top of page
  • robinapark

CRPS - Need to know (2024 update)

Updated: May 9


Definitions/Diagnostic criteria

(1) Pain is Out of proportion: CRPS describes an array of painful conditions that are characterized by a continuing (spontaneous and/or evoked) regional pain that is seemingly disproportionate in time or degree to the usual course of any known trauma or other lesion. There MUST be pain for the formal diagnosis.

(2) Variable location and presentiation: The pain is regional (not only defined in a specific nerve territory or dermatome - though at times there may be evidence of a precipitating injury to an underlying nerve (type 2 CRPS - see below)) and usually has a distal predominance of abnormal sensory, motor, sudomotor, vasomotor, and/or trophic findings. The syndrome shows variable progression over time (3)

(3) Four main features to find: Vasomotor, Sudomotor, Motor/Trophic and Sensory.

(3 on history and 2 on examination)


CRPS type 1 = No clear lesion (90%) Memory tip: (the most common comes first)

CRPS type 2 = Clear peripheral nerve lesion (10%)

Acute (<1 year) or Chronic (> 1 yr)

Budapest Criteria:

Grading severity



- Most common: Women aged between 40 and 60

Interesting caveat though... The female preponderance could have an artefact component because women suffer three times more radial fractures than men and this is a common precipitant event.

More severe injury = more risk:

The risk of CRPS seems to be higher for patients with complicated fractures, a rheumatological disease, or intense pain >1 week after trauma (possibly with higher opioid requirement). Incidence varies between 5 and 25 cases per 100,000 people per year.

Genetics? Maybe...

It is heavily debated whether there is a genetic disposition for CRPS. There are “CRPS families” with striking associations to migraine associations.


  • They recover better but have more relapses.

  • They have more of a warm phase change and less of the atrophy changes ongoing.

  • More dystonia and movement disorders.

  • Self harm is higher in this group.

  • Tend to be overachievers.

  • More lower limb in kids (75%) (sprained ankles), adults is older women (more likely wrist fractures).

Social factors?

  • The research on social factors in CRPS is in its infancy. Surprisingly, it is the affluent patients, who develop CRPS more often after distal radius fracture

Around forever?

CRPS changes after somewhat innocuous injury have been recorded even by the physician Galen in 210 AD!

Approximately 15% of sufferers will have unrelenting pain and physical impairment years after CRPS onset and are considered to have a long-term condition, although more patients will have a lesser degree of ongoing pain and dysfunction


Risk factors

  • Demographics: Female gender (3-4x more likely), especially if postmenopausal. Older age (30-70)

  • Biomedical: Migraine headache association, Distal radius fracture, ankle dislocation, or intra-articular fracture, Anesthesia lasting ≥ 120 minutes or extended use of a tourniquet during limb surgery.

  • Higher than usual pain is reported during the early phase of trauma or following removal of a cast.

  • Immobilization.

  • Mitochondrial genetic disorders

  • ACE inhibitor use (4)

  • Social: Affluence (maybe)

(Obesity and diabetes REDUCES the risk)



  • Genetic predisposition (CRPS families)

  • Excessive inflammatory response likely driven by innate and adaptive immune system overactivity

  • Decreased sympathetic activation causing warm phase

  • Sympathetic dysfunction - sympatho-afferent coupling (sympathetically maintained pain with catecholamines)

  • Endothelial dysfunction with NO and endothelin-1

  • Neuronal plasticity in the CNS - learned non-use

The suspected story...

A posttraumatic inflammatory reaction is driven inappropriately by activation of the innate and adaptive immune system. Without intervention this leads to central nociceptive sensitization, cortical reorganisation, and implicit maladptive behavioural processes such as rigidity, stiffness and disuse.

The first step involving all this uncontrolled inflammation and immune response generally leads to the "warm” CRPS common acute phase

The uncontrolled immune processes are interesting. Cytokines in the area possibly cause keratinocyte proliferation inappropriately and further inflammatory soup creation.

