Interventional and Neurostimulation Psychiatry (Overview)

What unifies a grab-bag of devices

Surveyed one at a time, the interventional modalities look like an unrelated assortment: a seizure induced under anesthesia, a magnet held to the scalp, an electrode wrapped around a nerve in the neck, a wire threaded into the brainstem's neighbors, a trickle of current between two sponges, a beam of focused sound. They differ in invasiveness, in evidence, in mechanism of delivery, and in the patients they suit. Yet they are not really six treatments for six problems. They are six ways of perturbing the same dysregulated circuitry, at different nodes, depths, and intensities — and, at the cellular level, they converge on a single final common pathway. This overview makes that unity explicit, organizes the modalities along the gradient that actually structures clinical decision-making, confronts the evidence pattern that recurs across the implanted therapies, and locates interventional pharmacology within the same logic.

The first convergence: circuits

Every modality in this series acts on overlapping frontolimbic and reward circuitry. ECT modulates it globally through a generalized seizure. TMS reaches the subgenual cingulate indirectly, by stimulating the dorsolateral prefrontal cortex node functionally connected to it. DBS reaches that same subgenual cingulate — and the ventral striatal reward circuitry — directly, with an implanted electrode. VNS influences the network from the periphery, via brainstem monoaminergic nuclei. tDCS nudges the same prefrontal target subthreshold. Ablative surgery and focused ultrasound interrupt or modulate the deep capsular fibers that connect the loop.

The striking fact is the recurrence of the subgenual cingulate — the structure that also anchors the neuroimaging biomarker literature and the default-mode network account. TMS aims to influence it from the surface; DBS stimulates it directly; ECT normalizes its connectivity. Three radically different interventions, one circuit. This is the strongest evidence that interventional psychiatry is fundamentally a circuit medicine: the modalities are alternative means of access to a shared anatomical target, and choosing among them is partly a question of how deep, how precise, and how invasive an access route a given patient's illness justifies.

The second convergence: plasticity

Beneath the circuit lies the cellular endpoint this entire library treats as its downstream hub. ECT produces the most dramatic pro-plasticity effects in clinical psychiatry — measurable hippocampal volume increase and BDNF elevation. TMS and tDCS induce LTP- and LTD-like synaptic change. DBS and VNS exert their slow, accruing benefits over timescales consistent with structural remodeling. Across the board, the interventional therapies appear to work by restoring neuroplasticity and BDNF signaling in circuits where chronic illness has degraded it.

This is why interventional psychiatry rhymes so closely with the rapid-acting pharmacology discussed below: the interventions cluster at the downstream end, forcing plasticity where the illness has suppressed it. Whatever else distinguishes a seizure from a magnetic pulse from a current, they share this destination.

The organizing gradient: invasiveness × precision × evidence

Clinically, the modalities arrange themselves along a gradient with three correlated axes — how invasive, how anatomically precise, and how strong the controlled evidence. Read in order of increasing invasiveness:

• tDCS — least invasive, surface-level, modest and sham-sensitive evidence, but uniquely scalable and home-deliverable.
• TMS — non-invasive, surface-to-network, moderate and FDA-cleared evidence, with targeting precision improving rapidly.
• ECT — non-incisional but requiring anesthesia and a seizure, global in action, and carrying the strongest efficacy evidence of any modality.
• VNS — surgically implanted, peripheral and diffuse in action, slow in onset, with equivocal controlled evidence but signals of durability.
• DBS — most invasive of the reversible options, maximally precise and deep, with negative controlled trials but rich open-label and connectomic promise.
• Ablative neurosurgery / focused ultrasound — the irreversible extreme, reserved for the most refractory illness, with low-intensity focused ultrasound as the non-invasive deep frontier.

Two patterns fall out of this ordering. First, precision and invasiveness trade off against each other: the only way to reach a deep target precisely, with current technology, is to be invasive — which is exactly the trade-off that low-intensity focused ultrasound threatens to dissolve. Second, and more uncomfortably, evidence quality does not rise with invasiveness. The single best-evidenced modality, ECT, is non-incisional; the two most invasive implanted therapies, VNS and DBS, have the most equivocal controlled evidence. Invasiveness buys precision, not proof.

