Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2026 Aug 18;9(8):e2628284. doi: 10.1001/jamanetworkopen.2026.28284

Transcranial Electrical Stimulation and Delirium Among Postoperative Patients

A Randomized Clinical Trial

Daria Antonenko 1,✉, Viktor Bublitz 2, Sophie Leroy 1,3, Ulrike Grittner 1,3,4, Stephan Kersting 5, Henry W S Schroeder 6, Martin Burchardt 7, Marek Zygmunt 8, Saskia Mohrhagen 1,2, Tabea Marie Franck 2, Paul Jakob Quartz 1, Pia Thomas 1, Anna E Fromm 1, Robert Fleischmann 1, Falk von Dincklage 2
PMCID: PMC13487549  PMID: 42611517

Key Points

Question

Does transcranial electrical stimulation (tES), applied to surgical patients in the immediate postoperative period, reduce the incidence of delirium?

Findings

In this randomized clinical trial that studied 225 patients 65 years and older, tES did not reduce the incidence of postoperative delirium, compared with sham. Preregistered subgroup analyses revealed that α-transcranial alternating current stimulation over the posterior parietal cortex may reduce postoperative delirium incidence for patients with longer surgery.

Meaning

These findings suggest no uniform effect of postoperative tES to prevent delirium, but rather a variable benefit dependent on the core risk factor of surgical duration.

Abstract

Importance

Postoperative delirium (POD) occurs as a significant complication after elective surgery.

Objective

To investigate the modulation of delirium incidence by transcranial electric stimulation (tES).

Design, Setting, and Participants

A double-blind, sham-controlled randomized clinical trial (Modulating Delirium Through Stimulation [MODEST]) was conducted at University Medicine Greifswald, Germany, from February 8, 2024, to January 15, 2025. Patients were aged 65 years or older who were scheduled to undergo elective surgery. Exclusion criteria were any history of neurological or psychiatric disease, or intake of central nervous system–active medication. Data were analyzed from June 10, 2025, to June 5, 2026.

Intervention

α-Transcranial alternating current stimulation (tACS) of the salience network, transcranial direct current stimulation (tDCS) of the frontoparietal network (dosing: 20 minutes, 2 mA), or sham, in the postanesthesia care unit (PACU) after major elective surgery.

Main Outcomes and Measures

The primary outcome was POD incidence in the ward during 1 of up to 5 postoperative days for the comparison of tACS vs the sham group. The secondary outcomes included POD in the ward for the comparison of tDCS vs sham group, POD in PACU, POD severity and motor type, and postoperative pain. Prespecified subgroup analyses opted to test the dependence of POD incidence on age, sex, duration, and type of surgery, baseline cognitive function, and preoperative pain levels.

Results

A total of 225 patients (mean [SD] age, 73.0 [6.0] years; 59.6% [134] male) were randomly assigned to tACS (n = 68), tDCS (n = 83), or sham stimulation (n = 74). No substantial between group differences were observed for POD incidence in ward (tACS, 11.5% [6 of 52]; tDCS, 14.3% [9 of 63]; sham, 9.2% [6 of 65]; tACS vs sham: odds ratio [OR], 1.60 [95% CI, 0.47-5.46]; P = .45; tDCS vs sham: OR, 1.51 [95% CI, 0.49-4.68]; P = .47) or in PACU (tACS, 14.6% [7 of 48]; tDCS, 21.4% [12 of 56]; sham, 21.7% [13 of 60]; tACS vs sham: OR, 0.95 [95% CI, 0.38-2.36]; P = .91; tDCS vs sham: OR, 0.91 [95% CI, 0.36-2.30]; P = .83). Further secondary analyses of POD severity and motor type, and modulation of pain levels, did not show any substantial difference between stimulation groups. Prespecified subgroup analyses revealed different modulation of POD incidence in PACU depending on duration of surgery (OR, 0.15 [95% CI, 0.02-0.97]; P = .046), indicating that tACS may reduce POD incidence for patients with longer surgery (absolute risk difference for POD in the tACS compared with the sham group, for a surgical duration of 3 to 5 hours of −26.6% to −46.5%). Ratings of adverse events did not differ between groups (incidence rate ratio, 2.3 [95% CI, 0.8-6.5]), and no serious adverse events were reported.

Conclusions and Relevance

In this randomized clinical trial of tES for POD, there was no evidence for a uniform effect of postoperative tES on the incidence of POD, but it suggested a benefit of parietal α oscillatory enhancement for patients with longer surgery. Future research has to test the specific effect on vulnerable groups, examine the optimal timing for tES application to modulate POD, and investigate the value of personalizing tES parameters.

Trial Registration

German Clinical Trials Register: DRKS00033703

This randomized clinical trial investigates the modulation of delirium incidence by transcranial electric stimulation in patients 65 years and older.

Introduction

The growing number of older patients worldwide requiring surgical interventions poses substantial challenges for perioperative care. Postoperative delirium (POD) affects 15% to 50% of older surgical patients depending on the type of surgery, imposing considerable health care costs through prolonged hospital stays and increased resource use.1,2 Although technical refinements in surgery and anesthesia have enhanced procedural safety, maintaining cognitive health during the perioperative period remains a major concern.3 Perioperative neurocognitive disorders, particularly POD, can diminish surgical benefits by impeding functional recovery.4 Clinically, POD presents with an acute onset of disturbances in attention and consciousness.5 Patients experiencing POD face increased risks for treatment-associated complications, cognitive decline, and long-term care dependency.6 Enabling neurocognitively safe surgery is therefore essential to enhance outcomes for older patients.7

The pathogenesis of delirium involves multiple interconnected mechanisms, including inflammatory processes, neuroendocrine stress activation, converging on impaired neurotransmitter homeostasis, and disrupted network connectivity.8 Emerging evidence characterizes POD as an acute network disorder, with electroencephalographic studies revealing abnormal patterns in α oscillations, impaired functional connectivity, and altered neural network topology.9,10 Beyond acute brain network dysfunction, delirium may induce lasting structural consequences, as indicated by increases in neurofilament light chain levels and microstructural changes.9 Current delirium management relies predominantly on symptom-oriented care, and, although evidence-based nonpharmacological interventions are effective, the burden of POD remains high, and there is little evidence that pharmacological interventions improve outcomes.11,12 Consequently, optimizing preventive strategies addresses a critical unmet need, given the limited evidence for outcome-improving interventions after POD onset.