The activated keratinocytes lay down connective tissue leading to contractures and local osteoblasts and osteoclasts cause local bone resorption.

The typical CGRP and Substance P are released locally as part of the inflammatory soup. These then sensitise local sympathetic afferents to respond inappropriately to adrenal hormones and also drives the pain sensitisation of local pain fibres. Subtance P particularly is known to promote hair growth and the CGRP enhances sweating.

We have found auto-antibodies within the area against adrenergic and cholinergic receptors making a auto-immune component of this presentation likely.

The question is - what happens in cold CRPS? Inflammation is less obvious in primarily cold CRPS, and investigations specifically for this subtype are sparse. Increased endothelin 1 and reduced nitric oxide probably contribute to the cold bluish skin.

Next steps after the initial processes have had their 'play'...

Nneuronal plasticity in the CNS plays a significant role in the chronicity of CRPS. It becomes increasingly difficult to modulate/treat after 6-12 months likely as these 'learning' processes are hard to 'unlearn' and behaviours become instinctual/habitual.

Another possibility is a reflex inhibition of movement mediated by the expectation of pain.89 This results in a pathological movement pattern (eg, while walking), which again increases the pain through eg, unphysiological muscle and joint loads.2

The perception of allodynia is a consequence of central (spinal) sensitization. Its presence has been verified through functional magnetic resonance imaging by activation of the “pain matrix” through painful touching of the affected but not by nonpainful touching of the unaffected hand

The hypothesis of “sympathetically maintained pain” is similar: nociceptors in the affected limb become sensitive to catecholamines.78 The presumably sympathetic symptoms were the motivations for the use of sympathetic blocks to treat CRPS

Depression and anxiety, which were assessed by self-reports, are not related to the development of CRPS

Surgery should be avoided on a CRPS-affected limb where possible, and be deferred where it cannot be avoided until 1 year after the active process has resolved. (2)

There is insufficient robust literature to predict outcome from amputation, with just two small series publishing comprehensive results: Deliessen et al (1995) reported disappointing results in their series of 28 patients, with recurrent symptoms at some level in all cases, and only two patients successfully wearing a prosthesis although the majority (24/28) reported overall satisfaction with the results.81(2)

The role of the sympathetic nervous system in CRPS is unclear; however, autonomic manifestations previously ascribed to sympathetic overactivity could be due to catecholamine hypersensitivity (3)

CRPS is more common in women, with a female-to-male ratio of 2:1 to 4:1 [22-26]. The incidence appears to be highest in postmenopausal women [23]. (3)

The most common inciting events leading to CRPS are fractures, crush injuries, sprains, and surgery. However, no precipitating factors are identified in up to 10 percent of patients (3)

CRPS in adults more commonly occurs in the upper limbs (3)


How do I now explain how CRPS occurs to patients?

I honestly find that patients get very confused if I talk about fibroblast activations and immunoinflammatory processes.

A simpler way I now explain CRPS is I say to patients:

"Usually what happens is when you hit your arm and get a bruise, it becomes red, swollen, hot and sore. This is normal, as your body brings in inflammation to heal your tissues. Now what should happen, is that once the tissues start to get healed, that inflammation starts to go away again. It becomes less red, less swollen and less sore.

But, in CRPS we think this 'off' switch doesn't happen properly. So instead of all that swelling and redness going away, it gets more and more and more even though the initial tissue injury might be pretty much healed. So the area becomes redder, hotter, more swollen, and the pain receptors continue to have the volume knob turned up to 12 out of 10. Now, because of that pain, the brain often doesn't want to move the body part anymore! Like if you put your hand in a fire, your body and brain will try to stop you doing that again. If you even think about putting your hand in the fire again, your heart rate may go up, you may become sweaty, and even feel a bit sick.

