Neuronal Activity Topography Reveals Neuronal Change in Comatose Patients Treated by tDCS.

Chiaki Takahashi^1,2*, Takashi Shibata³, Mieko Tanaka⁴, Yuri Watanabe⁴, Yohei Kobayashi⁴, Isao Matsushita¹

¹Department of Rehabilitation Medicine, Kanazawa Medical University.

²Department of Rehabilitation Medicine, Toyama Prefectural Rehabilitation Hospital & Support Center for Children with Disabilities.

³Department of Neurosurgery, Toyama Nishi General Hospital.

⁴Brain Functions Laboratory, Inc.

---

Abstract

BACKGROUND: In our hospital, transcranial direct current stimulation (tDCS) is applied in combination with occupational therapy (OT) for patients with persistent consciousness disorders. There have been reports from some research groups on Left dorsolateral prefrontal area (F3) anodal stimulation as wells as Frontal pole (Fpz) anodal stimulation under different conditions. We chose either anodal stimulation site depending on the patient's condition.

CASES: Nine patients underwent tDCS with Fpz or F3 anodal stimulation for 10 days, followed by a 2-day break, after which the stimulation sites were switched, and the patients were stimulated for another 10 days. Their Electroencephalography changes were evaluated by using visual images of neuronal activity topography (NAT). The asymmetry index (AI) was also calculated to compare lateral differences in neuronal changes in the frontal lobe. Seven out of nine cases started with Fpz anode, and two with F3 anode. In the former setting, six cases improved, two aggravated and one remained unchanged. In the latter, one case unchanged and one remained aggravated. After applying tDCS, NAT and AI of the improved cases showed a tendency to approximate the normal control group. Discussion: Among these cases, all but one showed improvement in CRS-R scores, especially in Fpz stimulation. Additionally, changes in the scores indicated that most cases showed improvements in responses to visual and auditory stimulation. Furthermore, the NAT system allows for the rapid and minimally invasive assessment of changes in neuronal activity. Consequently, it is suggested that this system may be valuable for evaluating the therapeutic effects of tDCS and identifying effective stimulation sites.

---

Introduction

The efficacy of transcranial direct current stimulation (tDCS) treatment for various diseases such as hemiparesis in cerebral infarction sequelae, stiffness in Parkinson disease, depression and chronic pain have been reported^{1, 2}. During this treatment, a weak direct current of about 1-2 mA is passed between an anode and a cathode placed on the scalp, which is said to affect the membrane potential of the neurons directly under the dura mater. The cortex under the anode enhances excitability induced by depolarization and the cathode enhances repressibility induced by hyperpolarization from the cell membrane³. There are also individual differences with respect to this, and cases with opposite behavior has been reported⁴. However, an accurate and inexpensive physiological test to evaluate the effectiveness of tDCS has not been established. Currently, there are several articles reporting changes in consciousness level evaluated using batteries such as Coma Recovery Score-Revised (CRS-R)^{5, 6}, and changes in regional cerebral blood flow evaluated using near-infrared spectroscopy (NIRS)⁷.

We adopted tDCS with rehabilitation with the aim of improving persistent consciousness disorders of the patients admitted to the rehabilitation ward who are in subacute or chronic phase. To evaluate the effectiveness of tDCS, we applied Neuronal Activity Topography (NAT) which can assess changes in the excitability of neurons by recording the Electroencephalogram (EEG) at rest⁸. There were differences among the nine cases in both the effects and data obtained by NAT, and the representative case presentations and contributing factors are reported here.

Cases

Methods

The subjects were nine cases who were admitted to Toyama Prefectural Rehabilitation Hospital & Support Center for Children with Disabilities for stroke or traumatic brain injury (TBI) between2019 and 2020, assessed with CRS-R at the time of admission, and diagnosed with unresponsive wakefulness syndrome (UWS) or minimally conscious state (MCS). Cases with intracranial hematoma remaining, cases in which ventriculo-abdominal shunts had been performed, and cases with a history of epilepsy were excluded.

