Effect of transcranial direct current stimulation on paroxysmal sympathetic hyperexcitability with acquired brain injury and cortical excitability: a randomized, double-blind, sham-controlled pilot study

Abstract

Paroxysmal sympathetic hyperexcitation (PSH) refers to a clinical syndrome characterized by a sudden increase in sympathetic excitability caused by severe brain injury. This study aims to investigate the effectiveness and practicality of combining transcranial direct current stimulation (tDCS) with medication to treat PSH and employ non-linear electroencephalography (EEG) to assess changes in cortical activation post-intervention. 40 PSH patients were randomly assigned to receive either active tDCS or sham tDCS treatment over an 8-week period. The tDCS stimulation targeted the prefrontal area, left frontal-temporal-parietal cortex, right frontal-temporal-parietal cortex, and left dorsolateral prefrontal cortex. Both patient groups also underwent medication and other conventional therapies. The Paroxysmal Sympathetic Hyperactivity Assessment Measure (PSH-AM), Coma Recovery Scale-Revised (CRS-R), medication dosage, and approximate entropy (ApEn) index were assessed before and after treatment. The active tDCS group exhibited more substantial improvements in changes of PSH-AM, changes of CRS-R, and medication reduction ratios compared to the sham tDCS group after the treatment. After treatment and during follow-up, a significantly greater number of patients in the active tDCS group demonstrated clinically important differences compared to the sham tDCS group. The active tDCS group showed significantly higher ApEn indices in the less affected frontal lobe compared to the control group. No significant differences in ApEn indices were noted in the sham tDCS group before and after treatment. Regression analysis revealed that the group (active tDCS/sham tDCS) was the primary factor associated with improving PSH-AM. Therefore, we believe that in patients with PSH, combining tDCS with medication therapy demonstrated superior clinical efficacy compared to medication therapy alone. Electrophysiological results also indicated enhanced cortical excitability. Therefore, this single-center pilot study suggests that multi-target, multi-session tDCS combined with medication may be an effective treatment protocol for PSH.

Introduction

Paroxysmal sympathetic hyperexcitation (PSH) is a complication arising from acute diffuse or multifocal brain injuries, most commonly due to traumatic brain injury (TBI), hypoxia, or cerebrovascular accidents [1]. This clinical syndrome manifests as sudden spikes in sympathetic excitability and is characterized by abrupt episodes of increased heart rate, rapid breathing, elevated blood pressure, profuse sweating, and sometimes accompanied by fever and dystonic posturing [2]. These symptoms may occur spontaneously, but PSH symptoms are more likely to be provoked by non-painful stimuli, persisting over weeks or months [3]. During the episodes, PSH can lead to hyperventilation and hypocapnia, which in turn may cause cerebral vasoconstriction and hypoperfusion [4]. Patients with PSH experience extended hospital stays [5, 6], and an increased risk of complications like heterotopic ossification [7], severe weight loss [8], and infections [5, 6]. Timely and appropriate PSH treatment is crucial, and we should pay more attention to the treatment protocol for PSH.

The primary objectives of PSH treatment involve avoiding triggers, reducing excessive sympathetic activity, and providing support [3]. Presently, there is no standardized protocol for treating PSH. Existing pharmacological interventions consist of a combination of several medication classes, including benzodiazepines, opioids, non-selective beta-blockers, and muscle relaxants, which only offer symptomatic relief [9]. However, their therapeutic efficacy is limited. Uncontrolled sympathetic hyperactivity unresponsive to typical medications can lead to hyperthermia, cardiac complications, and fatalities [2]. Moreover, some medications (e.g., benzodiazepines and muscle relaxants) have central depressant effects that can cause sedation or drowsiness [10, 11]. Given that PSH is more prevalent in individuals with disorders of consciousness (DOC), long-term use of these drugs can negatively impact the recovery of consciousness in brain injury patients. Hence, the pursuit of effective and more favorable therapies for long-term patient outcomes is imperative.

Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation method that has been proven effective and safe in numerous neurological disorders. A 2013 systematic review summarized six studies demonstrating that tDCS can modulate autonomic activity in healthy individuals and those with major depressive disorder, suggesting its potential impact on PSH [12]. tDCS induces and enhances neuroplastic changes, facilitating the recovery of cortical and subcortical impairments. In our prior investigations, we have utilized tDCS to improve unresponsive wakefulness syndrome (UWS) [13], psychomotor inhibition state [14], swallowing apraxia [15] and apraxia of speech [16].

