ORIGINAL ARTICLE

Are Manual Therapies Effective in Reducing Pain From Tension-Type Headache?

A Systematic Review
Cesar Fernandez-de-las-Penas, PT,* Cristina Alonso-Blanco, PT,* Maria Luz Cuadrado, MD, PhD,w Juan Carlos Miangolarra, MD, PhD,* Francisco J. Barriga, MD, PhD,w and Juan A. Pareja, MD, PhDw

Objectives: A systematic review was performed to establish whether manual therapies have specic ecacy in reducing pain from tension-type headache (TTH). Methods: Computerized literature searches were performed in MEDLINE, EMBASE, AMED, MANTIS, CINAHL, PEDro, and Cochrane databases. Papers were included if they described clinical (open noncontrolled studies) or randomized controlled trials in which any form of manual therapy was used for TTH, and if they were published after 1994 in the English language. The methodologic quality of the trials was assessed using the PEDro scale. Levels of scientic evidence, based on the quality and the outcomes of the studies, were established for each manual therapy: strong, moderate, limited, and inconclusive evidence. Results: Only six studies met the inclusion criteria. These trials evaluated dierent manual therapy modalities: spinal manipulation (three trials), classic massage (one trial), connective tissue manipulation (two trials), soft tissue massage (one trial), Dr. Cyriaxs vertebral mobilization (one trial), manual traction (one trial), and CV-4 craniosacral technique (one trial). Methodologic PEDro quality scores ranged from 2 to 8 points out of a theoretical maximum of 10 points (mean = 5.8 2.1). Analysis of the quality and the outcomes of all trials did not provide rigorous evidence that manual therapies have a positive eect in reducing pain from TTH: spinal manipulative therapy showed inconclusive evidence of eectiveness (level 4), whereas soft tissue techniques showed limited evidence (level 3). Conclusions: The authors found no rigorous evidence that manual therapies have a positive eect in the evolution of TTH. The most urgent need for further research is to establish the
Received for publication June 9, 2004; rst revision January 30, 2005; second revision May 13, 2005; accepted May 21, 2005. From the *Department of Physical Therapy, Occupational Therapy, Physical Medicine and Rehabilitation, Universidad Rey Juan Carlos (URJC), Alcorcon, Madrid, Spain; and wDepartments of Neurology of Fundacion Hospital Alcorcon and Medicine of the Universidad Rey Juan Carlos (URJC), Alcorcon, Madrid, Spain. Reprints: Cesar Fernandez de las Penas, Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos, Avenida de Atenas s/n, Alcorcon, Madrid, 28922 Spain (e-mail: cesarfdlp@yahoo.es or cesar.fernandez@urjc.es). Copyright r 2006 by Lippincott Williams & Wilkins

ecacy beyond placebo of the dierent manual therapies currently applied in patients with TTH. Key Words: tension-type headache, manual therapy, soft tissue techniques, spinal manipulation (Clin J Pain 2006;22:278285)

eadaches are one of the most common problems seen in medical practice. Among the many types of headache disorders, tension-type headache (TTH) is the most prevalent in adults. Population-based studies suggest 1-year prevalence rates of 38.3% for episodic TTH and 2.2% for chronic TTH.1 Recently, the second edition of the Classication of Headache Disorders of the International Headache Society (IHS) has kept the clinical criteria for the diagnosis of TTH2,3 but has withdrawn EMG or pressure algometry from the diagnostic features for subdivision, as only tenderness on manual palpation has proved useful to distinguish the dierent subtypes of TTH.3 Accordingly, the following subtypes of TTH are now considered: infrequent episodic TTH associated/not associated with pericranial tenderness; frequent episodic TTH associated/not associated with pericranial tenderness; chronic TTH associated/not associated with pericranial tenderness; and probable TTH (infrequent, frequent, or chronic TTH). Despite its scientic interest, the pathophysiology of TTH is not clearly understood. Bendtsen reported that both peripheral mechanisms (ie, myofascial tenderness of pericranial structures) and central mechanisms (ie, sensitization of supraspinal neurons and decreased antinociceptive activity from supraspinal structures) might explain some of the symptoms of TTH.4 There are many therapeutic approaches aimed at treating benign chronic and recurrent headaches such as TTH, including pharmacotherapy, cognitive therapy, relaxation therapy, biofeedback, and physical therapy.5 Headache suerers are also frequent users of complementary techniques such as manual therapies and chiropractic care.6 Vernon and McDermaid reported that upper cervical manipulation, soft tissue therapy, and myofascial trigger point treatment were the techniques
Clin J Pain  Volume 22, Number 3, March/April 2006

