comparative study on electromyography (emg) biofeedback...

I

“Comparative study on Electromyography (EMG)

biofeedback and Galvanic Skin Resistance (GSR)

biofeedback in Tension Type Headache”.

Thesis Submitted to

The KLE Academy of Higher Education and Research,

Belgaum

(KLE DEEMED UNIVERSITY)

[Declared as Deemed-to-be-University u/s 3 of the UGC Act, 1956 vide Govt. of India

Notification No.F.9-19/2000-U.3 (A)]

(Accredited ’A’ Grade by NAAC)

For the award of the Degree of

Doctor of Philosophy in the Faculty of

Medicine (Physiotherapy)

by

Mrs. Veena Bembalgi M.P.T. (Registration No: KLEU/PhD/08-09/DOUNO8032)

Under the Guidance of

Prof. Dr. Karkal Ravishankar Naik DM (Neurology), Professor and Head, Department of Neurology,

K.L.E. University’s Jawaharlal Nehru Medical College,

Belgaum – 590010. Karnataka, India

September - 2013

II

KLE ACADEMY OF HIGHER EDUCATION AND

RESEARCH,

(KLE DEEMED UNIVERSITY) [Declared as Deemed-to-be-University u/s 3 of the UGC Act, 1956 vide Govt. of India



BELGAUM

DECLARATION

I hereby declare that this thesis entitled “Comparative study on

Electromyography (EMG) biofeedback and Galvanic Skin Resistance

(GSR) biofeedback in Tension Type Headache” is a bonafide and

genuine research work carried out by me under the guidance of

Dr. Karkal Ravishankar Naik DM (Neurology), Professor and Head,

Department of Neurology, KLE University’s Jawaharlal Nehru

Medical College, Belgaum-590010 Karnataka, India. The thesis or

any part thereof has not formed the basis for the award of any

degree/fellowship or similar title to any candidate of any

University.

Date: Mrs. Veena Bembalgi M.P.T

Place: Belgaum KLES College of Physiotherapy,

Hubli

III


RESEARCH,




BELGAUM

This is to certify that the thesis entitled “Comparative study on


(GSR) biofeedback in Tension Type Headache” is a bonafide record

of original research carried out by Mrs. Veena Bembalgi for the

award of DOCTOR OF PHILOSOPHY (PHD) IN FACULTY OF

Medicine (Physiotherapy) under my supervision and guidance.

Date Dr. Karkal Ravishankar Naik DM (Neurology), Place: Belgaum Professor and Head, Department of Neurology,

KLE University’s

Jawaharlal Nehru Medical College,

Belgaum.

IV


RESEARCH,




BELGAUM



(GSR) biofeedback in Tension Type Headache” is a bonafide and

genuine research carried out by Mrs. Veena Bembalgi M.P.T under

the guidance of Dr. Karkal Ravishankar Naik DM(Neurology), Professor

and Head, Department of Neurology, KLE University’s

Jawaharlal Nehru Medical College, Belgaum, Karnataka, India.

Date: Dr. A. S. Godhi M.S, FICS Place: Belgaum Dean Faculty of Medicine,

K.L.E. University,

J. N. Medical College,

Belgaum -590010. Karnataka

V


RESEARCH,




BELGAUM

ENDORSEMENT BY THE PRINCIPAL



(GSR) biofeedback in Tension Type Headache” is a bonafide

research work done by Mrs. Veena Bembalgi under the guidance of

Dr. Karkal Ravishankar Naik DM(Neurology), Professor and Head,

Department of Neurology, KLE University’s Jawaharlal Nehru

Medical College, Belgaum, Karnataka, India.

Date: PRINCIPAL,

Place: Belgaum Dr. Sanjiv Kumar

KLEU Institute of

Physiotherapy, Belgaum

VI


RESEARCH,




BELGAUM

Copyright Declaration

We hereby declare that the KLE Academy of Higher Education and

Research, Belgaum, Karnataka, Belgaum, Karnataka shall have the

rights to preserve, use and disseminate this thesis in print or

electronic format for academic / research purpose.

Mrs Veena Bembalgi M.P.T

Dr. Karkal Ravishankar Naik DM(Neurology) Professor and Head,

Department of Neurology,

KLE University’s Jawaharlal Nehru Medical College,

Belgaum, Karnataka, India.

Date:

Place: Belgaum

© KLE ACADEMY OF HIGHER EDUCATION AND RESEARCH,

BELGAUM

VII

KLE ACADEMY OF HIGHER EDUCATION AND RESEARCH,

BELGAUM

UNDERTAKING

I, Mrs. Veena. A. Bembalgi hereby declare that the information

and the data mentioned in my thesis entitled “Comparative

study on Electromyography (EMG) biofeedback and

Galvanic Skin Resistance (GSR) biofeedback in Tension

Type Headache” belongs to me and is original.

I am aware of the definition of plagiarism as detailed below:

An act or instance of using or closely imitating the language

and thoughts of another author without authorization and the

representation of that author’s work as one’s own, as by not

crediting the original author.

A piece of writing or other work reflecting such unauthorized

use or imitation.

The deliberate or reckless representation of another’s words,

thoughts or ideas as one’s own without attribution in connection

with submission of academic work, whether graded or otherwise.

I hereby declare that the thesis prepared by me is original-one and

does not involve plagiarism anywhere. In case at a later stage it is

found that I have indulged in plagiarism, then I am solely

responsible for the same and the institution is at liberty to take

any disciplinary action against me including cancellation of

dissertation or any other penalties imposed by the university.

Date:

Place: Mrs. Veena Bembalgi M.P.T

VIII

Acknowledgement

I owe my accomplishment of the course to my late mother-in-

law, Mrs. Gurudevi Bembalgi, who was my first motivator to join the

PhD programme. It was solely her motivation and support that

inspired me to join this course. I owe my entire work to her.

I extend sincere gratitude to my guide Dr. Karkal Ravishankar

Naik for his meticulous guidance and encouragement through-out the

study. He was the main driving force for the completion of this study.

He is a perfectionist and very systematic in his work. Being his pupil

has helped me inculcate those qualities too.

I thank all my participants for their participation without which

it would be impossible to perform the study.

With great privilege, I take the opportunity to express my

sincere thanks to Honorable Chancellor KLE University & Chairman

of K.L.E Society, Dr. Prabhakar. B. Kore for providing us the

opportunity to accomplish the course.

I am grateful to Honorable Vice-Chancellor KLE University

Dr. (Prof) C. K. Kokate, Belgaum, for his timely motivation and

inspiration throughout the course.

IX

I am extremely happy to extend my heartfelt thanks to Dr. P. F.

Kotur, former Registrar, KLE’s University Belgaum, for his support

and encouragement.

I express sincere thanks to Dr. M.G.Mokashi, for his guidance

and support during my study and thesis writing phase.

I thank Dr. Prashant Mukkannavar and Prof. Steve. Simon

(U.K), who helped me tremendously with the statistical analysis.

My heartfelt thanks to my father-in-law, Dr. M.S. Bembalgi and

my husband Dr. Anilkumar Bembalgi and all my family members for

their constant support and encouragement throughout the course

period. I, also thank Dr. Anilkumar Bembalgi for helping me with the

analysis of data on drugs.

I express my gratitude to Dr. Suresh. DuganiMCH(Neurosurgery),

Dr. Rajendra. DuganiDM(Neurology) and all other physicians for their

guidance and for referring me my study participants.

Dr. Sanjivkumar, former Principal, KLES College of

Physiotherapy, was a source of constant support and motivation. I

thank him sincerely and am grateful to him.

I thank my colleagues Dr. Rashmi Saibannavar and Dr.

Shradhha Bhandari and my students Madhura Palande, Amruta

X

Deshpande, Richa Singh and Radhika Galgali for helping me with the

data recordings and calculation of data scores.

I am grateful to Mr. Siddappa and Mr. Basavraj Antannavar,

Assistants at KLES College of Physiotherapy, Hubli for keeping the

treatment room open and meeting the requirements for the therapy

even on holidays.

Sincere thanks to Prof. Richard Sherman, Director, Behavioral

Medicine Research and Training foundation, U.S.A, for answering my

queries and sharing some valuable information.

I also thank Mr. Girish Aladi, Mr. Kumaresh. Hundekar and

Mr. Halemani for their technical support and help.

I extend sincere gratitude to all who have knowingly or

unknowingly contributed in the completion of my study.