Electroencephalography (EEG) is a widely employed electrophysiological technique for real-time monitoring of brain function in clinical settings, offering the advantages of high temporal resolution, ease of use, and cost-effectiveness. Approximate entropy (ApEn), a nonlinear dynamics analysis (NDA) measure, can assess changes in cortical function by quantifying the regularity of the signal, providing insights into cortical excitability [17]. Previous research from our team has revealed that NDA can delineate alterations in brain function during unconscious states and potentially offer prognostic information for unconscious individuals [18].

Consequently, we postulated that tDCS could enhance the recovery of PSH by increasing cortical excitability and reinforcing inhibitory modulation through the descending pathway. This study aimed to investigate this hypothesis by randomly allocating subjects into experimental and control groups, administering active tDCS stimulation and sham stimulation, respectively. Other treatments included medications and conventional treatments. We assessed changes in clinical symptoms, medication reduction ratios before and after treatment, and the subjects’ ApEn to examine the effectiveness and alterations in cortical activation following tDCS.

Methods

Study design

This study was a prospective randomized controlled trial conducted at a single center, consisting of two groups: the active tDCS group and the sham tDCS group. The study obtained approval from the Ethics Committee of Wangjing Hospital, China Academy of Chinese Medicine Sciences (CACMS). The study design is depicted in Fig. 1.

Given that this is the pioneering clinical protocol to employ tDCS for the treatment of PSH, our study is designed as a pilot randomized controlled trial. The primary objective is to evaluate the feasibility and preliminary efficacy of the intervention. To achieve this, we have recruited a total of 40 participants, with 20 patients assigned to the experimental group receiving tDCS treatment and 20 patients assigned to the control group.

Fig. 1

Study design (tDCS: transcranial direct current stimulation; CRS-R: Coma Recovery Scale-Revised; PSH-AM: Paroxysmal Sympathetic Hyperactivity Assessment Measure; EEG: electroencephalography)

Participants

This study was carried out in the Department of Rehabilitation at CACMS. A total of 40 subjects met the inclusion criteria and participated in the study. The final group of patients analyzed comprised 19 individuals with TBI and 21 with stroke. The group consisted of 26 males and 14 females, ranging in age from 17 to 82 years, with disease durations ranging from 30 to 135 days (median, 89.50 days). All patients were right-handed, as determined by their spouses or guardians using the Edinburgh Handedness Inventory. Please refer to Fig. 2 for the Consort form.

The inclusion criteria were as follows: (1) All subjects had acquired brain injuries (such as TBI or stroke); (2) All subjects were diagnosed with UWS or minimally conscious state (MCS) according to the JFK Coma Recovery Scale-Revised (CRS-R) (CRS-R ≤ 21 points); (3) All subjects exhibited typical PSH cases (PSH-AM score ≥ 17 points). (4) The duration of the patient’s brain injury was between 1 and 6 months. (5) Brain lesions were diagnosed and localized prior to enrollment using computed tomography or magnetic resonance imaging.

The exclusion criteria were as follows: (1) Patients with DOC caused by primary brainstem injuries; (2) Presence of intracranial metal implants; (3) Local skin injury in the tDCS stimulation area; (4) Subjects with severe spasticity causing electromyographic artifacts, thereby affecting the recording of EEG. (5) Patients exhibiting functional communication or functional object use, qualifying as Emergence from Minimally Conscious State in the CRS-R assessment, are excluded.

