Abstract
Background Cervicogenic headache and migraine are common causes of visits to physicians and physical therapists. Few randomized trials utilizing active physical therapy and progressive therapeutic exercise have been previously published. The existing evidence on active treatment methods supports a moderate effect on cervicogenic headache.
Objective The aim of this study is to investigate whether a progressive, group-based therapeutic exercise program decreases the intensity and frequency of chronic headache among women compared with a control group receiving a sham dose of transcutaneous electrical nerve stimulation (TENS) and stretching exercises.
Design A randomized controlled trial with 6-month intervention and follow-up was developed. The participants were randomly assigned to either a treatment group or a control group.
Setting The study is being conducted at 2 study centers.
Patients The participants are women aged 18 to 60 years with chronic cervicogenic headache or migraine.
Intervention The treatment group's exercise program consisted of 6 progressive therapeutic exercise modules, including proprioceptive low-load progressive craniocervical and cervical exercises and high-load exercises for the neck muscles. The participants in the control group received 6 individually performed sham TENS treatment sessions.
Measurements The primary outcome is the intensity of headache. The secondary outcomes are changes in frequency and duration of headache, neck muscle strength, neck and shoulder flexibility, impact of headache on daily life, neck disability, fear-avoidance beliefs, work ability, and quality of life. Between-group differences will be analyzed separately at 6, 12, and 24 months with generalized linear mixed models. In the case of count data (eg, frequency of headache), Poisson or negative binomial regression will be used.
Limitations The therapists are not blinded.
Conclusions The effects of specific therapeutic exercises on frequency, intensity, and duration of chronic headache and migraine will be reported.
Headache is among the most common reasons for visits to primary care physicians and physical therapists among working women.1,2 Headaches affect almost half of the adult population worldwide.3–5 The estimated costs of migraine alone are €27 billion in Europe.5
Different types of headache often coexist, such as cervicogenic headache, tension-type headache, and migraine. The pathogenesis of headache is not yet unambiguously clear. Several explanatory factors have been proposed, including disorders in neural connections, soft tissues, bony structures, and body posture.2,6 Patients with cervicogenic headache typically also have neck symptoms, and physical therapy is the primary treatment.7 Patients with tension-type headache and migraine often receive medication as the first choice, supported by physical therapy.5,6
At present, there is scarce scientific evidence on various therapeutic modalities in patients with chronic headache and migraine.8 Specifically, more research is needed to determine which physical therapy approaches are effective.8 For patients with cervicogenic headache and tension-type headache, several components of physical therapy have been used, including ultrasound, transcutaneous electrical nerve stimulation (TENS), stretching, relaxation with active kinetic control movements, and manual treatment.7–12 On the other hand, aerobic training, relaxation therapy, and biofeedback have mainly been applied for patients with migraine.13,14 However, few randomized controlled trials (RCTs) have included large numbers of patients, particularly outpatients, and included a single-component approach such as active and progressive therapeutic exercise.2,15,16 The principles of therapeutic exercise in successfully treating patients with chronic headache consist of low-load and low-intensity proprioceptive training at the first phase and specific high-load and high-intensity exercises in the later phase of the program, or the former or the latter throughout the treatment period.2,10,14,17
In our 6-month RCT, we use a single-component progressive therapeutic exercise program among working women with chronic headache (cervicogenic headache or tension-type headache, with or without migraine). We study whether the program is effective in comparison with a control group receiving a placebo dose of TENS treatment. Our main hypothesis is that the specific therapeutic exercise training program will alleviate the intensity, frequency, and duration of headache. We also study how the therapeutic exercise program influences the function of the neck, the individual's quality of life, and the use of medication and other treatments.
Method
Study Design
A 6-month, 2-center RCT with 6-, 12-, and 24-month follow-up periods after the baseline was developed.
Eligibility Criteria
Women with ongoing chronic cervicogenic headache, tension-type headache, or migraine were recruited using newspaper advertising, invitations on the homepages of the 2 study centers, and advertisements placed with the occupational health services in the cities of Tampere and Helsinki.
