are thought to involve complex interactions among cortical, spinal, and peripheral processes, among which maladaptive neuroplasticity is considered a key contributing mechanism.88–90 Therefore, rTMS has been investigated as a potential therapeutic approach for the treatment of PLP.

Three RCTs evaluated rTMS for PLP, with variations in stimulation targets, frequency, and treatment duration.61–63 Two studies61,62 evaluated high-frequency (10–20 Hz) stimulation targeting M1, whereas the study by Vats et al63 investigated low-frequency (1 Hz) stimulation targeting the DLPFC. All three studies reported statistically significant pain reduction compared with sham. However, the durability of these effects varied, with two studies61,63 reporting benefits lasting up to two months after treatment, whereas the remaining study demonstrated significant between-group difference at 15 days but not at 30 days post-treatment.62

Taken together, these findings indicate promising short-term analgesic effects of HF-rtMS targeting the M1 or DLPFC, although the limited evidence does not allow the optimal protocol or long-term clinical benefits to be established.

Central Poststroke Pain

Central poststroke pain (CPSP) is a form of central NP that results from lesions affecting the somatosensory pathways following stroke.91,92 Pharmacological therapy, including antidepressants and anticonvulsants, is generally considered the first-line treatment for CPSP.93 However, treatment outcomes are often suboptimal, and a proportion of patients continue to experience persistent pain, highlighting the need for alternative therapeutic approaches such as rTMS.

Three RCTs investigated 10-Hz rTMS for CPSP, yielding inconsistent results across cortical targets.64–66 In separate trials, M1 stimulation significantly reduced pain compared with sham, whereas PMC/DLPFC stimulation showed no significant benefit.64,65 Ojala et al found no significant between-group differences among M1, the secondary somatosensory cortex (S2), and sham stimulation, despite a modest reduction in pain following S2 stimulation at one month after treatment.66 Adverse events were generally mild and transient, although increased spasticity during S2 stimulation and one episode of collapse during M1 stimulation were reported by Ojala et al.66

These mixed findings, together with small sample sizes and differences in stimulation targets and protocols, leave the analgesic efficacy and optimal stimulation strategy of rTMS for CPSP uncertain.

Neuropathic Pain After Spinal Cord Injury (SCI)

Spinal cord injury (SCI) is a severe neurological injury that frequently leads to long-term disability and a range of sequelae. Chronic pain affects approximately 81% of individuals with SCI, with NP representing nearly 86% of these cases.94 NP in patients with SCI is often refractory to conventional pharmacological treatments, prompting the exploration of alternative therapeutic approaches such as rTMS.67

Our literature search identified 10 RCTs67–76 evaluating the efficacy of rTMS for treating NP in patients with SCI. Of these, 6 studies67–72 compared active conventional rTMS with sham stimulation, 2 studies73,74 compared rTMS applied to different stimulation sites, and 2 studies75,76 compared iTBS with conventional rTMS.

Six RCTs reported inconsistent analgesic outcomes with conventional rTMS for SCI-related NP.67–72 Among the three trials using 10-Hz M1 stimulation, Sun et al and Zhao et al reported significantly greater pain reduction compared with sham,71,72 whereas Kang et al found no significant between-group differences in average pain intensity or pain interference.68 Both vertex-targeted studies67,69 reported pain reduction in active and sham groups; however, Defrin et al additionally found a significant increase in heat-pain thresholds with active rTMS and sustained improvements in pain scores during follow-up with only active rTMS. Nardone et al also observed significant pain reduction from baseline with 10-Hz left PMC/DLPFC stimulation, but not with sham.70

Two additional RCTs examined different cortical targets for SCI-related NP.73,74 Both reported analgesic benefits following active stimulation, but their comparative findings differed. Jetté et al found no significant difference between M1 hand and leg stimulation,73 whereas Wu et al reported greater pain reduction with combined M1 and PMC stimulation than with M1 stimulation alone.74

Two further RCTs compared iTBS-based protocols with conventional 10-Hz rTMS for SCI-related NP.75,76 Both reported significant pain reduction across active stimulation protocols. Yang et al found that iTBS priming followed by 10-Hz rTMS produced greater pain reduction than iTBS alone, but showed no significant difference from 10-Hz rTMS alone.75 Kim et al reported significant pain reduction with both iTBS and 10-Hz rTMS compared with sham, without detecting a statistically significant difference between the two active protocols.76

Overall, HF-rTMS targeting M1, particularly at 10 Hz, has been the most frequently studied approach for SCI-related NP, although analgesic outcomes remain inconsistent. iTBS may also warrant consideration in this population.

Other Neuropathic Pain Conditions

Beyond the aforementioned NP conditions, rTMS has also been investigated for NP associated with other disorders, including brachial plexus injury, lumbosacral radiculopathy, and mild traumatic brain injury (mTBI).

NP is a common and debilitating complication following brachial plexus injury, affecting approximately 54% of patients with this condition.95 In an RCT involving 20 patients, de Assis et al reported that 10-Hz M1-rTMS and tDCS significantly reduced continuous and paroxysmal pain, improved multidimensional pain outcomes, and reduced anxiety compared with sham stimulation, with no significant difference between the two active treatments.77 The authors suggested that tDCS may be preferred because of its practicality and accessibility.

Lumbosacral radiculopathy commonly causes radiating pain due to nerve root compression or irritation.96 Attal et al in 2016 reported that 10-Hz M1-rTMS produced significantly greater pain reduction than both tDCS and sham stimulation, whereas tDCS did not differ significantly from sham.78 Active rTMS lowered cold pain thresholds, which correlated with its analgesic effect, but had no significant impact on individual neuropathic symptoms.

NP can develop after mTBI as a result of central nervous system dysfunction affecting pain pathways. In 2018, Choi et al reported an RCT in 12 patients with central NP after mTBI assigned to active rTMS or sham groups.79 Ten-hertz rTMS was applied over the M1 of the affected hemisphere for 10 sessions over 2 weeks. Compared with sham, active rTMS significantly reduced pain intensity up to 4 weeks post-treatment and improved the physical component score of the Short Form-36 Health Survey (SF-36) from session 10 through the 4-week follow-up.

Chronic Primary Pain Conditions

According to the International Classification of Diseases, 11th Revision (ICD-11), chronic primary pain describes long-lasting or recurrent pain affecting one or more anatomical regions for at least 3 months, accompanied by significant emotional distress or functional disability that interferes with everyday activities and social roles and that cannot be explained by another chronic pain disorder.97 In this section, we review the application of rTMS in 2 chronic primary pain conditions, FM and complex regional pain syndrome (CRPS). Other pain conditions within the categories of chronic primary pain—related to headache, orofacial pain, visceral pain, and musculoskeletal pain—are discussed in subsequent sections. The included studies are summarized in Table 2.

Table 2Characteristics of Studies Evaluating Repetitive Transcranial Magnetic Stimulation (rTMS) for Some Chronic Primary Pain ConditionsNo.Author, YearStudy Design/ Group (n)Stimulation SiteCoil Type of Active rTMSActive rTMS ParametersSham ProcedureTreatment CourseFollow -UpSummary of FindingsSafety, n (Active/Sham)Jadad ScoreFibromyalgia53Carretero et al98 2009- Parallel- Active rTMS (n = 14), sham (n = 12)Right DLPFCFigure-of-eight coil− 1 Hz; 20 trains/session, 60s/train, ITI 45s, total 1200 pulses/session- 110% MTCoil positioned perpendicular to the skull at the stimulation site, then tilted 45° forward20 sessions (daily on weekdays)Evaluations at weeks 2,4, and 8Both active rTMS and sham groups showed improvements in Fibro Fatigue and Clinical Global Impression scores, with no between-group differences; no significant changes were observed in the Likert Pain ScaleHeadache/neck pain (6/2), worsening depression (1/0), nausea and increased fatigue (1/0)3/554Tanwar et al99 2020- Parallel- Active rTMS (n = 45), sham (n = 41)Right DLPFCFigure-of-eight coil− 1 Hz; 8 trains/session, 150 pulses/train, ITI 1 min, total 1200 pulses/session- 90% RMTDedicated inactive coil positioned over the same cortical target5 consecutive days/week for 4 weeks15 days, 3 months and 6 months post-treatmentThe active rTMS group showed significant improvements in pain and related symptoms after treatment, with effects lasting up to 6 months follow-up, whereas no significant changes in pain were observed in the sham rTMS groupHeadache (2/0), neck pain (0/2), mild dizziness (0/1)2/555Tiwari et al100 2024- Parallel- Active rTMS (n = 17), sham (n = 17)Right DLPFCFigure-of-eight coil− 1 Hz; 8 trains/session, 150 pulses/train, total 1200 pulses/session- 90% RMTActive coil positioned at a right angle (90°) to the scalp5 consecutive sessions/week, total 20 sessions over 4 weeksafter the 4-week treatment courseActive rTMS significantly reduced pain intensity compared with sham post-treatment. Other pain perception parameters, cognition, and sleep quality improved significantly only in the active rTMS group- Mild headache (2/0), neck pain (0/1), dizziness (0/1)- All adverse events were mild and transient4/556Passard et al101 2007- Parallel- Active rTMS (n = 15), sham (n = 15)Left M1Figure-of-eight coil− 10 Hz; 25 trains/session, 8s/train, ITI 52s, 2000 pulses/session- 80% RMTDedicated sham coil with appearance and acoustic output like the active coilDaily sessions for 2 weeks with a 2-day interval between weeks, total 10 sessionsAssessments on days 15, 30 and 60 from treatment initiationActive rTMS significantly reduced pain and improved several quality-of-life domains (fatigue, morning tiredness, activity, walking, and sleep), with effects lasting up to 2 weeks after treatmentHeadache (4/5), nausea (1/0), transient tinnitus (0/2), mild dizziness (0/1)4/557Mhalla et al102 2011- Parallel- Active rTMS (n = 16), sham (n = 14)Left M1Figure-of-eight coil− 10 Hz; 15 trains/session, 10s/train, ITI 50s, total 1500 pulses/session- 80% RMTDedicated sham coil identical in size, color, and shape, reproducing stimulation sounds- Induction: 5 daily sessions- Maintenance: 3 weekly, 3 fortnightly, 3 monthly sessions- Total: 14 sessionsDay 1 to week 21 and at week 25Active rTMS significantly reduced pain vs sham for up to 1 month after the last session, although the effects tended to decrease during the monthly phase. Active rTMS significantly reduced both sensory and affective pain dimensions, decreased some pain interference, and alleviated FM-related symptoms—including fatigue, stiffness, and morning tiredness—while also lowering pain catastrophizing, with no significant effects on anxiety or depression- Headache (6/5), transient, dizziness(1/0)- Headaches led to withdrawal in 2 patients (one per group).

Other headaches were mild and transient (under 3 hours)4/558Guinot et al103 2021- Parallel- Active rTMS + MT (n = 17), sham + MT (n = 19)M1MagVenture MAGPRO system (active coil type not specified)− 10 Hz; 2000 pulses/session, 20 min/session- 80% RMTDedicated sham coil identical in size, color, shape, and acoustic output- Induction: 10 sessions over 2 weeks (5 sessions/week)- Maintenance: 2 sessions in week 3, then weekly sessions in weeks 4, 6, 9, and 136 months post-treatmentNo significant difference in pain relief was observed between active rTMS and sham over time. Pain intensity, cardiorespiratory fitness, quality of life, depression, and catastrophizing improved significantly by week 14 and remained stable through week 40, while cardiac autonomic function and sleep efficiency showed no significant changesNo adverse effects recorded during the study4/559Lacroix et al104 2022- Parallel- Active rTMS (n = 41), sham (n = 37)Left M1Magstim Rapid (active coil type not specified)− 10 Hz; 20 trains/session, 10s/train, ITI 50s, total 2000 pulses/session- 90% RMTDedicated sham coil reproducing acoustic output and scalp sensations without inducing a magnetic field- Induction: 5 sessions/week for 3 weeks (15 sessions)- Maintenance: 1 session every 21 days up to 6 months (in 9 patients responding after induction phase)Assessments on day 21, day 42, and month 6A significant clinical response to rTMS was observed following the induction phase and was sustained over 6 months, particularly reflected by improvements in PGIC for pain, fatigue, and depression- Migraine (2/3)- No other adverse events reported4/560Silva et al105 2025- Parallel- Active rTMS (n = 52), sham (n = 49)Left M1Figure-of-eight coil− 10 Hz; 30 trains/session, 100 pulses/train, ITI 20s, total 3000 pulses/session- 80% RMTDedicated sham coils identical in size, color, and shape, producing similar sounds- Induction: 10 daily sessions- Maintenance: weekly over 6 weeks, extended maintenance: fortnightly over 8 weeksAssessment at weeks 8 and 16Adjunctive M1-rTMS reduced pain intensity up to 8 weeks; analgesia diminished at week 16. Functional improvements were maintained at week 16No severe adverse events, seizures, or deaths4/561Short et al106 2011- Parallel- Active rTMS (n = 10), sham (n = 10)Left DLPFCSolid focal coil− 10 Hz; 80 trains/session, 5s/train, ITI 10s, total 4000 pulses/session- 120% RMTShielded sham coil reproducing acoustic output, combined with synchronized electrical scalp stimulation10 sessions over 2 weeks (5 sessions/week)1 and 2 weeks post-treatmentNo significant difference between groups in pain relief was observedTwo participants reported headaches after the first session4/562Avery et al107 2015- Parallel- Active rTMS (n = 7), sham (n = 11)Left DLPFCNeuronetics coil (geometry unspecified)− 10 Hz; 75 trains/session, 4s/train, ITI 26s- 120% MTShielded sham coil with similar acoustic output and forehead electrical stimulation to reproduce scalp sensations15 sessions over 4 weeksAfter 5, 10, and 15 sessions; 1 week, 1 month, and 3 months post-treatmentBoth groups showed significant pain relief with no significant between-group differenceNeuropsychological and audiometric assessments showed no significant between-group differences5/563Fitzgibbon et al108 2018- Parallel- Active rTMS (n = 14), sham (n = 12)Left DLPFCFigure-of-eight coil− 10 Hz; 75 trains/session, 4s/train, ITI 21s, total 3000 pulses/session- 120% RMTDedicated sham coil reproducing sound, active coil angled 45° away from the scalp20 sessions (daily Mon–Fri for 4 weeks)Week 4, 1-month post-treatmentNo significant improvement in pain severity and impact between groups was observed, but active rTMS significantly increased the responder rate compared with shamHeadache (4/3), stimulation-site discomfort (4/1), neck pain (0/2), nausea (1/2), dizziness (1/0), and other adverse events (1/1)4/564Cheng et al109 2019- Parallel- Active rTMS (n = 9), sham (n = 11)Left DLPFCFigure-of-eight coil− 10 Hz; 40 trains/session, 4s/train, ITI 26s, total 1600 pulses/session- 100% MTDedicated placebo coil mimicking acoustic and somatosensory stimulation.10 sessions total, 5 sessions/week for 2 weeksAt week 1, and week 2 during treatmentActive rTMS significantly reduced pain over the 2-week treatment compared with sham.