Transcranial electrical stimulation (tES) represents a noninvasive neuromodulation technique delivering weak electrical currents through scalp electrodes, applying either transcranial alternating current stimulation (tACS) or transcranial direct current stimulation (tDCS) to cortical regions. By modulating activity in critical neural circuits, tES can influence both functional network dynamics and induce microstructural plasticity and has demonstrated therapeutic potential across neurological conditions.13,14,15 Conceptualizing delirium as a potentially reversible acute network disorder provides a mechanistic rationale for network-targeted interventions through tES. To date, 3 previous trials have reported the effect of brain stimulation on targeting neural plasticity in the prefrontal cortex using tDCS16,17 or transcranial magnetic stimulation.18

We conducted a randomized double-blind sham-controlled trial applying tES following elective surgery in older patients. The primary end point assessed whether α-tACS reduces POD incidence on the surgical ward compared with sham stimulation. Secondary analyses examined tDCS effects on ward POD incidence, POD occurrence in the postoperative anesthesia unit (PACU), delirium severity and motor type, and postoperative pain. Prespecified subgroup analyses explored potential effect modifiers including age, sex, surgical duration and type, baseline cognitive status, and preoperative pain levels. Our objective was to evaluate tES efficacy in preventing POD.

Methods

Study Design and Patients

The Modulating Delirium Through Stimulation (MODEST) study was a single-center, double-blind, randomized, sham-controlled trial performed in the departments of Anesthesiology and Neurology at the Greifswald University Hospital, Germany, comparing the effect of tES on the incidence of POD after elective surgery in older adults. The study was approved by the local ethics committee of the University Medicine Greifswald. This trial was designed, conducted, and reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) 2010 reporting guideline for randomized controlled trials.19 Design, statistical analysis plan, and data collection procedures were prospectively aligned with CONSORT requirements to ensure that all recommended methodologic details and end points were captured during data acquisition and fully incorporated into the analyses. The study was registered at the German Clinical Trials Register, and the full study protocol was published20 (trial protocol in Supplement 1).

Inclusion criteria were age 65 years or older, planned elective surgery including vascular, abdominal, thoracic, urological, gynecological, and spinal procedures planned to last more than 60 minutes, and the ability to provide written informed consent. Patients were excluded in cases of known neurological or psychiatric conditions, including a history of epileptic seizures, contraindications for tES application,21 long-term medication with nonopioid central nervous system active drugs, isolation due to multidrug-resistant germs, and inability to read or speak German. All patients gave written informed consent prior to participation. Figure 1 presents the study design.

Figure 1. Schematic Illustration of the Study Design and Simulation Illustration of Modeling on a Template Brain.

Two-panel schematic: study timeline and brain maps for tACS and tDCS fields. Panel A, labeled A with the title Study design. A left to right timeline begins with a rectangular box labeled Patients connected by a line to a circle labeled T0. Beneath T0, the text BL. A vertical line above the timeline at the T0 to T1 interval labeled Surgery. The timeline continues to a circle labeled T1 with text beneath reading 0 min. Above the segment spanning from near T1 toward T3, a horizontal bracket labeled PACU. Between T1 and T2, the label tES appears with a small lightning bolt symbol above the line, and a vertical line drops downward to the text tACS, tDCS, or sham. The timeline continues to a circle labeled T2 with text beneath reading 30 min, then to a circle labeled T3 with text beneath reading 60 min. To the right, above the final segment, the label Ward. The timeline ends at a circle labeled T4-8 with text beneath reading 1-5 d. Panel B, labeled B with the title Type of stimulation. Two semi-transparent head renderings with a colored cortical surface overlay appear side by side. The left rendering labeled tACS; the right rendering labeled tDCS. Between the two renderings, a vertical color bar labeled magnE with units V per m, ranging from 0 at the bottom to 0.5 at the top. The color scale transitions from dark purple at low values through blue, green, yellow, and red at the highest values. In both renderings, warmer colors cluster over superior and posterior dorsal cortical regions, with cooler blues and purples over other cortical areas. Each head includes a small, flat, bluish patch on the scalp near the upper lateral aspect.

Open in a new tab

A, Patients were assessed at baseline after informed consent, prior to surgery (T0), followed by their surgery and a postoperative assessment (T1) with subsequent application of brain stimulation (tES). Subsequent postintervention assessment time points were scheduled 30 minutes (T2) and 60 minutes (T3) after surgery. Patients were assessed in the hospital for a maximum of 5 days (T4 to T8). B, Active stimulation was applied either in an oscillatory α frequency over posterior parietal regions or as anodal direct current over left dorsolateral prefrontal cortex, for 20 minutes with a 2-mA zero-to-peak intensity. Computational modeling on a template brain shows the simulated distribution of current created with the software SimNIBS (simnibs.org). BL indicates baseline; magnE, magnitude of electric field; PACU, postanesthesia care unit; tACS, transcranial alternating current stimulation; tDCS, transcranial direct current stimulation.

Randomization and Blinding

Eligible patients were randomly allocated to 1 of 3 groups (eAppendix in Supplement 2). Patients and investigators were blinded to the stimulation condition by using sham stimulation in the control group: current was initially applied for 30 seconds to elicit the typical tingling sensation of active stimulation on the scalp and automatically turned off subsequently. Previous evidence proved the effectiveness and safety of this method.21

Trial Procedures

After randomization, patients underwent a baseline assessment, followed by their elective surgery, after which tES was applied in PACU during rest. Electrical stimulation was applied via a battery-driven stimulator (Neuroelectrics Starstim Home Research Kit) through 2 round saline-soaked sponge electrodes (5-cm diameter), placed in a neoprene cap, and located over P3 and P4 for tACS and F3 (anode) and Fp2 (cathode) for tDCS. For tACS application, the frequency of 9.5 Hz was chosen to target α oscillations.22 Current was delivered for 20 minutes with a 2-mA intensity (zero to peak) with 10 additional seconds of fade-in and fade-out. For sham control, current was delivered for 30 seconds (plus a 10-second ramp-up and ramp-down) with either the tACS (n/2) or tDCS (n/2) set-up. After the intervention, patients were asked whether they had experienced any adverse events, including itching, pain, burning, warmth, metallic taste, fatigue, or any other discomfort.