278

Clin J Pain

Volume 22, Number 3, March/April 2006

Manual Therapy for Tension-Type Headache

most often used by Canadian chiropractors in the management of episodic TTH.7 Although there is considerable evidence that behavioral strategies8 and pharmacologic treatment9 can be eective for headache disorders, the evidence for manual therapies is less clear. In a recent systematic review, Astin and Ernst reported that available data did not support that spinal manipulation was an eective treatment of headache.10 On the other hand, in a previous systematic review, Bronfort et al concluded that spinal manipulation seemed to be eective in the management of cervicogenic headache, but not TTH.11 In those papers only spinal manipulative therapy was evaluated, and trials included patients with dierent types of headache: migraine, cervicogenic headache, and TTH. In another systematic review, Vernon et al analyzed the eectiveness of complementary/alternative (CAM) therapies in the treatment of cervicogenic headache and TTH and concluded that some CAM might be useful in the treatment of these disorders.12 As the pathogenic mechanisms of cervicogenic headache and TTH are probably dierent, dierent therapeutic modalities might be indicated in either condition.13 Some manual therapies, such as soft tissue manipulation techniques, connective tissue treatment, transverse friction massage, and myofascial trigger point treatment, are being used in TTH even though their eectiveness in this setting has not been established. To our knowledge, a specic systematic review of manual therapies for TTH has not been published yet. To establish whether manual therapies have specic ecacy in the treatment of TTH and to update the literature on this subject, we undertook a systematic review of the eectiveness of manual therapies in the management of TTH.

facilities allowed search limits, searches were restricted to clinical or controlled trials. We also checked the reference lists of the papers that were identied in the database searches.

Article Selection
Articles chosen to go through the selection process were independently reviewed by two authors. Papers were included if they met the following inclusion criteria. First, they had to describe clinical (open noncontrolled studies) or randomized controlled trials in which any form of manual therapy (soft tissue manipulation techniques, spinal manipulation, ischemic compression technique, massage therapy, transverse friction massage, myofascial release, chiropractic manipulation, or spray and stretch technique) had been used for TTH. Comparative trials were included if at least one group received some form of manual therapy. Second, the study must have been published after 1994 in the English language. Published proceedings and abstracts were excluded.

Data Extraction
As with article selection, data from each study were extracted independently by two authors. A standardized data-extraction form that contained questions on population, interventions, methodology, results, and outcome measures was used, according to the CONSORT statement.15 For each study, the following data were taken: inclusion and exclusion criteria, design, randomization, description of dropouts and blinding, outcome measures, details of the therapeutic intervention, and results. A good interobserver reliability was veried with the kappa coecient (K = 0.79). Finally, both authors had to achieve a consensus on each item of the data-extraction form.

METHODS Data Sources

Computerized literature searches were performed for clinical/controlled trials and reviews of manual therapy in TTH, using the following databases: PubMed (from 1975), Ovid MEDLINE (from 1975), Ovid EMBASE (from 1975), the Cochrane Database of Systematic Reviews, the Cochrane Collaboration Trials Register (CCTR), AMED (Alternative Medicine), MANTIS (Manual Alternative and Natural Therapy), CINAHL (Cumulative Index to Nursing and Allied Health Literature), and PEDro (Physiotherapy Evidence Database). Medical Subject Headings (MeSH) and other key words for search were as follows: tension-type headache OR contraction headache combined with manual therapy treatment, soft tissue manipulation techniques, spinal manipulative therapy, spinal manipulation, ischemic compression technique, massage therapy, transverse friction massage, myofascial release, chiropractic manipulation, and spray and stretch technique. Search strategy followed the guidelines described by Greenhalgh.14 This search was supervised by an expert librarian scientist who helped us on each stage of the procedure. When database
r

Methodologic Quality Assessment

Many methods of quality scoring for clinical trials have been described.1618 In the current systematic review, the PEDro quality scoring method was used. This scale is based on the Delphi list and scores studies based on the presence or absence of 10 methodologic criteria19,20: random allocation, concealed allocation, baseline comparability, blinded assessors, blinded subjects, blinded therapists, adequate follow-up, intention-to-treat analysis,21 between-group comparisons, and point estimates and variability (Table 1).

Outcome Assessment
To determine the possible inuence of manual therapies on TTH, we calculated the eect size for each outcome measure that was mentioned in each study.22 If a trial included more than two manual therapy modalities, the eect size was calculated for the experimental group with respect to the reference group. The eect size was

2006 Lippincott Williams & Wilkins

279

Fernandez-de-las-Penas et al

Clin J Pain

Volume 22, Number 3, March/April 2006

TABLE 1. PEDro Score Items*

1. 2. 3. 4. 5. 6. 7. 8. Subjects were randomly allocated to groups. Allocation was concealed. The groups were similar at baseline regarding the most important prognostic indications. There was blinding of all subjects. There was blinding of all therapists who administered the therapy. There was blinding of all assessors who measured at least one key outcome. Measurements of at least one key outcome were obtained from more than 85% of the subjects initially allocated to groups. All subjects for whom outcome measurements were available received the treatment or control condition as allocated, or where this was not the case, data for at least one key outcome were analyzed by intention to treat. The results of between-groups statistical comparisons are reported for at least one key outcome. The study provides both point measurements and measurements of variability for at least one key outcome measure.

