Transcutaneous Auricular Vagus Nerve Stimulation for Recovery of Consciousness After Surgery for Spontaneous Intracerebral Hemorrhage

This study uses a core randomized controlled trial with nested multimodal mechanistic substudies. The primary clinical aim is to determine whether early taVNS improves recovery of consciousness after surgical treatment of spontaneous intracerebral hemorrhage (sICH). The mechanistic aims are to characterize treatment-related changes in residual arousal networks, cerebral perfusion and metabolism, neuroinflammation and neural injury, electrophysiological function, and autonomic regulation; to test whether these changes mediate clinical recovery; and to identify baseline phenotypes associated with treatment response.

Eligible participants will be randomized in a 1:1 ratio using a minimization procedure on postoperative day 5 (±1 day). The active group will receive left-ear taVNS at 25 Hz, a pulse width of 500 microseconds, and an intensity of 0.8-1.5 mA, using alternating 30-second on and 30-second off cycles. Stimulation will continue for 8 hours once daily for 6 weeks. The sham group will wear the same device with the electrode placed at the same left-ear location for 8 hours once daily for 6 weeks, but no current will be delivered. Other standard postoperative and rehabilitation care will be provided according to local clinical practice.

Consciousness will be assessed with the Coma Recovery Scale-Revised (CRS-R) at baseline and on postoperative days 28 and 90. The postoperative day-90 assessment will be performed by an independent blinded assessment team. CRS-R diagnostic categories will be used to classify participants as being in unresponsive wakefulness syndrome/vegetative state (UWS/VS), minimally conscious state minus (MCS-), minimally conscious state plus (MCS+), or emergence from the minimally conscious state (eMCS)/conscious.

Mechanistic assessments will be organized around a common schedule, with modality-specific windows and eligibility criteria. Reference time points are the pre-stimulation baseline on postoperative day 5 (±1 day), an early-treatment window, postoperative day 28, the end of the 6-week treatment period, and postoperative day 90. Not every participant is required to complete every mechanistic modality. Participation depends on clinical stability, contraindications, site capability, sample availability, and separate consent where required.

The neuroimaging platform may include noncontrast computed tomography to quantify hematoma location and volume, perihematomal edema, intraventricular extension, hydrocephalus, midline shift, residual hematoma, and secondary injury; computed tomography angiography when clinically obtained to characterize vascular features; and computed tomography perfusion to measure cerebral blood flow, cerebral blood volume, mean transit time, and time-to-maximum in the central thalamus, brainstem and basal forebrain arousal nodes, and distributed cortical networks. Magnetic resonance imaging may include T1- and T2-weighted imaging, fluid-attenuated inversion recovery, diffusion-weighted imaging, susceptibility-weighted imaging, diffusion tensor imaging, arterial spin labeling, and resting-state functional magnetic resonance imaging. Prespecified imaging features may include structural preservation of arousal-network nodes and pathways, white-matter integrity, regional perfusion, functional connectivity, network segregation and integration, and longitudinal change after treatment.

The electrophysiology and autonomic platform may include routine or high-density electroencephalography, spectral power, background organization, reactivity, complexity, functional connectivity, sleep-wake organization, and event-related responses where feasible. Electrocardiography-derived heart-rate variability, blood-pressure variability, and other prespecified autonomic indices may be used to assess engagement of vagal and central autonomic pathways.

The biospecimen platform may include serial blood collection for inflammatory, immune, endothelial, coagulation, metabolic, and neural-injury biomarker domains. Candidate measures may include blood-cell and routine biochemical indices, C-reactive protein, cytokines and chemokines, neurofilament light chain, glial fibrillary acidic protein, S100 calcium-binding protein B, neuron-specific enolase, and other prespecified markers documented in the laboratory analysis plan. Cerebrospinal fluid may be collected only when available from clinically indicated external ventricular drainage, lumbar drainage, or another clinically necessary procedure, unless an additional research-only procedure has received specific ethics approval and separate informed consent. Paired blood and cerebrospinal-fluid analyses may be used to characterize central and systemic inflammatory or neural-injury responses.

Mechanistic analyses will evaluate: (1) the effect of taVNS on longitudinal biomarker and network changes; (2) associations between these changes and CRS-R improvement or emergence from the minimally conscious state; (3) mediation pathways linking taVNS, target engagement, and recovery of consciousness; and (4) treatment-effect heterogeneity according to baseline structural preservation, perfusion, functional connectivity, electrophysiological patterns, autonomic function, and biomarker profiles. These analyses are exploratory unless otherwise prespecified in the statistical analysis plan.

Sample Size Calculation: The sample size was calculated based on the proportion of participants who achieve emergence from the minimally conscious state (eMCS) or regain a higher level of consciousness by postoperative day 90. Previous cohorts of patients with intracerebral hemorrhage or prolonged disorders of consciousness reported that approximately 59%-67% achieved eMCS during follow-up. Because the present trial enrolls a more severely affected population with a Glasgow Coma Scale score of 5-9 on postoperative day 5, the eMCS rate in the sham-control group was conservatively assumed to be 55%. The corresponding eMCS rate was assumed to be 80% in the active taVNS group, representing an absolute difference of 25 percentage points. With a two-sided alpha level of 0.05, 90% statistical power, and a 1:1 allocation ratio, 72 evaluable participants are required per group. Allowing for approximately 10% loss to follow-up, withdrawal, or missing primary outcome data, the final planned enrollment is 160 participants, with 80 participants in each group.

One formal interim analysis is planned for March 2027, provided that approximately 80 participants, representing 50% of the planned information size, have completed the 90-day primary-outcome assessment. If this information threshold has not been reached by March 2027, only a blinded review of recruitment, protocol adherence, data completeness, and an appropriate safety review will be conducted, and the formal interim efficacy analysis will be postponed until the prespecified information threshold is reached. The interim analysis will be performed by an independent statistician and reviewed by an independent Data Monitoring Committee. Interim assessments will address efficacy, safety, and futility. A Lan-DeMets alpha-spending function approximating O'Brien-Fleming boundaries will be used to preserve the overall two-sided type I error rate. A stringent stopping boundary will be applied for overwhelming efficacy, while a non-binding conditional-power boundary will be used for futility. Investigators, outcome assessors, and the sponsor's operational study team will remain blinded to comparative interim results. Based on the prespecified Data Monitoring Committee charter, the committee may recommend continuation, early termination for overwhelming efficacy or safety concerns, or discontinuation for futility.