The recurring evidence pattern: open-label promise, controlled-trial failure

The most important methodological lesson of this series is that the implanted, slow-onset modalities — VNS and DBS — share a signature: striking open-label results, failed randomized sham-controlled primary endpoints, and long-term follow-up data suggesting benefit accruing over time. VNS's RECOVER trial missed its primary endpoint amid an unexpectedly strong sham response; DBS's BROADEN and RECLAIM trials failed outright; both have open-label durability data that look considerably better.

There are two honest readings, and the truth may differ by modality. The skeptical reading is that the controlled trials are correct and the open-label data reflect expectancy and the natural course of illness. The charitable reading is that these trials were defeated by features specific to slow, invasive interventions: onset over months that short blinded endpoints cannot capture; the powerful placebo effect of an implanted device; the near-impossibility of true blinding; and heterogeneous targeting that averages real responders together with patients whose electrodes never engaged the right circuit. The DBS field's pivot toward individualized connectomic targeting is essentially a bet on the charitable reading — that the failures were failures of aim, not of premise. The disciplined stance holds both readings in view and resists collapsing the uncertainty in either direction.

Where interventional pharmacology fits

Interventional psychiatry as a clinical subspecialty has never been only about devices. It encompasses interventional pharmacology — treatments that, like the device therapies, are delivered in a clinical setting, act rapidly on circuits, and are reserved for more refractory illness. Ketamine and its enantiomer esketamine are the central examples, and they belong in this series because they converge on exactly the same endpoint as the device therapies by a pharmacological route.

Esketamine (Spravato) was FDA-approved in 2019 as adjunctive therapy for treatment-resistant depression, gained an indication for depression with acute suicidal ideation in 2020, and in January 2025 became the first and only monotherapy approved for treatment-resistant depression — administered in-clinic under a REMS program because of its dissociative and sedative effects. Its mechanism is the bridge that unifies this series with the glutamatergic etiology and rapid-acting pharmacology accounts: NMDA-receptor antagonism triggers a downstream AMPA-mediated surge in BDNF and mTOR signaling, driving rapid synaptogenesis — the same neuroplasticity endpoint that ECT achieves through a seizure and TMS through repetitive stimulation. Ketamine is, in effect, a pharmacological plasticity intervention that produces in hours what the device therapies build over weeks.

The RECOVER trial illustrates how thoroughly these approaches now interlock: its markedly treatment-resistant participants had, in the large majority, already received TMS, ECT, or esketamine before enrolling in a VNS study. Interventional psychiatry is not a set of competing silos but a layered toolkit for the refractory tail of the illness, unified by rapid action on circuits and convergence on plasticity. The device/drug distinction is less fundamental than the shared logic.

The shared future: biomarker-guided targeting

If circuits and plasticity are the present unity of interventional psychiatry, biomarker-guided precision is its future. The evidence is already accumulating: connectivity-guided TMS targeting (the basis of accelerated SAINT protocols) outperforms fixed-distance landmarks; connectomic, tractography-defined DBS targeting addresses the aiming problem that may have sunk the controlled trials; TMS-EEG offers a direct readout of target engagement; and closed-loop DBS, which stimulates only when it detects a neural signature of the depressed state, is the most literal possible fusion of biomarker and intervention — a diagnostic that gates a treatment in real time. The trajectory across every modality is the same: from one-size-fits-all delivery toward individualized, biomarker-informed, and ultimately adaptive treatment.

Honest limits

Four limits apply across the series. First, these are treatments for the refractory tail, not first-line care — their risk-benefit calculus is justified by the severity and treatment-resistance of the patients who reach them. Second, durability is a universal challenge: acute response, where achieved, relapses without a maintenance strategy, a problem as true of ECT as of ketamine. Third, mechanism is incompletely understood across the board — we know these interventions work, and roughly where and how, far better than we know precisely why, which should temper confident mechanistic storytelling. Fourth, access, cost, and stigma distort use in ways that are clinically and ethically significant — most starkly in the underutilization of ECT, the most effective option, and in the unregulated consumer market that shadows the most accessible one.

