The Role of Cervical Manipulation and /or Exercise in the Management
of Cervicogenic Headache
A Review of the Literature
Eileen Gurgone, PT
University of Medicine and Dentistry of New Jersey
School of Health Related Professions
Dr. Craig Scanlan
Chronic headache syndromes continue to represent significant diagnostic and treatment challenges for the medical community. Headache is responsible for 18 million outpatient visits per year and represents 1% of primary care physician visits (Silberstein, 2000). In 1988 the International Headache Society published its diagnostic criteria for the various headache types (IHS ,1988). Cervicogenic headache which was first described by Sjaastad, Saunte, Hovdal, Breivik and Gronbaek (1983) and later modified into its current description, is recognized by the IHS as a significant health problem (Sjaastad , 1998). It is estimated to affect approximately 2.5% of the population and be the leading primary cause of headache in 14% to 18% of those experiencing five or more headaches per month (Shofferman, 2002). In its latest publication the IHS has further expanded its diagnostic criteria and cervicogenic headache is now a subcategory of chronic daily headache (Piovesan, 2003). A comparison of the diagnostic criteria and clinical characteristics of the chronic daily headache types is available for review in Appendix A.
Currently, leading researchers are collaborating with the IHS and the World Cervicogenic Headache Society to further clarify the diagnostic criteria and contribute to the development of effective treatment interventions. These interventions may go beyond pharmacological pain management into the realm of joint manipulation and neuromuscular re education in order to impact upon the neural plasticity of centrally mediated headache pain. This review examines the role of such treatment and attempts to synthesize the available research to answer the following question : Does manual therapy consisting of manipulation and/or exercise affect headache parameters in adults with cervicogenic headache ?
Cervicogenic headache is a chronic, hemicranial pain syndrome (non lancinating) in which the primary source of the pain is in the cervical spine but the pain is perceived in the head ( Bogduk , 1993). Sources of this pain lie in the structures innervated by the C1 through C3 spinal nerves and include the upper cervical synovial joints, muscles of the sub cranial spine, discogenic (C2C3) , as well as the vertebral/internal carotid arteries and the dura (Bogduk , 2001). The headache literature is very clear that this is not occipital (C2) neuralgia, which is its own clinical entity and is characterized by lancinating pain. It is also not to be confused with tension type headache, which has a bilateral band like quality ( Nicholson , 2001). The physiological basis for this pain is via the trigeminocervical nucleus , which is the nociceptive nucleus of the head, throat and sub cranial spine ( O/A through C3 ). Pain is explained by the convergence of descending afferent trigeminal nerve input and afferent input from C1 thru C3 spinal levels synapsing at this brainstem nuclei. Convergence at the trigeminocervical nucleus may allow for referral of pain from the neck to the trigeminal sensory receptive fields in the face and head ( Bogduk , 1994). In addition, the principle of neuronal hyper excitability is appreciated as a central factor in which dynamic changes can occur within the central nervous system in response to pain (Jensen , 2001). Is then ,the cervicogenic headache amenable to manual treatments directed at impacting upon this complex interaction?
The treatment of cervicogenic headache can present a considerable challenge for clinicians. This is due in part to the anatomic and pathophysiologic complexities of the condition ,and also to the lack of high-level standardized evidence based principles of care. The comprehensive nature of manual therapy interventions often makes it difficult to isolate the value of one intervention versus another. A thorough review of the available literature reveals that the primary focus of study has been on the other forms of chronic headache, and even at that very few are the gold standard, randomized controlled trials. However , earlier systematic reviews by Bronfort, Assendelft, Evans, Haas, and Bouter (2001) and Astin and Ernst (2002) have identified support for spinal manipulation. Recent large scale systematic reviews concerning manipulative therapy for neck disorders (Gross, Hoving , & Haines, 2004) and non invasive physical treatments for chronic/recurrent headache (Bronfort, Nilson & Haas, 2004) have upon careful review identified evidence in support of manipulation and/or specific exercise programs in the management of this condition.
This review, for the purpose of clinical application, attempts to identify and assimilate the most relevant literature applicable to this patient population within the scope of physical therapy practice (i.e. non invasive, non pharmacological).
The search for relevant articles began with the use of OVID and the following databases via this source : Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Cochrane Central Register of Controlled Trials, Medline 1966 thru the present and CINHAL. PEDro was also accessed through the search engine GOOGLE to gain access to the Center of Evidence Based Physiotherapy. The APTA website was utilized for access to the Hooked on Evidence database. No restrictions were placed upon language. Due to the multidisciplinary nature of the subject , medical , chiropractic and allied health literature were all screened for inclusion. Subject headings (MeSH) and key words were initially derived from the question and included headache , cervicogenic and manipulation. Keywords were then identified from the initial articles including exercise, mobilization, pain and further headache specific language including migraine, trigeminocervical and nitric oxide. These terms assisted in identifying the research regarding pathophysiology and theory . The topical area of headache in general is quite broad, therefore Boolean AND was used to combine terms and identify articles specific to the topic of cervicogenic headache . A sample of the search history is available for review. (Appendix B).
The articles were then screened to identify original research of all types. The search could not be narrowed to randomized controlled trials only because insufficient numbers of this level of evidence are currently available. The focus was narrowed to articles that studied cervicogenic headache exclusively and in whom the subjects met the IHS diagnostic criteria. The various headache types cannot be taken collectively as a group, as differences exist between them , dictating differences in treatment. In order to determine the effectiveness of manipulation and/or exercise exclusively, studies that utilized modalities extensively were excluded. Modalities themselves impact the outcomes , making it difficult to ascertain which intervention was responsible for the result. An effort was also made to identify studies which assessed similar outcomes. The outcomes discussed in this review reflect the goals of treatment expressed most often by this patient population. These goals include a reduction in headache frequency, duration and intensity, as well as a decrease in medication use. The incidence of adverse reactions represents an important measure of success given the small but possible risk of vertebrobasilar injury with manipulation. In reviewing the available literature, specific muscular impairments in cervicogenic headache were also identified. These studies were included as well, which led to an additional outcome measure of muscular control referred to as the craniocervical flexion test (CCFT).
