Effect of Adjunctive Transcranial Direct Current Stimulation and Cognitive Behavior Therapy on Headache Disability in Episodic Frequent or Chronic Tension-Type Headache

A Pilot, Exploratory Study

Abstract

Background: 

Literature suggests that non-pharmacological treatments like non-invasive brain stimulation and cognitive behavioural therapy (CBT) may be used as augments in the management of Tension-Type Headache (TTH).

Aims and Objectives: 

To assess the efficacy of Transcranial Direct Current Stimulation (tDCS) and Cognitive Behavioral Therapy (CBT) compared to treatment as usual (TAU) on headache-related disability and the psychological impact of headache on life of patients suffering from TTH.

Materials and Methods: 

Thirty right-handed TTH patients HDI (headache disability index) > 30 were randomly allocated to group A (tDCS + TAU), group B (CBT + TAU), and group C (TAU). While the primary outcome measure was headache related disability (Headache Disability Index(HDI)), somatic symptoms (Somatic Symptom Severity (SSS-8)), anxiety & depression (Hospital Anxiety and Depression Scale (HADS)) and sleep quality (Pittsburgh Sleep Quality Index (PSQI). Assessments were done at baseline, end of the third and sixth month.

Results: 

We found a statistically significant reduction in HDI (F=12.419, p <0.001), SSS-8 (F=4.587, p=0.010), and HADS-anxiety (F=4.810, p=0.002). tDCS showed better efficacy than CBT, while these two were significantly better than TAU on HDI and SSS-8. On HADS-A also both tDCS and CBT were better than TAU, but showed no significant difference between them.

Conclusion: 

Supplementing non-pharmacological techniques like tDCS and CBT will be effective in reducing headache-related disability and associated psychological symptoms.

INTRODUCTION

Tension-type headache (TTH) is the most common headache in the population with high social impact leading to significant loss of effective working days. Due to its high lifetime prevalence in the population, episodic frequent or chronic TTH causes a high degree of disability.[¹] Therefore, effective treatment and prevention of relapses become important in patients of episodic frequent or chronic TTH. Studies done on the pattern of medical treatment in TTH suggests the overuse of analgesics and frequent use of over-the-counter medication with frequent relapses after withdrawal of the drug.[²] At the current time, the most effective nonmedical treatment used for patients with TTH is electromyography biofeedback and relaxation technique as per the available old studies.[³] Recently, other forms of nonpharmacological treatment such as cognitive behavior therapy (CBT) and neurostimulation techniques were also reported to have a positive impact on the management of chronic TTH, especially in those who do not prefer, are nonresponsive, or poorly tolerate medical treatment like pregnant women.[⁴⁵⁶] Although both pharmacological prophylaxis and nonpharmacological prophylaxis are recommended, the drug and nondrug trials suffer from a lack of controlled conditions and a lack of comparable efficacy measures.[⁵] Therefore, the necessity emanates from novel treatment strategies.

Transcranial direct current stimulation (tDCS) has recently been used nonpharmacological treatment modality in the last decade for the headache that can noninvasively modulate specific brain regions.[⁶] Its usage has been mostly studied in chronic migraine but not in episodic frequent or chronic TTH. Furthermore, there is a lack of head-to-head clinical trials within nonpharmacological therapies. Hence, citing the above reasons, we intend to study the comparison of add-on preventive effects of tDCS and CBT to treatment as usual (TAU) among episodic frequent or chronic TTH patients in a randomized controlled design. The ensuing objective of our study is to assess the concomitant effects on associated somatic symptom severity, sleep quality, depression, and anxiety.

SUBJECTS AND METHODS

Subjects

Right-handed patients with more than 18 years of age and minimum education up to 8^th grade were enrolled. Patients meeting the diagnostic criteria for episodic frequent or chronic TTH (TTH) as per the International Classification of Headache Disorders III were included.[⁷] Subjects obtaining a score of >30 (mild headache disability) on the Headache Disability Index (HDI) were included.[⁸] Exclusion criteria included any significant neurological (significant head injury or epilepsy) or medical illness, diagnosis of mental and behavioral disorder due to the use of psychoactive substances harmful use or dependence (except nicotine and caffeine), cardiac instrumentation, and shift work disorder. The study is registered in the Clinical Trials Registry India (No: CTRI/2021/09/036943). The study was conducted from October 2021 to March 2022.