Eligibility was assessed in 3 phases during the screening process. Women who were interested in participating in the study phoned or emailed a contact person at the Tampere study center (UKK Institute for Health Promotion Research, Tampere, Finland). In this first phase of screening, women were interviewed over the phone to assess their eligibility by a trained research nurse or physical therapist.
The inclusion criteria were: (1) women aged 18 to 60 years having chronic cervicogenic headache (with or without migraine) at least on 8 days during 4 consecutive weeks at baseline, (2) the candidates were currently working and aimed to continue working at the same place for at least 2 more years, (3) the intensity of the candidate's headache had been 40 mm or more (scale of 0–100) on a visual analogue scale (VAS)18,19 during the previous 2 months, and (4) the candidate's Headache Impact Test (HIT-6 Scoring Interpretation Finnish Version 1.1) indicated 56 points or more (scale of 36–78).20
The exclusion criteria were: (1) verified severe degenerative changes in the cervical vertebral or intervertebral disks, including disk prolapse; (2) a whiplash injury or similar type of trauma, which is a contraindication for specific exercise training; (3) treatment with manual therapy or physical therapy because of headache within the month prior to the baseline measurement; (4) a change in medication known to affect headaches during the month prior to the baseline measurement; (5) changed bifocals during the month prior to the baseline measurement; (6) habitual physical activity at least 3 times or more weekly; (7) alcohol abuse (questions from the Alcohol Use Disorders Identification Test [AUDIT], Finnish version, <6 points)21; and (8) depressive symptoms according to the Beck Depression Inventory (BDI-21, >19 of 63 points).22
Recruitment Procedure
After the telephone screening, potential participants received an information letter about the trial and an invitation to the second phase of the screening process. This phase began with a structured examination by the study's general practitioner. The examination included a review of the patient's general health status, medication, and earlier verified diagnoses of headache (International Statistical Classification of Diseases and Related Health Problems, 10th revision23 [ICD-10]) and a clinical examination to exclude participants with dysfunctions according to exclusion criteria or verified severe degenerative changes in the cervical spine. Then the physical therapist gave the participants instructions on the use of headache diaries (over a 4-week period) and 1-week physical activity monitoring using a 3-dimensional (3D) accelerometer (Hookie AM20, Traxmeet Ltd, Espoo, Finland) to determine the intensity of the headache and to evaluate physical activity at baseline.
Patient eligibility was confirmed in the third phase of the screening process, which was 4 weeks after the second screening appointment. The headache diaries were checked by the physical therapist. If a headache intensity of 40 mm or more was achieved on 8 days or more in the 4-week evaluation period and the potential participant gave her written informed consent, she was included in the 6-month trial.
Women who had the above-defined chronic cervicogenic headache, tension-type headache, or migraine were then randomly assigned to either a therapeutic exercise training group (THEX group) or a control group that received TENS. The active intervention period lasted 6 months and included measurements at baseline and after the 6-month intervention. Follow-up assessment was performed at 12 months after the baseline measurements, and 24-month follow-up assessment is ongoing. The study flow chart is presented in the Figure.
Flowchart of the study. THEX=therapeutic exercise, TENS=transcutaneous electrical nerve stimulation.
Outcome Measures
Table 1 shows the outcome measures, methods used, and time points of the measurements.
Outcome Variables and Methods and Schedule of the Assessmentsa
Primary outcome.
During the 6-month training period, all participants were requested to report each episode of headache every day, including its intensity, frequency, and duration (the time when the headache began and ended), using a diary. The intensity of headache was assessed using the VAS (0–100).18,19 The duration of headaches was calculated as weekly minutes.
Secondary outcomes.
The Headache Impact Test-6 (HIT-6),20 which consists of 6 items (pain, social functioning, role functioning, vitality, cognitive functioning, and psychological distress), was used to assess the impact of headache on daily life. The other secondary outcome measures were validated questionnaires (in Finnish): Neck Disability Index,24–26 Nordic Musculoskeletal Questionnaire27,28 with selected items for upper limb and neck complaints, Fear-Avoidance Beliefs Questionnaire (FABQ),29,30 Work Ability Index,31 a measure of quality of life (RAND 36-item Health Survey),32 Fatigue Impact Scale,33 and Beck Depression Inventory (BDI).22 In addition, the participants were asked single items about their physical activity, smoking, sleep, menstruation status, use of medication, absence from work, and the cost of headache treatments.