The same thing happens a little bit when you try to move your CRPS body part. So the muscles don't work as well as they used to and you can become weaker. Eventually, your brain doesn't like worrying about this body part all the time and starts to just block normal control signals for that area and some people can even forget their CRPS body part is even there!

So we need to work together to reduce down this significant inflammation and see if we can start to get this process turning off! We will work with you to do this by.... Etc."


Presentation (Birklein & Dirmova, 2017; Dynamed 2021)

The diagnosis is primarily clinical and usually occurs in the extremities

CRPS can ONLY be diagnosed after a normal healing time has elapsed

Pain is the most important symptom.

Pain is permanent or fluctuating and most often feels like it is in the deep tissue.

Positive symptoms include a burning (key symptom), stinging, or tearing sensation.

Negative symptoms include hypoesthesia and impairment of thermal perception.

Patients often report that their extremity no longer belongs to their body.

Allodynia is more severe in more activated cases. Pain is often worsened by movement. contact, stress, and changes in temperature at night.

Spontaneous CRPS is very rare but possible (e.g. no trauma whatsoever)

Repeated measurements of the skin temperature show dynamics, ie, changing temperature differences (warmer gets colder or vice versa) of >1°C.

There may be two subtypes, though this is debated - Hot and Cold.

  • ~70% of patients initially report a “primarily warm” subtype with increased skin temperature, erythema and swelling.

  • Remaining 30% report a “primarily cold” subtype

Two-thirds of patients with CRPS have some decreased muscle strength but also functional motor impairments related to pain. This is important as many people who experience pain have pain-induced muscle dysfunction, but this movement is often significantly weaker and prominent.

Pain-related immobility can lead to early contractures. Unfortunately, the pain may reduce and strength return, but if contractures occur, they may persist. Contractures occur more commonly in poorly treated CRPS but can also occur despite treatment attempts.

Other trophic changes can be found on the skin (e.g. ulcers), the nails, and the hairs (in acute CRPS increased and in chronic CRPS decreased growth). The main symptom of vascular dysfunction is the oedema which can grow to dramatic extents and is always found in the acute phase.

In chronic CRPS, patients report that they feel that their own extremity is thicker than it actually is. Fifty percent of the patients have sudomotor disturbances, mostly hyperhidrosis. All patients display a change in skin colour from reddish (“warm” CRPS) to blueish livid (“cold” CRPS) eventually. Skin temperature is different when comparing both sides. Rarer symptoms such as movement disturbance are possible including tremor, myoclonus, or fixed dystonia.

These symptoms spread distally and are not limited to innervation territories.

If CRPS is not improved in the acute phase and becomes chronic, the visible symptoms change throughout because of the changing pathophysiology; the pain, however, remains.

The main symptoms then include movement disorders, alternating skin temperature, sensory loss, hyperalgesia, and body perception disturbances. Psychological factors such as posttraumatic stress or pain-related fear may impact the course and the treatability of CRPS

For those in whom pain persists, psychological symptoms (anxiety, depression), and loss of sleep are likely to develop, even if they are not prominent at the outset.(2)




Differential diagnosis

  • Vascular insufficiency / clots (More swelling and likely acute onset -no clear trauma)

  • Cellulitis (Systemic features of infection more common. Localised)

  • Metalwear complications (Multiple and possible - clearly needs metalwear!)

  • Conversion disorder (Unlikely to give clear inflammatory signs)

  • Compressive neuropathy (Tinels sign, discrete nerve)

  • Peripheral nerve trauma (As above)

  • Demyelination / MS (Usually not overt local findings - more central issues)

  • Erythromelalgia (Much more erythema and heat and less the other findings)

  • Psychogenic purpura (Psychological history ++)



Specific images are rarely required but are used to rule out other differential diagnoses if at times of uncertainty.

Sensitivity of Nuclear Bone Scans for CRPS is approximately 80% and Specificity is 85%. Though studies are quite limited.

