What is Complex Regional Pain Syndrome – in plain English

What is Complex Regional Pain Syndrome? Well, complex regional pain syndrome (CRPS) is exactly thatResearchBlogging.org

Complex: Many, many different problems have been identified in people with CRPS. These problems can be severe and usually involve many of our biological systems.  This makes the situation complex.

Regional: The symptoms of CRPS are almost always confined to a particular region of the body. The arm of the leg is the most commonly involved region. Sometimes CRPS of one arm spreads to the leg on the same side of the body. Sometimes it spreads to the opposite arm, in which it usually ‘mirrors’ the other one.

Pain: CRPS is exquisitely, severely, painful. Like the other symptoms, the pain is confined to the affected limb or limbs.  Often the skin is too tender to touch and the whole limb is too painful to move. In fact, sometimes the limb is too painful to even imagine moving [1,2].

Syndrome: A syndrome is a pattern of symptoms that often occur together.  Normally, the reason that something is called a syndrome is that we do not know what causes it, we just know that this pattern of symptoms occurs together and usually at the same time.

So, CRPS is the name given to a pattern of symptoms that often occur together.

Diagnosis of CRPS

Because CRPS is not based on a particular injury, bacteria, virus, or other identifiable cause, diagnosis of CRPS relies on the symptoms and signs (what’s the difference between signs and symptoms? – see Note 1 at the end of the article) Not all the signs and symptoms are always present. So, the official position established and endorsed by the International Association for the Study of Pain is that we should use this checklist:

Symptoms and Signs of CRPS

Symptoms and Signs of CRPS

These are the symptoms (the things the person feels) 
Continuing pain that is disproportionate to the inciting eventy/n
The skin is painful to touch - this is called allodynia
The skin feels very sensitive - this is called hyperaesthesia
The limb feels colder or hotter than the other one (presuming that the other one is not affected)
The limb changes colour for example, it goes blue, or red, or white, or mottly
The limb feels swollen
The limb seems to sweat more than the other one
The limb doesn’t move properly
The limb shakes
The limb is very weak

The finger or toe nails grow faster than those on the other hand or foot
The hair on that limb grows faster than the hair on the other limb

These are the signs (the things someone else assesses) 
Increased pain caused by a pin prick (hyperalgesia)
Pain caused by gentle stroking of the skin (allodynia)
The limb looks to be a different colour from the other one
The limb is colder or warmer than the other one
There is current oedema, or clear evidence of past oedema
The limb is more sweaty than the other one
There is reduced range of motion
There is tremor (the limb shakes when the patient tries to move it)
There is fixed dystonia (the limb is stuck in an awkward posture)
The fingernails or toenails on the affected limb are different to those on the other one
The hair on the affected limb is longer or darker than the hair on the other one


To be diagnosed with CRPS, the following criteria have to be met:

  1. The compulsory item is ticked
  2. At least one symptom in each group of symptoms is ticked.
  3. At least one sign in 2 or more of the groups of signs is ticked.

So, this means that you can not really diagnose yourself with CRPS based simply on your symptoms.  There is no added diagnostic value of: x-rays, bone scans, blood tests, sympathic function tests, quantitative sensory testing, nerve conduction tests.

How many people have CRPS?

No one really knows how many people have CRPS, but some really good research studies give us a rough idea.  For example, one really good study, undertaken in The Netherlands [3] suggests that, every year, one person in every 4000 will develop CRPS.  That doesn’t really sound like many, but it means that about 5000 Australians, 20 000 Brits and 75 000 Americans, will get diagnosed with CRPS this year.  That is how many people will get it.  How many people will actually have it in any given year? In crude terms, you can just multiply that number by 5 – 25 000 Australians, 100 000 Brits and 400 000 Americans.  Now the numbers look big. They look even bigger when you consider that treatment for each patient with chronic CRPS costs about AU$12,000 per year and total lost income due to CRPS in Australia (a small country by population standards) exceeds $1 billion [4].   So, if you are someone with CRPS, you are not, by any means, alone. If you are someone trying to decide whether to fund CRPS research, you surely should!