Questionnaires and the clinical examination tests.
At baseline, information on age, education, and occupation was collected. The measurements and the first follow-up of the intervention took place after the 6-month intervention period. In addition, a postal questionnaire, 1-week headache diary, and accelerometer measurements for 7 days were performed at the 12-month follow-up and will be collected at the 24-month follow-up point.
The clinical examination was conducted at baseline and after intervention by a physical therapist. The range of active movements in the cervical spine (flexion, rotation, and lateral bending) was measured using a Myrin goniometer (model 927034, OB Vinkelmätare Myrin, Lic Rehab, Solna, Sweden) in a standardized way with the participant in a seated position. Isometric endurance of the flexor and extensor muscles of the cervical spine was performed according to Edmondston et al.34 The isometric flexion test was performed with the participant lying in a supine position on a plinth. The participant's head was in a neutral position and supported by physical therapist. The participant was then asked to slightly flex her upper neck and lift her head up from the therapist's hands while retaining upper neck flexion at the same time. Verbal feedback (“hold your head up”) was given to the participant if her head declined during the test. The test was terminated if the participant was unable to maintain the position of her head after she had been asked to maintain the stable head position twice. The hold time was measured in seconds.
The neck extensor test was performed according to Edmondston et al34 but without the extra weight. The participant was lying on a plinth in a prone position with her arms at her side with her head, supported by the physical therapist, over the plinth in a neutral sagittal-plane position. The participant then was asked to hold her cervical spine horizontal. The test was terminated if the participant was unable to maintain the position of her head after she had been asked to maintain the stable position twice. The hold time was measured in seconds.
The mobility of the shoulder and neck region was measured in a standing position with the participant's back against a wall. The participant was asked to lift her arms straight over her head to full flexion so that the back of her hands touched the wall. The classification was done separately for both arms according to the restrictions in the range of motion using a 3-point scale (0=severe restriction, 1=moderate restriction, 2=no restriction).35 Moreover, the alignment of the posture of the cervical spine was assessed as the participant stood back in a neutral upright position close to the wall. The distance from the tip of the ear to the wall was measured in centimeters.
Randomization
Randomization to treatment groups was stratified for the 2 study centers (UKK Institute for Health Promotion Research, Tampere, Finland [n=61] and Helsinki Metropolia University of Applied Sciences, Helsinki, Finland [n=61]). Block randomization with randomly selected block sizes of 2, 4, 6, and 8 was used to ensure the equality of the group sizes. A randomization schedule for each stratum was generated by a statistician who was not involved in the recruitment or treatment of the participants. The group assignments were placed in consecutively numbered sealed envelopes. As the participant was enrolled in the study, the next available envelope of her stratum was opened by the participant, and she was assigned to either the THEX group or the TENS group.
Intervention
THEX Group's Training Program
Training in the THEX group consisted of 6 progressive therapeutic exercise modules (THEX M1–M6; Tab. 2) including low-load proprioceptive craniocervical and cervical exercises progressing to exercises with specific high load for neck muscles during the 6-month intervention. The training protocol of the therapeutic exercises was based on Jull et al9 and Ylinen et al.25 However, we customized the previous training protocols and offered training in 6 different phases and modules to ensure progression of the training during the 6-month period (Tab. 2; eAppendix). The main goals of the therapeutic exercise program were (1) to improve postural and movement control of the cervical spine and shoulder region and (2) to strengthen the muscles important for body posture. Two experienced physical therapists, one at each study center, instructed and supervised the THEX program.
Phases of Therapeutic Exercise (THEX) Modules 1–6, With Specification of Execution, and Transcutaneous Electrical Nerve Stimulation (TENS) Treatmenta
Supervised training sessions (Tab. 2).
