Free Web Hosting Provider - Web Hosting - E-commerce - High Speed Internet - Free Web Page

The International Association of Pain committee on taxonomy defines pain as a sensory or emotional process associated with tissue damage Acute pain, such as that which occurs in trauma, alerts the patient to the presence of harmful or potentially harmful stimuli. Chronic pain is not merely a temporal extension of acute pain. Persistent tissue injury and activation of nociceptors, especially in the presence of inflammation, cause pharmacologic, physiologic, and anatomic changes in central nervous system processing of nociceptive information. Pain can also occur in the absence of ongoing nociceptive activity.

Acute pain may be recurrent in disease processes such as cancer, sickle cell anemia, AIDS, multiple sclerosis, and other disorders in which recurrent or progressive tissue injury, or both, is experienced. Chronic pain may no longer warn the patient of bodily injury and thus serves no useful function. Acute pain of severe intensity may be associated with anxiety and fear and is usually accompanied by physiologic activation of the sympathetic nervous system. The so-called chronic-pain state is used by some clinicians to describe patients who report pain on a long-term basis, even if there is no apparent tissue-injury component. Chronic pain is characterized by adaptation of sympathetic activity and the development of chronic vegetative symptoms, including a decrease in appetite, malaise, sleep disturbances, and irritability. The chronic-pain state is often associated with depression, anger, and other affective states.

Pain is the most common reason that patients visit a physician. New pain complaints account for 40 million visits annually. Nearly 64 million people suffer from trauma-related pain each year. Chronic pain is responsible for $4 billion in lost workdays. As much as $65 million a year is lost as a result of diminished work productivity.

Pain complicating acute and chronic disease or medical procedures is also common . Approximately 50% of patients with cancer report pain of at least moderate severity, and 46% of patients with cancer report pain that cannot be managed with analgesics. Nearly all patients with advanced cancer report severe pain. AIDS is associated with pain in 83% of patients.

Acute and chronic pain can be viewed as a public health problem. Fortunately, much has been learned about fundamental pain mechanisms. This chapter reviews advances in the neurobiology of nociception and current clinical practice guidelines for the asse ssment and treatment of acute and chronic pain, including so-called neuropathic pain, which is pain complicating neurologic injury or illness.

Traditionally, the pain associated with such diseases as cancer or multiple sclerosis has served as a model for acute and chronic pain. Several physiologic types of pain have thus been defined: somatic, visceral, neuropathic, and complex regional pain syndrome, formerly known as sympathetically maintained pain .

Somatic or nociceptive pain occurs as a result of activation of nociceptors in cutaneous and deep musculoskeletal tissues. Usually, the patient can accurately localize this pain to the site of pathology; it may be felt in superficial cutaneous or deeper musculoskeletal structures.

Visceral pain results from ischemia, infiltration, compression, distention, and torsion or stretching of thoracic, abdominal, and pelvic viscera. It is typically poorly localized and often associated with nausea, vomiting, and diaphoresis, particularly when it is acute. Visceral pain is often referred to cutaneous sites that are remote from the site of the lesion (e.g., jaw pain associated with the deep substernal pressure during myocardial infarction or shoulder pain with diaphragmatic irritation). Tenderness at the referred cutaneous site may occur. Visceral pain is mediated through both the sympathetic and the parasympathetic nervous systems. So-called silent nociceptors have been identified in visceral tissues. Silent nociceptors are not active until sensitized by inflammatory and other chemical stimuli; once sensitized, these receptors respond with many of the same neurophysiologic characteristics that nociceptors in somatic nerves do. The sensitization of silent visceral nociceptors has been implicated in the development of chronic abdominal pain associated with syndromes such as inflammatory bowel disease.

Neuropathic pain results from injury to the peripheral or cen tral nervous system and may occur without ongoing tissue damage. Pain resulting from lesions of the peripheral nerves (especially lesions that are traumatic in origin and that partially or completely interrupt afferent sensory transmission between the peripheral and central nervous systems) has sometimes been termed deafferentation pain. Pain resulting from injury to the spinal cord or brain, especially when complicating cerebrovascular, demyelinating, or traumatic CNS injury is involved, is usually termed central pain. Pain resulting from neural injury is often se vere, and the quality is different from somatic or visceral pain. Neuropathic pain is typically described as a constant dull ache, often with a pressure or viselike quality; superimposed paroxysms of burning or electric shock-like sensations are common. These paroxysms of pain may be associated with spontaneous and ectopic activity in the peripheral and central nervous systems.

A complete understanding of the pathophysiology of pain in specific patients is seldom possible; frequently, several different physiologic mechanisms give rise to pain in the same patient. Nonetheless, clinical inferences regarding pain mechanisms in specific circumstances often have direct diagnostic and therapeutic implications. For example, the presence of paroxysmal or lancinating pain usually leads a clinician to suspect a neuropathic etiology, even in the absence of demonstrable sensory or motor abnormalities or reflex changes. Tricyclic antidepressant or anticonvulsant med ications may successfully manage lancinating pain when traditional analgesics, such as opioids or nonsteroidal anti-inflammatory drugs (NSAIDs), have failed.

A new term, complex regional pain syndrome (CRPS), is now preferred to the term reflex sympathetic dystrophy or sympathetically maintained pain, to avoid any inferences regarding the presence or absence of sympathetic involvement in the associated pain state. PS is classified as two types: type I is the traditionally recognized type and does not involve primary injury to nerves; type II is associated with the historically termed causalgia syndromes to distinguish patients with nerve injury. Either type of CRPS may present with features of regional sympathetic hyperactivity and respond to sympathetic interruption and thus may be sympathetically mediated. However, either type of syndrome may respond to more conventional analgesic therapies in the absence of sympathetic nervous system interventions. Associated signs and symptoms of nerve injury or dystrophy (e.g., muscle atrophy, osteoporosis, and neurophysiologic and electrophysiologic signs of nerve injury) are not predictive in defining response to therapy in CRPS.