Date: Mrs. Veena Bembalgi

Place: Belgaum

XI

LIST OF ABBREVIATIONS:

AAPB: Association of Applied Psychophysiology and Biofeedback

AHS: American Headache Society

BF: Biofeedback

CDH: Chronic Daily Headache

CI: Confidence Interval

CTTH: Chronic Tension Type Headache

EMG: Electromyography

EMGa: EMG audio group

EMGav: EMG audiovisual group

EMGv: EMG visual group

GSR: Galvanic Skin Resistance

GSRa: GSR audio group

GSRav: GSR audiovisual group

GSRv: GSR visual group

ICHD: International Classification of Headache Disorders

IHS: International Headache Society

ISNR: International Society for Neurofeedback and Research

NIH: National Institute for Health

QoL: Quality of Life

REMG: Relative EMG

RMANOVA: Repeated Measures Analysis of Variance

XII

SEMG: Surface EMG

SF-36: 36-item Short form health survey

SPSS: Statistical Package for the Social Sciences

TTH: Tension Type Headache

WHO: World Health Organization

ηp2: Partial eta squared value

XIII

ABSTRACT

Background and purpose: Tension type headache (TTH) is the most widespread

and most common primary headache disorder accounting for nearly 90 % of all

headaches. Efficacy of electromyography (EMG) biofeedback (BF) in patients

with tension type headache has been extensively studied and proven. However,

efficacy of galvanic skin resistance (GSR) biofeedback has not been studied

adequately. So far there are no studies on the efficacy of isolated audio, visual and

combined EMG or GSR BF in TTH. The aim of the present study was to compare

the efficacy of electromyography biofeedback and galvanic skin resistance

biofeedback in patients with tension type headache and to study and compare the

efficacy of auditory, visual and audio-visual biofeedback in patients with TTH.

Methodology: This study was a randomized single blinded controlled prospective

study. Out of 232 recruited subjects, 211 (145 females and 66 males) were

randomly assigned to seven groups receiving electromyography feedback auditory

(EMGa) (n =27), visual (EMGv) (n=28), combined audio-visual (EMGav) (n=27),

galvanic skin resistance biofeedback auditory (GSRa) (n =26), visual (GSRv)

(n=29) and combined audio-visual (GSRav) (n=28) and a control group (n = 27).

Each subject (except the control group) received 15 sessions of respective

biofeedback for 30 minutes each in an isolated room. The control group received

only medication prescribed by their treating doctor. Each patient was blinded to

the type of biofeedback (EMG or GSR) being given. Pain variables (average

frequency, duration and intensity of headache per week), SF-36 quality of life

XIV

scores and analgesic usage were recorded at baseline, 1 month, 3 months, 6

months and 1 year after therapy.

Results: All groups showed a significant decrease in pain variables (p

XV

TABLE OF CONTENTS

Sl. No. Sections Page No.

1. INTRODUCTION 1-12

2. NEED OF THE STUDY 13-14

3. RESEARCH QUESTION 15

4. AIMS AND OBJECTIVES 16

5. REVIEW OF LITERATURE 17-27

6. METHODOLOGY 28-35

7. RESULTS 36-82

8. DISCUSSION 83-99

9. CONCLUSION 100

10. SUMMARY 101-103

11. BIBLIOGRAPHY 104-121

12. ANNEXURE – I PHOTOGRAPHS 122-124

13.

ANNEXURE – II

ETHICAL CLEARANCE, CONSENT FORM,

PROFORMA

125-134

14.

ANNEXURE – III SF-36 QUESTIONNAIRE

KANNADA

ENGLISH

15. ANNEXURE – IV PUBLICATIONS

16. ANNEXURE – V MASTER CHART

XVI

LIST OF TABLES

Table

No. Particulars Page No.

01 Demographic Data 39

02 Baseline pain variables in the study group 41

03 Baseline SF-36 variables in the study group 43

04 Consumption of analgesics 44

05 Intra and Inter group analysis of pain variables of

all groups 46

06 Intra group comparison of pain variables in EMG

groups 48

07 Intra group comparison of pain variables in GSR

groups 49

08 Intra group comparison of pain variables in the

control group 52

09 Intergroup comparison of pain variables at

1month post intervention 53

10 Intergroup comparison of pain variables at 3

months post intervention 54

11 Intergroup comparison of pain variables at 6


12 Intergroup comparison of pain variables at 1 year

post intervention 56

XVII

13

Intra and inter group analysis of SF-36 scores

a) Total scores b) Physical scores c) Mental scores

61

62

63

14 Intra group comparison of SF-36 variables in the

EMG groups 64


GSR groups 65


control group 66

17 Intergroup comparison of SF-36 scores at 1 month


18 Intergroup comparison of SF-36 scores at 3


19 Intergroup comparison of SF-36 scores at 6


20 Intergroup comparison of SF-36 scores at 1 year


21 Intra group comparison of analgesic consumption

for EMG groups 76


for GSR groups 77


for control group 78

24 Improvement in pain variables in percentage 80

25 Improvement in SF-36 scores in percentage 81

26 Effect size for all outcome measures at one year 82

XVIII

TABLE OF FIGURES

Figure No Particulars Page No

01 Flow chart of subjects through the trial 38

02 Means of pain variables at baseline 40

03 Means of SF-36 scores at baseline 42

04 Trend of average frequency of headache

through the time measures 47

05 Trend of average duration of headache


06 Trend of average intensity of headache


07 Trend of SF-36 total scores through the

time measures 60

08 Trend of SF-36 physical scores through the

time measures 72

09 Trend of SF-36 mental scores through the

time measures 74

XIX

LIST OF PHOTOGRAPHS

Photograph

No Particulars Page No

01 EMG biofeedback machine 122

02 GSR biofeedback machine 122

03 Placement of EMG biofeedback electrodes 123

04 Placement of GSR biofeedback electrodes 123

05 Subject receiving EMG audiovisual

biofeedback 124

06 Subject receiving GSR audiovisual

biofeedback 124

Introduction

1

Introduction

1

INTRODUCTION

Stress is a frequent occurrence in all our lives. It is a state of

physiological or psychological strain caused by undesirable stimuli, physical,

mental or emotional; internal or external that could likely disturb the functioning

of an individual.1 When stress is being experienced by a person constantly with no

relief or with increased frequency, it is termed as “distress”. Distress leads to

weakened cognitive and physiological control and, as a result, decreased

performance. It can lead to symptoms like headache, gastrointestinal disturbances,

elevated blood pressure, chest pain, insomnia, peptic ulcers, sexual dysfunction,

skin ailments, etc.

The physiological responses to stress may differ with regards to acute and

chronic stress. Acute stress generally is short lived and causes no actual damage,

whereas chronic stress can cause a sustained response to stress causing damage

and chronic pain.

Stress reactions (response to stress) cause amplification of physiological

parameters such as muscle tension, blood pressure, increased sweating, etc. This

causes disorders in the body like headaches, irritable bowel syndrome, ulcers,

hyperhidrosis, chest pain, etc. Eventually, it results into a vicious cycle wherein

stress causes pain or stress related disorders and increased pain or other

symptoms, which leads to further amplification of stress.

Stress related disorders are often termed as “psychosomatic” disorders

which involves the mind and body. These are the disorders in which the mind

makes the body vulnerable for disorders. Tension-type headache (TTH) is one of

Introduction

2

the common and chief diseases in psychosomatic medicine because of its

correlation with psychosocial factors.2

Mental stress and tension are the most frequently reported triggers of

tension-type headache.3,4

Genetic or family-related environmental factors are also

associated with TTH.5 Major life events such as surgery, divorce and deaths of

close family members induce major negative effect. Such events in the prior year

have been modestly related to the persistence of headache.6 In addition to physical

variables like muscle tension, electro-dermal activity, temperature, etc and other

demographic variables of pain, psychological risk factors have been empirically

associated with the occurance of headache. These comprise, social support,

hypnotizability, affect, life events, and negative thinking.7 Trait negative

affectivity is raised in chronic headache causing over reporting of somatic

symptoms like headache pain, irrespective of organic disease.8,9

This indicates

that mental health is largely affected in patients with TTH and therefore a good

deal of attention should be paid to the psychological component in terms of

assessing and taking measures to improve the mental health of patients with TTH.

Introduction- Biofeedback

3

BIOFEEDBACK

The term “Biofeedback” was voted against the term Autoregulation in

1969. The organization who coined this word was named the Biofeedback

Research Society (BRS). In 1976, the BRS was renamed Biofeedback Society of

America (BSA). The present name of the society, the Association for Applied

Psychophysiology and Biofeedback came into existence in 1989.

Edmund Jacobson, a physician was one of the earliest contributors in the

field of biofeedback. In 1938 he monitored electromyography (EMG) of patients

practicing progressive muscle relaxation to find out if the muscles actually

relaxed.

Previously, it was believed that autonomic responses could not be

controlled voluntarily. Miller and Leo DiCara in 1962 demonstrated that

curarized rats could learn to control their autonomic functions (breathing patterns,

muscle tone, blood pressure, salivation, GSR, etc).

In 1966, Joe Kamiya, who is popularly known as “the father of

biofeedback” found that some subjects could learn to discriminate the presence of

alpha waves when electroencephalography (EEG) was performed on them. He

also found that they could learn to manipulate their alpha frequency by about 1

Hz, thus establishing that subjects could control their own neuro-biological

rhythm.

Physicians Marinacci and Whatmore practiced biofeedback even before

the term was founded. They used EMG biofeedback to treat stroke patients. But


4

their work on neuromuscular re-education was not continued by others and

remained undeveloped till it was rediscovered.