A reasonable clinical heuristic follows from the gradient: match invasiveness and risk to severity and refractoriness; prefer reversible interventions over irreversible ones where both reach the target; prefer better-evidenced modalities over promising-but-unproven ones; and plan for maintenance from the outset rather than treating acute response as the goal.

Bottom line

Interventional and neurostimulation psychiatry is best understood not as a catalog of devices but as a single circuit-and-plasticity medicine expressed through many access routes. Every modality, from a current between two sponges to an electrode in the subgenual cingulate to a seizure under anesthesia, perturbs overlapping frontolimbic circuitry and converges on the restoration of neuroplastic signaling — and ketamine reaches the same endpoint pharmacologically, dissolving the device/drug boundary. The modalities array along a gradient where precision and invasiveness rise together but evidence quality does not, where the implanted therapies share a humbling pattern of open-label promise and controlled-trial failure that the field is trying to solve through better targeting, and where the future plainly lies in fusing intervention with the biomarker program. Used for the refractory patients they are meant for, matched to severity, preferring the reversible and the well-evidenced, and planned with durability in mind, these are among the most powerful tools in psychiatry.

Selected references

1. UK ECT Review Group. Efficacy and safety of electroconvulsive therapy in depressive disorders: a systematic review and meta-analysis. Lancet. 2003.
2. Blumberger DM, et al. Theta burst versus high-frequency rTMS in depression (THREE-D). Lancet. 2018.
3. Cole EJ, et al. Stanford Neuromodulation Therapy (SNT) for treatment-resistant depression. Am J Psychiatry. 2021.
4. Fox MD, et al. Efficacy of TMS targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry. 2012.
5. Conway CR, et al. Durability of the benefit of vagus nerve stimulation in markedly treatment-resistant major depression: a RECOVER trial report. Int J Neuropsychopharmacol. 2026.
6. Holtzheimer PE, et al. Subcallosal cingulate DBS for treatment-resistant depression (BROADEN). Lancet Psychiatry. 2017.
7. Riva-Posse P, et al. A connectomic approach for subcallosal cingulate deep brain stimulation surgery. Mol Psychiatry. 2018.
8. Scangos KW, et al. Closed-loop neuromodulation in an individual with treatment-resistant depression. Nat Med. 2021.
9. Woodham RD, et al. Home-based transcranial direct current stimulation for major depressive disorder: a fully remote phase 2 randomized sham-controlled trial. Nat Med. 2024.
10. Brunoni AR, et al. Trial of electrical direct-current therapy versus escitalopram for depression (ELECT-TDCS). N Engl J Med. 2017.
11. Kim SJ, et al. MR-guided focused ultrasound capsulotomy for treatment-refractory obsessive-compulsive disorder. Mol Psychiatry. 2018.
12. Legon W, et al. Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans. Nat Neurosci. 2014.
13. Berman RM, et al. Antidepressant effects of ketamine in depressed patients. Biol Psychiatry. 2000.
14. Zarate CA, et al. A randomized trial of an N-methyl-D-aspartate antagonist in treatment-resistant major depression. Arch Gen Psychiatry. 2006.
15. Daly EJ, et al. Efficacy of esketamine nasal spray plus oral antidepressant in treatment-resistant depression. JAMA Psychiatry. 2019.
16. Johnson & Johnson / US FDA. Esketamine (Spravato) approved as monotherapy for treatment-resistant depression. 2025.
17. Duman RS, Aghajanian GK. Synaptic dysfunction in depression: potential therapeutic targets. Science. 2012.
18. Mutz J, et al. Comparative efficacy and acceptability of non-surgical brain stimulation for major depressive episodes: network meta-analysis. BMJ. 2019.
19. Widge AS, et al. Deep brain stimulation for treatment-resistant psychiatric illnesses: what has gone wrong and what should we do next? Biol Psychiatry. 2016.
20. Williams NR, et al. Neurostimulation and the future of precision psychiatry. Neuropsychopharmacology (review). 2021.