Fig. 2

CONSORT 2010 flow diagram

Interventions

Transcranial direct current stimulation (tDCS)

We believe that the rationale of tDCS treatment for PSH is to promote the recovery of cortical and subcortical injuries. In our previous research, the efficacy of single target tDCS treatment for DOC was not satisfactory, we chose multi-target tDCS intervention for DOC [19]. Hence, we selected multi-target tDCS consistent with DOC treatment in this study, including the prefrontal cortex, left dorsolateral prefrontal cortex (DLPFC), left frontal-temporal-parietal cortex (FTPC) and right FTPC. Establishing of neural plasticity and functional reorganization requires a long time [20], seldom resulting in significant improvement over a short period. Therefore, in this study, we chose multi-session tDCS. A portable battery-driven device (IS200, Chengdu, China) delivered a constant direct current of 2.0 mA (0.056 mA/cm²) through a pair of surface sponge electrodes (5 cm × 7 cm) soaked in saline. The stimulation sites were determined based on our previous study [19] and the EEG international 10–20 system. The prefrontal area was located 3.5 cm above the FPz; the left DLPFC was situated at F3; and the left/right FTPC was positioned at the midpoints of C3-T3/C4-T4. The cathode electrode was placed over the neck (prefrontal area), F4 (right DLPFC), and the back of the opposite shoulder (bilateral FTPCs) (Fig. 3). The tDCS of all subjects was carried out in the order shown in Fig. 3, with a total of 8 weeks of intervention. The stimulation sites for each week are shown in Fig. 3. tDCS was administered twice daily for 20 min, five days a week.

The electrode size and placement for sham tDCS were identical to those of active tDCS. In the sham stimulation group, the stimulator was deactivated after 30 s to mimic the sensation of active current.

Fig. 3

Transcranial direct current stimulation electrode position (FTPC: frontal-temporal-parietal cortex; DLPFC: dorsolateral prefrontal cortex)

Medications and other treatments

We selected non-selective beta-blocker propranolol in combination with the benzodiazepine drug clonazepam due to their demonstrated efficacy in alleviating PSH symptoms. All subjects met the criteria for typical PSH cases (PSH-AM score ≥ 17 points) and were initially prescribed propranolol at 80 mg/day and clonazepam at 4 mg/day. The medication dosage is adjusted weekly based on the improvement of PSH symptoms, primarily the frequency and severity of PSH episodes. Additionally, personalized modifications are made according to individual patient responses to optimize symptom management. This approach ensures a balance between effective symptom control and the gradual discontinuation of medication, thereby enhancing the clinical applicability and reproducibility of the treatment regimen. The reduction ratio of medications before and after treatment served as a clinical efficacy indicator.

Both groups of patients underwent conventional treatments twice a day for 50 min following tDCS sessions. Conventional treatments encompassed standard bedside physical therapy, multimodal sensory stimulation, auditory stimulation, and environmental reinforcement therapy. Throughout the intervention, efforts were made to avoid triggers that might induce PSH attacks in both patient groups.

Blinding

In a 1:1 ratio, patients were randomly assigned to either active or sham tDCS stimulation. A randomization code was generated by an independent statistician using a randomization program and stored in opaque envelopes. Independent researchers recruited eligible subjects, and the envelopes were opened in the specified order. Patients were randomly allocated to either the experimental or control group. Specialized rehabilitation therapists administered tDCS treatment, while researchers evaluating the results remained unaware of the research hypothesis and group assignments. Therefore, the study designers, therapists, outcome assessors, subjects, and statisticians remained blinded to group allocation until the final statistical analyses were completed.

Clinical assessment

The primary outcome measure is the Paroxysmal Sympathetic Hyperactivity Assessment Measure (PSH-AM) score, comprising two components: the Diagnosis Likelihood Tool (DLT) for diagnostic probability assessment and the Clinical Feature Scale (CFS) for evaluating the severity of clinical features.

The diagnosis can be categorized as ‘unlikely’ (< 8 points), ‘possible’ (8–16 points), or ‘probable’ (≥ 17 points) by summing the CFS and DLT scores [2]. This scale is practical and easy to apply in clinical practice, serving as a valuable tool for PSH diagnosis [21]. A clinical important difference (CID) is defined as a change in diagnosis from ‘probable’ to ‘unlikely’.

Additional outcome measures include the CRS-R score and the reduction ratio of clonazepam and propranolol. The CRS-R scale, developed by JFK Medical Center, evaluates patients’ consciousness by assessing visual, auditory, motor, verbal functions, communication, and arousal [22]. The medication reduction ratio is calculated as follows: (initial dosage - dosage after treatment) / initial dosage.

