A Practical Guide for Complex Regional Pain Syndrome in the Acute Stage and Late Stage

Clinical Definitions

The diagnosis of CRPS is contingent on the presence of regional pain combined with autonomic dysfunction, atrophy, and functional impairment affecting musculoskeletal, neural, and vascular structures ( Fig. 53.1 ). Type 1, or “classic,” RSD is not associated with an identifiable peripheral nerve injury. Type 1 CRPS is initiated by trauma and is associated with pain that is out of proportion to the injury, swelling, color changes, skin changes, atrophy, and stiffness. Type 2 has similar signs and symptoms in addition to an identifiable peripheral nerve injury (causalgia). Type 3 includes nontraumatic causes producing extremity pain such as myofascial syndrome ( Table 53.1 ). The type 3 category of CRPS is controversial; however, the use of this category expands the concept to include all extremity pain pathologic conditions. Pain can be acute, chronic, or mixed and can be classified as nociceptive (pain that arises from actual or threatened damage to nonneural tissue and is secondary to the activation of nociceptors), neuropathic (pain initiated or caused by a primary lesion or dysfunction in the nervous system), or nocicplastic (pain that arises from altered nociception despite no clear evidence of actual or threatened tissue damage causing the activation of peripheral nociceptors or evidence for disease or lesion of the somatosensory system) in origin. In addition, pain may be sympathetically maintained or sympathetically independent. The traditional hallmark of CRPS is pain without evidence of actual or threatened tissue damage or evidence of a disease or lesion directly affecting the somatosensory system. This chapter discusses only CRPS type 1 and type 2.

Budapest Criteria and Definitions

The diagnostic criteria for CRPS have evolved over the years. A consensus workshop in 1994 decided to use the term CRPS rather than RSD and developed specific diagnostic criteria codified by the International Association for the Study of Pain (IASP) ( Box 53.2 ). These ISAP criteria had high diagnostic sensitivity but poor specificity for discriminating between diagnostic groups. In 2003, another meeting was held in Budapest to improve the ISAP criteria ( Box 53.3 ). The modified diagnostic criteria were called the Budapest Criteria and were based on empirically derived criteria reported in the literature. The four distinct CRPS components included in the Budapest Criteria retained the sensitivity of the IASP criteria and enhanced the specificity of the diagnostic criteria.

The distinction “major nerve damage” is confusing as a discriminator for type 2 versus type 1 CRPS. Injury to an identifiable nerve is appropriate to designate a diagnosis as type 2. Historically, causalgia was associated with an injury to a “major” mixed motor and sensory nerve (e.g., the median or sciatic nerve).

Neuropathic pain is defined as pain initiated or caused by one or more lesions or dysfunction of the peripheral nervous system or central nervous system (CNS). Nociceptive pain results from damage to tissues as a result of thermal, mechanical, chemical, or other irritants. Neuropathic pain may be associated with CRPS type 2, or it may be considered a distinct entity, depending on its source and clinical manifestations ( Table 53.2 ). Mononeuropathy, if symptoms and signs extend beyond the nerve distribution, may constitute CRPS type 2.

The diagnosis of CRPS is a clinical diagnosis based on the history and physical examination along with diagnostic testing; it is dependent on the experience and clinical judgment of the physician. Therefore criteria such as in the Budapest recommendations provide a valuable, but not an infallible, process to support the diagnosis. It is crucial to remain cognizant that CRPS is a heterogeneous process; symptoms and signs may fluctuate with time and may “resolve” with or without sequelae. Patients without pain who “no longer meet diagnostic criteria but [may be] still symptomatic” can have considerable disability and deformity associated with contractures, and so on.

Attempts to identify a pathognomonic marker for CRPS remain unsuccessful; there are no diagnostic tools, serum biomarkers, or laboratory findings that define the condition. ^, Early recognition along with prompt treatment of CRPS is important to minimize permanent loss of function. However, even with prompt recognition and treatment, patients with CRPS may experience permanent impairment and disability. Many patients also face work incapacity, changes in occupation, and/or psychological disorders. ^,

CRPS may include a component of sympathetically maintained pain (SMP) or sympathetically independent pain (SIP). These terms recognize the dynamic nature of dystrophic responses and stress the value of reliable and consistent clinical observations and descriptions. The diagnosis of SMP is based on pain relief with sympatholytic medications or sympathetic blocks ^, ; however, SMP may become SIP over time. Objective and reproducible methods to assess pain, quantify trophic changes, define autonomic dysfunction, and measure functional impairment are important to provide consistent treatment regimens and reliably assess outcomes. ^,

Late or Chronic CRPS

Patients with late, longstanding, or chronic CRPS may not continue to fulfill the diagnostic criteria for the process. However, they can experience residual deformity and/or debility.

