Recent Advances in Pain Research: Implications for Chronic Headache

Abstract

Within the last 2 decades there has been an explosion in new information on mechanisms underlying pain. Unfortunately this information has not resulted in a similar improvement of our handling of patients with chronic pain including chronic musculoskeletal pain. Neuronal hyperexcitability, which apparently is a key phenomenon in many (if not all) types of chronic pain results in changes in the nervous system from the level of the peripheral nociceptor to the highest cortical centers in the brain. The neuronal plastic changes in chronic pain conditions makes the nociceptive system amenable for treatment with several traditional as well as untraditional types of interventions. Two treatment areas that seem worth exploring within chronic pain including headache concerns preventive measures and endogenous pain modulation.

Introduction

In the last two decades there has been a considerable advance in our understanding of pain mechanisms. Molecular targets such as the vanilloid and opioid receptors together with different ion channels have been unraveled (1 –4), spinal neuroplastic changes following a sustained noxious input have been described (5) and it has been possible by new imaging techniques to get insight into the brain and observe the processing of acute and chronic pain (6). It is clear that acute and postoperative types of pain are now better treated than previously, but progress in the treatment of chronic pain conditions has been at a much slower pace. We are still left with many chronic pain patients, including musculoskeletal types of pain such as headache, for whom we are unable to provide any beneficial treatment. The expansion in terms of DNA sequencing of receptors and ion channels, and the development in understanding neuroplastic changes in the nervous system following noxious input, stands in rather sharp contrast to our treatment options for chronic pain patients. It is not at all clear that the number of chronic pain patients has been reduced and in many instances treatment of chronic pain is still a disappointing story (7).

The following three areas, where basic research may contribute to a better handling of chronic types of pain, including musculoskeletal pain, will be described: mechanisms of hyperexcitability and its treatment; prevention of chronic pain; and utilization of endogenous pain modulation.

Key findings in basic pain research

The pain signalling system was previously thought to be a hard-wired single line system connecting the periphery with centres in the brain. It is now recognized that the system analysing noxious information consists of several channels with many synaptic relays, feedback circuits and a high degree of inborn plasticity.

Acute nociception

In normal subjects, transduction of noxious mechanical, chemical and thermal information occurs via specialized receptors in the cell membranes. Heat detection, for example, takes place via the vanilloid receptor (VR-1), which gates cations across the cell and which also is activated by capsaicin the hot compound in chili pepper (for review see 1 –3). Mechanical information may take place, for example via acid sensing Na + channels that are particularly sensitive to acidification of the environment as seen during inflammation (8). The subsequent translation into action potentials results in the release of the excitatory amino acid glutamate from the primary afferent terminal in the spinal cord, which via non NMDA receptors (AMPA and kainate) generates a fast excitatory post-synaptic potential (9, 10). This is the background for the acute brief sensation of a noxious stimulus. Already under these conditions descending and segmental control systems modulate the noxious event. If the noxious stimulus becomes more prolonged or intense other substances such as the neuropeptides (substance P, neurokinin A CGRP) and other neuromodulators are co-released and a subsequent recruitment of the NMDA receptor systems takes place, giving rise to activation of voltage gated calcium channels with more prolonged depolarization of the cell. These phenomena may explain the gradual wind-up of the cell after a sustained c-fibre input and the temporal summation of the human pain experience after repetitive stimulation at noxious inntensity (1, 2, 4, 5).

Early sensitization

Peripherally

In the periphery the sensitivity of the cellular receptors can be changed by various agents such as bradykinin, serotonin, adrenaline neurotrophins and adenosine produced during tissue injury, e.g. inflammation. Phosphorylation of channels or receptors in the cell membrane is responsible for a change in membrane excitability. For example, by activation of intracellular proteinkinases (PKA and PKC) phosphorylation of the so-called SNS (sensory specific sodium) channels belonging to the TTX resistant type occurs. It has been shown that after injury to nerve axons there is a clear change of sodium channel expression, with downregulation of the SNS/PN3 sodium channel genes and a corresponding upregulation of a previous silent type III channel gene (11). The end result is a change of the cation current through the channel. Other receptors have been cloned, including the P₂X₃ present in small sensory neurones (12). Mediators like PGE2, 5-HT and adenosine are able to alter voltage thresholds of certain ion channels and thereby facilitate neuronal transmission. Importantly, sensitizing molecules may affect different receptors, so eliminating one sensitizing agent will probably not block peripheral sensitization, but only one aspect of it.