Significant contributions to this field have been made by researchers in the

clinical aspects like (a) Basmajin, who used surface EMG to study role of

different muscles in movements and used the information for rehabilitation, (b)

A.Kegel, who used pneumatic biofeedback devices to train pelvic floor muscles,

(c) Johan Stovya used biofeedback for treating anxiety and (d) Thomas

Budzynski used SEMG for treatment of headaches.10

The Association for Applied Psychophysiology and Biofeedback (AAPB),

the Biofeedback Certification Institute of America (BCIA), and the International

Society for Neurofeedback and Research (ISNR) convened a task force of

renowned scientists and clinicians in late 2007 who worked together to craft a

standard definition for biofeedback. They defined biofeedback as " a process that

enables an individual to learn how to change physiological activity for the

purposes of improving health and performance.”.11

Precise instruments measure physiological activity such as, heart function,

, muscle activity, breathing, electroencephalogram, skin temperature etc. These

biofeedback instruments rapidly and accurately "feed back" information to the

user. The use of this information, often in combination with changes in thinking,

emotions, and behavior supports needed physiological changes.11

Patients with the

use of this information (biofeedback) learn enhanced control over the

physiological process (operant learning model).12

Over time, these changes can be

preserved without continued use of an instrument.10

Any learning is facilitated by


5

feedback. The same principle is used in biofeedback therapy whose main aim is to

assist the patients in self regulation of psycho-physiological factors, thereby

allowing them to gain voluntary control over physiological parameters.

Learned behavioral control over physiological responses was first

published in 1961. In the 1960s and 1970s, human studies revealed that through

operant feedback methods, voluntary control could be learnt over many

physiologic responses (e.g., heart rate, blood pressure, skin conductance, muscle

tension, skin temperature, evoked potentials and various rhythms of EEG).13

Biofeedback therapies are non-pharmacologic treatments that use scientific

instruments to measure, amplify, and feed back physiological information to the

patient being monitored, thereby promoting control and manipulation of

physiological parameters. It is virtually free of any adverse side effects and

therefore seemingly the preferable choice for treatment of psychosomatic

disorders.14

Biofeedback therapy has evolved over the last 30 years, and today there

are innumerable disorders for which biofeedback therapy has been used.

Biofeedback therapy is now used for a variety of disorders, such as headache

(migraine, tension and mixed), urinary incontinence, essential hypertension etc

with reliable results.

Introduction- Tension Type Headache

6

TENSION TYPE HEADACHE

Headache is a clinical syndrome affecting 91% of all males and 96% of all

females at some time during their life span.15

The World Health Organization

recognized that primary headaches are among the first 20 major causes of

disability.16

In the primary care practice, tension type headache is the most

commonly diagnosed variety of primary headache .17

Tension type headache, formerly called tension headache or muscle

contraction headache is the most frequently occuring headache disorder.18

It is the

commonest among primary headaches. It is the most dominant and costly

headache.19

Tension type headaches are responsible for nearly 90% of all

headaches. As per the International Headache Society (IHS), its lifetime

occurance in the general population ranges in different studies from 30 to 78%.

Inspite of its high prevalence and regardless of the fact that it has the highest

socio-economic impact, it is still the least studied of the primary headache

disorders.20

Population based studies have established that 24–37% of the adult

populations have TTH several times a month; 10% have it weekly; and 2–3%

have chronic TTH, usually lasting for many years. 21, 22

A study of the global

prevalence and burden of headaches showed that the community burden resulting

from disability caused by TTH is greater than that of migraine.19

Tension type headache is more common in women, in a ratio of 1.5:1,23

whereas other studies have shown that the female to male ratio of TTH is 5:4.19, 24

Published estimates of the prevalence of tension type headache vary over a wide

range from 1.3% to 65% in men and 2.7% to 86% in women.18, 25-35


7

A World Health Organization (WHO) statement released in 2000 on

headache disorders and public health quotes that the onset of TTH is often in the

teen and prevalence peaks in the fourth decade and subsequently declines23

,

whereas the average age of onset of TTH was found to be 25–30 years in cross-

sectional epidemiological studies21

. The prevalence peaks between the age of 30

to 39 and decreases slightly with age.22

Some of the risk factors for developing TTH have been reported to be poor

self-rated health, unable to relax after work and sleeping few hours per night.22

Two Danish studies have shown that the number of workdays missed was three

times higher for TTH than for migraine in the population, 21, 22

and a US study has

also found that absenteeism because of TTH is considerable.36

In a study by Fuh et al, 2008 where a cohort study was conducted to study

the outcome of elderly patients with chronic tension type headache (CTTH) in a

span of 13 years, the authors found 30% of patients with CTTH evolved to

chronic migraine (CM) or episodic migraine.37

Therefore it is important to curb

the tension type headache before it transforms to migraine which could lead to

difficulty in treating due to its complex nature.

Tension type headache is clinically and patho-physiologically

heterogeneous. The complex interrelation of the various pathophysiological

factors of TTH; makes this disorder often difficult to treat. Various therapeutic

measures have been recommended to be used in sequence or in combination.

Therapies for TTH can be subdivided into short term, abortive treatment of each

attack (mainly pharmacological) and long term, prophylactic treatments

(pharmacological and/or non-pharmacological).38


8

Several non-pharmacological treatments have been recommended for

management of TTH, them being physical therapy,39,40,41

craniocervical training,42

oro-mandibular treatment43,44

acupuncture,45,46,47

relaxation therapies,48,49

cognitive-training50,51

biofeedback52,53

etc. However, the scientific evidence for

efficacy of most treatment modalities is sparse.54, 55, 56, 57

Biofeedback is one of the most prominent behavioural headache

treatments. It is an established non-pharmacologic technique commonly used in

the treatment of migraine and tension type headaches.58

Several published studies

have suggested that biofeedback is effective in reducing the frequency and

severity of headaches, thereby limiting the patient’s dependence on medication.

Conforming to this, studies have also proposed that biofeedback may effect a

reduction in medical utilization in headaches.59, 60, 61, 62

Introduction- Biofeedback in TTH

9

BIOFEEDBACK IN TENSION TYPE HEADACHE

Enormous research carried out in the mid-20th century on stress and

illness, formed the basis needed to establish headache as a psycho-physiological

disorder, thereby justifying the application of contemporary behavioral headache

treatments. Over the past thirty years, these behavioral treatments for headache

have gathered a sizeable evidence base.

Budzynski and colleagues were the first to publish demonstration of

biofeedback for tension headache treatment. They developed the EMG

biofeedback model and protocol for tension headache63

and pursued to

demonstrate initial headache improvements in uncontrolled64

and controlled

experiments.65

The American Headache Society recognized biofeedback as a valid form

of headache therapy in 1978.66

The U.S. Headache Consortium, which was a

multi-disciplined assemblage of seven professional practice organizations also

endorsed behavioral therapy including biofeedback for headache as important

evidence based treatment.12

Meta-analytic reviews of the literature consistently have shown behavioral

interventions to yield 35% to 55% improvements in migraine and tension-type

headache and that these outcomes are significantly superior to control conditions.

The positive evidence from these studies has lead many professional practice

organizations to recommend use of behavioral headache treatments alongside

pharmacologic treatments for primary headache.12

A Task Force of the Association for Applied Psychophysiology and

Biofeedback and the Society for Neuronal Regulation was formed in 2001, which


10

developed guidelines for the evaluation of the clinical efficacy of psycho-

physiological interventions. The criteria for levels of evidence of efficacy was

laid down and approved by both associations. These criteria were used to assign

efficacy levels for the vast number of conditions for which biofeedback has been

used. Use of biofeedback for headaches in adults was awarded

“level 4-efficacious”, the criteria for which were:

a. In a comparison with a no-treatment control group, alternative treatment group,

or placebo control utilizing randomized assignment, the investigational

treatment is shown to be statistically significantly superior to the control

condition, or the investigational treatment is equivalent to a treatment of

established efficacy in a study with sufficient power to detect moderate

differences, and

b. The studies have been conducted with a population treated for a specific

problem, for whom inclusion criteria are delineated in a reliable, operationally

defined manner, and

c. The study used valid and clearly specified outcome measures related to the

problem being treated, and

d. The data are subjected to appropriate data analysis, and

e. The diagnostic and treatment variables and procedures are clearly defined in a

manner that permits replication of the study by independent researchers, and

f. The superiority or equivalence of the investigational treatment has been shown

in at least two independent research settings. 67

The task force included all studies in which biofeedback was used and

found the therapy efficacious.