The first 2 sessions of the THEX modules (M1 and M2 for postural improvement) included individual supervision. The interval between the sessions from THEX M1 to M2 and from M2 to M3 was 2 weeks, respectively. The remaining 4 THEX modules (M3–M6) were conducted for 6 weeks each, with one supervised group session at the beginning of each module. These group sessions were organized in groups of a maximum of 10 participants to ensure that the proper technique of the new exercises was learned for each module.
In detail, 3 different exercises in THEX M1 and M2 concentrated on the postural and motor control (5- to 10-seccond contraction, 5–10 seconds of relaxation, 5–10 repetitions). Progressively, 2 additional cervical and scapular exercises aiming at postural and motor control were taught in THEX M3. In addition, muscle strengthening exercises with rubber bands (Thera-Band, yellow and red, HCM-Hygenic Corp, Perak Darul Ridzuan, Malaysia)25 were included in the program (Tab. 2, eAppendix). The aim of the rubber band exercises was eventually to achieve a load of 80% of maximal voluntary contraction (so that participants would be able to perform 10–15 repetitions during a single set). Furthermore, the control of proper upper body posture was supervised and controlled in THEX M4 to M6 (Tab. 2).
Home training.
After the supervised session at the beginning of the THEX M1 to M4 sessions, the participants were prescribed to perform the exercises at home. The exercises took place once a day, 6 days a week. Thereafter, 4 weekly home training sessions were prescribed during THEX M5 and M6, and these mainly consisted of high-load strengthening exercises (Tab. 2, eAppendix). In addition to verbal supervision for home training, all participants received all of the THEX module exercises as a printed booklet and as a video on a USB flash memory stick (SanDisk Cruzer USB2.0 Flash Drive, Shanghai, China) to allow the participants to follow the protocol and to ensure high-quality performance of the training movements. After the 6-month training period, the participants were encouraged to continue exercising at home on their own.
TENS Group's Program
The participants randomly assigned to the TENS group (control group) received 6 individual 45-minute placebo treatment sessions at same time points as the THEX group (Tab. 2), where TENS of less than 10 mA (phase width of 150 microseconds and 100-Hz constant frequency) was given by a physical therapist who was not involved with the THEX group (Tab. 2). The placement of electrodes is presented in the Appendix.
In addition, 3 supervised stretching exercises for the neck muscles were prescribed for the participants in both groups as home exercises (eAppendix). The participants in the TENS group were given a similar booklet and video on a USB flash memory stick as in the THEX group, but the booklet and video contained only the stretching exercises. Moreover, participants in both study groups were encouraged to maintain their habitual physical activity level throughout the 6-month intervention period.
Statistical Methods
Power calculations.
The sample size of the trial was estimated based on the change in headache intensity measured by VAS in the study by Ylinen et al.25 A sample size of 49 people in each group was calculated to detect a decrease of 20 mm in VAS scores in the THEX group versus no change in the TENS group at 6 months to achieve 90% power. The calculations were based on a 2-sided t test using a significance level (alpha) of .05, assuming a standard deviation of 30 mm for both groups. A dropout rate of 25% was expected because the participants were outpatients and because only 6 supervised therapeutic exercise training sessions were offered for participants in the THEX group and 6 TENS placebo treatments with stretching exercises were offered in the TENS group during the 6-month intervention. Thus, we aimed to recruit 122 participants (61 participants per study group).
Statistical analyses.
Between-group differences will be analyzed separately at 6, 12, and 24 months from baseline using generalized linear mixed models, with appropriate distribution and link functions confounding variables as the covariates. In the case of count data (eg, frequency of headache), Poisson or negative binomial regression will be used. The results will primarily be analyzed according to the intention-to-treat principle. These analyses allow the incorporation of incomplete longitudinal data into the models. The effect size will be calculated by the method of Cohen d (mean baseline scores minus mean follow-up scores, divided by the pooled standard deviation).