9. 10.

quality randomized controlled trials with generally consistent outcomes. 4. Inconclusive evidence: only one relevant, low-quality randomized controlled trial, no relevant randomized controlled trials or randomized controlled trials with inconsistent outcomes. A study was considered relevant when at least one of the outcome measures concerned headache intensity, headache frequency, and headache duration (headache diary). For being generally consistent, at least three fourths of the trials that analyzed the same manual therapy had to have the same result (positive, neutral, or negative). A randomized controlled trial was considered to be of high quality if the methodologic score of PEDro was greater than 5 and of low quality if the PEDro score was 5 or less.

*To apply PEDro quality scoring, eligibility criteria of the clinical trial must be specied. Total score ranges from 0 to 10.

RESULTS Selected Articles

Fifty-ve potentially relevant articles were identied. Of those 55 articles, 40% were identied on MEDLINE, 31% on CINAHL, and 29% on AMED. Eighteen papers appeared in all databases, so that 37 articles were reviewed. Thirty-one articles were subsequently excluded because no form of manual therapy was actually used. Finally, only six articles fullled all criteria for selection.2833 Two of these papers were overlapping publications, as they included the same patients and involved data from the same trial.31,32

calculated by the following formula23: ES Mean of experimental group Mean of control reference group= Standard deviation of control reference group A positive value of ES indicates a better outcome for the experimental group, and a negative value indicates a better outcome for the control (reference) group. According to Thomas and Nelson,24 an eect size greater than 0.8 is large, around 0.5 is moderate, and less than 0.2 is small. We needed the standard deviation (SD) to be included in the trials, because the formula we used was based on the SD. In those studies in which the eect size could not be calculated, an alternative method was used to formulate conclusions on the eectiveness of each intervention.25 In those cases, outcomes were dened as positive (the experimental group obtained a signicantly greater improvement than the control [reference] group, neutral (there were no signicant dierences between groups), or negative (when the control [reference] group obtained a signicantly greater improvement than the experimental group). P<0.05 was used to dene a signicant outcome measure. This method, used in previous systematic reviews,26,27 reached four levels of scientic evidence based on the quality and the outcomes of the trials: 1. Strong evidence: multiple, relevant, high-quality randomized controlled trials with generally consistent outcomes. 2. Moderate evidence: one relevant, high-quality randomized controlled trial AND one or more relevant, lowquality randomized controlled trials with generally consistent outcomes. 3. Limited evidence: one relevant, high-quality randomized controlled trial OR multiple relevant, low-

Included Clinical Trials

The six studies that met the inclusion criteria of this review evaluated dierent manual therapy modalities: spinal manipulation, classic massage, connective tissue manipulation, soft tissue massage, Dr. Cyriaxs vertebral mobilization, manual traction, and CV-4 craniosacral technique (Table 2). The use of a single modality was investigated in four studies,29,30,32,33 and a combination of various manual therapies was analyzed in three papers.28,31,33 One trial compared the use of a single manual therapy versus the combination of two manual therapies.33 In two trials in which spinal manipulation was evaluated,30,33 therapists were instructed to focus their manipulation eorts on the cervical spine, but they were free to manipulate any area of the spine (thoracic or lumbar regions) they determined by palpation and spinal motion tests to be clinically relevant. In the third trial that analyzed the eectiveness of spinal manipulation,28 the experimental group only received joint manipulation along the cervical spine, but dierent cervical vertebrae could be manipulated. Due to the substantial heterogeneity among dierent manual therapies, a metaanalysis could not be performed. The total sample size was 405 patients (mean SD = 60.6 48.4). Bove and Nilsson included only patients with episodic TTH (n = 75),28 whereas Demirturk et al included patients with chronic TTH (n = 30)31 according to the criteria set by the IHS.2 In the
r