Twenty-six original research articles were ultimately identified based on this criteria. Sixteen were excluded when upon closer review the primary outcome measures focused on neck pain without sufficient headache data and/or primarily focused on the other forms of chronic daily headache . One was excluded due to a primary intervention involving modalities, which was not apparent from an initial review of the abstract. Ten studies were then included as the focus of this review. Three are randomized controlled trials (Jull et al.,2002 ; Nilsson, Christensen, & Hartvigsen,1995,1997). One is a single group design with repeated measures (Schoensee, Jensen, Nicholson, Gossman, & Katholi, 1995). One is a single subject design with repeated measures (Beeton & Jull,1994). One is a prospective case series (Whorton & Kegerreis, 2000). Two comprehensive case reports have been included, one retrospective (Sizer, Phelps, & Brismee, 2002) and the other prospective (Petersen, 2003). Two cross sectional diagnostic trials have been included for their relevance to the discussion of muscular dysfunction and implications for treatment and outcome measurement ( Jull, Barrett, Magee, & Ho 1999; Watson & Trott , 1993 ). The subjects shared similar baseline characteristics and were primarily female , aged 18-60. This corresponds to the population typically seen in the clinical setting.
The discussion that follows seeks to clarify the role of cervical manipulation, address the role of exercise and the value of a combined program of care.
The highest level of evidence obtained for this review was the work of Jull et al., (2002) who conducted a multi center , prospective, randomized controlled trial to investigate the role of cervical manipulation and low load exercise, alone and in combination , in the management of cervicogenic headache versus a control group receiving no treatment. A 2x2 factorial design was utilized and analysis was by intention to treat. Two hundred subjects were chosen based on a power of .80 to detect a 50% reduction in headache frequency 6 months post treatment with an alpha of .05. Patients were randomized by an independent assessor to 4 groups : manipulative therapy , exercise therapy, combined therapy and a control group. Treatment consisted of 8 to 12 , 30-minute sessions over 6 weeks. Data was collected at baseline , at 7 weeks and at 3,6 and 12 months. The nature of the treatment precluded blinding of the subjects or treating therapists , however the outcome assessors were blinded for collection of non-self reported data. All groups were permitted to take pain medication as needed and all recorded headache parameters in a headache log. The primary outcome measure was a change in headache frequency. Secondary outcome measures included changes in headache duration , intensity and assessment of deep cervical flexor muscle function and medication usage. When compared to the controls all the treatment interventions demonstrated improved outcomes with regard to frequency and intensity. Manipulation plus exercise was found to be superior to exercise alone in regard to duration. Only the groups including exercise demonstrated improved cervical muscle function.
Nilsson et al., ( 1995 ) conducted a randomized controlled trial of 39 subjects comparing manipulation of the cervical spine with a control group. The control group received soft tissue massage and sham laser treatment. Each group received treatment twice per week over a 3-week period. Blinded outcome observers analyzed data regarding intensity, duration and medication usage. A one-week follow up was conducted. The differences between the groups failed to reach statistical significance.
Nilsson et al., (1997) conducted a second arm of the initial trial described above. The authors felt that although the previous trial had failed to reach statistical significance the trend toward improvement had been identified and warranted further study. Fifteen additional subjects were added and the authors reported on 54 subjects. The study was conducted as described above and during this trial statistically significant results were found. The manipulation group reported a 36 % decrease in headache intensity and a 69% decrease in headache duration.
Schoensee et al., (1995) conducted a quasi-experimental single group design with repeated measures to assess the effectiveness of cervical mobilization on headache . Ten volunteers were medically cleared to participate. No attempt at blinding was made. In this ABA design, phase A (4-6 weeks) served as a baseline for data collection of headache frequency, intensity and duration . During phase B ( 4-5 weeks) the subjects received 9-11 sessions of joint mobilization. In the final phase A (4-6 weeks) the subjects received no treatment. Four data collection points occurred during the study (at the beginning of each phase and at the end of the final phase). The authors report a statistically significant reduction in all parameters at the conclusion of the treatment period, with only decreased frequency noted at the follow up assessment.
Whorton and Kegerreis (2000) reported a prospective case series on the use of manual therapy and exercise in the treatment of cervicogenic headache. Six cases were described in which the subjects received both manipulation and exercise . They recorded their headache parameters over an average treatment span of 3.8 weeks. Each subject received an average of 5.9 treatments. Data was collected at a six-month follow up. Five of the 6 subjects reported a statistically significant improvement with treatment. One subject left the trial before manipulation and exercise were introduced.
Sizer et al., (2002) published a retrospective case report of a 45-year-old female with a 20-year history of cervicogenic headache . The patient received a combined program of manipulation and exercise for a total of 24 treatments over a 3-month period. Pre and post treatment data collection of headache parameters as well as a 4 and 12 month follow up were collected. The authors reported a significant improvement in all parameters, which was maintained at follow up. Blinded outcome assessors were utilized .
Beeton and Jull (1994) in a prospective, quasi-experimental single case study of a 32-year-old male investigated a program of manipulation and exercise on cervicogenic headache. The study was an ABC design with repeated measures. The program consisted of 3 six-week phases with data collection at 2-week intervals, this included a 6-week follow up. The A phase was for data collection, the B phase represented treatment. The C phase consisted of home exercise only. There was blinding of the outcome assessor and principal investigator. This subject demonstrated a significant improvement in all headache parameters, which was maintained at follow up.