Analgesic response varied by depression severity: in mild-to-moderate depression, active rTMS produced consistently lower pain than sham across the 2-week treatment period, whereas in severe depression, no significant between-group difference was found, but pain decreased over time with active rTMS and worsened in the sham groupOne participant reported mild dizziness; no seizures or headaches reported5/565Bilir et al110 2021- Parallel- Active rTMS (n = 10), sham (n = 10)Left DLPFCFigure-of-eight coil− 10 Hz; 5s/train, ITI 25s, total 1500 pulses/session- 90% RMTReversely positioned coil delivering 1% RMT to produce clicking sounds without intended cortical stimulation- Induction: 5 sessions for 2 weeks- Maintenance: weekly sessions for 4 weeks- Total: 14 sessionsAssessments in weeks 2 and 6No significant differences in pain score or FSS were observed between or within groups over time. VAS-stiffness and FIQ improved in week 2 vs baseline in the active rTMS group but without significant differences between groups; no significant changes were observed in HADS or cognitive outcomesNo adverse events reported4/566Ackermann et al111 2025- Parallel- Active rTMS (n = 14), sham (n = 13)Left DLPFCFigure-of-eight coil− 10 Hz; 40 trains/session, 4s/train, ITI 26s, total 1600 pulses/session- 100% RMTActive coil positioned at a 45° angle to minimize cortical stimulation10 daily sessions (once daily on weekdays)Clinical follow-up 8 weeks post-treatmentPain scores improved over time in both groups, with no significant between-group difference, indicating a strong placebo response with symptom reduction in both groups- Active: local pain, tingling, facial muscle contractions, and transient headaches (lasting up to a few hours)- Sham: scalp pressure- No serious adverse events reported3/567Lee et al112 2012- Parallel- 1-Hz rTMS (n = 5), 10-Hz rTMS (n = 5), sham (n = 5)− 1 Hz: Right DLPFC- 10 Hz: Left M1Figure-of-eight coil− 1 Hz: 2 trains/session, 800 pulses/train, ITI 60s; total 1600 pulses/session- 10 Hz: 25 trains/session, 80 pulses/train, ITI 10s, total 2000 pulses/session- 1 Hz: 110% RMT; 10 Hz: 80% RMTActive figure-of-eight coil positioned perpendicular (90°) to the skull, using the low-frequency stimulation protocol10 consecutive sessions (5 sessions/week over 2 weeks)Immediate and 1-month post-treatmentWithin-group differences in VAS and BDI were significant immediately post-treatment, with the largest improvement in 1-Hz rTMS, but these effects were not sustained at 1 monthOne high-frequency rTMS participant experienced an epileptic-like event after the sixth session and discontinued treatment; EEG did not show significant epileptiform discharges2/568Kankane et al113 2024- Parallel1-Hz rTMS (n = 30), 10-Hz rTMS (n = 30), sham (n = 30)− 1 Hz: Right DLPFC- 10 Hz: Left DLPFCFigure-of-eight coil− 1 Hz: 150 pulses/train, ITI 60s, 10 sessions- 10 Hz: 10 pulses/train, ITI 20s, 10 sessions- 80% RMTDedicated sham coil5 sessions/week, total 10 sessions over 2 weeksImmediately, 1 month, and 3 months post-treatmentBoth 1-Hz and 10-Hz rTMS significantly improved NPRS, HAM-A, HDRS, and FIQR scores immediately after treatment and up to 3 months, with no significant difference between stimulation frequenciesDizziness: high-frequency n=2, low-frequency n=1; Headache: n=1 in each active group3/569Altas et al114 2019- Parallel- M1 rTMS (n = 10), DLPFC rTMS (n = 10), sham (n = 10)Left M1, left DLPFCFigure-of-eight coil− 10 Hz; 60 trains/session, 2s/train, ITI 28s, total 1200 pulses/session- 90% RMTReversely positioned coil over the vertex at 1% RMT, producing clicking sounds without intended cortical stimulation5 sessions/week, total 15 sessions over 3 weeksImmediately after the 3-week treatment coursePain decreased significantly within all groups after treatment; between-group differences were significant only for M1 vs sham, with greater pain reduction in the M1 groupNo adverse events reported4/570Tilbor et al115 2024- Parallel- Active dTMS (n = 8), sham (n = 11)ACC–mPFCH7− 20 Hz; 50 trains/session, 2s/train, ITI 20s, total 2000 pulses/session- Up to 100% MTIntegrated H7 coil used in reverse configuration within the same helmet5 sessions/week, total 20 sessions over 4 weeksAssessments during the 4-week treatment course: day 10, day 20A significant group × time interaction for SF-MPQ scores indicated that pain-related sensory and affective symptoms decreased significantly more in the dTMS group than in the sham group.Treatment was reported to be safe and well tolerated3/571Maestú et al116 2013- Parallel- Very low intensity rTMS (n = 28), sham (n = 26)Whole head (non-focal stimulation)Custom-built multicoil EEG-cap system− 8 Hz (square wave)- Magnetic field approximately 43 nT at 1 cm, 0.9 nT at 4 cm, current amplitude 545 µA, very low-intensity rTMSIdentical inactive magnetic stimulation device and EEG cap; no magnetic field generated20 min/session, once per week, 8 consecutive weeksWeekly assessments throughout 8 weeks of treatmentActive stimulation significantly improved pain thresholds after the first session and perceived chronic pain, daily activities, and sleep quality after 6 weeks, with no significant effects on depression, fatigue, headache severity, or serotonin levels compared with shamNo adverse effects reported4/572Forogh et al117 2021- Parallel- rTMS (n = 15), tDCS (n = 15)Left DLPFCFigure-of-eight coil− 10 Hz; ITI 15s, total 1000 pulses/session- 100% RMTNo sham3 sessions over 1 week (every other day)Immediately, 6 and 12 weeks post-treatmentrTMS produced greater and more sustained VAS score reduction than tDCS up to 12 weeks after course termination, with significantly higher responder rates. The interaction between time and group was not significant for DASS-21 or FIQRMild, transient, self-limiting headache in two rTMS participants3/573Izquierdo-Alventosa et al118 2021- Parallel- Active rTMS (n = 17), PE (n = 16), control group (n = 16)Left M1Figure-of-eight coil− 10 Hz; 60 trains/session, 5s/train, ITI 15s, 50 pulses/train; total 3000 pulses/session- 80% RMTNo sham10 sessions over 2 weeks (5 sessions/week)Immediately after treatment (week 2 for TMS; week 8 for the other groups)Both PE and 10-Hz rTMS improved pain, FM impact, physical fitness, and emotional outcomes, with 10-Hz rTMS producing greater improvements in emotional outcomesTreatment-related adverse events not specifically reported3/5Complex Regional Pain Syndrome74Pleger et al119 2004- Crossover- Active rTMS, sham (n = 10)M1Figure-of-eight coil− 10 Hz; 10 trains/session, 1.2s/train, ITI 10s- 110% RMTActive coil tilted 45°, with only its edge resting on the scalpSingle session per conditionImmediately, 30 seconds, 15, 45, and 90 minutes after each sessionSeven of the 10 patients reported pain relief following stimulation.

Pain relief began 30s after stimulation, peaked at 15 min, and reintensified approximately 45 min after rTMS; sham stimulation was ineffectiveParesthesia (n=2), Hemihyperthermaesthesy (n=2), tiredness (n=1), dizziness (n=1), mild headache (n=1)3/575Picarelli et al120 2010- Parallel- Active rTMS (n = 11), sham (n = 11)M1Figure-of-eight coil− 10 Hz; 25 trains/session, 10s/train, ITI 60s, total 2500 pulses/session- 100% RMTDedicated inactive figure-of-eight sham coil, producing sound similar to the active coil10 daily sessions (once daily on weekdays)Daily during 10 sessions; 1 week and 3 months post-treatmentActive rTMS significantly reduced VAS during the stimulation period compared with sham from session 2 to session 10; this difference did not persist at 1-week or 3-month follow-upGeneralized seizure (active: n=1); headache (6/4), neck pain (2/4), nonpainful scalp sensations (4/2), dizziness (1/1)4/576Qiu et al121 2025- Parallel- rTMS + MLD (n = 17), rTMS alone (n = 18), sham (n = 18)M1Figure-of-eight coil− 10 Hz; 1.2s/train, ITI 10s, total 1200 pulses/session- 100% MTActive figure-of-eight coil tilted 45°, with only the coil edge touching the scalp15 sessions, daily on weekdays for 3 weeksImmediately after the 3-week interventionrTMS significantly improved CRPS severity compared with sham; rTMS + MLD significantly improved pain, hand edema, motor function, CRPS severity, anxiety, and depression. Compared with rTMS alone, rTMS + MLD provided additional improvements in all outcomesAdverse events not specifically reported3/5Abbreviations: ACC, anterior cingulate cortex; BDI, Beck Depression Inventory; CRPS, complex regional pain syndrome; DASS-21, Depression Anxiety Stress Scale-21; DLPFC, dorsolateral prefrontal cortex; dTMS, deep transcranial magnetic stimulation; EEG, electroencephalography; FIQ, Fibromyalgia Impact Questionnaire; FIQR, Revised Fibromyalgia Impact Questionnaire; FM, fibromyalgia; FSS, Fatigue Severity Scale; HAM-A, Hamilton Anxiety Rating Scale; HDRS, Hamilton Depression Rating Scale; ITI, intertrain interval; M1, primary motor cortex; MLD, manual lymphatic drainage; mPFC, medial prefrontal cortex; MT, motor threshold; nT, nanotesla; NPRS, Numeric Pain Rating Scale; PE, physical exercise; PGIC, Patient Global Impression of Change; RMT, resting motor threshold; rTMS, repetitive transcranial magnetic stimulation; SF-MPQ, Short-Form McGill Pain Questionnaire; tDCS, transcranial direct current stimulation; VAS, visual analog scale.

Fibromyalgia

FM is a chronic pain condition marked by persistent widespread musculoskeletal pain and a wide range of associated symptoms, including fatigue, sleep disturbance, cognitive dysfunction, and mood abnormalities.122 FM affects approximately 1.4% of the general population worldwide.123 FM management typically includes nonpharmacological interventions such as exercise and cognitive behavioral therapy, as well as pharmacological treatments including antidepressants and anticonvulsants.124 However, a proportion of patients exhibit an inadequate response to pharmacological treatment. In such cases, alternative therapeutic approaches, including rTMS, may be considered. rTMS has been widely studied in patients with FM because it may not only alleviate pain but also improve associated symptoms.

Our literature search identified 21 studies98–118 evaluating the analgesic efficacy of rTMS in patients with FM. Fourteen studies98–111 compared conventional rTMS with sham stimulation, 2 studies112,113 examined different stimulation frequencies, 1 study114 compared stimulation at different cortical targets, 1 study115 investigated dTMS, 1 study116 evaluated very low-intensity rTMS, and 2 studies117,118 compared rTMS with other therapeutic interventions.

