Approach

The main goal of treatment is to reduce pain and increase limb function and mobility. Improvement in associated abnormalities will follow if pain reduction and increased mobility are successful. Unfortunately, uncertainties in diagnosis, epidemiology, pathophysiology, and natural history have confounded treatment analysis. There are many treatments offered but little robust evidence to support their use.​[36]​​[51]​​[52]​​

Many of the pharmacological agents used (e.g., antidepressants and anticonvulsants) are approved in some countries for non-pain conditions but have been found to be effective in certain neuropathic pain states such as diabetic polyneuropathy and postherpetic neuralgia. Therefore, they may be useful in CRPS. All therapies are empirical, and treatment should be individualised.

CRPS: early and late

The concept of staging CRPS is no longer used, but treatments and therapies offered to patients in the earlier stages and patients with established chronic CRPS may differ, as may responses to these interventions. For example, the condition of patients with early CRPS may improve naturally, so drugs or nerve blocks that are effective or can be used safely only for a few weeks or months may be appropriate for this group, but not for patients with chronic CRPS.[51]​​ There is no definitive marker between these two groups, but patients with symptoms of <6 months' duration may be considered to have early CRPS compared with those who have had >6 months of symptoms. There is usually an inciting event, when symptoms start, that the patient can pinpoint. Late CRPS is more likely to exhibit physical signs such as smooth, shiny skin or radiographical evidence of osteopenia (which takes time to develop).

Physiotherapy and rehabilitation

The occupational therapist initially assesses the impact of CRPS on activities of daily living, as well as the degree of oedema, sensory disturbance, and range of motion.[36]​​[51]​​​​​ The aims of therapy are to reduce oedema, decrease muscle guarding, and improve function in daily activities. This may involve bandaging, desensitisation techniques, and active and passive movement exercises. Desensitisation techniques are varied but may include repetitive graded stimulation of the area with soft fabrics, which is repeated and built up to harsher textures. Stress loading encourages active movement and should be encouraged as early as possible. This can be adapted to individual capabilities and can progress to include the carrying of small weights (for the upper limb) and increased weight-bearing on the affected side when walking (for the lower limb).

The effectiveness of various rehabilitation techniques is uncertain. One systematic review found no evidence to support some of the frequently used physiotherapy recommendations for CRPS.[53] However, other reviews report some evidence for efficacy, while acknowledging that the data on physiotherapy/rehabilitation techniques are limited and individual responses vary.​​[51]​​[54][55]​​​

Functional therapy

Subsequent functional therapy concentrates on improving the range of motion and functional dexterity required for activities of daily living.[36]​​[51]​​​​[56] The time frame is extremely varied, and the patient may require intensive sessions 3 to 5 times a week initially. Interventional approaches, such as repeated sympathetic nerve blocks, pharmacological therapy, and pain psychology may be utilised to facilitate patient participation in physiotherapy programmes.[51]​​​​[57]

Rehabilitation training

Graded motor imagery (GMI) and mirror visual feedback (MFV) therapy are two broadly similar types of rehabilitation training that consist of phased limb laterality recognition, imagined movement, and mirror movement phases over a period of time.

GMI and MFV therapy reduced pain and increased movement in the affected limb in small numbers of patients, mostly those with upper limb involvement.[51]​​[55]​​ Although the mechanisms behind this therapy are unknown and the quality of the evidence is low, larger-scale, high-quality randomised controlled trials are warranted.[55]​​[58][59]

Psychological therapies

Studies have suggested that patients with CRPS do not differ psychologically from other patients with chronic pain.[60] There is evidence that psychological therapies may be useful as part of a comprehensive interdisciplinary treatment programme.[16][51][60]

Psychosocial and behavioural management plays an important role in a holistic multi-modal approach to the management of CRPS. The goals are to identify psychological factors perpetuating pain and disability, treat anxiety and depression, and facilitate an environment to reduce barriers and promote healthy functioning.[16][36][61]

Corticosteroids

A short course of corticosteroids may be offered to patients (without contraindications) in the early stages of CRPS, with the aim of decreasing inflammation in the affected limb.[62]​ Relative contraindications include diabetes, osteoporosis, glaucoma, cataracts, peptic ulcers, and hypertension. Reduced doses and length of treatment might be reasonably trialled to assess symptom response. Intravenous therapies have been offered, with improvement reported up to 12 months after therapy.[63] Corticosteroids have not been reported to be effective in chronic CRPS. ​