Limb magnetic resonance imaging helps to exclude differential diagnoses like rheumatic diseases or infections;

X-rays in direct side-to-side comparison are not sensitive but can prove a patchy osteoporosis or may help to make differential diagnoses such as pseudoarthrosis after fracture;


Treatment (Updated in May 2024)

Rob's summary (what may I do?)

  • Maximise all non-pharm managements (goes without saying...)

  • Pharmacological

    • Paracetamol and COX-2 NSAID use seems logical and low-risk

    • Prednisone if seen in the early phase. 30 mg for 3 days and then down by 5 mg every three days until ceased (seems safest on the balance of things but still reasonable evidence of benefit for some patients)

    • Oral Alendronate course for eight weeks - 35-40 mg daily for 8 weeks

    • Gabapentin - May consider using this as an adjunct if the pain remains severe

    • I may use intravenous ketamine if there are no significant psychological concerns and they have psychological support.

    • Opioids - used very sparingly - ideally sublingual buprenorphine PRN or possibly Tapentadol IR 50 mg PRN.

Iolascon, G., Snichelotto, F., & Moretti, A. (2024). An update on the pharmacotherapeutic options for complex regional pain syndrome. Expert Review of Neurotherapeutics, 24(2), 177–190.

Ferraro, M. C., Cashin, A. G., Wand, B. M., Smart, K. M., Berryman, C., Marston, L., ... & O'Connell, N. E. (2023). Interventions for treating pain and disability in adults with complex regional pain syndrome‐an overview of systematic reviews. Cochrane Database of Systematic Reviews, (6).)

Her, Y. F., Kubrova, E., Dombovy-Johnson, M., ElSaban, M., Mostert, K., & D’Souza, R. S. (2024). Complex Regional Pain Syndrome: Updates and Current Evidence. Current Physical Medicine and Rehabilitation Reports, 1-21.




NB: The correct exam answer is that there is no clear evidence of benefit for rehabilitation medicine and psychological therapies (But I personally would take this with a grain of salt - same for almost all pain conditions - just hard to make it statistically significant)

Physical and occupational therapies reduce pathologic movement patterns and movement limitations and train physiologic use of the extremities. However, according to Cochrane, the actual evidence of benefit remains elusive.

Smart KM, Ferraro MC, Wand BM, O'Connell NE. Physiotherapy for pain and disability in adults with complex regional pain syndrome (CRPS) types I and II. Cochrane Database Syst Rev. 2022 May 17;5(5):CD010853. doi: 10.1002/14651858.CD010853.pub3. PMID: 35579382; PMCID: PMC9112661.

GMI and Mirror therapy works best with acute CRPS and CRPS after stroke (small RCT, only for poststroke CRPS). There appears to be a small statistical signal; however, it has not been universally reproducible, so clear evidence of its benefit over other options remains uncertain. There is very little risk of harm, though, so I would recommend it at this time (2024).

Shafiee E, MacDermid J, Packham T, Grewal R, Farzad M, Bobos P, Walton D. Rehabilitation Interventions for Complex Regional Pain Syndrome: An Overview of Systematic Reviews. Clin J Pain. 2023 Sep 1;39(9):473-483. doi: 10.1097/AJP.0000000000001133. PMID: 37224001.

Graded Exposure” (GEXP) treatment has shown good evidence for efficacy in CRPS. For this approach, a psychologist identifies and classifies fear-triggering situations (eg, pain induction through certain movements and situations). Patients are then gradually exposed to these situations by a physiotherapist. There is still a widespread misconception that patients with CRPS should avoid pain to prevent aggravation; this is not valid. Contractures follow quickly if the extremity is not moved during the inflammatory phase when a proliferation of connective tissue cells occurs.



Gabapentin: A very mild effect on allodynia was shown for gabapentin (RCT; secondary endpoint). It is justified to assume that this might also be valid for pregabalin, but the evidence is poor. The only real RCT I could find was from 2004, and the benefit was mixed. A head-to-head trial in 2016 showed similar benefits between amitriptyline and gabapentin in children.