Details on a device malfunction identified during the conduct of the study, the verification procedure used to identify affected patients, and the corrective measures implemented are provided in the eAppendix in Supplement 2. Preoperative assessments included clinical and demographic data as well as baseline cognitive function using the Montreal Cognitive Assessment (MoCA).23 Assessment of POD was performed using the German version of the 3-minute diagnostic interview Confusion Assessment Method (3D-CAM).24,25 The 3D-CAM-S score assessing delirium severity was calculated from the 3D-CAM assessment.26 The 3D-CAM combines brief cognitive testing and clinical observations to assess 4 delirium features: acute change or fluctuating course, inattention, disorganized thinking, and altered level of consciousness. These features are combined using the CAM algorithm to produce a binary result: delirium present or absent. POD motor type was assessed using the Richmond Agitation-Sedation Scale (RASS).27 Self-reported pain intensity was assessed using the Numeric Rating Scale (NRS).28

General anesthesia was administered according to the hospital’s standard procedures. Induction was performed with propofol and sufentanil, followed by maintenance with either inhaled sevoflurane or continuous intravenous propofol infusion via a syringe infusion pump. Intraoperative analgesia was provided by continuous or bolus administration of opioids in combination with nonopioid analgesics, at the discretion of the attending anesthesiologist. Muscle relaxation was achieved with neuromuscular blocking agents as required and antagonized, when necessary, at the end of surgery. Airway management used either an endotracheal tube or a laryngeal mask. Standard intraoperative monitoring included hemodynamic and ventilatory parameters, end-tidal sevoflurane concentration, and frontal electroencephalography monitoring. Postoperatively, pain was treated according to the World Health Organization analgesic ladder, and excessive shivering and agitation were managed with clonidine.

Outcome Measures

The primary outcome was defined as the incidence of POD (tACS vs sham)—having at least 1 positive 3D-CAM score across all available assessments (up to 10 assessments over up to 5 postoperative in-hospital days, performed twice daily). Secondary outcome measures were incidence of POD as defined by having at least 1 positive 3D-CAM within the 10 visits in the 5 postoperative in-hospital days (tDCS vs sham), the incidence of delirium in the PACU as defined by having at least 1 positive 3D-CAM score in the PACU (after the intervention), maximal POD severity (3D-CAM-S), delirium motor type (RASS), and postoperative pain in the PACU and in the 5 first postoperative days (NRS).

Statistical Analysis

Sample size estimation was conducted based on the study by Tao et al,16 as detailed in the published trial protocol20 and also provided in Supplement 1. Group-specific multiple imputation of missing values was used based on outcome variables, age, sex, and baseline measures of particular outcomes (eAppendix in Supplement 2). Missing data occurred predominantly because study staff were unable to reach patients at scheduled assessment time points (eg, patient temporarily out of the ward for diagnostic procedures or rehabilitation) or because of limited study-personnel availability. A multiple binary logistic regression was used to test differences between treatment (tACS) and control (sham) in the incidence of POD in the ward (primary outcome analysis), with POD incidence in the ward entered as the dependent variable, and group (tACS or sham), stratification variables (age, sex) included as independent variables. Similar multiple binary logistic regression models using multiple imputed data were used to test differences for tDCS or sham in POD incidence in the ward, and differences for tACS or sham and tDCS or sham in POD incidence in the PACU. Analyses of POD severity, motor type, and pain are presented in the eAppendix in Supplement 2. Subgroup analyses were performed by including interaction terms between subgroups and treatment group in the same adjusted logistic regression models that were used for the main outcomes. Odds ratios (ORs), mean differences, and marginal estimates for subgroups and corresponding 95% CIs are reported. For interactions reaching the prespecified significance threshold, absolute risk differences between treatment and sham at specific values of the subgroup variable were additionally derived as marginal estimates from the same model. No adjustment for multiple testing was applied. All statistical tests were 2 sided, and P < .05 was considered statistically significant. Data were analyzed from June 10, 2025, to June 5, 2026.

Results

From February 8, 2024, to January 15, 2025, 892 patients were screened at clinical admission. A total of 225 patients (mean [SD] age, 73.0 [6.0] years; range, 65-90 years; 76 female patients) were randomly assigned to 1 of 3 groups (tACS [n = 68], tDCS [n = 83], or sham stimulation [n = 74]) (Figure 2). Overall, 45 patients dropped out after consent due to postponed or cancelled surgery, withdrawal, postoperative infection, intraoperative ketamine, further surgery, or other reasons such as logistic issues, change of anesthesia type (eg, local instead of general anesthesia), or no extubation immediately after surgery. Therefore, 180 patients (91 female patients [50.6%]) were included in the primary analysis. Baseline and clinical characteristics are summarized in Table 1. Intraoperative and postoperative data are shown in Table 2.

Figure 2. Flow Diagram of Study Population.

Flow diagram of participant screening, randomization, and analysis across three groups. Top center, a light gray rectangle labeled 892 Assessed for eligibility. A vertical line descends to a right-pointing arrow leading to a light gray rectangle at upper right labeled 667 Excluded, with two indented lines beneath: 303 Did not meet inclusion criteria and 192 Declined to participate. The main vertical line continues downward to a large light gray oval centered on the page labeled 225 Randomized. From this level, a short horizontal connector extends to the right to a light gray rectangle labeled 172 Excluded because of technical issues with device, with a small letter a at the end of device. The central vertical line continues to a second large light gray oval labeled 172 Additionally randomized. Three diagonal connectors extend from this oval to three side-by-side group boxes. Left group box: 68 Randomized to tACS intervention; next line 52 Received the allocated intervention; next line 16 Did not receive the allocated intervention with a small letter b. Middle group box: 83 Randomized to tDCS intervention; next line 63 Received the allocated intervention; next line 20 Did not receive the allocated intervention with a small letter c. Right group box: 74 Randomized to sham intervention; next line 65 Received the allocated intervention; next line 9 Did not receive the allocated intervention with a small letter d. Beneath each group box, a downward arrow leads to a follow-up box. Left follow-up box: 0 Discontinued intervention; 0 Lost to follow-up for primary outcome; 48 Postoperative follow-up in ward with parenthetical due to early discharge and a small letter e. Middle follow-up box: 0 Discontinued intervention; 0 Lost to follow-up for primary outcome; 61 Postoperative follow-up in ward with parenthetical due to early discharge and a small letter f. Right follow-up box: 0 Discontinued intervention; 0 Lost to follow-up for primary outcome; 62 Postoperative follow-up in ward with parenthetical due to early discharge and a small letter g. At the bottom, three final light gray rectangles align under the groups: left 52 Analyzed for primary outcome; middle 63 Analyzed for primary outcome; right 65 Analyzed for primary outcome.