Pain is a subjective experience that has both sensory and affective components. Experiencing pain involves a series of complex neurophysiologic processes that have four distinct components: transduction, transmission, modulation, and perception. Transduction is the process by which a noxious stimulus is converted to an electrical impulse in appropriate sensory nerve endings. Transmission is the conduction of impulses that arise from the painful stimulus along nerves to the CNS with major connections in the thalamus and projections to the cingulate, insular, and somatosensory cortex. Modulation alters pain transmission. Ther e is evidence of both inhibitory and excitatory mechanisms that modulate nociceptive impulse transmission in both the peripheral and central nervous systems. Pain perception probably lies at the level of the thalamus, with the cortex being essential for the discrimination of specific sensory experiences. Not all of these steps need to occur for a person to experience pain. For example, trigeminal neuralgia is thought to be caused by midaxonal discharges initiated at the site of a compressed or demyelinated nerve; this process occurs in the absence of transduction of a chemical stimulus at a nociceptor. As another example, modulation of a pain impulse may not occur if specific nervous system tracts are injured.

Primary afferent nociceptors (PANs) are unspecialized free nerve endings found in nearly all tissues that convert noxious mechanical, chemical, and thermal stimuli to an electrical impulse within the nerve . Nociceptors normally are not spontaneously active but may show sensitization and become active, particularly after thermal injury to the skin. Nociceptors can be activated or sensitized by endogenous chemicals that arise from damaged tissue and cells of the immune system; for example, bradykinin, serotonin, histamine, and potassium ions can activate nociceptors, and interleukin-1, prostaglandin E ₂, leukotriene B₄, and nerve growth factor can sensitize nociceptors. Pain impulses are transmitted via unmyelinated (C fibers) and thinly myelinated (A-delta) fibers. Approximately 10% of all myelinated cutaneous peripheral nerve fibers carry nociceptive information, and more than 90% of all unmyelinated fibers are nociceptive. tivation of a single A-delta fiber is sufficient to cause pain, and direct electrical stimulation of unmyelinated (C) nociceptors is associated with a dull, burning, or aching pain. Positive symptoms and signs of peripheral nerve disease (e.g., Tinel's sign or positive straight leg raising in S1 root compression) are associated with spontaneous activity in peripheral nerve. Paresthesias are represented by ectopic paroxysmal activity and high-frequency (> 220 Hz) discharges in peripheral nerve.

Repetitive stimulation of a nociceptor can cause tissue damage, lowering the activation threshold and causing primary hyperalgesia (enhanced pain). Pain caused by stimuli that would not normally produce pain, such as pressure or simple movements, is probably the result of sensitized nociceptors or a sensitized nervous system.

Secondary hyperalgesia refers to the expansion of the zone of cutaneous hyperalgesia outside of the area of injury . It is caused by CNS changes resulting from the initial injury. Secondary hyperalgesia probably underlies so-called spreading pain that occurs in the absence of ongoing or recurrent injury or disease.

One of the most common causes of pain is inflammation. Among the immune mediators, leukotriene B₄, acting through newly described intracellular pathways, may play a critical role in inflammation, including determining the extent and duration of an inflammatory response. It is also likely that nerves contribute to the ongoing inflammatory process (i.e., neurogenic inflammation). Sensory neurons may release substance P and calcitonin gene-related peptide, and sympathetic afferent neurons can release prostaglandin E₂. These processes may perpetuate inflammation, which could explain chronic inflammatory conditions of the skin, joints, gastrointestinal tract, and brain.

A cascade of events is largely responsible for pain associated with such inflammatory conditions as arthritis, cancer, infection, and postsurgical pain. The steps of the cascade that involve the conversion of arachidonic acid via cyclooxygenases to prost aglandins are therapeutically important because they are inhibited by various NSAIDs. Spinal administration of NSAIDs blocks the hyperalgesia produced by formalin injection into the hind paw of animals, indicating that prostaglandins are important to the central as well as the peripheral anti-inflammatory actions of NSAIDs. Steroids may block the inflammatory process at other steps in the inflammatory cascade.

Opioids can have a peripheral site of action in the presence of inflammation. Studies have shown that opioids administered topically to peripheral tissues at sites of inflammation reduce excitability of the afferent fiber and are antinociceptive. Inflammation may disrupt the perineurium and expose preexisting opioid receptors to the external environment. Opioid receptors are also present on cells that migrate to inflamed tissue. Consistent with observations made in animals, intra-articular injection of morphine is effective for the treatment of postoperative pain in humans. Therefore, local application of opioids that are polar and that do not cross the blood-brain barrier may provide potent analgesia at peripheral sites of inflammation without adverse CNS effects, such as sedation, nausea, respiratory depression, and mental clouding.

The primary afferent fibers, with their cell bodies lying in the dorsal root ganglion, connect with a second neuronal cell, located within the dorsal horn of the spinal cord. Afferent fibers from nociceptors enter the spinal cord laterally in the dorsal root and ascend or descend several segments in Lissauer's tract before synapsing in the dorsal horn.

The dorsal horn consists of six laminae. Laminae I and II are the sites of termination of primary afferent C fibers; together, they make up the substantia gelatinosa, which is important for integration and modulation of incoming nociceptive information. Lamina V is the site of second-order wide dynamic range (WDR) and nociceptive-specific (NS) neurons, which receive input from both nociceptive and nonnociceptive primary neurons. The NS neurons respond only to intense noxious stimulation in the peripheral receptive field; the WDR neurons respond to innocuous and noxious stimuli of many types. Both types of neurons are thought to be important in the encoding of nociceptive information: the NS response signals the presence or absence of a tissue-damaging stimulus, whereas the WDR response provides information about stimulus quality and per haps location. The variety of chronic-pain states can be explained in terms of the inputs to these cells and their supraspinal connections.