Historical Review

• Ambrose Paré, 16th century: Charles IX experienced burning pain after phlebotomy

• Percivall Pott, 1771: described pain after nerve injury

• Silas Weir Mitchell, 1864: classic description of causalgia in the upper extremity with a brachial artery injury and partial injury of a median nerve

• Sudek, 1900: bone demineralization associated with posttraumatic pain

• Leriche, 1916: posttraumatic “burning” pain

• Evan, 1947 and Bonica, 1973; term reflex sympathetic dystrophy

• Proceedings of the VI Congress on Pain, 1991: introduced the term complex regional pain syndrome

• Budapest criteria, 2007: diagnostic criteria based on symptoms and signs with improved specificity

Physiology of Pain

Normally, pain is perceived only in the presence of actual or impending cellular tissue damage; persistent pain is pathologic in the absence of continued trauma. Painful peripheral nociceptive experiences that include cellular damage produce secondary inflammation by activating and sensitizing polymodal low-threshold mechanoreceptors and nociceptor afferent neurons, which, in theory, produce ectopic chemosensitivity to α-adrenergic agonists. This information is relayed via the process of transduction through small myelinated (Aδ), large myelinated (Aβ), and small unmyelinated (C) afferent fibers to the dorsal horn of the spinal cord, where sensitization of wide–dynamic range (WDR) neurons contributes to central nociceptive discharge ( Fig. 53.2A ). Within the dorsal horn, excitatory amino acids serve as the principal neurotransmitters. Transmitters include N -glutamate, aspartate, α-amino-3-hydroxy-5-methyl-4-isoxazopropionic acid (AMPA), N -methyl- d -aspartate (NMDA), and substance P. Of these, the NMDA receptor-transmitter interaction produces long-lasting potentials, is refractory to stimulation, and is theorized to play an important role in CRPS types 1 and 2. Nociceptive input is modulated via descending pathways, and both peripheral and central factors are required for the perception of pain (see Fig. 53.2B ). Pain intensity is determined by the magnitude and extent of the initiating/ongoing event, afferent input, efferent modulation, and CNS interpretation. Conscious appreciation of nociceptive (painful) experiences is dependent on a complex interplay of afferent and efferent information modulated and balanced by physiologic adaptations. Vasomotor disturbances may result from a variety of mechanisms, including antidromic vasodilation, vasoparalytic dilation, normal somatosensory reflexes, and denervation super sensitivity. Responses to nociceptive stimuli can vary considerably among individuals. The presence of nociceptive-induced inappropriate transmitter-receptor activity can affect peripheral microcirculatory control and, thereby, result in impaired nutritive blood flow and may sensitize CNS pathways. ^, Autonomic nervous system involvement in the pathophysiology of CRPS is acknowledged.

Acute Versus Late or Chronic Pain

Acute pain is initiated during tissue injury or destruction, and the presence of acute pain may be beneficial or harmful. Beneficial effects include physiologic responses for maintenance of blood pressure, cardiac output, intravascular volume, and appropriate homeostasis. Acute pain warns the host of danger, prevents inappropriate motion of an injured extremity, and can diminish additional harm from repetitive injury. Persistence of pain beyond the need for protective action is unpleasant and can induce hypertension, tachycardia, coagulopathy, hyperglycemia, anxiety, fear, and chronic pain. Chronic pain that occurs in the absence of ongoing tissue destruction or that provides an inappropriate reflection of the intensity, magnitude, or duration of the tissue damage/compromise is pathologic. Although the pathophysiology of chronic pain is incompletely understood, the following processes can contribute to its establishment: persistent mechanical irritation of peripheral neural structures, incomplete regeneration of peripheral nerves, abnormal neurotransmitter activity, nutritional deprivation secondary to abnormal arteriovenous shunting, and central imprinting.