Device: DC-STIMULATOR PLUS (NeuroCare Group GmbH, Munich, Germany)

Target sites: (1) Anode: Frontal pole (Fpz), Cathode: Left motor cortex (C3) (Fig 1A)　(2) Anode: Left dorsolateral prefrontal area (F3), Cathode: right anterior frontal lobe (Fp2) (Fig 1B)

Stimulus intensity: 2mA with a 10 second ramp at the beginning and end of the stimulation period

Time duration: 20 minutes daily for 10 days (from Monday to Friday) at each setting

Electrodes: saline-soaked surface sponge electrodes (7 cm × 5 cm)

Figure 1A: A picture of the electrode placement. The left image shows the frontal view and the right image shows the top view. Red rectangle indicates the anode and blue one the cathode. Anode was placed on Fpz and cathode was on C3.

In all cases, tDCS was used in conjunction with occupational therapy (OT). During the stimulation, range of motion (ROM) exercises, listening to the patient's favorite music, and viewing photos of family and friends were conducted. The time of day that occupational therapy was given varied from day to day, and 20 to 60 minutes of exercise and swallowing therapy was given daily for all cases.

Seven cases started treatment with setting (1) and two cases with (2). Regarding the selection of stimulation sites, we initially choose Fpz anodal stimulation. However, if the evaluation of MRI or NAT visible images, or clinical symptoms such as spasticity, before stimulation indicated that F3 anodal stimulation was more appropriate based on the distribution of neural activity, we opted for F3 anodal.

After 10 days of stimulation, there was a rest period for two days. This was followed by another 10 days of stimulation during which the electrode placements were changed to the other setting.

We evaluated NAT in each patient before and after the first stimulation, and after 10 days and 20 days of stimulation. In addition, consciousness level was evaluated using CRS-R at the same time. The CRS-R is a scale specifically for the evaluation of chronic consciousness disorder. The lower the score, the more serious the consciousness disorder⁵.

We chose electrode locations based on previously reported research. Thibaut et al. reported that F3 stimulation for anode showed evidence of effectiveness⁶. On the other hand, experimental conditions of Fpz stimulation for anode reported by Naro et al⁹ was different than that of Thibaut et al. Thus, we chose either of those stimulation sites, depending on the location of the patient's brain damage and neurological symptoms.

The approval was obtained by Toyama Prefectural Rehabilitation Hospital & Support Center for Children with Disabilities Ethics committee to report these cases. (No.93 approved on March13, 2020)

Since the device DC-STIMULATOR PLUS is for off-label use, written informed consent was obtained from nine patients or family members before treatment. Furthermore, it includes permission to use anonymized personal data including MRI images and health records.

Information on NAT

The EEG represents the potential fluctuation induced by neuronal activity. NAT is a technology developed by Musha and colleagues, which enables the visualization of changes in neuronal activity calculated by standardizing and analyzing information obtained from EEG. Various factors such as age, medication, and brain damage due to stroke can cause changes in the intensity, amplitude, and frequency of local cerebral potentials. However, NAT can display this information with a high level of sensitivity.