Centrally

At the central terminal of the primary afferent a similar phosphorylation of ion channels and receptors occurs because of activation and mobilization of intracellular facilitating systems from a persistent peripheral noxious input. Both NMDA-dependent and AMPA receptors are involved in this facilitation of cells which in part involves an increased influx of calcium into the cell. The increased NMDA receptor gating may also be mediated indirectly via various G-protein-coupled receptors such as NK, metabotropic and tyrosine kinase receptor systems that exert their activity via second messenger systems. The results of this process termed central sensitization are: (i) a reduction of threshold to depolarize the cell; (ii) a cellular activity that outlasts the peripheral nociceptive input; and (iii) a spread of cellular activity to neighbouring cells (2, 5, 10).

Long-term sensitization

There is gradual change from transient to persistent sensitization. These phenomena can occur both during inflammatory and neuropathic pain conditions but are most readily seen in the latter types. There are differences between the changes seen after inflammation and after nerve injury.

Inflammation is due to release of a cascade of substances from various cells in the vicinity of the injury. Macrophages release cytokines (IL1, IL6 and TNF alpha and nerve growth factor). Damaged cells release ATP and protons, mast cells expell histamine, serotonin, prostaglandins and other arachidonic acid metabolites with an ability to sensitize the nerve terminals. During inflammation there is an upregulation of a series of receptors (VR1, SNS, SNS-2) and peptides (CGRP, substance P) These molecules are transported antegrade to the injury region, where they increase sensitivity to inflammatory mediators and when transported to the central terminal they cause central sensitization. Another important aspect of the persistent sensitization after inflammation involves the expression of neuropeptides in large A-fibre cells, which normally are only expressed in small c-fibre DRG cells. This ‘phenotypic switch’ of A-fibre cells into c-fibre mimicking cells means that A-fibre input now may be translated into C-fibre responses (non-noxious touch is experienced as pain).

Nerve injury

Nerve injury causes marked changes in the periphery and centrally that are different from those seen after inflammation. This change is due mainly to a loss of fibres, with a reduction of substance P, CGRP, VR1 and SNS. A phenotypic switch of cells is also seen with a changed expression of receptors, ion channels and transmitters (2). Loss of small afferent c-fibres causes sprouting of large A-beta fibres from deeper laminae and into the outer laminae containing normal noxious responding cells. This spinal reorganization with altered connectivity may be one reason for touch-evoked pain that is resistant to pharmacological modulation.

Neuronal hyperexcitability: clinical significance

A puzzle in current pain research is the transition of acute pain into chronic pain states. In headache this distinction between acute and chronic states is clearly reflected in the major headache conditions: migraine with its repetitive acute pain attacks, always with return to normal activity within hours; and chronic tension-type headache, which is a classical example of a chronic pain condition.

It is not yet clear what determines the transition of acute into chronic pain, and why patients may differ in their responses and behaviour despite an apparent similar initial noxious event.

The translation of cellular and molecular events following persistent noxious activation is the development of a neuronal hyperexcitability with pain elicited by low threshold non-noxious input that persists after cessation of stimulation and a spread of pain to uninjured areas (13 –17).

Neuronal hyperexcitability is today considered a core phenomenon in patients with chronic pain irrespective of involved tissue (musculoskeletal, visceral, nervous tissue). Several clinical examples of such neuronal hyperexcitability are present in the clinical literature. For example, in tension-type headache, fibromylagia and other musculoskeletal disorders, tender points are present outside the exact injury side and so is the cutaneous allodynia. In musculoskeletal pain hyperexcitability can also be demonstrated, for example by the presence of a lowering of threshold to activate neurones and by a spread of pain not only outside the involved area, but also from muscle to distant skin areas within the same dermatome (18 –20). In patients with tension-type headache and in patients with cluster headache, cuataneous hyperalgesia may be present during periods with ongoing pain. In different types of muscle pain, such as fibromyalgia, tension type headache and chronic whiplash syndrome, a lowering of the pain threshold can also be demonstrated locally and in some studies even also more generally (21 –23).

A prerequisite for selection of the optimal treatment requires an exploration of the pain, its manifestations, location, cause and possible mechanisms. At present the pharmacological possibilities for treating musculoskeletal pain, including headache, are few and it has until now mainly been limited to classical analgesic compounds such as NSAIDs, paracetamol and opioids. There have, however, been a few attempts to modulate chronic muscle pain by utilizing the NMDA antagonist ketamine. It is expected that new pharmacological agents will target specific molecular mechanisms in the traumatic injury or inflammation cascade.

Can chronic pain be prevented?

A central point in current pain research and treatment is the transition of acute pain into chronic pain states, the identification of risk factors for such transition and whether it in fact can be prevented. The nociceptive system is not a fixed static system, but a dynamic neuronal network that continuously alters its response characteristics depending on the prior exposure to noxious stimuli (for review see 1 –3, 5, 10).