11

Later, in 2008 a review article on efficacy of Biofeedback in TTH quoted

that BF in TTH can be supported as an efficacious and specific treatment option,

which according to the Association of Applied Psychophysiology and

Biofeedback (AAPB) and International Society for Neurofeedback and Research

(ISNR), criteria this constitutes the highest level of evidence (Level 5), reserved

for psycho-physiological interventions, that have established Level 4 evidence and

have shown additional superior treatment results in comparisons to credible sham

therapy or alternative bona fide treatments.68

The past three decades have shown, behavioral interventions (chiefly

relaxation, biofeedback, and stress management) to have become standard

components in the varied choice of treatments for management of migraine and

tension type headaches. Meta-analytic literature reviews have consistently

demonstrated clinically significant reductions in recurrent headache through

behavioral interventions . Behavioral interventions have yielded approximately

35-50% reduction in migraine and tension type headache activity. Although

comparisons between standard drug and non-drug treatments for headache have

been initiated only recently, the available evidence suggests that the level of

headache improvement with behavioral interventions may prove beneficial over

those obtained with widely used pharmacologic therapies in representative patient

samples. In recent years, some efforts have been made to increase the availability

and cost effectiveness of behavioral interventions through alternative delivery

formats and mass communications.13,52

Biofeedback treatments for TTH provide patients with feedback of

physiological processes, thereby assisting them to gain voluntary control over


12

bodily functions by manipulation of physiological parameters (e.g., to reduce

dysfunctional muscle tension, increase skin resistance etc) and thereby augment

self efficacy in dealing with pain episodes.53

Of all biofeedback therapies or techniques, EMG biofeedback has been

extensively reviewed and used the most. A recent extensive and thorough meta-

analysis including 53 studies concluded that biofeedback has a medium to large

effect and the effect was found to be long lasting up to 15 months.68

GSR BF has been used in treatment of stress69

and related psychosomatic

disorders like hypertension1, hyperhidrosis,

70 Raynaud’s disease,

70 epilepsy

71 etc

but has been infrequently used in management of TTH.12, 48, 68

Need of the Study

2

Need for the study

13

NEED FOR THE STUDY

Pharmacotherapy remains the mainstay of treatment for all types of

headaches and vast amounts of prescription and over-the-counter medications are

used. Side effects frequently occur with these medications which at times can be

life-threatening. The medications themselves often contribute to the reduced

productivity among headache sufferers.14

A number of treatment strategies used in the treatment of TTH, either in

isolation or combined with one another. These consist of pharmacological

treatment, physical therapy, acupuncture, relaxation therapy or alternative

medicine. Biofeedback though has proven its efficacy in the treatment of TTH, is

not widely used in India by health professionals. Biofeedback therapy is still a

novel and an infrequently used therapy for treating practitioners as well as the

common people in India. Therefore, a study of this nature was required to create

an awareness of this field of therapy among the health practitioners as well as the

community. Of all biofeedback techniques, EMG biofeedback has been

extensively studied in the management of TTH. However, despite the fact that

GSR biofeedback is used in many other disorders with psychosomatic components

in the pathogenesis like hypertension, epilepsy, hyperhidrosis, etc it has been

infrequently evaluated and used in the management of TTH.

Therefore, there was a need to find the efficacy of GSR biofeedback in

TTH and also compare its efficacy with EMG biofeedback in patients with TTH.

A novel attempt has also been made in studying and comparing the

efficacy of auditory, visual and combined biofeedback in both EMG and GSR

Need for the study

14

biofeedback in TTH subjects. The implications of this could be used in designing

and manufacturing of EMG and GSR biofeedback units for therapeutic benefits.

Research Question

3

Research Question

15

RESEARCH QUESTION

Which type of biofeedback (electromyographic or galvanic skin resistance) and in

which form, is effective in the treatment of tension type headache?

HYPOTHESIS

The working hypotheses of this study were:

1. Electromyography (EMG) biofeedback is more effective than Galvanic

Skin Resistance (GSR) biofeedback in the treatment of tension type

headache (TTH).

2. There is no difference in effectiveness of auditory, visual or combined

EMG biofeedback in the treatment of TTH.

3. There is no difference in effectiveness of auditory, visual or combined

GSR biofeedback in the treatment of TTH.

Aims & Objectives

4

Aims and Objectives

16

AIMS AND OBJECTIVES

Aims

To study and compare the efficacy of EMG biofeedback with GSR

biofeedback in patients with Tension Type Headache.

To study and compare the efficacy of auditory feedback, visual feedback

and both visual and auditory feedback together.

Objectives

To examine the efficacy of EMG biofeedback in tension type headache

patients.

To examine the efficacy of GSR biofeedback in tension type headache

patients.

To compare the efficacy of EMG biofeedback and GSR biofeedback in

tension type headache patients.

To compare the effectiveness of isolated visual, isolated auditory or

combined visual and auditory biofeedback together of EMG biofeedback

in tension type headache patients.

To compare the effectiveness of isolated visual, isolated auditory or

combined visual and auditory biofeedback together of GSR biofeedback in

tension type headache patients.

Review of

Literature

5 5

Review of literature

17

REVIEW OF LITERATURE

Pharmacotherapy has remained the main mode of treatment of people

affected with TTH. Though pharmacotherapy is modestly effective in the reducing

the frequency and intensity of TTH 72

, it has some evident drawbacks. First, the

widely used antidepressant medications are associated with multiple potential

adverse effects. Second, chronic TTH (CTTH) constitutes a risk factor for

analgesic medication overuse and the development of medically induced headache

in addition to the adverse effects due to long term consumption.73, 74

Behavioural treatments (Biofeedback)

Behavioural treatments have been recommended as an adjunct to

pharmacotherapy. Of these behavioural treatments, biofeedback has formed an

evidence based treatment option for TTH. These treatments involve the patient’s

active role in preventing and management of headache episodes and thereby

improving the coping with the physiological and psycho-physiological

consequences of pain.53

Previous quantitative reviews and meta-analyses have assessed the

outcome of various behavioural treatments for TTH, such as cognitive therapy,

biofeedback, relaxation and hypnotherapy75,76,77,78,79

and have shown average

improvement rates that exceeded those of no treatment conditions.76

The

maximum treatment gains were achieved for electromyography feedback (EMG-

FB) alone or in combination with relaxation, which is the predominantly applied

biofeedback modality for TTH. Average improvement scores ranged from 46% 78

to 61%75

.


18

One meta-analysis investigated psychological headache treatments and

provided standardized measures of treatment success, which resulted in a medium

to large average effect size for EMG-FB in adults.79

Investigations of specific

versus nonspecific effects of biofeedback in comparison with other behavioural

headache treatments were not meta-analytically integrated in that study. Also, the

long-term effects of the efficacy on different outcome variables, and treatment

mediators of biofeedback were not systematically analyzed. These limitations

were targeted in another meta-analysis whose objective was to present an up to

date and comprehensive evaluation of the efficacy of biofeedback for TTH. Its

first aim was to establish the short and long term efficacy of biofeedback as well

as the treatment specificity in comparison to various control groups. Another aim

was to determine, differential treatment effects in the form of pain measures and

of psychological, behavioural, and physiological outcome categories. In

continuation, analyses of potentially moderating effects of treatment and patient

characteristics were incorporated. Lastly, specific meta-analytical techniques were

applied to control for possible confounding effects of selective publication,

dropout, and study validity. Meta-analytic integration of 53 studies resulted in a

significant medium to large effect size (d=0.73; 95% confidence interval=0.61,

0.84) that was stable over an average follow up phase of fifteen months.

Biofeedback proved to be more effective than headache monitoring, relaxation

therapies and placebo. The strongest improvements resulted for frequency of

headache episodes. In addition significant effects were observed for self-efficacy,

muscle tension, depression, symptoms of anxiety and analgesic medication. The

meta-analysis also found that biofeedback for TTH can be supported as an


19

efficacious and specific treatment option. According to the AAPB and ISNR

criteria this constitutes the highest level of evidence (Level 5), reserved for psycho

physiological interventions.68

Andrasik80

reviewed meta-analyses and evidence-based reviews of

behavioural treatments for headaches in adults. After considering all meta-

analyses to date he concluded that the effects of behavioural treatments were

superior to various control and placebo conditions and similar to current

medications for both migraine and tension type headache. Combining behavioural

and pharmacological treatments may increase effectiveness even further.

Several reports of unsuccessful biofeedback training have appeared in the

research literature since the inception of biofeedback training three decades ago.

Many of the unsuccessful studies conducted in the early development of the field

reflect failure to thoroughly train the patients. For example, some unsuccessful

studies provided only minimal training with the biofeedback instrumentation

(often one to four sessions of brief duration), offered little coaching, involved no

home practice, and failed to conform to clinical criteria.12

Biofeedback is a time tested therapy for psychosomatic disorders. Various

forms of biofeedback targeting the psycho-physiological parameters either directly

or indirectly have been used in the treatment of TTH, namely EMG biofeedback,

temperature biofeedback, GSR biofeedback, blood volume pulse biofeedback and

electroencephalogram (EEG) biofeedback. Of these EMG BF is the most

frequently used one and GSR BF has been used seldom.68


20

EMG Biofeedback

A recent review article including 11 studies concluded that owing to low

power of studies, the evidence to support or refute EMG biofeedback’s role in

TTH compared to placebo or any other treatments is conflicting.56

Since the

conclusion was drawn from a small set of studies, its results cannot be

generalised. Another critical review by Krishnan and Silver81

on the meta-analysis

by Nestrouric et al53

found insufficient evidence to determine whether EMG

biofeedback is effective in treating CTTH. It is difficult to comment on their

outcome, since the conclusion was drawn from critical analysis of one meta-

analysis only.

