Subscription is FREE for qualified healthcare professionals in the US.
11 Articles in Volume 14, Issue #5
DEA and Doctors Working Together
Working With Law Enforcement and DEA
Demystifying CRPS: What Clinicians Need to Know
Glial Cell Activation and Neuroinflammation: How They Cause Centralized Pain
History of Pain: The Treatment of Pain
Spirituality Assessments and Interventions In Pain Medicine
The Stanford Opioid Management Model
We Need More “Tolerance” in Medical Pain Management
Treating Rebound or Chronic Daily Headaches
Buprenorphine With Naloxone for Chronic Pain
More on Nitrous Oxide and Meperidine in Pain Care

Demystifying CRPS: What Clinicians Need to Know

Complex regional pain syndrome (CRPS) often has baffled the medical community. However, improved knowledge of pro-inflammatory cytokines and central sensitization of pain are unraveling some of the secrets behind CRPS.

Complex regional pain syndrome (CRPS) is a debilitating neurologic syndrome characterized by pain and hypersensitivity, vasomotor skin changes, and functional impairment. The condition also is marked by various degrees of trophic changes, including alternations in hair pattern and in nail consistency.

CRPS generally follows a fracture, but also can be triggered by other types of trauma—surgery, inflammation, stroke, crush injury, myocardial infarction, neoplasms, immobilization, and sprains.1 Psychological stressors and poor coping skills can influence the natural history and severity of CRPS.

At least 50,000 new cases of CRPS occur annually in the United States, although the true incidence of CRPS is uncertain. A 2003 study of residents in Olmsted County, Minnesota, found an incidence of 5.46 per 100,000 person years.1 However, a 2007 study by de Mos et al found an incidence of 26.2 per 100,000 person years.2 In both studies, females were 3 to 4 times more likely than males to be affected with the condition. Upper extremities are more likely than lower extremities to be implicated in CRPS, with fractures often precipitating the onset of the syndrome.

Studies have identified potential links between CRPS and several other conditions, including asthma, migraine, and osteoporosis among patients who develop the pain syndrome later in life. A significant association exists between concurrent use of angiotensin converting enzyme (ACE) inhibitors and the risk for CRPS, although the relationship has not been proven to be causal.

What is CRPS?   

In 1994, a panel working for the International Association for the Study of Pain (IASP) developed the umbrella term CRPS encompassing 2 conditions: reflex sympathetic dystrophy (RSD) and causalgia. Under the new classification, RSD became CRPS Type I and causalgia, CRPS Type II.

Although the IASP definition of CRPS is extremely sensitive, it has been criticized for leading to over diagnosis of the condition. Recognizing this problem, a panel of experts met in Budapest, Hungary, in 2003 to resolve the issue. Their modified criteria, published in 2007, are generally considered to have improved the diagnosis of CRPS (Table).3

Both forms of CRPS include an initial “warm” phase, which likely is due to extravasation of plasma and protein in the affected area, followed by vasodilation. The etiology of this phenomenon, while uncertain, may be the release of pro-inflammatory cytokines (interleukins [IL] and tumor necrosis factor [TNF]-alpha) and neuropeptides (calcitonin gene-related peptide and substance P). Tests of blister fluid, plasma, and cerebrospinal fluid of patients with CRPS have revealed elevated levels of these cytokines.

Patients with either form of CRPS also experience hyperesthesia, allodynia, and/or hyperalgesia. Hyperesthesia is the increased sensitivity to stimulation; allodynia is the pain associated with stimulus that normally provokes no pain; hyperalgesia is an exaggerated painful response to a painful stimulus.

CRPS Type I is marked by a lack of a specific nerve lesion. Patients with Type II CRPS, on the other hand, have clear evidence of nerve injury (causalgia), although symptoms may extend beyond the course of the affected peripheral nerve, differentiating the injury from isolated mononeuropathy.