Pharmacological analgesic therapy

Analgesia will almost certainly be required, and there are several agents that can be tried. Specific evidence for CRPS is limited, but trials for neuropathic pain may be informative. There is no order of preference, and it may be necessary to try several agents and/or employ a combination in order to achieve satisfactory analgesia and allow continuation of physiotherapy. Patient considerations and preferences should guide choice of therapy.[36]​​[51]​​ Particular care should be taken when treating women of childbearing potential, and the latest data on teratogenicity should be consulted.[64][65]

  • Non-steroidal anti-inflammatory drugs (NSAIDs) are often used as first-line treatment for CRPS, especially at initial presentation. However, they remain poorly studied specifically for CRPS, with no guidance on choice of NSAID therapy. Most guidelines recommend moderate to higher doses for 2-4 weeks, at which time response of medication can be assessed. Long-term use at these doses is generally not advised. There are no specific data for selective cyclo-oxygenase-2 (COX-2) inhibitors as an alternative option for CRPS.[51]​​[62]​​

  • Bisphosphonates have shown promise in several randomised controlled trials in patients with evidence of osteopenia or osteoporosis in the affected limb.[51][62][66] Both oral and intravenous therapy has been tried, but there is no evidence for superiority of a particular regimen. Potential (rare) adverse effects of long-term use of bisphosphonates include jaw necrosis and atypical femur fractures.[67]

  • Anticonvulsants are of benefit in neuropathic pain and have been used with some success in the treatment of CRPS. Gabapentin is the most widely tested and used; pregabalin may be better tolerated but its effectiveness has not been studied in randomised controlled trials.[51]​​ Topiramate, lamotrigine, or levetiracetam may be tried if gabapentin or pregabalin are ineffective or not tolerated.[51]​​ Anticonvulsants can be trialled in either early or late CRPS.

  • Tricyclic antidepressants are effective in several neuropathic pain states.[51]​ Amitriptyline is the most widely tested, but nortriptyline and desipramine may be more effective due to their more selective noradrenergic activity, and nortriptyline has fewer adverse effects. Tricyclic antidepressants can be useful for the treatment of pain in early or late CRPS. For patients who do not tolerate or respond to, or are not candidates for, tricyclic antidepressants, serotonin-noradrenaline reuptake inhibitors (SNRIs) may be considered similar to other neuropathic pain states, although there are no specific studies for CRPS.[51]​​[68]​​

  • Alpha-adrenergic antagonists and agonists, in particular alpha antagonists (e.g., prazosin, phenoxybenzamine) and alpha-2 agonists (e.g., clonidine), are possible treatments for sympathetically mediated pain in CRPS; however, evidence is based on case series.[51]​​[54][69]

  • Topical anaesthetics (e.g., lidocaine alone or with prilocaine) can be applied to affected areas.[51]​ These may be used in combination with other pharmacological therapies and may be used in both early and late CRPS.

  • The use of opioid analgesics in neuropathic pain states, including CRPS, is controversial. Dependency behaviour is a risk, and mortality has been demonstrated with persistent use of strong opioids, with a dose-dependent relationship. Opioids should only be considered once other options have been tried, and if the expected benefits are anticipated to outweigh risks to the patient.[70] Tramadol, methadone, oxycodone, and transdermal fentanyl have been studied in neuropathic pain states. However, this does not necessarily indicate superiority over other opioids for treating refractory CRPS, and the choice of opioid and dosing should be made after careful consideration of patient comorbidities and risk factors.[51]​​[71]

  • Ketamine intravenous infusions have been used for the treatment of chronic CRPS, with systematic reviews suggesting that sub-anaesthetic doses may have low to moderate evidence of efficacy.[51]​​[54][72]​​[73][74] Ketamine is a potent N-methyl-D-aspartate (NMDA) antagonist, with potential mechanisms to alter and change neuronal plasticity and nociceptive signaling in the central nervous system at sub-anaesthetic doses. Studies have not used a standardised dosing regimen, making comparing and interpreting the data difficult. Ketamine infusions may be used after failure of multi-modal therapy, including interventional therapy, with careful consideration of patient comorbidities and risk for psychotomimetic and cardiac adverse effects, along with appropriate monitoring during and after treatment.[51]​​