Brown S, Johnston B, Amaria K, et al. A randomised controlled trial of amitriptyline versus gabapentin for complex regional pain syndrome type I and neuropathic pain in children. Scand J Pain. 2016;13:156–63.

Amitriptyline: Sedative tricyclic antidepressants should be used in particular if sleeping problems prevail. Although there are no controlled studies, analgesic drugs according to the World Health Organization analgesic ladder can be tested, especially in very acute phases(1).


Opioid-insensitive pain might be frequent because of decreased central opioid receptor availability in CRPS. (1) Efficacy and safety of opioids in CRPS is questionable, and long-term opioid use might even worsen troublesome symptoms, such as allodynia and hyperalgesia so their use remains controversial and unclear. Only one RCT has been conducted studying controlled-release morphine on pain reduction in CRPS and found that there was no difference between the opioid and placebo groups.

Shermon S, Fazio KM, Shim R, Abd-Elsayed A, Kim CH. Prescription Trends in Complex Regional Pain Syndrome: A Retrospective Case-Control Study. Brain Sci. 2023 Jun 30;13(7):1012. doi: 10.3390/brainsci13071012. PMID: 37508943; PMCID: PMC10377480.

Corticosteroids: Supporting evidence for using oral glucocorticoids in CRPS is very weak but there are suggestions of reduced pain particularly early on. Anecdotally Stanford professors use 30 mg daily and decreasing by 1 tablet daily every three days until ceased. It has been stated that the possible risk of side effects and long-term toxicities outweigh the potential benefits, but this remains unclear.

Rob's take: I was initially concerned about prednisone in high doses precipitating mania and possible steroid-induced psychosis. It seems that the symptoms are really rare, below 40 mg per day (see below). It also seems that weaning and ceasing the prednisone tends to resolve the symptoms, so I'm less worried about it now.

The Boston Collaborative Drug Surveillance Program monitored 676 consecutive hospitalised patients who received prednisone therapy and recorded a 1.3% (6/463) incidence of psychiatric disturbances in patients receiving 40 mg/d or less, a 4.6% (8/175) incidence in patients receiving 41 to 80 mg/d, and an 18.4% (7/38) incidence in patients receiving more than 80 mg/d.

Huynh G, Reinert JP. Pharmacological Management of Steroid-Induced Psychosis: A Review of Patient Cases. J Pharm Technol. 2021 Apr;37(2):120-126. doi: 10.1177/8755122520978534. Epub 2020 Dec 2. PMID: 34752563; PMCID: PMC7953074.

Atalay NS, Ercidogan O, Akkaya N, Sahin F. Prednisolone in complex regional pain syndrome. Pain Physician. 2014 Mar-Apr;17(2):179-85. PMID: 24658479.

Bisphosphonates: Cochrane review in 2023 suggests there is likely some benefit in reducing pain intensity after treatment but are often associated with some side effects. It seems unclear whether IV or oral is better and exactly which agent. There is reports of an IV Zoledronic acid trial that was ceased but it was unclear why. Some studies have shown oral alendronate given over 8 weeks showed benefit in pain, pressure tolerance, and joint mobility.

The main concern is osteonecrosis of the jaw. Looking through the stats, this is far less likely with oral therapy than an infusion. It appears that 1 in 100,000 patient years is a reasonable estimate of the incidence of osteonecrosis of the jaw in patients receiving oral nitrogen-containing bisphosphonates for osteoporosis. The risk of developing ONJ for patients taking alendronate, the most commonly prescribed oral bisphosphonate, has been estimated to occur in approximately 0.7 per 100,000 persons per years’ exposure 38; on the other hand, the incidence of ONJ for risedronate and ibandronate cannot yet be quantified because too few cases have been reported (12 cases for risedronate and one for ibandronate).

Masoodi, N. A. (2009). Oral bisphosphonates and the risk for osteonecrosis of the jaw. British Journal of Medical Practitioners, 2(2).