Patients with Type I (but not Type II) CRPS sometimes experience spreading of their pain to areas beyond the intial locus of pain; in order of prevalence, this spreading can be contiguous (enlargement of the affected area), independent (symptoms in a distant non-contiguous location), or mirror-image (symptoms in a region opposite of the area of the initial presentation).4

There is no medical cure for CRPS. The vast majority of CRPS patients have some degree of pain and dysfunction for greater than 5 years after onset, whereas approximately 15% of CRPS patients develop intractable pain and physical impairment that impacts their ability to work and function normally.5 As with other chronic conditions, patients with CRPS commonly experience psychological symptoms related to their condition that clinicians also must address. These include depression, anxiety, fear, disuse of the affected body part, and social withdrawal.

The value of CRPS staging has become a matter of dispute among pain specialists, and clinicians tend to ignore them in practice. Further, a prospective study involving more than 800 patients with CRPS failed to cinform a sequential progression of the syndrome.6

Stage I CRPS traditionally has been thought to manifest as acute symptoms marked by sensory and/or vasomotor changes, and possibly sudomotor changes. Stage II involves increased pain, as well as vasomotor, motor, and trophic changes. Patients in Stage III experience a diminution in pain, significantly increased motor and/or trophic changes, and continued changes in vasomotor activity.

How to Diagnose CRPS

Because no definitive test exists for CRPS, the diagnosis must be made by exclusion. The list of conditions with symptoms that overlap with those of CRPS is long, and includes entrapment neuropathies, thoracic outlet syndrome, discogenic disease, deep vein thrombosis, cellulitis, vascular insufficiency, lymphedema, and erythromelalgia.7

The hallmark of diagnosis is a thorough clinical evaluation of symptoms and signs. A variety of laboratory and vascular studies can be helpful in excluding other pain diagnoses. These tests including electromyelography and nerve conduction testing to rule out peripheral neuropathy; magnetic resonance imaging (MRI) and x-rays to identify any soft tissue trauma, disc disease, central canal stenosis, or bone disorders; vascular testing to rule out deep vein thrombosis; and blood testing to rule out infectious causes of pain, cellulitis, and rheumatologic disease. However, outcomes studies have failed to support the value of such testing.

Evidence that CRPS is sympathetically mediated comes from observed changes in sympathetic activity or disturbances in blood flow. Several other tests can aid in the diagnosis of CRPS, although their utility is in dispute. Thermography using an infrared thermometer can be used to measure temperature differences across affected areas. The triple-phase bone scan has been found to be an excellent tool for ruling out CRPS Type I, owing to its greater sensitivity and higher negative predictive value than both MRI and plain x-rays.8 The triple-phase bone scan measures increased uptake of technetium into the bones of the affected limb and is thought to indicate when a patient has entered the third stage of CRPS. Sudomotor testing to assess resting and provoked sweat output of the painful limb versus unaffected limbs may indicate abnormal sympathetic activity. And the use of sympathetic nerve blocks, including stellate ganglion blocks and lumbar sympathetic blocks, can indicate the presence of sympathetically maintained pain (SMP) and can help facilitate functional restoration. An infusion of phentolamine (a mixed alpha1- and alpha2-adrenergic antagonist) may be more specific for SMP than nerve blocks.


Given the incomplete understanding of the causes of CRPS, the pathophysiology of the condition also is uncertain. However, data indicate that tissue injury to an extremity causes nerve trauma that, in turn, elicits the release of cytokines and neuropeptides (TNF-alpha, IL-1b, IL-2, substance P) in the affected area. These molecules produce peripheral sensitization. Genetic factors may exaggerate this response in certain people, although genes for CRPS have not been identified.

Nerve trauma may cause reduced density of nociceptive fibers, altering sweat glands and hair follicles. In addition, nociceptive fibers begin to express adrenergic receptors; the sympathetic nervous system (SNS) and catecholamines can trigger nociceptive firing. Decreased SNS outflow after the initiating trauma leads to vasodilatation, which reduces blood flow to the affected region and triggers local hypoxia and resultant trophic changes.