Petersen (2003) in a prospective case report of a 27-year-old female investigated the role of manipulation and exercise over an 8-week treatment period. The patient received 8 treatment sessions. Pre and post treatment data was collected in regard to headache parameters. There was no follow up or blinding of assessors. The author reported a significant improvement for all parameters at the conclusion of the trial.
Watson & Trott (1993) and Jull et al., (1999) performed quasi experimental cross sectional controlled diagnostic trials which identified the deep cervical flexor muscle group as dysfunctional in cervicogenic headache patients. Watson & Trott initially identified poor endurance and weakness in a study of 60 subjects . Thirty controls were compared to 30 patients with headache. The control group demonstrated no deficits in these cervical muscles. Jull et al., reproduced these findings in a study of 30 subjects. Again fifteen subjects with headache were found to have deep cervical flexor dysfunction while 15 age and gender matched controls did not.
It is important to note at the outset of this discussion that manipulative techniques are employed by a variety of health care practitioners. These include physical therapists, osteopaths, chiropractors and physiatrists. Each discipline may utilize different language when defining these techniques . Reviews of the current research identifies that a lack of standardized language still exists. The Guide to Physical Therapist Practice (2003) endorses interchangeable use of manipulation and mobilization which are defined collectively as “ a manual therapy technique comprising a continuum of skilled passive movements to the joints and/or related soft tissue that are applied at varying speeds and amplitudes , including a small amplitude /high velocity therapeutic movement”. The Manipulation Education Committee of the American Physical Therapy Association defines thrust manipulation as “ high velocity , low amplitude therapeutic movements within or at end range of motion” (2003). Historically, in physical therapy education the Maitland grades of mobilization separate the terminology as follows: grades one through four are described as mobilization, and grade five is defined as thrust manipulation (Maitland, 1964). When reviewing the literature caution must be utilized so as not to extrapolate the use of one proven technique to all available techniques. For the purpose of this review, every attempt was made to clarify the techniques used in each study. Of the eight articles addressing treatment, two exclusively utilized thrust technique ( Nilsson et al., 1995,1997) ; three exclusively utilized Maitland non thrust mobilization ( Schoensee et al. ,1995 ; Whorton & Kegerreis, 2000 ; Petersen , 2003); and three used a combination of thrust and non thrust (Jull et al., 2002; Beeton & Jull, 1994 ;Sizer et al.,2002). The motion segment to which the treatment is applied is also important. Again these articles were chosen to focus the intervention to the implicated segments of the sub cranial spine ( Bogduk, 2001). However it is important to note that both Sizer (2002) and Whorton (2000) addressed the upper thoracic restrictions, which may have contributed to the outcomes to a greater degree. The upper thoracic spine makes a significant functional contribution to cervical movement (particularly rotation) as well as to end range upper extremity elevation. Dysfunction of the scapulothoracic complex adds stress to the midcervical spine particularly with sustained functional movement of the upper extremities (Janda,1994; Watson & Trott, 1993).
Impact on Outcome Measures
Recent systematic reviews have attempted to determine the effectiveness of spinal manipulation for the treatment of headache disorders. Earlier studies have supported manipulation and tension type headache (Vernon, McDermaid , and Hagino, 1999; Bronfort et al., 2004) and preliminary support for manipulation and migraine exists ( Akbayrak, Citak, Demirturk and Akarcali, 2001; Tuchin & Pollard , 1998). A high quality RCT ( Hoving et al. , 2002) identified non thrust manipulation as superior to physical therapy and general practitioner in the management of neck pain which included headache in as many as 50% of the 183 subjects. However, headache data was not reported separately. A high quality cohort study( Wang et al. ,2003) also identified success with manipulation in 57 subjects with neck pain and concurrent headache ( 50 to 70 % of subjects ).Neck pain was the primary outcome measure and headache specific data was not available for review. The question remains, what is the role of manipulation in the management of cervicogenic headache exclusively? Caution must be exercised so as not to generalize results between headache types. Despite the shared neural pathway much is yet unknown and research indicates that headaches can co exist (Sjaastad , Fredriksen ,Pareja, Stolt-Nielsen & Vincent, 1999). In addition, different biochemical processes may occur between the headache types ( Narin, Erbas, Ozturk & Idimin, 2003;Koseoglu, Akboyraz, Soyuer & Ersoy , 2003).
This section identifies three studies, which exclusively examine manipulation and cervicogenic headache (Nilsson, 1995, 1997 and Schoensee et al. 1995) as well as a high quality RCT that examines a manipulation only treatment arm in the context of the larger study (Jull et al., 2002). These studies utilized outcome measures with the greatest relevance to patient’s goals. This included the headache parameters of intensity, duration and frequency as recorded via the headache log and visual analog scale (VAS). The IHS regards a 50% reduction in headache frequency to be a clinically relevant benchmark for patient care ( Jull et al ., 2002). Use of analgesics and report of adverse reactions also constitutes an important safety profile for these interventions.
Nilsson et al., (1995) published a randomized controlled trial of 39 patients in which no statistically significant difference between the treatment and control groups was identified after three weeks of thrust manipulation and upon a one week follow up. The authors reported at that time that they believed the study was underpowered and had begun to identify trends toward improvement. In 1997, they subsequently recruited 15 additional subjects for a total of 54 and identified a statistically significant difference between the groups (Nilsson et al. , 1997). The authors reported a 36% decrease in headache intensity and a 69% decrease in duration versus the control group. This would constitute a short-term benefit with only a limited follow up of one week. Frequency was not reported on. Additionally the validity of the study is threatened by adding subjects over time. The control group actually did receive soft tissue techniques and sham laser, which did not appear to improve their outcomes either by direct effect or placebo. Despite these drawbacks the subjects were a homogenous sample, and randomization /blinding of assessors improves generalizability.