An effect size of 0.20 through 0.49 will be considered a small effect, 0.50 through 0.79 will be considered a medium effect, and 0.80 or more will be considered a large effect. Cost-effectiveness calculations will be based on absence from work (the number of days off from work because of headache), costs of physicians' and physical therapists' appointments, pharmaceutical claims for headache, and the adverse events of training. The analyses will be conducted by using IBM SPSS software package version 22 (IBM Corp, Armonk, New York).
Ethics
The procedures and design of the trial were approved by the Ethical Committee of Pirkanmaa Hospital District, Finland (Code R12006), and the trial was prospectively registered in ClinicalTrials.gov Protocol Registration System (identifier: NCT01664585). This study is carried out according to the guidelines of good scientific practice and provisions of the Finnish Medical Research Act (Helsinki Declaration). Any protocol modifications are reported to the Ethical Committee and to the trial registry. The study was funded by The Social Insurance Institution of Finland (36/26/2011). This funding source had no role in the design of this study and will not have any role during its execution, analyses, interpretation of the data, or decision to submit results.
Discussion
Significance of the Study
Women with chronic headache are frequent clients of public health services, and they often visit several physicians and other health care providers in order to get relief for their headache, causing a major burden to the health care system. This trial is targeted at women who have severe chronic headache. We aimed to recruit participants with headache intensity of 40 mm or more on the VAS, which means quite a disruptive intensity of headache. Moreover, we were interested in patients whose chronic headaches occur often (ie, at least on 8 days during the 4-week screening period at baseline). In addition, we included women whose headaches were classified as migraine but who also had neck pain. The reason for this decision is that sometimes the differentiation among migraine, tension-type, and cervicogenic headaches can be challenging, as patients can have several headache diagnoses.9 To date, there are only few RCTs in which the effects of specific therapeutic exercises have been studied among patients with migraine. Moreover, there is preliminary evidence that patients with mild migraine may benefit from aerobic training.14,36,37 Also, it will be interesting to look at the headache diaries throughout the 6-month intervention period. To our knowledge, there are no neck-specific therapeutic exercise training studies for patients with headache where they have kept daily headache diaries for 6 months. The diaries may reveal detailed information on the training dose and response in terms of different therapeutic exercise movements, loads, and intensities at different phases of the intervention.
Aspects for Clinical Practice in Physical Therapy
Patients with chronic headache and neck symptoms form a heterogeneous group who frequently use many types of conservative treatment. The most commonly used treatment is medication,38 even though various nonpharmaceutical treatments have been proposed to treat patients with cervicogenic headache or tension-type headache. For example, passive modes of physical therapy treatments, acupuncture, manipulative therapy, chiropractic, and homeopathy are commonly used.10,38 Active rehabilitative training, however, also can provoke headache symptoms, which might be a reason for the low number of trials to date. Therefore, patients with severe chronic headache and migraine present a major challenge for the design of the exercise training program.
The training protocol of this study is based on studies by Jull et al9 and Ylinen and colleagues.2,25 The THEX modules that we developed, utilizing an upper neck–specific exercise regimen (eAppendix), have been pilot tested with clinical patients. As a leading principle, the progression of a specific upper-neck exercise training regimen among women with cervicogenic headache was aimed to be mild. Due to the varying health status and functioning of the participants, the selected movements in the 6 different modules needed careful planning. Mild progression also was important because of the potential risk of exercise-induced headaches and migraines, which often result in decreased training motivation later on. Therefore, the first 2 THEX modules were of very low intensity in order to minimize discontinuation of training and to maximize training adherence. Moreover, the first 3 modules of the therapeutic exercises were mainly aimed at improving the participants' awareness of postural control in the craniocervical and cervicothoracic regions. In the successive modules from 4 to 6 of the progressive training program, we emphasized increasing the endurance of the neck muscles. The participants in this study are outpatients, and the procedure of recruitment to the study was close to ordinary clinical practice. From an ethical point of view, instead of traditional “experiment versus control” design, we offered the control group an equal number of therapeutic sessions, albeit with a noneffective dosage of TENS. Thus, the same amount of care was offered, although the treatment itself was different.
