Transcranial magnetic stimulation: a working reference for referring clinicians

Overview

TMS is a non-invasive cortical stimulation technique with clear FDA indications, real-world response and remission data, and a rapidly evolving protocol landscape. The useful questions for a referring clinician are usually practical: what protocols are available, what the evidence is for a given indication, and what the contraindications look like.

How we got here

TMS moved from a laboratory curiosity to an FDA-cleared treatment inside a single professional lifetime. Each major leap came from a circuit-level insight about brain networks — not from a chemistry insight. That pattern continues to shape where the field is going.

• 1985Barker
  Anthony Barker and colleagues at the University of Sheffield demonstrate the first TMS-evoked motor response in humans. A magnetic pulse over motor cortex produces a contralateral hand twitch — non-invasive, painless, and reproducible. The essential ingredients for clinical translation are in place.
• 1990sDrevets, George
  PET imaging reveals metabolic abnormalities in the subgenual anterior cingulate cortex (Brodmann area 25) in depression. At the NIH and later at MUSC, Mark George conducts the first human trials targeting left prefrontal cortex with TMS for depression — pointing to a circuit-based framework rather than a neurotransmitter-based one.
• 2005Mayberg
  Helen Mayberg's work reconceptualizes depression as dysfunction of a distributed network, with sgACC as a pathologic hub. Deep brain stimulation of sgACC produces sustained remission in some patients with otherwise-intractable depression (Mayberg 2005). The network framing redirects attention from “chemical imbalance” toward circuit dysfunction.
• 2007–2010OPT-TMS, FDA
  O'Reardon 2007 and George 2010 (OPT-TMS) establish TMS efficacy in pivotal sham-controlled trials. The FDA clears TMS for MDD in adults in October 2008 (NeuroStar). TMS enters clinical practice as an FDA-cleared treatment for treatment-resistant depression.
• 2012Fox
  Michael Fox and colleagues demonstrate that TMS efficacy tracks the degree of functional anticorrelation between the DLPFC stimulation site and the sgACC (Fox 2012). This is the conceptual foundation for fMRI-guided personalized targeting — different individuals have the right target in different cortical locations.
• 2018Blumberger
  The THREE-D randomized non-inferiority trial (n=414) shows that intermittent theta burst stimulation (iTBS, ~3 minutes) is non-inferior to standard 10 Hz rTMS (~37 minutes) in treatment-resistant depression (Blumberger 2018, Lancet). Shorter sessions dramatically expand clinic throughput without compromising outcomes.
• 2022Cole
  Stanford Neuromodulation Therapy (SAINT/SNT) combines accelerated iTBS with fMRI-guided individualized targeting and shows ~79% remission in a randomized sham-controlled trial (Cole 2022). FDA-cleared under K213481. Notably, SAINT's lead investigators (Williams and Bentzley) trained at MUSC under Mark George and Baron Short — continuing a regional research lineage.
• 2024–2025Adolescent clearances
  NeuroStar (March 2024), Magstim (March 2025), MagVenture (August 2025), and BrainsWay (November 2025) each receive FDA clearance for adolescent MDD in the 15–21 age range — the first major expansion of TMS indications in a vulnerable population that has few well-tolerated alternatives.

Protocols

Protocols differ in frequency, burst pattern, laterality, session duration, and therapeutic intent. Most share the same underlying logic; they differ in how stimulation is packaged and delivered.

Coil types

Coil geometry determines the field's focality and depth — two properties that trade off against each other. A more focal coil targets cortex precisely but doesn't reach deep structures; a less focal coil reaches deeper but stimulates a larger volume. No single coil is “better” — each is a deliberate choice in that tradeoff (Deng 2014).

The rhythms, visualized

Protocol names proliferate; the underlying differences are mostly about how stimulation is packaged in time. The figures below show a few seconds of each protocol’s pulse structure at a consistent scale. They are schematic, not to clinical scale, but make the differences between protocols immediately visible.

Accelerated schedule: SAINT

Four protocols, one principle

The proliferation of names can obscure what’s actually varying. Across rTMS, iTBS, dTMS, and accelerated protocols, the underlying logic is the same: repeated, targeted stimulation of a cortical node in a dysfunctional network, summed over a course, to produce downstream circuit-level change. What changes between protocols is dose packaging — pulse shape, burst structure, session length, coil geometry, and schedule.

Typical course and taper

A standard clinical course of conventional rTMS or iTBS runs 5 days per week for 6 weeks (30 sessions), followed by a 6-session taper spread over the following 3 weeks for a total of 36 sessions. The bulk of the treatment burden is in the first 6 weeks; the taper is designed to support consolidation of response. Accelerated protocols compress this entire course into 5 days.

Devices

FDA has cleared multiple manufacturers' TMS systems for depression, with newer indications added over time. Devices differ in coil design, session length options, navigation capability, and business model. None is categorically “best” — the right choice depends on the protocols you plan to run and the indications you want to treat.

Some common TMS devices

Specialized / newer platforms

A handful of systems occupy specific niches beyond these:

• Magnus Medical (SAINT/SNT) — licenses the SAINT protocol and uses MagVenture hardware combined with Soterix/Brain Science Tools neuronavigation. Monopoly position on FDA-cleared accelerated fMRI-guided iTBS. Costs are substantially higher than conventional TMS; typically not covered by insurance.
• Nexstim NBT — Finnish system originally designed for pre-surgical mapping, with SmartFocus MRI-based neuronavigation. FDA-cleared for MDD.
• Apollo (MAG & More) — German system using the HANS head-and-neck positioning approach. FDA-cleared for MDD, with adolescent MDD clearance added in 2025.
• Soterix / Neurosoft — US-bundled system that pairs the Neurosoft stimulator with Soterix's own neuronavigation that does not require line-of-sight.

Choosing a device: what actually matters

Disclosure

The Prisma Health Neuromodulation Program (Brownell Center, Greenville, SC) currently uses NeuroStar, with Magstim and BrainsWay platforms being added. This expansion broadens the protocol options available locally — notably adding BrainsWay's H-coil deep TMS and Magstim's additional protocol options. This page is independent of any manufacturer and is not endorsed by or affiliated with any TMS device company.

Targeting

Where the coil goes matters. Depression outcomes have been shown to vary meaningfully with targeting precision, and targeting has been one of the most active areas of TMS development in the past decade.

Motor threshold (MT)

Stimulation intensity is individualized to each patient's resting motor threshold — the minimum intensity over M1 that reliably produces a contralateral hand muscle response. Treatment doses (commonly 100–120% MT for prefrontal protocols) are scaled from that reference.