An earlier single group repeated measures design by Schoensee et al. (1995) investigated non-thrust mobilization on 10 subjects during a three week treatment period and a six week follow up period. Like Nilsson, this still constitutes a short-term follow up limiting the external validity. Statistically significant improvements were achieved for frequency, duration and intensity during the treatment period. Frequency remained improved at the six-week follow up period. There are questions regarding the homogeneity of the sample as the data reveals that seven of the 10 subjects demonstrated symptoms consistent with migraine headache.
The highest quality study with a manipulation only treatment arm was the RCT by Jull et al. (2002) in which the 200 subjects were randomized to four equal groups to investigate the role of cervical manipulation and low load exercise alone and in combination versus a control group receiving no intervention. Statistical analysis allowed for each treatment to be compared to the controls and for investigation of both treatment effects simultaneously via the factorial design. Furthermore, summary comparisons were made for both short term (7 weeks) and long term (12 month follow up). Headache frequency was the primary outcome measure. Maitland thrust and nonthrust was utilized. Treatment occurred over 6 weeks for all the groups. Independent assessors randomized the subjects and outcome assessors were blinded as well. As in the Nilsson (1995,1997) and Schoensee et al. trials (1995) the nature of the interventions precluded the blinding of the patients and treating therapists. The study by Jull et al., with a large homogenous sample, control group and intention to treat analysis resulted in strong internal and external validity and subsequent generalizability. It is of significance that when compared to the controls all the treatment groups demonstrated improved outcomes with regard to frequency and intensity. None of the interventions were significantly different from each other in regard to frequency and intensity. The improvements were maintained at the 12-month follow up. So, although manipulation alone was not better than the other interventions it was statistically significant over no treatment and the treatment demonstrated long-term effects. In all of the treatment groups at 7 weeks, 76% gained the benchmark of 50%, while 35% had complete relief. At 12 months, 72 % maintained 80-100% relief. It is important to note that duration of headache was not improved with manipulation and these implications will be discussed later. This study strongly supports the use of manipulation not only for the short term, but long term as well.
Outcome assessment also demonstrated decreased use of analgesics (Nilsson et al., 1997; Jull et al., 2002). Given the risk of rebound headache and gastrointestinal complications (7600 deaths per year due to gastrointestinal bleeding ) with non-steroidal anti-inflammatory drugs, decreased medication usage is a clinically relevant outcome (Tarone et al., 2004). Nilsson et al., reported a 36% drop in the manipulation group while Jull et al. , reported that the manipulation group had the best outcome with 100% resolution versus a 33% increase in the controls.
None of the interventions examined in this review, even those which utilized thrust technique (Nilsson et al., 1995,1997; Jull et al., 2002; Beeton & Jull, 1994; Sizer et al. 2002) resulted in any vertebrobasilar injury. Given the anatomic course of the vertebral arteries through the cervical spine, appropriate skill and screening for contraindications to mobilization and manipulation remains warranted. Catastrophic incidents remain rare, from 1 per 20,000 to 1 per 1 million (Assendelft, 1996) to 1 per 400,000 to 3-6 per 10 million (Haldeman, Kohlbeck and McGregor, 2002). According to a retrospective case series by Haldeman et al. (2002) one hundred and seventeen incidents of post manipulative vascular injury have been documented since 1934. This review did not identify any adverse reactions in either the low or high velocity groups. None of the subjects in this review experienced anything other than treatment soreness, which resolved within 24 –48 hours.
To summarize the assessment of manipulation only, of the three RCT’s, two support the use of manipulation in the short term (Nilsson et al., 1997; Jull et al., 2002) and one of particularly strong quality also supports its use for the long term (Jull et al., 2002). Schoensee et al. (1995) although a lower level of evidence also supports manipulation for short-term outcome.
Questions still exist regarding enhancing long-term outcomes in the cervicogenic headache population. While reviewing the treatment arms and additional outcome measures of muscular performance, Jull et al. (2002) have identified that although the manipulative therapy group fared well for intensity and frequency this group failed to improve when the deep cervical flexor muscles were re assessed for strength and endurance. This indicates that there was no spontaneous return of this muscle function without the use of exercise despite significant pain relief. This has previously been identified to be the case regarding neck flexor recovery post whiplash (Jull et al., 2000) and multifidii recovery post lumbar spine injury (Hides, Richardson, & Jull, 1996). Close inspection of the data provided by Beeton and Jull (1994) demonstrated that the improvements in the headache parameters did not begin until the muscular re education exercise were added. This then raises the question, what is the role of exercise directed at the deep cervical flexors and postural muscles in the long-term management of cervicogenic headache? Can muscular retraining affect neuronal hyper excitability in the centrally mediated pain of headache, much like manipulation produces a short-term neurophysiological effect?
Altered neuro motor function of the neck flexor synergy has been identified in the cervicogenic headache population (Jull et al., 1999; Watson & Trott, 1993). The deep cervical flexor (DCF) group is composed of the longus capitis,longus colli, rectus capitis, and lateralis. Neck flexor synergy refers to the ability to utilize the DCF to produce craniocervical flexion and maintain intersegmental stability for the midcervical muscles to act, primarily the sternocleidomastoid and scalenes (Janda, 1994; Jull et al., 1994).
Watson and Trott (1993) via a cross sectional trial quantified the diagnostic criteria of impaired muscle function in cervicogenic headache with a simple inexpensive test of the DCF muscles. By measuring posture and DCF strength and endurance in headache subjects versus controls a pattern began to emerge. Using a simple pressure gauge apparatus while performing OA flexion, isometric strength and endurance was tested. This trial identified that in persons with cervicogenic headache forward head posture was present and that the DCF muscles were weaker and had significantly lower endurance. This was a strong study based on a homogenous sample of 60 subjects. It identified the role of endurance of the DCF and implications for treatment (endurance training /low load) . The cross sectional trial performed later by Jull et al. (1999) supported the finding that DCF function was significantly poorer in the headache group .The reader is cautioned that cause and effect regarding posture and DCF function has not yet been established. Postural photography was utilized by Jull et al., (2002) and despite significant improvements in headache pain, posture remained unchanged in all the groups. It may be the dynamic control of the cervical spine versus static posture that has a greater impact upon headache (Jull et al.).