An assessment conducted by National Institutes of Health (NIH) panel on

efficacy of behavioural and relaxation therapies in chronic pain, established that

EMG biofeedback was more effective than psychological placebo but equivalent

to relaxation therapy for TTH.82

When analysis was done to no treatment or to

pseudo-placebo therapy, EMG BF alone or combined with relaxation therapy

were found to be more superior in a meta-analytic review of 78 articles with 2866

patients receiving cognitive feedback, relaxation therapy, EMG BF or EMGBF in

combination with relaxation therapy.76

Another meta-analysis found a 48% decrease in headache activity in which

EMG biofeedback/relaxation therapy was used with limited therapist contact.83

Reducing therapist contact time with no loss in efficacy is an important

consideration in terms of cost to the patient and in improving the confidence of the

patient to cope with headache.


21

Maintenance of the therapeutic effects of EMG BF is an important

consideration, which should not be overlooked. Nine TTH patients receiving

EMG BF/relaxation therapy were followed up to 5 years after treatment. A

headache index computed from the headache diary found that 78% of the patients

remained improved.84

GSR Biofeedback

A thorough literature search yielded two studies using GSR biofeedback

in tension type headache85,86

. The study by Collet et al85

was a comparative study

between GSR feedback (n = 16) and Schultz relaxation (n = 15) in patients with

tension headaches. They found no significant improvement in the group treated by

relaxation at the end of the treatment whereas the group treated by GSR feedback

showed significant improvements with respect to frequency and intensity of

headaches and to anxiety as measured by subjects' self-evaluation (p


22

investigators found that there was a significant increase in electrodermal activity

(p


23

pharmacological treatments. Their main intention is to improve the design of trials

and the reporting of results of studies on behavioural interventions for headache.

Measuring sites for EMG and GSR biofeedback

The American Headache Society (AHS) recommends measurement of GSR

from middle or distal phalanx of any two fingers or the palm.88

Similarly, the

psycho-physiological guidelines for techniques in measurement of electro-dermal

activity, also recommends that electro-dermal resistance be measured form either

middle or distal phalanges of two fingers, or thenar or hypothenar eminence of

palmar surface.89

Many researchers have preferred middle phalanx to distal as it is

thought to have less scarring and movement as compared to distal phalanx90,91,92,93

.

In a review of articles from 1985 to 1990 in the Psychophysiology journal,

the reviewers found that among 53 studies that used skin conductance as their

dependant variable 18 (34%) measured skin resistance from distal phalanx, 10

(19%) from the middle phalanx, 14 (26%) from the palm and three (6%) from

other sites; the remaining eight studies (15%) did not mention the site of

recording.94

Therefore it was seen that maximum studies used either distal or

middle phalanx for recording skin resistance. The variation of electrodermal

responses recorded from the distal and middle phalanx was studied by the same

group, which showed that distal phalanges were more electrodermaly active than

middle phalanx and that distal phalanx provide a more sensitive measure of electro

dermal activity than the middle phalanx.

On the contrary, Edelberg suggested that there are greater skin responses

from areas of greater tactile sensitivity.95

Ruch found that two-point discrimination

is higher (distance between the two points to be discriminated/recognised as two


24

points) in fingertips meaning that, finger tips are more sensitive than middle

phalanges.96

Therefore these areas show greater skin responses. In addition to this

there are greater numbers of sweat glands present on distal phalanx therefore

giving a greater skin conduction response.97

Therefore in this study the distal

phalanx of index and ring fingers are being used to assess the GSR activity, as

used by investigators in a recent study in which GSR BF was used in stressed

individuals.1

The most commonly used and popular placement of surface EMG

(SEMG) electrodes for TTH is on the frontalis muscle.98

In our study, the EMGBF

electrodes were placed 2.5cm above the centre of each eyebrow after appropriate

skin preparation as was followed in a study by Cohen et al.99

Outcome variables

As per the recommendation of the American Headache Society

behavioural clinical trials workgroup, 88

measure of frequency of headache should

be reported as the primary dependent variable. The recommendation for headache

frequency reporting is consistent with the IHS guidelines for controlled trials of

drug treatments100,101

and will facilitate meta-analyses and other comparisons

across studies of various interventions.102,103

The IHS clinical trials guidelines for

migraine prefer headache frequency to headache index because there is no

consistent definition of headache index, and changes in this measure can be more

difficult to interpret clinically.99

The guidelines have also mentioned a high desire for trials to include a

sufficiently wide spectrum of secondary outcome measures to capture possible

differential outcomes of drug and behavioural therapy. This is because, even if


25

drug and behavioural therapies are found to exert a similar impact on the primary

outcome variable, the impact of these two treatment modalities may differ on

other measures of functioning (e.g., psychological symptoms, quality of life,

efforts to manage headaches, etc).102,104,105,106

Identification of potential

differences in the impact of the two therapy modalities requires the inclusion of

outcome measures that assess a range of outcomes likely to be impacted by either

therapy modality.107,108

Moreover, studies have shown that genetic or family related environmental

factors are associated with about 50% of all cases.109

Major life events such as

surgery, divorce and deaths of close family members induce negative effect. Such

events in the prior year have been modestly related to the persistence of

headache.6 In addition to physical variables like muscle tension, electro-dermal

activity, temperature, etc and other demographic variables of pain, psychological

risk factors have been empirically associated with the incidence of headache.

These include life events, social support, hypnotisability, affect and negative

thinking.7 Trait negative affectivity is elevated in chronic headache and has been

related to over reporting of somatic symptoms like headache pain, independent of

organic disease.8, 9

. Inspite of TTH being the most prevalent type of headache,

little has been published about effect or burden of TTH on the health related

quality of life (HRQoL). Most of the research on HRQoL has been focussed on

migraine.110

Only a few studies pertaining to HRQoL have been done on chronic

TTH patients in general population111

or in specialised headache clinics112

.

Pain variables like frequency, intensity and duration of pain, can only

provide information about pain but not the impact that it has on the patient’s life.


26

Above review suggests that TTH has an impact on physical and mental health of

the patients, thereby affecting the quality of life in social life, at work place etc

causing impairment in overall functioning.

Therefore, in lieu of this secondary outcome variables chosen in our study

SF-36 QoL (Quality of life) total, physical and mental scores could cover a

broader aspect of the physical, mental and social aspects of the patient which

would represent the global health of the patient and impact of the headache on the

patients’ health.

In a population based study in Spain, the investigators found SF-36 to be a

reliable and valid tool for measuring the health related quality of life of patients

with chronic daily headache (CDH) and that it could be used as an effective tool

in studying the effectiveness of therapeutic agents for CDH.112

Another, similar

study done in the Italian population, found SF-36 tool was sensitive to clinical

changes in patients with primary chronic daily headaches.113

A first of its kind

survey, done on 901 patients with CDH, demonstrated that the SF-36 scores

varied among headache diagnoses. The SF-36 scores were greatly influenced by

psychological distress, as well as the percentages of the types of patients. Their

findings also suggested that improvement in the pain profile as well as

psychological well-being could predict a generalized improvement in the SF-36

scales in headache patients.114

Considering these studies, SF-36 QoL tool seems to

be a fairly reliable tool to assess effectiveness of intervention in TTH patients as

well.

The American Headache Society guidelines also urged the investigators to

employ a daily self-report headache diary as their principal dependent measure for


27

assessing treatment outcome whenever possible. Accordingly, a headache diary

was given to all our subjects in which they reported the frequency, duration and

intensity of every headache episode. They were asked to report the consumption

of medications as well, which we reported as one of our secondary outcome

measure.

Concurrent v/s terminal biofeedback:

Biofeedback may be concurrent or terminal. Concurrent BF is information

that is present all the time a person is receiving BF therapy. Terminal BF is

knowledge of results coming after an action, and is more likely to assist in

learning than is concurrent feedback.115

Giving concurrent feedback only

intermittently to supplement terminal knowledge of results appears to be of

probable value in permanent learning.116

Methodology

6

Methodology

28

METHODOLOGY

Study design: This study was a randomized, single blinded, prospective

controlled trial.

Source of data: Data was obtained from the subjects recruited from various

neurology clinics and subjects referred by neurologists to the outpatient

department of KLES College of Physiotherapy, Hubli for biofeedback therapy.

Study duration: Subjects were recruited from January 2009 up to August 2011

and followed up till August 2012.

Informed consent: Subjects were recruited in the trial only after obtaining

informed consent from them. (Informed Consent Form approved by the ethical

committee attached)

Ethical clearance: Ethical clearance was granted by the ethical committee formed

by KLE University, Belgaum. (Ref: KLEU/08-09/D-10502)

Sampling design: Simple random sampling was used with lottery method for

allocation of subjects to seven groups.

Subjects with TTH were enrolled in the study. Subjects who did not

consent and who did not meet the eligibility criteria were excluded from the study.

The rest of the subjects were randomized using the lottery method for allocation.

Allocation procedure: Chits numbered one to seven were placed in a bowl and

the subjects were asked to pick the chits. Subjects with the following chit numbers

were allocated to the corresponding groups:

1 : EMG auditory biofeedback (EMGa) group

2 : EMG visual biofeedback (EMGv) group

3 : EMG auditory +visual (EMGav) group

Methodology

29

4 : GSR auditory (GSRa) group

5 : GSR visual (GSRv) group

6 : GSR auditory +visual (GSRav) group

7 : Control group

Sample size: Sample size was calculated using the following formula:

N=2(Zα+Zβ)2×pq/(p1-p2)

2

Probability of Type I error was set at 0.05

Power of the study was set at 80% (0.8)

p1= 1.0 and p2=0.75 were the mean differences of pre and post (baseline to one

year) average frequency of headache per month in the EMG biofeedback training

group and pain management group respectively from a study by Mullay et al

2009117

.

p=0.875 was calculated as (p1+p2)/2 and q=0.125 was calculated as 1-p.

