The primary difference between complex regional pain syndrome (CRPS) types I and II is that there is an identifiable peripheral nerve injury in type II.
The diagnosis of CRPS is based on clinical findings, but a positive third-phase bone scan provides objective corroboration for the clinical diagnosis.
Radiographs are used to determine whether osteopenia is present, and laser Doppler fluxmetry measurements are used to evaluate cutaneous perfusion.
Decreased pain after stellate ganglion block is presumptive evidence that pain is sympathetically maintained, whereas no reduction in pain suggests that pain processing is sympathetically independent.
Medical management and surgical management vary depending on the clinical presentation and the outcome of pharmacologic agents. However, hand therapy is considered an important aspect of nonsurgical management.
Reflex sympathetic dystrophy (RSD) is a generic term used to describe post-traumatic pain accompanied by inappropriate autonomic activity and impaired extremity function ( Box 115-1 ). In Europe, this condition is called algodystrophy. However, the term complex regional pain syndrome (CRPS), as proposed by the International Association for the Study of Pain in 1993, provides descriptive terminology based on clinical features, location, and specifics of the injury, without implying mechanism, cause, or sympathetic maintenance. CRPS type I (traditional RSD or algodystrophy) occurs without an identifiable peripheral nerve injury; CRPS type II, causalgia , is clinical RSD with a peripheral nerve abnormality. Either form may be sympathetically maintained or sympathetically independent. The former is defined as pain occurring with relief or improvement of symptoms during or after sympatholytic blockade. CRPS is not synonymous with sympathetically maintained pain (SMP), in contradistinction to the consensus recommendation of the American Association for Hand Surgery in 1991. Physiologic manifestations of CRPS are normal responses to an initial noxious insult that are prolonged abnormally and persist in the absence of ongoing or impending cellular damage. Patients presenting with CRPS have incapacitating pain and functional compromise. In CRPS, a complex series of peripheral and central events affect peripheral autonomic control, modify central nervous system (CNS) activity, and produce pain, autonomic dysfunction, trophic changes, and functional impairment. Pain may be nociceptive or neuropathic. The former originates from a mechanical source in the absence of an identifiable nerve injury. Neuropathic pain emanates from an injury or dysfunction of a peripheral nerve combined with trophic changes, autonomic dysfunction, and functional impairment. Thus, a localized neuroma or neuroma-in-continuity, a cause of neuropathic pain, is not per se CRPS type II. The presence of a nociceptive focus (e.g., mechanical irritation of the wrist secondary to an unstable distal radioulnar joint or injury to a peripheral nerve) serves as a trigger to further exaggerate physiologic responses mediated through inappropriate α-adrenergic activity within the dorsal horn of the spinal cord. After a peripheral injury, excitation and sensitization of wide dynamic range neurons (internuncial connections in the dorsal horn of the spinal cord) increase the conscious appreciation of pain. SMP occurs when receptor disease [receptor disease refers to the concept that increased sensitivity of adrenergic receptors is responsible for CRPS] predominates and sympatholytic intervention (e.g., intravenous phentolamine or stellate ganglion block) decreases α-adrenergic tone with subsequent decrease in pain, normalization of autonomic function, and increased physical capability. Over time, physiologic and/or anatomic adaptation occurs or permanent structural damage may alter peripheral and CNS architectural structures. When this occurs, sympathetic blockade is no longer effective, and pain becomes sympathetically independent.
Acute atrophy of bone
Chronic traumatic edema
Major traumatic dystrophy
Minor traumatic dystrophy
Pain dysfunction syndrome
Painful post-traumatic osteoporosis
Although CRPS can develop in any patient, epidemiologically, white women who smoke cigarettes are affected most frequently. The estimated female preponderance of CRPS varies from 1 : 1.6 to 4.5 : 1. In the Netherlands, the incidence of CRPS was reported as 26.2 per 100,000 in a series of patients studied from 1996 through 2005.
