Physical therapy and exercise in headache


  • C Fernández-de-las-Peñas

    Corresponding author
    1. Department of Physical Therapy, Occupational Therapy, Physical Medicine and Rehabilitation, and
    2. Esthesiology Laboratory, Universidad Rey Juan Carlos, Alcorcón, Madrid, Spain
    Search for more papers by this author

César Fernández de las Peñas PT, PhD, Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos, Avenida de Atenas s/n, 28922 Alcorcón, Madrid, Spain. Tel. + 34 91 488 88 84, fax + 34 91 488 89 57, e-mail


To date the effects of physical therapy or exercise in headaches has hardly been analysed. Conflicting results in previous studies might be due to the low number of clinical trials plus an indiscriminate application of different techniques. As different pathogenic mechanisms seem to be involved in each headache, the most appropriate techniques might differ in each case. Future research should address the evaluation of specific physical therapy procedures for each headache disorder, and the identification of those subgroups of patients who are most likely to benefit from a particular intervention.


Several non-pharmacological approaches have been proposed for the management of headaches (exercise, physical therapy, cognitive therapy, relaxation therapy, or biofeedback), with physical therapy being one of the most widely used. While there is sufficient evidence that pharmacological or behavioural strategies can be effective for headaches, the evidence for the effectiveness of physical/manual therapy remains to be elucidated. Here we shall briefly review current scientific evidence for physical therapy in headaches as well as the effects of physical therapy and exercise on pain mechanisms. We shall also outline a clinical approach for the management of tension-type headache based on recent studies. Finally, we shall raise some points for future research.

Scientific evidence for physical therapy in headache disorders

To date the effectiveness of physical therapy or exercise in headaches has hardly been analysed. A systematic review focused on tension-type headache concluded that there was insufficient evidence to support or refute the efficacy of physical therapy, exercise or spinal manipulation for this headache disorder (1). In contrast, a recent review about manual therapy in cervicogenic headache found strong evidence that spinal manipulative therapy was effective for reducing headache intensity, headache duration and medication intake in this type of headache (2). To our knowledge there are no published reviews on the effectiveness of physical therapy or exercise in migraine, though some studies suggest that spinal manipulation might be also effective for migraine headache. Previous conflicting results might be due to the low number of trials plus an indiscriminate application of techniques. As different pathogenic mechanisms seem to be involved in each headache, the most appropriate techniques might be different in each case. For instance, cervicogenic headache is apparently related to joint dysfunctions in the upper cervical segments. Therefore, spinal manipulation would conceivably be the choice for this headache. On the other hand, there are suggestions that myofascial trigger points might be involved in the genesis of tension-type headache, so that soft tissue approaches could be more appropriate for this disorder. A strategy aimed at potential pathogenic mechanisms would probably make physical therapy more effective in different headache disorders. Moreover, the results could improve further if we get to identify those subgroups of patients who are most likely to benefit from these interventions.

Neurophysiological effects of physical therapy and exercise

Physical therapy and exercise have multiple physiological effects. Nevertheless, there are two main mechanisms that may account for the relief of pain in headache patients: reduction of peripheral sensitization, and activation of descending inhibitory pathways.

It is well known that central sensitization is one of the mechanisms that lead to chronicity in headache. Sensitization of the central nervous system has been related either to prolonged nociceptive inputs from the periphery (due to the release of algogenic substances in the vicinity of peripheral nociceptors) or to impairment of the central descending inhibitory system. Shah et al. demonstrated that muscle trigger points might contribute to peripheral sensitization when they found higher levels of algogenic substances in active trigger points (those points whose referred pain patterns are responsible for the patient's symptoms) as compared with latent trigger points or control points (3). As central sensitization is a dynamic condition under the influence of peripheral nociceptive inputs (4), it would probably decrease if peripheral inputs could be identified and subsequently eliminated. Accordingly, one mechanism by which physical therapy could be effective would be the reduction of peripheral sensitization (and, consequently, the reduction of central sensitization) induced by active muscle trigger points.

The second mechanism is the activation of descending inhibitory pathways by exercise. Köseoglu et al. found that exercise induced an increase in β-endorphin levels in patients with migraine (5). Similar results have been reported by other authors who have demonstrated that exercise has a general hypoalgesic effect. Physical therapists should include the prescription of exercise programs for their headache patients in order to activate the descending inhibitory pathways.

Clinical approach for physical therapy in chronic tension-type headache

Our research group has demonstrated that myofascial trigger points are conspicuous in patients with tension-type headache. A myofascial trigger point is a tender spot located within a taut band of a skeletal muscle, whose stimulation (maintained pressure or palpation during both contraction and stretching of the muscle) elicits a referred pain in a distant region. Trigger points are considered active if such referred pain reproduces the patient's symptoms, whereas latent trigger points provoke a non-familiar pain. We have found active trigger points in the upper trapezius (6), suboccipital (7), sternocleidomastoid (8), temporalis (9) or superior oblique (10) muscles in patients with chronic tension-type headache. The referred pain evoked from those active trigger points upon examination resembled the usual head pain. Furthermore, the patients with active trigger points in the above-mentioned muscles showed more severe headache parameters (greater headache intensity and frequency, or longer headache duration) than those patients with latent trigger points (6–9). These results have led us to formulate a pain model in which tension-type headache can be explained to some extent by the activity of trigger points in the cervical, head and shoulder muscles (11).

Patients with chronic tension-type headache also show motor control impairments in the deep flexor muscles (12), and selective reduced cross-sectional area in the suboccipital muscles (13). In addition, a recent study found that atrophy in the suboccipital muscles was associated with active TrPs in the same muscles (14). As all these muscle disorders may perpetuate central and peripheral sensitization and all of them may be connected (14), their management should be considered in physical therapy programmes for patients with tension-type headache.

Future research: clinical prediction rules

Based on current knowledge, new trials should assess the effectiveness of those physical therapy techniques that may influence the pathogenic mechanisms operating on each headache disorder. What is more, future clinical studies should be conducted on particular subgroups of patients rather than heterogeneous headache populations. This strategy could allow a better understanding of the results obtained with physical therapy programmes. The development of clinical prediction rules can assist us in the selection of the patients. Clinical prediction rules combine several variables from medical histories, clinical examinations and self-report outcomes in order to classify patients according to their expected response to treatment. Several clinical prediction rules have been found to be useful in classifying patients with neck or low back pain who are likely to benefit from a particular intervention, e.g. spinal manipulation. However, there is a lack of this kind of research in headache disorders (15). Our group has recently developed a clinical prediction rule that provides the potential to identify those patients with chronic tension-type headache who are likely to experience short-term and 1-month follow-up success with a muscle trigger point therapy approach (16). An increase in chance of experiencing a successful intervention (+LR 5.9) from 54% to 87.4% occurred when patients satisfied the following four variables: less headache duration (< 8.5 h per day), less headache frequency (< 5.5 days/week), lower body pain assessed by the SF-36 questionnaire (< 47/100 points) and higher vitality (> 47.5/100 points). Future studies are now required with larger series of patients.

Conflicts of interest

The author has declared no conflicts of interest.