The first site of modulation of nociceptive information in the CNS occurs in the dorsal horn of the spinal cord. Central sensitization is dependent on N-methyl-d-aspartate (NMDA) receptor activation by excitatory amino acids such as glutamate. Glutamate and NMDA receptors are found in high concentration in the dorsal horn. Blockade of NMDA receptors by such drugs as ketamine, dextrorphan, phencyclidine, and MK-801 reduces the progressive increase in the number of depolarizations evoked by a single stimulus in response to repeated stimulation of peripheral C fibers and can also prevent or terminate central sensitization once it has begun.

Nitric oxide (NO) is an excitatory neurotransmitter. Because it is a gas, it can diffuse across the synaptic cleft to induce retrograde effects, such as release of glutamate and other neurotransmitters. It also appears to interact with adjacent neurons and glia. Thus, interference with the synthesis of NO has important, widespread ramifications, including the reversal of morphine tolerance. The synthesis of NO involves activation of the enzyme NO synthetase, which is an NMDA-dependent process . This enzyme is activated by Ca²⁺, which gains access to the cell via activation of the NMDA receptor, a calcium channel ionophore.

Axons from WDR and NS neurons in laminae I and V decussate in the central gray matter of the spinal cord and become the ascending projections of the neospinothalamic and paleospinothalamic tracts. The neospinothalamic tract projects monosynaptically to the ventroposterolateral (VPL) nucleus of the thalamus and probably encodes for stimulus localization and intensity. The paleospinothalamic tract sends multiple projections to the brain stem reticular formation and medial regions of the thalamus before termination in VPL nucleus. Several other ascending spinal tracts have been identified, including the spinohypothalamic tract, spinoreticular tract, spinopontoamygdala tract, and the dorsal column tract. The spinoreticular and spinocervicothalamic tracts ascend in the ipsilateral dorsolateral quadrant (separate from the spinothalamic tracts in the anterolateral quadrant) but join with the spinothalamic tract in the medial lemniscus in the brain stem. Dorsal column fibers transmit both proprioceptive and nociceptive information and ascend in the ipsilateral medial lemniscus to the thalamus; stimulation of these fibers is a target for pain relief. The dorsal column tracts may be responsible for the recurrence of painful symptoms after relatively selective ablative procedures on the spinothalamic tract (e.g., cordotomy).

Pain from head and neck structures is processed through cranial nerves V, VII, and IX and the first three cervical spinal nerves. The spinal trigeminal nucleus and adjacent reticular formation are the equivalent of the spinal dorsal horn. This nucleus has laminar structure and afferent synaptic connections similar to those of the spinal dorsal horn. Second-order neurons send axonal projections to the other side of the brain stem and ascend in the contralateral medial lemniscus, eventually terminating in the ventroposteromedial (VPM) thalamic nucleus. The ventrobasilar complex (VPL, VPM) axons project to the parietal somatosensory cortex, and the medial thalamic nuclear groups project to the striatum and cerebral cortex.

Head and Holms postulated that the thalamus and cortex were essential for pain perception and discrimination. Large lesions of the cerebral cortex that destroy the somatosensory area may produce minimal or no pain, whereas small and highly localized cortical lesions outside of this area can be associated with spontaneous pain and increased pain perception. Penfield's classic cortical mapping studies demonstrated that cortical stimulation produces pain, though infrequently. Patients with lesions of the insular cortex may manifest a syndrome called pain asymbolia. These patients do not generate emotional pain responses despite normal thresholds to cutaneous pain stimuli contralateral to the insular lesion. Further support for the specific role of the cortex in the perception and processing of painful stimuli comes from animal studies that demonstrate NS neuronal responses identical to those found in the spinal cord dorsal horn.

A lesion anywhere in the spinothalamic tract and thalamocortical projections may result in central pain. Hypoalgesia, the reduced appreciation of painful stimuli, such as pinprick or thermal sensation (especially cold), is an almost invariable accompaniment of central pain disorders. Postulated mechanisms for hypoalgesia include a loss of inhibitory influences on excitatory pathways as well as the emergence of spontaneous thalamic activity (primarily in VPL and VPM) in patients with central pain syndromes after stroke.

Brain imaging with positron emission tomography (PET) and functional magnetic resonance imaging enables visualization of cortical and thalamic structures associated with processing of experimental and natural pain stimuli. The cortical and subcortical areas that are activated depend on the type of pain stimulus. In general, brain-imaging studies demonstrate consistent activation of the thalamus, anterior cingular gyrus, and insular cortex. Primary and secondary somatosensory cortices are less consistently activated. The PET studies, however, indicate that the areas of cortical activation seen in experimentally induced pain do not necessarily match the findings in patients with chronic neuropathic pain.

PET images of normal persons subject to experimentally induced pain have shown that regional cerebral blood flow in the contralateral posterior thalamus is consistently decreased. By contrast, increased regional cerebral blood flow is observed in the bil ateral anterior insular and posterior parietal cortices, bilateral inferior lateral prefrontal cortices, and cerebellar vermis.

The anterior cingulate cortex has long been recognized as a region of pain perception; lesions within this region result in diminished pain. This area of cortex receives major input from the medial thalamus. Recent evidence from PET studies has confirmed the crucial role of the anterior cingulate cortex in subserving the affective components of the pain experience. The anterior cingulate cortex is selectively activated by painful thermal stimuli. Metabolic activity is attenuated in the anterior cingulate cortex when patients are subject to psychological techniques that produce analgesia, such as hypnosis and active distraction. Interest in the cingulate cortex as a neuroablative target for intractable pain is based in part on observations suggesting that the cingulate cortex plays a key role in the processing of nociceptive information.

