Complex Regional Pain Syndrome: Diagnosis and Treatment

Complex regional pain syndrome (CRPS) is essentially a clinical diagnosis. Two terms used to describe the associated pain-allodynia and hyperalgesia-are notable in the criteria for both types of CRPS. Allodynia is pain caused by a stimulus that is not usually painful and is commonly the most dramatic presenting symptom of these disorders. Patients may wear loose fitting clothing to limit contact between the fabric and the skin in the affected area. In more severe cases, patients may complain that even having bedsheets touching the body part is intolerable. In hyperalgesia, a normally painful stimulus causes more discomfort than expected. Both allodynia and hyperalgesia are covered by the more general term “hyperesthesia”-an increased sensitivity to stimulation. 

The only other common pain condition in which allodynia frequently occurs is postherpetic neuralgia (PHN), which is easy to differentiate from CRPS because of both the location and the usual presence of infection before the onset of PHN along the distribution of a nerve in the trunk or head. My recommendation is that when patients have allodynia following trauma to the affected part of the anatomy, usually an extremity or, less frequently, the genitalia, they should be treated as if they have CRPS unless there is some other clear diagnosis.

In most cases of CRPS, sudomotor and vascular changes will occur, most notably edema and changes in blood flow resulting in skin temperature changes in the affected body part. Many, but far from all, patients develop dystrophic changes but these do not usually occur until several months after the onset of the pain. At this stage, x-ray films may show a “ground-glass” over the painful area. A triple-phase bone scan where there is increased pooling present in the early bone uptake phase is considered to be confirmatory of the diagnosis. However, while it is fairly specific its sensitivity appears to be somewhat limited; the absence of these changes on this test do not rule out the diagnosis.

Treatment
The earlier CRPS is treated the greater the likelihood of at least reduction in the pain if not complete resolution of it. Unfortunately, as noted in Part I, in most patients the diagnosis is not made until substantial time has passed. Many different treatments, pharmacologic and nonpharmacologic, have been tried for CRPS, but few have demonstrated much impact. A recently updated guideline is based on a comprehensive review of treatments for the disorder and makes several recommendations based on the current literature.¹

The guideline’s review found that of greatest importance is treatment that focuses on functional restoration of the affected limb primarily through physical and occupational therapies. Patients often protect and limit use of an affected limb either because of pain or the fear of pain, which can rapidly lead to atrophy and additional difficulties, such as increasing weakness that result in a downward spiral of increased pain and decreased function. As is seen with chronic pain related to other disorders, there is little chance of improvement unless the patient tries and is encouraged and supported to remain as active and functional at as high a level as possible.

Pharmacologic. There have been a surprisingly limited number of rigorous studies to evaluate management of CRPS with medications. The best the recent guideline could say based on the available evidence is that because CRPS appears to be primarily a neuropathic form of pain, the medications that have been found to be optimal for neuropathic pain are recommended for initial treatment. These are primarily the serotonin-norepinephrine reuptake inhibitor (SNRI) antidepressants, including the tricyclics, duloxetine, and venlafaxine, and anticonvulsants, most notably pregabalin and gabapentin.

Opioids also may be beneficial for pain from CRPS as they are for other neuropathies. As to which opioids are most likely to work, the best recommendation the guideline was able to provide was to try either tramadol, because it is not only contains an opioid but also an SRNI, and methadone, because of its additional N-methyl-D-aspartate (NMDA) receptor antagonistic effect. I agree with the use of methadone, because I believe it is an excellent analgesic. However, as studies have demonstrated that most of the analgesic effects of tramadol appear to be related to its SNRI activity my preference is to use one of the antidepressants that produce greater levels of this activity.
 
Topical analgesics may also be beneficial. The lidocaine 5% patch not only provides pain relief with a minimal risk of adverse effects but the patch itself also may function as a barrier against environmental stimuli that may exacerbate the pain. Although the review also reported the benefits of capsaicin, I find that the heat commonly experienced when the cream is applied may acutely exacerbate CRPS pain because of the associated allodynia.

Other medications that might be beneficial in at least some cases include the NSAIDs and corticosteroids when swelling is present.
 
Interventional. Many interventional modalities have been tried for CRPS with mixed results, at best. When the disorder was considered to be the result of dysfunction in the autonomic nervous system, sympathectomies were frequently performed. The results from this surgery generally proved to be disappointing, however. The fact that it was irreversible and not uncommonly left patients with even worse pain has mostly eliminated it as a treatment option for CRPS.
 
Sympathetic blocks, stellate ganglion blocks for upper extremity pain and lumbar sympathetic blocks for lower extremity pain, have been and continue to be widely used for the treatment of CRPS. The guideline found that these blocks may be best at differentiating CRPS from a more specific sympathetically mediated pain and found that there is limited evidence to support their value for treatment of CRPS. 
 
Sympathetic blocks, based on my own experience, are worth trying for CRPS if they can be administered within 6 to 8 weeks after the initial onset of the pain. I refer all patients who fit into this group in whom I suspect CRPS for these blocks. After this relatively brief period, the success rates of the blocks appear to markedly decline. Because so very few patients are fortunate enough to have their CRPS diagnosed so quickly, most will be treated with these blocks months if not years after the pain began when the likelihood they will have any impact is limited.
 
Many different types of intravenous regional anesthesia agents have been tried, but the review found little evidence to support their use.

Psychotherapy. Psychological therapy for patients with CRPS may be overlooked as an option but the guideline review found this intervention to have a potentially significant impact on CRPS pain.  Because many people, including health care professionals, are often skeptical about the pain reported by patients with CRPS, patients may resist psychotherapy, fearing that their pain is being discounted as “not real” or “all in their heads.” It is important to disabuse them of this perception and provide reassurance and support. Biofeedback and cognitive-behavioral therapy, in which patients’ negative beliefs about their pain and responses to it are explored and modified, appear to be especially beneficial. 

It is important to note that because of the severity of the pain associated with CRPS, the difficulty obtaining a diagnosis with frequent dismissals by health care professionals, and the not infrequent failure of treatments, many patients do suffer comorbid depression and it is important that this be treated. 

Course of CRPS
The course of CRPS is variable. Apart from treating it as early as possible and maintaining the highest level of functioning despite the pain, there are few useful predictors for the course of CRPS. Even in those who receive optimal treatment, CRPS may not improve or may even worsen over time while patients who receive less or no specific treatment may get better. CRPS can spontaneously resolve even years after initial onset. I tell all patients with it that although I cannot say with any degree of certainty what the outcome will be, there is always the chance it can improve and not to give up hope.
 
1. Harden RN, Oaklander A, Burton AW, et al. Complex regional pain syndrome: practical diagnostic and treatment guidelines, 4th edition. Pain Med. 2013;14:180-229. 

Click here for Part I, Understanding Complex Regional Pain Syndrome