Targeting methods, from simplest to most precise

Targeting error adds up. Work from the Brainstim / Acacia group and others has shown that scalp-based methods can miss the intended DLPFC target by ~30 mm on average, and that more precise, connectivity-based targeting improves remission rates in head-to-head comparisons (Cash 2021; Li 2020).

Brain networks

Understanding why L-DLPFC is the standard depression target requires stepping back from the coil and thinking about circuits. Depression is increasingly conceptualized as a disorder of large-scale brain networks rather than a single region or neurotransmitter — and TMS works by modulating a specific node within those networks.

Three canonical large-scale networks are implicated in depression, and their interaction is thought to be where the pathology lives:

The rhythms within the networks

Network anatomy is only half the story. The DMN, CEN, and salience networks don’t just have locations — they have rhythms, and those rhythms are how the networks communicate. Each network is associated with a characteristic dominant oscillatory frequency, and information moves between regions when their rhythms align in particular ways. In depression, both the rhythms themselves and the way they couple across networks are dysregulated. TMS appears to work, in part, by retuning that rhythmic architecture.

The animation below shows what this looks like across three states — a healthy brain, a depressed brain, and a brain after a course of TMS. Each colored ring is a region; the pulsing represents the local oscillation; the lines between regions represent functional connectivity, with thickness proportional to coupling strength. The thought-text streaming from the DMN reflects what those network dynamics produce subjectively.

Click through the three tabs and watch what changes. In the healthy state, the regions oscillate at distinct frequencies with moderate amplitudes. The dashed line between DMN and CEN represents anticorrelation — when one ramps up, the other steps back, with the salience network arbitrating between them. The sgACC is present but not dominant, and thought content is mundane and outward-facing.

In the depressed state, three things change at once. The sgACC’s rhythm slows and amplifies — a finding consistent with the thalamocortical dysrhythmia framing of mood disorders, in which pathologically slow, monotonous oscillations replace flexible faster rhythms. The DMN–sgACC connection thickens and turns red, reflecting the abnormal hyperconnectivity Liston and colleagues documented (Liston 2014; Greicius 2007). Meanwhile the CEN ring shrinks — the dorsolateral prefrontal cortex is hypoactive (the cognitive symptoms of depression map directly onto this) — and its anticorrelation with the DMN breaks down. Cognitive control can no longer suppress the negative ruminative loop. Thoughts pour out of the sgACC-dominated DMN faster than the CEN can intercept them, and the content shifts to the cognitive distortions clinicians recognize: I’m a failure. Nothing will change. I ruin everything.

In the after TMS state, the 10 Hz coil over the dlPFC pulses at treatment frequency and the picture reorganizes. The CEN ring brightens and accelerates — the network is being entrained. The DMN–sgACC line thins back toward baseline (the “TMS attenuates abnormal sgACC hyperconnectivity in responders” finding from Liston 2014, with related connectivity-decrease-tracks-response findings in Baeken 2014; Taylor 2018 importantly observed that sgACC connectivity dropped with response in both active and sham arms, so the connectivity change tracks clinical improvement rather than active stimulation per se). The CEN–sgACC line turns teal: top-down control is restored. Rhythms re-synchronize, and the thought stream slows and shifts: I can try. This might pass.

The “feeling versus knowing” gap

Cognitive therapists have long described a clinical phenomenon that has resisted easy explanation: a patient can recite a perfectly sound rational reframe of their depressive cognition — I know I’m not actually a failure; I have evidence to the contrary; I’m loved by my family — and walk out of the session feeling exactly the same. They know the cognitive content. They cannot feel that they believe it. Stott (2007) named this rational-emotional dissociation (RED) — the “head versus heart” gap that limits how much pure cognitive intervention can accomplish in moderate-to-severe depression.

The neural mechanism is becoming clearer, and it lives in the cross-frequency structure of the rhythms above. In simplified form: theta is a slow carrier wave (4–8 Hz) and gamma is the activity that carries specific content (30–100 Hz). For cortical content generated in the dlPFC to integrate with limbic representations of self-worth, threat, or felt safety, the gamma packets must arrive on a particular phase of the theta cycle — the rising phase. This is theta-gamma phase-amplitude coupling (PAC), and it has been shown causally to gate emotional integration: Bramson and colleagues (2020) demonstrated that targeting long-range theta-gamma PAC between anterior PFC and sensorimotor cortex causally improved emotional-action control in healthy humans, with the effect dependent on the relative phase of stimulation. Alekseichuk (2016) showed in a parallel paradigm that working memory only improved when gamma bursts were phase-locked to theta peaks — phase-mismatched stimulation didn’t work.

In depression, this carrier-wave architecture is broken. Frontal-midline theta — the rhythm Cavanagh and Frank (2014) identified as the cognitive-control signature of midcingulate cortex — is altered in mood pathology; cross-frequency theta-gamma coupling is reduced; the sgACC dominates with low-frequency power that is uncoupled from the gamma content the dlPFC is generating. The result is rational-emotional dissociation at the network level: the cognitive reframe is produced as gamma activity, but without a properly entrained theta carrier to deliver it into limbic representation, the affective system never receives the update.

The animation below makes this visible. The dlPFC at the top continuously generates gamma packets carrying CBT-style cognitive content. The slow purple wave between cortex and limbic is the theta carrier, with a glowing window that follows the rising phase of each cycle — this is when the door is open. Watch what happens to the packets in each state.

In the depressed state, the theta wave is shallow and irregular; the lock-window glow flickers in and out. Most packets miss the rising phase entirely and scatter when they hit the wave — visible as red rings dissipating without ever reaching the limbic receiver. The integration meter sits around 12%. The limbic region’s caption stays still feels like a failure. This is RED at the rhythm level: the cognitive content is being generated faithfully, but the bridge to felt belief is broken.

In the CBT alone state, the theta carrier strengthens somewhat — the rhythmic stabilization that comes from sustained cognitive engagement — and a portion of packets begin to phase-lock with the rising window. You’ll see occasional teal landings: moments where a reframe actually integrates as felt belief. The limbic caption shifts to glimpses of belief. Integration meter rises to roughly 32%. This matches the clinical picture: CBT in moderate-to-severe depression produces real but inconsistent integration of the cognitive content it generates.

