2 Pain Mechanisms and Their Importance in Clinical Practice and Research
The Massachusetts General Hospital Handbook of Pain Management
Pain Mechanisms and Their Importance in Clinical Practice and Research
Isabelle Decosterd and Clifford J. Woolf
After great pain, a formal feeling comes
The Nerves sit ceremonious, like Tombs
The stiff Heart questions was it He, that bore,
And Yesterday, or Centuries before?
— Emily Dickinson, 1830–1886
I. Fundamental pain mechanisms
1. Response to acute painful stimuli
2. Peripheral sensitization
3. Central sensitization
5. Structural reorganization
II. Toward a new conceptual approach for the understanding of pain
III. Implications for therapeutic approaches
IV. Implications for evaluation of efficacy of new therapies
It has become increasingly clear from animal models and from preclinical and clinical studies that multiple mechanisms operating at different sites and with different temporal profiles induce chronic pain syndromes. The identification of these mechanisms may provide the best lead to effective pain treatment, especially in the case of novel treatments. Whereas primary disease factors initiate pain mechanisms, it is the pain mechanisms, not the disease factors, that produce chronic pain. Identifying the causes of diseases is important, but it is also essential to differentiate them from pain mechanisms. Because a particular disease may activate several different pain mechanisms, a disease-based classification is useful primarily for disease-modifying therapy, but less for pain therapy. Similarly, symptoms are not equivalent to mechanisms, although they may reflect them. The same symptom may be produced by different mechanisms and a single mechanism may elicit different symptoms.
In this chapter, we propose a new way of analyzing pain, based on the current understanding of pain mechanisms, and we show the implications of this for assessing pain in individual patients and for evaluating new forms of diagnosis and therapy.
I. FUNDAMENTAL PAIN MECHANISMS
1. Response to acute painful stimuli
Acute pain is initiated by a subset of highly specialized primary neurons, the high-threshold nociceptors, innervating peripheral tissues (skin, muscle, bone, viscera). The peripheral terminals of these sensory neurons are adapted so as to be activated only by intense or potentially damaging peripheral noxious stimuli. These receptors are functionally distinct from the low-threshold sensory fibers, which are normally activated only in response to nondamaging low-intensity innocuous stimuli. Nociceptor transduction mechanisms involve activation of any of the following:
Temperature-sensitive receptor ion-channel sensors [such as vanilloid (capsaicin) receptor subtype 1 (VR1) and vanilloidreceptor–like protein 1 (VRL1)]
Channels (yet to be identified) sensitive to intense mechanical deformation or stretch of the membrane
Chemosensitive receptors [such as VR1, dorsal root acidsensing ionic channel (DRASIC) and ATP-gated ion channel type 3 (P2X3)] activated by protons, purines, amines, peptides or growth factors, and cytokines released from damaged tissue or inflammatory cells
2. Peripheral sensitization
The sensitivity of the peripheral terminal is not fixed, and its activation either by repeated peripheral stimulation or by changes in the chemical milieu of the terminal increases the excitability of the terminal and decreases the threshold for initiation of an action potential in the primary sensory neuron. This phenomenon is referred to as peripheral sensitization. Peripheral sensitization reflects changes in the channel kinetics caused both by transduction in ion channels themselves (autosensitization, resulting from prior activation) and by an increase in excitability of the terminal membrane (heterosensitization, initiated by sensitizing stimuli such as inflammatory mediators that do not activate the usual pain transducers).
Autosensitization of vanilloid receptors (VR1, VRL1), for instance, may represent both (a) conformational changes of the receptor secondary to the external heat stimuli and (b) the entry of calcium through the transducer itself, leading to activation of protein kinase C, which phosphorylates VR1. Heterosensitization is driven by sensitizing agents such as prostaglandin E2, histamine, bradykinin, serotonin, and neurotrophic factors that can activate intracellular kinases. Intracellular kinases have the ability to phosphorylate and change the activity state of voltage-gated sodium channels such as SNS (the sensory-neuron–specific sodium channel PN3/Nav1.8).