Open in a new tab

tACS indicates α-transcranial alternating current stimulation; tDCS, transcranial direct current stimulation.

aPatients who did not receive proper stimulation due to technical issues with the stimulation device were categorized as a “historical study population” and excluded from the full analysis set (the modified full analysis set appears in Supplement 1).

bDue to postponed or cancelled surgery (n = 3), withdrawal of consent (n = 6), postoperative infection (n = 1), or other reasons (n = 6).

cDue to postponed or cancelled surgery (n = 2), withdrawal of consent (n = 4), intraoperative ketamine (n = 1), further surgery (n = 1), or other reasons (n = 12).

dDue to withdrawal of consent (n = 2) or other reasons (n = 6).

eSix patients at T4, 7 at T5, 8 at T6, and 9 at T7.

fNine patients at T4, 15 at T5, 10 at T6, and 8 at T7.

gNine patients at T4, 10 at T5, 13 at T6, and 12 at T7.

Table 1. Baseline Demographic and Clinical Characteristics.

Characteristic Group Full sample (N = 180)
tACS (n = 52) tDCS (n = 63) Sham (n = 65)
Age, mean (SD), y 73.0 (7.0) 73.0 (5.8) 73.0 (5.4) 73.0 (6.0)
Sex, No. (%)
Female 23 (44.2) 26 (41.3) 27 (41.5) 76 (42.2)
Male 29 (55.8) 37 (58.7) 38 (58.5) 104 (57.8)
BMI, mean (SD) 28.3 (5.6) 27.4 (4.8) 28.8 (5.8) 28.2 (5.4)
Educational level >12 y, No. (%) 16 (30.80) 28 (44.4) 16 (45.7) 68 (58.1)a
ASA classification, No. (%)
II 30 (57.7) 38 (60.3) 41 (63.1) 109 (60.6)
III 21 (40.4) 25 (39.7) 20 (30.8) 66 (36.7)
NRS score, median (IQR) 3 (0-5) 2 (0-5) 1 (0-4) 2 (0-5)b
MoCA score, median (IQR) 27 (24.5-29) 27 (23.5-28) 27 (25-29) 27 (24-28.8)c
Open in a new tab

Abbreviations: ASA, American Society of Anesthesiologist physical status classification; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NRS, Numerical Rating Scale for pain levels; tACS, transcranial alternating current stimulation; tDCS, transcranial direct current stimulation.

a

For 117 patients.

b

For 92 patients.

c

For 102 patients.

Table 2. Surgical and Anesthetic Characteristics.

Characteristic Group Full sample (N = 180)
tACS (n = 52) tDCS (n = 63) Sham (n = 65)
Duration of surgery, mean (SD), min 125 (92) 133 (95) 142 (111) 134 (100)
Duration of anesthesia, mean (SD), min 193.4 (108) 189.8 (111) 206.1 (130) 193.4 (117)
Induction dose, mean (SD)
Sufentanil, μg 22.8 (6.4) 22.6 (6.1) 22.0 (4.9) 22.4 (5.7)
Propofol, mg 152.3 (44.3) 154.8 (47.0) 143.0 (35.2) 151.1 (42.8)
Type of surgery, No. (%)
General surgery 19 (36.5) 29 (46.5) 28 (43.1) 76 (42.2)
Neurosurgery 17 (32.7) 18 (28.6) 16 (24.6) 51 (28.3)
Thoracic surgery 0 0 1 (1.5) 1 (0.6)
Urology 10 (19.2) 10 (15.9) 9 (13.8) 29 (16.1)
Gynecology 6 (11.5) 5 (7.9) 10 (15.4) 21 (11.7)
Vascular surgery 0 1 (1.6) 1 (1.5) 2 (1.1)
Open in a new tab

Abbreviations: tACS, transcranial alternating current stimulation; tDCS, transcranial direct current stimulation.

Primary Outcome Measure

For the primary outcome POD in the ward, the incidence rates were 11.5% in the tACS group (6 of 52), 14.3% in the tDCS group (9 of 63), and 9.2% in the sham group (6 of 65) (Figure 3A). Binary logistic models after multiple imputation of missing values revealed no substantial difference between tACS and sham groups (tACS vs sham: OR, 1.60 [95% CI, 0.47-5.46]; P = .45).

Figure 3. Bar Graphs of Postoperative Delirium (POD) and Line Graph of Duration of Surgery.

Three-panel figure: two P O D bar charts and a surgery-duration interaction line plot. Three panels arranged left to right and labeled A, B, and C in small boxed letters near the upper left of each panel. Panel A title at top: Ward. Vertical axis label: P O D, percent, with tick marks at 0, 10, 20, and 30. Horizontal axis label: Type of stimulation, with three categories: t A C S, t D C S, and Sham. Three teal vertical bars: t A C S approximately 11 percent, t D C S approximately 14 to 15 percent, and Sham approximately 9 percent. Panel B title at top: P A C U. Vertical axis label: P O D, percent, with tick marks at 0, 10, 20, and 30. Horizontal axis label: Type of stimulation, with categories t A C S, t D C S, and Sham. Three teal vertical bars: t A C S approximately 15 percent, t D C S approximately 21 to 22 percent, and Sham approximately 21 to 22 percent. Panel C title at top: Duration of surgery. Vertical axis label: P O D risk difference, percent, spanning from minus 90 at the bottom to 30 at the top, with a dashed horizontal reference line at 0. Horizontal axis label: Duration of surgery, h, with tick marks at 1, 2, 3, 4, and 5. Two lines with circular markers and shaded confidence bands: an orange series labeled t A C S in an inset legend near the lower left, and a dark teal series labeled t D C S. Orange markers: about plus 10 at 1 hour, about minus 5 at 2 hours, about minus 25 at 3 hours, about minus 45 at 4 hours, and about minus 60 at 5 hours; orange shaded band widens with duration, extending to roughly minus 85 near 5 hours. Dark teal markers: about plus 3 at 1 hour, about plus 3 at 2 hours, about minus 2 at 3 hours, about minus 10 at 4 hours, and about minus 15 at 5 hours; dark teal shaded band broadens upward and downward with duration, reaching roughly plus 25 at 5 hours and roughly minus 30 at 5 hours.