Patients were randomly allocated to tDCS, CBT, and TAU groups. A total of 55 patients were screened and 35 were enrolled for treatment (the rest 20 were having one or more exclusion criteria). Written and signed informed consent was taken from the patient and caregivers before enrollment. Out of these, five patients dropped out of the study, two in tDCS, two in CBT, and one in the TAU group. Thirty patients completed the study. All 35 patients were included in the final analysis [Figure 1]. The current trial has followed the CONSORT statement and checklist.[⁹]

Figure 1: CONSORT diagram

All secondary causes (e.g., refraction error) were ruled out by appropriate history and clinical evaluation and wherever feasible neuroimaging. Any other investigation was pursued per the treating physician's clinical discretion. The follow-up assessments were done at the end of the 3^rd and 6^th months from the baseline. In all the three groups, the patient was under TAU as per clinician discretion.

Intervention

Subjects in all three arms were allowed to continue on their pharmacological treatment either acute or prophylactic (TAU) in a naturalistic manner. Subjects in Group A were delivered tDCS and were continued on TAU. Group B subjects received CBT sessions along with TAU, whereas Group C subjects were continued solely on TAU.

Transcranial direct current stimulation (Group A)

Brain driver TM v2.1 tDCS device was used for brain stimulation. The standardized international 10–20 EEG system was used for positioning the tDCS small electrodes (3.5 cm²) which deliver low levels of constant electric current. Anode was placed in the center of the frontal pole of the subdominant hemisphere (Fp2); a cathode was placed two cm higher than the ipsilateral mastoid process.[¹⁰] Thrice a week single session per day was delivered for 12 weeks (a total of 36 sessions). In a single session, 2 mA constant current for 20 min was given.[¹¹]

Cognitive behavior therapy (Group B)

Pain-specific CBT sessions were delivered. CBT sessions followed a structured manual and covered psycho-education, exercising modules including specific coping strategies for pain and stress, and goal-setting skills. CBT session was given once a week for 12 weeks, each session lasted for 30 min. Twelve sessions were divided into three phases: the initial phase (1–3 sessions), the middle phase (4–9 sessions), and the terminal phase (10–12 sessions). The initial phase did include psycho-education about the relationship between pain and bio-psychological factors, the influence of psychological factors (poor sleep quality and anxious or depressed affect) on chronic headache, the bio-psycho-social model of pain, and fear and avoidance model, and the effects of conditioning (and expectations). The middle phase included analysis of the situations that induced headache or stress and identifying dysfunctional thinking patterns. This phase also addressed inappropriate beliefs about headache and maintaining coping strategies into constructive cognition. Introduction to Progressive Muscle relaxation technique and conducted under the supervision of a therapist which was also given as homework. In the termination phase, focus was placed on goal setting and desired behavior to reduce headache. Booster sessions were planned at the end of the 3^rd and 6^th months to review implemented goals.[¹²]

Clinical measures

In this study, headache impact was rated on HDI. HDI is the 25-item beta version questionnaire used to identify the impact of headache and its treatment items are subgrouped into functional and emotional subscales.[⁸] Sleep quality was measured by the global score of the Pittsburgh Sleep Quality Index (PSQI).[¹³] The somatic symptoms burden was evaluated by the 8 item-Somatic Symptom Severity (SSS-8). SSS-8 is a brief self-report questionnaire.[¹⁴] Concomitant anxiety and depression were assessed by the Hospital Anxiety and Depression Scale (HADS).[¹⁵] HADS comprises seven questions for anxiety and depression each. Scores <7 for each anxiety and depression indicate noncases.[¹⁶] Subjects were rated on clinical scales at baseline and at the end of the 3^rd and 6^th months.

Blinding procedure

Subjects were rated by an independent rater who was blind to the randomization. The patients were sequentially and randomly assigned to the three groups using blocks of random number sequences. The numbers were written in sealed opaque envelopes. For each patient, the envelope was opened immediately before the commencement of the first session by the clinician administering the TTH. The patients were also blind to the arm of treatment.

Statistical analysis

Appropriate statistical analysis is conducted by using the IBM SPSS for Windows, version 28 (IBM Corp., Armonk, New York, USA). The primary outcome variables were HDI scores. Secondary outcome variables were SSS-8, HADS, and PSQI scores. The normal probability plots, as well as the Kolmogorov–Smirnov test, were used to verify the assumption of normality. The differences in sample characteristics within the two groups were examined using one-way ANOVA and Chi-square test (wherever applicable). We conducted an intention-to-treat analysis and included all the patients enrolled in the study, despite whether they completed the study or not. The missing values were replaced by the method of last observation carried forward. The primary analysis (mixed model ANOVA) was the effect of interaction over some time (baseline, at the end of the 3^rd month, and the end of the 6^th month) within the three groups in the double-blind phase trial. The between-subject factor is the treatment arm (Group A vs. B vs. C) and the within-subject factor is time, i.e., pretreatment at the end of the 3^rd month and the end of the 6^th month. We also conducted a supplementary per-protocol analysis to include only those patients who completed the study.