280

2006 Lippincott Williams & Wilkins

Clin J Pain

2006 Lippincott Williams & Wilkins

TABLE 2. Manual Therapy Trials Included in This Systematic Review

Study Bove et al28 RCT Design PEDro Score 8/10 Diagnosis Episodic TTH Sample Size 75 (26 men/49 women) Treatments (n patients) A) Spinal manipulation+soft tissue therapy (n = 38) B) Soft tissue therapy+placebo laser (n = 37) A) Spinal manipulation (n = 15) B) Spinal manipulation+cervical manual traction (n = 15) A) Connective tissue manipulation (n = 15) B) Dr. James Cyriax vertebral manipulation (n = 15) A) CV-4 craniosacral technique (n = 20) B) Protraction-retraction neck exercises (n = 20) C) No treatment (n = 20) A) Spinal manipulation (n = 70) B) Amitriptyline doses between 10 mg to 30 mg/per day (n = 56) Connective tissue manipulation Results Both groups obtained a signicant reduction in outcome measures. No dierences between groups Group A obtained a greater improvement than group B No signicant dierences between groups were found Group A obtained a greater improvement than group B and C (P<0.05) No signicant dierences between groups at the end of treatment. At 4-weeks follow up group A showed a greater improvement than group B Patients showed a signicant improvement in all the outcome measures (P<0.05)

Volume 22, Number 3, March/April 2006

Donkin et al33

RCT (No placebo group)

7/10 (rated by the authors) 7/10 (rated by the authors) 6/10 (rated by the authors)

TTH (subtype was not specied) Chronic TTH

30 (10 men/20 women)

Demirturk et al31

RCT (No placebo group)

30 women

Hanten et al29

RCT

TTH (subtype was not specied)

60 (17 men/43 women)

Boline et al30

RCT

5/10

TTH (subtype was not specied) TTH (subtype was not specied) Number Sessions

126 (gender was not specied) 20 women Follow Up 14 weeks after the intervention 4 weeks after the intervention 4 weeks after the intervention (Immediate eects) 4 weeks after the intervention 6 months after the intervention

Akbayrak et al32 Study Bove et al28 Donkin et al33 Demirturk et al31 Hanten et al29 Boline et al30

Open non-controlled study Outcome Measures Daily hours of headache, headache intensity (VAS), daily analgesic use Headache diary, McGill Pain Questionnaire, Neck Disability Index, Numerical Pain Rating Scale Headache index value, active CROM, PPT Headache pain intensity (VAS), aective component of pain (VAS) Headache pain intensity (VAS), weekly headache frequency, over the counter medication usage, functional health status (SF-36) Headache intensity, frequency and duration, analgesic use and associated symptoms

2/10 (rated by the authors)

8 sessions in 4 weeks (two sessions per week) 9 sessions in 5 weeks (two sessions per week) 20 sessions in 4 weeks (5 sessions per week) 1 Group A received 12 sessions in 6 weeks (two sessions per week) Group B took a daily dose of amitriptyl. during 6 weeks 20 sessions in 4 weeks (5 sessions per week)

Manual Therapy for Tension-Type Headache

Akbayrak et al32

RCT, randomized controlled trial; TTH, tension type headache; PPT, pressure pain threshold; CROM, cervical range of motion; amitriptyl, amitriptyline; VAS, visual analogue scale.

281

Fernandez-de-las-Penas et al

Clin J Pain

Volume 22, Number 3, March/April 2006

rest of studies (300 subjects), episodic and chronic TTH were analyzed together without distinction between them.29,30,32,33 The mean duration of TTH episodes was specied in three studies28,32,33 and ranged from 1.5 to 5.2 hours per day (mean = 3.2 1.8). In the study by Boline et al, the duration of TTH episodes was not mentioned, but the authors analyzed the weekly headache frequency.30 Hanten et al analyzed pain intensity and the aective component of pain, but the duration and frequency of TTH episodes were not considered.29 Demirturk et al31 analyzed pain frequency and pain intensity but not the duration of TTH episodes. Treatment duration in the included studies ranged from only one session to 6 weeks of treatment (mean = 3.6 1.9 weeks). The number of sessions ranged from 1 to 20 (mean = 11.6 7.3). Further analysis did not reveal any impact on the outcomes depending on the intensity of the treatment. The follow-up period diered among studies. In one trial the eect was assessed only immediately after treatment.29 The rest of trials included long-term followups, which were performed at 4 weeks,30,31,33 12 weeks,28 or 6 months32 after the intervention. The nature of control/reference groups also varied considerably among the trials. Studies that analyzed spinal manipulation had dierent control/reference groups: one study compared spinal manipulation to drug therapy (amitriptyline),30 another one compared spinal manipulation alone to the combination of spinal manipulation and manual traction,33 and the last one compared spinal manipulation and soft tissue massage to soft tissue massage combined with placebo laser therapy.28 One study compared connective tissue manipulation technique to Dr. Cyriaxs vertebral mobilization group.32 The last one compared CV-4 craniosacral technique to protractionretraction exercises of the neck (rest position group) and to a control group (no treatment).29 Undoubtedly, this diversity of control/reference groups makes it dicult to establish the ecacy of manual therapies beyond placebo therapy in the treatment of patients with TTH.