EEG recording and nonlinear dynamics analysis

EEG signals were recorded using a wireless digital electroencephalogram system (ZN16E, Chengdu, China) under eyes-closed conditions and painful stimulation. Sixteen EEG electrodes were positioned according to the international 10–20 system, with earlobe electrodes serving as the initial reference. The EEG recording and artifact-free epoch selection methods followed previous studies [14, 19].Preprocessing of EEG data involved band-pass filtering within the range of 0.3–100 Hz. To minimize electrical noise, a 50-Hz notch filter was applied. Throughout the EEG recording period, participants remained awake in a quiet ward to ensure data reliability.

To optimize cortical activation, the Han’s acupoint nerve stimulator (HANS) simultaneously stimulated bilateral acupoints, including Quchi (LI11), Neiguan (PC6), Hegu (LI4), Waiguan (SJ5), Zusanli (ST36), Sanyinjiao (SP6), Taichong (LR3), and Yongquan (KI1). The 16-channel EEG recording was conducted concurrently, ensuring uniform current intensity across all acupoints.

Due to the varying affected sides of each subject’s brain injury, we employed descriptions of the more affected or less affected sides to replace the traditional EEG electrode names. The terms “more affected/less affected” refer to: (1) based on the patient’s imaging evidence (CT or MRI), the side that is more affected is termed “more affected”, and the side that is less affected is termed “less affected”. For instance, the side affected by a hematoma in TBI patients or by a stroke lesion can be identified through imaging evidence, and this side is referred to as “more affected”. (2) For patients who are difficult to distinguish by imaging evidence, we use EEG to distinguish between the two hemispheres, such as TBI patients with diffuse axonal injury or stroke patients with bilateral hemisphere injury. The hemisphere with a lower average ApEn index at the EEG site is called “more affected”, while the hemisphere with a higher average ApEn index is called “less affected”. Therefore, the names of the electrodes are referred to as FP_L, FP_M, F_L, F_M, AT_L (anterior temporal), AT_M, C_L, C_M, MT_L (middle temporal), MT_M, P_L, P_M, PT_L (posterior temporal), PT_M, O_L, and O_M.

ApEn assigns a non-negative value to a time series, where higher values indicate greater complexity or irregularity in the data. Increased irregularity signifies heightened time series complexity, reflecting greater non-linear cell dynamics or cortical network interaction (increased ApEn) [14]. ApEn serves as a measure of system complexity and was calculated to assess cortical excitability, with details of the algorithm described by Wu et al. [18].

In addition, to avoid immediate effects, EEG collection and clinical evaluation of all subjects after treatment were conducted 24 h after the last tDCS stimulation.

Statistical analysis

Data analysis was performed using SPSS version 26.0 (SPSS, USA). Baseline characteristics of the active tDCS and sham tDCS groups were compared using independent samples t-tests for continuous variables and the Chi-squared test for dichotomous variables. When the data displayed a normal distribution and homogeneity of variance, independent samples t-tests were utilized for between-group comparisons; otherwise, the Wilcoxon signed-rank test was employed. Changes in ApEn before and after treatment within the same group were assessed using paired sample t-tests. Additionally, the Chi-square test was used to compare the number of patients with a CID between the two groups. A linear regression model was established using the forward method. In the regression analysis, subjects’ clinical characteristics (age, gender, lesion, duration, and type) and group (active tDCS/sham tDCS) were treated as independent variables, while the improvement in PSH-AM was the dependent variable. Additional analyses examined if the effects of intervention were moderated by 2 subgroups: TBI or stroke (lesion). Statistical significance was indicated by p < 0.05. We applied the Hochberg procedure using R (version 4.4.1) to adjust the p-values for multiple comparisons in the statistical analysis of the ApEn difference. This method reduces the overall risk of false positives across multiple hypothesis tests, providing a more rigorous approach to evaluating the statistical significance of our findings. The adjusted p-values are reported in the results section.

Results

Table 1 summarizes the basic characteristics of the participants. There were no significant differences in age, gender, lesion type, duration, or CRS-R and PSH-AM scores between the active tDCS group and the sham tDCS group. No participants withdrew from the study. Only three participants in the active tDCS group displayed local skin redness beneath the tDCS electrode, and no other significant adverse events were observed.