Based on a series of experimental studies on neuronal hyperexcitability and the induction of sensitization the question was raised in 1988 to a clinical audience (24) of whether neuronal hyperexcitability and pain could be prevented by an analgesic treatment before injury as opposed to after the injury. This issue, termed pre-emptive analgesia, has been explored extensively experimentally and clinically within the last decade (for review see 5).

In classical studies Woolf and others observed that the functional plastic changes with sensitization of second order neurones could be prevented by a local anaesthetic blockade or by an opioid administration before, but not after, injury. While the experimental evidence for prevention is quite strong the evidence in clinical pain is much weaker (25 –27). In fact careful controlled clinical trials addressing the challenging concept that sensitization and hence pain can be prevented by pre-treating the individual before the insult instead of after have failed to show long-lasting and clinical relevant reduction of pain (28 –30). What are the reasons for this apparent discrepancy between basic and clinic? The following factors have been suggested to play a role (25):

• Clinical pain is more complex than experimental pain, often involving different mechanisms.

• Patients are heterogenous in terms of their pain.

• Sensitization may be present before pre-emptive treatment is started.

• Pre-emptive analgesia is insufficient or lasts for too short a time.

The strong experimental evidence for a pre-emptive effect, together with the increasing understanding of neuroplasticity of the nervous system, calls for a re-examination of the pre-emptive phenomena in appropriate clinical conditions. In musculoskeletal pain the evidence for a preventive effect is as weak as it is in the neuropathic pain litterature. However, it is tempting to suggest that preventive measures are what is done in our everyday living when continually exercizing and moving and thus preventing static muscle activity.

Endogenous pain modulation

The failure of many pharmacological agents to relieve chronic pain (31 –34), together with a complexity of regulation and side-effects of several current existing drugs, raises the question of whether alternative procedures for pain control can be brought into action clinically.

Historically, experimental and clinical pain has been regarded as a result of tissue damage, with a close relationship between noxious stimulus intensity and degree of perceived pain. This concept is a direct consequence of a hard-wired, single line labelled relation between noxious stimulus and pain signal in the brain. However, the input and its consequence is not fixed, but can be modified by a range of psychological and emotional factors, such as attention, distraction, anxiety and hypnosis in both experimental and clinical conditions. An extensive literature indicates that this psychological influence on pain is best described in a framework of a powerful neuronal network originating in different brain structures with multiple targets and a faceted pharmacology (35 –38). For example, the cognitive-behavioural model of pain proposes that pain can not be understood only on the basis of tissue damage, but that cognitive factors (expectation, reflection, memory, etc.) and behavioural factors (social environment, work, physical activity, etc.) influence our response to the noxious stimulus (39). Along these lines it is necessary when looking at improvement also to consider social and psychological outcome parameters. Psychologists use this in their everyday work when seeing chronic pain patients, each arriving at the pain clinic with their own combination of expectations, attitudes, beliefs, past memories and their own set of resources. This recognition should make everyone humble when seeing a chronic pain patient, but also optimistic because it represents a challenge and leaves a set of treatment possibilities that goes beyond the potential benefit of drugs and various surgical procedures.

One challenge is to understand in more detail the neurobiology of the psychological modulation. In fact that work started 3–4 decades ago by analysing the anatomy and physiology of pain-modulating networks. Briefly these modulatory systems consist of both facilitating and inhibitory networks, which via up- and down-stream pathways in the brainstem control pain-responding neurones in the spinal cord and in the brain (36 –38). At the spinal level intrinsic interneurones and descending systems from the brain control pain transmission. At other pain relays in the brainstem, thalamus and cortex, direct and indirect inhibitory and facilitating systems are present. The inhibitory role of the endogenous pain modulation is indicated by the recruitment of pain inhibition during various psychological conditions, such as hypnosis, expectations or placebo analgesia (40), and it is also known that experimental and clinical pain can be modified by cognitive processes (35). These findings suggest that the nociceptive system has an inborn plasticity that permits it to operate within a spectrum ranging from intense hyper-senstivity to intense hyposensitivity. The latter has, in chronic pain, including that seen in chronic headache, only been utilized to a limited extent and there is a need to explore the potential of this type of pain modulation in further detail.

Conclusion

Our understanding of pain mechanisms has increased considerably, but this progress has so far not been paralleled by a similar advance in our treatment of clinical pain. A rethinking is necessary and examination of untraditional ideas may reveal novel findings about how chronic pain can be explored, treated and even prevented.