Open in a new tab

Incidence of POD (3D-CAM) in the ward (A) and PACU (B). Stimulation groups did not differ in POD in the ward (primary) or in PACU. C, Marginal effects of the duration-of-surgery × tACS interaction on POD incidence in PACU, derived from a multiple binary logistic regression model adjusted for age and sex (interaction term: odds ratio, 0.15 [95% CI, 0.02-0.97]; P = .046). Estimates are model-derived adjusted values on the probability scale, presented to illustrate the clinical magnitude of the interaction at different surgical durations. Shaded areas indicate 95% CIs. CAM indicates confusion assessment measure; PACU, postanesthesia care unit; POD, postoperative delirium; tACS, transcranial alternating current stimulation; and tDCS, transcranial direct current stimulation.

Secondary Outcome Measures

Binary logistic models showed no substantial difference between tDCS and sham groups for the incidence of POD in the ward (tDCS vs sham: OR, 1.51 [95% CI, 0.49-4.68]; P = .47). For POD in PACU, the incidence rates were 14.6% in the tACS group (7 of 48), 21.4% in the tDCS group (12 of 56), and 21.7% in the sham group (13 of 60) (Figure 3B). Binary logistic models after multiple imputation of missing values revealed no substantial difference between tACS or tDCS and the sham group (tACS vs sham: OR, 0.95 [95% CI, 0.38-2.36]; P = .91; tDCS vs sham: OR, 0.91 [95% CI, 0.36-2.30]; P = .83).

For POD severity, maximum scores in the ward did not differ between groups (mean [SD] score, 1.75 [1.68] for tACS, 1.45 [1.44] for tDCS, and 1.76 [1.83] for sham; regression estimates after multiple imputation of missing values and log-transformed data: tACS vs sham: ratio of geometric means, 1.03 [95% CI, 0.84-1.26]; P = .79; tDCS vs sham: ratio of geometric means, 0.92 [95% CI, 0.77-1.11]; P = .40). Similarly, POD severity in PACU did not differ between stimulation groups (mean [SD], 2.08 [2.01] for tACS, 2.33 [2.38] for tDCS, and 2.44 [2.66] for sham; tACS vs sham: ratio of geometric means, 0.95 [95% CI, 0.72-1.24]; P = .68; tDCS vs sham: ratio of geometric means, 1.01 [95% CI, 0.79-1.27]; P = .96) (eFigure 1A in Supplement 2). Similarly, groups did not differ in POD motor type (eFigure 1B and eTable 1 in Supplement 2) and in pain levels (eFigure 2 and eTable 2 in Supplement 2).

Prespecified Subgroup Analyses

Prespecified exploratory subgroup analyses revealed different modulation of POD incidence in PACU depending on duration of surgery (interaction term: OR, 0.15 [95% CI, 0.02-0.97]; P = .046), indicating that tACS may reduce POD incidence for patients with longer surgery. The significant interaction between tACS and duration of surgery indicated that the potential protective effect of tACS against PACU delirium increased with longer operations. Marginal estimates derived from the same model indicated that the absolute risk difference between tACS and sham was −26.6% (95% CI, −46.8% to −6.4%; P = .04) at a surgical duration of 3 hours, and −46.5% (95% CI, −73.0% to −19.9%; P = .006) at 4 hours, suggesting a potential surgical duration-dependent reduction in delirium risk after the intervention (Figure 3C). All other subgroup analyses did not yield substantial interactions. eTable 3 in Supplement 2 reports POD in ward and eTable 4 in Supplement 2 reports POD in PACU.

Adverse Events

No severe adverse events were reported. Nonsevere adverse events such as itching, transient pain, warmth, metallic taste, and fatigue were reported by 10 patients in the tACS group, 17 patients in the tDCS group, and 8 patients in the sham group. The incidence of adverse events did not differ substantially between groups (incidence rate ratio, 2.3 [95% CI, 0.8-6.5]) (eTable 5 in Supplement 2). After stimulation, patients were asked to guess their group assignment at randomization (eTable 6 in Supplement 2). The James blinding index29 was 0.338 (95% CI, 0.237-0.439) for the active intervention groups and 0.744 (95% CI, 0.659-0.829) for the sham control group, indicating successful blinding (eTable 7 in Supplement 2).

Discussion

This randomized clinical trial investigated whether tES applied postoperatively reduces the incidence of POD in older patients undergoing elective surgery. We found no evidence that a single session of α tACS over parietal cortices prevents POD in the ward. There were no group differences between active (tACS, tDCS) and sham groups in any of the secondary outcomes, either, including POD in ward, POD in PACU, POD severity and motor type, and self-reported pain intensity. Preregistered subgroup analyses provided preliminary evidence that tACS may reduce POD incidence for patients with longer surgery, although this exploratory finding requires confirmation in adequately powered trials.

Conceptualizing POD as a disorder that is substantially driven by acute network dysfunction provides the rationale for the application of network-targeted interventions through tES.18,30 As global α band dysconnectivity with a core hub within the posteromedial cortex has been reported as a key aspect of the network dysfunction,10 we chose to target posterior parietal regions with α tACS. However, the incidence of POD did not significantly differ between the tACS and the sham group, providing no evidence for the efficacy of this intervention. The responsiveness and general susceptibility to brain stimulation may considerably vary depending on the individual’s presence and extent of combinations of predisposing and precipitating factors. The effects of an intervention may specifically vary depending on factors such as head and brain anatomy,31 age, education, and baseline network functioning, among others.32,33 Further, residual effects of general anesthetics and analgesics may alter cortical network states during early postoperative recovery. These drug-induced network changes may reduce interventional effects targeting cortical plasticity and responsiveness to weak external electric fields.

