Dunning JR, Butts R, Mourad F, Young I, Fernandez-de-las Peñas C, Hagins M, Stanislawski T, Donley J, Buck D, Hooks TR, Cleland JA.  BMC Musculoskeletal Disorders. 201;17(64). doi: 10.1186/s12891-016-0912-3

Abstracted by: Kristen Fisher, SPT from Missouri State University, Springfield, MO.

This randomized clinical trial, which was performed in multiple sites around the United States, evaluates the efficacy of treating cervicogenic headaches with manipulation versus mobilization with implementation of an exercise program. Cervicogenic headaches (CH) is reported to appear in approximately 4.1% of the population and is the cause of 16-20% of chronic headaches (Sjaastad & Bakketeig, 2008). The International Headache Society has outlined several criteria for a chronic headache to be diagnosed as CH. The following are criteria of CH: 1) pain referred from regions of the neck and felt in one or more areas in the head; especially around the occulo-fronto-temporal area, 2) clinical/imaging evidence of a lesion within the cervical spine or musculature that is a valid cause of headaches, 3) Evidence that the lesion is causing the pain by pain with activation, or decreased pain with inhibition, 4) Pain resolves within 3 months post successful treatment (Van Suijlekom et. al, 2010). Physical therapy intervention for cervicogenic headaches has included postural stabilization, therapeutic exercise, mobilization, manual therapy, and manipulation of the cervical spine. Many studies have been performed to analyze the effectiveness of a combination of interventions in treating cervicogenic headaches. However, no study to date has investigated the benefits of manipulation alone versus mobilizations.

The study initially included 251 patients who presented to 8 different outpatient therapy clinics with complaints of headaches were screened for eligibility for the study. The administering therapists utilized the criteria set in place for Cervicogenic Headaches International Study Group including 1) unilateral head pain that starts in posterior neck 2) pain triggered by neck movement/sustained positions 3) reduced range of motion in the cervical spine specifically less or equal to passive range of 32° in the Flexion-Rotation Test 4) pain elicited by pressure over one of the upper cervical joints and 5) moderate – severe non-throbbing pain. Inclusion criteria also included headache frequency of at least 1 per week and experiencing the headache for a minimum of 3 months. Exclusion criteria included bilateral headaches, presence of migraine headaches, presence of hyperreflexia, sensory disturbances, or muscle wasting, and any red flag criteria including tumor, fracture, metabolic disease, rheumatoid arthritis, osteoporosis, and resting hypertension. After screening for inclusion and exclusion criteria, 110 patients were eligible to continue with the study. Participants were randomly divided into a manipulation intervention group (n = 58) and into the mobilization and exercise intervention group (n= 52).

Outcome measurements for the study were dependent on self-reporting of headache intensity on the Numeric Pain Rating Scale (NPRS) and frequency of headaches. The Neck Disability Index (NDI) was a secondary outcome measurement tool. Treatment specifics for the manipulation groups included C1-2 and/or T1-2 high velocity low amplitude (HVLA) thrust performed on at least one of the 6-8 treatment sessions at the therapist’s discretion. At the other sessions, those segments could be revisited for manipulation or other hypomobile segments could receive a HVLA thrust manipulation. The mobilization and exercise group were treated with joint mobilization of C1-2 and/or T1-2 and at least one other spinal articulation in the cervical through thoracic segments also at the therapist’s discretion on at least one of the 6-8 sessions with incorporation of cranio-cervical flexion exercises and 10 minutes of progressive resistance exercise of the shoulder girdle musculature.

Patients were followed up at 1 week, 4 weeks and 3 months post baseline assessment. The overall outcome is that the manipulation group experienced higher statistically outcomes than the mobilization and exercise group. Subjects in the manipulation intervention decreased scores on the NPRS at all of the follow-up periods with a between-group percentage change in headache intensity of 36.58%. The NDI proved to be a grander secondary source with a percentage chain in disability of 35.56% from baseline to 3 months. It was also shown that the manipulation group experienced less frequency of headaches at each follow-up period as well.

Personal Commentary:

The authors mentioned that until recently, HVLA thrust manipulations of the upper cervical spine have been avoided and considered taboo in practice due to increased risk for severe adverse effects. A manipulation at the upper cervical segments increases the risk for vertebrobasilar stroke and other complications working so close to the brain stem. This article suggests that a well-educated and practiced clinician can decrease cervicogenic headache intensity, frequency, and associated disability through a HVLA manipulation greater than any other skilled intervention. The article also mentioned that other studies have been completed to assess the effectiveness of manipulation and mobilization, but that the current study was better as it separated the two.

This article lacks longevity and fails to answer the question why manipulation works better than any other skilled intervention. Manipulation addresses hypomobile segments and stimulates inhibition of pain fibers by a quick stretch to the nociceptive fibers. The subjects do experience decrease in pain and headache intensity, but manipulation as a single intervention does not address how to keep that segmental mobility the HVLA thrust created. It is essential for a skilled intervention to address strength, endurance, and decreased mobility deficits that a manipulation does not attend to. The importance of including manipulation in a plan of care is greatly suggested by this article, however the plan of care should be extended to postural control and therapeutic exercise as well in order to avoid re-occurrence of cervicogenic headache symptoms.


Sjaastad, O. & Bakketeig, L.S. (2008). Prevalence of cervicogenic headache study of headache epidemiology. Acta Neurologica Scandinavica, 117: 173-180. doi: 0.1111/j.1600-0404.2007.00962.x

Van Suijlekom, H. Van Zundert, J., Narouze, S., van Kleef, M., & Mekhail, N. (2010). Cervicogenic Headaches. World Institute of Pain, 10 (2): 124-130. doi: 10.1111/j.15332500.2009.00354.x