The EEG recording used for NAT analysis is taken for five minutes while subjects are awake in a resting state with their eyes closed and is derived from 21 electrodes (international 10-20 system). EEG data were band-pass filtered to pick up signals in the 4-20 Hz frequency band, divided into 0.64s segments, and subjected to Fourier analysis. The calculated discrete Fourier coefficients are used to obtain the discrete power spectrum (PS), which consists of 11 frequency components (from 4.68 (3×1.56) Hz to 20.34 (13×1.56) Hz) in the signal channels from 21 electrodes. The normalized power spectrum (NPS) is obtained by standardizing this variable. This standardization can reduce individual differences in power spectra and allow for comparison among different cases. Furthermore, sNAT is obtained by removing offset values from NPS. The sNAT is a multi-dimensional marker and consists of 210 submarkers with reference to the ten frequency components on each of the 21 signal channels and they cover the θ, α and β frequency bands. Each submarker also has its own role in characterizing the cerebral neuronal activities. Moreover, the Z score is used to construct a visible image on NAT mappings. It shows distance from the average of the normal group. On the image, when compared to the normal group, it is presented in green if it is closer, with warm colors indicating hyperactivity, and cool colors indicating hypoactivity according to each frequency band⁸. The normal group consisted of 52 subjects (age 71.9 ± 5.9 years, 28 male and 24 female, MMSE 29.1 ± 1.1) who had normal cognitive function, MRI, and EEG based on the University of Tsukuba Epidemiological Survey in Tone Town, Ibaraki Prefecture¹⁰. To compare the level of consciousness and neuronal activity in normal group without therapeutic intervention to the temporal changes observed in the stimulated patient group, we refrained from administering tDCS to the normal group. Administering stimulation to the normal group could potentially introduce alterations or biases in neuronal activity, thereby complicating the assessment of their baseline state. Asymmetry Index (AI) value represents the balance between left and right brain activity and is easily calculated using the following equation¹¹. We selected F3 and the right dorsolateral prefrontal area (F4) as a region of interest.

AI (F3, F4) = NPS(F3)-NPS(F4)

We created line graphs with AI values on the vertical axis and frequency on the horizontal axis for the periods before stimulation, immediately after, 10 days after, 20 days after, and for the healthy control group. We then visually observed the changes in the graphs after stimulation.

Case Presentation

The overview of nine cases is presented. (Table 1) Six out of nine cases had stroke while three had brain injury. Out of the nine cases, we initially chose Fpz anodal stimulation for seven cases. However, for Case No1, the NAT visible images before stimulation showed hyperactivity of neurons in frontal lobe. To adjust this, we first selected F3 anodal stimulation, planning to place the cathode at that site. Regarding Case No5, decreased spontaneity and indifference of the patient due to a weakened emotional response suggested symptoms similar to apathy, so we expected that F3 anodal stimulation would improve the attentional function.

They were divided into three groups: improved, aggravated, and unchanged, based on the changes in CRS-R scores before and after tDCS. In the initial setting, six cases improved (all cases were Fpz for anode), two were aggravated (one case each for Fpz and F3 anode) and one remained unchanged (F3). Of the two aggravated cases, one improved after changing the anodal site.

The following are three representative cases that demonstrated different responses to tDCS treatment. (No.2, 3, 5 in Table 1).

Table 1: Overview of nine cases.

No	Disease	Side	Age	Duration from onset to start of tDCS (month)	Types of consciousness disorders	Change of CRS-R		Initial Anodal sight
1	SAH	R	72	5.5	UWS	6→6→6	F3
2	SAH	L	72	5	UWS	10→12→13	Fpz
3	SAH	Bil	28	2.5	UWS	2→1→8	Fpz
4	Cerebellar hemorrhage	L	87	3	MCS	12→15→22	Fpz
5	DAI		76	3	MCS	18→13	F3
6	TBI	R	54	3.5	MCS	21→23→23	Fpz
7	SAH	R	62	4	UWS	6→9→15	Fpz
8	TBI		83	0.6	MCS	17→21→22	Fpz
9	ICH	R	72	3	MCS	14→15→18	Fpz

Case 1: improvement

A 72-year-old woman suffered from subarachnoid hemorrhage with subcortical hematoma extending over a large area of the right hemisphere due to a ruptured right internal carotid artery aneurysm and was transferred to our hospital (Figure 2A). She had complete paralysis of the left side of her body, severe consciousness disturbance with tracheostomy, and a state of unresponsive wakefulness syndrome (UWS). At first, we could not communicate with her and did not observe her purposeful movement. In this case, stimulation of tDCS (Fpz for anode, C3 for cathode) was performed during OT rehabilitation while she was looking at photos of her family and with ROM training. Two weeks after the start of stimulation until discharge, she was able to open and close her fingers and could hold a photo frame according to instructions. The CRS-R score improved from 10 to 13 and AI showed a tendency to approximate the value of the normal group as the number of stimulations increased (Figure 2B). The visible images of sNAT in the alpha and theta band demonstrated a tendency towards bilateral symmetry after 2 weeks of stimulation (Figure 2C).