In addition, the duration, extent, and type of surgery may impact the a priori probability to develop POD and thus the effects of an intervention.2 Prespecified subgroup analyses provided preliminary evidence that tACS may reduce POD incidence for patients with longer procedures. As longer surgeries are associated with greater systemic stress and more pronounced network vulnerability, they may offer a larger therapeutic window for tACS. This finding supports the hypothesis that tES effects depend critically on the degree of underlying network disruption and that benefit may be restricted to higher-risk subgroups. Extended surgical procedures may compromise neural network integrity, thereby increasing susceptibility to external oscillatory modulation.34 Here, targeted α-frequency tACS over parietal cortices may restore optimal network connectivity patterns and improve cognitive reserve, thereby potentially preventing POD manifestation.35,36,37 This potential dependency of tACS effects on surgery duration further supports the notion of patient- and situation-specific effects, suggesting that network support mechanisms may be more critical for preventing an episode of POD in certain contexts than in others. These findings highlight the need for further studies with targeted application strategies.

To evaluate the replicability of tDCS effects on POD incidence, we included a tDCS group, implementing anodal stimulation over the dorsolateral prefrontal cortex using parameters similar to those reported by Tao and colleagues.16 In contrast to Tao et al16 who demonstrated a substantial reduction in POD incidence (sham, 20%; anodal, 5%), our study did not replicate this protective effect on POD incidence. While the study by Tao et al16 focused exclusively on patients undergoing major lower limb arthroplasty, our study intentionally included diverse surgical procedures, encompassing general, vascular, thoracic, neurological, urological, and gynecological operations, to enhance external validity and generalizability. The diversity in surgical interventions may introduce varying degrees of tissue trauma and inflammatory responses, which may influence delirium risk through the release of damage-associated molecular patterns.38 More extensive tissue damage results in greater damage-associated molecular pattern release, subsequently triggering more pronounced inflammatory cascades that may contribute to POD development. In a second clinical trial conducted by the same group, Li and colleagues,17 who investigated patients undergoing laparoscopic colorectal cancer surgery exclusively, reported similar baseline delirium rates with approximately 25% incidence in the sham group vs 8% in the active stimulation group. Notably, their protocol incorporated 2 tDCS sessions, which may have enhanced therapeutic efficacy through cumulative neuromodulatory effects.14,39 This dosing difference represents another potential factor contributing to the discrepant outcomes between studies.

Notably, the POD incidence we observed reduced the sensitivity of our study to detect intervention effects on POD incidence significantly; the statistical power to detect a POD reduction of the extent as shown in a previous study16 was at only 46%, and even stronger effect sizes leading to only 0 or 1 POD case in the tACS group would have been detected at a statistical power below 80%. Future research should account for the substantial variability of POD incidence reported in the literature and observed in our trial, in which the sham-group incidence (9.2%) differed by approximately 13 percentage points from the rate anticipated on the basis of an earlier study at the same institution (22.2%2). Sample size calculations for subsequent trials should therefore plan conservatively for this uncertainty, for example, by powering for the lower end of plausible baseline incidence ranges or by using adaptive designs that allow re-estimation after an internal pilot. Future studies should also explore alternative temporal windows of stimulation (preoperative, intraemergence, or during manifest delirium),40 higher-risk surgical populations, and individualized stimulation parameters guided by neuroimaging or electrophysiological biomarkers.33,41

In sum, our results underscore the need for future investigations that stratify patients by relevant clinical variables, including surgical duration, patient age, and surgical type. Additionally, identifying individual biomarkers or clinical estimates of tES responsiveness could facilitate the development of personalized neuromodulation protocols for POD prevention. Such targeted approaches may optimize therapeutic outcomes while minimizing unnecessary interventions in patients unlikely to benefit from tES treatment.

Limitations

Several limitations warrant consideration when interpreting our findings. As a single-center investigation, our results may not fully capture the diversity of clinical practices and patient populations across different health care settings. Our study encompassed a heterogeneous spectrum of surgical interventions, introducing substantial variability in postoperative trauma severity and associated delirium risk. Lower-trauma surgical procedures may result in reduced pain levels and inflammatory responses, potentially creating floor effects that mask therapeutic benefits. However, in contrast to the more restrictive surgical cohorts examined by Tao et al16 and Li et al,17 this heterogeneity may enhance the generalizability and external validity of our findings. The assessment of tES effects relied primarily on clinical and cognitive scoring systems, which may lack sufficient sensitivity to detect subtle neuromodulatory changes. Finally, the observed POD incidence was substantially lower than anticipated for our sample size calculations, resulting in a relatively small cohort of patients with delirium. Although this reduction reflects improvements in perioperative care, this limitation may have reduced our statistical power to detect clinically meaningful differences between treatment groups, particularly in subgroup analyses.

Conclusions

In this randomized clinical trial of tES for POD, postoperative single-session tES, administered as either α-frequency tACS or anodal tDCS, did not reduce the incidence of POD in older adults undergoing elective major surgery. Against an incidence already substantially lower than anticipated in the sham group, no clinically meaningful effect was detectable.

Supplement 1.

Trial Protocol and Statistical Analysis Plan

jamanetwopen-e2628284-s001.pdf (200.7KB, pdf)
Supplement 2.

eAppendix

eFigure 1. Severity (A) and motor type (B) of POD. POD, post-operative delirium

eFigure 2. Pain levels for pain in rest (A), in motion (B), and in perception (C). tACS, transcranial alternating current stimulation

eTable 1. Number and proportion of missing values for the primary outcome, secondary outcomes, and reported covariates, overall and by randomization arm (full analysis set, N=180).

eTable 2. Group comparisons (results from statistical models) for POD motor type in PACU (note, for POD motor type in ward no statistical comparison was performed as nearly all values were 0).

eTable 3. Group comparisons (results from statistical models) for pain levels at rest in ward and in PACU

eTable 4. Subgroup analyses (model interactions) for POD in the ward

eTable 5. Subgroup analyses (model interactions) for POD in the PACU

eTable 6. Self-reported incidence of adverse events

eTable 7. Number of patients by group and assignment guess

eReferences

jamanetwopen-e2628284-s002.pdf (884KB, pdf)
Supplement 3.