Inhibitory and excitatory neuroanatomic pathways originate in the brain stem ventoreticularis median nucleus and descend in the dorsal longitudinal fasciculus to modulate nociception and pain perception. Several important neurotransmitters have been identified in these descending pathways, including norepinephrine, serotonin, and endogenous opioid compounds (e.g., methionine-enkephalin, leucine-enkephalin, and b-endorphin). The inhibitory pathways involve the off cells, which are activated by opioids, and the excitatory pathways involve the on cells, which are inhibited by morphine. The relative activity of on and off cells determines the degree of inhibition occurring in this endogenous pain suppression pathway. This mechanism is thought to provide the physiologic basis for the action of morphine (and other opioids) as well as other analgesic drugs, such as tricyclic antidepressants and alpha-adrenergic agonists.

Tricyclic antidepressant compounds, such as amitriptyline and desipramine, block the presynaptic reuptake of serotonin and norepinephrine, thus augmenting their postsynaptic actions in descending pain suppression pathways. Tricyclic antidepressant drugs have been shown to have analgesic activity independent of their antidepressant effects, especially in such neuropathic pain states as diabetic neuropathy and postherpetic neuralgia.

Drugs that are alpha₂ agonists at adrenergic receptors in the spinal cord (e. g., clonidine) produce analgesia on administration into the epidural space. Phentolamine, an alpha₂-adrenergic receptor antagonist, has also been used to manage pain in regional pain syndromes, particularly when sympathetically maintained pain is suspected. The effect of phentolamine may predict responsiveness to anesthetic sympathetic blockade.

Activation of this descending pathway by the action of endogenous opioids may account for the phenomenon of placebo and acupuncture analgesia. The inconsistent reversibility of placebo and acupuncture analgesia by opioid antagonists (e.g., naloxone) casts doubt on this theory, however.

In a prospective study of cancer pain management, 63% of 1,170 physicians surveyed indicated that inadequate pain assessment skills was the primary reason for suboptimal outcomes and generally incomplete pain management in their practice settings.

Successful management of pain starts with a thorough and accurate assessment of the complaint . The history should consider the location, quality, duration, intensity, temporal pattern, and aggravating or relieving factors associated with the pain. The clinician should obtain information regarding the extent of pathology or tissue damage associated with the specific disease or disorder. In many diseases (e.g., cancer, AIDS, and sickle cell anemia), there is a rough correlation between increasing pain and progression of disease. The patient's psychological state and coping style and degree of family or social support should be assessed, as these factors may influence the meaning and perception of pain and the ability to comply with therapeutic recommendations. For patients with potentially life-threatening diseases, a discussion about prognosis between the primary treating physician and the patient and family is usually invaluable in evaluating the meaning of pain to the patient.

A detailed physical examination, including a focused neurologic examination, is always important, particularly when neuropathic pain is suspected. The mental status of the patient should be evaluated to determine the reliability of his or her reports of pain.

Diagnostic testing is usually necessary to determine the etiology of the pain complaint. It is important to review radiologic studies to ensure that the appropriate anatomy has been imaged and to assess the relevance of any abnormal findings. For example, a study of low back pain showed that many abnormal MRI findings thought to be related to pain were not associated with a patient's pain complaints. Abnormalities in disease-specific laboratory parameters (e.g., tumor markers in cancer patients, hemoglobin levels and presence of hemolysis in sickle cell patients in vaso-occlusive crises, and CD4 counts in AIDS patients) are sensitive indicators of d isease progression and may be correlated with pain before other objective physical signs or radiographic findings develop. It is helpful to make clinical inferences about the pathophysiologic mechanisms of pain, as these may guide diagnostic evaluations and even selection of analgesic and other therapies for pain. Critical to a successful treatment outcome is the clinician's commitment to reassess the patient at specific intervals, especially when pain is unresolved.

Accurate assessment of medical and psychological factors relevant to the individual patient with pain should lead to appropriate treatment. There are several basic strategies for pain management . Although pharmacotherapy constitutes the mainstay of treatment for most pain, experienced pain clinicians successfully individualize and integrate the relevant therapeutic modalities for each patient. Analgesic drugs should be prescribed only after an appropriate assessment of pain to achieve the best balance between pain relief and maximizing of function, in accordance with goals set by the patient.

The World Health Organization (WHO) developed a three-step analgesic ladder for the use of analgesics to treat cancer pain. This algorithm has been extended to other conditions. For example, experienced clinicians advocate the use of opioids for chronic pain of nonmalignant origin (e.g., arthritis, sickle cell anemia, low back pain associated with osteoarthritis, and osteoporosis) in carefully selected patients. Generally, long-term opioid therapy for nonmalignant pain is considered when continuation of analgesic therapy provides consistent pain relief and improves patient function.

The WHO analgesic ladder matches the patient's report of pain intensity to specific types of analgesics. Patients with mild pain can be managed with a nonopioid analgesic, such as acetaminophen, aspirin, or an NSAID . For moderate pain that cannot be effectively managed by nonopioid medications, step II of the WHO ladder recommends the prescription of so-called weak opioids (e.g., codeine, hydrocodone, or oxycodone) . Theoretically, opioids have no ceiling regarding their analgesic efficacy. These drugs are classified as weak, however, because side effects, such as nausea and sedation, often limit dosage escalation, and they are often formulated in fixed combinations with acetaminophen or aspirin that further limit dosing. For severe pain that is not relieved by weak opioids, a so-called strong opioid (e.g., morphine, hydromorphone, methadone, or fentanyl) can be used, either alone or in combination with a nonopioid (step III) . At any step of the ladder, certain adjuvant analgesics may be used to treat specific pain syndromes or to counteract side effects of opioid analgesics . Patients presenting with severe pain should be started at the third step of the ladder and not walked up the ladder from the first step. This approach was validated in studies of the management of cancer pain. One study involved 2,118 patients with more than 140,000 treatment days. A positive outcome resulted in 88% of patients.