Age at onset is most frequently between 30 and 55 years (mean, 40 years); however, individuals of any age may be affected. An identifiable nociceptive injury is diagnosed in less than 50% of cases. CRPS is observed frequently (20%–40%) after fracture of the distal radius. In addition, traumatic or iatrogenic injuries to the following nerves have been documented to contribute to and to precipitate CRPS: the palmar cutaneous branch of the median nerve, the median nerve at any level, the dorsal branch of the ulnar nerve, the superficial radial nerve, the ulnar nerve at the elbow, and the posterior interosseous nerve. An association between CRPS after simultaneous median nerve decompression and palmar fasciectomy for Dupuytren’s disease has been reported. Familial or genetic factors have been implicated as risk factors for CRPS. Patients with CRPS do not report higher pain scores or lower levels of psychological distress than do patients with back pain or local neuropathy. Self-induced disorders such as factitious disorders may be confused with CRPS. See Chapter 136 for more information on self-inflicted disorders.
The diagnosis of CRPS is based on clinical findings. In patients with CRPS, pain is initiated in the periphery by a noxious incident(s), is influenced by a post-traumatic event(s), is exacerbated by physiologic and/or anatomic variables, and is determined, in part, by congenital or genetic factors. There are no pathognomonic tests for CRPS; onset is frequently delayed and the average diagnosis is made 2 to 12 weeks after injury.
Symptoms and Signs
Pain, a prerequisite of CRPS, often is described as burning, throbbing, tearing, cutting, searing, shooting, and aching. Characteristic types of pain in CRPS include hyperalgesia, allodynia, and hyperpathia. Hyperalgesia, pain that is greater than expected for a given painful stimulus, is considered primary when it affects the immediate area surrounding the injury. Hyperalgesia is termed secondary when it causes distant discomfort in nontraumatized skin either proximal or distal to the initial injury area. Pain secondary to normally nonpainful stimuli is called allodynia. Hyperpathia is delayed pain that typically outlasts the initiating stimulus and spreads beyond normal dermatomal borders. Cold sensitivity (a painful response to cold exposure) is also commonly experienced by CRPS patients.
Physiologic changes include trophic disease and vascular events. Trophic changes associated with CRPS include stiffness, edema, and atrophy of the hair, nails, and skin. Hyperkeratosis of the skin may occur. Symptoms of vasomotor and/or autonomic nervous system dysfunction occur in 80% of patients. Various testing methods can be used to demonstrate abnormal autonomic function in most patients.
The subjective symptoms of patients with CRPS may be quantified by the use of validated instruments that evaluate pain, cold sensitivity, and numbness. These instruments include variations of the McGill Pain Questionnaire, the Carpal Tunnel Instrument, and the McCabe Cold Sensitivity Severity Scale. Extremity function may be analyzed using the Carpal Tunnel Instrument Function Scale and the Disabilities of the Arm, Shoulder, and Hand (DASH) form from the American Academy of Orthopaedic Surgery. The RAND SF-36 Health Survey ( www.rand.org/health/surveys_tools/mos/mos_core_36item.html ) may be used to measure the health-related quality of life of CRPS patients. This instrument assesses physical, social, and emotional functioning; perceived health; overall life satisfaction; perceived pain; and work performance.
The physical examination must include an assessment of the whole patient, with an emphasis on the cervical spine, thoracic spine, shoulder girdle, involved extremity, contralateral limb, and both lower extremities. A careful neurologic assessment is required to determine the presence or absence of discogenic or degenerative cervical disease, peripheral neuropathy, arthritis, and/or arthrofibrosis. Restrictive shoulder range of motion (ROM), or adhesive capsulitis, is also a common finding. This condition, often called shoulder–hand syndrome, negatively affects health-related quality of life and requires specific treatment modalities. Shoulder–hand problems often are overlooked without a careful examination of the shoulder. The involved extremity must be assessed for sensibility, hyperpathia, allodynia, discoloration, swelling, atrophy, vasomotor and autonomic tone, neurologic function, vascular status, grip, and pinch. In addition, any nociceptive foci should be noted. Reevaluation after or during sympatholytic treatment may reveal additional findings, may facilitate the identification of trigger areas or delineate underlying inflammatory processes, and may clarify any structural injury. If these conditions are correctable, their treatment may have a positive impact on outcome.
There is no pathognomonic marker for CRPS; therefore, tests augment and/or quantify historical data and clinical findings. The clinical evaluation may be aided by information from validated instruments evaluating health-related quality of life, symptoms, and function. Combined with a standardized examination, this information provides reproducible and quantifiable data. Specific testing modalities provide an analysis of anatomic integrity, physiologic performance, and functional capacity. Categorization of the CRPS patient using reproducible physiologic assessments guides treatment decisions and documents disease status. Standardized tests are used to evaluate bone density and osteopenia, sudomotor performance, vasomotor and thermoregulatory control, components of blood flow, and endurance.