Quality of Trials
Methodologic PEDro quality scores ranged from 2 to 8 points out of a theoretical maximum of 10 points (mean = 5.8 2.1) (Table 3). Four trials28,29,31,33 gained a high-quality rating (6 to 8 points), whereas the remaining two30,32 received a low score (5 points, 2 points). The most common aws were lack of blinding assessors (ve trials) and failure to explicitly use an intention-to-treat analysis (ve trials). In three papers subjects were described as being blind to the treatment condition, whereas the assessors were blind to the treatment condition in only one study. Although therapists were never described as blinded, that situation cannot be considered a aw because using real blinded therapists is impossible in this kind of trial. In that situation, the evaluation should be addressed in another way, such as comparing subjects and therapists expectations for success across treatment arms. Five trials used random assignment of patients, as well as an adequate method of allocation concealment. In these trials a comparison between groups at baseline (pretreatment) and at the end of therapy (posttreatment) was performed. Follow-up was reported in ve trials. However, one trial with a 4-week postintervention follow-up30 did not satisfy PEDro criteria for follow-up because that period was not considered appropriate for a 6-week intervention. Table 3 shows the details of the PEDro scale items and the total score of all studies. The articles are ordered by their methodologic scores, according to the checklist.

Effect Size and Treatment Efficacy

The eect size was calculated for two studies.29,31 Moreover, in their study Donkin et al reported the within-group (posttreatment vs. pretreatment) eect size of the outcome measurements in percentages.33 Conversely, the eect size was not specied and could not be calculated in three trials.28,30,32 In two of these studies,28,30 95% condence intervals, but not the mean and standard deviation, were given for pretreatment and

TABLE 3. PEDro Score Rated Details of the Trials Included in This Review
Study Bove et al28 Donkin et al33* Demirturk et al31* Hanten et al29* Boline et al30 Akbayrak et al32* Study Bove et al28 Donkin et al33* Demirturk et al31* Hanten et al29* Boline et al30 Akbayrak et al32* Random Allocation Yes Yes Yes Yes Yes No Follow-up Yes Yes Yes No No Yes Concealed Allocation Yes Yes Yes Yes Yes No Intention-toTreat Analysis Yes No No No No No Baseline Comparability Yes Yes Yes Yes Yes No Blind Assessors Yes No No No No No Blind Subjects No Yes Yes Yes No No Points Estimates and Variability Yes Yes Yes Yes Yes Yes Blind Therapist No No No No No No Total Score 8/10 7/10 7/10 6/10 5/10 2/10

Between-Group Comparisons Yes Yes Yes Yes Yes No

*PEDro score calculated by the authors of this review.

282

2006 Lippincott Williams & Wilkins

Clin J Pain

Volume 22, Number 3, March/April 2006

Manual Therapy for Tension-Type Headache

posttreatment data in each group. The last one was an open noncontrolled study, so there was not a comparison between groups.32 The most commonly used outcome measurements were headache intensity (n = 6),2833 headache frequency (n = 4),3033 and the number of analgesics per day (n = 3).28,30,32 Various additional measures were also obtained (pressure pain threshold, cervical range of motion, Short-Form McGill Pain Questionnaire, Neck Disability Index, and Health-Related Quality of Life Scale [SF-36]), but only in one study each. The ES could not be estimated in the trials where the use of analgesic medication was analyzed,28,30,32 but it was calculated for headache intensity and headache frequency in the rest of studies. Demirturk et al31 found a moderate ES (0.3) on the headache index value (frequency of headache and severity of pain) in favor of the Dr. Cyriaxs vertebral mobilization group. Furthermore, a moderate ES (0.49) on the pressure pain threshold and a small ES (0.1) on active cervical range of motion in favor of the same group were obtained. Hanten et al29 found a large size ES (0.84) on the pain intensity and pain aect outcomes in the experimental group (CV-4 craniosacral technique) in relation to the reference group (protractionretraction exercises of the neck) and the control group (no treatment). Donkin et al33 reported a moderate withingroup ES (3863%) for subjective results (intensity, frequency, and duration of headache) in the experimental group (manipulative group), whereas the reference group (manipulative plus manual traction group) showed a small to moderate within-group ES (1247%). The results of the three studies in which the eect size could not be calculated are controversial.28,30,32 Bove and Nilsson28 reported that spinal manipulation, as an isolated intervention, did not seem to have a positive eect on episodic TTH. In this trial soft tissue therapy was combined either with spinal manipulation (experimental group) or with placebo laser therapy (reference group). After treatment, the decrease in headache frequency and in analgesic use was similar in both groups. Therefore, outcomes from this study were dened as neutral. In the study by Boline et al,30 both groups (spinal manipulation group and drug therapy group) obtained a comparable improvement on headache frequency and intensity, analgesic use, and quality of life (SF-36) at the end of treatment. However, after a 4-week postintervention follow-up, the spinal manipulation group showed a signicantly greater improvement in all the outcome measures. Outcomes from this study were dened as neutral at the end of the treatment phase and as positive at the 4-week postintervention follow-up in favor of the spinal manipulation group. The third study32 lacked a control group, so the eectiveness of the applied technique (connective tissue manipulation) could not be determined.