The sample size thus calculated was 26.6 per group. To accommodate for drop

outs the sample size was chosen as 30 per group.

STUDY POPULATION

Inclusion criteria: Subjects included in the study were:

• Subjects with headache fulfilling the criteria for TTH laid down by

International Headache Society.20

• Both males and females between 18 to 65 years.

Exclusion criteria: Subjects excluded from the study were:

• Subjects who underwent complementary alternative medicine

interventions in the past 6 months.

Methodology

30

• Subjects with other headache types as described in International

Classification of Headache Disorders (ICHD) - II classification.

• Subjects with the presence of more than one type of headache in

addition to tension type headache.

• Subjects whose headache began after the age of 50 years.

• Subjects with serious somatic or psychiatric disease.

• Subjects with history of drug abuse or use of analgesics and triptans

>10 days per month.

Intervention: After allocation of subjects to the seven groups, all subjects were

informed about the treatment procedure in detail. Biofeedback training was given

in an isolated room in KLES College of Physiotherapy research laboratory, which

had minimal lighting and external noise, to facilitate relaxation. All subjects

underwent respective (EMG/GSR) BF training for 30 minutes per session for 15

sessions. Subjects underwent 15 biofeedback sessions with one session per day. If

the subject missed a session, the biofeedback session was provided when the

subject reported for therapy again, avoiding interval more than two days between

the sessions to avoid unlearning and deconditioning. The EMG BF was provided

using EMG-IR Retrainer (Chattanooga group Inc, U.S.A.) and GSR BF was

provided by GSR biofeedback Biotrainer GPF-2000 (Biotech, India).

EMG BF machine provided auditory and visual feedback. Auditory

feedback was in the form of clicks which increased in frequency and became a

continuous sound with increase in frontalis muscle tension and to no sound with

relaxation of frontalis muscle. Visual feedback on the display monitor was in the

form of glowing bars along with a numerical display which displayed the relative

Methodology

31

EMG activity of frontalis muscle in figures. The number of glowing bars was

directly proportional to tension in the frontalis muscle.

GSR BF machine similarly provided visual and auditory feedback. Visual

display was in the form of glowing bars and numerical display of real time skin

resistance in kilo-Ohms. The increase in number of red glowing bars depicted

increase in tension (fall in skin resistance) and decrease in the number of red bars

and increase in number of green bars indicated decrease in stress or tension

(increase in skin resistance). Auditory feedback was similar to EMG i.e. in the

form of clicks which became a continuous noise with increase in stress and to no

sound with relaxation. The training was given at 2% sensitivity with actual GSR.

No changes in sensitivity were made throughout the training sessions.

Skin preparation was done prior to attachment of electrodes by cleaning

the skin using spirit soaked cotton pad. After skin preparation surface EMG

electrodes (Ag-AgCl, triode electrodes) were applied 2.5cm above the centre of

each eyebrow118

and the GSR electrodes were applied on the middle phalanx of

the index and ring finger.89

Both EMG and GSR BF electrodes were placed on all

the subjects including the control group irrespective to which group they belonged

or what training they were to get. The subjects were unaware to whether they were

receiving EMG BF or GSR BF. The investigator was aware of the group the

subject belonged to and instructed the subject accordingly.

Both EMG and GSR BF auditory groups received only respective auditory

feedback. The subjects were instructed to reduce the tone and frequency of the

sound which would help them achieve relaxation. During the session the monitor

Methodology

32

on which the visual display was present was moved away from the field of vision

of the subject.

Similarly, both EMG and GSR BF visual groups were instructed to reduce

the number of glowing bars. In case of GSR, to decrease the number of red bars

and increase number of green bars to indicate relaxation. During the treatment

session the volume of the auditory tone was muted.

Both EMG and GSR audiovisual groups were instructed to reduce the tone

and frequency of sound as well as decrease the number of bars simultaneously.

Subjects in the control group were not asked to manipulate either the

visual or auditory display. They were only informed that their stress levels were

being recorded through the machines.

The subjects were instructed to practice relaxation at home, both during

the course of therapy and at the end of 15 sessions, in a way similar to the

relaxation during the biofeedback therapy sessions. However, compliance of the

subjects in the home program was not monitored. All subjects were allowed to

take the medications prescribed by their treating physicians especially if they were

preventive/prophylactic medications. They were requested to avoid taking any

analgesic / abortive / palliative medication unless the headache was unbearable.

Outcome parameters:

Primary variables:

As per the recommendations of the American Headache Society

Behavioral Clinical Trials Workgroup, 2005, 88

the primary variables selected for

our study were:

Average frequency of headache per week.

Methodology

33

Average duration of headache per week.

Average intensity of headache per week.

The recommendation for headache frequency reporting is consistent with

the IHS guidelines for controlled trials of drug treatments100, 101

and will facilitate

meta-analyses and other comparisons across studies of various interventions.102, 103

Secondary variables: The secondary variables considered in our study were

SF-36 Quality of Life – total, physical and mental scores.

Analgesic consumption.

These secondary variables are also termed as “secondary non headache

measures” by the American Headache Society Behavioral Clinical Trials

Workgroup, 2005.88

Assessment of outcome variables:

The demographic data in regards to age, gender and chronicity of headache

was collected from all the subjects in the trial.

Pain Diary119: As per the guidelines of the American Headache

Society Behavioral Clinical Trials Workgroup, 200588

, a pain diary was given to

all the subjects in which they were asked to note down the headache episodes,

duration and intensity of headache they experienced in a week. At the end of the

week, the averages of the headaches in that week were calculated. The variables

were recorded as:

1. Average frequency of headache per week: number of headache

episodes per week.

2. Average duration of headache per week: total hours of all episodes

of headache that week divided by the number of episodes in that week.

Methodology

34

3. Average intensity of headache per week: average of the ten-point

visual analogue score (VAS) per headache that week.

The subjects were also requested to note down the use of analgesics during

any of the pain episodes.

To assess the secondary variables, a licensed SF-36 questionnaire was

procured from Quality Metric Incorporated, USA in the regional language

(Kannada) and in English. It is a multi-purpose, short-form health survey with 36

questions and yields a psychometrically based physical and mental health

summary measures and a total score. The SF-36 was judged the most widely

evaluated generic patient assessed health outcome.120

Its reliability has been

estimated using both internal consistency and test-retest methods. Most of the

published reliability statistics have exceeded 0.80. Validity studies generally

support the intended objective of high and low SF-36 scores as documented in the

original user’s manuals.112,111,121

Analgesic consumption was recorded from the pain dairy of the subjects as

well as from the prescriptions of medications given to the subjects by their

treating physician.

Data collection:

All data was collected at the following time measures:

Baseline: scores of primary and secondary variables the week prior to

the start of the treatment.

Scores at one, three, six months and one year were the scores of the

last week of the corresponding month.

Methodology

35

Statistical Analysis

SPSS version 16 was used for analysis of data. Repeated measures analysis

of variance (RMANOVA) was performed since the data was collected over

various time measures. Data spherecity was checked using Mauchly test and when

significant differences were found, data was corrected using Greenhouse Geisser

correction. Post hoc analysis was performed with modified Bonferroni correction

to find the point of significance for intragroup comparison and intergroup

comparison. Consumption of analgesics over one year was analysed using

Kruskall Wallis test and when significant differences were found Wilcoxson’s

signed rank test was done to find the point of significance. Partial eta squared

value (ηp2) was calculated for all primary and secondary variables to find the

effect size at one year inter group analysis. Percentage improvement was

calculated by subtracting the monthly scores from the baseline and dividing it by

baseline scores. This end product was then converted to percentage by multiplying

it by 100. Significance for the results was set at p< 0.05.

Results

7

Results

36

RESULTS

A total of 232 subjects were enrolled in the study. Twenty-one subjects

were excluded because they did not meet the eligibility criteria or did not consent

to be a part of the study or quoted as the place and timing of treatment being

inconvenient to them. 211 subjects were randomized using lottery method for

allocation as described in the methodology. Nineteen subjects failed to complete

the treatment (thirteen subjects did not report for their treatment schedules, four

subjects switched over to other complementary therapies, two reported

inconvenient place for treatment). There were five dropouts in the first follow-up

(one month), of which two developed another health problem not related to

headache and three switched to alternative medicine. Additional ten were lost to

follow-up in the second follow-up at three months, of which four did not report

for follow-up, two subjects had a change of residence and four subjects switched

to alternative medicine. We lost ten subjects in the third follow-upat six months,

of these three subjects could not be contacted due to change in phone number, five

switched to alternative medicine, one subject developed a health problem and one

failed to report for follow up. Further sixteen subjects were lost to follow-up in the

fourth follow-up at one year, of whom seven subjects could not be contacted due

to change in phone numbers and residence; three subjects reported other health

problems and six subjects switched to alternative medicine. Overall in the study

period, eighteen subjects switched to alternative therapy (Homeopathy, Ayurveda,

Accupuncture, etc) at various follow up periods. At one month (3 subjects;

EMGv=2, EMGav=1), at three months (4 subjects; EMGav=2, GSRa=1,

Results

37

control=1), at 6 months (5 subjects; EMGa=1, EMGv=1, GSRav=1, GSRv=2) and

at one year (6 subjects; EMGa=1, EMGav=1, GSRav=2 and Control=2).