Table 1 Demographic characteristics of the patients

Clinical assessment

Table 2 displays the differences in clinical assessments between the two groups after the treatment. Compared to the sham tDCS group, patients in the active tDCS group exhibited more significant improvements in PSH-AM, CRS-R scores, and greater reduction ratios for clonazepam and propranolol. Table 3 outlines the number of patients with PSH-AM score variations after the treatment and at follow-up. In comparison to the sham tDCS group, the active tDCS group had a significantly higher number of patients with CID after the treatment (100% vs. 0%) and at follow-up (100% vs. 40%).

Table 2 Clinical assessment comparison between the active and Sham tDCS groups after treatment

Table 3 PSH-AM assessment after tDCS and at follow-up

Non-linear EEG analysis

Table 4 presents the alterations in EEG ApEn between the two groups following the treatment, for both painful stimulus conditions and eyes-closed conditions. The ApEn values of F_L significantly increased after active tDCS treatment in comparison to the sham tDCS group.

The differences in EEG ApEn within each group before and after the treatment are illustrated in Table 5. In the active tDCS group, ApEn values were significantly higher at sites of FP_M, F_M, P_M, AT_M,PT_M, FP_L, F_L and AT_L after the treatment. In the sham tDCS group, there were no significant differences in ApEn at any site before and after the treatment.

Table 4 Changes in ApEn difference between painful stimulus conditions and eyes-closed conditions after treatment in the active tDCS group and the Sham tDCS group

Table 5 Changes in ApEn difference between painful stimulus conditions and eyes-closed conditions before and after treatment in the active tDCS group and Sham tDCS group

Linear regression analysis

Table 6 outlines the findings related to the improvement of PSH-AM scores. Linear regression analysis revealed a correlation between the group (active tDCS) and the improvement in PSH-AM.

Follow-up

Figures 4 and 5, and Fig. 6 respectively show the intergroup differences of PSH-AM, clonazepam dosage, and propranolol dosage between the active tDCS group and the sham tDCS group before treatment, after treatment, and at follow-up. Figure 4 shows that there are significant differences in PSH-AM score between the two groups after treatment (p < 0.001) and at follow-up (p < 0.001). Figure 5 shows that there were significant differences in clonazepam dosage between the two groups after treatment (p < 0.001) and at follow-up (p < 0.001). Figure 6 shows that there were significant differences in propranolol dosage between the two groups after treatment (p < 0.001) and at follow-up (p < 0.001). In Fig. 6, the propranolol dosage in the active tDCS group was recorded as 0 after treatment and during follow-up, as propranolol had been discontinued at these time points.

Table 6 Factors related to PSH-AM improvement based on multivariate linear regression analysis

Fig. 4

Differences in PSH-AM between active tDCS group and sham tDCS group before treatment, after treatment and at follow-up. **p < 0.01 (PSH-AM: paroxysmal sympathetic hyperactivity assessment measure)

Fig. 5

Differences in dosage of clonazepam between active tDCS group and sham tDCS group before treatment, after treatment and at follow-up. **p < 0.01

Fig. 6

Differences in dosage of propranolol between active tDCS group and sham tDCS group before treatment, after treatment and at follow-up. **p < 0.01

Subgroup analysis

Figure 7A illustrates the mean differences (MDs) in PSH-AM scores between active tDCS and sham tDCS at various time points. 1)Pre: There is no significant difference for either subgroups or all patients, as the MDs include 0. 2)Post: A notable negative MDs indicates that active tDCS is more effective than sham tDCS in reducing PSH-AM scores after treatment. The MDs in all subgroups and all patients are entirely below 0, signifying statistical significance. 3)Follow-up: The results indicate that active tDCS is superior to sham tDCS, with MDs also entirely below 0. It’s noteworthy that higher PSH-AM scores indicate more pronounced symptoms of PSH, hence the labels “Active Better” and “Sham Better” at the bottom of the graph are reversed compared to other charts. Figure 7B depicts the MDs in CRS-R scores between active and sham tDCS. 1)Pre: There is no significant difference for either subgroups or all patients, as the MDs include 0. 2)Post: The MD of TBI patients includes 0, indicating no significant difference in CRS-R scores between the two groups of TBI patients. However, for stroke and all patients, active tDCS is more effective than sham tDCS in improving CRS-R scores. The MDs of all subgroups and all patients are positive and do not cross 0, indicating that active tDCS is associated with a statistically significant greater reduction ratio of clonazepam compared to sham tDCS (Fig. 7C). The MD for all subgroups and all patients do not cross 0, signifying statistical significance about reduction ratio of propranolol for these groups (Fig. 7D).