The craniocervical flexion test (CCFT) was introduced as a low load test targeting the tonic holding capacity of the deep cervical flexors (Beeton & Jull, 1994). The subject was instructed to perform OA flexion with objective measurements recorded via pressure sensor biofeedback. This measurement tool has subsequently been utilized for diagnostic purposes (Jull et al., 1999) and for training purposes ( Jull et al., 2002; Petersen, 2003). Endurance was measured by the ability to perform ten – 10 second holds of 10mmHg pressure. Neck flexor synergy was assessed by then asking the patient to progress to midcervical flexion and hold as long as possible.
Impact on Outcome Measures
Very little available research to date has examined the role of low load exercise or any exercise program in isolation for the treatment of cervicogenic headache. As previously discussed in the study by Beeton and Jull (1994), the headache log revealed that the complete resolution of headache at 6 weeks co incided with the time frame when DCF training was initiated. Also CCFT assessment at the conclusion of the trial demonstrated that the subject had in fact achieved the goal of ten- 10 second holds and increased synergy hold time from 2 seconds to 105 seconds. The strength of this single subject design was improved via repeated measures and blinding of the independent assessor and principal investigator. Jull et al. (2002) did have an exercise only arm which identified that the exercise group, in fact ,demonstrated the same statistically significant outcomes in regard to headache frequency and intensity and this was maintained at 12 month follow up. Of great importance as it relates to long-term outcome, only the groups including exercise and exercise alone-demonstrated improved performance on CCFT. Due to the fact that exercise, which resulted in no adverse reactions, produced improvement it would be a safe option for those in which manipulation and mobilization are contraindicated. Additionally, Jull et al. ,identified that the exercise only group also had 100% resolution of analgesic use (same as manipulation group).
Petersen (2003) utilized the CCFT and also demonstrated that the subject only had the ability to perform one- 10-second hold with considerable substitution of the SCM/scalenes initially. At the conclusion of the trial the subject was able to demonstrate ten-10 second holds. No assumptions can be made regarding exercise alone however as the treatments were applied concurrently.
Of note, in each of the studies which utilized low load exercise instruction as part of a formal home exercise program the subjects reported that the exercise component was an effective management tool for aborting early headache symptoms (Beeton & Jull, 1994; Petersen, 2003; Sizer et al., 2002; Whorton & Kegerreis, 2000). These studies as single subject and case series designs have limited generalizability .
The combined programs document care as it is generally applied in a clinical setting, concurrently. This makes it difficult to determine the effects in isolation. These well documented case reports (Beeton & Jull, 1994; Petersen, 2003; Sizer et al., 2002; Whorton & Kegerreis, 2000) contribute to the body of knowledge regarding cervicogenic headache. The work of Jull et al., (2002) provides the highest level of evidence regarding the impact of the combined program.
Jull et al. (2002) reported that manipulation plus exercise resulted in statistically significant improvements in headache frequency, intensity and duration when compared to the controls and the effects were maintained at 12 months. Interestingly in this study only the combination of manipulation (low and high velocity) and low load exercise had a positive influence on the duration of headache. Although it may seem surprising that the combined treatment did not show greater benefit for all parameters than the individual treatments, it does show the complexity of headache parameters and demonstrates the importance of each type of treatment upon each headache parameter. The combined treatment maintained the outcome at the 12-month follow up as well.
Whorton and Kegerreis (2000) demonstrated significant improvement in 5 of 6 subjects, which was maintained at 6 months. Three of 5 achieved complete resolution while 2 demonstrated a 50 % reduction in intensity and an 80-90 % reduction in frequency. This study has several confounding variables, which would warrant further study. In addition to non-thrust cervical manipulation, upper thoracic and soft tissue techniques were utilized in a secondary fashion. Most notable here is the additional use of aerobic exercise. Unfortunately no separate or detailed data is available to extrapolate any specific impact of the aerobic component on outcome. Beeton and Jull (1994) have identified the resolution of symptoms with the combined phase with only one headache recurring at the 6 week follow up. Close review of the data also reveals that although no manipulative techniques were applied to segments C3C4 through T5T6 , movement was restored to normal –perhaps the exercise intervention was responsible. Sizer et al. (2002) presented a thorough case report in which a combined program also demonstrated long-term improvement at 4 and 12 months. The subject, with a 24-year headache history achieved full resolution and reported the low load home exercise program as critical to maintaining the headache free state. Beeton and Jull (1994) and Whorton and Kegerreis (2000) have also reported on the integral role of the home program. Confounding variables exist in the work of Sizer et al. (2002) as preparatory electrical stimulation and upper thoracic techniques were utilized as needed. Petersen (2003) replicated the intervention of Beeton and Jull (1994) and Jull et al. (2002). This subject, over an 8-week period reduced headache frequency from 7 per week to 0/1 per week, with intensity decreased by 90 %. However the lack of follow of diminishes the generalizability of this work.
Medication usage was significantly diminished with the combined program as well. Jull et al. (2002) demonstrated a 93 % decrease. Whorton and Kegerreis (2000) provided extensive data regarding a variety of medications utilized by the subjects and identified a 50-100% drop in use. Of note, in addition to over the counter analgesics and prescription anti-inflammatories, one subject was also using Imitrex (triptan) and Amitriptyline (tricyclic). These medications are typically used for migraine relief. Triptan drugs, due to their non-selective powerful vasoconstrictor action, can induce cardiac complications (Mauskop, 2001). This subject was able to terminate the use of Imitrex thru the 6-month follow up period. Whorton and Kegerreis (2000) reported on the loss of one subject who withdrew after receiving only soft tissue techniques and experienced increased pain. Only treatment soreness was reported in response to the combined program with no vertebrobasilar injury. The combined programs demonstrated a positive influence on the headache parameters, medication use and demonstrated a strong safety profile. However only one controlled study has been done (Jull et al., 2002). Case reports especially with confounding variables limit generalizability although the evidence appears similar from case to case when the protocols are compared.