RESULTS

Sample characteristics

No significant difference was found in the sample characteristics among the three groups. Gender, family type, habitat, education status, presence of psychosocial stressors, past psychiatric and medical illnesses, family psychiatric and medical illnesses, and sleep cycle duration were comparable except for marital status (X² (6, 35) =10.165, P = 0.041). The three groups were comparable in the duration of illness (F = 0.514, P = 0.603) but not in age (F = 3.714, P = 0.035) [Table 1].

Table 1: Comparison of the sociodemographic profile and clinical variables as the intention-to-treat analysis across three groups (n=35)

Safety and side effects

None of the subjects reported any major adverse effects with tDCS. Two patients complained of headaches during the few initial sessions. One patient complained of itching at the local site. No patient discontinued the study due to the side effects.

Outcome measures

At the baseline all clinical measures, i.e., HDI (F = 0.141, P = 0.869), SSS-8 (F = 0.279, P = 0.759, HADS-depression (F = 1.453, P = 0.249), HADS-anxiety (F = 0.445, P = 0.645), and PSQI (F = 2.568, P = 0.092) were comparable among the three groups [Table 2].

Table 2: Interaction effect of intervention between the three arms (Groups A, B, and C) (intention-to-treat analysis) across baseline, end of the 3^rd month, and end of the 6^th month using mixed model analysis (n=35)

Headache Disability Index

At baseline, mean values of HDI were 71.17 ± 11.900 in Group A (tDCS), 68.08 ± 4.228 in Group B (CBT), and 69.73 ± 16.402 in Group C (TAU). At the end of the 3^rd month, HDI mean values were 50.42 ± 15.042 in Group A (tDCS), 56.50 ± 14.010 in Group B (CBT), and 67.09 ± 15.496 in Group C (TAU). At the end of the 6^th month, HDI mean values were 25.50 ± 23.27 in Group A (tDCS), 44.83 ± 18.862 in Group B (CBT), and 66.45 ± 12.421 in Group C (TAU). Patients subjected to tDCS (Group A) showed a statistically significant reduction in HDI at the end of 6 months (F = 12.419, P < 0.001) as compared to CBT and TAU.

Somatic Symptom Severity (SSS-8)

At baseline, the mean values of SSS-8were 16.58 ± 8.404 in Group A (tDCS), 14.83 ± 4.73 in Group B (CBT), and 15.09 ± 4.44 in Group C (TAU). At the end of the 3^rd month, SSS-8 mean values were 11.92 ± 5.680 in Group A (tDCS), 12.83 ± 3.762 in Group B (CBT), and 13.73 ± 3.04 in Group C (TAU). At the end of the 6^th month, SSS-8 mean values were 10.42 ± 3.895 in Group A (tDCS), 10.42 ± 3.895 in Group B (CBT), and 12.64 ± 3.641 in Group C (TAU). Patients subjected to tDCS (Group A) showed a statistically significant reduction in somatic severity at the end of 6 months (F = 4.587, P = 0.010) as compared to CBT and TAU.

Hospital Anxiety Depression Scale (subscale depression)

At baseline, mean values of HADS-D were 9.08 ± 3.728 in Group A (tDCS), 9.25 ± 3.82 in Group B (CBT), and 11.55 ± 3.984 in Group C (TAU). At the end of the 3^rd month, HADS-D mean values were 8.25 ± 3.696 in Group A (tDCS), 7.50 ± 3.261 in Group B (CBT), and 10.73 ± 2.970 in Group C (TAU). At the end of the 6^th month, HADS-D mean values were 3.83 ± 3.22 in Group A (tDCS), 6.25 ± 3.108 in Group B (CBT), and 8.73 ± 3.663 in Group C (TAU). Patients subjected to tDCS (Group A), CBT (Group B), and TAU (Group C) overall did not show any statistically significant reduction. In comparison to CBT and TAU, patients in tDCS (Group A) showed a marginal reduction in depressive features (F = 2.606, P = 0.057) at the end of 6 months.

Hospital Anxiety Depression Scale (subscale anxiety)

At baseline, mean values of HADS-A were 9.08 ± 4.055 in Group A (tDCS), 9.92 ± 2.43 in Group B (CBT), and 10.36 ± 3.264 in Group C (TAU). At the end of the 3^rd month, HADS-A mean values were 8.75 ± 3.494 in Group A (tDCS), 7.75 ± 2.050 in Group B (CBT), and 10.09 ± 2.844 in Group C (TAU). At the end of the 6^th month, HADS-A mean values were 4.25 ± 3.194 in Group A (tDCS), 6.25 ± 2.14 in Group B (CBT), and 9.00 ± 2.864 in Group C (TAU). Patients subjected to tDCS (Group A) showed a statistically significant reduction in anxiety features at the end of 6 months (F = 4.810, P = 0.002) as compared to CBT and TAU.