What causes CRPS?

We don’t know this either.  One theory suggests that something goes wrong with the inflammatory response after a minor injury so that it just winds up and up until the whole protective system is very very sensitive.  Another theory is that CRPS results from psychological weakness and another that it relates to certain personality types.  The last two theories have been proven wrong – in the largest study, these things were not related to having CRPS: pre-injury psychological profile (including ‘personality’), biomechanical factors (the specific size, shape, configuration and movement of your body parts), injury characteristics (how and what you injured), surgical approaches (what operation, if any, they did), compensation and previous illness history [3].  The first theory has neither been proven wrong, nor proven right. Both seem possible.  It is also possible that there are several potential causes of CRPS. We really don’t know.

What we do know is that the response to tissue trauma (injury) is excessive.  We know that the autonomic nervous system (the system that controls blood flow to your body) is involved, most probably by altered sensitivity to adrenaline.   We know that the immune system is involved. We know that the system that controls movement is involved.  Last, but not least, we know that the pain system is involved.

Acute CRPS seems different to chronic CRPS

In acute CRPS, temperature changes almost always involve the limb being hotter than normal and hotter than the other one, and being redder than normal.  As CRPS progresses, a cold and white limb becomes more common and people often notice that the hair or nails, or both, on that limb, are growing faster than normal and faster than the other limb.  Chronic CRPS is also associated with other evidence of dysfunction within the central nervous system, including the brain.  For example, there is some evidence that people with chronic CRPS think that their affected limb is bigger than it really is; they often feel as though the limb doesn’t belong to them anymore; the sensory system can develop errors so that one can’t be sure exactly where on the limb one has been touched unless they are looking at it; it can become more and more difficult to move, develop the shakes, or become ‘stuck’.  These are all things that are distressing but they are also things that we think can be treated with appropriate rehabilitative treatments – although the evidence is weak because it is a very recent direction in treatment.

What is the best treatment for CRPS?

In order to determine if a treatment is effective, it must be tested with the strongest experiments possible.  Although many treatments have been used for CRPS, only a small few have been tested with strong experiments.  We can categorise treatments according to the results of experiments (See Table in Note 2):

Treatments for acute CRPS

  1. Anti-inflammatory therapies: Some are good, some a probably no good and some are definitely no good.
  2. Analgesics: Some might be good (eg ketamine intravenously or gabapentin) but there are no good experiments on it; some are no good (eg local anaesthetics, anti-epileptics, serotonin reuptake inhibitors).
  3. Spinal cord stimulators are probably good, although remember that there are risks with inserting a stimulator onto your spinal cord and many need revision or relocation.
  4. TENS is no good.
  5. Activation & mobilisation (eg physical therapies) are probably good as long as they promote movement and function, not targeting pain relief.
  6. Mirror therapy is probably good.

Treatments for chronic CRPS

  1. Spinal cord stimulators are probably good although remember that there are risks with inserting a stimulator onto your spinal cord and many need revision or relocation.
  2. Graded motor imagery is good (and also see graded motor imagery for pathologic pain)
  3. Tactile discrimination training might be good but there are no good experiments on it (see also our other paper on tactile discrimination).
  4. The other things that are listed for acute CRPS are not effective for chronic CRPS.

Where can I tell my doctor or physiotherapist to look for more information on treatment guidelines?

There are official guidelines developed in The Netherlands. The guidelines are a little out of date but the best around (http://pdver.atcomputing.nl).  The RSDSA association has a bundle of resources for clinicians and patients. BodyInMind has resources for clinicians, including teaching materials and publications.

What research is being done?

The most impressive concerted research effort involving CRPS is going on in the Netherlands. It is called TREND and you can have a look here. Also, the National Institute of Neurological Disorders and Stroke (NINDS) and other institutes of the National Institutes of Health (NIH) conduct research relating to CRPS in laboratories at the NIH and also support additional research through grants to major medical institutions across the country. NINDS-supported scientists are studying new approaches to treat CRPS and intervene more aggressively after traumatic injury to lower the chances of developing the disorder.  The Australian National Health & Medical Research Council funds clinical studies into CRPS after wrist fracture.