After trauma or surgery, a transient period of dystrophic extremity function is normal. However, it is abnormal for hyperpathia (heightened pain experienced by a given stimulus), allodynia (a painful response to normally nonpainful stimuli), vasomotor disturbances, and functional deficiencies to persist. If untreated, these conditions may progress to permanent compromise of the extremity. Importantly, interventions modify progression, and treatment may temporarily relieve symptoms if/when they recur. Posttraumatic alterations in extremity physiology follow a variable time course. Consequently, abnormal prolongation of these otherwise normal responses is pathologic, and, over time, irreversible changes in anatomic structures or physiologic processes may occur ( Fig. 53.3 ). When these irreversible changes occur, pain may persist, decrease, or become minimal and/or permanent deformity may occur. Therefore CRPS may be considered an abnormally severe or prolonged manifestation of a normal postinjury response. Abnormally prolonged dystrophic events may damage or compromise the arteriovenous shunt mechanism, ^{, , ,} produce arthrofibrosis, ^, cause excessive osteopenia, alter neuroreceptor function, and result in central pain imprinting and/or any combination of these sequelae.

Demographics of Complex Regional Pain Syndrome

In 2003, the incidence of CRPS in Olmsted County, Minnesota, was 5.46 per 100,000 person-years, with four times as many female as male patients affected. From the electronic medical records of more than 600,000 patients listed in the Dutch Health Care System, the overall incidence rate of CRPS in the Netherlands was higher at 26.2 per 100,000 person-years, with females affected three times more often than males. In the Dutch population-based study, the upper extremity was more frequently involved compared with the lower extremity; fractures were identified as the most common precipitating event. ^, The mean age at onset was 46 years in the Minnesota study and 53 years in the Dutch study. ^, CRPS can also develop in children and adolescents with the majority of cases occurring in girls under the age of 12 involving the lower extremity. A favorable outcome was noted in most of pediatric CRPS patients. The risk of developing CRPS in a second anatomic location in patients with a history of CRPS is greater than the risk occurring in the general population. The incidence of cigarette smoking is higher in RSD patients, and cigarette smoking is statistically linked to RSD. Data suggesting a possible familial or genetic predisposition to CRPS is postulated but not confirmed. A predisposition to CRPS type 1 has been suggested on the involved side of hemiplegic stroke patients. There are frequent references that the majority of patients in whom RSD/CRPS was diagnosed within 1 year of injury (with treatment initiated) will improve significantly, ^, and patients with untreated symptoms (50% or more) lasting longer than 1 year will experience profound residual impairment. Based on these generalizations, early diagnosis and treatment portends a better but not guaranteed outcome. ^, The direct and indirect costs of CRPS are substantial; protracted medical care is expensive; loss of wages is major, and malpractice claims—when awarded—are greater. ^,

Fracture of the distal end of the radius and ulna is one of the most common injuries associated with CRPS, and its occurrence complicates postinjury recovery in a measurable percentage of patients. The natural history of CRPS after distal radius fractures includes finger stiffness or “poor function” at 3 months after injury and correlates with residual RSD at 10 years. CRPS is more likely to occur after distal radius fractures in patients managed with tight casts or following a nociceptive or neuropathic event such as compression of the median nerve, overdistraction, instability of the distal radioulnar joint, or ulnar fracture. Dystrophic pain after surgical release of the carpal tunnel is commonly encountered in referral centers. ^,

Common nerve injuries occurring during operative procedures that can precipitate CRPS include: (1) injury to the palmar cutaneous branch of the median nerve during carpal tunnel surgery, (2) damage to the superficial branch of the radial nerve during surgical approaches to the first and second dorsal compartments for tenosynovitis, and (3) trauma to the dorsal branch of the ulnar nerve during a surgical approach to the distal ulna. ^, Reports that CRPS can produce devastating consequences in association with simultaneous median nerve decompression at the wrist and partial palmar fasciectomy for Dupuytren contracture are not correct, and therefore these procedures can be performed safely.