low-quality trial.30 These studies used relevant outcome measures (ie, headache intensity, headache frequency, and/or headache duration). However, results were not generally consistent, as one trial reported positive results,33 another one reported neutral results,28 and the last one reported neutral results at the end of treatment and positive results at follow-up.30 Spinal manipulation could have achieved a moderate evidence of ecacy (level 2). Nevertheless, due to the inconsistency among the results of dierent trials, spinal manipulative therapy obtained an inconclusive evidence of eectiveness (level 4). The remaining manual therapies showed limited evidence of ecacy in reducing pain from TTH (level 3). Connective tissue manipulation was analyzed in one highquality trial31 and in an open noncontrolled study without a reference/control group.32 CV-4 craniosacral technique was analyzed in just one high-quality trial.29 All these papers also used relevant outcomes (ie, headache intensity, headache frequency, and/or headache duration). However, consistency of results could not be established, as there was only one study assessing each form of manual therapy.

DISCUSSION
The main nding of the current systematic review is that there are just a few randomized controlled trials evaluating the eectiveness of manual therapies in the management of TTH. Moreover, the substantial heterogeneity of the techniques used and the small sample sizes in most trials make it dicult to draw denite conclusions. Results did not provide any rigorous evidence that manual therapies have a positive eect in the evolution of TTH. The most urgent need for further research is to establish the ecacy beyond placebo of the dierent manual therapies currently used to treat patients with TTH. In our analysis, spinal manipulative therapy showed inconclusive evidence (level 4) of eectiveness, as there have not been generally consistent results among studies: one trial reported positive results,33 another one reported neutral results,28 and the last one reported neutral results at the end of treatment and positive results at follow-up.30 These results are in agreement with previous systematic reviews reporting that available data did not support that spinal manipulative therapy was an eective treatment of TTH.10,11 Therefore, the hypothesis that spinal manipulative therapy may have specic ecacy in reducing pain from TTH is neither supported nor refuted by the research to date. On the other hand, the eectiveness of connective tissue manipulation,31,32 Dr. Cyriaxs vertebral mobilization,31 and CV-4 craniosacral technique29 showed limited evidence (level 3) of eectiveness in the management of TTH. It has been proposed that the most prominent clinical nding in TTH patients is a considerably increased tenderness to palpation of pericranial myofascial tissues.3436 Moreover, some authors have claimed that myofascial tissues might play an important role in

Levels of Scientific Evidence

The eectiveness of spinal manipulation was analyzed in three studies: two high-quality trials28,33 and one
r

2006 Lippincott Williams & Wilkins

283

Fernandez-de-las-Penas et al

Clin J Pain

Volume 22, Number 3, March/April 2006

the genesis of TTH.37 Because the aim of soft tissue therapies is to alter mechanical stress caused by myofascial tissue disorders, these techniques might be an eective therapeutic choice for TTH. Still, additional trials are required before any soft tissue technique can be considered eective for the management of TTH. We believe that both self-reported and objective measurements are needed for the evaluation of therapeutic eects. Self-reported measurements related to pain and disability levels were the most commonly used in the included trials. Among the self-reported outcome measures, the most frequently assessed were headache intensity (all trials) and headache frequency (four trials).3033 Other types of self-reported and disability outcomes, such as the Short-Form McGill Pain Questionnaire, the Neck Disability Index, and the SF-36, were also obtained in some studies. On the other hand, objective outcomes such as the cervical range of motion or the pressure pain threshold were seldom assessed.31,33 The consumption of analgesics was assessed in three studies.28,30,32 However, it is dicult to classify this outcome as objective or subjective. Although the number of analgesics per day is an objective measure, it clearly depends on the subjective perception of headache intensity. To achieve a comprehensive knowledge of the eect of manual therapies on TTH, other outcomes might be also assessed. As the most prominent clinical nding in TTH patients is an increased tenderness to palpation of pericranial myofascial tissues, changes in tenderness of these structures should be analyzed (eg, by means of the Total Tenderness Score38 or the Pressure Pain Thresholds39). Undoubtedly, it would be interesting to consider changes in tenderness of myofascial tissues and their relationship to changes on subjective measures after the interventions with soft tissue therapies. To our knowledge, this is the rst systematic review that has focused on TTH, but other systematic reviews have also analyzed the eectiveness of spinal manipulative therapy and complementary/alternative therapies for chronic headaches. Astin and Ernst10 and Bronfort et al11 reported that available data did not support that spinal manipulation was an eective treatment of most chronic headaches, whereas Vernon et al concluded that some complementary/alternative therapies might be useful in the treatment of cervicogenic headache and TTH.12 The inconsistency of ndings of previous systematic reviews may be a consequence of the dierent quality-assessment methods that have been used. Most reviews have used the Jadad scale,18 whereas we used the PEDro score. The reliability of the PEDro score has been established by Mahel et al as fair to good (ICC = 0.68, 95% CI = 0.570.76),40 which was similar to that reported by Jadad et al for the Jadad Scale (ICC = 0.59, 95% CI = 0.460.74).18 The PEDro score appears to have sucient consistency for use in systematic reviews of physical therapy. In the current review the most common aws of the included trials were lack of blinding assessors (ve papers) and failure to explicitly use an intention-totreat analysis (ve studies). We believe that better-