In the TMS-restored state, the theta carrier becomes robust and stable — the rhythmic substrate has been retuned. Packets now consistently catch the rising phase, ride the wave down, and land as integrating bursts in the limbic region. Noda (2017) documented this directly: a course of high-frequency left dlPFC rTMS in depressed patients produced significantly increased resting gamma power and increased theta-gamma coupling at frontal sites. The limbic receiver in the animation transforms from coral to teal as the integration crosses 50%, and the caption shifts to and now I feel it. Patients describe this transition with phrases like “I knew it before, but now I feel it.”

This framework also explains why iTBS works. The protocol literally delivers gamma-frequency bursts (3 pulses at 50 Hz) on a theta-frequency carrier (5 Hz repetition) — an external imposition of the exact theta-gamma coupling pattern depression has lost. Closed-loop EEG-synchronized rTMS protocols, currently in clinical trials (NCT02920840; George 2023), are testing whether actively timing pulses to the patient’s ongoing oscillatory phase can drive this restoration even more effectively.

What this means clinically

The rhythm-level framing reframes some familiar clinical observations. Patients who tell you they “know all the right things to think but cannot feel them” are not failing at CBT — they are describing a network state in which the cognitive content is being generated but not delivered. Patients who report sudden moments of felt insight during a TMS course (“something shifted yesterday and I can’t describe it, but the same thoughts feel different now”) may be describing the moment when carrier-wave integration was restored. And the empirical observation that TMS responders often describe their pre-treatment cognitions as more accurate than their post-treatment cognitions — without remembering having “decided” that — makes sense if the mechanism is rhythm restoration rather than belief replacement. The cortical content didn’t change. The brain’s ability to feel what it already knew did.

Can CBT during or right after TMS make the gamma packets land?

The mechanistic picture above predicts something specific and clinically actionable: if TMS works partly by restoring the theta carrier and the integration window, then cognitive content delivered during or immediately after a session should integrate more efficiently than the same content delivered weeks later in a clean therapy office. The packets aren’t just being generated — they’re being delivered into a transiently more plastic, better-coupled receiver. There is now a small but real literature testing this prediction.

The plasticity-window basis. Kozyrev and colleagues (2018, PNAS) showed directly in animal cortex that high-frequency 10 Hz TMS produces a transient cortical state of increased excitability and response variability — a permissive period during which directed input gets preferentially imprinted onto the local map. In their paradigm, 30 minutes of passive visual exposure to a single orientation, applied during this post-TMS window, produced a stable shift in cortical orientation preference toward the trained input that was not seen without the TMS priming. The general principle is well-established in spike-timing-dependent plasticity: when TMS modulates a circuit and behaviorally relevant input arrives within a defined post-stimulation window, the synaptic changes that get consolidated reflect the input. This is the bench-science basis for the idea that what you do during and right after a TMS session may shape what gets retuned.

The clinical signal in depression. Donse and colleagues (2018, Brain Stimulation) reported on 196 patients with treatment-resistant MDD receiving simultaneous rTMS and CBT in a large naturalistic cohort — therapy delivered during the stimulation session itself, with the cognitive content scheduled to land inside the post-pulse plasticity window. 66% of patients responded and 56% remitted at end of treatment, with 60% sustained remission at 6-month follow-up. The authors noted these rates are relatively high compared to previous findings in RCTs, though the design was open-label and naturalistic rather than randomized. Xu and colleagues (2023, Brain Sciences) meta-analyzed 27 RCTs of rTMS combined with psychological interventions across multiple conditions and found that active rTMS plus psychological intervention significantly outperformed sham rTMS plus the same psychological intervention on overall clinical symptoms (k=16, SMD 0.31, p<0.01). For the CBT-specific subgroup with 10 or more sessions, the effect was smaller but still significant (k=3, SMD 0.21, p<0.01).

The honest limitation. A more recent scoping review and meta-analysis (Giron 2025) made the more rigorous comparison — active rTMS plus an active psychological intervention versus active rTMS plus sham intervention — and found no significant added benefit across the four controlled trials that enabled this contrast (p = 0.96). The Adu 2023 randomized pilot of rTMS plus internet-delivered CBT in treatment-resistant depression also showed no advantage over rTMS alone. The directionally positive Donse data and the Xu meta-analytic signal are real, but the cleaner controlled question — does adding therapy to TMS outperform TMS by itself? — remains genuinely open. Possible reasons: unguided digital CBT may be insufficient; the optimal timing window may be tighter than current trials have tested; therapy modality and patient-therapist alignment matter; and existing trials have been small.

The PTSD case is instructive. Isserles and colleagues (2013) ran a pilot in PTSD pairing brief script-driven trauma exposure with deep TMS over the medial PFC and showed reductions in intrusive symptoms not seen in either control arm. The follow-up multisite RCT (Isserles 2021, Biological Psychiatry, n=125) tested the same combination more rigorously and found that both groups improved substantially, with the sham-TMS-plus-exposure arm doing slightly better than active-TMS-plus-exposure. The authors’ interpretation: brief script-driven exposure may itself be a viable PTSD treatment, and mPFC stimulation may have interfered with the trauma memory–mediated extinction the exposure was driving. The lesson isn’t that combination doesn’t work — it’s that combinations are sensitive to mechanism. Pairing stimulation with cognitive content that engages the same circuit isn’t automatically additive; the timing, frequency, and target need to match the kind of learning the therapy is trying to produce.

What this means for practice. The mechanism-level argument for offering CBT alongside a TMS course is strong: the brain is in a transiently more plastic state during and after sessions, the cognitive content that drives clinical change in CBT depends on the integration architecture TMS is restoring, and the largest naturalistic cohort with concurrent treatment reports better-than-typical outcomes. The randomized data on incremental benefit over TMS alone is mixed and the best-controlled comparisons are still small. A reasonable clinical posture: where patients are already engaged in CBT or interested in starting it, scheduling sessions during or shortly after the TMS course is biologically sensible and unlikely to be harmful, and may help. Whether it adds enough over TMS alone to justify recommending it for patients who are not otherwise interested is not yet settled.

This framework explains several clinical observations: why precision matters (Cash 2021 — scalp-based targeting misses the individual optimal DLPFC site by ~30 mm on average), why individualized fMRI-guided targeting improves outcomes (the DLPFC locus most anticorrelated with sgACC varies substantially between individuals), and why response prediction may become possible (baseline network connectivity patterns may identify who will respond to which protocol).

Evidence by indication

Evidence strength varies substantially by indication, and — importantly — does not always track with FDA clearance. Data below reflects meta-analyses and large real-world registries where available.