Psychological Factors and Effects

CRPS is not a psychogenic condition. Extensive reviews of the existing literature do not support a psychological causation or an associated personality disorder. However, chronic pain is known to play a role in psychological well-being, with 58% of 283 consecutive patients admitted to pain centers, regardless of cause, fulfilling the criteria for personality disorders. ^, Dependent, passive-aggressive, and histrionic personality disorders are common in patients with chronic pain and are seen in patients with CRPS. Behavioral responses that reduce extremity use exacerbate edema and atrophy.

Symptoms and Signs

The pain associated with CRPS is observed in a variety of clinical manifestations ( Table 53.3 ). The pain may be described as burning, throbbing, pressing, cutting, searing, shooting, or aching (or any combination of these descriptors). Hyperalgesia, or the perception of pain greater than would be expected, may be primary affecting the area of injury or secondary affecting nontraumatized surrounding areas or the entire limb. Identification of the zone of primary pain can aid in the localization of a nociceptive focus; however, diagnosis of the etiologic trigger event may not be possible until successful sympatholytic management of the secondary hyperalgesia is complete. Allodynia, or perception of pain initiated by normally innocuous stimuli, is a characteristic of sympathetically maintained CRPS. Hyperpathia, or pain produced by painful stimuli that appears with a delay, outlasting the initiating stimulus and spreading beyond the normal neural distribution, is frequently encountered. It is important to differentiate spontaneous pain from stimulus-evoked pain . Difficulty sleeping because of burning pain is common and can be an early portent of progressive symptoms and signs.

Quantifying subjective complaints of pain with the use of standardized and validated instruments or questionnaires allows an objective analysis of symptoms. ^, Useful instruments to evaluate symptoms and health status include the visual analog scale (VAS) for pain, the 36-item health status questionnaire (SF-36), the McGill Short Form Pain Questionnaire, and self-administered questionnaires designed to assess upper extremity symptoms/function. ^, The self-administered questionnaire used for assessment of the severity of symptoms and functional status for patients with carpal tunnel syndrome is applicable to dystrophic states caused by carpal tunnel surgery or injury to the median nerve at the wrist and allows the examiner to graphically follow the patient’s health-related quality of life ( Fig. 53.4 ). ^{, ,} Cold intolerance, or pain on exposure to cold, can be analyzed by the McCabe Cold Sensitivity Severity Scale. The importance of standardized physiologic testing and instruments to document signs and symptoms is supported by the lack of reproducibility of isolated clinical evaluations. However, the use of published criteria and scoring systems for CRPS can be helpful. The Budapest criteria facilitate the diagnostic process.

Trophic changes —stiffness; edema; osteopenia; atrophy of the hair, nails, or skin; hypertrophy of the skin (or hyperkeratosis); or any combination of these changes—may be present in patients with CRPS (see Fig. 53.1 ). Changes in the skin, hair, or nails are seen within 10 days of onset in 30% of extremities with CRPS type 1 (RSD). Despite treatment that reduces pain, cold intolerance, pain after use of the affected extremity, joint stiffness, nail and hair growth abnormalities, sensory disturbances, loss of finger extension/flexion, decreased grip strength, and shoulder stiffness are frequent and may persist. Osteopenia is common, involves both cortical and cancellous (trabecular) bone, and requires significant demineralization for visualization on plain radiographs ( Fig. 53.5 ). Objective analysis of bone demineralization requires dual-photon absorptiometry or quantitative scintigraphy. ^, Stiffness and atrophy of joints, muscles, and tendons may become apparent during endurance testing. ^, Movement disorders and dystonic posturing can also occur.

Physical Examination

The clinical characteristics of CRPS are related to altered extremity physiology. The magnitude of symptoms and signs is a reflection of the initiating event, postinjury response, inherent patient adaptability, interventional modalities, and time. The classic dystrophic progression from acute (<3 months) to dystrophic (3 to 6 months) to chronic (>6 months) occurs infrequently because of individual variability and the effects of partial treatment. Patients with CRPS who are evaluated for a painful, hot, swollen, stiff hand are easily identifiable; however, the precipitating nociceptive trigger (e.g., contusion of the superficial radial nerve) may not be identified until the SMP is managed and the dystrophic manifestations are treated. Once allodynia and hyperpathia are alleviated, careful examination can delineate an underlying disorder that may be amenable to treatment. ^, Ineffectively or partially treated patients with CRPS or variants of CRPS can be labeled as having “poor results” or as “noncompliant patients.” In the postoperative or posttraumatic period, unusual swelling, stiffness, pain, and restlessness can represent a dystrophic response.