CRPS is a disease of the central nervous system (CNS) as well as the SNS. Changes in the nervous system include ongoing nociceptive input that produces central sensitization, and altered afferent input from the affected extremity, which contributes to reduction in somatosensory representation in the brain.

Some evidence suggests that patients with CRPS suffer somatosensory impairment that affects their ability to process tactile, thermal, and noxious stimuli.9 Patients experience impaired tactile sensation, for example, as the intensity of their pain increases, and they develop hyperalgesia. Several cortical activity mapping studies, such as functional MRI (fMRI), show cortical reorganization in patients with CRPS.8 Follow-up of these patients shows reversal of cortical changes in those who achieve pain reduction through therapy. Finally, at least one study indicates that CRPS may have an autoimmune element. Kohr et al found that patients with CRPS have serum autoantibodies with surface antigens of autonomic neurons.10

 Treatment Options

Optimal treatment for CRPS is multimodal and multidisciplinary. An effective approach should include management of pain as well as attention to psychosocial distress. The medical team should, when appropriate, consist of physicians in pain medicine and specialists in rehabilitation physiotherapy and psychology.11 

Pharmaceutical Management

Aggressive pharmacologic therapy should begin as early as possible after the onset of symptoms. The goals of treatment are analgesia and restoration of normal functional and psychological conditions. Clinicians should consider a wide range of medications, including anticonvulsants, corticosteroids early in treatment, topical agents, opioid and nonopioid analgesics, bisphosphonates, sympatholytic agents, and tricyclic antidepressants.

Few placebo-controlled trials have shown therapeutic efficacy of analgesics for patients with CRPS, regardless of the route of administration or drug delivery technique.

Steroids have proven effective for improving the clinical status of patients with acute CRPS. Christensen et al found that 10 mg of prednisone 3 times daily improved the clinical status of patients with less than 13 weeks of CRPS symptoms.12 Intranasal calcitonin (Fortical, others, approved for postmenopausal osteoporosis) 3 times daily also has been found to reduce pain associated with CRPS, as has intravenous sodium clodronate (300 mg daily).13 Several studies suggest that the bisphosphonate alendronate (Fosamax, others) and clodronate, which is typically used to treat symptoms of bone cancer, can reduce pain and inflammation from CRPS and improve range of motion in patients with acute CRPS.14-16

Transdermal clonidine (Catapres, others), an alpha-2 agonist that prevents release of catecholamines, may be helpful with small areas of hyperalgesia.17

Nonpharmaceutical Management

Nonpharmacologic therapies, such as sympathetic nerve blocks, can provide relief of pain and restoration of motor function throughout physical therapy (PT). A 2002 meta-analysis found weak evidence for sympathetic blockade as a therapeutic modality, with less than one-third of patients obtaining complete relief.18 Epidural and brachial plexus blocks that interrupt the sympathetic nerves also can help facilitate PT. Clinicians should exercise care with potential motor block and PT.

Evidence supporting neuromodulation and intrathecal infusions for CRPS is promising. For neuromodulation, one randomized controlled trial and several prospective studies without matched controls, as well as 8 retrospective studies, indicate that spinal cord stimulation (SCS) helps reduce pain in patients with CRPS Type I, especially when combined with PT, with benefits lasting up to 3 years. The proposed mechanism for the effect is the ability of SCS to suppress vasoconstrictor activity, activate A-alpha and A-delta afferent fibers that trigger spinal inhibitory interneurons, and release serotonin and norepinephrine into the dorsal horn to reduce pain transmission pre- and post-synaptically.

SCS also appears to improve functional status in CRPS patients, and is associated with reductions in mood and anxiety.19 In some of the studies reviewed, the benefits of SCS diminished with time. However, it is now believed that the key to long-term efficacy of SCS is early implantation and strict patient selection.20 In fact, some are even advocating that clinicians consider SCS much earlier (at 3 months) as a treatment for CRPS.21 Even the National Institute of Clinical Excellence and the British Pain Society agree that strong clinical evidence supports the use of SCS for CRPS and other neuropathic pain conditions.20

Surgical sympathectomy may be considered but only in patients who first respond to chemical sympathetic blockade. Long-term successful outcomes have been reported in 70% to 85% of patients if the surgery is performed within 1 year of diagnosis, but practical experience suggests less positive results are more likely.

