For years there have been two schools concerning the pathophysiology of CRPS 1. One
school considered CRPS, to be a dysregulation of the sympathetic nervous system, the other
considered CRPS, to be a chronic infl ammation.
One of the problems in studying CRPS is the lack of a gold standard. For clinical use it is
important to work with a standardized set of criteria. The criteria used nowadays, such as
the Veldman criteria 1, the IASP criteria 2 and the Bruehl criteria 3 are not always easy to
apply. Studies on CRPS should focus on developing markers to make the diagnosis in a
more sensitive and specifi c way, and to follow the development of the disease or the effect
Temperature is one of the diagnostic tools used in CRPS. Temperature at the surface of
an extremity is mainly the result of skin blood fl ow and this is the result of a complex
combination of central and local regulation systems. The tympano thermometer is nowadays
the most popular device to monitor temperature in research on CRPS.4 It is cheap, easy to
use, non invasive and reliable. We have however demonstrated the limitations of the use of
this device in CRPS. Other investigators have studied the usefulness of videothermography
as a diagnostic tool for CRPS.5 With videothermography we are able to see differences
in local temperature, but this is a qualitative and mainly subjective interpretation. A
videothermographic image is build up of thousands of pixels. Calculation of the mean of
the measured values of these pixels results in a loss of important information by regression
to the mean. Therefore we proposed a new calculation method which quantifi es the
difference in what we call an asymmetry factor.6 Compared to the use of the tympano
thermometer for diagnosis of CRPS, there is a clear increase in sensitivity and specifi city.
Further research is needed to evaluate the validity of this method. It would be interesting,
to see whether a temperature stress test with videothermograpy, as performed by Wasner et
al. with a tympanothermometer, 7 can further improve the sensitivity and specifi city.
In the clinical picture of CRPS rubor, calor, dolor and functio laesa are an indication that at
least in part an infl ammatory process is playing a role in the pathophysiology of CRPS.
Paul Sudeck, a German surgeon, was the fi rst to hypothyse on the pivoting role of an
infl ammatory process.8 Jan Goris, a Dutch trauma surgeon, supported this notion by
reporting many studies focussing on a prolonged infl ammation.9 Veldman and colleagues
described the signs and symptoms of 829 patients with a relatively short-lasting CRPS.
This group described the clinical picture of regional infl ammation, which increases after
muscle exercise.1 Heerschap et al. investigated the lower leg skeletal muscles of 11 patients
with CRPS at rest with 31P nuclear resonance spectroscopy. An impairment of the high-
energy phosphate metabolism was found. It was suggested that this may be caused by
cellular hypoxia or diminished oxygen utilisation. This could be the result of a reduced
oxygen extraction in CRPS extremities, a phenomenon which is classically detectable in
areas of infl ammation.10 A scintigraphic study of van Oyen et al. demonstrated that there is
a vascular leakage of macromolecules in the acute phase of CRPS 1, which also suggests
an infl ammatory process.11
Birklein et al. showed that neuropeptides are increased in plasma of patients with CRPS.12
Leis et al. showed that administration of neuropeptides in CRPS extremities results in an
exacerbation of the disease.13
In contrast, in several other studies no indication of an infl ammatory immune response
Ribbers et al. could not fi nd any differences between lymphocytes and activated T cells in
blood from patients with CRPS 1 and healthy controls.14
Van de Beek et al. analysed blood plasma levels of the cytokines interleukin-1β, interleukin-
6, interleukin-8, interleukin-10 and TNFα and found no differences between patients with
CRPS and healthy controls.15
If an infl ammatory immuno-respons plays a pivoting role, the substances, (e.g. neuropeptides,
cytokines and eicosanoids) which attribute to the process should be detected in increased
amounts in the involved extremity. Secondly, there has to be a worsening of the signs and
symptoms of CRPS after administration of these substances to patients with CRPS. Finally,
the therapeutic activity of specifi c antagonists or synthesis inhibitors for these substances
has to be proven.16