Major depressive disorder

Largest evidence base in the field. Multiple RCTs, meta-analyses, and a large real-world registry (NeuroStar, n=5,010) support both high-frequency left-sided and iTBS protocols for MDD. Key numbers worth knowing:

Durability

Two key studies on how long response lasts:

• Dunner 2014 — Multisite naturalistic observational study of 257 patients with pharmacoresistant MDD. 53% met response or remission criteria after the acute series. Among initial responders, 62.5% continued to meet response criteria throughout 12-month follow-up. ~36% of patients received some form of reintroduction TMS during follow-up.
• Senova 2019 — Systematic review and meta-analysis of depression outcomes at 3, 6, and 12 months after a TMS course. Among initial responders, sustained response rates were 66.5% at 3 months, 52.9% at 6 months, and 46.3% at 12 months. The authors concluded that maintenance treatment may enhance durability and should be considered in clinical practice.

Meta-analytic effect size

Schutter 2009 meta-analysis of double-blind sham-controlled studies found an overall weighted mean effect size of d = 0.39 (95% CI 0.25–0.54, p<0.0001) for high-frequency rTMS over left DLPFC — comparable in magnitude to antidepressant medications. Vida 2023 reported a risk ratio of 2.25 for response and 2.78 for remission with rTMS as add-on vs. standard pharmacotherapy in patients with two prior treatment failures (19 RCTs, 854 patients).

Low-frequency right-sided: a viable alternative

Low-frequency right-sided rTMS performs comparably to high-frequency left-sided treatment, with a possible advantage in side-effect profile and seizure risk:

• Berlim 2013 — Meta-analysis of 8 RCTs (n=263). Low-frequency right-sided rTMS: 38.2% response (vs. 15.1% sham; OR 3.35) and 34.6% remission (vs. 9.7% sham; OR 4.76). Conclusion: LF-rTMS is a promising treatment for MDD, with clinically meaningful benefits comparable to standard antidepressants and HF-rTMS.
• Chen 2013 — Meta-analysis directly comparing HF-left vs LF-right rTMS. No significant difference in efficacy (OR 1.15; 95% CI 0.65–2.03). Authors noted LF right-sided rTMS produces fewer side effects and is more protective against seizures.

iTBS vs. conventional rTMS

Bulteau 2022 (THETA-DEP) directly compared iTBS to 10 Hz rTMS in treatment-resistant unipolar depression. Response rates were 36.7% (iTBS) vs. 33.3% (10 Hz rTMS); remission rates were 18.5% vs. 14.8%. The conclusion: iTBS is as effective as rTMS and likely more cost-effective — a major reason iTBS has become a workhorse protocol.

Obsessive–compulsive disorder

FDA-cleared in 2018 for deep TMS with H-coil targeting mPFC/ACC. The pivotal trial and subsequent data:

• Carmi 2019 — Multi-center, randomized, double-blind, sham-controlled trial of dTMS for OCD. 29 sessions over 6 weeks. Active treatment: 20 Hz dTMS at 100% MT, 50 trains of 2-second pulse trains with 20-second intervals, totaling 2,000 pulses per session. Mean Y-BOCS change: −6.0 points active vs. −3.3 sham (effect size 0.69). Primary response (6 weeks): 38.1% active vs. 11.1% sham; at 1-month follow-up: 45.2% vs. 17.8%.
• Gersner 2019 — Pooled multicenter + open-label data (n=68). Mean Y-BOCS reduction of 8.6 points (30.1%) through week 6. 60.3% partial response (≥20% Y-BOCS decrease); 47.1% full response (≥30% decrease).

Smoking cessation

FDA-cleared in 2020. Deep TMS approach targeting bilateral insula and prefrontal cortex. Useful alternative for patients who have failed pharmacotherapy and behavioral approaches.

MDD with anxious distress / anxiety

FDA-cleared in 2021–22 for MDD with comorbid anxious distress (not a standalone anxiety indication).

Cox 2022 — Systematic review and meta-analysis of rTMS for GAD and panic disorder (13 studies, 677 patients). In GAD patients with or without comorbidities, rTMS produced significant improvements in anxiety (SMD 1.45; p<0.001) and depression scores (SMD 1.65; p<0.001) regardless of rTMS parameters. For panic disorder, overall severity improvement did not reach statistical significance. Authors conclude rTMS is a safe and effective treatment for improving anxiety scores in GAD.

Off-label and emerging indications

TMS has been studied in multiple conditions beyond its FDA-cleared indications. Evidence quality varies widely, and for most of these indications the literature consists of small trials, heterogeneous protocols, and meta-analyses with significant methodological limitations. The table below summarizes what the strongest available meta-analyses show, with notes on the limitations that matter clinically.

How to read this table. Effect sizes reported here come from published meta-analyses, not from individual trials. They are not directly comparable to the response/remission rates quoted for MDD because the underlying studies use different outcome measures and report statistics differently. For off-label use, “there is a signal” is a more accurate characterization than specific response-rate numbers. Rigorous head-to-head RCTs specifically designed for each of these indications — rather than secondary analyses of depression trials — remain the research priority.

FDA clearance history

How TMS works — mechanism

The exact mechanism by which TMS alleviates depression remains incompletely understood. Current evidence suggests it restores aberrantly functioning neurocircuits rather than producing a single-pathway pharmacologic effect. Several lines of evidence converge:

• Subgenual cingulate connectivity — Weigand 2018 demonstrated prospectively that functional connectivity between the individual cortical target (DLPFC) and the subgenual cingulate predicts antidepressant response to rTMS. This finding underpins fMRI-guided targeting approaches and the SAINT protocol.
• Structural change — Furtado 2013 showed increased left amygdala volume in responders and decreased left hippocampus volume in non-responders on neuroimaging, suggesting rTMS may promote neurogenesis or other effects favoring neuronal plasticity.
• Cortico-striatal-thalamic-cortical loop — Peters 2016 describes the CSTC loop as a central pathway implicated in psychiatric disease and its response to neuromodulation.
• Salience network disruption across diagnoses — McTeague 2017 analyzed 5,493 patients across MDD, bipolar, schizophrenia, anxiety, and SUDs vs. 5,728 controls and found shared abnormal activation in the anterior cingulo-insular or “salience network.” The authors describe this as a transdiagnostic vulnerability — networks intrinsic to adaptive, flexible cognition are vulnerable to broad-spectrum psychopathology.