3. Central sensitization
In addition to changes in the sensitivity of the nociceptor peripheral terminal, post-injury pain hypersensitivity is also an expression of modulation of nociceptive synaptic transmission in the dorsal horn of the spinal cord. This is called central sensitization. Input from nociceptors to the spinal cord evokes an immediate sensation of pain that lasts for the duration of the noxious stimulus and also induces an activity-dependent functional plasticity in the dorsal horn that outlasts the stimulus. The increased excitability is triggered by peripheral nociceptor input, releasing excitatory amino acids and neurotransmitters that act on spinal cord postsynaptic receptors to produce inward currents, as well as activating signal transduction cascades in the neuron.
These processes result in activation of both serine/threonine kinases and tyrosine kinases, which, by phosphorylating membrane proteins, particularly the receptors for N-methyl-D-aspartate glutamate (NMDA) and alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)/kainate, increase membrane excitability by changing ion channel properties. This boost in excitability recruits existing subthreshold inputs to the dorsal horn neurons, thereby amplifying responses to noxious and non-noxious stimuli. The changes may be restricted to the activated synapse or spread to the adjacent synapse, and they are responsible for pain produced by low-threshold afferent inputs and the spread of pain hypersensitivity to regions beyond the tissue injury (secondary hyperalgesia).
Central sensitization is a major contributor to inflammatory and neuropathic pain, producing a largely NMDA-dependent, brush- or pinprick-evoked secondary hyperalgesia and a tactile allodynia. In inflammation, this activity-dependent central plasticity is driven by input from sensitized afferents innervating the inflamed tissue. After nerve injury, central sensitization can be driven by the ectopic activity in the injured fibers resulting from changes in the expression, distribution, or activity of ion channels. These central functional changes are contributed to by changes or switches in the phenotype of sensory neurons. Up to 30 molecules, mainly neuromodulators [such as galanin, vasoactive intestinal peptide (VIP), cholecystokinin (CCK), neuropeptide Y (NPY), brain-derived neurotrophic factor (BDNF), and nitric oxide synthase (NOS)] that alter synaptic drive and modify the response to basal stimulation, are regulated after nerve injury. In addition to the change in gene expression of the level of neuromodulators, novel expression also occurs, so that subpopulations of dorsal root ganglion (DRG) cells that do not normally express a neuromodulator, such as substance P or BDNF, begin to do so.
For example, substance P, which is normally expressed only in nociceptors, begins to be expressed in low-threshold sensory neurons after both inflammation and nerve injury. This means that although central sensitization is normally evoked only by nociceptor input, input from A fibers can also produce this phenomenon after nerve injury or inflammation. One example of this is the development of a progressive tactile pain hypersensitivity with the repeated touch of inflamed skin.
In addition to the activity-dependent increase in membrane excitability triggered by peripheral input, a decrease in phasic and tonic inhibition can also produce changes in dorsal horn excitability. This disinhibition may result from a down-regulation of inhibitory transmitters or their receptors, and from a disruption of descending inhibitory pathways. Furthermore, nerve injury, by virtue of injury discharge and ectopic activity, may lead to cell death in the superficial lamina of the dorsal horn, where inhibitory interneurons are concentrated.
5. Structural reorganization
After nerve injury, another anatomic change occurs: the structural reorganization of central connections. This involves the sprouting or growth of the central terminals of low-threshold mechanoreceptors from their normal termination site in the deep dorsal horn into lamina II (See Chapter 1, figure 2), the site of termination of nociceptor C-fiber terminals. The sprouted low threshold A fibers make synaptic contact in lamina II with neurons that normally receive nociceptor input, and this new pattern of synaptic input provides an anatomic substrate for tactile pain hypersensitivity.
A complex system of mechanistic changes occurs then, following the activation of the somatosensory pathways by both peripheral inflammatory and nerve lesions. An increase in the gain of the nociceptive system, in the periphery and in the central nervous system, is caused by activity-dependent plasticity, and it manifests as a widely distributed but transient pain hypersensitivity. With time, the changes evolve so that a number of different mechanisms that induce pain hypersensitivity are recruited. Three different forms (activation, modulation, and modification) of neural plasticity that produce pain hypersensitivity are summarized in Figure 1. Activation is directly linked to the noxious stimuli and it involves transduction and transmission of the signal. Modulation involves the peripheral and central sensitization processes. Modification of the system includes gene regulation, altered connectivity, and cell death. Persistent pain states may be associated with mechanisms that involve changes that are irreversible, such as cell death.