This review has attempted to identify the role of manipulation and /or low load exercise in the management of cervicogenic headache. Manipulation appears to be supported as a safe and effective intervention in the studies reviewed. Certainly it is worth noting that this is based on only two randomized controlled trials (Nilsson et al. 1997; Jull et al. 2002) and a single group design (Schoensee et al., 1995). Jull et al. (2002) were able to identify long-term control. Given the safety profile, affect on headache parameters and drop in medication use , recommendations for use in a clinical program can be made.
Two cross sectional studies (Watson & Trott 1993; Jull et al., 1999) have significantly added to the body of knowledge regarding muscular control and cervical dysfunction. They have identified the role of the deep cervical flexors in cervicogenic headache as well as supporting previous work identifying that spontaneous muscle recovery does not occur without exercise. This has resulted in a specific low load exercise program. Additionally, Jull et al. (2002) demonstrated that exercise alone also resulted in short and long-term improvement, thereby an effective intervention can be made in patients for whom manipulation is contraindicated. Other than the exercise only treatment arm in the RTC by Jull et al. (2002) no other study has investigated the impact of exercise alone in the management of cervicogenic headache, so this warrants further study.
Manipulation plus exercise demonstrated improved outcomes on all parameters as well . Each of the case reports and case series showed significant improvement utilizing similar programs. The programs were well tolerated and the subjects experienced no adverse reactions. The home exercise program directed at the deep cervical flexors was identified as an integral part of not only maintenance but also as an abortive measure of headache control.
Manipulation alone and in combination with low load exercise does appear to be a safe and effective non-pharmacological intervention in the cervicogenic headache population. Additional research regarding exercise alone would be necessary to draw any final conclusion regarding its use in isolation.
An important issue to address is the role of placebo. Placebo effect refers to an improvement in the patients’ condition, which is not directly attributable to the treatment. It is not yet known why this occurs but without a control group for comparison it is hard to know if a placebo effect is taking place. The improvement may occur because the patient had a “belief” in the treatment and /or confidence in the practitioner. The improvement may also have occurred as a natural course of the condition regardless of the intervention. This is an important consideration as only the work of Jull et al. (2002) and Nilsson et al.(1995 and 1997) utilized control groups for comparison.
Recommendations for the Future
Management of pain with effective, non-pharmacological, long term control is of great importance in our patient populations. These studies certainly support continued research to improve upon interventions directed at CNS modulation outside of invasive or medicinal options.
Interventions, which impact upon the trigemino cervical nucleus, may have implications for other diagnostic categories including migraine. Vascular headache is also mediated by the trigeminal and cervical systems (May, 1999). In the migraine scenario the initial constriction of blood vessels (aura) subsequently leads to reflex dilation as the vessels work to restore blood flow to oxygen deprived areas. Thus the throbbing nature of the headache. This may explain why cervicogenic headache generally described as dull can progress to throbbing. There appears to be a shared network as the dural vessels and the dura themselves are innervated by branches of the trigeminal nerve (May, 1999). No study has yet been done which examines the effect of aerobic exercise in isolation on cervicogenic headache. Significant research is ongoing regarding the impact of aerobic exercise on migraine with very promising results (Lockett, 1992; Tuchin, 1998; Mauskop, 2001). Inquiring into the exact mechanism of action is ongoing as well (Narin, 2003; Koseoglu, 2003). This has dual implications – perhaps aerobic exercise would have an additive effect upon manipulation and DCF exercise and perhaps migraine patients can benefit from non-invasive physical therapy interventions.
Given the multi factional nature of pain mechanism in partner with the complex biomechanical and neural control of cervicoscapular movement, the task of studying cervicogenic headache is challenging. However given its incidence and often disabling nature pursuit of the optimal treatment is warranted.
References
American Physical Therapy Association.(2003). Guide to Physical Therapist Practice
(2nd ed.). Alexandria, Va.
Akbayrak, T., Citak, I., Demirturk, F., & Akarcali, I. (2001). Manual therapy and pain
changes in patients with migraine –an open pilot study. Advances in Physiotherapy,
3(2), 49-52.
Assendelft, W.J., Bouter, L.M., & Knipschild, P.G. (1996) Complications of spinal
manipulations: a comprehensive review of the literature. Journal of Family
Practice, 42,475-480.
Astin, J.A., & Ernst, E.(2002). The effectiveness of spinal manipulation for chronic
headache disorders : a systematic review of randomized clinical trials. Cephalagia,
22, 617-623.
Beeton, K., & Jull, G. (1994). Effectiveness of manipulative physiotherapy in the
Management of cervicogenic headache: a single case study. Physiotherapy, 80(7),
417-423.
Biondi, D.M. (2001). Cervicogenic headache : diagnostic evaluation and treatment
strategies. Current Pain & Headache Reports, 5(4), 361-368.
Bogduk, N. (1993). Headache of cranial nerve and cervical nerve origin other than tic. In
Jolesen, Tfelt – Hansen & KMA Welch (Eds.), The Headaches (pp.765-772). New
York: Raven Press.
Bogduk, N. (1994). Cervical causes of headache and dizziness. In JD Boyling & N
Palastanga (Eds), Grieves Modern Manual Therapy (pp 317-322). New York:
Churchill Livingstone Inc.
Bogduk, N. (2001). Cervicogenic headache: anatomic basis and pathophysiologic
mechanism. Current Pain & Headache Reports, 5(4), 382-386.