Cutting edge & prediction

Three trends define where the field is moving: precision targeting (fMRI-guided individualization), accelerated protocols (days, not weeks), and prediction (biomarkers and AI to identify who will respond and personalize stimulation). Each is summarized below, followed by a candid look at why some patients still don't respond — and how close we are to predicting that before the first session.

fMRI-guided targeting

Preliminary work from the Acacia and Harvard collaborative group (n≈195, preprint) compared standard scalp-based targeting to individualized fMRI-guided targeting using the sgACC-anticorrelation principle. Early numbers suggest response rates may rise substantially with fMRI-guided targeting compared with scalp-based methods, with median targeting error under scalp methods of ~30 mm. If replicated, this would support fMRI-guided targeting as a meaningful advance in precision TMS. Specific percentages circulating in industry communications (e.g., 62% vs. 77.5% response, OR 2.3) should be confirmed against the peer-reviewed publication before being cited clinically.

SAINT / SNT

Stanford Neuromodulation Therapy: 5-day accelerated iTBS with individualized fMRI-based targeting. In the pivotal randomized trial, roughly 78.6% of participants met remission criteria after the active arm (Cole 2022). FDA-cleared under K213481. Real-world replication is ongoing; outcomes in community settings remain an active question.

ONE-D

Shorter-course accelerated protocol under active development and study. Preliminary reports (Vaughn 2025) describe encouraging response rates on HDRS and BDI measures in early cohorts, but peer-reviewed publication is pending.

SWIFT

BrainsWay accelerated deep TMS protocol. Industry registration data describe strong response and remission outcomes; FDA clearance has been reported. The specific percentages circulating in manufacturer communications should be confirmed against the peer-reviewed publication and the FDA 510(k) summary before being used in clinical documentation or patient counseling.

Why some patients don't respond

Response rates of 60–80% mean 20–40% of patients don't get better from a standard TMS course. The clinically useful question isn't whether TMS works — the evidence is settled there — but why it doesn't work for some patients. Non-response is not a single phenomenon, and it is not a moral failing of the treatment or the patient. In practice, when a course doesn't produce the expected result, the cause usually falls into one of three categories — and the distinctions matter, because the right next step differs substantially across them.

Clinically, this framework shifts the first question after a limited response from “was the TMS protocol adequate?” to “which of these three factors is most likely in play, and what's the next reasonable move?” The answer is usually not more pulses at a higher intensity.

Can we predict response before the first session?

Prediction research has moved substantially in the past five years. Three signal classes are emerging — imaging-based, EEG-based, and multimodal machine learning — that may let us stratify patients to likely-responder and likely-non-responder groups before a course begins. Each has real data behind it and real limitations worth naming.

AI in TMS today

AI's role in TMS isn't a future promise — it's already shaping how the field researches, delivers, and personalizes treatment. The thread running through all of it: TMS is heterogeneous (patients, brains, responses), and AI is how we're finally engaging that complexity at scale. Three places it lives in current practice:

What this means for practice today

None of these prediction tools are yet standard of care, and none should be used to deny a reasonable TMS trial to a patient who would otherwise qualify. But the framing has shifted. The sequence — treat first, then learn retrospectively who responded — is beginning to flip toward stratify first, then treat.

For the referring clinician today, the practical implications are modest but real:

• A pre-TMS workup that screens for — and treats — medical contributors (thyroid, sleep, B12, pain, substance use) is a higher-yield intervention than protocol tuning in many limited-response cases.
• A social and environmental assessment matters. If the life circumstance is actively driving the pathology, a stimulation-only plan is likely to disappoint regardless of targeting precision.
• When a patient has not responded to one TMS protocol, the alternative is rarely more of the same. Switching laterality (HF-left → LF-right), switching to iTBS or deep TMS, moving to an accelerated or fMRI-guided protocol, or reconsidering ECT or ketamine on mechanism-diversity grounds are all better-supported moves than simply extending the original course.
• Biomarker-guided stratification — particularly EEG-based measures like iAF that are cheap and repeatable — is the most plausible near-term clinical entry point. Worth tracking. Not yet ready to deploy.

Contraindications

The most common reason a referral is declined or delayed. Worth screening for before sending the patient.

Absolute contraindications

• Ferromagnetic metal in or near the head — aneurysm clips, cochlear implants, intracranial stents or coils, deep brain stimulators, shrapnel, or other non-removable ferromagnetic hardware within roughly 30 cm of the coil.
• Active implanted devices in the stimulation field — cochlear implants, vagus nerve stimulators (VNS), and intracranial electrodes.
• Cardiac pacemakers or ICDs where the device or leads are within the magnetic field.
• Medication infusion pumps (e.g., intrathecal pumps) that could be disrupted by the field.

Relative contraindications / caution

• History of seizure or lowered seizure threshold — personal seizure history, recent CNS lesions, recent significant head trauma, withdrawal states, or concurrent proconvulsant medication. TMS lowers the seizure threshold; overall risk is low but not zero.
• Active mania or rapid cycling — risk of treatment-emergent mood elevation, particularly in bipolar-spectrum patients.
• Active substance use — complicates risk assessment, adherence, and interpretation of response.
• Pregnancy — limited data; not an absolute contraindication but warrants case-by-case review with OB and psychiatry.
• Significant hearing loss or inadequate ear protection — coil clicks can reach intensities requiring protection; pre-existing hearing concerns need documentation and appropriate ear protection.
• Unstable medical or psychiatric comorbidity — acute suicidality, unstable cardiac disease, or uncontrolled medical illness may warrant stabilization before starting a course.

Common side effects (not contraindications)

Headache, scalp irritation or discomfort at the stimulation site, transient lightheadedness, and fatigue after sessions. These are usually self-limited and improve over the first week of treatment.

Referral tools

Practical guidance for referring clinicians.

What to include in the referral note

The items below are the clinical documentation that supports a good TMS evaluation — and, independent of any specific program's intake process, they are the elements payers typically require for prior authorization.

• Current PHQ-9or equivalent severity score.
• Antidepressant trials in the current episodefor each, include class, dose, duration, and reason for stopping or insufficient response. Payers typically require two or more adequate trials from different classes.
• Psychotherapy trialCBT or equivalent, with duration and outcome.
• ECT statusconsidered, offered, or reason not appropriate.
• Contraindication screendocumented, including implants and seizure risk.
• Diagnosis and severitydocumented in the current note, not just the problem list.

When to refer

What we can say confidently

Six statements that hold up across the current evidence base. Useful when talking to patients, families, and skeptical colleagues.

References

Full reference list for material cited above, in AMA format with DOIs and PMIDs where available. Citations marked (verify) are newer or preprint-stage and should be independently confirmed before being cited in clinical documentation.