In other infl ammatory processes, e.g. rheumatoid arthritis, rhinitis and psoriasis, the
substances which play a role in infl ammation are locally detectable only. Therefore we
focussed on a method whereby we could sample material from the involved area. Especially
in dermato-immunologic research, the induction of artifi cial suction blisters is a well known
and reliable method to harvest samples of interstitial fl uid.17
In a group of 9 patients with a relatively short duration of CRPS we sampled venous
blood and fl uid from artifi cial suction blisters in the involved and uninvolved extremity in
order to measure the cytokines IL-6, IL-1β and TNFα, the neuropeptides NPY and CGRP,
and prostaglandin E2. In plasma no differences in these mediators of infl ammation were
observed. In the blister fl uid of the involved extremity, however, we found a signifi cantly
higher level of IL-6 and TNFα.18 This was the fi rst time that the involvement of mediators
in CRPS 1 has been demonstrated directly. Later we confi rmed this observation in two
additional studies in larger groups of patients.19,20
Cytokines are produced by many different types of cells such as T-lymphocytes, monocytes,
macrophages and skin resident cells (like mast cells). Mast cell activity can be monitored
relatively easy with tryptase measurement.21 In 20 CRPS patients we induced artifi cial
blisters and detected signifi cantly higher levels of tryptase in the blister fl uid of the involved
extremity. This is direct evidence for involvement of mast cells in CRPS 1. A limitation of this
study is that we did not include a control group with neuropathic pain only. Consequently,
doubt remains as to whether the higher tryptase levels are uniquely characteristic for CRPS
1, or just for a neuropathic injury. Further investigations in patients with neuropathic pain
only have to be conducted before defi nite conclusions can be drawn. In this study there
was a signifi cant correlation between IL-6 and TNFα levels found in blister fl uid of the
involved extremity, but no signifi cant correlation between tryptase and IL-6 and tryptase and
TNFα. This confi rms our hypothesis that also other cells than mast cells must be involved.
More studies are needed to detect the involvement of other cells. A signifi cant correlation
between the Visual Analogue Scale and tryptase levels in the involved extremity was found.
Based on these fi ndings there is a possible place for pharmacological intervention studies
in patients with CRPS 1 with mast cell specifi c anti-allergic drugs, such as anti-histaminic
and mast cell stabilizing drugs.19
If TNFα plays an important role in CRPS, anti-TNF could have a positive effect on the
infl ammatory aspects of CRPS 1. We have started an open label study which includes 8
patients. The fi ndings in two patients are very promising.22 Patient 1, who has had the disease
for 5 years, has now been treated during 3 time periods. As expected and comparable
with anti-TNF treatment in other diseases (e.g. rheumatoid arthritis and Crohn’s disease)
there is a need to repeat the treatment. Anti-TNF has a long half-life time, but in chronic
diseases, treatment has to be repeated about 4 times a year. Little is known about the long-
term treatment with anti-TNF. The second patient, who had a relatively short duration of
CRPS, did not need repeated treatments and recovered almost completely. The question
remains whether this result can be attributed to the anti-TNF treatment or whether it is just
a fortunate natural course. Completion of the open label study followed by a double-blind
randomised controlled trial is necessary to draw more defi nite conclusions.
We also need to know how the cytokines behave during the course of the disease. Do
cytokines always play a role in long-standing CRPS 1, or do they only play a role in a
subgroup, as described by Bruehl et al, which shows the full blown picture of CRPS 1.23
Further research is necessary to gain more insight. Without doubt not only proinfl ammatory
substances but also anti-infl ammatory substances (like IL-10) are involved. Due to the lack
of a suffi cient amount of blister fl uid, the blister fl uid samples have a volume of about 300
ul or less, we were not able to answer these questions in the performed studies. Moreover,
it takes many hours to produce an artifi cial blister and although it is not harmfull it does
leave small scars. It would be unethical to induce more blisters in one patient, therefore
new patients are required to address this question.