1. Signs and Symptoms

Signs are things that someone else can observe. Symptoms are things that you feel. For example, feeling cold is a symptom.  Actually being cold is a sign.

2. Evidence that something works

CategoryWhat it means
Level 1-2 evidence of efficacyStrong experiments show that the treatment is good
Level 3 evidence of efficacyNot very strong experiments suggest the treatment might be good, but we need strong experiments to really know
Level 4 evidence Some experts think the treatment is good but there are no experiments, so we really don’t know
No evidenceNo one really knows

Evidence that something doesn’t work

CategoryWhat it means
Level 1 - 2 evidence of no effectStrong experiments show that the treatment is not better than a placebo
Level 3 - 4 evidence of no effectNot so strong experiments suggest the treatment is no better than a placebo,which means it is almost certainly no better than placebo
Level 1-4 evidence of harmStrong experiments, weak experiments or expert opinion suggest the treatment is harmful


  1. Moseley, G.L. (2004). Imagined movements cause pain and swelling in a patient with complex regional pain syndrome. Neurology 62, 1644.
  2. Moseley, G.L., Zalucki, N., Birklein, F., Marinus, J., Hilten, J.J.v., and Luomajoki, H. (2008). Thinking about movement hurts: The effect of motor imagery on pain and swelling in people with chronic arm pain. Arthritis Care & Research 59, 623-631.
  3. de Mos, M., de Bruijn, A.G.J., Huygen, F.J.P.M., Dieleman, J.P., Stricker, B.H.C., and Sturkenboom, M.C.J.M. (2007). The incidence of complex regional pain syndrome: A population-based study. Pain 129, 12-20.
  4. Kemler, M.A., and Furnee, C.A. (2002). The Impact of Chronic Pain on Life in the Household. Journal of Pain and Symptom Management 23, 433-441.

Lorimer Moseley (2009). What is Complex Regional Pain Syndrome – in plain English BodyInMind


  1. Mark Brennan says:

    Is there not a differentiation between type one and two anymore?

    Lorimer Reply:

    Yeah there is – as it was, type 2 if one can demonstrate a frank peripheral nerve lesion, type 1 if not. It is a bit dodgy though because whether one can or not depends on equipment and experience to some extent. Still, I should change that – nice pick up.

    Lorimer Reply:

    We received this question via facebook: how do you feel about stellate ganglion blocks? Here’s a response to that question:

    Check out this review:
    This is what the New Zealand Accident Compensation Corporation
    “One guideline, three systematic reviews and one experimental study provided information on the effectiveness of sympathetic ganglion nerve block for complex regional pain syndrome (CRPS).

    One high quality systematic review was identified evaluating the use of stellate and lumbar sympathetic ganglion blocks (Cepeda, 2002). It included 29 studies, three of which were randomised controlled trials. The majority of case series were retrospective and included small numbers of participants. The review included a meta-analysis of pain outcomes pooled from a sub-group of 14 studies in which the magnitude of treatment effects was quantified. Less than one third of people with CRPS (29%) obtained pain relief by three quarters of pre-treatment levels. The authors noted that the rate of success may be acceptable to many patients and practitioners but was consistent with a placebo response. Two other systematic reviews concluded that the limited amount and quality of the available evidence did not allow definite conclusions on the treatment’s effectiveness (Forounzanfar, 2002; Kingery, 1997). The guideline (Sanders, 1999) concluded that nerve blocking procedures should be limited to those patients in whom functional capacity can also be improved.

    One randomised controlled trial was identified which was not included in the above Cepeda review (Tran, 2000). Seventeen people with CPRS I or II received a lumbar sympathetic nerve block using bupivacaine with iohexol (a contrast dye) or one using bupivacaine with saline. Both types of blocks resulted in significant pain relief from baseline to 1 hour post-block, and for six (iohexol / bupivacaine group) or three (saline / bupivacaine group) days afterwards.”