Three parallel RCTs98–100 evaluated the analgesic effects of LF-rTMS (1 Hz) applied to the right DLPFC in patients with FM. In 2009, Carretero et al found both active and sham groups showed improvements in fatigue and global clinical status with no between-group differences, whereas pain scores remained unchanged.98 In contrast, a larger RCT by Tanwar et al demonstrated that active rTMS significantly reduced pain and related symptoms after treatment, with effects sustained for up to 6 months after treatment; no significant pain reduction was observed with sham stimulation.99 More recently, Tiwari et al also observed greater pain reduction with active rTMS than with sham, alongside improvements in cognition and sleep quality within only the active group.100

In this review, 5 studies101–105 evaluated 10-Hz M1-rTMS in patients with FM, with generally favorable but inconsistent findings. Passard et al, and Mhalla et al reported significantly greater pain reduction with active rTMS than sham, lasting up to 2 weeks and 1 month after treatment, respectively.101,102 Lacroix et al reported significant clinical benefits of active rTMS following induction, with a trend toward sustained benefits over six months of maintenance treatment, whereas Silva et al found significantly greater pain reduction with active rTMS than sham at week 8, but not at week 16.104,105 In contrast, Guinot et al found that active rTMS plus a multicomponent rehabilitation program (MT) showed no significant difference in pain relief compared with sham rTMS plus MT.103 rTMS appears to improve fatigue and reduce the impact of pain on daily life; however, its effects on other symptoms, particularly mood-related symptoms such as depression, appear inconsistent.

Our review also revealed 6 studies106–111 evaluating the effects of HF-rTMS targeting the left DLPFC in patients with FM. Four studies found no significant difference in pain reduction between active and sham rTMS.106,107,110,111 Fitzgibbon et al similarly found no significant difference in pain severity but reported a higher responder rate with active rTMS.108 By contrast, Cheng et al in 2019 reported that active rTMS significantly reduced pain over a 2-week treatment period compared with sham stimulation in patients with FM and comorbid MDD.109 Interestingly, analgesic response varied according to baseline depression severity. Among patients with mild-to-moderate depression, active rTMS produced consistently lower pain scores than sham throughout the 2-week treatment period.

In contrast, among patients with severe depression, no significant between-group differences were observed; however, pain decreased over time in the active rTMS group but worsened in the sham group.

Two studies112,113 compared different rTMS frequencies in patients with FM. Lee et al compared 1-Hz over right DLPFC with 10-Hz over left M1 stimulation, reporting significant immediate within-group pain reduction, greatest with 1-Hz rTMS, but not sustained at 1 month after treatment.112 More recently, Kankane et al evaluated 1-Hz rTMS over the right DLPFC and 10-Hz rTMS over the left DLPFC, reporting improvements in pain and related symptoms lasting up to 3 months after treatment, with no significant difference between the stimulation frequencies.113

In our review, five remaining studies examined distinct stimulation approaches in patients with FM.114–118 Altas et al compared 10-Hz rTMS targeting M1 and the left DLPFC with sham stimulation and reported that pain decreased significantly in all three groups, only M1 stimulation produced significantly greater pain reduction than sham.114 Tilbor et al also showed that active dTMS targeting the anterior cingulate cortex (ACC)–medial PFC significantly reduced sensory and affective pain symptoms compared with sham.115 Maestú et al found that very low-intensity rTMS significantly reduced subjective chronic pain versus sham after 6 weeks.116 Compared with tDCS, Forogh et al reported that rTMS produced greater and more sustained VAS score reduction than tDCS up to 12 weeks after course termination, with significantly higher responder rates.117 Compared with physical exercise (PE), Izquierdo-Alventosa et al also reported that both PE and 10-Hz rTMS improved pain, FM impact, physical fitness, and emotional outcomes, with 10-Hz rTMS producing greater improvements in emotional outcome than PE.118

Overall, the findings suggest potential immediate and short-term analgesic benefits of rTMS in patients with FM, with HF-rTMS for M1 or DLPFC stimulation. However, substantial heterogeneity in stimulation targets, treatment duration, maintenance schedules, and patient characteristics, particularly comorbid depression, may contribute to inconsistent findings. Follow-up periods also varied considerably, with many studies assessing outcomes only during treatment or shortly thereafter, leaving long-term efficacy uncertain.

Further well-designed trials are needed to determine the optimal rTMS protocol in this population.

Complex Regional Pain Syndrome

CRPS is a chronic, debilitating pain condition accompanied by sensory, motor, autonomic, and trophic abnormalities.125 CRPS is thought to arise from multiple interacting pathophysiological mechanisms, including central sensitization, dysfunction of descending pain modulatory pathways, and maladaptive neuroplasticity.126 Based on these alterations in pain processing pathways, rTMS has been explored as a potential therapeutic approach.

Three RCTs119–121 in our review investigated the efficacy of rTMS for treating CRPS. Pleger et al conducted a crossover study involving 10 right-handed patients with CRPS type I who received a single session of 10-Hz rTMS applied to the M1 contralateral to the painful side.119 Seven of the 10 patients reported pain relief following stimulation. The analgesic effect began approximately 30 seconds after stimulation, peaked at 15 minutes, and pain re-intensified by 45 minutes, whereas sham stimulation was ineffective.

Picarelli et al similarly observed greater pain reduction with active 10-Hz M1-rTMS than sham during a 10-session treatment course, although benefits were not sustained at subsequent follow-ups.120 More recently, Qiu et al found that rTMS with manual lymphatic drainage improved pain and functional outcomes more than rTMS alone in patients with poststroke CRPS.121

Taken together, these findings suggest potential immediate and short-term analgesic benefits with CRPS type I, but evidence for sustained effects remains limited, and the contribution of adjunctive interventions requires further investigation.