Vitamin C IS NOT A TREATMENT but rather can be used prophylactically from limb fractures to reduce the incidence of CRPS. It can be used in patients at higher risk (e.g. previous CRPS). It is reasonable to supplement for six to eight weeks after distal radius fractures in patients with poor baseline nutritional status or those who cannot comply with whole-food intake. A typical dose is 500 mg daily.

Topical Agents: Three RCTs have looked at 64 patients comparing EMLA cream (local anaesthetic), ketamine, and fatty cream with dimethylsulfoxide (DMSO) to placebo. After two months, DMSO showed mild improvement and the others were non-signficant.

There might be a reduction in pain for up to 3 months after intravenous ketamine (continuous infusion for 4 days; maximum 30 mg/h for a 70-kg patient).



Spinal cord stimulation (SCS) seems to be an alternative to treat CRPS pain but not function in the lower extremity for up to 5 years (RCT, no active control). (1)

Interventional procedures for the treatment of pain related to CRPS include trigger/tender point injections, regional sympathetic nerve block, spinal cord stimulation, epidural clonidine, and chemical or mechanical sympathectomy, among others (3)

The outcome of the DRG stimulation regarding pain reduction and quality of life was superior to SCS.(1)

Pharmacological treatment maybe particularly effective in acute stages and includes steroids, bisphosphonates, and dimethylsulfoxide cream. Common anti-neuropathic pain drugs can be recommended empirically. Intravenous long-term ketamine administration has shown efficacy in randomised controlled trials, but its repeated application is demanding and has side effects. Important components of the treatment include physio- and occupational therapy including behavioural therapy (eg, graded exposure in vivo and graded motor imaging). If psychosocial comorbidities exist, patients should be appropriately treated and supported. Invasive methods should only be used in specialised centres and in carefully evaluated cases (1)

he limited evidence base for sympathetic block as a treatment for CRPS is conflicting. Small randomized trials comparing nerve block with sham/placebo or other active comparators have failed to show a difference in short-term pain reduction [59], but nonrandomized studies have suggested benefit in selected patients [60]. Despite weak supporting evidence in the literature, it is the author's experience, and that of many interventional pain clinicians, that the aforementioned procedures could be beneficial for many patients and life changing for some. (3)



Within the first year, 70% improved, especially in the function of the extremity and the visible symptoms (edema, skin color, and sweating). However, 25% of the patients still fulfilled the Budapest Criteria and only 5% were without complaints.

High levels of anxiety worsens the prognosis (1)

Management should include reassurance that the pain will resolve, either completely or partially, in up to 80% of cases, although ongoing lesser pain and motor dysfunction with limited limb disability may be more common. (2)

Patients are often scared that their CRPS spreads. These patients can be reassured that true CRPS spread is rather rare, occurring in about 7% of all cases, although transient pains in other limbs may be more common. (2)

Recurrence of CRPS is not uncommon; estimates of recurrence range from approximately 10 to 30 percent, with the higher rates occurring in younger patients, including children (3)

In a study of 1183 consecutive patients with CRPS, recurrences were seen in 10 percent of patients (3)

Early mobilization after limb injury may also reduce the risk of CRPS, though there are no high-quality data to confirm benefit.(3)


Golden pearls




References / Articles / Resources

  1. Birklein, F., & Dimova, V. (2017). Complex regional pain syndrome–up-to-date. Pain reports, 2(6).

2. UpToDate - Accessed March 2021

3. Dynamed - Accessed March 2021

4. de Mos, M., Huygen, F. J. P. M., Stricker, B. C., Dieleman, J. P., & Sturkenboom, M. C. J. M. (2009). The association between ACE inhibitors and the complex regional pain syndrome: Suggestions for a neuro-inflammatory pathogenesis of CRPS. PAIN®, 142(3), 218-224.,BK%20and%20SP%20%5B26%5D.

129 views0 comments

Recent Posts

See All


bottom of page