    Mark Brennan Reply:

    Cheers, I have a guy at the mo who is awaiting nerve conduction tests to determine ulnar nerve function and although he does not fit the distinct criteria for CRPS I feel he is on a continuum. Could there be elements of peripheral and central changes though therefore elements of type 1 and 2 and therefore graded Motor imagery and mirror box may help and could be part of overall management?. I feel its worth trying.

  2. jenny crane says:

    Hi Lorimer
    After reading your blog and comment re TENS and CRPS I wanted to respond as I have seen TENS like stimulation be affective when incorporated into the overall management plan for people with CRPS. Our company distributes the Interx device which we often descrive as “TENS” like, even though the application is very different.
    It may well be considered that TENS is not indicated for the treatment of pain relating to the condition CRPS due the lack of long term benefit or inconsistency of results that are commonly experienced. However, it may be necessary to look further into the way the technology is applied to determine whether or not it has a place in the treatment of such difficult conditions. TENS is a broad category which incorporates many different technologies applied in a variety of ways . The way a modality is used can have a significant affect on the outcomes achieved.

    Recent analysis of the effect that treatment parameters of TENS have on pain relief clearly demonstrates that it is possible to optimize those parameters to achieve more pain relief. The importance of treatment amplitude, pulse frequency and treatment location have all been shown in recent literature and even in articles dated as far back as 1975 (Melzack et al) .

    Based upon my experience of the benefit of InterX therapy and the successful treatment of very difficult conditions including CRPS, Allodynia, Phantom Pain and others , I suggest that the technology of TENS which has barely evolved in the last 30 yrs limits the ability of a therapist to optimize treatment parameters for the effective relief of pain from CRPS. The protocols used with InterX dictate both an extra segmental and segmental approach for the treatment of CRPS. This includes contralateral treatment, treatment of the whole spine and the face and scalp. The crucial technological aspect of InterX is that it allows the identification of optimal treatment points within chosen treatment areas while also delivering a significant dose of stimulation (typically 45mA). The identification of treatment points is based upon patients’ physiological response (increased sympathetic skin response) , and not arbitrary or anatomical points. The dose is an important factor and the InterX amplitude is approximately three times higher than Bjordal et al considered high amplitude and thus important for the success of a treatment.

    If this scientific application of a modality excludes InterX from the classification of TENS then so be it, but it does not fall into the category of electroacupuncture either as points of low impedance are not always acupuncture points. As such InterX (Interactive Neurostimulation) may be considered in any of the above categories or in one of its own as a scientific delivery of high amplitude stimulation to points relating to a patients sympathetic nerve activity. Either way, it should be noted that even if “TENS” is not indicated for the treatment of CRPS, it does not mean that TENS-like technologies when scientifically and correctly applied cannot offer sufferers of CRPS significant relief of their pain. This relief can bring them back into functional rehabilitation and ultimately to achieve long term relief with the aid of an integrated approach. With spinal cord stimulation considered the next best option, surely a non-invasive, evidence based approach to transcutaneous neurostimulation should be offered first.

    Lorimer Reply:

    Hi Jenny –
    Thanks a million for taking the time to contribute! It is great to hear of your successes relating to the management of CRPS. I am not, obviously, very familiar with InterX, aside from our very brief conversations about it – there is clearly much to know! I am, equally obviously, very keen to pursue better ways to treat people with CRPS and if InterX ends up being excellent than that would be excellent. I guess, however, that we need the evidence. This is not trivial because evidence is tricky to collect, but here is what I say to anyone who sees real successes in their clinical interactions: “If it really is as good as you think it is, then we all should know about it. If it really isn’t as good as you think it is, then you should know about it”. So, let’s find out – I would be happy to help here if I can. In the meantime, we rely on what evidence there is and the evidence we currently have for TENS as a way to treat CRPS, is not good – remembering CRPS is not ‘just another chronic pain’. We can also utilise what we know about the mechanisms that underpin CRPS and here I am a little lost. I don’t know how it could work. This reflects my ignorance perhaps, but I am, as I said, very keen to pursue the truth, wherever it may lie. How do you reckon it would work in CRPS? Again, thanks for bothering to contribute – we appreciate it. Feel free to pose an answer to that one won’t you?