The nonopioid class of analgesics includes acetaminophen, aspirin, and NSAIDs. Unlike the opioid analgesics, these drugs do not produce physical dependence or tolerance, but they have an intrinsic ceiling on their analgesic efficacy. Both the minimum effective dose and the ceiling dose may vary among persons. Dosage titration within a narrow range may be necessary in some patients.

The NSAIDs inhibit the enzyme cyclooxygenase (COX), thereby inhibiting prostaglandin synthesis. Prostaglandins sensitize and activate nociceptors and also have an effect on prostaglandin biosynthesis at CNS sites. Because NSAIDs and opioids have different mechanisms of action, coadministration of both drug types to provide additive analgesia has been recommended in WHO and other guidelines for pain management. This additive effect has not been confirmed in more recent meta-analysis of repeated dose studies, however.

Adverse effects are common with use of most nonopioid analgesics. All may be associated with GI toxicity; the most serious complications are ulceration and bleeding. The risks of GI hemorrhage are not the same for all drugs. This may be because some NSAIDs selectively inhibit the COX-2 isoform, which is induced in the presence of inflammation and is not critical to maintenance of renal blood flow and maintenance of the GI mucosa. Relatively selective COX-2 inhibitors (e.g., eto dolac and nabumetone) have fewer GI mucosa and renal side effects.

The regular, long-term use of either acetaminophen or NSAIDs confers a significantly increased risk of end-stage renal disease. Acetaminophen is equipotent to aspirin in terms of analgesic efficacy, is generally very well tolerated, and is not associated with the risks of GI hemorrhage that are associated with NSAIDs. Acet aminophen is a weak inhibitor of COX, which presumably explains its poor anti-inflammatory potency.

Opioid analgesics are the mainstay of therapy for severe acute and chronic pain. These are classified as pure agonists or agonist-antagonists on the basis of their interactions with opioid receptors and the effects that are produced . The agonist-antagonist drugs include butorphanol, nalbuphine, pentazocine, and dezocine. This class of opioids are partial agonists at kappa opioid receptors and antagonists at mu opioid receptors. Because they are partial agonists, they have a ceiling on their analgesic efficacy. They commonly produce psychoto mimetic effects by interacting with kappa opioid receptors. Finally, they may precipitate withdrawal when administered to patients physically dependent on mu agonists such as morphine. For these reasons, agonist-antagonist opioids are not recommend ed for the treatment of patients with chronic pain. They are useful for one-time or short-term treatment of acute pain in opioid-naive patients.

Certain principles should be followed when using opioid analgesics in the treatment of cancer pain. When pain is continuous, opioids should be administered on an around-the-clock basis. As-needed administration often results in the patient's experiencing multiple episodes of pain during the day; such administration is not recommended except for the first 24 to 48 hours after initiation of opioid therapy to determine the daily dosage requirement. Once the optimum daily dose is determined, the opioid should be prescribed by the clock, preferably in a long-acting (i.e., oral sustained-release or transdermal) formulation. Many patients experience so-called breakthrough pain or transitory increases in pain above a baseline level. For breakthrough pain and as needed, a rescue dose of a short-acting opioid should be available, equal to approximately 10% to 20% of the total scheduled daily dose.

There is substantial variation from person to person in the dosage that is necessary for adequate analgesia, even in patients with similar pain syndromes. The variability in opioid responsiveness has pharmacokinetic and pharmacodynamic components. Genetic factors have been shown to be important in determining opioid responsiveness. Approximately 15% of whites lack the oxidative phosphorylative enzyme CPDY211, which is needed to metabolize codeine to morphine, and therefore, they require higher does of codeine. In a retrospective analysis, Kaiko and coworkers noted that blacks and Orientals participating in single-dose clinical trials to determine opioid efficacy reported twice the analgesic effect as whites. Sex is also a factor in opioid responsiveness. Women who received the kappa agonist drug butorphanol reported greater analgesia than men.

Morphine is the most widely used agent for moderate to severe pain because of its availability, well-defined pharmacology, and relatively low cost. Selection of an opioid, however, should be based on such factors as convenience of administration, cost, compliance, and side-effect profile for the individual patient. The long-term use of meperidine is not recommended, because accumulation of the toxic metabolite normeperidine may occur, and agitation or seizures may result.

When a patient develops intolerable side effects from one opioid, it may be useful to switch to another opioid analgesic to obtain a more favorable balance between analgesia and side effects. Knowledge of the equianalgesic doses of the different opioids facilitates switching from one drug to another . Because cross-tolerance among the opioids is not complete, one half of the calculated equianalgesic dose is recommended as the starting dose for the new drug.

Certain adverse effects are common to all opioid analgesics. Constipation is nearly universal, and every patient should be given a prophylactic bowel regimen. A typical program includes daily use of a senna-based laxative, supplemented as needed by lactulose or milk of magnesia. Sedation is common at the start of treatment or after a dosage increase but often resolves within a few days. Persistent sedation may be treated by increasing the dosage of caffeine in the diet or by use of a prescription psycho stimulant, such as methylphenidate or dextroamphetamine, in low doses . Nausea and vomiting occur in as many as 30% of patients taking opioids but may resolve within a few days. When they are unresolved, the use of an antiemetic drug (e.g., phenothiazine or metoclopramide) may be required.

Confusion and hallucinations are much less common (< 15% of patients) but usually necessitate dosage reduction or switching to another opioid analgesic. Confusion and delirium have many other potential causes in patients with medical illness. Therefore, when a change in mental status occurs in a patient on a stable opioid regimen, reevaluation of the disease status should be considered before the confusion is attributed to the opioid. Other relatively infrequently reported adverse effects include dry mouth, urinary hesitancy, and pruritus.