Eventually, 151 subjects were analysed (Figure 1). In total, sixty subjects

dropped out from the initially recruited and randomised 211 subjects with 28.4%

drop out rate, which was more than the acceptable norms of 20% for randomised

controlled trials. However, anticipating a large dropout in the longitudinal study,

this limitation of dropouts was accommodated by increasing the sample size by 21

subjects more than the total calculated sample size.

Results

38

Figure 1: Flow chart of subjects through the trial

232 subjects assessed for eligibility Excluded n= 21, Not

meeting inclusion Criteria (n=9)

Declined to participate (n=7), Inconvenient

timings and place (n= 5)

Allocated to EMGa (n=30) Failed to receive treatment (n=3)

Allocated to EMGv (n=30) Failed to receive treatment (n=2)

Allocated to EMGav (n=30) Failed to receive treatment (n=3)

Group EMG,n= 90, Electromyography Biofeedback + Progressive relaxation

Group GSR, n= 90, Galvanic skin resistance Biofeedback + Progressive relaxation

Group C, Control group, n= 31 Progressive relaxation only

Allocated to GSRa (n=30) Failed to receive treatment (n=4)

Allocated to GSRv (n=30) Failed to receive treatment (n=1)

Allocated to GSRav (n=30) Failed to receive treatment (n=2)

211 subjects randomised

Allocated to Control (n=31) Failed to receive treatment (n=4)

(n=27) (n=26),Lost to follow up (n=2) (failed to report for follow-up)

(n=26), Lost to follow up (n=1) (failed to report for follow-up)

(n=26)


(n=28)


(n=25)Lost to follow up (n=2) (failed to report for follow-up)






(n=25)Lost to follow up (n=1) failed to report for follow-up)

(n=25)

n=24) Lost to follow up (n=1)

(n=23) Lost to follow up (n=2)












Analysed n=23 Analysed n=21 Analysed n=22 Analysed n=21 Analysed n=20 Analysed n=22 Analysed n=22

Eval

uate

d at

6

mon

ths

Eval

uate

d at

1

mon

th

Eval

uate

d at

1

yea

r Ev

alua

ted

at

3 m

onth

s En

rollm

ent

Allo

catio

n

Results

39

Demographic data:

Demographic data in terms of age, gender, chronicity of headache and education level was collected from the subjects in the trial.

There was no significant difference between any of the groups in regards to age and chronicity of headache (p>0.05). Number of females

in all groups was more in comparison to males (Table 1). Education level too showed an insignificant difference between groups

(p=0.49).

Table 1: Demographic Data [Mean (95%CI)]

Variables EMGa

n=27

EMGv

n=28

EMGav

n=27

GSRa

n=26

GSRv

n=29

GSRav

n=28

Control

n=27 P

Age (years) 36.7

(32.2-39.3)

39.2

(36.2-44.1)

37.4

(33.5-41.3)

35.2

(30.1-36.2)

38.1

(32.4-40.3)

35.6

(32.5-38.7)

37.3

(33.5-41) 0.43

Chronicity of

headache

(years)

11.9

(8.2-13.9)

13.9

(12.7-15.0)

12.5

(9.6-14.45)

14.2

(11.7-16.6)

11.7

(9.2-14.13)

10.9

(7.4-14.3)

11.2

(8.6-15.3) 0.55

Gender

no. females(%) 22(81.4) 22(78.5) 19(70.3) 21(80.7) 20(72.4) 21(75) 20(74) --

Results

40

Baseline variables:

I] Pain variables:

Average frequency of headache per week was lower inGSRv (4.6+2.3)

than all the other groups. At baseline itself there were significant

differences between EMGv (p=0.04) and EMGav (p=0.05) when

compared to GSRv. Significant difference was found in average duration

of headache per week between EMGa and control (p=0.01) at baseline.

There were significant differences noticed in average intensity of headache

per week between EMGv v/s GSRav (p=0.01) and EMGv v/s control

(p=0.007) at baseline.(Figure 2, Table 2)

Figure 2: Means of pain variables at baseline.

Results

41

Table 2: Baseline pain variables in the study group

Comparison of

variables at baseline p value (Mean difference)

Pain variables

Average

frequency

of headache

Average

duration

of headache

Average

intensity

of headache

EMGa vs EMGv 1.0(-.93) 1.0(2.6) 1.0(-.60)

EMGav 1.0(-.63) 1.0(.56) 1.0(-0.6)

GSRa 1.0(1.2) 1.0(2.3) 1.0(.76)

GSRv 0.12(2.5) 1.0(1.3) 1.0(.06)

GSRav 1.0(1.0) 0.38(3.6) 1.0(1.2)

Control 1.0(.13) 0.01(4.1) 0.40(.93)

EMGv vs EMGav 1.0(.30) 1.0(-2.0) 1.0(.53)

GSRa 0.19(2.1) 1.0(-.33) 0.08(1.3)

GSRv 0.04(2.0) 1.0(-1.3) 1.0(.66)

GSRav 0.30(1.2) 1.0(-1.0) 0.01(1.8)

Control 1.0(1.0) 1.0(1.5) 0.007(1.5)

EMGav vs GSRa 0.76(1.8) 1.0(1.7) 0.86(.83)

GSRv 0.05(1.9) 1.0(.76) 1.0(.13)

GSRav 0.12(1.7) 1.0(3.1) 0.36(1.2)

Control 1.0(.76) 0.56(3.5) 0.44(1.0)

GSRa vs GSRv 1.0(.70) 1.0(-.96) 1.0(-.70)

GSRav 1.0(-.13) 1.0(1.3) 1.0(.43)

Control 1.0(-1.0) 1.0(1.8) 1.0(.16)

GSRv vs GSRav 1.0(-.83) 1.0(2.3) 0.75(1.1)

Control 0.07(-1.7) 1.0(2.8) 1.0(.86)

GSRav vs Control 1.0(-.93) 1.0(.46) 1.0(-.26)

Results

42

II] SF-36 scores

Significant differences were found in total SF-36 scores at baseline when

pair wise comparisons were made between GSRa v/s EMGv (p=0.05),

EMGav (p=0.03), GSRv (p=0.007) and GSRv v/s control (p=0.02).No

significant differences were found in the physical and mental scores of SF-

36 between any comparisons; therefore in this regard all groups were

homogenous. (Figure 3, Table 3)

Figure 3: Means of SF-36 scores at baseline

Results

43

Table 3: Baseline SF-36 variables in the study group

Comparison of

variables

at Baseline

p value (Mean difference)

SF-36 Total score Physical score Mental score

EMGa vs EMGv 1.0(-2.1) 1.0(-5.2) 1.0(-5.1)

EMGav 1.0(-2.4) 1.0(-1.6) 1.0(-2.6)

GSRa 1.0(5.9) 1.0(3.2) 1.0(2.2)

GSRv 1.0(-3.4) 1.0(1.7) 1.0(-3.9)

GSRav 1.0(-.26) 1.0(.96) 1.0(-1.8)

Control 1.0(5.5) 1.0(4.4) 1.0(2.9)

EMGv vs EMGav 1.0(-.3) 1.0(3.6) 1.0(2.4)

GSRa 0.05(8.0) 0.27(8.4) 0.22(7.3)

GSRv 1.0(-1.3) 0.55(6.9) 1.0(1.1)

GSRav 1.0(1.8) 0.50(6.1) 1.0(3.2)

Control 0.46(7.6) 0.06(9.6) 0.15(8.0)

EMGav vs GSRa 0.03(8.3) 1.0(4.8) 0.46(4.9)

GSRv 1.0(-1.0) 1.0(3.3) 1.0(-1.3)

GSRav 1.0(2.1) 1.0(2.5) 1.0(.82)

Control 0.12(7.9) 0.25(6.0) 0.78(5.5)

GSRa vs GSRv 0.007(-9.3) 1.0(-1.5) 0.31(-6.2)

GSRav 0.14(-6.1) 1.0(-2.2) 1.0(-4.1)

Control 1.0(-.36) 1.0(-1.1) 1.0(6.4)

GSRv vs GSRav 1.0(3.2) 1.0(-.73) 1.0(2.1)

Control 0.02(9.0) 1.0(2.7) 0.16(6.8)

GSRav vs Control 0.45(5.8) 1.0(3.4) 0.71(4.7)

Results

44

III] Analgesic consumption:

Kruskall Wallis H test found no significant differences between groups at baseline on analgesic consumption

[H(6,N=151)=5.36,p=0.49]. (Table 4)