Headache

Headache is a major public health problem affecting a substantial proportion of the global population and contributing significantly to years lived with disability.127,128 According to the International Classification of Headache Disorders, 3rd edition (ICHD-3), headache disorders are broadly classified into primary headache disorders, secondary headache disorders, and painful cranial neuropathies and other facial pain and other headaches.129 In this review, rTMS has been investigated in both primary and secondary headache disorders. The primary headache conditions included migraine, tension-type headache, and cluster headache, whereas the secondary headache condition examined was posttraumatic headache following mild traumatic brain injury (mTBI); detailed characteristics of the included studies are summarized in Table 3. Table 3Characteristics of Studies Evaluating Repetitive Transcranial Magnetic Stimulation (rTMS) for Headache or Orofacial PainNo.Author, YearStudy Design/ Group (n)Stimulation SiteCoil Type of Active rTMSActive rTMS ParametersSham ProcedureTreatment CourseFollow -UpSummary of FindingsSafety, n (Active/Sham)Jadad ScoreMigraine77Teepker et al130 2010- Parallel- Active rTMS (n = 14), sham (n = 13)VertexCircular coil− 1 Hz; 2 trains/session, 500 pulses/train, ITI 1 min, 1000 pulses/session- Visual MT − 2%Dedicated figure-of-eight sham coil producing a sound and feeling like active stimulationOnce daily for 5 consecutive daysDaily during 5 treatment days; 8 weeks post-treatmentNo significant differences between active and sham group in pain intensity, migraine attack frequency, or analgesic use- During treatment: Discomfort during motor threshold assessment (5/4), prolonged sitting discomfort (1/1), sleepiness (1/1), headache (0/2)- After treatment: Amyostasia (1/1), testiness (1/1), Vigorous dreams (1/0), and phonophobia (1/0)2/578Misra et al131 2013- Parallel- Active rTMS (n = 47), sham (n = 48)Left frontal cortexFigure-of-eight coil− 10 Hz; 10 trains/session, 60 pulses/train, ITI 45s, total 600 pulses/session- 70% MTDedicated sham stimulator with identical appearance and similar acoustic output, without magnetic stimulation3 sessions on alternate days1-month post-treatmentActive rTMS yielded significantly higher ≥50% responder rates for both headache frequency and VAS versus shamTransient drowsiness (active/sham: 1/0); discomfort during stimulation (all active patients/ 7 sham patients)4/579Conforto et al132 2014- Parallel- Active rTMS (n = 9), sham (n = 9)Left DLPFCFigure-of-eight coil− 10 Hz; 32 trains/session, 5s/train, ITI 30s, total 1600 pulses/session- 110% RMTActive coil positioned perpendicular to the vertex15 sessions in the first 4 weeks, 23 sessions total over 8 weeksAssessments at weeks 4 and 8 during treatmentSham reduced headache days significantly more than active rTMS, suggesting a strong placebo effect; pain intensity improved similarly in both groups- Headache worsening or onset (active/sham: 7/3), sleepiness (7/7), difficulty concentrating (3/0), and transient trigeminal hypoesthesia (1/0)- Two withdrawals due to procedural pain; transient trigeminal-distribution hypesthesia in one participant.- No seizures or syncope reported4/580Kumar et al133 2021- Parallel- Active rTMS (n = 10), sham (n = 10)Left M1Figure-of-eight coil− 10 Hz; 10 trains/session, ITI 60s, total 600 pulses/session- 70% RMTActive figure-of-eight coil positioned perpendicular to the vertex at minimum stimulator output5 sessions/week, 10 sessions total over 2 consecutive weeksImmediately and 1-month post-treatment; headache-related disability additionally assessed at 3 monthsActive rTMS significantly reduced VAS pain intensity and headache frequency compared with baseline, with effects maintained at 1-month follow-up; sham showed no significant changeNo adverse effects reported4/581Brighina et al134 2004- Parallel- Active rTMS (n = 6), sham (n = 5)Left DLPFCFigure-of-eight coil− 20 Hz; 10 trains/session, 2s/train, ITI 30s- 90% MTActive coil positioned perpendicular to the scalp over the left DLPFC, using the same stimulation schedule12 sessions on alternate days (excluding weekends)Post-intervention and 1-month post-treatmentActive rTMS significantly reduced attacks, headache index, and abortive medication use; sham showed no significant improvementNo adverse effects reported4/582Rapinesi et al135 2016- Parallel- Add-on dTMS + SDT (n = 7), SDT (n = 7)Left DLPFCH1 coil− 10 Hz; 10 trains/session, 2s/train, ITI 20s, total 360 pulses/session- 100% MTNo sham3 sessions/week on alternate days, 12 sessions over 4 weeksWeeks 2 and 4 during treatment; 1-month post-treatmentAdd-on dTMS significantly reduced pain intensity, headache frequency, rescue medication use, and depressive symptoms during treatment and at 1 month compared with baseline and the control groupdTMS was well tolerated; no detailed adverse event counts reported2/583Shehata et al136 2016- Parallel- rTMS (n = 14), BTX-A injection (n = 15)Left M1Figure-of-eight coil− 10 Hz; 20 trains/session, 100 pulses/train, ITI 10s, total 2000 pulses/session- 80% MTNo sham3 sessions/week for 1 month, 12 sessions totalWeeks 4, 6, 8, 10, and 12 from baseline; week 4 corresponds approximately to treatment completionBoth BTX-A and rTMS significantly reduced headache frequency and severity at week 4 with no between-group difference; effects persisted to weeks 6–8 in both groups but were sustained to weeks 10–12 only with BTX-A- rTMS: worsening headache (n=2; both withdrew), transient tinnitus (n=1)- BTX-A: injection-site pain (n=5), hematoma (n=2), blepharoptosis (n=1)- No serious adverse events4/584Sahu et al137 2019- Parallel- Active iTBS (n = 20), sham (n = 21)Left DLPFCFigure-of-eight coil- Triplet 50 Hz bursts, repeated at 5 Hz; 20 trains/session, 2s/train, ITI 8s, total 600 pulses/session- 80% AMTActive figure-of-eight coil positioned perpendicular to the scalp over the left DLPFC2 sessions daily, 10 sessions total over 5 consecutive days2, 4, 6, and 12 weeks post-treatmentActive iTBS significantly reduced headache frequency, duration, and severity vs sham; analgesic effect most pronounced at 2 weeks- Transient mild scalp discomfort during active iTBS, which resolved spontaneously after stimulation- No major adverse events reported2/585Kalita et al138 2021- Parallel- rTMS (n = 41), rTMS + amitriptyline (n = 42)Left frontal cortexFigure-of-eight coil− 10 Hz; 10 trains/session, 60 pulses/train, ITI 45s- 70% RMTNo sham3 sessions on alternate days per month, 9 sessions total over 3 monthsAssessments at months 1, 2, and 3 during the repeated monthly treatmentrTMS + amitriptyline yielded significantly higher ≥50% responder rates for headache frequency and severity compared with rTMS alone at 3 monthsNoise-related discomfort and headache during rTMS; dry mouth was more common with adjunctive amitriptyline3/586Naji et al139 2024- Parallel- rTMS (n = 36), tDCS (n = 36)Left DLPFCFigure-of-eight coil− 10 Hz; 10 trains/session, 60 pulses/train, ITI 45s, total 600 pulses/session- 70% RMTNo sham3 sessions1-month post-treatmentBoth groups showed reduced anxiety at follow-up, but only the rTMS group demonstrated significant improvement in depression; no significant between-group differences were observed in pain intensity, anxiety, or headache impact on daily lifeNo adverse events were reported in either group3/5Tension-Type Headache87Mattoo et al140 2019- Parallel- Active rTMS (n = 15), sham (n = 15)Right DLPFCFigure-of-eight coil− 1 Hz; 8 trains/session, 150 pulses/train, ITI 1 min, total 1200 pulses/session- 110% RMTActive coil positioned perpendicular to the right DLPFC at minimal stimulation intensity, retaining acoustic outputOnce per weekday, 5 sessions/week, 20 sessions total over 4 weeks4 weeks post-treatmentActive rTMS significantly reduced pain intensity and increased nociceptive flexion reflex thresholds compared with shamNo specific events were documented in the reported results3/588Rajain et al141 2023- Parallel- Active rTMS (n = 10), sham (n = 10)Right DLPFCFigure-of-eight coil− 1 Hz; Stimulation for 25 min- 110% RMTActive figure-of-eight coil positioned perpendicular to the skull, preventing intended cortical stimulation5 sessions/week, 10 sessions total over 2 weeksWithin 24 hours after the 2-week treatment courseActive rTMS significantly reduced pain intensity, pericranial muscle activity, and headache impact while increasing nociceptive thresholds compared with shamNo adverse events reported3/5Cluster Headache89Portocarrero-Sánchez et al142 2025- Crossover- Active rTMS, sham (n = 8)M1YINGCHI M-100 Ultimate− 10 Hz; 10 trains/session, 60 pulses/train, ITI 60s, total 600 pulses/session- 70% RMTSame device and stimulation schedule without active magnetic stimulation, reproducing the appearance and acoustic outputDaily sessions over 10 consecutive working days per condition, 1-month washoutAssessments at 2, 4 and 12 weeks after the treatment sequenceActive rTMS showed no significant effect on attack frequency, attack intensity, duration, or rescue medication compared with sham; 2 patients achieved transient remission during active rTMS, but symptoms recurred after washoutTwo mild, transient adverse events, with no adverse event-related withdrawals; group-specific safety data were not provided4/5Posttraumatic Headache After mild Traumatic Brain Injury90Leung et al143 2016- Parallel- Active rTMS (n = 12), sham (n = 12)Left M1Figure-of-eight coil− 10 Hz; 20 trains/session, 100 pulses/train, ITI 1s, total 2000 pulses/session- 80% RMTActive coil reversed by 180°, with magnetic shielding facing the scalp, while preserving the treatment setup and neuronavigation procedures3 sessions, intersession interval 24–72 h1 and 4 weeks post-treatmentActive rTMS produced greater reductions in persistent headache intensity than sham at 1 and 4 weeks after treatment; significantly more patients in the active rTMS group achieved ≥50% pain reduction at 1-week post-treatment; rTMS significantly reduced functionally disabling headache exacerbations at 4 weeks versus sham, with no sustained neuropsychological changes- Transient local tenderness (active/sham: 1/0; resolved in a few hours post-treatment) and mild dizziness (1/1)- No long-term neurological complications4/591Leung et al144 2018- Parallel- Active rTMS (n = 14), sham (n = 15)Left DLPFCFigure-of-eight coil− 10 Hz; 20 trains/session, 100 pulses/train, ITI 1s, 2000 pulses/session- 80% RMTActive coil reversed by 180° and covered with two layers of magnetic shielding film, retaining similar procedural conditions4 sessions, intersession interval 24–72 h1 and 4 weeks post-treatmentActive rTMS produced greater reductions in persistent headache intensity and prevalence than sham in 1 and 4 weeks after treatment; depressive symptoms improved significantly at 1 week in the rTMS group but not in the sham groupTransient elevation in a neuropsychological perseveration score in the active group; no other side effects reported4/592Leung et al145 2026- Parallel- Active rTMS (n = 31), sham (n = 38)Left M1Figure-of-eight coil− 10 Hz; 20 trains/session, 100 pulses/train, ITI 2s- 80% RMTIntegrated active/placebo figure-of-eight coil, with sham stimulation reproducing acoustic and procedural characteristics; no additional electrical scalp stimulation10 sessions, intersession interval 24–72 h1 week, 1 month, 2 months and 3 months post-treatmentActive rTMS significantly reduced persistent headache frequency and duration of disabling headaches at 2- and 3-month follow-up, accompanied by improvements in daily function, attention, and memoryTransient headache exacerbation (3/6), abnormal nonpainful sensations (8/4), fatigue (4/2), nausea (3/0), and other mild adverse events; no significant between-group differences4/5Orofacial Pain93Fricová et al146 2013- Parallel- Active rTMS (n = 13), sham (n = 10)M170-mm air-cooled coil (shape not specified)− 20 Hz; 20–36 trains/session, ITI 1.9s, 720 pulses/session- 95% MTDedicated inactive sham coil placed over the same target, reproducing acoustic and scalp sensations1 session daily for 5 daysWithin 15 min after each of five sessions; 2 weeks post-treatmentActive rTMS significantly reduced pain intensity compared with sham; analgesic effect observed during treatment persisted up to 14 days after stimulation; 20-Hz rTMS showed greater analgesic effect than previous 10-Hz stimulation data- Well tolerated- No serious adverse effects reported4/594Lindholm et al147 2015- Crossover- S2 rTMS, S1/M1 rTMS, sham (n = 16)Right S2, S1/M1Figure-of-eight coil− 10 Hz; 50 pulses/train, ITI 10s, total 1000 pulses/session, 15-min break mid-session- 90% RMTActive coil separated from the scalp by a 75-mm plastic spacer, minimizing the induced cortical electric fieldSingle session per condition, 4-week washout between periodsUp to 4 weeks after the final interventionActive rTMS over the right S2 cortex produced greater analgesic effects than S1/M1 stimulation or sham, suggesting that S2 may be a promising target for neuropathic orofacial pain treatment- No serious adverse effects- Unpleasant temporal muscle contractions in two patients during active stimulation2/595Kohútová et al148 2017- Parallel- Active iTBS (n = 10), sham (n = 9)M1Air-cooled, 70-mm coil− 50 Hz bursts (3 pulses/burst) repeated at 5 Hz, 20 trains/session, 2s/train, ITI 8s, 600 pulses/session- 90% MTIntermediate theta-burst stimulation (imTBS) used as sham, with the same target and intensity as active iTBS but a different train structureSingle sessionWithin 60 min after the single session and 2 weeks post-treatmentiTBS significantly reduced pain post-stimulation compared with sham, but no significant difference was observed at 2-week follow-up- Mild, transient headache- No serious adverse effects reported4/596Henssen et al149 2019- Crossover- Bilateral rTMS, unilateral rTMS (n = 12)M1Figure-of-eight coil− 10 Hz; 10 trains/session, 10s/train, ITI 50s, 1000 pulses/hemisphere- 80% RMTActive coil tilted 90° away from the scalp for sham stimulationSingle session of bilateral and unilateral stimulation2 weeks after each stimulation sessionBilateral rTMS produced significantly greater pain reduction than unilateral rTMSAdverse events not specifically reported2/597Säisänen et al150 2022- Crossover- 10-Hz rTMS, 20-Hz rTMS (n = 19)M1Nexstim navigated TMS system− 10 Hz: 6s/train, ITI 24s, total 2400 pulses/session- 20 Hz: 4s/train, ITI 46s, 10-min break, 3600 pulses/session- Maintenance: 10 Hz, 6s/train, ITI 24s, 3000 pulses/session- 90% RMTNo sham− 5 consecutive days per session; single session per condition, 6-week washout- Maintenance for responders: once daily for 5 consecutive days, repeated every 2–3 months depending on clinical response and patient willingnessDuring therapy and 2 weeks after each interventionHF-rTMS reduced pain intensity, and repeated treatment enhanced the analgesic effect, whereas higher stimulation parameters did not provide additional benefitFatigue (n=3; two related to 20 Hz), pain in the head (n=6), dizziness (n=2), and headache, nausea, poor sleep, ophthalmic tingling, hand twitching, hand tingling, and chest pain (n=1 each)2/598Liu et al151 2025(Trigeminal neuralgia)- Parallel- M1 rTMS (n = 13), DLPFC rTMS (n = 10), sham (n = 11)M1, DLPFCFigure-of-eight coil− 10 Hz; 25 trains/session, 100 pulses/train, 10s/train, ITI 50- 80% RMTActive coil rotated 90° to remain perpendicular to the M1 target; an additional group received active DLPFC stimulation5 consecutive daily sessions × 2 blocks separated by a 2-day intervalImmediate and 1-month post-treatmentPain, anxiety, and depression decreased significantly over time in all groups, but no significant differences were observed between groupsNo specific treatment-emergent adverse-event results reported3/599Herrero Babiloni et al152 2024(Temporomandibular disorders)- Crossover- Active rTMS, sham (n = 41)M1Figure-of-eight coil− 20 Hz; 30 trains/session, 50 pulses/train, ITI 30s, total 1500 pulses/session- 80% RMTDedicated sham coilSingle session per condition, 7-day washoutTwice a day, for 21 consecutive daysActive rTMS induced a modest immediate analgesic effect compared with sham, with no sustained effects over 7 days; no significant between-group differences were found in pain interference, sleep quality, or negative affect- Mild, transient headaches (3 active only, 2 sham only, and 3 both conditions), resolving within hours without medication- Transient generalized fatigue in one participant after both conditions.

No other adverse events reported4/5100Umezaki et al1532016(Burning mouth syndrome)- Parallel- Active rTMS (n = 12), sham (n = 8)Left DLPFCFigure-of-eight coil− 10 Hz; 5s/train, ITI 10s, 15 min/session- 110% RMTShielded sham coil combined with synchronized electrical scalp stimulation to reproduce the sensation of active rTMS5 sessions/week over 2 weeksAssessments on days 8, 15, 30, and 60 from treatment initiationIn the active rTMS group, pain intensity decreased by 67%, with 75% of patients achieving >50% pain reduction; active rTMS over left DLPFC significantly reduced pain after 1 week of treatment, whereas sham did not; functional status showed a similar trend, but mood and the affective component of pain remained unchanged- Mild, transient headache (7/5)- No withdrawals attributed to adverse effects3/5Abbreviations: AMT, active motor threshold; BTX-A, botulinum toxin type A; DLPFC, dorsolateral prefrontal cortex; dTMS, deep transcranial magnetic stimulation; HF-rTMS, high-frequency repetitive transcranial magnetic stimulation; iTBS, intermittent theta-burst stimulation; ITI, intertrain interval; M1, primary motor cortex; MC, motor cortex; MT, motor threshold; RMT, resting motor threshold; rTMS, repetitive transcranial magnetic stimulation; S1, primary somatosensory cortex; S2, secondary somatosensory cortex; SDT, standard drug therapy; tDCS, transcranial direct current stimulation; VAS, visual analog scale.

Migraine

Migraine is one of the most common and disabling primary headache disorders, characterized by recurrent attacks of moderate-to-severe headache, typically unilateral and pulsating in nature, and often accompanied by nausea, photophobia, and phonophobia.129,154 Migraine involves alterations in brain excitability, repeated activation and sensitization of the trigeminovascular system, and subsequent structural and functional brain changes in susceptible individuals.155 Based on these pathophysiological mechanisms, rTMS has emerged as a promising therapeutic option for migraine.

Our literature search identified 5 RCTs130–134 comparing conventional rTMS with sham stimulation in patients with migraine. The study by Teepker et al was the only trial in this review that investigated LF-rTMS.130 After 5 consecutive sessions, no significant between-group differences were observed compared with sham stimulation in pain intensity, migraine attack frequency, or analgesic use. Four studies131–134 included in this review compared HF-rTMS with sham stimulation.

Misra et al, Kumar et al, and Brighina et al reported favorable outcomes with active stimulation, including reductions in headache frequency and pain intensity.131,133,134 In contrast, Conforto et al found that sham stimulation reduced headache days significantly more than active rTMS, whereas pain intensity improved to a similar extent in both groups, suggesting a substantial placebo effect.132

Five RCTs evaluated alternative stimulation approaches and comparative treatments for migraine.135–139 Rapinesi et al reported that add-on dTMS significantly reduced pain intensity, headache frequency, rescue medication use, and depressive symptoms during treatment and at 1 month compared with baseline and the control group.135 In the same year, Shehata et al found that both botulinum toxin type A (BTX-A) injection and rTMS significantly reduced headache frequency and severity at week 4, with no between-group difference; effects persisted to weeks 6–8 in both groups but were sustained to weeks 10–12 only with BTX-A.136 Sahu et al concluded that active iTBS significantly reduced headache frequency, duration, and severity compared with sham.137 The analgesic effect was most pronounced at 2 weeks and persisted for up to 12 weeks after treatment. In 2021, Kalita et al showed that rTMS combined with amitriptyline resulted in significantly higher responder rates for headache frequency and severity than rTMS alone at 3 months during the repeated monthly treatment.138 Naji et al in 2024 reported no significant differences between rTMS and tDCS in pain intensity, anxiety, or headache-related disability. Anxiety improved in both groups, whereas depression improved significantly only in the rTMS group.139

Taken together, available RCTs suggest that rTMS—particularly HF-rTMS protocols targeting the left M1 or left DLPFC—may provide therapeutic benefits for patients with migraine, potentially lasting up to 3 months after treatment, although findings remain heterogeneous.