Respiratory depression is an uncommon complication in patients taking opioids on a long-term basis. Pain acts as a physiologic antagonist of the CNS-depressant effects of opioids. Tolerance also develops to the respiratory-depressant effects of opioids. Patients at risk for respiratory depression are those with recent large increases in their opioid dosage (typically > 50%); the risk is greater when sedatives are coadministered or when a pain-relieving anesthetic or neurosurgical procedure removes the stimulant effect of pain. The opioid naloxone is often used inappropriately to reverse respiratory depression and sedation in cancer patients taking opioids on a long-term basis. Simple measures, such as vigorous physical stimulation to awaken the patient, should be attempted and other causes of respiratory depression excluded before return of pain is produced by rapid I.V. injection of naloxone and before an iatrogenic opioid a bstinence syndrome is initiated. When naloxone is needed for patients who have been taking opioids on a long-term basis, a 0.4 mg ampule should be diluted in 10 ml of saline and administered by slow I.V. push.

Opioids may be administered by continuous subcutaneous (S.C.) or I.V. infusion in patients unable to take oral opioids, such as patients with intractable nausea and vomiting or who have nonfunctional GI tracts. The S.C. route has been recommended because the pharmacokinetics of morphine and hydromorphone via the S.C. route are similar to the pharmacokinetics via the I.V. route, and S.C. administration provides equally effective analgesia more conveniently and at a lower cost.

Tolerance and physical dependence, two predictable pharmacologic consequences of long-term opioid administration, should not be confused with psychological dependence or addiction. Tolerance refers to the need to increase the dosage of the drug to mainta in the same analgesic effect. In the clinical setting, tolerance to the analgesic effects of opioids is rarely a major problem in patients with stable pain syndromes. Usually, the need to increase opioid dosage to maintain analgesia implies worsening pain as a result of progressive disease. An increased-dosage requirement should not simply be attributed to opioid tolerance, which is signaled by a progressive shortening of the duration of analgesia, implying a change in responsiveness of the receptor to the opioid. Physical dependence refers to the development of a withdrawal syndrome with abrupt discontinuance of therapy or substantial dosage reduction. This phenomenon may occur when repeated doses of opioids are administered for more than a few days.

Tolerance and physical dependence do not imply psychological dependence or drug addiction, although these often coexist. Addiction is a psychological and behavioral syndrome characterized by compulsive drug use, loss of control, and continued drug use despite harm to the patient. the treatment of cancer pain with long-term opioid therapy, addiction develops extremely rarely, although physical dependence is almost universal. Exaggerated fears of addiction on the part of the patient, physician, and public at large, as well as fears on the part of the physician about scrutiny of his or her medical practices by drug and medical board regulators, are barriers to adequate treatment of pain.

The adjuvant analgesics are a diverse group of drugs indicated for use in certain circumstances and as a means to counteract the adverse effects of opioids in other circumstances . These drugs may be used alone or in combination with opioids and nonopioid analgesics at any step of the WHO analgesic ladder.

With the exception of the tricyclic antidepressants and anticonvulsants, the use of adjuvant drugs in cancer pain is empirical. Few randomized, controlled clinical trials exist to document the indications for their use in patients with pain. Amitriptyline, imipramine, nortriptyline, and desipramine have all been demonstrated to have some analgesic efficacy in chronic pain, especially pain of neuropathic origin. The new selective serotonin reuptake inhibitors, although attended by fewer side effects than the tricyclic antidepressants, have varied efficacy as analgesics. A randomized, controlled trial in which fluoxetine was used in diabetic neuropathy could not demonstrate an analgesic effect in nondepressed patients. Paroxetine, however, was reported effective in the management of painful diabetic neuropathy.

The anticonvulsant drugs may be effective in the management of neuropathic pain, particularly for those syndromes associated with paroxysmal or lancinating pain. Clonazepam may be useful in managing myoclonus that occurs as a toxic effect of high-dose opioid therapy.

I.V. lidocaine often provides rapid relief of neuropathic pain. Although the effect is short-lived, it may persist longer than the duration of the infusion. In a randomized, controlled trial of cancer-related neuropathic pain, however, I. V. lidocaine was no better than placebo. Mexiletine, an oral analogue of lidocaine, has been used to manage chronic neuropathic pain.

Peripheral neuropathic pain syndromes sometimes respond to topical agents. Capsaicin cream (0.075% concentration) has been shown to reduce diabetic neuropathy, postherpetic neuralgia, and postmastectomy pain.

Corticosteroids may rapidly relieve metastatic bone pain, nerve compression pain, headache associated with brain tumors, and epidural spinal cord compression. Because there is potential for serious adverse effects with prolonged use, corticosteroids are best reserved for short-term use or for patients with advanced disease.

Psychostimulants, such as caffeine, methylphenidate, pemoline, and dextroamphetamine, are also useful in the treatment of cancer pain. Methylphenidate has been shown to enhance the analgesic effect of opioid drugs and to decrease the sedation associated with opioid use. These stimulant medications are most useful in patients who are receiving adequate analgesia from opioids but whose quality of life is compromised by excessive sedation.

Opioid analgesics can be introduced into the epidural or intrathecal space for the management of select patients with cancer pain. Small doses of opioids administered by these routes are delivered in close proximity to their receptors in the dorsal horn of the spinal cord; high local concentrations of opioids are thus achieved. Because the amount of drug administered is reduced, good analgesia with fewer side effects may result. Spinally administered opioids should be considered for patients whose pain is at least partially opioid responsive but who cannot tolerate the side effects of oral or parenteral opioids.