All patients should receive cognitive behavioral therapy to improve coping and minimize the impact of the condition.

Emerging Therapies

Several novel treatments for CRPS have shown promise in controlling various aspects of the syndrome. These include both pharmacologic and nonpharmacologic approaches. Mirror box therapy—in which patients are trained to move an unaffected limb in front of a mirror—was first described for the treatment of phantom limb pain22 and appears to provide similar benefits for CRPS Type I.23 The rationale for mirror box therapy in these patients is that CRPS degrades proprioceptive feedback of the motor commands to the affected limb, which may increase pain and changes in the primary somatosensory cortex. Indeed, fMRI evidence suggests that there are changes in the somatosensory cortex in CRPS Type I patients. Moving the unaffected limb in front of a mirror causes cortical reorganization of the sensory homunculus by using visual feedback as a substitute for missing proprioceptive feedback.

Intravenous ketamine, an N-methyl-D-aspartate receptor antagonist, may reduce pain in patients with refractory CRPS in both anesthetic and subanesthetic doses. Case reports of ketamine “coma” have found that patients with severe CRPS report less pain without neurocognitive impairment after the 5-day course of treatment.24 Similarly, cases series and reports suggest that subanesthetic doses of ketamine reduce pain and symptoms associated with CRPS. Topical ketamine also shows promise in attenuating allodynia and hyperalgesia associated with the syndrome.25

Finally, a transdermal low-current electrostimulation therapy device, Calmare, has been approved in the United States for management of chronic neuropathic pain and cancer pain.26 It has been used in Europe for the treatment of chemotherapy-induced peripheral neuropathy (CIPN). Studies using Calmare, also known as Scrambler therapy, for management of pain following CIPN have demonstrated dramatic relief without adverse effects.27 The device is marketed for neuropathic pain conditions such as CRPS, postherpetic neuralgia, and phantom limb pain. Calmare transmits 16 sequences of low frequency electrical stimulation via electrodes that are placed in dermatomal distribution above and below the painful areas.27 Patients report an average relief lasting 3 to 4 months, although repeat sessions may be required. The cost of treatment is $2,500 for 10 sessions; it typically is not covered by insurance.


Rehabilitation is the mainstay of CRPS treatment. Physiotherapy should be based on 3 pillars: motivation, desensitization, and mobilization.28  Motivation should stress the development of a therapeutic alliance through encouragement and education to overcome movement phobia in the affected limb. Desensitization involves working with the patient to control allodynia, which may be a limiting factor in the improvement of symptoms. Doing so often requires cutaneous desensitization measures, including progressive use of corse textures, proprioreceptive challenge with scrubbing, and weight-bearing. Progressive pain management measures to provide adequate analgesia will be needed to facilitate the desensitization process. The third component, mobilization, involves the progressive increase from gentle range-of-motion exercises to stress loading, isometric strengthening, and aerobic conditioning. Pain management and psychological therapies facilitate progression through the rehabilitation pathway, and vocational and functional therapies are critical to a successful outcome.


Some evidence suggests that CRPS may be preventable. A review of the literature by Peres et al describes that vitamin C (500 mg/d for 50 days after the injury) appeared to prevent CRPS Type I in patients with wrist fractures. Avoiding relapse also may be possible by postponing surgery on the affected extremity until signs of CRPS are minimal, the use of regional anesthetic techniques such as epidural or brachial plexus blockade if surgery is unavoidable, and the administration of salmon calcitonin (100 IU/d subcutaneously) from the date of surgery or trauma until 4 weeks postoperatively.28

Last updated on: May 19, 2015
close X