Radiographs may be used to assess the amount of regional osteopenia, a finding observed in 70% to 80% of CRPS patients. The classic radiographic finding associated with CRPS is periarticular osteopenia; however, CRPS may affect both cortical and cancellous bone. Genant and colleagues described five patterns of resorption that occur in CRPS: patchy, irregular trabecular bone and subperiosteal, intracortical, endosteal, and subchondral and juxtachondral surface erosions. However, radiographic changes occur relatively late in the progression of CRPS ( Fig. 115-1 ).
Three-phase bone scans “should not be used as a major criterion in diagnosing reflex sympathetic dystrophy.” A third-phase scan with positive findings is not a prerequisite for the diagnosis of CRPS or SMP. The diagnostic importance of abnormal phases of bone scans is a subject of debate, and abnormal findings on scans in any phase assist in the corroboration of the diagnosis of RSD. Although it has been stated that only a third-phase scan correlates with RSD, abnormal findings in any phase of a three-phase scan document abnormal physiology of blood flow, bone turnover, or both. Unfortunately, bone scan findings do not correlate with the traditional staging criteria for RSD, do not predict recovery, do not determine the potential for response to treatment, and do not necessarily revert to normal after successful treatment. However, the presence of positive third-phase bone scan findings provides objective corroboration for the clinical diagnosis of CRPS ( Figs. 115-2 and 115-3 ).
Sudomotor performance may be evaluated using resting sweat output, galvanic skin response, peripheral autonomic surface potentials, sympathetic skin response, and quantitative sudomotor axon reflex test. These tests provide an objective measure of sweat function by indirect or direct means.
Vasomotor and thermoregulatory control can be assessed by monitoring digital pulp temperature(s), laser Doppler fluxmetry measurements of cutaneous perfusion ( Fig. 115-4 ), or laser Doppler perfusion imaging of cutaneous perfusion profiles. These studies indirectly evaluate cutaneous perfusion and provide reproducible profiles of physiologic capacity if combined with a physiologic stressor.
Under normal conditions, 80% to 90% of total finger blood flow is involved in thermoregulation. However, in patients with CRPS, the percentage of blood flow involved in thermoregulatory activity may vary. Thermoregulatory flow is calculated by subtracting the nutritional flow from the total flow. Nutritional flow may be measured by using vital capillaroscopy. This technique measures nutritional blood flow of the nail fold capillaries by direct evaluation of red blood cell flow through a single capillary loop by using an epi-illumination microscope and computerized analyses of flow ( Fig. 115-5 ). Patients with CRPS have reduced nutritional flow and are unable to modulate flow compared with normal patients. Decreased nutritional flow occurs in patients with longstanding CRPS and may be secondary, in part, to irreversible changes in arteriovenous shunt mechanisms and/or arteriovenous shunt control.
Endurance testing may be performed by using a variety of computerized diagnostic techniques that quantify muscle energy, strength, and endurance. Subtle abnormalities in function may be assessed only by careful analysis of these test results.
Diagnostic blockade may be used to determine whether CRPS is sympathetically maintained or sympathetically independent. The most common diagnostic blockade procedures performed are stellate ganglion block, a phentolamine test, epidural injection, and controlled trials of oral sympatholytic medications. Diagnostic blockade is performed classically with stellate ganglion block or phentolamine ; however, phentolamine currently is not available in the United States.
Stellate ganglion block is performed by injecting the cervical sympathetic trunk with a short- to medium-acting stabilizing agent (e.g., lidocaine) ( Fig. 115-6 ). Decreased pain after injection is presumptive evidence that pain is sympathetically maintained. The inability to obtain pain reduction after or during the test suggests either an alternative diagnosis or that irreversible peripheral changes have occurred that have changed the previously SMP to sympathetically independent pain. A brief response to oral sympatholytic medications is presumptive evidence of complex regional pain that is sympathetically maintained.
Of CRPS patients treated within the first year of injury, 80% will show significant improvement; only 50% of those treated after 1 year will improve. However, patients with CRPS after distal radius fractures have a poorer prognosis; stiffness and decreased finger function at 3 months correlate with residual impairment and morbidity at 10 years. Although early intervention is important, it is not always possible, and many patients are treated inadvertently before diagnosis. Thus, the natural history may be improved by medical intervention despite an incorrect or incomplete diagnosis.