Fig. 7

Sub-analyses for the PSH-AM, CRS-R, reduction ratio of clonazepam and reduction ratio of propranolol, Overall and by Lesion (traumatic brain injury or stroke). (TBI: traumatic brain injury)

Discussion

This study introduces a novel randomized controlled protocol employing multi-target and multi-session tDCS for patients with PSH following brain injury. The results revealed that active tDCS treatment, as compared to sham tDCS, led to a more pronounced improvement in PSH symptoms and a reduction in medication usage among the subjects. The active tDCS group exhibited a significant increase in CID patients during treatment and follow-up when compared to the sham tDCS group. Since we did not set up subgroups in advance, based on the results of subgroup analysis, we believe that the significant differences in the therapeutic effect of active tDCS on PSH patients between different subgroups (stroke or TBI) require further research. Future research should focus on identifying factors that influence the responsiveness to tDCS and exploring the long-term implications of these treatment outcomes. This could lead to more targeted and effective use of tDCS in clinical settings, optimizing therapeutic benefits for patients with stroke and TBI. Furthermore, after treatment, the ApEn index of less affected frontal lobe in the active tDCS group was significantly higher than in the sham tDCS group. Some ApEn indices in the active tDCS group significantly increased after treatment, while no significant differences were observed in the sham tDCS group. Regression analysis highlighted the critical role of the group (active tDCS) in improving PSH-AM.

The follow-up results demonstrate that the PSH symptoms of the active tDCS group patients improved more significantly than those of the sham tDCS group patients after treatment and during follow-up, and the dosage of clonazepam and propranolol in the active tDCS group patients were also significantly lower than that in the sham tDCS group patients.

The effects of medications and tDCS on clinical symptoms in PSH patients

Currently, there is no standardized medication treatment protocol for patients with PSH, and no single medication effectively treats all PSH symptoms. As a result, most patients require the use of multiple potentially complementary medications for treatment [3]. Propranolol, owing to its lipophilicity, is capable of penetrating the blood-brain barrier and addressing tachycardia, hypertension, sweating, reducing metabolic rate, and potentially improving dystonia [23, 24]. Clonazepam has a notable impact on agitation, hypertension, tachycardia, and dystonia [25]. Hence, we chose a combination of these two medications as the therapy protocol in our study.

PSH is predominantly observed in individuals with DOC. PSH arises from severe acquired brain injury, resulting in damage to the inhibitory pathways, causing an exaggerated response by the sympathetic nervous system to internal and external stimuli [4]. tDCS has the potential to aid in the recovery of cortical and subcortical injuries by regulating membrane potential and synaptic plasticity [26]. We posit that tDCS serves as a top-down therapeutic approach for the etiology of PSH. In contrast, medications aim to alleviate PSH symptoms rather than addressing its root cause. Moreover, commonly used medications like benzodiazepines and muscle relaxants have central inhibitory effects, and their prolonged use has adverse effects on improving patient consciousness. Hence, we hypothesized that combining tDCS with medications could provide comprehensive relief for PSH symptoms while simultaneously enhancing patient awareness.

According to the findings of this study, the combination of tDCS and medications more effectively and rapidly alleviates PSH symptoms in subjects and also contributes to a certain degree of improvement in their awareness. Additionally, benzodiazepines and non-selective beta-blockers exhibit superior reduction rates and speed. However, medication therapy can only provide partial relief for PSH symptoms in subjects, with a slower rate of improvement and less satisfactory results, and the use of benzodiazepines in the sham tDCS group is of longer duration. These factors not only heighten the risk of complications and prolong hospitalization but also do not favor the enhancement of patient awareness, potentially impacting long-term patient outcomes negatively. Regression analysis results further indicate a significant correlation between the relief of PSH symptoms and tDCS treatment, with the active tDCS group demonstrating more pronounced relief of PSH symptoms.