Table 4: Consumption of analgesics. [n(percent)analgesic usage, Mean rank]

p value calculated by Kruskall Wallis test

Groups Baseline 1 month 3 months 6 months 1 year p value

EMGa(n=27) 23(100), 54.5 20(75), 44 15(60), 37.7 15(53.3), 34.9 11(45.4), 31.5 0.003

EMGv (n=28) 24(95.8), 56.6 20(70), 45.4 17(58.8), 40.5 14(42.8), 33.6 12(41.6), 33.1 0.002

EMGav(n=27) 24(87.5), 58.8 21(90.4), 60.3 23(60.8), 45.8 18(50), 40.5 12(33.3), 32.3 0.001

GSRa(n=26) 23(91.3), 53.5 20(85), 50.6 19(68.4), 47.9 15(40), 36.2 14(50), 34.7 0.01

GSRv(n=29) 20(100), 40.5 14(85.7), 35.8 12(66.6), 29.6 10(60), 27.5 9(44.4), 22.4 0.007

GSRav(n=28) 22(95.4), 50.3 18(100), 40.5 15(85.7), 37.8 14(71.4), 36.6 12(50), 33.7 0.08

Control (n=27) 17(82.3), 42.2 16(68.7), 38.1 14(57.1), 33.1 12(83.3), 24.6 10(60), 30.9 0.04

p value 0.49 0.71 0.99 0.04 0.98

Results

45

Repeated Measures Analysis of variance (RMANOVA):

Primary variables:

1. Average frequency of headache/week:

Intra group: All groups showed a significant reduction in average

frequency of headache/week at the end of one year except control group

[F(1,21)=2.98, p=0.06, ηp2=0.09] (Table 5, Figure 4). A post-hoc analysis

to find the point of significance, revealed significant differences after six

months of intervention in all EMG groups (Table 6), whereas GSRv and

GSRav found reductions in the first and third month itself, the reduction

then plateaued thereafter up to one year. GSRa showed a difference only at

one year (p=0.01) (Table 7).

Inter group: ANOVA findings on intergroup analysis showed a

significant difference between all groups at one month [F(1,11)=610,

p=0.001, ηp2=0.98], three months [F(1,8)=157,p=0.001, ηp

2=0.95],

six months [F(1,5)=57.6,p=0.001,ηp2=0.92] and one year

[F(1,2)=61.1,p=0.01, ηp2=0.96] (Table 5). A post-hoc showed the point of

significance between EMGav and GSRv (p=0.04, 95%CI:-1.2-6.8) at one

month (Table 9), between EMGa v/s control (p=0.02, 95%CI:-4.2-6.8),and

EMGavv/s GSRv (p=0.05, 95%CI:3.9-8.4) at three months (Table 10),

EMGv v/s EMGav (p=0.05, 95%CI:-5.8-0.4) and control (p=0.05,

95%CI:-4.2-1.4), EMGav v/s control (p=0.02, 95%CI:-2.2-4.5) and

GSRav v/s control (p=0.05, 95%CI:-5.3-1.2) in the sixth month (Table 11)

and between five pairs with EMGav showing most of the differences at

one year. (Table 12)

Results

46

Table 5: Intra and Inter group analysis of pain variables of all groups. Mean+SD(95%CI)

Variables EMGa EMGv EMGav GSRa GSRv GSRav Control p value

Frequency Baseline 7.1± 2.6(5.9-8.2) 7.6±1.6(6.8-8.4) 7.5±3.2(6.1-9) 5.4±2.8(4.1-6.7) 4.6±2.3(3.5-5.7) 5.9±2.4(4.8-6.9) 6.2±2.1(5.2-7.1) 0.01

1 month 6± 2.4 (4.9-7) 6.2±2.9(4.8-7.5) 6.5±2.8(5.2-7.7) 5.1±2.5(3.9-6.2) 3.1±1.6(2.3-3.9) 4.4±2(3.5-5.3) 5.4±1.6(4.7-6.2) 0.001

3months 5.3± 3.1 (3.9-6.7) 6.2±3.8(4.3-8) 5.8±2.7(4.6-7) 3.8±1.3(3.2-4.4) 3.7±0.9(2.2-3.1) 3.8±1.9(2.9-4.7) 5±2.6(3.8-6.2) 0.001

6months 3.7±1.7 (2.9-4.4) 4.6±2.1(3.6-5.6) 4.5±2.8(3.2-5.7) 4.1±2.2(3.1-5.2) 3.7±2.2(2.6-4.7) 3.9±2.4(2.8-4.9) 5.3±2.4(4.2-6.4) 0.001

1 year 3±2.2 (2.1-4) 3.5±1.5(2.8-4.2) 2.9±1.7(0.5-2) 2.7±1.5(2-3.4) 2.9±1.9(2-3.8) 2.6±1.5(0.9-2.3) 4.6±2.4(3.5-5.7) 0.01

p value 0.005 0.04 0.003 0.006 0.004 0.02 0.06

Duration Baseline 15.1±5.4(12.8-7.5) 11.4±7.1(8-14.7) 13.2±6.4(10.4-16.1) 12.3±6.7(9.3-15.4) 14.2±6.3(11.2-17.1) 11±6.6(8.1-14) 10.2±4.7(8.1-12.3) 0.03

1 month 11.6±6(9-14.2) 8±4.5(5.8-10.1) 9.2±4.8(7-11.3) 8.6±5.5(6.1-11.1) 9.9±2.8(5.5-8.2) 7.5±5.2(5.1-9.8) 7.4±3.4(5.8-8.9) 0.001

3months 8.1±4.7(6-10.2) 6.7±3.2(5.1-8.2) 7.5±3.8(5.8-9.2) 5.6±3.7(3.9-7.3) 5.1±3(3.6-6.5) 5±3.1(3.6-6.4) 5.3±2.8(4.1-6.6) 0.001

6months 5.3±3.1(3.9-6.6) 5.2±2.5(4-6.4) 5.5±2.7(4.3-6.6) 5±3.5(3.4-6.6) 6.9±5.3(4.4-9.4) 4.9±3(3.5-6.2) 6.5±3.4(4.9-8) 0.001

1 year 3.6±2.4(2.5-4.6) 4.4±2.2(3.3-5.4) 4.2±1.6(0.5-1.9) 3.3±1.7(2.5-4) 5±2(2.1-3.9) 3±1.8(1.2-2.9) 5.2±3.3(3.7-6.7) 0.01

p value 0.005 0.004 0.002 0.006 0.005 0.003 0.002

Intensity Baseline 6.9±1.6(6.2-7.6) 7.5±1.3(6.8-8.1) 6.7±1.4(6.1-7.3) 5.9±2(5-6.8) 6.4±2.1(5.4-7.4) 5.5±2.3(4.5-6.6) 5.8±1.9(5-6.7) 0.02

1 month 5.7±1.1(5.2-6.2) 5±1.2(4.4-5.6) 4.7±2(3.7-5.6) 4.4±1.8(3.6-5.3) 3.2±1(2.7-3.6) 4±1.8(3.2-4.8) 4.2±1.6(3.5-5) 0.001

3months 3.6±1.8(2.8-4.3) 3.3±0.9(2.8-3.7) 3.7±1.7(2.9-4.5) 3.9±1.2(3.3-4.4) 2.6±1(2-3.1) 3.3±1.9(2.4-4.1) 3.3±1.4(2.6-3.9) 0.001

6months 3.1±1.4(2.5-3.8) 3.3±1(2.8-3.7) 2.7±1.6(2-3.5) 2.7±1.2(2.1-3.2) 2.6±1.9(1.7-3.5) 2.4±1.7(1.6-3.2) 3.4±1.4(2.7-4) 0.001

1 year 1.9±1.3(1.3-2.5) 2.5±1.4(1.8-3.1) 1.9±1.1(0.3-1.3) 2.2±1.3(1.6-2.8) 1.9±1.2(1.3-2.4) 2.2±1.2(0.8-1.9) 3.1±1.4(2.5-3.7) 0.001

p value 0.006 0.01 0.01 0.01 0.03 0.002 0.05

Results

47

Figure 4: Trend of average frequency of headache through the time measures.

Results

48

Table 6: Intra group comparison of pain variables in EMG groups

p value (Mean difference)

Pain variables

Average

frequency of

headache

Average

duration of

headache

Average

intensity of

headache

EMGa baseline v/s 1 month .16(1.1) .001 (3.1) .04(1.2)

3 months .20(1.7) .00 (7.0) .00(3.3)

6 months .00(3.3) .00(9.8) .00(3.7)

1 year .00(4.0) .00(11.9) .00(5.0)

EMGa1 month v/s 3 months 1.0(.65) .01(3.5) .00(2.1)

6 months .001(2.2) .00(6.3) .00(2.5)

1 year .001(2.9) .00(8.0) .00(3.8)

EMGa 3 months v/s 6 months .006(1.6) .00(2.8) 1.0(.43)

1 year .006(2.2) .001(4.5) .007(1.6)

EMGa 6 months v/s 1 year 1.0(.65) .14(1.6) .02(1.2)

EMGv baseline v/s 1 month .20(1.4) .10(3.4) .00

comparative study on electromyography (emg) biofeedback...

Documents