Abstract
Despite the tremendous socoeconomic impact of tension-type headache surprisingly little is known about the underlying pathophysiology and treatment. Existing evidence and theories about the relation between central and peripheral mechanisms are discussed. Central sensitization is probably the most important key to understand this widespread disorder. An effective prevention or reversal of this central sensitization will probably be of major importance in future treatment strategies.
Introduction
Tension-type headache is the most prevalent type of headache and represents therefore one of the most costly diseases in modern society (1, 2). Surprisingly little research on this disease has actually been carried out, and knowledge about key pathophysiological issues such as the nature and site of the noxious stimulus is extremely limited. The main reason for this is the previous lack of a proper classification. Before 1988, no precise definition of tension-type headache was available, and several terms such as muscle contraction headache, tension headache, psychogenic headache, psychomyogenic headache, or stress headache were used. Furthermore, because mental stress and tension are the most frequently reported precipitants of tension-type headache (3, 4), many scientists equated this with causative factors and did not try to describe the pathophysiology and mechanisms behind the disorder.
In the International Headache Classification in 1988 (5), tension-type headache was precisely classified and defined by means of operational criteria. The subdivision into an episodic and a chronic form, and into types with and without a muscular factor, was developed mainly on the basis of the experience of a group of experts and not on the basis of scientific evidence, which at that time was very limited. For decades, it has been a matter of debate whether the pain originates from the periphery or from the central nervous system (6–8). Clinical and laboratory investigations to substantiate any of these hypotheses are few. Although the pain clinically resembles pain from the myofascial tissues, modern pain physiology indicates that both peripheral and central mechanisms may contribute. In the following, the current pathophysiological theories will be presented.
P eripheral mechanisms
The increased tenderness that is the most pronounced and consistent finding in these patients probably represents the activation of peripheral nociceptors (6–10). This increase in tenderness with increasing frequency and intensity in tension-type headache but not in migraine is the crucial factor in the understanding of the pathophysiology. Human experimental models of peripheral muscle pain are actually few but they have demonstrated that myofascial tenderness can precede the headache and is therefore very likely to be involved in the underlying mechanism (11). Nevertheless, local experimental pain models where different algogenic substances are injected into the trapezius muscles demonstrate that patients with a history of episodic tension-type headache develop significantly more local pain than healthy controls but that none of the groups develops headache (12). The underlying pain mechanisms for tension-type headache may therefore be an effect of either temporal or spatial summation of peripheral stimuli or both in predisposed individuals.
The texture of pericranial, shoulder and chewing muscles is often altered in tension-type headache, with generalized increased consistency. Such findings have previously only been detected by manual palpation, but a newly invented and validated instrument, a hardness meter, has confirmed this observation (13–15). The pathophysiological background and significance of these findings have not yet been clarified.
C entral mechanisms
Tension-type headaches are generally reported to occur in relation to emotional conflict and psychosocial stress, but the cause–effect relationship is not clarified. Stress and mental tension are the most frequently reported precipitating factors, but they occur with similar frequency in tension-type headache and in migraine (3, 4). These results are in concordance with the findings of generally normal personality profiles in individuals with episodic tension-type headache, whereas studies of subjects with the chronic form often reveal a higher frequency of depression and anxiety (16–18). As in other chronic pain disorders, psychological abnormalities in tension-type headache may rather be viewed as secondary rather than primary (16), and anxiety and depression are probably comorbid with chronic tension-type headache.
The decreased pain, thermal and electrical thresholds that have been reported in chronic tension-type headache patients (19–21) probably represent a central misinterpretation of the incoming signals. A recent study of the stimulus–response function to mechanical pressure (22) demonstrated for the first time that chronic tension-type headache has a physiological basis and is caused at least partly by qualitative changes in the central processing of sensory information, i.e by a central sensitisation, as with other chronic pain conditions.
On the basis of measurements of the so-called exteroceptive second silent period, it was previously suggested that the limbic pathways to the brain stem were involved (23, 24). The exteroceptive second silent period may provide important information about the central mechanisms but the applied methodology has been challenged, and negative studies have recently been published (25, 26). Biochemical defects either in the opioid system or in the production of neurotransmitters has also been suspected because the nociceptive flexor reflex, a spinally organized reflex, is decreased in chronic tension-type headache (27), but no recent studies have confirmed these findings. Studies of neuropeptides and endorphins in these patients have mainly been negative (28, 29), and only one former study (30) has noted increased metenkephalin in the cerebrospinal fluid in chronic tension-type headache. It is very likely that an impaired supraspinal modulation of the repeated peripheral stimulation may play a part in these chronic pain disorders but a precise molecular identification is lacking and the cause–effect relationship to pain continuous for decades is yet unclear. It is therefore most likely that these various abnormalities may result in or be a function of the disturbed balance between peripheral input and central modulation. The primary eliciting cause and the evolution of pain are, however, still unknown.
Nitric oxide (NO) plays an important role in the pathophysiology of primary headaches including chronic tension-type headache. An NO synthase inhibitor thus reduced headache and muscle hardness in a recent study (31, 32), whereas the NO donor glyceryl trinitrate caused more headache in patients with chronic tension-type headache than in healthy controls. Patients also developed a delayed headache of the tension type (33). These findings suggest that glyceryl trinitrate-induced delayed headache can be used as a valuable human model of tension-type headache and that NO-related central sensitization may be an important factor in the underlying pathophysiology (33, 34).
In contrast to clinical observations, epidemiological studies document that migraine and tension-type headache are different disorders although they coexist in many patients (4,35,36). As episodes of tension-type headache are more pronounced and frequent in individuals with coexisting migraine than in nonmigraineurs (4, 35), it indicates that migraine can be a precipitating factor in tension-type headache in genetically predisposed individuals. A genetic predisposition to chronic tension-type headache, reflected in a 3.18-fold increased risk in first-degree relatives compared with the general population, has been published (37, 38), but the mode of transmission seems to be complex.
On this basis, it can be concluded that the underlying pain mechanisms in tension-type headache are highly dynamic, as tension-type headache represents a wide variety of frequency and intensity, not only between individuals, but also within the individual subject over time. The initiating stimulus may be a condition of mental stress, unphysiological motor stress, a local myofascial release of irritants, or a combination of these. Secondary to the peripheral stimuli, the supraspinal pain- perception structures may become activated, and because of the central modulation of the incoming stimuli, a self-limiting process will be the result in most individuals. As chronic tension-type headache usually evolves from the episodic form (39, 40), an effective prevention of this evolution from a peripheral mechanism in episodic to a central mechanism in chronic tension-type headache will therefore be of major importance in future treatment strategies.
Regrettably, there have been no new developments in the treatment possibilities. As a result of this and the lack of scientific interest from the pharmaceutical companies, the treatment is widely nonspecific, very often ineffective and consists mainly of simple analgesics. Therefore, our present pathophysiological knowledge hopefully may lead to increased scientific interest and development of effective treatment strategies in the future.