The analysis is notable for being, according to its authors, the first unified comparative framework to rank nine electrical stimulation modalities as a distinct therapeutic category rather than evaluating them in isolation. Previous meta-analyses had concentrated mainly on non-invasive brain stimulation techniques such as repetitive transcranial magnetic stimulation and transcranial direct current stimulation, leaving clinicians without head-to-head comparative evidence across the wider spectrum of bioelectronic approaches. The nine modalities examined spanned a remarkable range of invasiveness and mechanism: four variants of electroconvulsive therapy differentiated by electrode placement and dose, transcranial direct current stimulation, transcranial alternating current stimulation, cranial electrotherapy stimulation, implantable vagus nerve stimulation, non-invasive transcutaneous auricular vagus nerve stimulation, and transcutaneous acupoint electrical stimulation.

When the researchers ranked the interventions using the surface under the cumulative ranking curve, or SUCRA, a statistic that quantifies the probability of each treatment ranking best, transcutaneous auricular vagus nerve stimulation emerged at the top for clinical response rate with a SUCRA value of 99.9 percent. Compared with sham stimulation, taVNS carried an odds ratio of 4.12 (95 percent CI 3.08 to 5.16) for achieving a treatment response, and it also outperformed transcranial alternating current stimulation directly, with an odds ratio of 3.26 (95 percent CI 1.72 to 4.80). The authors caution, however, that SUCRA reflects ranking probability rather than proof of clinical superiority, and the effect estimate was imprecise, resting on a limited number of trials and low certainty of evidence.

The biological plausibility of the taVNS ranking rests on well-mapped neuroanatomy. Vagal afferent fibres project to the nucleus tractus solitarius, which in turn modulates the locus coeruleus noradrenergic system, the dorsal raphe nucleus serotonergic system, and limbic structures including the amygdala and hippocampus. Chronic intermittent vagus nerve stimulation has been shown to induce sustained neuroplastic changes, including increased hippocampal brain-derived neurotrophic factor expression and enhanced neurogenesis, which may explain the progressive and delayed therapeutic response characteristic of vagal stimulation. The evidence in this network largely reflects the non-invasive auricular form of the technique, typically delivered at a frequency of 20 Hz, an intensity of 4 to 6 mA, and a pulse width of 0.2 milliseconds.

For continuous symptom severity, a different modality took the lead. Transcranial direct current stimulation ranked highest for reducing HAMD scores, with a SUCRA of 71.3 percent and a standardised mean difference of minus 1.70 versus sham (95 percent CI minus 2.80 to minus 0.60). tDCS delivers low-intensity direct current, generally 1 to 2 mA, through scalp electrodes to produce polarity-dependent, subthreshold modulation of cortical excitability. The authors attribute its strong showing partly to statistical precision: tDCS was the most extensively studied modality in the network, contributing the largest pooled sample size for the HAMD outcome. Its mechanism involves NMDA receptor-dependent long-term potentiation-like plasticity, and repeated sessions produce cumulative neuroplastic after-effects mediated by BDNF secretion and calcium-dependent synaptic remodelling that persist beyond the stimulation period itself.

Interestingly, tDCS ranked appreciably lower for the categorical clinical response rate than for the continuous severity score, sitting at a SUCRA of 53.6 percent on the response endpoint. The authors interpret this discrepancy as suggesting that tDCS may produce moderate but broadly distributed symptom improvements across a study population rather than driving a subset of patients across formal response thresholds. This pattern has appeared in previous tDCS trials, where effect sizes on continuous severity scales were more robust than those on response-rate measures, and it may reflect the relatively low stimulation intensities and short treatment durations used in some included trials. Parameter optimisation, including higher current intensities, twice-daily regimens, more sessions, and MRI-guided individualised targeting, may improve the capacity of tDCS to elicit formal clinical responses.