Remarkable is the fact that Birklein et al. found higher levels of neuropeptides in plasma,
but could not fi nd a difference between the involved and the uninvolved extremity. They
suggested an impaired metabolism of neuropeptides as a reason for the higher levels.12
We doubt this explanation. The question arises whether the venous blood circulation
compartment which they were looking at is able to show differences in plasma neuropeptide
levels between the involved and uninvolved extremity. It is possible that differences are
mixed before detection. In blister fl uid, however, we also found no signifi cant differences in
neuropeptide levels between the involved and uninvolved extremity. After treatment with
capsaicin we showed an improvement in the clinical picture, but we could not demonstrate
any signifi cant changes in the local levels of CGRP in blister fl uid.20 Thus the question arises
how important neuropeptides are in the pathophysiology of CRPS. We think this indicates
that neuropeptides are only secondarily involved in CRPS 1. Supplementary to this idea
is the observation of Birklein et al. that the levels of neuropeptides are higher in patients
with CRPS 2 in comparison to the levels in patients with CRPS 1.12 Further investigation is
necessary to elucidate the role of neuropeptides in CRPS. Interesting for future research
is the role of NK-1 receptor antagonists in CRPS. Animal studies have shown that in the
chronic constriction injury model an NK-1 receptor antagonist is effective.24 There are,
however, two problems: fi rst, the fact that this model is probably more representative for
CRPS 2 and how comparable a relatively short lasting model can be for a chronic disease.
Secondly, there is a difference in activity of NK-1 receptor antagonists between species.25
Until now the effects of NK-1 receptor antagonists in humans are rather disappointing.
Nevertheless it would be interesting to study the effect of an available NK-1 receptor
antagonist in patients with CRPS.
Another question whch remains is why do some people with the same trauma or same kind
of surgery develop CRPS and others not? Is there a genetic predisposition for susceptibility
for CRPS 1. Earlier studies reported differences in the HLA system between CRPS patients
and healthy controls.25,26,27 This is an indication of the involvement of changes in the
immune system. Recently, Vaneker et al. reported the involvement of the TNF2 allele in
CRPS 1.29 Humans with this allele synthesize more TNFα in infl ammatory situations. This
phenomenon could be an explanation for the higher levels of TNFα we detected in CRPS
Another interesting fi nding was made by van der Vusse et al.30 They reported an increase
in antibody levels against the Parvo B19 virus. Perhaps this results in an acquired immune
alteration. This could change the susceptibility and could also explain why some patients
are more vulnerable to CRPS than others.
This thesis does not focus on the other important aspect of CRPS, namely sensitisation.
Sensitisation is the process in which temporary or defi nite changes in the nervous system
occur resulting in stronger or weaker transport of impulses through pain pathways or
changes in, for example, the function of the sympathetic nervous system. Sensitisation is
often a result of a combination of chronic nervous system stimulation, release of peptides
and an infl ammatory process. It is responsible for especially the chronic signs and symptoms
of the disease, such as allodynia and dystonia, and in part responsible for the sympathetic
dysfunction. Of course much research is also needed on this aspect of the disease. CRPS is
a model in vivo as to how our nervous system can act in chronic pain states.
Nowadays, the two schools of thought regarding the pathophysiology on CRPS are
communicating with each other more and more. Almost everyone is convinced that a
dysregulation of the sympathetic system and an infl ammatory process are both part of
CRPS. Consequently, solutions should be sought in combinations of therapies focussed on
For researchers, CRPS 1 is a very challenging disease. For patients, physicians and other
health care workers, however, it can be a very frustrating disease. Due to developments
in research, slowly but surely we are understanding more about CRPS 1. However, we still
have a long way to go.
Hypothesis pathophysiology CRPS 1 and CRPS 2
Higher levels of IL-6, TNFα and tryptase (Mast cells) in the involved extremity
1. Veldman PHJM, Reynen HM, Arntz IE, Goris RJA. Signs and symptoms of refl ex sympathetic
dystrophy: prospective study of 829 patients. Lancet 1993; 342: 1012-6.
Stanton Hicks M, Janig W, Hassenbusch S, Haddox JD, Boas R, Wilson P. Refl ex sympathetic
dystrophy : changing concepts and taxonomy. Pain 1995 ; 63 :127-33.
Bruehl S, Harden RN, Galer BS, Saltz S, Bertram M, Backonja M, Gayles R, Rudin N, Bhugra
MK, Stanton-Hicks M. External validation of IASP diagnostic criteria for Complex regional pain
Syndrome and proposed research diagnostic criteria. Pain 1999; 81; 147-54.