Data Sharing Statement

jamanetwopen-e2628284-s003.pdf (15.4KB, pdf)

References

  • 1.Gou RY, Hshieh TT, Marcantonio ER, et al. ; SAGES Study Group . One-year Medicare costs associated with delirium in older patients undergoing major elective surgery. JAMA Surg. 2021;156(5):430-442. doi: 10.1001/jamasurg.2020.7260 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Müller J, Nowak S, Weidemeier M, et al. Duration of surgery and intraoperative blood pressure management are modifiable risk factors for postoperative neurocognitive disorders after spine surgery: results of the prospective CONFESS study. Spine (Phila Pa 1976). 2023;48(16):1127-1137. doi: 10.1097/BRS.0000000000004722 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Mahanna-Gabrielli E, Schenning KJ, Eriksson LI, et al. State of the clinical science of perioperative brain health: report from the American Society of Anesthesiologists Brain Health Initiative Summit 2018. Br J Anaesth. 2019;123(4):464-478. doi: 10.1016/j.bja.2019.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Kunicki ZJ, Ngo LH, Marcantonio ER, et al. Six-year cognitive trajectory in older adults following major surgery and delirium. JAMA Intern Med. 2023;183(5):442-450. doi: 10.1001/jamainternmed.2023.0144 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Association AP. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association; 2013. doi: 10.1176/appi.books.9780890425596 [DOI] [Google Scholar]
  • 6.Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories after postoperative delirium. N Engl J Med. 2012;367(1):30-39. doi: 10.1056/NEJMoa1112923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Deeken F, Sánchez A, Rapp MA, et al. ; PAWEL Study Group . Outcomes of a delirium prevention program in older persons after elective surgery: a stepped-wedge cluster randomized clinical trial. JAMA Surg. 2022;157(2):e216370. doi: 10.1001/jamasurg.2021.6370 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wilson JE, Mart MF, Cunningham C, et al. Delirium. Nat Rev Dis Primers. 2020;6(1):90. doi: 10.1038/s41572-020-00223-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Safavynia SA, Arora S, Pryor KO, García PS. An update on postoperative delirium: clinical features, neuropathogenesis, and perioperative management. Curr Anesthesiol Rep. 2018;8(3):252-262. doi: 10.1007/s40140-018-0282-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Fleischmann R, Traenkner S, Kraft A, Schmidt S, Schreiber SJ, Brandt SA. Delirium is associated with frequency band specific dysconnectivity in intrinsic connectivity networks: preliminary evidence from a large retrospective pilot case-control study. Pilot Feasibility Stud. 2019;5:2. doi: 10.1186/s40814-018-0388-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Burton JK, Craig L, Yong SQ, et al. Non-pharmacological interventions for preventing delirium in hospitalised non-ICU patients. Cochrane Database Syst Rev. 2021;11(11):CD013307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Burry L, Hutton B, Williamson DR, et al. Pharmacological interventions for the treatment of delirium in critically ill adults. Cochrane Database Syst Rev. 2019;9(9):CD011749. doi: 10.1002/14651858.CD011749.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Lefaucheur JP, Antal A, Ayache SS, et al. Evidence-based guidelines on the therapeutic use of transcranial direct current stimulation (tDCS). Clin Neurophysiol. 2017;128(1):56-92. doi: 10.1016/j.clinph.2016.10.087 [DOI] [PubMed] [Google Scholar]
  • 14.Polanía R, Nitsche MA, Ruff CC. Studying and modifying brain function with non-invasive brain stimulation. Nat Neurosci. 2018;21(2):174-187. doi: 10.1038/s41593-017-0054-4 [DOI] [PubMed] [Google Scholar]
  • 15.Antonenko D, Fromm AE, Thams F, Grittner U, Meinzer M, Flöel A. Microstructural and functional plasticity following repeated brain stimulation during cognitive training in older adults. Nat Commun. 2023;14(1):3184. doi: 10.1038/s41467-023-38910-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Tao M, Zhang S, Han Y, et al. Efficacy of transcranial direct current stimulation on postoperative delirium in elderly patients undergoing lower limb major arthroplasty: A randomized controlled trial. Brain Stimul. 2023;16(1):88-96. doi: 10.1016/j.brs.2023.01.839 [DOI] [PubMed] [Google Scholar]
  • 17.Li C, Tao M, Chen D, et al. Transcranial direct current stimulation for anxiety during laparoscopic colorectal cancer surgery: a randomized clinical trial. JAMA Netw Open. 2024;7(4):e246589. doi: 10.1001/jamanetworkopen.2024.6589 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zhou C, Gao YN, Qiao Q, et al. Efficacy of repetitive transcranial magnetic stimulation in preventing postoperative delirium in elderly patients undergoing major abdominal surgery: A randomized controlled trial. Brain Stimul. 2025;18(1):52-60. doi: 10.1016/j.brs.2024.12.1475 [DOI] [PubMed] [Google Scholar]
  • 19.Schulz KF, Altman DG, Moher D; CONSORT Group . CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c332. doi: 10.1136/bmj.c332 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Leroy S, Bublitz V, Grittner U, Fleischmann R, von Dincklage F, Antonenko D. Modulating delirium through stimulation (MoDeSt): study protocol for a randomized, double-blind, sham-controlled trial assessing the effect of postoperative transcranial electrical stimulation on delirium incidence. Trials. 2025;26(1):4. doi: 10.1186/s13063-024-08699-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Antal A, Alekseichuk I, Bikson M, et al. Low intensity transcranial electric stimulation: safety, ethical, legal regulatory and application guidelines. Clin Neurophysiol. 2017;128(9):1774-1809. doi: 10.1016/j.clinph.2017.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Leroy S, Bublitz V, von Dincklage F, Antonenko D, Fleischmann R. Normative characterization of age-related periodic and aperiodic activity in resting-state real-world clinical EEG recordings. Front Aging Neurosci. 2025;17:1540040. doi: 10.3389/fnagi.2025.1540040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Nasreddine ZS, Phillips NA, Bédirian V, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695-699. doi: 10.1111/j.1532-5415.2005.53221.x [DOI] [PubMed] [Google Scholar]
  • 24.Palihnich K, Gallagher J, Inouye S, Marcantonio E. The 3D CAM Training Manual for Research, Version 4.1. Hospital Elder Life Program Boston; 2016. [Google Scholar]
  • 25.Olbert M, Eckert S, Mörgeli R, Marcantonio E, Spies C. 3D-CAM: Delir-Testinstrument für deutschsprachigen Raum übersetzt. Anasthesiol Intensivmed Notfallmed Schmerzther. 2018;53(11-12):793-796. [DOI] [PubMed] [Google Scholar]
  • 26.Vasunilashorn SM, Guess J, Ngo L, et al. Derivation and validation of a severity scoring method for the 3-minute diagnostic interview for confusion assessment method–defined delirium. J Am Geriatr Soc. 2016;64(8):1684-1689. doi: 10.1111/jgs.14234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166(10):1338-1344. doi: 10.1164/rccm.2107138 [DOI] [PubMed] [Google Scholar]
  • 28.Hartrick CT, Kovan JP, Shapiro S. The numeric rating scale for clinical pain measurement: a ratio measure? Pain Pract. 2003;3(4):310-316. doi: 10.1111/j.1530-7085.2003.03034.x [DOI] [PubMed] [Google Scholar]
  • 29.James KE, Bloch DA, Lee KK, Kraemer HC, Fuller RK. An index for assessing blindness in a multi-centre clinical trial: disulfiram for alcohol cessation—a VA cooperative study. Stat Med. 1996;15(13):1421-1434. doi: 10.1002/(SICI)1097-0258(19960715)15:13<1421::AID-SIM266>3.0.CO;2-H [DOI] [PubMed] [Google Scholar]
  • 30.Antonenko D, Leroy S, Müller J, et al. Large-scale network mechanisms underlying postoperative cognitive improvement after spine surgery. Int J Surg. 2025;111(2):2317-2320. doi: 10.1097/JS9.0000000000002210 [DOI] [PubMed] [Google Scholar]
  • 31.Antonenko D, Grittner U, Saturnino G, Nierhaus T, Thielscher A, Flöel A. Inter-individual and age-dependent variability in simulated electric fields induced by conventional transcranial electrical stimulation. Neuroimage. 2021;224:117413. doi: 10.1016/j.neuroimage.2020.117413 [DOI] [PubMed] [Google Scholar]
  • 32.Fromm AE, Trujillo-Llano C, Grittner U, Meinzer M, Flöel A, Antonenko D. Increased variability in response to transcranial direct current stimulation in healthy older compared to young adults: A systematic review and meta-analysis. Brain Stimul. 2025;18(4):1257-1265. doi: 10.1016/j.brs.2025.06.005 [DOI] [PubMed] [Google Scholar]
  • 33.Miniussi C, Bortoletto M. Harnessing neural variability: implications for brain research and non-invasive brain stimulation. Neurosci Biobehav Rev. 2025;176:106312. doi: 10.1016/j.neubiorev.2025.106312 [DOI] [PubMed] [Google Scholar]
  • 34.Belrose JC, Noppens RR. Anesthesiology and cognitive impairment: a narrative review of current clinical literature. BMC Anesthesiol. 2019;19(1):241. doi: 10.1186/s12871-019-0903-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Fröhlich F, Sellers KK, Cordle AL. Targeting the neurophysiology of cognitive systems with transcranial alternating current stimulation. Expert Rev Neurother. 2015;15(2):145-167. doi: 10.1586/14737175.2015.992782 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Grover S, Nguyen JA, Reinhart RMG. Synchronizing brain rhythms to improve cognition. Annu Rev Med. 2021;72:29-43. doi: 10.1146/annurev-med-060619-022857 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Herrmann CS, Rach S, Neuling T, Strüber D. Transcranial alternating current stimulation: a review of the underlying mechanisms and modulation of cognitive processes. Front Hum Neurosci. 2013;7:279. doi: 10.3389/fnhum.2013.00279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Hovens IB, van Leeuwen BL, Mariani MA, Kraneveld AD, Schoemaker RG. Postoperative cognitive dysfunction and neuroinflammation; cardiac surgery and abdominal surgery are not the same. Brain Behav Immun. 2016;54:178-193. doi: 10.1016/j.bbi.2016.02.003 [DOI] [PubMed] [Google Scholar]
  • 39.Stephens JA, Berryhill ME. Older adults improve on everyday tasks after working memory training and neurostimulation. Brain Stimul. 2016;9(4):553-559. doi: 10.1016/j.brs.2016.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.van der A J, Lodema Y, Ottens TH, et al. Delirium treatment with transcranial electrical stimulation (DELTES): study protocol for a multicentre, randomised, double-blind, sham-controlled trial. BMJ Open. 2024;14(11):e092165. doi: 10.1136/bmjopen-2024-092165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Van Hoornweder S, Stagg CJ, Wischnewski M. Personalizing transcranial electrical stimulation. Trends Neurosci. 2025;48(9):663-678. doi: 10.1016/j.tins.2025.07.007 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