The dose-response relationship of tDCS is a critical factor in determining its therapeutic efficacy. Parameters such as stimulation intensity, session frequency, electrode placement, and stimulation duration may influence clinical outcomes. In this study, we selected a current intensity of 2.0 mA, an 8-week intervention period, and a twice-daily, 20-minute stimulation regimen based on prior research and clinical experience in DOC treatment [13, 14, 19]. Our findings suggest that this multi-target, multi-session approach yields superior symptom relief and improved awareness compared to sham tDCS.

However, variations in tDCS dosing could yield different clinical outcomes. Previous studies have demonstrated a non-linear relationship between tDCS dose and its effects, with an optimal dosing range required for maximal efficacy [27, 28]. Future studies should explore whether increasing session frequency, extending treatment duration, or adjusting electrode configurations could enhance the clinical effectiveness of tDCS in PSH patients. Additionally, individualized dosing strategies based on patient responsiveness should be investigated to optimize outcomes.

The electrophysiological effects of tDCS and possible mechanisms for improving PSH

From an electrophysiological perspective, this study utilized ApEn indices to assess the cortical excitability of subjects. After treatment, in comparison to the sham tDCS group, the active tDCS group exhibited cortical activation in the less affected frontal lobe under painful stimulation. Moreover, in the active tDCS group, an extensive increase in cortical activation was observed post-treatment as compared to pre-treatment, while no such activation was evident in the sham tDCS group. Consequently, the more notable and rapid improvement in PSH symptoms in the active tDCS group may be attributed to the broad enhancement of cortical excitability.

Sympathetic nervous responses involve various cortical and subcortical structures, and diffuse or multifocal brain injury is more likely to precipitate PSH [3]. According to the excitability-to-inhibitability ratio model, in PSH patients, multiple cortical and subcortical structures lose control over brainstem regulation, leading to the disruption of inhibitory drive originating from the brainstem nucleus for the spinal cord reflex arc. This disruption ultimately affects the balance between inhibitory and excitatory interneurons of the motor and sympathetic nerves [29, 30]. But there are also theories that suggest that the potential mechanism of PSH is diffuse or focal brain injury, which leads to the disconnection of the inhibitory center of the cerebral cortex from the hypothalamic, diencephalic, and brainstem centers [31].In our study, patients in the active tDCS group displayed substantial enhancement in cortical excitability, particularly in the less affected cerebral cortex. tDCS facilitates the recovery of cortical and subcortical injuries, thus promoting the rehabilitation of pathways controlling sympathetic nerves. Therefore, we believe that the widespread enhancement of cortical excitability may be the fundamental reason for the more pronounced and expeditious improvement in PSH symptoms in patients treated with active tDCS.

In a prior study of ours, we used tDCS with the same stimulation sites to treat patients with DOC. We observed an enhancement of the prefrontal-parietal and temporo-parietal associative cortical networks in the less affected hemisphere, which correlated with improved patient consciousness [19]. This suggests that the cortical network associated with PSH may overlap with the network linked to consciousness. In our current study, while PSH symptoms improved, there was notable cortical activation in the less affected frontal lobe. Consequently, the prefrontal cortex of the less affected hemisphere may be common regions in both PSH-related cortical networks and awareness-related cortical networks.

Limitations

Firstly, the EEG nonlinear analysis in this study solely assessed the ApEn index and could only analyze local cortex excitability. Future studies should incorporate cross-approximate entropy analysis to explore the connections between cortical networks. Secondly, the follow-up in this study was relatively simple, involving only the PSH-AM score and subjects’ medication dosages, lacking objective assessments like EEG. The study’s duration is relatively short, without the evaluation of patients’ long-term consciousness prognosis. In addition, we designed an 8-week intervention protocol and did not evaluate it during the treatment process, making it impossible to confirm how early the treatment effect occurred. Therefore, future research can further explore this issue. What’s more, as a pilot study, this research has inherent limitations, including a small sample size and a single-center design, which may restrict the generalizability of the findings. Future large-scale, multi-center studies are needed to validate these results and enhance their external validity.

Conclusion

Multi-target and multi-session tDCS, in combination with medications, offer a more effective and faster means of improving PSH symptoms, along with the rapid reduction of medications affecting patient consciousness recovery. This improvement may be linked to the extensive enhancement of cerebral cortex excitability in patients. Consequently, the combination of tDCS and medications could serve as an effective treatment approach for PSH patients.