Bronfort, G., Assendelft, W., Evans, R., Hass, M., & Bouter, L. (2001). Efficacy of
spinal manipulation for chronic headache : a systematic review. Journal of
Manipulative and Physiological Therapeutics, 24(7), 457-466.
Bronfort, G., Nilsson, N., Hass, M., Evans, R., Goldsmith, C., Assendelft, W., et al.
(2004). Non invasive physical treatments for chronic/recurrent headache. The
Cochrane Library, 3, 1-167.
Gross, A.R., Hoving, J.L., Haines, T.A., Goldsmith, C.H., Kay, T., Aker, P., et al. (2004).
Manipulation and mobilization for mechanical neck disorders. The Cochrane Library,
3, 1-99.
Haldeman, S., Kohlbeck, F.J., & McGregor, M. (2002). Unpredictability of
cerebrovascular ischemia associated with cervical spine manipulation therapy.
Spine, 27(1), 49-55.
Hides, JA., Richardson, CA., & Jull, GA. (1996). Multifidus muscle recovery is not
automatic after resolution of acute, first-episode low back pain. Spine 21(23), 276-
279.
Hoving, J.L., Koes, B.W., de Vet, H., van der Windt, D., Assendelft, W.J., Mameren,
H.V., et al. (2002). Manual therapy, physical therapy or continued care by a general
practitioner for patients with neck pain. Annals of Internal Medicine, 136(10), 713-
722.
International Headache Society Classification Committee. (1988). Classification and
diagnostic criteria for headache disorders, cranial neuralgias, and facial pain.
Cephalagia, 8, 9-96.
Janda, V. (1994). Muscles and motor control in cervicogenic disorders: assessment and
management. In R Grant (Eds), Physical Therapy of the Cervical and Thoracic Spine
(pp 95-216). New York: Churchill Livingstone.
Jensen, T.S. (2001). Recent advances in pain research : implications for chronic
headache. Cephalagia, 21, 765-769.
Jull, GA. (2000). Deep cervical neck flexor dysfunction in whiplash. Journal of
Musculoskeletal Pain, 8, 143-154.
Jull, G., Barrett, C., Magee, R., & Ho, P. (1999). Further clinical clarification of muscle
dysfunction in cervical headache. Cephalalgia, 19, 179-185.
Jull, G., Trott, P., Potter, H., Zito, G., Niere, K., Shirley, D. et al. (2002). A randomized
controlled trial of exercise and manipulative therapy for cervicogenic headache.
Spine, 27 (17), 1835-1843.
Koseoglu, E., Akboyraz, A., Soyuer, A., & Ersoy, A. O. (2003). Aerobic exercise and
plasma beta-endorphin levels in patients with migrainous headache without aura.
Cephalalgia, 23(10), 972-976.
Lockett, D. M., & Campbell, J.F. (1992). The effects of aerobic exercise on migraine.
Headache, 32(1), 50-54.
Maitland, GD.(1977). Vertebral Manipulation (4th ed.). London:Butterworth & Co.
Mauskop, A. (2001). Alternative therapies in headache . Is there a role?. Medical Clinics
of North America, 85(4), 1077-1084
May, A., & Goadsby, PJ. (1999). The trigeminovascular system in humans:
pathophysiologic implications for primary headache syndromes of the neural
influences on the cerebral circulation. Journal of Cerebral Blood Flow &
Metabolism, 19 (2), 115-127.
Narin, S.O., Pinar, L., Erbas, D., Ozturk V., & Idimin F., (2003). The effects of exercise-
related changes in blood nitric oxide level on migraine headache. Clinical
Rehabilitation, 17(6), 624-630.
Nicholson, G.G., & Gaston, J. (2001). Cervical Headache. The Journal of Orthopedic &
Sports Physical Therapy, 31(4), 184-193.
Nilsson, N., Christensen, H., & Hartvigsen . (1995). A randomized controlled trial of the
effect of spinal manipulation in the treatment of cervicogenic headache. Journal of
Manipulative and Physiological Therapeutics, 18, 293-297.
Nilsson , N., Christensen, H., & Hartvigsen. (1997). The effect of spinal manipulation in the
treatment of cervicogenic headache. Journal of Manipulative and Physiological Therapeutics,
20(5), 326-330.
Petersen, S.M. (2003). Articular and muscular impairments in cervicogenic headache: a
case report. Journal of Orthopedic & Sports Physical Therapy, 33(1), 30-32.
Piovesan, EJ., & Kowacs, PA. (2003). International headache Society criteria (IHC-2003)
What will be changed in the primary headaches classification? Migranes Cefaleias,
6(2), 38-44.
Schoensee, S.K., Jensen, G., Nicholson, G., Gossman, M., & Katholi, C. (1995). The
effect of mobilization on cervical headaches. Journal of Orthopedic & Sports Physical
Therapy, 21(4), 184-196.
Schofferman, J., Garges, K., Goldwaite, N., Koestler, M., & Libby, E. (2002). Upper
Cervical Anterior Diskectomy and Fusion Improves Discogenic Cervical Headaches.
Spine, 27(20), 2240-2244.
Silberstein, S., Lipton, RB., Goadsby, PJ., & Smith, RT. (2000). Headache in Primary
Care (1st ed.). Philadelphia: Taylor & Francis.
Sizer, P.S., Phelps, V., & Brismee, J. (2002). Diagnosis and management of cervicogenic
headache and local cervical syndrome with multiple pain generators. Journal of
Manual and Manipulative Therapy, 10(3), 136-152.
Sjaastad, O., Fredriksen, T., Pareja, J.A., Stolt-Nielsen, A., & Vincent, M. (1999).
Coexistence of cervicogenic headache and migraine without aura. Functional
Neurology, 14(4), 209-281.