  • Intranasal calcitonin has been evaluated in a limited number of studies.[51]​​[75]​​​​ Results are mixed, but it offered significant pain relief in some patients.[76] The mechanism of action is unclear, use is controversial, and it is rarely available.[51]​​ The European Medicines Agency (EMA) recommends that calcitonin is given at the lowest possible effective dose for the shortest possible time, due to a possible increased risk of cancer with long-term use.[77]

Transcutaneous electrical nerve stimulation

Although formal evidence for the effectiveness of transcutaneous electrical nerve stimulation (TENS) is lacking, clinical experience suggests a beneficial effect for some CRPS patients.[78][79]​​ At the start of TENS treatment pain may increase for a few seconds to minutes, followed by a more prolonged decrease with continued use. TENS is safe to use in combination with pharmacological therapies.

Interventional analgesic therapies

Interventional therapies are usually reserved for the treatment of chronic CRPS when other treatments have failed to provide sufficient relief to allow physiotherapy and functional rehabilitation. Several methods are available, and their use depends on local practice and individual patient preference.

Sympathetic block and sympathectomy

  • Perhaps the greatest controversy in the CRPS literature regards the efficacy of local anaesthetic sympathetic blockade at the stellate ganglion for the upper limb or at the lumbar sympathetic ganglion for the lower limb. There are significant concerns regarding the exact role of the sympathetic nervous system in CRPS, the indications for blocking sympathetic nerves, efficacy, lack of sufficient randomised controlled trials, placebo effect, and risk potential.[46][80][81][82][83]​​

  • Meta-analyses have concluded there is insufficient evidence to support the use of local anaesthetic sympathetic block for CRPS.[84][85][86]

  • Sympathetic blockade may allow short-term clinical improvement in some patients (who are then thought to have sympathetically maintained pain), permitting more active physiotherapy during that period.[47] However, a subsequent study found little benefit of thoracic sympathetic block in the short term, but significantly reduced pain at 12 months.[87] Sympathetic block using local anaesthetic may be tried as an adjunct in early CRPS.

  • If a diagnostic sympathetic block is successful, a series of 3-6 blocks is administered. If these are successful, sympathectomy may be recommended in some centres.[88] However, there is very little high-quality evidence for its efficacy; there are significant risks to the procedure, as well as the possibility of recurrent pain 6-12 months later.[86]

Intravenous regional anaesthetic blockade

  • This allows delivery of medication directly into the affected extremity. Controversy exists over the efficacy of this procedure, and various agents have been tried, such as guanethidine, bretylium, clonidine, phentolamine, or reserpine. Combination therapy may be more effective than single-agent therapy, but the evidence is insufficient.[47]

Somatic nerve blocks and epidural infusions

  • Nerve blocks with local anaesthetics, clonidine, and opioids, either alone or in combination, as a single dose or a continuous infusion, are commonly used with some success.[47] The block may include a brachial plexus block or epidural infusion. These are an option for patients with chronic CRPS that is not responding to pharmacological therapy. The main limitation to continuous infusions is the high infection rate with indwelling lines.

Implantable stimulators

  • The implantation of peripheral nerve stimulators (peripheral nerve stimulation [PNS]) and spinal cord stimulators (spinal cord stimulation [SCS]) in patients with refractory CRPS has become more common, although the exact mechanisms of action remain unclear.[89]

  • PNS is indicated when symptoms are confined to one peripheral nerve distribution or respond to selective nerve blockade/regional block. However, newer PNS systems may allow stimulation of multiple peripheral nerves utilising a single system.[89][90]

  • SCS may be considered for patients with diffuse symptoms (i.e., not confined to one anatomical region). If trial stimulation with temporary electrodes results in significant pain reduction and increased function over several days, permanent electrode placement is considered.[91] Clinical evidence from randomised controlled trials supports the use of SCS in the management of CRPS; possible complications include infection, nerve damage, lead migration and breakage, and implantable pulse generator site pain.[51][89][92][93][94][95] ​SCS is usually reserved for patients who have not obtained relief with pharmacological and less invasive interventional therapies.[51]​​[89]​​

  • Dorsal root ganglion (DRG) stimulation is another option, with some evidence that it may offer better outcomes than traditional SCS (dorsal column stimulation).[51]​​[89][96]

Use of this content is subject to our disclaimer