Headaches from neck pain: should physical therapy for cervicogenic headaches look beyond the cervical spine?
Headaches from neck pain are one of the most frustrating presentations in the clinic. Cervicogenic headache physical therapy has traditionally focused on the cervical spine — and for good reason. Manual therapy to the cervical joints works, and exercises targeting the deep neck flexors help too. The research supports both.
But what if the neck isn't the whole story?
A 2005 case report published by McDonnell, Sahrmann, and Van Dillen explored whether impairments in the scapulothoracic and lumbar regions — not just the cervical spine — could be contributing factors in cervicogenic headache. More specifically, they asked whether treating those regions could meaningfully reduce a patient's headache frequency and intensity.
The study
This was a case report, meaning it describes the examination and treatment of a single patient. Case reports sit at the lowest level of the evidence hierarchy — they cannot establish cause and effect, and their findings cannot be generalized. With that context in mind, this paper is best read as a clinical reasoning framework rather than proof of efficacy.
The patient was a 46-year-old male with a 7-year history of cervicogenic headache. His average pain was 5 out of 10, with peaks of 10 out of 10, and he reported constant symptoms. His Neck Disability Index score was 31 out of 50, placing him in the severe disability category. His symptoms were aggravated by using his arms — whether working with horses or sitting at a computer — and he frequently woke up at night from headache pain.
One important detail: one week before starting physical therapy, he received trigger point injections to the posterior cervical and upper trapezius region. These provided 24 to 48 hours of complete pain relief, and his pain had dropped to 3 out of 10 by the time he arrived for his first PT visit.
The examination
The treating therapist — working within Sahrmann's Movement Impairment Syndrome framework — assessed the cervical, scapulothoracic, and lumbar regions together rather than in isolation.
Key findings included forward head posture with excessive upper cervical extension, marked scapular abduction and depression bilaterally, thoracic kyphosis, and significant weakness of the middle and lower trapezius, rhomboids, and lower abdominals. The deep neck flexors were so weak that formal strength testing wasn't even possible at baseline.
One of the more clinically interesting findings came from a simple test: the examiner manually elevated and adducted the patient's scapulae while he performed cervical rotation. His range of motion increased by 10 degrees in both directions and his headache pain decreased. This suggested that scapular positioning was directly influencing his cervical symptoms — and became the basis for treatment.
The intervention
The patient was seen 7 times over 3.5 months. No manual therapy was used at any point.
Treatment focused entirely on active exercise and functional modification. The exercise program targeted abdominal strength and control, deep neck flexor activation, scapular retraction and elevation, and full shoulder range of motion without compensatory neck or lumbar movement. Functional instruction emphasized supporting the weight of the arms throughout the day — resting forearms on a desk, placing hands in pockets while standing — to reduce the constant downward pull on the cervical spine.
The results
The results across those 7 visits were notable.
By the fourth visit — just 25 days in — the patient was going several days at a time without any headache. By the seventh visit, headaches had dropped to once per week at an intensity of 1 out of 10. His NDI score fell from 31 to 11, shifting from severe to mild disability. Cervical rotation improved from 39 degrees (painful) to 50 degrees (pain-free), and cervical extension went from 25 degrees (painful) to 40 degrees (pain-free). Scapular position improved from 17.8 cm lateral to 11.4 cm lateral from midline.
At a 5-month follow-up phone call, he was sleeping through the night without medication, performing all activities of daily living, and managing flare-ups independently with his exercises.
Limitations of the study:
This is a single case report. There is no control group, no blinding, and no way to determine whether improvement was due to the intervention, natural history, regression to the mean, or the trigger point injections administered the week prior. The trigger point injection in particular is a notable confound — the patient was already improving before PT began.
No independent outcome assessor. The treating therapist was also involved in measuring and reporting outcomes, which introduces the potential for bias.
The primary outcomes — headache frequency and intensity — relied on patient self-report without a structured diary or validated headache frequency tool.
The passive correction of scapular position test used in the examination has not been formally validated as a diagnostic tool.
With only one patient, there is no way to know which components of the intervention — the scapular exercises, the lumbar work, the abdominal training, the functional modifications — were actually driving the improvement, or whether any of them were.
Clinical implications for cervicogenic headache physical therapy
At first glance, the results are compelling. A patient with 7 years of constant, debilitating headaches from neck pain improved dramatically with just 7 visits — and without a single manual therapy technique.
But the more important takeaway may be the clinical reasoning behind the approach.
The idea that scapular depression and abduction can transfer the weight of the upper extremities through the cervicoscapular muscles onto the posterior cervical spine is biomechanically plausible. The idea that lumbar extension drives compensatory cervical extension throughout the day is similarly logical. These are not novel concepts — regional interdependence has been discussed extensively in the PT literature — but applying them systematically to cervicogenic headache is less common in practice.
The passive correction of scapular position test is particularly worth noting. It takes seconds to perform and gives the clinician immediate feedback about whether scapular mechanics are contributing to cervical symptoms. If range of motion improves and pain decreases with manual scapular support, that's a meaningful clinical signal.
What this study cannot tell us is whether adding scapulothoracic and lumbar treatment to a standard cervical approach actually improves outcomes. That question has not been answered in a controlled trial. And given that manual therapy to the cervical spine has a solid short-term evidence base for CH, it is also worth asking: would this patient have done just as well — or better — with the addition of manual therapy?
Final thoughts
This is not a study you should use to change your protocol. It is something that should prompt you to look further than the cervical spine when you assess a patient with headaches from neck pain.
The cliche “shoulders down and back” is not always the answer. In this case, “shoulders down”, or scapular depression, may actually exacerbate cervicogenic symptoms.
Are the scapulae sitting low and wide? Is the patient extending through the neck every time they raise their arm? Is there a lumbar alignment that is feeding into prolonged cervical extension posture throughout the day? These are questions worth asking. The evidence for acting on the answers is still limited — but the clinical logic is sound.
At The LAB Doctors of Physical Therapy, we take a whole-body approach to cervical spine conditions, building treatment programs grounded in both the evidence and the individual in front of us. Click here to learn more about our services.
Citations
1.) McDonnell MK, Sahrmann SA, Van Dillen L. A specific exercise program and modification of postural alignment for treatment of cervicogenic headache: a case report. J Orthop Sports Phys Ther. 2005;35(1):3–15.