Adverse effects of spinal opioids are thought to result in part from supraspinal redistribution of drug. These include pruritus, urinary retention, nausea, vomiting, and respiratory depression. Respiratory depression may occur early (within 1 to 2 hours) or late (within 6 to 24 hours), but the risk of respiratory depression is quite low in patients who are not opioid naive. Tolerance to the analgesic effects of spinal opioids may develop rapidly in some patients.

Catheter placement is associated with a low but definite risk of epidural infection. Nevertheless, long-term epidural analgesia is safe and effective when patients are monitored carefully and receive prompt treatment if signs of infection develop. Ideally, when spinal opioids are considered for the treatment of back pain in cancer patients, MRI of the spine should precede placement of an epidural catheter to avoid unexpected neurologic deterioration from injection in the presence of an epidural tumor.

Other drugs may be administered into the epidural space as an adjunct to opioid analgesia. Epidural clonidine may be effective in select patients with cancer pain, particularly neuropathic pain, which often responds poorly to spinal opioids. Coadministration of local anesthetics, such as bupivacaine (0.025% concentration), and opioids into the epidural space may also enhance analgesia in select cases, particularly when opioid tolerance develops.

Peripheral nerve blocks are most useful for the management of well-localized somatic pain. Examples include intercostal nerve block for chest wall pain, gasserian ganglion block for craniofacial pain, and paravertebral block for radicular pain. Temporary blockade of peripheral nerves by injection of a short-acting local anesthetic is often done as a screen. In patients responding well to temporary block, neurolytic block with absolute alcohol or phenol may provide sustained relief. Subarachnoid neurolytic block can be effective in patients with advanced disease and pain that is limited to a few spinal segments. Neurolytic procedures are most appropriate for patients with a limited life expectancy. Neurolytic blocks carry a delayed development of deafferentation pain within 6 to 12 months.

Blockade of the sympathetic nervous system may provide effective pain relief in select cases. A recent meta-analysis showed that celiac plexus block is a highly effective technique for managing pain related to pancreatic and other intra-abdominal cancers. Superior hypogastric plexus block is effective in some patients with intractable pelvic pain of malignant origin. Stellate ganglion block is sometimes useful in treating pain of upper extremity or head and neck malignancies, whereas lumbar sympathetic block is occasionally effective for pain associated with pelvic tumors or lumbosacral plexus infiltration. Studies of the comparative efficacy of these procedures and aggressive pharmacotherapy are lacking, however.

Neuroablative procedures provide pain relief by modifying pain pathways. These should be considered for patients with cancer pain when pharmacotherapy is inadequate or associated with unacceptable side effects. They can also benefit patients disabled by so-called incident or movement-related pain, which is notoriously difficult to manage pharmacologically. In general, neuroablative procedures are most appropriate for patients whose life expectancy is short, because delayed recurrence of pain, which is sometimes more severe than the initial pain, may occur secondary to deafferentation.

Cervical cordotomy interrupts the ascending lateral spinothalamic tract. This widely used neuroablative procedure is often performed percutaneously by creating a radiofrequency lesion. Cordotomy is indicated for patients with unilateral lower extremity pain of malignant origin. It is less successful in man aging midline or bilateral pain. Nociceptive pain generally responds best to cordotomy, whereas neuropathic or deafferentation pain is less reliably relieved. Bilateral cordotomies may result in bladder and sexual dysfunction or respiratory compromise and are not recommended.

Commissural myelotomy interrupts the fibers of the spinothalamic tracts because they decussate in the midline. It may be effective for selected patients with intractable midline or bilateral lower extremity pain of malignant origin. Myelotomy carries a relatively low risk of producing disturbances of bowel, bladder, or motor function. It requires an open surgical procedure, however, which generally entails a multilevel laminectomy. A more limited myelotomy, in which lesions in the dorsal columns are selectively made, has been advocated for the relief of intractable pelvic visceral pain. This procedure is based on the recent finding of a specific nociceptive anatomic pathway originating in the pelvic viscera and ascending in the dorsal columns.

Selective dorsal root entry zone (DREZ) rhizotomies have been used to treat specific pain syndromes. Dorsal rhizotomy interrupts the posterior sensory nerve roots, either by operative nerve section or by chemical neurolysis. It can be effective for intra ctable chest wall pain caused by tumor invasion or for intractable pain in the head and neck region related to invasive tumors. DREZ ablation is advocated to treat painful brachial plexus avulsion; this condition has been associated with foci of hyperexc itable cells in the superficial layers of the DREZ that are thought to be responsible for the lancinating pain that is so prominent in this disorder. As many as 30% of afferent fibers enter the spinal cord through the ventral root, however, and this may be the reason why such neuroablative procedures as creation of DREZ lesions and dorsal rhizotomies so often fail to produce permanent pain relief. As for any neuroablative procedure, dorsal rhizotomy is most appropriate for patients with a limited life expectancy but may benefit some patients with nonmalignant, particularly posttraumatic, pain syndromes.

Cingulotomy has been used to manage diffuse and midline pain of bony and visceral origin. The mechanism of pain relief is unknown.

Ablation of the pituitary gland by chemical, radiofrequency, or radiosurgical means may produce widespread analgesic effects in patients with metastatic bone pain refractory to primary chemotherapy, radiation therapy, or aggressive analgesic interventions. The mechanism of analgesia is unknown, but pain relief may be achieved in both hormone-dependent and hormone-independent tumors.

Nonpharmacologic approaches are almost always adjunctive to pharmacologic pain management but may be useful in all patients. Local application of heat or cold may relieve pain from muscle spasm. Massage can aid management of myofascial pain. Counterstimulation techniques, including transcutaneous electrical nerve stimulation (TENS), are thought to modify transmission of painful impulses and may reduce pain intensity in some patients. A randomized, controlled trial of TENS in chronic low back pain did not demonstrate efficacy, however.