Tension-Type Headache

Tension-type headache (TTH) is characterized by recurrent episodes of mild-to-moderate bilateral head pain, usually described as a pressing or tightening sensation that is not aggravated by normal daily activities.156 The pathophysiology of TTH is thought to be multifactorial, involving both peripheral and central mechanisms. One proposed mechanism underlying TTH is increased sensitivity of the central nervous system, particularly within the trigeminal system.157 Therefore, rTMS has been explored as a therapeutic approach for TTH.

Our review revealed 2 RCTs140,141 investigating the effect of rTMS in treating patients with TTH. Both studies evaluated 1-Hz rTMS targeting the right DLPFC and reported significantly greater reductions in pain intensity compared with sham stimulation.140,141 Rajain et al additionally reported improvements in pericranial muscle activity, nociceptive thresholds, and headache impact of active rTMS compared with sham stimulation.141

Overall, two small RCTs suggest potential analgesic benefits of 1-Hz rTMS targeting the right DLPFC in patients with TTH. However, evidence remains insufficient to establish the clinical significance and durability of these effects.

Cluster Headache

Cluster headache (CH) is a primary headache disorder characterized by recurrent attacks of severe unilateral pain, typically located in the periorbital region, and frequently accompanied by ipsilateral autonomic symptoms such as lacrimation, nasal congestion, or ptosis.158 The mechanism of CH is considered multifactorial, involving hypothalamic dysfunction, abnormalities in the trigeminovascular system, and dysregulation of the trigeminal–autonomic reflex.159 rTMS has been investigated as a potential treatment for cluster headache; however, the number of available studies remains limited.

Our review revealed only 1 study142 by Portocarrero-Sánchez et al in 2025 that evaluated 10-Hz rTMS targeting the M1 contralateral to the painful side in patients with CH. No significant differences between active and sham stimulation were observed in attack frequency, intensity, duration, or rescue medication use. Although two of the eight patients experienced transient remission, their symptoms recurred after washout.

Current evidence remains insufficient to establish the analgesic efficacy or durability of rTMS in patients with CH.

Posttraumatic Headache After Mild Traumatic Brain Injury (mTBI)

mTBI is a neurological condition typically resulting from external mechanical forces to the head and is often associated with symptoms including headache, cognitive impairment, and sleep disturbance.160 Posttraumatic headache after mTBI is thought to arise from multiple interacting mechanisms, including abnormalities in central pain processing, neuroinflammation, and structural and functional brain changes.161 rTMS has also been applied as a potential treatment for posttraumatic headache following mTBI.

Our review identified 3 RCTs143–145 evaluating the efficacy of rTMS for posttraumatic headache following mTBI. All three investigated 10-Hz rTMS, with two targeting the left M1 and one targeting the left DLPFC. The earlier studies reported greater reductions in headache intensity than sham for up to 4 weeks, with additional improvements in ≥50% pain response, disabling headache exacerbations, and short-term depressive symptoms following active rTMS.143,144 A subsequent M1-targeted RCT demonstrated reductions in persistent headache frequency and disabling headache duration at 3 months post-treatment, accompanied by improvements in daily functioning and some cognitive outcomes.145

These findings are generally consistent, indicating that rTMS may provide short-term benefit for posttraumatic headache after mTBI. However, the small number of RCTs precludes firm conclusions regarding the optimal target and durability of benefit.

Orofacial Pain

Orofacial pain (OFP) is a common condition worldwide, affecting approximately 10–15% of the adult population.162 According to the International Classification of Orofacial Pain (IOP), OFP disorders are categorized into several groups, including pain attributed to disorders of dentoalveolar structures, myofascial orofacial pain, temporomandibular joint pain, pain attributed to lesions or diseases of cranial nerves, pain resembling primary headaches, and idiopathic OFP.163 In this review, 8 studies investigated the effects of rTMS in patients with OFP. Among these, 5 studies146–150 examined OFP in general, while the remaining studies151–153 focused on specific ICOP categories, including trigeminal neuralgia (pain attributed to cranial nerve lesions), temporomandibular disorders (TMD, temporomandibular joint pain), and burning mouth syndrome (BMS, idiopathic OFP). The detailed characteristics of the included studies are summarized in Table 3.

Evidence for rTMS in orofacial pain remains limited and heterogeneous, but most studies suggest short-term pain relief with high-frequency stimulation.146–150 Fricová et al reported that five sessions of 20-Hz rTMS over the contralateral M1 significantly reduced refractory facial pain compared with sham, with analgesic effects lasting up to 14 days after treatment; 20-Hz stimulation also showed greater analgesic effect than previous 10-Hz stimulation data.146 However, another study found that both 10- and 20-Hz M1 rTMS reduced pain intensity, with repeated treatment enhancing analgesia but no clear additional benefit from higher stimulation parameters.150 Regarding stimulation site, Lindholm et al found that right secondary somatosensory cortex (S2) stimulation produced greater analgesic effects than sensorimotor cortex (S1/M1) or sham stimulation.147 In painful trigeminal neuropathy, bilateral 10-Hz M1 rTMS also produced greater pain reduction than unilateral stimulation.149 Beyond conventional HF-rTMS, M1 iTBS stimulation produced a significant immediate reduction in pain intensity compared with sham, although this effect was not maintained at the 2-week after treatment.148

Beyond these studies, rTMS has also been evaluated in several other orofacial pain conditions. In trigeminal neuralgia, a three-arm RCT found that pain, anxiety, and depression improved over time after 10-Hz rTMS targeting either M1 or the DLPFC, but neither active protocol differed significantly from sham.151 In TMD, a single session of 20-Hz M1 rTMS produced only a modest immediate analgesic effect, which was not sustained over 7 days.152 In BMS, 10-Hz left DLPFC rTMS significantly reduced pain after 1 week, whereas no significant improvement was observed with sham stimulation.153 In the active rTMS group, pain intensity decreased by 67%, with 75% of patients achieving more than a 50% reduction in pain. Functional status showed a similar trend, whereas mood and the affective component of pain remained unchanged.

Overall, rTMS may provide short-term analgesic benefits in orofacial pain, but responses vary across conditions and protocols. Differences in stimulation target and laterality may influence outcomes, while the optimal protocol remains uncertain.

Musculoskeletal Pain

Musculoskeletal pain is highly prevalent and represents one of the leading causes of disability worldwide.164 In this review, rTMS has been investigated as a therapeutic approach for several musculoskeletal pain conditions, including knee osteoarthritis (KOA), low back pain (LBP), poststroke shoulder pain (PSSP), and myofascial pain syndrome (MPS). The detailed characteristics of the included studies are summarized in Table 4. Table 4Characteristics of Studies Evaluating Repetitive Transcranial Magnetic Stimulation (rTMS) for Musculoskeletal PainNo.Author, YearStudy Design/ Group (n)Stimulation SiteCoil Type of Active rTMSActive rTMS ParametersSham ProcedureTreatment CourseFollow -UpSummary of FindingsSafety, n (Active/Sham)Jadad scoreKnee Osteoarthritis101Cheng et al165 2024- Parallel- Celecoxib group (n = 75), HAE + active rTMS (n = 75)M1Circular coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s- 90% RMTNo sham5 sessions/week over 4 weeksAfter 4 weeks of treatmentAfter 4 weeks, the combined treatment group showed significantly greater reduction in WOMAC scores than the celecoxib group, with greater improvements in knee function, range of motion, and biomechanical parameters.

Treatment-related adverse events were less frequent in the combined treatment group and were all mildCombined treatment/celecoxib: gastrointestinal reactions (0/6), cardiovascular events (0/2), neurological reactions (2/0)3/5102Chang et al166 2025- Parallel- Active rTMS + QSE (n = 16), sham + QSE (n = 12)M1Figure-of-eight coil− 10 Hz; 30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- 90% RMTIdentical-looking inactive sham coil producing clicking soundsTwice weekly for 6 weeks (12 sessions), rTMS immediately before 30 min supervised strengthening exerciseImmediate and 3 months post-treatmentBoth groups showed significant pain reduction post-intervention but did not maintain it at 3-month follow-up on NRS. Function improved significantly only in the rTMS + QSE group post-intervention- No reported adverse events in either group; transient tooth-filling and facial sensations occurred during active stimulation- One sham-group participant withdrew following acute knee pain exacerbation after the first session, attributed to strengthening exercise3/5103Dantas et al167 2025- Parallel- Active rTMS (n = 12), active rTMS + MT (n = 12), sham rTMS + MT (n = 12)M1Figure-of-eight coil− 10 Hz; 24 trains/session, 5s/train, ITI 45s- 90% RMTDisconnected coil with recorded stimulation soundsSingle sessionImmediately, 5, 15 and 30 days post-treatmentMT applied immediately after rTMS attenuated the analgesic effect, whereas a single rTMS session reduced pain intensity in patients with KOANo side effects reported in any group3/5104Hodkinson et al168 2025- Parallel- Accelerated iTBS (n = 31), sham(n = 12)Left DLPFCFigure-of-eight coil− 3 pulses/burst at 50 Hz, bursts every 200 ms, 1800 pulses/block, 5 blocks/day, interblock interval 50 min, total 9000 pulses/day- 80% RMTCoil positioned perpendicular to the scalp at reduced output, reproducing noise and vibration4 consecutive daysImmediately and at weeks 4, 8, 12, and 16Pain relief occurred immediately after treatment and persisted for up to 16 weeks with clinically meaningful reduction in pain, although no significant between-group differences were found- No adverse events or serious adverse events in either group- Six active participants (19.4%) required reduced stimulation intensity due to mild discomfort. All sham participants tolerated the full stimulation dose3/5105Wang et al169 2025- Parallel- Active rTMS + QSE (n = 22), sham + QSE (n = 22)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 7s/train, ITI 55s, total 1400 pulses/session- 80% RMTSham coil positioned perpendicular to the scalp, rotated 90° away to reproduce stimulation noiseOnce daily, 5 days/week for 12 weeksAssessments after 12 weeks of treatment; no separate post-treatment follow-upActive rTMS + QSE significantly reduced pain more than sham + QSE at 12 weeks.

WOMAC scores and several biomechanical outcomes improved in both groups, with significantly greater improvements in the experimental group- Transient scalp pain (1/0)- No serious adverse events3/5Low Back Pain106Ambriz-Tututi et al170 2016- Parallel- Active rTMS (n = 41), PT (n = 12), sham (n = 26)Left M1/S1Figure-of-eight coil− 20 Hz; 10 trains/session, 10s/train, ITI 28s- 95% RMTActive coil directed 180° away from the scalp at reduced output5 daily sessions in week 1, 8 maintenance rTMS sessions at weeks 3, 4, 6, 8, 12, 20, 28, and 36, 1-week post-protocol treatment inversionAssessments extending to week 36, including maintenance stimulationBy week 3, rTMS reduced pain by nearly 80% from baseline, significantly greater than reduction with PT; sham stimulation showed no significant effect until patients subsequently received rTMSMild headache and fatigue (3 cases in total; group-specific counts are not reported)2/5107Bursali et al171 2021- Parallel- Active rTMS (n = 10), sham (n = 10)M1Figure-of-eight coil− 5 Hz; 20 min/session, 1000 pulses/session- 70% RMTInactive device with recorded stimulation soundsDaily sessions, 5 days/week for 2 weeks, total 10 sessionsTreatment days 5 and 10; 1 and 3 months post-treatmentActive rTMS significantly improved DN4, ODI, BDI, and PSQI scores compared with the sham group. Although both groups showed reductions in VAS scores, improvement in the sham group was limited to the first month, whereas rTMS benefits were sustained for up to 3 months- Mild headache (1/0) recovered spontaneously- No other reported adverse events4/5108Li et al172 2024- Parallel- Active rTMS (n = 8), SE (n = 10), active rTMS + SE (n = 10)M1Figure-of-eight coil− 20 Hz; 1s/train, ITI 28s, session duration 20 min- 120% RMTNo sham5 sessions/week, total 10 sessions over 2 weeksImmediately after 2 weeks of treatmentCombined SE + rTMS improved clinical and neuroadaptive outcomes compared with rTMS alone but showed no significant difference vs SE aloneNot reported3/5109Chang et al173 2026- Parallel- M1 rTMS (n = 12), DLPFC rTMS (n = 12), sham (n = 12)Left M1, right DLPFCFigure-of-eight coil- M1: 10 Hz; 30 trains/session, 50 pulses/train, ITI 25s, 1500 pulses/session- DLPFC: 1 Hz; 150 trains/session, 10 pulses/train, ITI 2, 1500 pulses/session- 80% RMTDedicated sham coil reproducing auditory and sensory effects5 sessions/week, total 10 sessions over 2 weeksImmediate and 1-month post-treatmentBoth M1 rTMS and DLPFC rTMS significantly reduced pain vs sham post-treatment. DLPFC rTMS may confer additional advantages for addressing sleep disturbanceAcross all three groups: headache (10 events), dizziness (5), scalp pain (5), and nocturnal arousal (4).