Figure 1. The three forms of neural plasticity that can produce pain hypersensitivity are summarized, highlighting the molecular and cellular changes implicated in pain mechanisms. (Modified from Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–1769, with permission.)
II. TOWARD A NEW CONCEPTUAL APPROACH FOR THE UNDERSTANDING OF PAIN
The current clinical evaluation of pain uses an etiologic or disease-based approach. This approach, however, should be modified to incorporate a mechanism-based diagnosis of pain. Identifying the causative disease is essential, particularly when disease modifying treatment is required, but in the vast majority of patients with persistent pain, the disease or pathology cannot be treated, and the injury is not reversible. In these cases, it is helpful to consider pain as the disease, and to attempt to identify mechanisms responsible for the pain rather than to categorize the patient primarily on the basis of underlying disease.
Given that mechanisms that produce pain in normal and pathologic conditions are being identified with increasing frequency in the laboratory, it is appropriate to begin to assess how such mechanisms fit into the overall schema of pain production. The notion of basal pain sensitivity, a term that represents the current status of an individual’s pain sensitivity, is fundamental. Basal pain sensitivity represents the pain experienced either spontaneously (i.e., in the absence of any identifiable stimulus) or evoked directly by, and within a short period of, a defined stimulus. In normal situations, there is no spontaneous or background pain, and pain is elicited only by intense or noxious stimuli. The amplitude of the pain, beyond a clear threshold level, is determined by the intensity of the stimulus, and the localization and timing of the sensation reflects the site and duration of the stimulus. This constitutes a state of pain normosensitivity. Normosensitivity is distinct from:
Pain hypersensitivity, in which pain may arise spontaneously, apparently in the absence of any peripheral stimulus
Hyperalgesia, in which the response to noxious stimuli is exaggerated
Hyperpathia, in which the pain may persist, radiate, or become excessively amplified
Allodynia, in which normally innocuous stimuli may produce pain
Normosensitivity is also distinct from those situations in which pain sensitivity is reduced, pain hyposensitivity, where suprathreshold noxious stimuli fail to elicit any pain response.
The aim of a mechanism-based approach is to first evaluate basal pain sensitivity by eliciting key aspects of the nature of the patient’s symptoms. Figure 2 shows how basal pain sensitivity can be qualitatively assessed by selectively eliciting the nature of symptoms. This can be accomplished using a relatively brief, semidirected interview (together with simple sensory testing to evoke symptoms) designed specifically to establish whether the patient’s basal pain sensitivity is normal, above normal, or below normal, and the extent to which the pain is spontaneous or evoked. The goal of the assessment is to characterize the clusters of symptoms, their onset and evolution, and to identify when possible the mechanisms responsible for the symptoms. Careful questioning, rather than the usual global assessments, will produce a new sort of clinical pain record based on the nature of the reported pain, to supplement the standard history (Chapter 4) and physical examination. Of course, this new approach needs to be validated, but its simplicity is likely to be its strength, increasing its usefulness beyond tertiary referral centers. The approach may be adopted in the future to aid treatment selection, especially when treatment efficacy is closely correlated with pain mechanisms (see Section IV).
Figure 2. Canvas for an interview-based qualitative assessment of pain. (Modified from Woolf CJ, Decosterd I. Implications for recent advances in the understanding of pain pathophysiology for the assessment of pain in patients. Pain 1999;6:S141–S147, with permission.)