Trial Protocol and Statistical Analysis Plan

jamanetwopen-e2628284-s001.pdf (200.7KB, pdf)
Supplement 2.

eAppendix

eFigure 1. Severity (A) and motor type (B) of POD. POD, post-operative delirium

eFigure 2. Pain levels for pain in rest (A), in motion (B), and in perception (C). tACS, transcranial alternating current stimulation

eTable 1. Number and proportion of missing values for the primary outcome, secondary outcomes, and reported covariates, overall and by randomization arm (full analysis set, N=180).

eTable 2. Group comparisons (results from statistical models) for POD motor type in PACU (note, for POD motor type in ward no statistical comparison was performed as nearly all values were 0).

eTable 3. Group comparisons (results from statistical models) for pain levels at rest in ward and in PACU

eTable 4. Subgroup analyses (model interactions) for POD in the ward

eTable 5. Subgroup analyses (model interactions) for POD in the PACU

eTable 6. Self-reported incidence of adverse events

eTable 7. Number of patients by group and assignment guess

eReferences

jamanetwopen-e2628284-s002.pdf (884KB, pdf)
Supplement 3.

Data Sharing Statement

jamanetwopen-e2628284-s003.pdf (15.4KB, pdf)

Articles from JAMA Network Open are provided here courtesy of American Medical Association

RESOURCES