Strengths and Weaknesses of the Study
We expected that the participants independently committed to the training at home and followed the 6 different training modules and the progression of the program, which may result in reduced headache. Moreover, we utilized group training to enhance the cost-effectiveness of the program, but we decided to include 2 individually supervised training sessions in the beginning of the intervention to ensure the learning of the right training technique. In terms of cost-effectiveness, only 6 sessions were supervised altogether, and for rest of the time, participants trained at home on their own. This new approach was chosen to get results at a low cost. To improve training adherence at home, participants in the THEX group were carefully instructed to perform the exercises, and a video and booklet of the exercises were produced and offered to the participants to help in their training. We also collected data on the frequency, intensity, and duration of headache and on the use and dose of medications using the diaries during the 4-week period at baseline and throughout the 6-month period. Although demanding for patients, this kind of information has seldom been collected and reported. In addition, we will investigate whether any changes occur in objectively measured daily physical activity among the patients with chronic headache. Physical activity data were collected during the 6-month training period and 12-month follow-up period and will be collected at the 24-month follow-up point using a 3D accelerometer device that enables the measurement of different doses and intensities of physical activity throughout the day for 7 days. Our physical activity data among women with headache will add knowledge on levels of physical activity, on which there is little information. Furthermore, we will follow the participants up to 24 months, which is a strength of this study and gives a long-term perspective to the effectiveness of the treatments should a short-term effect be achieved.
This study design has a few limitations. The first limitation is the inability to perform blinding. A therapeutic exercise training regimen is easy to be recognized as an active treatment; therefore, the participants cannot be blinded. The physical therapists who performed the careful assessments were recruited on the basis of being the best in their field. However, a lack of resources in this multicenter study forced us to use the same physical therapy experts to perform both clinical assessments and to instruct and supervise the rehabilitative exercise training. Thus, we were unable to blind the physical therapists who conducted the clinical measurements at baseline and at 6, 12, and 24 months. Another limitation is the multicenter setting compared with a single-center study. Multicenter studies require careful monitoring to ensure that all study personnel follow the protocol exactly and that standardized assessment techniques are used in this study. A third limitation is that progressively increasing the intensity of the therapeutic exercise regimen may temporarily cause increased headache as a side effect. This and other facts formed a challenge in recruiting participants who would be willing to commit to long-term training and follow-up. Therefore, participants may drop out of the trial. However, the physical therapists gave realistic information to the participants about the type of sensations that may occur when starting the more intensive training. The participants also were allowed to seek medical or other treatments if needed during the study period.
Future Research
We are about to complete the 24-month follow-up, but we have not yet collated the data. With this trial, we hope to succeed in reducing the intensity, frequency, and duration of chronic headaches with our low-cost progressive rehabilitative training program in a clinically relevant way. Also, the dose-response relationship of specific training to reduce chronic headache should be investigated. If positive findings are observed, it would be interesting to investigate the mechanisms behind the pain relief in the future.
Appendix.
Description of Transcutaneous Electrical Nerve Stimulation (TENS) With Sham Electrode Placement Used in 6 Treatment Sessions for Control Group
Footnotes
Dr Rinne, Professor Häkkinen, Dr Ylinen, and Professor Nikander provided concept/idea/research design. All authors provided writing and data analysis. Dr Rinne, Mrs Garam, and Dr Kukkonen-Harjula provided data collection. Dr Rinne and Professor Nikander provided project management, fund procurement, and institutional liaisons. Dr Rinne provided participants. Dr Rinne, Mrs Garam, and Professor Nikander provided facilities/equipment. Dr Ylinen, Professor Häkkinen, and Professor Nikander provided consultation (including review of manuscript before submission). The authors thank Matti Pasanen, MSc, for his valuable expert help in conducting power calculations; Kari Tokola, MSc, for predetermining the statistical analyses; and Iida Siro, MSc, PT, who posed for the exercise photos.
- Received May 17, 2015.
- Accepted October 4, 2015.
- © 2016 American Physical Therapy Association
References
Issue highlights