Effective treatment of CRPS requires recognition and prompt intervention. Before initiating treatment, it is appropriate to stage the CRPS patient by determining whether pain is sympathetically maintained or sympathetically independent, total blood flow is high or low, evidence of abnormal arteriovenous shunting and/or nutritional deprivation exists, and permanent structural damage has occurred. The degree of soft tissue contracture should be assessed because patients with significant arthrofibrosis and atrophy often have sustained irreversible trophic changes that indicate a less favorable prognosis. Nociceptive foci should be identified and treated. Effective sympatholytic treatment often requires the use of multiple modalities.
Hand therapy is an important aspect of management of CRPS affecting the upper extremity. The active involvement of a hand therapist in the management of a CRPS patient provides an independent assessment of progress, continuity of care, and a vital feedback loop among the patient, physician, insurance carriers, and rehabilitation personnel. Pain modulation techniques used in hand therapy include contrast baths, desensitization, Fluidotherapy (Chattanooga Medical, Chattanooga, TN), static orthotic positioning, electrical stimulation, and ultrasound. Edema is controlled by using manual lymphatic drainage techniques or light compression garments. ROM activity can be aided by ultrasound and paraffin baths along with dynamic orthotic positioning and continuous passive motion equipment as long as the patient’s pain does not increase.
The use of oral, topical, and parenteral pharmacologic intervention is well documented in the management of CRPS. The most commonly used drugs include antidepressants, anticonvulsants, membrane-stabilizing agents, adrenergic agents, and steroids. The use of these medications is largely empirical; however, theoretical mechanisms support their use in chronic pain. Oral medications (e.g., anticonvulsants, local anesthetics) may stabilize hypersensitive membranes, may provide competitive inhibition of neurotransmitters (e.g., bretylium), may block or influence neurotransmitter activity and/or receptor affinity or both, or may increase nutritional flow and/or decrease inappropriate arteriovenous shunting.
The use of almost all the agents discussed are unlabeled indications according to the U.S. Food and Drug Administration. Therefore, the individual practitioner must be familiar with the pharmacology, physiologic actions, side effects, potential complications, indications, potential drug interactions, and contraindications of any medications prescribed.
In recent studies, vitamin C, at a dose of 500 mg/day, has been observed to decrease the incidence of CRPS after distal radius fractures and may provide a prophylactic advantage. A tricyclic antidepressant often is used in conjunction with a serotonin reuptake inhibitor. For example, amitriptyline and paroxetine may be used simultaneously ( Tables 115-1 and 115-2 ). Anticonvulsants are commonly used because of their theoretical effect on membrane stabilization ( Table 115-3 ). The use of oral membrane-stabilizing agents such as mexiletine and tocainide is limited because of the numerous side effects associated with these agents. Steroids are used frequently and are advocated by many as a primary oral medication ( Table 115-4 ).
|Drug||Mechanism of Action||Dose (Range, mg)||Common Side Effects|
|Anticholinergic Effects||Seizures||Orthostatic Hypotension||Conduction Abnormalities||Sedation|
|Imipramine (Tofranil, SK-Pramine)||Blocks reuptake of amines:||50–75 (50–300)||+++||+++||++++||++++||++|
|Amitriptyline (Elavil, Endep)||Blocks reuptake of amines:||25–75 (50–300)||+++++||+++||+++||++++||+++++|
|Doxepin (Sinequan, Adapin)||Blocks reuptake of amines:||50–75 (50–300)||+++||+++||++||++||++++|
|Desipramines (Norpramine, Pertofrane)||Blocks reuptake of amines:||50–75 (50–300)||++||++||+++||+++||++|
|Nortriptyline (Aventyl, Panclar)||Blocks reuptake of amines:||25–50 (50–150)||+++||++||+||+++||+++|
|Protriptyline (Vivactil)||Blocks reuptake of amines:||10–20 (15–60)||+++||++||++||++++||+|
|Maprotiline (Ludiomil)||Blocks reuptake of amines:||50–75 (50–225)||+++||++++||++||+++||+++|
|Trazodone (Desyrel)||Blocks reuptake of amines:||50–150 (50–600)||+||++||+++||+||+++|