Psychological support is important for cancer patients with pain. Education regarding the cause of the pain, the relation of the pain to the underlying cancer, and the expected future course of the pain is crucial. The patient should understand the rationale for the analgesic treatments offered. The preempting of fear related to the use of narcotics improves compliance. Emotional or psychological distress affects the patient's experience of pain. Some patients may require supportive psychotherapy or ps ychiatric referral for evaluation and treatment of depression or other disorders. Relaxation techniques, mental imagery, and hypnosis may be introduced as an adjunct at any stage of pharmacologic pain management. These psychological and behavioral inter ventions may enhance the effects of analgesics and reduce anxiety and emotional distress by increasing the patient's sense of autonomy and control over his or her pain.

Many chronic, incapacitating, or ultimately fatal disease pro cesses may be accompanied by severe pain. Examples include the pain associated with AIDS, arthritis, and multiple sclerosis; central pain after stroke; and musculoskeletal pain secondary to joint immobility complicating such neurodegenerative illnesses as Parkinson's disease, amyotrophic lateral sclerosis, and Alzheimer's disease. Birth defects related to prematurity and birth-related trauma that produce catastrophic neurologic deficits may be associated with pain in the neonate that is particularly problematic to assess and treat. Investigation of the physiology of nociception in the developing nervous system is just beginning, and the assessment of pain in preverbal children is difficult.

Pain is a significant problem in AIDS. About 50% of pain syndromes in AIDS patients are directly related to HIV infection, and 30% are related to the therapy for HIV disease. Neuropathic symptoms in AIDS patients with peripheral neuropathy are frequent. Acute and chronic polyneuropathy, symmetrical distal sensory neuropathy, brachial plexopathy, postherpetic neuralgia, cranial neuropathy, and headaches from acute and chronic meningitis are common. A central pain syndrome complicating toxoplasma abscess in the thalamus has also been reported. Other complications of AIDS include infectious GI disease causing esophagitis and abdominal pain, chest pain from pneumocystic pneumonia, and generalized myalgias.

Hemoglobinopathies producing the sickle cell phenotype produce acute episodic and chronic pain. Pain is an indicator of disease severity; patients who experience more than three vaso-occlusive episodes in a year have a higher mortality than patients who experience vaso-occlusive episodes less frequently. Episodes of vaso-occlusion with ischemia and infarction result in severe acute focal bone, muscle, and visceral pain that characterizes the homozygous recessive form of the disease.

Between 55% and 82% of patients with multiple sclerosis complain of pain.Ectopic spontaneous discharges in central demyelinated axons are proposed as the cause of paroxysmal pain, such as trigeminal neuralgia, which frequently complicates multiple sclerosis. The periorbital pain with eye movement of optic neuritis may be the result of traction on the meninges that envelops the swollen optic nerve. Extremity pain, especially with a dysesthetic component and painful electric shock-like sensations (e.g., Lhermitte's sign associated with neck flexion) is thought to result from posterior column demyelination and ephaptic conduction of impulses. Other causes of extremity and low back pain in multiple sclerosis are attributed to musculoskeletal abnormalities secondary to the abnormal gait and posturing typical of more severely affected patients.

Central poststroke pain develops in 1% to 2% of stroke patients. Decreased temperature sensation on the affected side nearly always accompanies the central pain. The central pain syndrome may be delayed for several years after the stroke. The pain may be superficial or deep, is often severe, and can be accompanied by hyperalgesia and allodynia. Paroxysms of severe pain may occur, sometimes elicited by emotional distress or movement. Another common cause of pain in the hemiplegic poststroke patient is mechanical shoulder pain unrelated to the central injury.

Central pain may occur from a traumatic injury at any level of the spinal cord, usually in the area of the spinal cord or root somatosensory deficit. If the nerve root is injured, neuropathic pain in a radicular distribution may add to the central pain of spinal cord origin. The pathophysiology of central pain caused by a spinal cord lesion is not known, and not everyone with similar lesions develops pain. Traumatic spinal cord lesions that have been reported to cause central pain include hemisection of the cord, cervical hematomyelia, vascular lesions, and syringo myelia. High-dose corticosteroids administered within hours of the injury may alleviate acute pain and aid recovery of neurologic function.

Although not a common problem in palliative medicine, the assessment and treatment of pain in the very young child pose significant challenges. The neuroanatomic pathways necessary for pain perception are developed in the human fetus, and analysis of the behavioral and physiologic responses to putatively painful stimuli suggests that human neonates perceive pain. Analgesic and anesthetic agents should be used in procedures and operations that are known to be painful in older persons or when the infant is responding in a way that is consistent with discomfort. The pharmacokinetic and pharmacodynamic responses to analgesic and anesthetic agents are unique in this age group.

Chronic pain occurs in 20% of patients older than 65 years. The elderly have a greater sensitivity to medications with sedative effects (e.g., opioids, tricyclic antidepressants, anticonvulsants, and benzodiazepines). Body-fluid volume decreases with age, and this may affect the distribution and concentration of drugs such as morphine.

Many medications may induce acute confusional states in the elderly and especially in the demented patient. In the elderly, enhanced adverse effects of the opioids include constipation and hypotension. With certain opioids, such as meperidine, toxic meta bolites accumulate, especially in the setting of renal dysfunction, that may cause convulsions, myoclonic jerks, hallucinations, agitation, and psychotomimetic effects. There is also an increased risk of drug interactions. In general, opioids and other d rugs with short plasma half-lives (e.g., morphine, hydromorphone, or oxycodone) are preferable to drugs with longer plasma half-lives (e.g., methadone or levorphanol); with a drug that has a short plasma half-life, there is a lower probability of accumul ation with repetitive dosing and therefore less chance that that drug will cause confusion.