Figure 2A: MRI-FLAIR image of case 1. It shows widespread damage of the left hemisphere due to subarachnoid hemorrhage (SAH).

Figure 2B: The vertical axis shows AI and horizontal shows frequency of EEG. Error bars for the normal group indicate standard deviation (SD). As the number of stimulations increases, the AI tends to be closer to that of the normal group.

Figure 2C: The visible images of sNAT in case 1. The right column shows changes in the θ band and the left shows in the α band over time. The lateral differences are considered to be improving based on the color of the NAT image.

Case 2: aggravation

A 76-year-old man suffered from diffuse axonal injury due to a traffic accident and his MRI image only showed a contusion on the splenium of corpus callosum (Figure 3A). He could respond with only simple words such as “yes” and “hello”, and could not understand instructions, and required assistance to eat meals as well. In this case, tDCS (F3 for anode, right forehead for cathode) was performed while listening to music and looking at photos of his family with ROM training during OT rehabilitation. A week after starting the treatment, he became drowsy and was unable to eat, therefore we decided to stop the stimulation. Thereafter, his condition reverted within a few days. CRS-R score aggravated from 18 to 13 and AI showed a tendency to deviate greatly from the value of the normal group as the number of stimulations increased (Figure 3B). In the visible images of sNAT, no changes were observed in the alpha band, while there was an increase in the theta band wave that moved from the occipital lobe to the precuneus after two weeks of stimulation. (Figure 3C).

Figure 3A: MRI-FLAIR image of case 2. Arrow heads indicate the contusion area of corpus callosum.

Figure 3B: As the number of stimulations increases, the AI tends to diverge from that of the normal group.

Figure 3C: The visible images of sNAT in case 2. No marked changes in both bands on the image.

Case 3: improvement after electrode placement change

A 28-year-old man suffering from Subarachnoid Hemorrhage (SAH) due to a ruptured right internal carotid artery aneurysm with extensive damage of right hemisphere and infarction of left internal capsule was in a state of Minimally Conscious State (MCS) and his limbs showed tetraparesis (Figure 4A). Moreover, he had a tracheostomy. He could communicate “yes” and “no” by opening and closing his left hand only when his condition was well. In this case, tDCS (Fpz for anode, C3 for cathod) was performed while listening to music and looking at photos with ROM training during OT rehabilitation. Two weeks after starting the treatment, he became drowsy and did not open his eyes, and we observed increased sputum. Therefore, we changed the place of both electrodes (F3 for anode, right forehead for cathode). One week after the change, he gradually awakened, and communication improved. At first, CRS-R score aggravated from 2 to 1, but subsequently improved from 1 to 8. AI also showed an initial tendency to deviate significantly from the value of the normal group, which then gradually moved towards the normal group’s value (Figure 4B). The visible images of sNAT in the alpha band showed decreasing of neuronal activity on the occipital lobe, and the theta band also showed a significant decrease on the left temporal and frontal lobe after 2 weeks of stimulation. These findings improved after changing electrode locations. (Figure 4C).

Figure 4A: MRI-FLAIR image of case 3. It shows widespread damage of the right hemisphere and infarct of the left posterior limb of the internal capsule due to severe SAH.

Figure 4B: At first, AI shows divergence from that of normal group after stimulation commenced. It then tends towards that of the normal group after the stimulation area was changed.

Figure 4C: The visible images of sNAT in case 3. Decrease in neuronal activity on the occipital lobe in the alpha band and on the left temporal and frontal lobe in the theta band were observed after 2 weeks of stimulation. These findings improved after changing electrode locations.

Discussion

We treated nine cases using tDCS, and variations in effectiveness were divided into 3 patterns according to the change of CRS-R score: improved, aggravated, and unchanged. At present, some cases show the tendency of improving with Fpz for anode and one with F3, therefore it is not conclusive as to which is more effective. After administrating tDCS, the cases described improving tendency showed better balance between the left and right hemispheres on NAT imaging and AI, on the other hand, the cases described aggravating tendency showed worse balance. The single unchanged case did not show any tendency.