designed trials are required to investigate the eectiveness of manual therapies (ie, soft tissue techniques), in the management of TTH. There are a few limitations of the current review. First, we included only papers written in English, so it is likely that we have missed some relevant studies written in other languages. Moher et al41 reported that trials published in French, Spanish, Italian, or German were of similar methodologic quality to those published in English. Therefore, the quality of missed trials evaluating manual therapies in TTH would have conceivably been analogous to what we found in English studies. However, according to Egger et al,42 there is a chance that by restricting the review to English papers the trials may overrepresent positive results (publication bias). Moreover, no eort was made to identify unpublished research, which is more likely to have negative outcomes.43 Second, the ndings of the current review are limited by study heterogeneity and the methodologic aws of the included trials. Failure to adequately blind patients and assessors was particularly common in the reviewed papers. It is certainly hard to blind patients in clinical trials that analyze the eectiveness of manual therapies, but failure to do so is, nonetheless, a potential source of bias. Third, two included papers were actually overlapping publications. The extended trial by Demirturk et al31 was somewhat unorthodox in that the decision to recruit a greater number of patients was made after the original analysis of data of the former trial.32 No specications were made regarding separate analyses, and therefore one must be concerned about the possibility of type 1 errors. However, the exclusion of that overlapping publication would have not altered the results of the present review. The main conclusion of this review is that there are only a few randomized controlled trials testing the eectiveness of manual therapies in the management of TTH. Spinal manipulation showed inconclusive evidence of eectiveness (level 4), whereas soft tissue manipulation techniques showed limited evidence of eectiveness (level 3). Additional well-designed clinical trials are required before manual therapies can be suciently validated for the treatment of patients with TTH. REFERENCES
1. Schwartz BS, Stewart WF, Simon D, et al. Epidemiology of tensiontype headache. JAMA. 1998;279:381383. 2. Classication and Diagnostic Criteria for Headache Disorders. Cranial neuralgias and facial pain, 1st ed. International Headache Society. Cephalalgia. 1988;8(Suppl 7):2934. 3. The International Classication of Headache Disorders. Headache classication subcommittee of the International Headache Society. Cephalalgia. 2004;24(Suppl 1):8152. 4. Bendtsen L. Central sensitization in tension-type headache: possible pathophysiological mechanisms. Cephalalgia. 2000;29:486508. 5. Bogaards MC, Ter Kuile MM. Treatment of recurrent tension headache: a meta-analytic review. Clin J Pain. 1994;10:174190. 6. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United States, 199097: results of a follow-up national survey. JAMA. 1998;280:15691575. 7. Vernon H, McDermaid C. Chiropractic management of episodic tension-type headache: a survey of clinical specialist. J Can Chiro Ass. 1998;42:209216.
r