Group-specific counts were not reported; no serious adverse events5/5110Li et al174 2025- Parallel- Active rTMS + rPMS (n = 15), sham rTMS + rPMS (n = 15)M1YRD-CCY-IA stimulation system- rTMS: 20 Hz; 1200 pulses/session- rPMS: 20 Hz; 4000 pulses/session- rTMS: 90% RMT; rPMS: 35–40% MSOSham rTMS using a placebo coilOnce daily, 5 days per week, total 10 sessions over 2 weeksImmediately after 2 weeks of treatmentVAS and ODI decreased significantly in both groups post-treatment, with no significant difference between groups. The active group increased the C-fiber pain threshold on the unaffected side and demonstrated enhanced activation of the SAC as well as improved functional connectivity between SAC and M1 compared with the control group- Mild headache (2/0)- No serious adverse events4/5111Yang et al175 2025- Parallel- iTBS + CSE (n = 20), sham + CSE (n = 20)Left DLPFCFigure-of-eight coil- Triplet 50 Hz bursts at 5 Hz, 2s on/8s off, 600 pulses/session- 80% RMTActive figure-of-eight coil tilted 90° relative to the scalp16 sessions over 4 weeks (4 sessions/week)Immediately after the 4-week treatment course; no later follow-upActive iTBS produced greater pain reduction than sham, with a higher proportion of patients achieving clinically meaningful improvement. iTBS also significantly reduced fear-avoidance beliefs and enhanced functional connectivity between the left DLPFC and cerebellum and occipital cortex; stronger connectivity was associated with greater pain reduction- Transient headache (1/0)- No serious adverse events4/5112Patricio et al176 2026- Parallel- Active rTMS (n = 34), active rTMS + MCE (n = 36), sham rTMS (n = 34), sham rTMS + MCE (n = 36)M1Figure-of-eight coil− 10 Hz; 40 trains/session, 5s/train, ITI 25s, 2000 pulses/session- 95% RMTDedicated sham coil reproducing auditory and sensory effects; magnetic shield prevented cortical depolarization10 sessions over 8 weeks: induction phase of 3 sessions within the first 8 days, followed by 1 session/week for 7 weeks; MCE groups received 30-min exercise immediately after rTMS4, 8, 12, and 24 weeks from baseline; primary endpoint at 8 weeksPain intensity decreased significantly over time with no significant between-group differences. Active rTMS combined with MCE did not provide greater pain reduction than other groups, and no differences were observed between active and sham rTMS with or without MCE- One generalized seizure occurred during session 4 in the active rTMS + MCE group.- Minor rTMS-related adverse events occurred in 58 participants, most commonly headache (n = 45; 32%) and neck pain (n = 12; 9%)- Three participants discontinued active/sham rTMS due to adverse events.4/5Poststroke Shoulder Pain113Choi et al177 2018- Parallel- Active rTMS (n = 12), sham (n = 12)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 5s/train, ITI 55s, total 1000 pulses/session- 90% RMTActive figure-of-eight coil positioned perpendicular to the skullOnce daily, 5 days per week, total 10 sessions over 2 weeksOne, two, and four weeks after treatmentActive rTMS over the affected M1 significantly reduced pain until 4 weeks after treatment, with no significant change in the sham group.

No significant changes were observed in passive shoulder range of motion, MI-UL, or MBC in either groupNo adverse effects reported3/5114Aydın et al178 2024- Parallel- Active (n = 7), sham (n = 11)M1Figure-of-eight coil− 5 Hz; 5s/train, ITI 25s, total 1000 pulses/session- 90% RMTReversely positioned coil over M1 at 1% RMT, producing clicking sounds without intended cortical stimulation5 sessions per week, total 15 sessions over 3 weeksAfter sessions 5, 10, and 15; 4 weeks post-treatmentNo significant difference between or within rTMS and sham groups in NRS score. rTMS over the affected M1 showed no significant benefits over sham in pain, daily functioning, disability, psychological symptoms, neurophysiological measures, or passive ROMNo adverse effects reported4/5115Zhu et al179 2025- Parallel- Sinew acupuncture (n = 30), 1-Hz rTMS (n = 30), sinew acupuncture + 1-Hz rTMS (n = 30)M1Figure-of-eight coil− 1 Hz; 30 pulses/train, ITI 30s, 900 pulses/session- 90% RMTNo shamOnce daily, 5 days per week, total 20 sessions over 4 weeksAfter 4 weeks of treatmentAfter treatment, all 3 groups showed significant pain reduction compared with baseline; the combined therapy group demonstrated greater improvement in pain relief, upper limb function, and daily activities than either monotherapy groupTreatment-specific adverse-event counts were not reported2/5Myofascial Pain Syndrome116Dall’Agnol et al180 2014- Parallel- Active rTMS (n = 12), sham (n = 12)Left M1Figure-of-eight coil− 10 Hz; 16 trains/session, 10s/train, ITI 26s, total 1600 pulses/session- 80% RMTDedicated inactive sham coil positioned over the same M1 targetOnce daily, total 10 sessionsFollow-up varied by outcome, extending up to 12 weeks post-treatmentrTMS significantly reduced VAS compared with sham, with a relative mean pain reduction of 30.21% at 12-week follow-up. rTMS analgesia in chronic MPS involved top-down modulation enhancing corticospinal inhibitory pathways, possibly via BDNF regulationNo adverse events observed4/5117Medeiros et al181 2016- Parallel- rTMS + DIMST (n = 11), rTMS + sham-DIMST (n = 12), sham-rTMS + DIMST (n = 12), sham-rTMS + sham-DIMST (n = 11)Left M1Figure-of-eight coil− 10 Hz; 600 pulses/session, 20 min/session- 80% RMTA sham-rTMS coil was used10 sessionsAfter each of 10 treatment sessionsAll sham-rTMS + DIMST and active interventions significantly reduced pain compared with sham during the 10-day treatment; no synergistic effect of combining rTMS + DIMST was observed- No serious or moderate adverse effects were reported- Two participants withdrew due to treatment inefficacy (one from rTMS group, one from DIMST group)4/5Abbreviations: BDI, Beck Depression Inventory; BDNF, brain-derived neurotrophic factor; CSE, core stabilization exercise; DIMST, deep intramuscular stimulation therapy; DLPFC, dorsolateral prefrontal cortex; DN4, Douleur Neuropathique 4 questionnaire; HAE, hip abductor exercise; iTBS, intermittent theta-burst stimulation; ITI, intertrain interval; KOA, knee osteoarthritis; M1, primary motor cortex; MBC, modified Brunnstrom classification; MCE, motor control exercise; MI-UL, Motricity Index–Upper Limb; MPS, myofascial pain syndrome; MSO, maximum stimulator output; MT, manual therapy; NRS, numeric rating scale; ODI, Oswestry Disability Index; PSQI, Pittsburgh Sleep Quality Index; PT, physical therapy; QSE, quadriceps strengthening exercise; RMT, resting motor threshold; ROM, range of motion; rPMS, repetitive peripheral magnetic stimulation; rTMS, repetitive transcranial magnetic stimulation; S1, primary somatosensory cortex; SAC, somatosensory associative cortex; SE, sling exercise; VAS, visual analogue scale; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.

Knee Osteoarthritis

KOA is among the most prevalent degenerative conditions and represents a leading cause of pain and functional impairment in older adults worldwide.182 The pathophysiology of KOA is complex and involves not only structural degeneration of articular cartilage and subchondral bone but also mechanisms related to central sensitization.183,184 Therefore, rTMS is considered a promising therapeutic approach for KOA treatment.

Our narrative review includes 5 RCTs, with most studies combining 10-Hz M1 stimulation with exercise-based rehabilitation.165–169 Of these, three studies combined 10-Hz M1 rTMS with exercise-based rehabilitation and generally reported greater post-treatment pain relief than comparator groups, with additional improvements in function or biomechanical outcomes depending on the study.165,166,169 The remaining two trials evaluated different strategies.167,168 Dantas et al reported that manual therapy applied immediately after rTMS attenuated the analgesic effect, whereas a single rTMS session alone sufficiently reduced pain intensity in patients with KOA.167 Accelerated iTBS over the left DLPFC produced clinically meaningful pain relief that persisted for up to 16 weeks, although outcomes did not differ significantly from sham.168

Overall, current evidence suggests potential immediate post-treatment analgesic and functional benefits of rTMS in KOA, particularly when combined with exercise-based rehabilitation. However, the number of available studies remains small, and findings are not fully consistent across trials.

Low Back Pain

LBP affects a large proportion of the global population and is recognized as one of the leading causes of years lived with disability.185 Central sensitization is considered one of the key mechanisms underlying LBP.186 Therefore, rTMS, which can modulate cortical excitability and central pain processing, has been explored as a potential therapeutic approach for LBP management.

Our literature review identified 7 RCTs170–176 evaluating rTMS in patients with CLBP. In 2016, Ambriz-Tututi et al reported a marked reduction in pain after 20-Hz rTMS over M1/S1, greater than that observed with physical therapy.170 Importantly, sham stimulation alone produced no significant improvement, whereas patients subsequently showed significant pain reduction after crossing over to active rTMS. Bursali et al also found that M1 rTMS improved pain and functional outcomes compared with sham, with benefits maintained for up to 3 months post-treatment.171 More recently, Chang et al showed that both M1- and DLPFC-targeted rTMS reduced pain compared with sham in patients with LBP and comorbid insomnia, with DLPFC stimulation providing additional benefits for sleep disturbance.173

Evidence was less consistent when rTMS was incorporated into multimodal rehabilitation. Combining rTMS with sling exercise improved outcomes compared with rTMS alone but not compared with sling exercise alone.172 Similarly, adding active rTMS to rPMS did not provide additional reductions in pain or disability over sham rTMS plus rPMS, despite associated neurophysiological changes.174 In contrast, Yang et al found that left DLPFC iTBS combined with core stabilization exercise produced greater pain reduction than sham and also reduced fear-avoidance beliefs, with associated increases in functional connectivity between the DLPFC and cerebellar/occipital regions.175 Most recently, Patricio et al reported that pain improved over time across all groups, but there was no significant advantage of active over sham rTMS, either alone or combined with motor control exercise.176

Overall, rTMS shows potential for short-term pain relief in LBP, particularly when targeting M1 or DLPFC; however, its added value alongside rehabilitation interventions remains uncertain, and heterogeneity in stimulation protocols and follow-up limits firm conclusions.

Post-Stroke Shoulder Pain

PSSP is one of the most common complications following stroke, with a reported prevalence of approximately 30–50% among stroke survivors.187 The pathophysiology of PSSP is multifactorial and involves both peripheral mechanisms, such as shoulder subluxation, and central mechanisms, including central sensitization.188 Accordingly, rTMS has attracted interest as a neuromodulatory strategy for treating PSSP.

Evidence for rTMS in PSSP remains mixed across 3 RCTs.177–179 HF-rTMS over the M1 produced conflicting results. Choi et al reported that 10-Hz rTMS significantly reduced pain, and the analgesic effect persisted for up to 4 weeks after treatment, whereas no significant change was observed in the sham group.177 By contrast, Aydın et al found no significant differences in pain intensity or overall clinical outcomes between active 5-Hz rTMS applied to the M1 and sham stimulation.178 Beyond rTMS monotherapy, Zhu et al evaluated 1-Hz rTMS in combination with sinew acupuncture and found greater pain relief with the combined treatment than with either intervention alone.179

Overall, the therapeutic value of rTMS for PSSP remains uncertain, with benefits varying across stimulation protocols and treatment combinations.

Myofascial Pain Syndrome

MPS is a prevalent musculoskeletal disorder characterized by the presence of myofascial trigger points—hyperirritable sites within taut bands of skeletal muscle.189 MPS is a multifactorial condition that may involve muscle overuse, microtrauma, and peripheral and central sensitization.189,190 Given the role of altered central pain processing in its pathophysiology, rTMS has been explored as a potential neuromodulatory approach for MPS management.

Two RCTs180,181 evaluated the analgesic effects of 10-Hz rTMS over left M1 in patients with chronic MPS. Dall’Agnol et al showed that active rTMS significantly reduced pain compared with sham, with benefits persisting up to 12 weeks post-treatment.180 The authors suggested that the analgesic effects of rTMS in MPS involved top-down modulation that enhanced corticospinal inhibitory pathways, possibly through BDNF regulation. Subsequently, Medeiros et al found that both rTMS- and DIMST-containing interventions reduced pain versus sham, but combining rTMS with DIMST provided no additional synergistic benefit.181

Overall, preliminary evidence suggests analgesic benefits of 10-Hz rTMS over left M1 in MPS, but further well-designed studies are needed to confirm its effectiveness.

Postsurgical Pain

Postoperative pain is a common clinical problem following surgery. rTMS has been investigated for both acute postoperative pain and chronic postsurgical pain conditions.