1. Schutter DJ. Antidepressant efficacy of high-frequency transcranial magnetic stimulation over the left dorsolateral prefrontal cortex in double-blind sham-controlled designs: a meta-analysis. Psychol Med. 2009;39(1):65-75. doi:10.1017/S0033291708003462. PMID: 18447962.
2. Schutter DJ. Quantitative review of the efficacy of slow-frequency magnetic brain stimulation in major depressive disorder. Psychol Med. 2010;40(11):1789-95. doi:10.1017/S003329171000005X. PMID: 20102670.
3. O’Reardon JP, Solvason HB, Janicak PG, et al. Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial. Biol Psychiatry. 2007;62(11):1208-1216. doi:10.1016/j.biopsych.2007.01.018. PMID: 17573044.
4. George MS, Lisanby SH, Avery D, et al. Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial. Arch Gen Psychiatry. 2010;67(5):507-516. doi:10.1001/archgenpsychiatry.2010.46. PMID: 20439832.
5. Berlim MT, Van den Eynde F, Jeff Daskalakis Z. Clinically meaningful efficacy and acceptability of low-frequency repetitive transcranial magnetic stimulation (rTMS) for treating primary major depression: a meta-analysis of randomized, double-blind and sham-controlled trials. Neuropsychopharmacology. 2013;38(4):543-551. doi:10.1038/npp.2012.237. PMID: 23249815.
6. Chen J, Zhou C, Wu B, et al. Left versus right repetitive transcranial magnetic stimulation in treating major depression: a meta-analysis of randomised controlled trials. Psychiatry Res. 2013;210(3):1260-1264. doi:10.1016/j.psychres.2013.09.007. PMID: 24113125.
7. Dunner DL, Aaronson ST, Sackeim HA, et al. A multisite, naturalistic, observational study of transcranial magnetic stimulation for patients with pharmacoresistant major depressive disorder: durability of benefit over a 1-year follow-up period. J Clin Psychiatry. 2014;75(12):1394-1401. doi:10.4088/JCP.13m08977. PMID: 25271871.
8. Deng ZD, Lisanby SH, Peterchev AV. Coil design considerations for deep transcranial magnetic stimulation. Clin Neurophysiol. 2014;125(6):1202-1212. doi:10.1016/j.clinph.2013.11.038. PMID: 24411523.
9. Peters SK, Dunlop K, Downar J. Cortico-striatal-thalamic loop circuits of the salience network: a central pathway in psychiatric disease and treatment. Front Syst Neurosci. 2016;10:104. doi:10.3389/fnsys.2016.00104. PMID: 28082874.
10. McTeague LM, Huemer J, Carreon DM, Jiang Y, Eickhoff SB, Etkin A. Identification of common neural circuit disruptions in cognitive control across psychiatric disorders. Am J Psychiatry. 2017;174(7):676-685. doi:10.1176/appi.ajp.2017.16040400. PMID: 28320224.
11. Weigand A, Horn A, Caballero R, et al. Prospective validation that subgenual connectivity predicts antidepressant efficacy of transcranial magnetic stimulation sites. Biol Psychiatry. 2018;84(1):28-37. doi:10.1016/j.biopsych.2017.10.028. PMID: 29274805.
12. Furtado CP, Hoy KE, Maller JJ, Savage G, Daskalakis ZJ, Fitzgerald PB. An investigation of medial temporal lobe changes and cognition following antidepressant response: a prospective rTMS study. Brain Stimul. 2013;6(3):346-354. doi:10.1016/j.brs.2012.06.006. PMID: 22784443.
13. Senova S, Cotovio G, Pascual-Leone A, Oliveira-Maia AJ. Durability of antidepressant response to repetitive transcranial magnetic stimulation: systematic review and meta-analysis. Brain Stimul. 2019;12(1):119-128. doi:10.1016/j.brs.2018.10.001. PMID: 30344109.
14. Filipčić I, Šimunović Filipčić I, Milovac Ž, et al. Efficacy of repetitive transcranial magnetic stimulation using a figure-8-coil or an H1-coil in treatment of major depressive disorder; a randomized clinical trial. J Psychiatr Res. 2019;114:113-119. doi:10.1016/j.jpsychires.2019.04.020. PMID: 31059991.
15. Carmi L, Tendler A, Bystritsky A, et al. Efficacy and safety of deep transcranial magnetic stimulation for obsessive-compulsive disorder: a prospective multicenter randomized double-blind placebo-controlled trial. Am J Psychiatry. 2019;176(11):931-938. doi:10.1176/appi.ajp.2019.18101180. PMID: 31109199.
16. Gersner R, Sisko E, Tendler A. How much can patients expect to improve with six weeks of deep TMS for OCD? Brain Stimul. 2019;12(2):525.
17. Li CT, Cheng CM, Chen MH, et al. Antidepressant efficacy of prolonged intermittent theta burst stimulation monotherapy for recurrent depression and comparison of methods for coil positioning: a randomized, double-blind, sham-controlled study. Biol Psychiatry. 2020;87(5):443-450. doi:10.1016/j.biopsych.2019.07.031. PMID: 31563272.
18. Cole EJ, Stimpson KH, Bentzley BS, et al. Stanford Accelerated Intelligent Neuromodulation Therapy for treatment-resistant depression. Am J Psychiatry. 2020;177(8):716-726. doi:10.1176/appi.ajp.2019.19070720. PMID: 32252538.
19. Sackeim HA, Aaronson ST, Carpenter LL, et al. Clinical outcomes in a large registry of patients with major depressive disorder treated with transcranial magnetic stimulation. J Affect Disord. 2020;277:65-74. doi:10.1016/j.jad.2020.08.005. PMID: 32799106.
20. Cash RFH, Cocchi L, Lv J, Fitzgerald PB, Zalesky A. Functional magnetic resonance imaging–guided personalization of transcranial magnetic stimulation treatment for depression. JAMA Psychiatry. 2021;78(3):337-339. doi:10.1001/jamapsychiatry.2020.3794.
21. Bulteau S, Laurin A, Pere M, et al. Intermittent theta burst stimulation (iTBS) versus 10 Hz high-frequency repetitive transcranial magnetic stimulation (rTMS) to alleviate treatment-resistant unipolar depression: a randomized controlled trial (THETA-DEP). Brain Stimul. 2022;15(3):870-880. doi:10.1016/j.brs.2022.05.011. PMID: 35609816.
22. Caulfield KA, Fleischmann HH, George MS, McTeague LM. A transdiagnostic review of safety, efficacy, and parameter space in accelerated transcranial magnetic stimulation. J Psychiatr Res. 2022;152:384-396. doi:10.1016/j.jpsychires.2022.06.038. PMID: 35816982.