Hsu et al. reported that one-third of their cases showed inhibitory effects under the anodal side⁴, suggesting that case 2 may be applicable to this phenomenon. It is well known that precuneus blood flow reduction is observed in Alzheimer disease¹², and it is possible that in case 2, increase of theta wave in this area caused transient advanced Alzheimer's-like symptoms to appear. With regard to the changes in level of consciousness in case 3, it is possible that the left hemisphere which was less damaged and therefore predominantly active with respect to arousal, was suppressed by the cathode placed at C3. Subsequently, we consider that the change in electrode location released the inhibition to the cortex of the left hemisphere and enabled its return to the original state. On the other hand, improved cases tended to approximate the values of the normal group regarding AI. While there are likely to be other effects, such as problems with the damaged area of the brain itself and damage to the neural connections, it is possible that the imbalance of neural activity between the left and right sides of the brain is a cause of the impaired consciousness. From these findings, we believe that careful consideration of electrode location is important. Therefore, it may be advisable to evaluate the visible image of NAT and AI after first stimulation of tDCS to check the balance of neuronal activity between the left and right hemispheres, in addition to adjustment of electrode placement. This would facilitate tailor-made stimulation for each patient and improve treatment efficiency.

Meanwhile, we also observed cases in which consciousness improved in a stepwise manner as treatment progressed. We believe that this was due to the cumulative effect of treatment. In particular, the cerebral cortex showed huge permittivity (relative permittivity 10E8), and we therefore considered that the repeated use of tDCS resulted in a cumulative effect¹³. An induction of neuronal plasticity based on Hebb’s rule¹⁴ is observed in the rehabilitation method such as repetitive facilitation exercises (RFEs)¹⁵ and Constraint-Induced movement (CI) therapy¹⁶. The same applies to tDCS, and we believe that rehabilitation that can promote neuroplasticity related to cognitive function can be considered under circumstances where electricity is stored in the brain for a few weeks. Furthermore, we expect that periodic tDCS at a time when the cortical store of electricity is expected to be exhausted would allow effectiveness to be maintained.

Although we set the washout period to two days, it is possible that this duration was insufficient for the excitability of neurons induced by the after-effects of tDCS to fully subside. Many authors adopted one to two weeks¹⁷, and some also reported that even a 10-day wash-out period may be insufficient as the effects of stimulation could persist¹⁸. The actual required duration has not been examined and remains unclear. For future consideration, it is necessary to observe changes in neural activity for a certain period after the stimulation to determine the required duration.

This paper summarizes case reports on the use of tDCS for the treatment of consciousness disorders. Consequently, while specific stimulation sites and intensities were designated, there was no strict protocol in place. Variations existed among cases, including the duration between onset and initiation of treatment, types of disorders, lateral differences in lesions, which were not standardized. Additionally, the presence of various physical therapies and swallowing therapies, which were tailored to each case, as well as the non-uniform timing of the stimulation. These factors likely influenced the therapeutic outcomes and constitute a limitation of this study. For example, in cases where the period between onset and the start of treatment was short, the observed improvements in consciousness levels might have been due to the natural course of recovery.

We analyzed the EEG responses of patients with consciousness disorder treated with tDCS and found some trends in improved cases.

Based on the trends observed in the nine cases reported here, we would like to plan future clinical studies to find answers to the questions which of the two electrode locations is more effective, and to identify the specific types of cases that benefit the most from each position. Furthermore, by using the NAT system, it may be possible to rapidly and minimally invasively evaluate changes in neuronal activity in response to tDCS. By using this system, it is possible to quickly detect cases where stimulation acts inhibitively even under anodal stimulation of tDCS, and adjusting the stimulation site early may lead to better therapeutic outcomes. We aim to advance verification of this simultaneously.

Acknowledgements: We would like to thank the occupational therapists and clinical laboratory technicians of Toyama Prefectural Rehabilitation Hospital & Support Center for Children with Disabilities. We thank Genius Plus for English proofreading.