The physical examination of suspected CRPS patients should include a neurologic assessment and evaluation of the cervical and thoracic spine. The presence of cervical disease (either discogenic or degenerative) may exacerbate CRPS, a form of “double-crush syndrome.” Preexisting or acquired thoracic outlet or distant compression neuropathies can represent a “triple or more” crush syndrome. Cervical spine and shoulder range of motion should be recorded, and the brachial plexus should be evaluated. The entire arm should be assessed to rule out evidence of vascular or neurologic compression within the thoracic outlet. An adhesive capsulitis, or shoulder-hand syndrome, is a frequent sequela that can be overlooked unless the shoulder is examined. Hypersensitivity, vascular adequacy, edema, sensibility, joint range of motion, motor function, grip, pinch, fibrosis, sweating, and vasomotor tone of the entire extremity must be evaluated. Grip strength is frequently abnormal and should be assessed carefully. Currently, there are no objective laboratory tests to aid in the diagnosis of CRPS.

Mechanical Nociceptive Focus

Identification of nociceptive foci, which include peripheral nerve lesions and mechanical derangements, is important. Evaluation before and after sympatholytic intervention may delineate “trigger events” that might have initiated or propagated the dystrophic symptoms. ^, Nociceptor input may be associated with SMP or SIP. However, early diagnosis plus prompt surgical correction of SMP associated with peripheral nerve injury is an effective management approach.

Diagnostic Testing

Bone Scan (Scintigraphy)

Historically, three-phase technetium 99m bone scanning (TPBS) was performed commonly and had a major role in the diagnosis of RSD. ^, With TPBS, the first, or “dynamic,” phase lasts 2 to 3 minutes and provides an assessment of digital perfusion. The second phase, or “blood pool image” or “tissue” phase, allows an assessment of total perfusion over a 3- to 5-minute period. The third phase, a standard bone scan, evaluates uptake of radiotracer in bony structures ( Fig. 53.6 ). In patients with CRPS, the use of a forearm or arm tourniquet or blood pressure cuff may affect accumulation of radiotracer in the hands or digits during phases I and II. Traditionally, scans have been considered “positive” if asymmetric flow is noted in phases I, II, or III. However, reports suggest that the diagnostic yield of phase III alone equals the diagnostic yield of the three-phase scan ( Fig. 53.7 ). Mackinnon and Holder have stated that a “strictly interpreted” phase III scan demonstrating “diffuse increased tracer uptake in the delayed image is diagnostic for RSD.” Although a positive phase III scan provides specific corroboration of the clinical diagnosis of CRPS, the sensitivity of scintigraphy is insufficient to provide standalone criteria, and therefore it should be used to corroborate clinical suspicion. In the majority of studies, a “positive” scan has high specificity but poor sensitivity. ^{, , ,} Quantitative scintigraphy provides useful physiologic data, allows quantitative assessment of regional bone loss associated with complex regional pain, and adds useful diagnostic information. ^{, ,} However, bone scan data do not correlate with symptoms, do not provide prognostic information, and are unable to aid in the prediction of which interventional approaches might be successful. ^,

In the majority of patients, the abnormal bone turnover reflected in the TPBS is diffuse throughout the hand and wrist. The segmental distribution of RSD isolated to a digit, a single ray, or multiple rays can be evaluated by a bone scan to delineate the affected areas ( Fig. 53.8 ). Although TPBS is used frequently, there is no evidence to support (1) diagnostic accuracy in variant, partially treated, or late-stage CRPS ; (2) prognostic data that predicts outcome ^, ; or (3) value in determining management decisions. ^{, ,}

At this time, a positive phase III bone scan is not a prerequisite for the diagnosis of CRPS or SMP. Bone scans do not correlate with traditional staging criteria for RSD, are not part of the diagnosis of SMP or CRPS, do not predict recovery, and will not predict the potential for response to treatment. ^{, ,} However, the presence of a positive phase III bone scan provides objective support for the clinical diagnosis of CRPS.