284

2006 Lippincott Williams & Wilkins

Clin J Pain

Volume 22, Number 3, March/April 2006

Manual Therapy for Tension-Type Headache

8. Haddock CK, Rowan AB, Andrasik F, et al. Home-based behavioural treatments for chronic benign headaches: a metaanalysis of controlled trials. Cephalalgia. 1997;17:113118. 9. Holroyd KA, Penzien DB. Pharmacological versus non-pharmacological prophylaxis of recurrent migraine headache: a meta-analytic review of clinical trials. Pain. 1990;42:113. 10. Astin JA, Ernst E. The eectiveness of spinal manipulation for the treatment of headache disorders: a systematic review of randomized clinical trials. Cephalalgia. 2002;22:617623. 11. Bronfort G, Assendelft JJ, Evans R, et al. Ecacy of spinal manipulation for chronic headache: a systematic review. J Manip Physiol Ther. 2001;24:457466. 12. Vernon H, McDermaid CS, Hagino C. Systematic review of randomized clinical trials of complementary/alternative therapies in the treatment of tension-type and cervicogenic headache. Complement Ther Med. 1999;7:142155. 13. Nillson N. Evidence that tension-type headache and cervicogenic headache are distinct disorders. J Manip Physiol Ther. 2000;23:288289. 14. Greenhalgh T. How to read a paper: the Medline database. Br Med J. 1997;315:180183. 15. Moher D, Schulz F, Altman D. The CONSORT statement: revised recommendations for improving the quality of reports of parallelgroup randomized trials. JAMA. 2001;285:19871991. 16. Oxman AD, Guyatt GH. Validation of an index of the quality of review articles. J Clin Epidemiol. 1991;44:12711278. 17. Verhagen AP, De Vet HCW, De Bie RA, et al. Balneotherapy and quality assessment interobserver reliability of the Maastricht criteria list for blinded quality assessment. J Clin Epidemiol. 1998;51:335341. 18. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17:112. 19. Verhagen AP, De Vet HCW, De Bie RA, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidemiol. 1998;51:12351241. 20. Moseley AM, Herbert RD, Sherrington C, et al. Evidence for physiotherapy practice: a survey of the Physiotherapy Evidence Database (PEDro). Aust J Physiother. 2002;48:4349. 21. Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomized controlled trials. Br Med J. 1999;319:670674. 22. Anderson R, Meeker WC, Wirick BE, et al. A meta-analysis of clinical trials of spinal manipulation. J Manip Physiol Ther. 1992;15:181194. 23. Rosenthal R. Meta-analytic Procedures for Social Research. Beverly Hills: Sage Publication; 1984. 24. Thomas J, Nelson J. Research Methods in Physical Activity, 2nd ed. Illinois: Human Kinetics; 1985. 25. Slavin RE. Best evidence synthesis: an intelligent alternative to meta-analysis. J Clin Epidemiol. 1995;48:918. 26. Van Tulder MW, Koes BW, Bouter LM. Conservative treatment of acute and chronic nonspecic low back pain: a systematic review of

27. 28. 29. 30.

31. 32. 33.

34. 35. 36.

37. 38. 39. 40. 41.

42. 43.

randomized controlled trials of the most common interventions. Spine. 1998;18:21282156. Ezzo J, Berman B, Hadais VA, et al. Is acupuncture eective for the treatment of chronic pain? A systematic review. Pain. 2000;86:217225. Bove G, Nilsson N. Spinal manipulation in the treatment of episodic tension-type headache. A randomized controlled trial. JAMA. 1998;280:15761579. Hanten WP, Olson SL, Hodson JL, et al. The eectiveness of CV-4 and resting position techniques on subjects with tension-type headaches. J Man Manip Ther. 1999;7:6470. Boline P, Kassak K, Bronfort G, et al. Spinal manipulation versus amitriptyline for the treatment of chronic tension-type headache. A randomized controlled trial. J Manip Physiol Ther. 1995;18: 148154. Demirturk F, Akarcali I, Akbayrak T, et al. Results of two dierent manual therapy techniques in chronic tension-type headache. Pain Clin. 2002;14:121128. Akbayrak T, Akarcali I, Karabudak R, et al. The results of connective tissue manipulation in the treatment of tension-type headache. Pain Clin. 2002;13:343347. Donkin RD, Parkin-Smith GF, Gomes AN. Possible eect of chiropractic manipulation and combined manual traction and manipulation on tension-type headache. A pilot study. J Neuromusculoskeletal System. 2002;10:8997. Jensen R, Rasmussen BK, Pederken B, et al. Muscle tenderness and pressure pain threshold in headache. A study population. Pain. 1993;52:193199. Jensen R, Olensen J. Initiating mechanism of experimentally induced tension-type headache. Cephalalgia. 1996;16:175182. Lipchik GL, Holroyd KA, Talbot F, et al. Pericranial muscle tenderness and exteroceptive suppression of temporalis muscle activity: a blind study of chronic tension-type headache. Headache. 1997;37:368376. Simons DG, Travell J, Simons LS. Myofascial Pain and Dysfunction. The Trigger Point Manual. Vol. 1, 2nd ed. Baltimore: Williams & Wilkins; 1999. Langermark M, Olensen J. Pericranial tenderness in tension headache. A blind controlled trial. Cephalalgia. 1987;7:249255. Fischer AA. Application of pressure algometry in manual medicine. J Manual Med. 1990;5:145150. Maher CG, Sherrington C, Hebert RD, et al. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther. 2003;83:713721. Moher D, Fortin P, Jadad AR, et al. Completeness of reporting of trials published in languages other than English: implications for conducting and reporting of systematic reviews. Lancet. 1996;347:363369. Egger M, Zellweger-Zahner T, Scheneider M, et al. Language bias in randomized controlled trials published in English and German. Lancet. 1997;350:326329. Cook DJ, Guyatt GH, Ryan G, et al. Should unpublished data be included in meta-analyses? Current convictions and controversies. JAMA. 1993;269:27492753.

2006 Lippincott Williams & Wilkins

285