Pittsburgh Sleep Quality Index

At baseline, mean values of PSQI were 11.00 ± 6.33 in Group A (tDCS), 7.08 ± 3.12 in Group B (CBT), and 10.55 ± 3.643 in Group C (TAU). At the end of the 3^rd month, PSQI mean values were 9.08 ± 5.230 in Group A (tDCS), 7.50 ± 3.261 in Group B (CBT), and 10.18 ± 3.093 in Group C (TAU). At the end of the 6^th month, PSQI mean values were 6.42 ± 5.501 in Group A (tDCS), 4.92 ± 2.712 in Group B (CBT), and 8.73 ± 3.47 in Group C (TAU). Patients subjected to tDCS (Group A), CBT (Group B), and TAU (Group C) not showed any statistically significant reduction in improving quality of sleep (F = 1.51, P = 0.224) at the end of 6 months.

Overall findings suggested that tDCS (Group A) showed a statistically significant difference in all clinical measures, except in PSQI, at the end of 6 months, as compared to CBT and TAU. Finding in a post hoc analysis, tDCS (Group A) had shown better effects than Group B (CBT) over headache disability and somatic severity.

DISCUSSION

Our study is one of the rare randomized controlled trials comparing the add-on preventive efficacy (long term) of tDCS and CBT to TAU for treating headache disability in episodic frequent or chronic TTH.

In this study, we had not only taken the HDI as an outcome parameter but also studied the somatization index, associated anxiety/depression, and sleep quality to have a complete idea about headache-related disability. It has been seen that many patients of TTH have no major disability due to the intensity of pain but have more problems due to associated illness anxiety and somatization.[¹⁷] Associated poor quality of sleep and masked depression further increase the problem in TTH patients. Therefore, addressing all the issues together will show a holistic response to treatment in TTH patients.

HDI was the primary outcome variable in our study to assess headache-related problems in TTH patients. Reduction in HDI was the maximum when tDCS was combined with medical treatment. It has been shown in previous studies that stimulation techniques reduce headache intensity, frequency, duration, and need for rescue medication.[¹⁰¹⁸] It is also evident that the effect of tDCS was continued even after discontinuation of medical treatment after 3 months which suggests that this technique also causes a reduction in relapses in TTH.

Patients with chronic headache commonly have associated illness anxiety and somatization due to fear of serious underlying illness.[¹⁹] It has also been concluded that timely assessment of somatization and evaluation of mental status can help to improve headache outcomes.[¹⁹]

In the current study, results showed that somatization index and anxiety were significantly lower in the tDCS group as compared to behavioral therapy and medical treatment only. Effect of depressive symptoms and quality of sleep was also better with tDCS although not statistically significant. These findings suggested that direct current stimulation can be a wonderful option for the comprehensive effects in all parameters associated with chronic headache.[¹⁹]

The tDCS protocol used in our study is similar to one previous study done on 86 patients with primary headache.[¹⁰] The recommended pattern of electrode placement by Pinchuk et al. was anode in the prefrontal (FP1) and cathode over the mastoid process.[¹⁰] The author reported 50% of improvement in the number of headache days, but the duration of stimulation was more prolonged, i.e., 30–45 min rather than the conventional 20 min used in our study. Concomitant improvement in the anxious-depressive profile could be due to the tDCS-induced right orbitofrontal inhibition and induced hemispheric frontal lobe activation.[¹⁰] Furthermore, tDCS had been shown to lower sympathetic activity in TTH patients.[¹⁹]

CBT, although not as good as tDCS, was better than medical treatment alone in terms of controlling HDI, somatic amplification, and anxiety in TTH. Similar to our study, other studies also found that CBT helps reduce headache intensity, frequency, and headache-free days.[²⁰²¹]

Treatment with medical treatment only is associated with poor control of comorbidity like somatic amplification and anxiety in patients with TTH. Furthermore, patients had a frequent relapse of headache symptoms once they are off medication. Therefore, supplementing medical treatment with tDCS is the best option for patients with TTH.

The long duration of follow-up (up to 6 months) in our study is a distinction. The current study suffers from some important limitations such as lack of other headache-associated outcomes (headache intensity, recording headache diary, and the use of pain medications), and lack of neuronavigation to target a particular site of the cortex.

CONCLUSION

The current study concluded that medical treatment alone is not sufficient for treating headache in long term. Reduction in HDI, associated anxiety, and somatic amplification was the maximum when tDCS was combined with medical treatment. Supplementing standard medical treatment along with tDCS or CBT is effective and holds a lot of potential in reducing headache-related disability, somatic severity, improving quality of sleep, and associated anxiety-depressive features. Future research can be done to utilize the potential of these techniques in chronic headache management.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.