This review identified 6 studies191–196 investigating the application of rTMS in postoperative settings (Table 5). Most studies investigated rTMS over the left PFC/DLPFC, whereas fewer targeted M1 or used iTBS. Across the early gastric bypass trials, single-session PFC rTMS reduced postoperative opioid requirements by approximately 36–40% over the first 44–54 hours after surgery.191,192 These opioid-sparing effects were not reproduced in a larger trial, in which neither 1 nor 2 active DLPFC rTMS sessions reduced opioid consumption; however, 2 active sessions were associated with lower sensory and affective pain scores than sham at several postoperative time points.193Table 5Characteristics of Studies Evaluating Repetitive Transcranial Magnetic Stimulation (rTMS) for Postsurgical Pain, Cancer-Related Pain, and Other Pain ConditionsNo.Author, YearStudy Design/ Group (n)Stimulation SiteCoil Type of Active rTMSActive rTMS ParametersSham ProcedureTreatment CourseFollow -UpSummary of FindingsSafety, n (Active/Sham)Jadad ScorePostsurgical Pain118Borckardt et al191 2006- Parallel- Active rTMS (n = 10), sham (n = 10)Left PFCFigure-of-eight coil− 10 Hz; 10s/train, ITI 20s, total 4000 pulses/session- 100% RMTDedicated sham figure-of-eight coil, identical in appearance, with an internal aluminum shield preventing cortical stimulationSingle session immediately postsurgeryPain and mood assessed twice daily; morphine use monitored for 44 hours after surgeryActive prefrontal rTMS reduced postoperative morphine consumption by 40% compared with sham within 44 h after surgery, with the greatest effect during the first 24 h; no significant effect of rTMS on mood and pain intensity change was observed- Headache (5/2); nausea (2/2)- No vomiting or severe headache reported3/5119Borckardt et al192 2008- Parallel- Active rTMS (n = 12), sham (n = 8)Left PFCFigure-of-eight coil− 10 Hz; 10s/train, ITI 20s, total 4000 pulses/session- 100% RMTDedicated sham coil identical to the active coil, incorporating an aluminum shield to block magnetic stimulationSingle session immediately after surgeryPain and mood assessed twice daily; morphine use monitored throughout hospitalization, approximately 54 hoursActive rTMS reduced cumulative morphine use by 36% and significantly decreased postoperative pain, and improved worst mood ratings compared with shamHeadache (1/0); fever (3/2); nausea (2/1)3/5120Borckardt et al193 2014- Parallel- 2 sessions rTMS (n = 28), 1 session rTMS + sham (n = 27), sham + 1 session rTMS (n = 25), 2 sessions sham (n = 28)Left DLPFCSolid focal coil− 10 Hz; 5s/train, ITI 10s, total 4000 pulses/session- 110% RMTDedicated shield sham coil reproducing stimulation sounds2 sessions: immediately postsurgery + 4 h laterRepeated pain during hospitalization, approximately 48 hours after surgeryNo difference in opioid use between groups; however, patients receiving 2 active rTMS sessions reported significantly lower sensory and affective pain scores than the sham–sham group at several postoperative time pointsNo treatment-related adverse events reported in any group5/5121Cheng et al194 2023- Parallel- DLPFC iTBS (n = 15), M1 iTBS (n = 15), sham (n = 15)Left M1, DLPFCFigure-of-eight coil− 3 pulses/burst at 30 Hz, burst at 6 Hz, 2s on/8s off, 600 pulses/session- 100% RMTActive coil positioned 90° to the scalpSingle session postsurgery1, 6, 24, and 48 hours after the single sessionDLPFC-iTBS significantly reduced postoperative PCA pump attempts at 6, 24, and 48 h compared with sham, whereas M1 stimulation showed no effect; no significant between-group differences were observed in pain ratings or total anesthetic consumptionVomiting: M1/sham (1/1)4/5122Kataria et al195 2024- Parallel- Active rTMS (n = 17), sham (n = 17)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 60 pulses/train, ITI 60s, total 1200 pulses/session- 80% RMTActive figure-of-eight coil elevated and tilted away from the scalp to prevent cortical stimulation5 consecutive days/week for 3 weeks, total 15 sessionsImmediately after 15 sessions over 3 weeks; no subsequent follow-uprTMS showed beneficial effects on pain, quality of life, and thermal sensitivity in patients with PMPSLow proportion of side effects reported; no adverse event-related withdrawals3/5123Feng et al196 2025-Parallel- ERT (n = 31), ERT + rTMS (n = 31)The side: 6 cm from the top of the head; the front: 1 cm from the same ear root connectionDouble-coil cross-center point− 1 Hz; 50 trains/session, 10 pulses/train, 10s/train, ITI 20s- 40% MTNo sham5 days/week, rest for 2 days for 8 weeksAssessment after the 8-week intervention; inflammatory markers assessed 3 days post-treatmentActive rTMS combined with early rehabilitation significantly reduced pain frequency and intensity, and improved knee function scores and inflammatory markers compared with rehabilitation aloneAdverse events not reported2/5Cancer-Related Pain124Khedr et al197 2015- ParallelActive rTMS (n = 15), sham (n = 15)M1Figure-of-eight coil− 20 Hz; 10 trains/session, 10s/train, ITI 30s, total 2000 pulses/session- 80% RMTActive coil elevated and angled away from the head, using the same stimulation parameters5 consecutive days/week for 2 weeks, total 10 sessionsAfter sessions 1, 5, and 10; 15 days and 1-month post-treatmentActive rTMS produced greater reductions in pain scores and depression scores than sham up to 15 days after treatment; effect not maintained at 1 monthNo stimulation-related side effects recorded2/5125Tang et al198 2022- Parallel- Active rTMS (n = 20), sham (n = 19)Left DLPFCFigure-of-eight coil− 10 Hz; 15 trains/session, 1.5s/train, ITI 3s, total 1500 pulses/session- 80% RMTSame coil with no magnetic output, producing similar sounds at different coil anglesOnce daily, 5 days per week for 3 weeksDays 3 and 7, and weeks 2 and 3 during treatment; no post-treatment follow-uprTMS improved pain control, reduced opioid dosage, and enhanced mood and quality of lifeTransient scalp numbness and facial muscle twitching reported in the active group; event counts and sham-specific events NR.

No serious adverse events3/5Pain Associated with Parkinson’s Disease126Li et al199 2020- Parallel- Active rTMS (n = 24), sham (n = 24)M1Figure-of-eight coil− 20 Hz; 100 pulses/train, 20 min/session- 80% RMTActive figure-of-eight coil oriented at 90° to the midsagittal plane, reproducing some sensory effects without intended cortical stimulation5 consecutive daily sessionsAfter session 5; 2 and 4 weeks post-treatmentOnly the rTMS group, but not the sham group, demonstrated significant improvements in pain, motor symptoms, depression, anxiety, and overall disease severityHeadache (3/3), dizziness (2/2), tinnitus (2/1), transient worsening of gait disturbance (1/0). One participant in each group withdrew due to headache or dizziness; two others withdrew due to leg pain (active) and fever (sham). No serious adverse events.3/5127Barboza et al200 2024- Parallel- Active rTMS (n = 14), sham (n = 11)Right PSIDouble-cone coil− 10 Hz; 30 trains/session, 10s/train, ITI 20s, total 3000 pulses/session- 80% RMTOrthogonally positioned active figure-of-eight coil, with the double-cone coil switched off during sham stimulation- Induction: 5 daily sessions in 1 week- Maintenance: once weekly for 7 weeks- Total 12 sessions over 8 weeksDays 7, 30, and 60 from treatment initiationNo significant differences were observed between active and sham rTMS in overall pain reduction; post hoc analysis showed greater response to active stimulation in the nociceptive pain subtypeHeadache (2/0), mild to moderate and lasting less than 4 hours.

No seizures or adverse event-related withdrawals3/5128Yang et al201 2025- Parallel- LD + sham (n = 24), LD + EN + sham (n = 24), LD + EN + rTMS (n = 24)Bilateral M1Round coil− 10 Hz; 50 trains/session, 14 pulses/train, 5.6s on/30s off, total 700 pulses/session- 90% RMTDedicated placebo coil with identical appearance and sound, delivering < 10% of the magnetic field intensity without cortical activation, positioned 15° away from the scalp3 sessions/week for 24 weeks (total 72 sessions)After 6 months of treatment; no separate post-treatment follow-upPain improved in all groups, but LD + EN + rTMS showed the most favorable outcomesLD + Sham/ LD + EN +Sham/ LD + EN +rTMSScalp discomfort (1/0/5), headache (2/1/4), dizziness (1/2/1), and nausea (3/4/3). One in LD + EN +Sham withdrew due to persistent nausea unrelated to the sham stimulation4/5Pain Associated with Multiple Sclerosis129Dahshan et al202 2025- Parallel- Active rTMS (n = 10), tDCS (n = 10), sham (n = 10)M1Figure-of-eight coil− 10 Hz; 20 trains/session, 50 pulses/train, ITI 20s- 80% RMTCoil positioned over M1 without magnetic pulse deliveryOnce daily for 5 consecutive daysAfter the five-day treatmentBoth tDCS and rTMS groups showed significant improvements in fatigue, pain, and depression compared with sham; tDCS may be suitable for home use, whereas rTMS may suit clinical settingsAdverse events not reported2/5Spasticity130Gharooni et al203 2018- Crossover- Active iTBS, sham(n = 10)M1Circular coil− 3 stimuli at 50 Hz repeated every 200 ms for 2s, ITI 8s, 600 pulses/session- 80% RMTActive coil rotated 90° about its vertical axis, using identical iTBS parameters10 sessions over 2 weeks per condition, 2-week washoutAssessed before and after each treatment periodAlthough upper-limb spasticity was reduced, the reduction was insufficient to improve perceived spasticity, functional outcomes, sensorimotor function, or pain- Interscapular tightness (1/0) is considered unrelated to stimulation. - No serious treatment-related AEs3/5131Korzhova et al204 2019- Parallel- 20-Hz rTMS (n = 12), iTBS (n = 12), sham (n = 10)M1Figure-of-eight coil− 20-Hz rTMS: 2s/train, ITI 28s, total 1600 pulses/session- iTBS: 5 Hz bursts (3 pulses at 35 Hz per burst), with 1200 stimuli per session- 80% MSOSham procedure reproduced using a separate coil located approximately 1 m from the participant, preserving the visual and acoustic characteristics of stimulation5 consecutive days over 2 weeksImmediately after 10 sessions; 2 and 12 weeks post-treatmentBoth 20-Hz rTMS and iTBS significantly reduced spasticity as assessed by the MAS, whereas the sham group showed no significant change. iTBS appeared to produce longer-lasting improvements in patient-reported spasticity, while 20-Hz rTMS was associated with reductions in pain and fatigue.No serious adverse events reported3/5Cervical Dystonia132Bradnam et al205 2016- Parallel- Active cerebellar iTBS + MT (n = 8), sham iTBS + MT (n = 8)Bilateral cerebellumFigure-of-eight coil- Bursts of 3 pulses at 50 Hz, repeated at 5 Hz;2s iTBS trains repeated every 10s, total 600 pulses/session- 80% MSOIdentical sham coil producing similar acoustic output without brain stimulation10 sessions over 10 working daysAfter 5 and 10 treatment days; quality of life additionally assessed at 4 and 12 weeks post-treatmentCerebellar iTBS may alleviate cervical dystonia symptoms and enhance upper-limb motor control and quality of life, potentially through mechanisms related to cerebellar neuroplasticity, although the underlying neurophysiology remains unclearNo adverse effects reported3/5Irritable Bowel Syndrome133Melchior et al206 2014- Crossover- Active rTMS, sham (n = 15)M1Figure-of-eight coil− 20 Hz; 20 pulses/session, 5s/train, ITI 55s, 2000 pulses/session- 80% RMTDedicated placebo coil resembling the active coil and producing similar sounds5 consecutive daily sessions per condition, 2-month washoutOn day 5 after the last rTMS sessionPain improved in both groups with no significant difference between active and sham rTMS in pressure pain threshold, maximum tolerated volume, or rectal compliance; however, active rTMS increased the volume threshold in patients with marked rectal hypersensitivityHeadache (3/1; mild and transient), hand paresis (0/1), severe anxiety (1/0). The latter two resolved within a few days but led to study withdrawal4/5Bladder Pain Syndrome/Interstitial Cystitis134Cervigni et al207 2018- Crossover- Active rTMS, sham (n = 13)M1H-coil− 20 Hz; 30 trains/session, 50 pulses/train, ITI 30s, total 1500 pulses/session- 110% RMTDedicated sham H-coil integrated into the same helmet, reproducing acoustic output and facial muscle activation while inducing a negligible electric field5 consecutive days/week for 2 weeks per condition, 6-week washoutDuring both treatment periods and at 3 and 6 weeks after the final treatment periodActive rTMS significantly reduced pain and urinary LUTS, and improved QOL compared with shamMild headache (n=2) and a lipothymic episode (n=1) during active rTMS; no serious adverse events reported4/5Abbreviations: DLPFC, dorsolateral prefrontal cortex; EN, entacapone; ERT, early rehabilitation therapy; iTBS, intermittent theta-burst stimulation; ITI, intertrain interval; LD, levodopa; LUTS, lower urinary tract symptoms; M1, primary motor cortex; MAS, Modified Ashworth Scale; MSO, maximum stimulator output; MT, motor training; PCA, patient-controlled analgesia; PFC, prefrontal cortex; PMPS, postmastectomy pain syndrome; PSI, posterior superior insula; QOL, quality of life; RMT, resting motor threshold; rTMS, repetitive transcranial magnetic stimulation; tDCS, transcranial direct current stimulation.