Regulation of Microvascular Flow

Under normal conditions, total digital blood flow is composed of 80% to 95% thermoregulatory flow and 5% to 20% nutritional flow; the distribution of nutritional versus thermoregulatory flow is governed by complex factors that control arteriovenous shunting. ^{, ,} Abnormal autonomic control of microvascular perfusion in patients with CRPS is reflected, in part, by ischemia secondary to inappropriate arteriovenous shunting and decreased nutritional flow. ^{, , , , , ,} The importance of nutritional flow in the pathogenesis of CRPS is supported by the segmental distribution of trophic events, the rapidity of action of successful sympatholytic interventions, and the similarity of pain patterns in patients with contrasting clinical findings. Abnormal arteriovenous shunting in bone has been documented in patients with CRPS. Nutritional deprivation may exist in both a “warm, swollen” hand with high total flow and a “cold, stiff” hand with low total flow. The common underlying finding in both clinical situations is nutritional deprivation resulting from inappropriate arteriovenous shunting ( Fig. 53.9 ). Pain relief in CRPS is not dependent on peripheral vasodilation (thermoregulatory flow) after spinal cord stimulation, thus suggesting the importance of nutritional flow. In addition, there is growing evidence that physiologic subgrouping of CRPS patients, in contradistinction to staging based on timing, is important. For example, “warm” CRPS responds differently to intervention than “cold” CRPS does.

Total digital blood flow—both the thermoregulatory and nutritional components—can be analyzed by digital temperature measurements, laser Doppler fluxmetry, and vital capillaroscopy. The use of a stressor to perturb physiologic homeostasis by thermal, emotional, or ischemic methods is necessary to obtain reproducible information, to evaluate dynamic response patterns over time, and to maximize sensitivity ( Fig. 53.10 ). ^, One technique that combines a stressor with physiologic measurements is isolated cold stress testing. ^, In this procedure, total digital blood flow and microvascular perfusion can be evaluated by monitoring digital temperature and laser Doppler fluxmetry before, during, and after controlled thermal stress. The heterogeneity of digital microvascular perfusion may be analyzed by laser Doppler techniques and is reflected in average pulp temperature readings. Nutritional flow may be measured directly by vital capillaroscopy ( Fig. 53.11 ). ^, Analysis of temperature, laser Doppler fluxmetry, microvascular perfusion, and nutritional flow permits an objective determination of autonomic vasomotor stability and physiologic staging of autonomic sympathetic performance, defines the adequacy of nutritional flow, and provides an assessment of the effects of interventions. ^{, ,} For example, the effectiveness of sympathetic blockade can be assessed by analysis of digital temperature and microvascular perfusion during the application of a physiologic stress ( Fig. 53.12 ). Laser Doppler perfusion imaging provides an assessment of microvascular perfusion over a 12- × 12-cm area and can be used to evaluate the effects of sympathetic blockade on skin blood flow.

Sudomotor Function

Sudomotor activity may be assessed by cumulative unstimulated resting sweat output, QSART, galvanic skin response, combined temperature and peripheral autonomic surface potential reflex response, and analysis of sympathetic skin response evoked by activation of sympathetic unmyelinated efferent fibers. These techniques can be used to obtain quantitative measurements of sweat function, which then can be contrasted to contralateral or ipsilateral areas or normal controls (or both). When combined with microvascular perfusion indices of temperature or laser Doppler fluxmetry, these data supplement the clinical examination, document autonomic dysfunction, and provide objective evidence of the effects of intervention. ^,

Staging

The concept of “acute stage” and “late-stage” CRPS can be helpful, but many patients do not follow a consistent progression from one stage to another. The length of the acute phase of CRPS is variable and is affected by the type of injury, patient coping mechanisms, and treatment. The acute stage encompasses treatment from onset to the development of atrophy and fixed contracture. Treatment of the acute stage is designed to alleviate pain, decrease abnormal arteriovenous shunting, manage edema, and correct nociceptive and/or neuropathic abnormalities. Management of patients with late stage CRPS is divided into (1) pain palliation using chronic medications (e.g., ketamine) or peripheral nerve or centrally implanted devices (e.g., dorsal column stimulators) and (2) correction of the fixed deformity or contracture.