III. IMPLICATIONS FOR THERAPEUTIC APPROACHES
The conventional assessment of pain syndromes includes the causative disease, the anatomic referral pattern of the pain, and a quantitative evaluation [such as the visual analog scale (VAS)]. This approach groups patients into categories based on their disease syndromes, such as neuropathic pain, headache, osteoarthritic pain, or cancer pain. Contemporary preclinical basic science has successfully elucidated the molecular mechanisms of action of current analgesics (opiates, nonsteroidal anti-inflammatory drugs, and sodium channel blockers) and their effects on pain mechanisms. Yet there is an extremely poor correlation between the efficacy of analgesics and pain syndromes. The increasingly popular measure of the number needed to treat (NNT) is an efficacy index representing the number of patients who need to be treated with a certain drug to obtain one patient with a defined degree of pain relief. The NNT, which has been studied in different pain categories, is a good example of the lack of specificity and predictive value of the current pain classifications. The NNT measure of efficacy does not reveal any consistent differences across different pain conditions for distinct drug classes observed. The goal of a mechanism-based assessment of pain is to provide a classification in which the categorization of patients into mechanism-based subpopulations will aid the rational treatment of pain. Dividing pain into components that reflect some of the major pain mechanisms may help identify how and why certain treatments work, thus revealing useful correlations between pain mechanisms and treatments.
IV. IMPLICATIONS FOR EVALUATION OF EFFICACY OF NEW THERAPIES
A major problem in clinical studies of pain is that the high intraand interpatient variability in pain scoring using global outcome measures makes it very difficult to evaluate the efficacy of novel analgesics. The usual explanations for this variability are the complexity of pain mechanisms, changes in the primary disease, and psychological factors. Another approach is therefore called for, one that provides new clinical outcome measures that enable an evaluation of whether new analgesics have an action on particular pain mechanisms.
If a new therapy is given to patients selected only on the basis of a particular disease (e.g., diabetic neuropathy), and the clinical outcome measure is a simple global pain measure (e.g., a VAS score of pain at rest), it is simply not possible to assess whether the treatment acts on a particular mechanism (e.g., central sensitization) and reduces a particular symptom (e.g., tactile or cold allodynia). Because the degree of central sensitization may differ considerably in this cohort of patients, any treatment that acts only on central sensitization will produce highly varied responses across the population. Once drugs are available that act specifically on novel pharmacologic targets such as the receptors and ion channels of DRG-specific VR1, P2X3, DRASIC, and SNS/SNS2 (SNS2 is sensory neuron-specific voltage-gated sodium channel NaN/Nav1.9), patients will need to be selected on the basis of a reasonable assessment that their pain involves one of these targets. For example, since VR1 is involved in encoding heat pain, a VR1 antagonist would not be expected to have any effect on a patient with tactile allodynia. Selection of patients on the basis of categories, instead of on the basis of mechanisms, is likely to result in a cohort of patients whose pain mechanisms are quite different. Only a limited number of these patients can be expected to respond to a mechanism-specific drug treatment. Patients who do not respond to the treatment produce a false-negative result by diluting the benefit in a subgroup with the targeted mechanism.
In the last decade, neurobiology research has enormously increased our knowledge of the fundamental mechanisms responsible for producing chronic pain. On the other hand, changes in clinical pain management have been slow. The challenge now is to bridge the large gap between basic research and clinical practice by utilizing new inputs from basic science in the classification, assessment, diagnosis, and treatment of pain.
McCleskey EW, Gold MS. Ion channels of nociception. Annu Rev Physiol 1999;61:835–856.
Mogil JS, Yu L, Basbaum AI. Pain genes? Natural variation and transgenic mutants. Annu Rev Neurosci 2000;23:777–811.
Sindrup SH, Jensen TS. Efficacy of pharmacological treatments of neuropathic pain: an update and effect related to mechanism of drug action. Pain 1999;83:389–400.
Woolf CJ, Bennett GJ, Doherty M, et al. Towards a mechanism-based classification of pain? Pain 1998;77:227–229.
Woolf CJ, Decosterd I. Implications for recent advances in the understanding of pain pathophysiology for the assessment of pain in patients. Pain 1999;6:S141–S147.
Woolf CJ, Mannion RJ. Neuropathic pain: Aetiology, symptoms, mechanisms, and management. Lancet 1999;353:1959–1964.
Woolf CJ, Max MB. Mechanism-based pain diagnosis: Issues for analgesic drug development. Anesthesiology 2001 (in press).
Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–1769.
Yaksh TL. Spinal systems and pain processing: Development of novel analgesic drugs with mechanistically defined models. Trends Pharmacol Sci 1999;20:329–337.