23. Cole EJ, Phillips AL, Bentzley BS, et al. Stanford Neuromodulation Therapy (SNT): a double-blind randomized controlled trial. Am J Psychiatry. 2022;179(2):132-141. doi:10.1176/appi.ajp.2021.20101429. PMID: 34711062.
24. Cox J, Thakur B, Alvarado L, Shokar N, Thompson PM, Dwivedi AK. Repetitive transcranial magnetic stimulation for generalized anxiety and panic disorders: a systematic review and meta-analysis. Ann Clin Psychiatry. 2022;34(2):e2-e24. doi:10.12788/acp.0050. PMID: 35550035.
25. Watts D, Pulice RF, Reilly J, Brunoni AR, Kapczinski F, Passos IC. Predicting treatment response using EEG in major depressive disorder: a machine-learning meta-analysis. Transl Psychiatry. 2022;12(1):332. doi:10.1038/s41398-022-02064-z. PMID: 35961967.
26. Vida RG, Sághy E, Bella R, et al. Efficacy of repetitive transcranial magnetic stimulation (rTMS) adjunctive therapy for major depressive disorder (MDD) after two antidepressant treatment failures: meta-analysis of randomized sham-controlled trials. BMC Psychiatry. 2023;23(1):545. doi:10.1186/s12888-023-05033-y. PMID: 37501135.
27. Voetterl HTS, Sack AT, Olbrich S, et al. Alpha peak frequency-based Brainmarker-I as a method to stratify to pharmacotherapy and brain stimulation treatments in depression. Nat Mental Health. 2023;1:1023-1032. doi:10.1038/s44220-023-00160-7.
28. Drysdale AT, Grosenick L, Downar J, et al. Resting-state connectivity biomarkers define neurophysiological subtypes of depression. Nat Med. 2017;23(1):28-38. doi:10.1038/nm.4246. PMID: 27918562.
29. Dinga R, Schmaal L, Penninx BWJH, et al. Evaluating the evidence for biotypes of depression: methodological replication and extension of Drysdale et al. (2017). NeuroImage Clin. 2019;22:101796. doi:10.1016/j.nicl.2019.101796. PMID: 30935858.
30. Moser P, Reishofer G, Prückl R, Schaffelhofer S, Freigang S, Thumfart S, Mahdy Ali K. Real-time estimation of the optimal coil placement in transcranial magnetic stimulation using multi-task deep learning. Sci Rep. 2024;14(1):19361. doi:10.1038/s41598-024-70367-w. PMID: 39169126.
31. DeSouza DD, Meng NF, DeGaetano NP, Nakano E, Hoang V, Nguyen T, Ling DS, Carreon DM, Siddiqi SH. Naturalistic outcomes with fMRI-guided and non-fMRI-guided accelerated TMS for depression. medRxiv preprint, posted September 5, 2025. doi:10.1101/2025.09.03.25334813. (preprint)
32. Kishi T, Ikuta T, Sakuma K, Hatano M, Matsuda Y, Kito S, Iwata N. Repetitive transcranial magnetic stimulation for bipolar depression: a systematic review and pairwise and network meta-analysis. Mol Psychiatry. 2023;28(3):857-861. doi:10.1038/s41380-023-02045-8.
33. Vaughn DA, Marino B, Engelbertson A, Dojnov A, Johnson L, Stine M, Vila-Rodriguez F, Weiss N, Nanos G, Downar J. Real-world effectiveness of a single-day regimen for transcranial magnetic stimulation using Optimized, Neuroplasticity-Enhanced techniques in Depression (ONE-D): an open-label case series. Transcranial Magnetic Stimulation. 2025. doi:10.1016/j.transm.2025.100200.
34. BrainsWay Ltd. Landmark data validate BrainsWay’s SWIFT™ Deep TMS: accelerated 6-session protocol achieves comparable response (87.8%) and remission (78.0%) to standard 20-session protocol in a multisite randomized non-inferiority trial. FDA clearance announced September 2025; data published in two peer-reviewed manuscripts in Brain Stimulation, 2026. (industry press release; peer-reviewed publication forthcoming)
35. Ventura F, Frias P, Rodrigues da Silva D, McGirr A, Cotovio G, Oliveira-Maia AJ. Efficacy, effectiveness, and safety of transcranial magnetic stimulation for bipolar depression: a systematic review and meta-analysis. Biol Psychiatry Glob Open Sci. 2026;6(1):100172. doi:10.1016/j.bpsgos.2025.100172.
36. Brown R, Cherian K, Jones K, Wickham R, Gomez R, Sahlem G. Repetitive transcranial magnetic stimulation for post-traumatic stress disorder in adults. Cochrane Database Syst Rev. 2024;8(8):CD015040. doi:10.1002/14651858.CD015040.pub2. PMID: 39092744.
37. Mehta DD, Praecht A, Ward HB, et al. A systematic review and meta-analysis of neuromodulation therapies for substance use disorders. Neuropsychopharmacology. 2024;49(4):649-680. doi:10.1038/s41386-023-01776-0. PMID: 38086901.
38. Parikh TK, Strawn JR, Walkup JT, Croarkin PE. Repetitive transcranial magnetic stimulation for generalized anxiety disorder: a systematic literature review and meta-analysis. Int J Neuropsychopharmacol. 2022;25(2):144-146. doi:10.1093/ijnp/pyab077. PMID: 34791275.
39. Yan T, Xie Q, Zheng Z, Zou K, Wang L. Different frequency repetitive transcranial magnetic stimulation (rTMS) for posttraumatic stress disorder (PTSD): a systematic review and meta-analysis. J Psychiatr Res. 2017;89:125-135. doi:10.1016/j.jpsychires.2017.02.021. PMID: 28278422.
40. Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of human motor cortex. Lancet. 1985;1(8437):1106-1107. doi:10.1016/s0140-6736(85)92413-4. PMID: 2860322.
41. Mayberg HS, Lozano AM, Voon V, et al. Deep brain stimulation for treatment-resistant depression. Neuron. 2005;45(5):651-660. doi:10.1016/j.neuron.2005.02.014. PMID: 15748841.
42. Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A. Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry. 2012;72(7):595-603. doi:10.1016/j.biopsych.2012.04.028. PMID: 22658708.
43. Berlim MT, Van den Eynde F, Tovar-Perdomo S, Daskalakis ZJ. Response, remission and drop-out rates following high-frequency repetitive transcranial magnetic stimulation (rTMS) for treating major depression: a systematic review and meta-analysis of randomized, double-blind and sham-controlled trials. Psychol Med. 2014;44(2):225-239. doi:10.1017/S0033291713000512. PMID: 23507264.