Electroconvulsive therapy, split into four separate nodes for bilateral temporal, bilateral frontal, high-dose unilateral, and low-dose unilateral electrode configurations, presented a more complicated picture. High-dose unilateral ECT ranked highest among the ECT variants for the HAMD score, with a SUCRA of 64.0 percent, and its clinical response rate was comparable to that of bilateral placements, consistent with recent evidence that dose optimisation against individual seizure thresholds can approach bilateral efficacy while preserving cognitive tolerability. Bilateral temporal ECT performed relatively poorly on both efficacy outcomes, but the authors urge careful interpretation, noting probable indication bias: bilateral ECT tends to be prioritised for patients with more severe illness, psychotic or catatonic features, or acute suicide risk, populations in whom achieving an adequate response is inherently harder. Across all four variants, ECT as a class remained among the most effective acute interventions for major depressive disorder.

Transcranial alternating current stimulation demonstrated moderate efficacy on both endpoints, ranking third for clinical response with a SUCRA of 67.3 percent and third for HAMD scores at 62.3 percent. Recent electroencephalography studies support its premise that 10-Hz stimulation applied to the left dorsolateral prefrontal cortex can enhance alpha power and functional connectivity within fronto-limbic networks, addressing the left frontal alpha asymmetry characteristic of depression. However, the heterogeneity of stimulation parameters across the included tACS trials, which spanned 10-Hz, 40-Hz, and individualised alpha-frequency protocols, limits the precision of the pooled estimates. Cranial electrotherapy stimulation and transcutaneous acupoint electrical stimulation, both less extensively investigated in controlled trials, achieved only moderate rankings and did not reach statistical significance in most pairwise comparisons, making their positions provisional.

Methodologically, the review was registered with PROSPERO (CRD420251103264) and conducted in accordance with the PRISMA 2020 statement and its network meta-analysis extension. The team searched 13 databases from inception to 16 July 2026, identifying 8,594 records, of which 33 studies published between 2002 and 2026 ultimately met inclusion criteria requiring a psychiatrist-established diagnosis under ICD-11 or DSM-5-TR, randomised designs, and sham or active electrical stimulation comparators. Risk of bias was assessed with the Cochrane RoB 2 tool, with 28 of 33 studies judged at low risk overall and five raising some concerns in randomisation and deviations from the intended intervention. Certainty of evidence was graded with the GRADE framework, and most comparisons were rated low to moderate, with methodological limitations and imprecision the primary reasons for downgrading. Sensitivity analyses left the results materially unchanged, and Egger’s tests found no statistically significant publication bias for any of the three outcomes.

The authors are candid about the limitations that temper these headline findings. The evidence network remains sparse, with few trials supporting some interventions such as implantable vagus nerve stimulation and transcutaneous acupoint electrical stimulation, and individual sample sizes generally modest. Splitting ECT into four nodes further reduced the studies available per variant, limiting statistical power. Stimulation parameters, including frequency, intensity, and pulse width, exert decisive influence on neurophysiological outcomes but could not be accounted for in the available primary data, potentially homogenising genuinely heterogeneous interventions. Safety outcomes were reported for only a subset of the network, covering tACS, the three ECT variants analysed, and tDCS, precluding a comprehensive comparative safety assessment. Meta-regression found no significant effect modification by intervention type, depression subtype, stimulation intensity, sex ratio, age, illness duration, or treatment course. The overall message is one of cautious optimism: electrical stimulation is associated with improved response rates and reduced symptom severity with acceptable tolerability, taVNS and tDCS each lead on different endpoints, but confirmation in larger, higher-quality randomised controlled trials with standardised parameters and routine adverse-event reporting is needed before these rankings can guide routine clinical decision-making.

Subject of Research: Comparative efficacy and safety of nine electrical stimulation modalities for depression

Article Title: Effects of different electrical stimulation on depression: a systematic review and network meta-analysis

Article References: Chai, J., Li, S., Huang, J., Yao, S., Zhou, L., Chen, X., Liu, F., Ji, C., Chen, H., Tan, S., Li, R., Ma, R., Li, S., & Zhao, S. (2026). Effects of different electrical stimulation on depression: a systematic review and network meta-analysis. Frontiers in Psychiatry, 17, 1962608. https://doi.org/10.3389/fpsyt.2026.1962608

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1962608

Keywords: depression, electrical stimulation, network meta-analysis, taVNS, tDCS, electroconvulsive therapy, tACS, vagus nerve stimulation, systematic review, neuromodulation, HAMD, randomised controlled trials