Tarone, R.E., Blot, W.J., & McLaughlin, J.K. (2004). Nonselective , non –aspirin, non-steroidal
anti-inflammatory drugs and gastrointestinal bleeding: relative and absolute risk estimates
from recent epidemiologic studies. American Journal of Therapeutics, 11(1), 17-25.
Tuchin, P.J., & Pollard, H. (1998). Does classic migraine respond to manual therapy- a
case series. Physical Therapy Reviews, 3(3), 149-162.
Vernon, H., McDermaid, C.S., & Hagino, C. (1999). Systematic review of randomized
clinical trials of complementary/alternative therapies in the treatment of tension-type
and cervicogenic headache. Complementary Therapies in Medicine, 7, 142-155.
Wang, W., Olson, S.L., Campbell, A.H., Hanten, W.P., & Glesson , P.B. (2003). Effectiveness of
physical therapy for patients with neck pain. American Journal of Physical Medicine and
Rehabilitation, 82,203-218.
Watson, D.H. & Trott, P.H. (1993). Cervical Headache :an investigation of natural head
posture and upper cervical flexor muscle performance. Cephalalgia, 13(8), 272-284.
Whorton, R.L., & Kegerreis, S. (2000). The use of manual therapy and exercise in the
treatment of chronic cervicogenic headaches: a series of case studies. Journal of
Manual and Manipulative Therapy, 8(4), 193-203.
Appendix A
Diagnostic Criteria – Chronic Daily Headaches
|
Cervicogenic |
Migraine |
Tension - Type |
|
A. Pain is localized to the neck and occipital region and may project to forehead, orbital region, temples, vertex and ears. B. Pain is precipitated or aggravated by special neck movements or sustained neck posture. C. At least 1 of the following:
D. Radiological exam reveals at least 1:
Clinically: unilateral without sideshift episode or chronic, 2-3 wk. may last hrs to days non throbbing moderate to severe does not respond to migraine medication may have ipsilateral UE radicular signs |
Without aura (“common”) A. At least 5 attacks lasting 4-72 hrs. with at least 2 of:
B. During headache at least 1:
With Aura (“classic”) A. At least 2 attacks with 3 of:
B. At least 1 aura features:
Clinically: unilateral with sideshift frontal/retro orbital, temporal, occipital 1 x per year to several per Wk food triggers
|
(“muscle tension”) A. Headache lasting 30 minutes to 7 days B. At least 2:
C. Both:
Clinically: bandlike (frontal) quality often relieved with activity responds to migraine meds. |
Note. Developed from “International Headache Society criteria (IHC-2003)” by E.J. Piovesan and P.A. Kowacs, 2003, Migranes Cefaleias, 6(2), 38-44. “Cervicogenic Headache” by D.M. Biondi, 2001, Current Pain and Headache Reports, 5(4), 361-368.
Appendix B
Database: CDSR, ACP Journal Club, DARE, CCTR
Search Strategy:
--------------------------------------------------------------------------------
1 cervicogenic headache.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (16)
2 cervical manipulation.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (24)
3 exercise.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (17575)
4 aerobic exercise.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (547)
5 pain.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (31257)
6 headache.mp. [mp=ti, ot, ab, tx, kw, ct, sh, hw] (6735)
7 from 1 keep 1-2,6,9 (4)
8 from 2 keep 4 (1)
9 3 and 6 (311)
10 from 9 keep 31,63 (2)
11 4 and 6 (7)
12 3 and 5 (1683)
13 12 and 4 (53)
14 7 or 8 or 10 (7)
15 from 14 keep 1-7 (7)
Database: Ovid MEDLINE(R) <1966 to September Week 4 2004>
Search Strategy:
--------------------------------------------------------------------------------
1 headache.mp. or exp HEADACHE/ or HEADACHE DISORDERS/ (30885)
2 cervic$.tw. (93580)
3 1 and 2 (1007)
4 cervical manipulation.mp. or exp Manipulation, Spinal/ (731)
5 migraine.mp. or exp MIGRAINE/ (14687)
6 aerobic exercise.mp. or exp Exercise/ (31136)
7 3 and 4 (40)
8 from 7 keep 1-2,20,28 (4)
9 3 and 6 (6)
10 3 and 5 (169)
11 from 10 keep 9,38,47,50,53,64,111 (7)
12 5 and 6 (25)
13 from 12 keep 2-3,8,10,22 (5)
14 nitric oxide.mp. or exp Nitric Oxide/ (63612)
15 14 and 1 (188)
16 from 15 keep 78 (1)
17 exp Muscle, Skeletal/ or myofascia.mp. or exp Muscles/ (400042)
18 17 and 3 (68)
19 from 18 keep 7,11,18-19,27 (5)
20 8 or 11 or 13 or 16 or 19 (22)
21 from 20 keep 1-22 (22)
Database: CINAHL - Cumulative Index to Nursing & Allied Health Literature <1982 to September Week 4 2004>
Search Strategy:
--------------------------------------------------------------------------------
1 exp Headache/ or cervicogenic headache.mp. (3137)
2 cervical manipulation.mp. (76)
3 from 2 keep 20 (1)
4 cervical headaches.mp. (4)
5 from 4 keep 4 (1)
6 exp Physical Therapy/ or cervical mobilization.mp. (23934)
7 1 and 6 (101)
8 from 7 keep 15,18,24,48-49,69,89,93 (8)
9 aerobic exercise.mp. or exp Aerobic Exercises/ (5154)
10 1 and 9 (8)
11 Pain/ or pain modulation.mp. (9967)
12 9 and 11 (90)
13 from 12 keep 52 (1)
14 EXERCISE/ or THERAPEUTIC EXERCISE/ or exercise.mp. (19600)
15 Trigeminal Nerve/ or trigeminocervical.mp. (59)
16 from 15 keep 8 (1)
17 3 or 5 or 8 or 13 or 16 (12)
18 from 17 keep 1-12 (12)