Recently, neuromodulation protocols beyond conventional rTMS have been explored. Cheng et al in 2023 showed that DLPFC-iTBS significantly reduced postoperative patient-controlled analgesia pump attempts at 6, 24, and 48 hours compared with sham, whereas M1 stimulation showed no effect; no significant between-group differences were observed in pain ratings or total anesthetic consumption.194

Postmastectomy pain syndrome (PMPS) is a common complication following breast surgery. Although its exact pathophysiology remains unclear, neuralgia of the intercostobrachial nerve is currently considered one of the most likely underlying mechanisms.208 In 2024, Kataria et al reported that 15 sessions of 10 Hz rTMS over the M1 had beneficial effects on pain, quality of life, and thermal sensitivity in patients with PMPS.195

The only RCT in postoperative sports knee injury found that adding 1-Hz M1 rTMS to early rehabilitation significantly reduced pain frequency and intensity, improved knee function, and inflammatory markers compared with rehabilitation alone.196

Overall, rTMS shows promising short-term effects in postoperative pain, although evidence across different surgical conditions remains limited.

Cancer-Related Pain

Pain is highly prevalent among patients with cancer. Nearly half of cancer survivors with solid tumors report persistent pain after completing curative treatment.209 The mechanisms underlying cancer-related pain are complex and involve both peripheral and central sensitization.210 Therefore, rTMS has been considered a promising neuromodulatory approach for the management of cancer-related pain.

This review identified 2 RCTs197,198 that investigated the application of rTMS in the treatment of pain in patients with cancer (Table 5). First, Khedr et al showed that 10 sessions of 20-Hz rTMS over the M1 produced significantly greater reductions in pain scores and depression than sham up to 15 days after treatment, though effects were not maintained at 1 month.197 Similarly, 10 Hz rTMS applied over the left DLPFC in patients with advanced non-small-cell lung cancer and cancer-related pain also significantly reduced pain intensity compared with sham, lowered opioid requirements, and improved mood and quality of life.198

Overall, preliminary evidence suggests that rTMS can reduce cancer-related pain and may also improve associated outcomes such as opioid use, depression, and quality of life; however, the evidence is limited to 2 small RCTs with relatively short follow-up.

Other Pain Conditions

In addition to the conditions discussed above, rTMS has been investigated for the treatment of various other disorders, including Parkinson’s disease (PD), multiple sclerosis (MS), spasticity, cervical dystonia, irritable bowel syndrome (IBS), and Bladder pain syndrome/interstitial cystitis (BPS/IC). The studies included in this section are presented in Table 5.

Pain is a common nonmotor symptom in PD, affecting approximately 40% of patients, and nearly half of those experiencing pain remain undertreated.211,212 According to the Ford classification, PD-related pain can be categorized into musculoskeletal pain, radicular or NP pain, dystonia-related pain, akathisia-related discomfort, and central Parkinsonian pain, arising from both peripheral and central mechanisms.211–213 In this review, 3 RCTs199–201 evaluated the efficacy of rTMS for pain in patients with PD. Li et al reported that with musculoskeletal pain related to PD, only the active rTMS group, not the sham group, demonstrated significant improvements in pain, motor symptoms, depression, anxiety, and overall disease severity.199 Subsequently, Barboza et al found that no significant differences were observed between active and sham rTMS in overall pain reduction, although post hoc analysis suggested a greater response in patients with nociceptive pain.200 More recently, in 2025, Yang et al compared levodopa + sham, levodopa + entacapone + sham rTMS, and levodopa + entacapone + 10-Hz bilateral M1 rTMS; pain improved in all groups, with the most favorable outcomes in the combined pharmacotherapy + rTMS group.201 Current evidence suggests that rTMS may have analgesic effects in PD, and its efficacy may vary depending on the pain subtype.

Pain is common in patients with multiple sclerosis (MS), occurring in up to 50–75% of individuals during the disease course and even presenting as an initial symptom in approximately 10% of cases.214 In 2025, Dahshan et al reported that both 10-Hz M1 rTMS and tDCS groups showed significant improvements in pain, fatigue, and depressive symptoms compared with sham. tDCS may show potential for home use, whereas rTMS may suit clinical settings.202

Spasticity is a common motor disorder characterized by increased muscle tone and exaggerated tendon reflexes due to upper motor neuron lesions and is often accompanied by significant pain.215 In this review, 2 studies203,204 investigated the application of rTMS for the management of pain and spasticity. Both protocols showed potential to reduce spasticity. In incomplete cervical SCI, M1 iTBS was safe and feasible, with preliminary findings suggesting a reduction in upper-limb spasticity; however, improvements were insufficient to affect perceived spasticity, functional outcomes, sensorimotor function, or pain.203 Subsequently, Korzhova et al reported that both 20 Hz rTMS and iTBS significantly reduced spasticity as assessed by the Modified Ashworth Scale, whereas no significant change was observed in the sham stimulation group.204 Additionally, iTBS appeared to produce longer-lasting improvements in patient-reported spasticity, whereas 20 Hz rTMS was associated with reductions in pain and fatigue.

Current evidence remains limited and inconsistent, making it difficult to draw definitive conclusions about the effectiveness of rTMS in reducing pain in patients with spasticity; however, rTMS may have the potential to reduce spasticity.

Cervical dystonia is a focal movement disorder characterized by involuntary contractions of the neck muscles that lead to abnormal head postures and often significant neck pain.216 Evidence for cervical dystonia is limited to a single RCT, in which cerebellar iTBS combined with motor training reduced cervical dystonia symptoms and enhance upper-limb motor control and quality of life, potentially through mechanisms related to cerebellar neuroplasticity, although the underlying neurophysiology remains unclear.205

IBS is a common functional gastrointestinal disorder characterized by recurrent abdominal pain associated with alterations in bowel habits; dysregulation of the brain–gut axis and central pain processing plays a key role in its pathophysiology.217 Melchior et al found pain improved in both groups, with no significant difference between active and sham rTMS in the pressure pain threshold, maximum tolerated volume, or rectal compliance; active stimulation increased rectal tolerance in patients with marked hypersensitivity.206

BPS/IC is a chronic condition characterized by lower urinary tract symptoms and pelvic pain. Its pathophysiology is multifactorial, involving chronic inflammation, mast cell activation, urothelial dysfunction, autoimmune dysfunction, afferent sensory hyperactivity, and central nervous system sensitization.218 Cervigni et al found that M1 dTMS significantly reduced pain and improved urinary symptoms and quality of life compared with sham.207

Clinical Considerations and Future Directions

Beyond the evidence for individual pain conditions, several cross-cutting issues are important for interpreting the current literature and translating rTMS into clinical practice. These include the design and effects of sham stimulation, safety, durability of analgesic effects and maintenance treatment, multimodal treatment integration, and remaining research priorities.

Sham stimulation represents an important source of methodological heterogeneity across rTMS trials. Among the sham-controlled studies included in this review, the predominant approaches were dedicated or integrated sham coils and repositioning of the active coil by tilting, rotating, or separating it from the scalp; some systems additionally reproduced scalp sensations using electrical stimulation. These approaches differ in their ability to reproduce the auditory, cutaneous, and muscular sensations of active rTMS, which may influence blinding and treatment expectancy.

Importantly, sham stimulation was not always without clinical effects. Several studies reported improvements in pain from baseline in the sham group, and some found no significant difference in clinical outcomes between active and sham stimulation. Adverse events were also reported in sham groups, including headache, scalp or neck discomfort, dizziness, fatigue, and nausea.

This observation is consistent with a previous meta-analysis of 93 placebo-controlled TMS trials, in which 13.6% of sham-treated participants experienced at least one adverse event, supporting the presence of nocebo phenomena during TMS.14 Overall, these placebo and nocebo responses may affect trial outcomes and complicate the interpretation of differences between active and sham rTMS.

rTMS was generally well tolerated across the included trials. Most reported adverse events were mild and transient, most commonly headache, scalp or stimulation-site discomfort, pain exacerbation, dizziness, neck pain, fatigue, nausea, or paresthesia. Some participants were unable to complete stimulation or withdrew from the study because of adverse events.

Serious adverse events were uncommon, although 2 generalized seizures were reported.120,176 A notable serious adverse event was reported by Picarelli et al, in which a patient with CRPS and no prior history of seizures or epilepsy developed a generalized seizure after the seventh session of active 10-Hz M1 rTMS.120 The patient experienced no further seizures during 18 months of follow-up. The authors suggested that the event may have been facilitated by the high total pulse number (2,500 pulses/session) and concomitant use of tricyclic antidepressants, which can lower the seizure threshold. In another trial, Patricio et al reported a generalized seizure during the fourth session of active 10-Hz M1 rTMS in a patient with chronic low back pain.176 The patient had no history of seizures and was not taking medications known to increase rTMS-related seizure risk, while subsequent EEG and brain imaging were normal.

Together, these rare events highlight the importance of careful screening, adherence to established stimulation parameters, and consideration of concomitant medications when assessing seizure risk during rTMS.

Follow-up duration varied substantially across studies, ranging from assessments immediately after individual stimulation sessions or at the end of the treatment course to several months thereafter, with the longest post-treatment follow-up extending to 6 months. This variability makes it difficult to determine how long the analgesic effects of rTMS persist after treatment. In parallel, maintenance stimulation was inconsistently applied across studies.

Together, these findings indicate that both the long-term durability of analgesia and the optimal maintenance strategy remain insufficiently established.

rTMS has been evaluated as an adjunct to exercise, acupuncture, pharmacotherapy, and other treatments, especially in musculoskeletal pain conditions, but its added value within multimodal pain management remains uncertain. Some studies reported greater improvements in pain or function with combined treatment, whereas others found no additional or synergistic benefit. Overall, the effectiveness of multimodal rTMS appears to depend on the underlying pain condition, the type of adjunctive therapy, and the specific treatment protocol, highlighting the need for further studies to identify the most effective combinations and treatment strategies.

Future research should prioritize adequately powered multicenter RCTs using standardized stimulation and outcome-assessment protocols to define condition-specific protocols, establish the durability of analgesic effects, and improve comparability across studies. Direct head-to-head comparisons of M1, DLPFC, and other candidate targets, together with systematic evaluation of stimulation dose, session frequency, and induction and maintenance schedules are needed. Emerging approaches such as iTBS and accelerated stimulation warrant further evaluation because of their potential to reduce treatment burden.

Given the inconsistent findings of combined interventions, the optimal integration of rTMS with pharmacotherapy, rehabilitation, or psychological treatments also requires clarification. Additionally, emerging evidence suggests that baseline neurophysiological, neuroimaging, and molecular characteristics may help identify patients more likely to benefit from treatment.219–221 This also represents a potentially promising research direction that warrants further investigation in larger, well-designed studies. Economic considerations are also important for the broader clinical use of rTMS, particularly because analgesic effects are often short-lived and repeated or maintenance sessions may be required to sustain benefit; therefore, health-economic evaluation should be considered an important direction for future research.

Limitations

The findings of this review should be interpreted in light of several limitations. First, the search was restricted to PubMed and English-language publications, which may have led to the omission of relevant studies and may have increased the risk of language bias. In addition, as a narrative review, the findings should be interpreted as a broad overview rather than a comprehensive systematic synthesis.

Second, we used the Jadad scale to assess the methodological quality of the included RCTs with respect to randomization, blinding, and withdrawals/dropouts. While this approach provides a practical and standardized assessment, it is less comprehensive than the Cochrane Risk of Bias 2 tool and may therefore not capture all potential sources of bias. Third, many included trials had small sample sizes and potential challenges related to blinding and sham stimulation, which should be considered when interpreting the findings.

Fourth, given the large number of included studies and the substantial variability in outcome measures across trials, quantitative measures such as effect sizes, confidence intervals, and responder rates were not systematically extracted. Instead, this review focused on the direction and statistical significance of the principal outcomes reported in the original studies. Therefore, statistically significant findings should be interpreted cautiously and should not be assumed to represent clinically meaningful benefit.

Conclusion

Overall, the evidence for rTMS in pain management remains heterogeneous. Our review suggests that rTMS shows promising analgesic effects across several pain conditions, including general NP, SCI-related pain, FM, migraine, orofacial pain, KOA, and LBP; however, the evidence remains inconsistent across studies. Relatively consistent but still limited evidence suggests potential analgesic benefits of rTMS in PDN, PHN, PLP, CRPS, TTH, post-mTBI headache, MPS, cancer-related pain, and PD-related pain.

Evidence for CPSP, PSSP, postsurgical pain, and spasticity-related pain remains limited and heterogeneous, while several other conditions—including brachial plexus injury, lumbosacral radiculopathy, mTBI, cluster headache, MS, cervical dystonia, IBS, and BPS/IC—are supported only by single studies, precluding firm conclusions. Future research should prioritize RCTs with standardized protocols, direct comparisons of stimulation targets and other protocol parameters, evaluation of accelerated iTBS and emerging rTMS variants, biomarker-guided patient selection, and health-economic assessment. Given the off-label use and heterogeneous evidence base, rTMS should currently be regarded as an experimental adjunct for refractory pain, with full informed consent and use in specialized centers.