Conflicts of Interest Disclosure: The authors declare no conflict of interest concerning this manuscript.

Funding: The authors report no funding.

References:

1. Begemann MJ, Brand BA, Ćurčić BB, et al. Efficacy of non-invasive brain stimulation on cognitive functioning in brain disorders: a meta-analysis. Phychol Med 2020; 50: 2465-2486.
2. Antal A, Alekseichuk I, Bikson M, et al. Low intensity transcranial electric stimulation: Safety, ethical, legal regulatory and application guidelines. Clin Neurophysiol 2017; 128: 1774-1809.
3. Yamaguchi T, Moriya K, Tanabe S, et al. Transcranial direct-current stimulation combined with attention increases cortical excitability and improves motor learning in healthy volunteers. J Neuroeng Rehabil 2020. doi: 10.1186/s12984-020-00665-7.
4. Hsu TY, J CH, Tseng P. Individual Differenced and State-Dependent Responses in Transcranial Direct Current Stimulation. Front Hum Neuro. 2016. doi: 10.3389/fnhum.2016.00643.
5. Giacino JT, Kalmar K, Whyte J. The JFK Coma Recovery Scale–Revised: Measurement Characteristics and Diagnostic Utility. Arch Phys Med Rehabil 2004; 85: 2020-2029.
6. Thibaut A, Bruno MA, Ledoux D, et al. tDCS in patients with disorders of consciousness: sham controlled randomized double-blind study. Neurology 2014; 82: 1112-1118.
7. Ronak P, Aleksander D, Ara D, et al. Systematic review of combined functional near-infrared spectroscopy and transcranial direct-current stimulation studies. 2020. doi: 10.1117/1. NPh.7.2.020901.
8. Musha T, Matsuzaki H, Kobayashi Y, et al. EEG markers for characterizing anomalous activities of cerebral neurons in NAT (Neuronal Activity Topography) method. IEEE TRANS BIOMED ENG 2013; 60: 2332-2338.
9. Naro A, Calabro RS, Russo M, et al. Can trans cranial direct current stimulation be useful in differentiating unresponsive wakefulness syndrome from minimally conscious state patients? Restor Neurol Neurosci 2015; 33: 159-176.
10. Ikejima C, Hisanaga A, Meguro K, et al. Multicentre population-based dementia prevalence survey in Japan: a preliminary report. PSYCHOGERIATRICS 2012; 12: 120-123.
11. Shibata T, Musha T, Kosugi Y, et al. Boundary EEG Asymmetry Is Associated to Linguistic Competence in Vascular Cognitive Impairments. Front Hum Neurosci 2018. doi: 10.3389/fnhum.2018.00170.
12. Minoshima S, Foster NL, Kuhl DE. Posterior cingulate cortex in Alzheimerʼs disease. Lancet 1994; 344: 895.
13. Miyakawa H, Aonishi T. Apparent extracellular current density and extracellular space: basis for the current source density analysis in neural tissue. arXiv, q-bio.NC 2012;1209:4722.
14. Hebb DO. The Organization of Behavior. In A Neuropsychological Theory. Wiley, New York, 1949; 335.
15. Kawahira K, Shimodozono M, Etoh S, et al. Effects of intensive repetition of a new facilitation technique on motor functional recovery of the hemiplegic upper limb and hand. Brain inj 2010; 24: 1202-13.
16. Taub E, Miller NE, Novack TA, et al. Technique to improve chronic motor deficit after stroke. Arch Phys Med Rehabil 1993; 74: 347-54.
17. Spieser L, Wildenberg W, Hasbroucq T, et al. Controlling Your Impulses: Electrical Stimulation of the Human Supplementary Motor Complex Prevents Impulsive Errors. J Neurosci 2015; 18: 3010-3015.
18. Cavinatoa M, Gennab C, Formaggioa E, et al. Behavioural and electrophysiological effects of tDCS to prefrontal cortex in patients with disorders of consciousness. Clin NeuroPhysiol 2019; 130: 231-238.