44. Blumberger DM, Vila-Rodriguez F, Thorpe KE, et al. Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic stimulation in patients with depression (THREE-D): a randomised non-inferiority trial. Lancet. 2018;391(10131):1683-1692. doi:10.1016/S0140-6736(18)30295-2. PMID: 29726344.
45. George MS, Huffman S, Doose J, et al. EEG synchronized left prefrontal transcranial magnetic stimulation (TMS) for treatment resistant depression is feasible and produces an entrainment dependent clinical response: a randomized controlled double blind clinical trial. Brain Stimul. 2023;16(6):1753-1763. doi:10.1016/j.brs.2023.11.010. PMID: 38040143.
46. Liston C, Chen AC, Zebley BD, et al. Default mode network mechanisms of transcranial magnetic stimulation in depression. Biol Psychiatry. 2014;76(7):517-526. doi:10.1016/j.biopsych.2014.01.023. PMID: 24629537.
47. Baeken C, Marinazzo D, Wu GR, et al. Accelerated HF-rTMS in treatment-resistant unipolar depression: insights from subgenual anterior cingulate functional connectivity. World J Biol Psychiatry. 2014;15(4):286-297. doi:10.3109/15622975.2013.872295. PMID: 24447053.
48. Taylor SF, Ho SS, Abagis T, et al. Changes in brain connectivity during a sham-controlled, transcranial magnetic stimulation trial for depression. J Affect Disord. 2018;232:143-151. doi:10.1016/j.jad.2018.02.019. PMID: 29494898.
49. Mulders PC, van Eijndhoven PF, Schene AH, Beckmann CF, Tendolkar I. Resting-state functional connectivity in major depressive disorder: a review. Neurosci Biobehav Rev. 2015;56:330-344. doi:10.1016/j.neubiorev.2015.07.014. PMID: 26234819.
50. Greicius MD, Flores BH, Menon V, et al. Resting-state functional connectivity in major depression: abnormally increased contributions from subgenual cingulate cortex and thalamus. Biol Psychiatry. 2007;62(5):429-437. doi:10.1016/j.biopsych.2006.09.020. PMID: 17210143.
51. Stott R. When head and heart do not agree: a theoretical and clinical analysis of rational-emotional dissociation (RED) in cognitive therapy. J Cogn Psychother. 2007;21(1):37-50. doi:10.1891/088983907780493313.
52. Canolty RT, Knight RT. The functional role of cross-frequency coupling. Trends Cogn Sci. 2010;14(11):506-515. doi:10.1016/j.tics.2010.09.001. PMID: 20932795.
53. Alekseichuk I, Turi Z, Amador de Lara G, Antal A, Paulus W. Spatial working memory in humans depends on theta and high gamma synchronization in the prefrontal cortex. Curr Biol. 2016;26(12):1513-1521. doi:10.1016/j.cub.2016.04.035. PMID: 27238283.
54. Bramson B, den Ouden HE, Toni I, Roelofs K. Improving emotional-action control by targeting long-range phase-amplitude neuronal coupling. eLife. 2020;9:e59600. doi:10.7554/eLife.59600. PMID: 33161846.
55. Noda Y, Zomorrodi R, Saeki T, et al. Resting-state EEG gamma power and theta-gamma coupling enhancement following high-frequency left dorsolateral prefrontal rTMS in patients with depression. Clin Neurophysiol. 2017;128(3):424-432. doi:10.1016/j.clinph.2016.12.023. PMID: 28160748.
56. Cavanagh JF, Frank MJ. Frontal theta as a mechanism for cognitive control. Trends Cogn Sci. 2014;18(8):414-421. doi:10.1016/j.tics.2014.04.012. PMID: 24835663.
57. Kozyrev V, Staadt R, Eysel UT, Jancke D. TMS-induced neuronal plasticity enables targeted remodeling of visual cortical maps. Proc Natl Acad Sci U S A. 2018;115(25):6476-6481. doi:10.1073/pnas.1802798115. PMID: 29866856.
58. Donse L, Padberg F, Sack AT, Rush AJ, Arns M. Simultaneous rTMS and psychotherapy in major depressive disorder: clinical outcomes and predictors from a large naturalistic study. Brain Stimul. 2018;11(2):337-345. doi:10.1016/j.brs.2017.11.004. PMID: 29174304.
59. Isserles M, Shalev AY, Roth Y, et al. Effectiveness of deep transcranial magnetic stimulation combined with a brief exposure procedure in post-traumatic stress disorder — a pilot study. Brain Stimul. 2013;6(3):377-383. doi:10.1016/j.brs.2012.07.008. PMID: 22921765.
60. Isserles M, Tendler A, Roth Y, et al. Deep transcranial magnetic stimulation combined with brief exposure for posttraumatic stress disorder: a prospective multisite randomized trial. Biol Psychiatry. 2021;90(10):721-728. doi:10.1016/j.biopsych.2021.04.019. PMID: 34274108.
61. Xu X, Xu M, Su Y, Cao TV, Nikolin S, Moffa A, Loo C, Martin D. Efficacy of repetitive transcranial magnetic stimulation (rTMS) combined with psychological interventions: a systematic review and meta-analysis of randomized controlled trials. Brain Sci. 2023;13(12):1665. doi:10.3390/brainsci13121665.
62. Adu MK, Shalaby R, Eboreime E, et al. Apparent lack of benefit of combining repetitive transcranial magnetic stimulation with internet-delivered cognitive behavior therapy for the treatment of resistant depression: patient-centered randomized controlled pilot trial. Brain Sci. 2023;13(2):293. doi:10.3390/brainsci13020293. PMID: 36831836.
63. Giron CG, Tang AHP, Jin M, Kranz GS. Antidepressant efficacy of administering repetitive transcranial magnetic stimulation (rTMS) with psychological and other non-pharmacological methods: a scoping review and meta-analysis. Psychol Med. 2025. doi:10.1017/S0033291725000315.

All numbered citations have been verified in PubMed/published form. Citations marked (verify) in entries 30–33 are drawn from recent conference proceedings, preprints, or preliminary industry communications and should be independently confirmed before use in patient-facing documentation or consult notes. References 46–63 (added in v4) cover the network-rhythm and theta-gamma coupling mechanism content and the CBT-during-TMS literature.

Last updated: May 2026 · This page is educational and does not constitute individual clinical advice.