Oerlemans HM, Graff MJ, Dijkstra-Hekkink JB, Boo de T, Goris RJ, Oostendorp van RA. Reliability
and normal values for measuring the skin temperature of he hand with an infrared tympanic
thermometer: a pilot study. J Hand Ther 1999; 12: 284-90.
Karstetter KW, Sherman RA. Use of thermography for initial detection of early refl ex sympathetic
dystrophy. J Am Podiatr Med Assoc 1991; 81: 198-205.
Huygen FJPM, Niehof S, Klein J and Zijlstra FJ. Computer assisted skin videothermography is a
high sensitive tool in the diagnosis and monitoring of complex regional pain syndrome type 1.
Eur J Appl Physiol 2004; 19: 516-24
Wasner G, Schattschneider J, Baron R. Skin temperature side differences-a diagnostic tool for
CRPS? Pain 2002; 98: 19-26
Sudeck P. Uber die acute entzundliche knochenatrophie. Arch Klin Chir 1900; 62: 147.
Goris RJ. Refl ex sympathetic dystrophy; model of a severe regional infl ammatory respons
syndrome. World J Surg 1998; 22: 197-202.
10. Heerschap A, Hollander den JA, Reynen H, Goris RJ. Metabolic changes in refl ex sympathetic
dystrophy: a 31P NMR spectroscopy study. Muscle nerve 1993; 16: 367-73.
11. Oyen WJ, Arntz IE, Claessesns RM, Meer van der JW, Cortsens FH, Goris RJ. Refl ex sympathetic
dystrophy of the hand: an excessive infl amatory response? Pain 1993; 55: 151-7.
12. Birklein F, Schmelz M, Schiffer S, Weber M. The important role of neuropeptides in complex
regional pain syndrome. Neurology 2001; 57: 2179-84.
13. Leis S, Weber M, Isselmann A, Schmelz M, Birklein F. Substance P-induced protein extravasation
is bilaterally increased in complex regional pain syndrome. Exp Neurol 2003; 183: 197-204.
14. Ribbers GM, Oosterhuis WP, Limbeek van J, Metz de M. Refl ex Sympathetic Dystrophy: Is the
Immune System involved? Arch Phys Med Rehabil 1998;79: 1549-52.
15. Beek van de WJ, Remarque EJ, Westendorp RG, Hilten van JJ. Innate cytokine profi le in patients
with complex regional pain syndrome is normal. Pain 2001; 91: 259-61.
16. Huygen FJPM, Bruijn de AGJ, Klein J, Zijlstra FJ. Neuroimmune alterations in the complex regional
pain syndrome type 1. Eur J Pharmacol 2001; 429: 101-13.
17. Michiels JJ and Zijlstra FJ. Prostaglandin cylo-oxygenase products but not tromboxane A2 are
involved in the pathogenesis of erythromelalgia in trombocythaemia. Mediat Infl amm 1993; 2;
18. Huygen FJPM, Bruijn AGJ, Bruin de MT, Groeneweg JG, Klein J, Zijlstra FJ. Evidence for local
infl ammation in complex regional pain syndrome type 1. Mediators Infl lamm 2002; 11:47-51
19. Huygen FJPM, Ramdhani N, Toorenenbergen van A, Klein J, Zijlstra FJ. Mast cells are involved
in infl ammatory reactions during Complex Regional Pain Syndrome type 1. Immunology letters
2004; 91: 147-54
20. Huygen FJPM, Schoneveld JM, Kjaerness J, Klein J, Zijlstra FJ. The use of topical capsaicin in the
treatment of Complex Regional Pain Syndrome type 1; an open label study in 14 patients. 2004
21. Deleuran B, Kristensen M, Larsen CG, Matsson P, Enander I, Andersson A, Thestrup-pedersen K.
Increased tryptase levels in suction-blister fl uid from patients with urticaria. Br j dermatol 1991;
22. Huygen FJPM, Niehof S, zijlstra FJ, Hagen van PM, Daele van PLA. Successful treatment of CRPS
1 with anti-TNF. J Pain Symp Manag 2004; 27: 101-3.
23. Bruehl S, Harden RN, Galer BS, Saltz S, Backonja M, Stanton Hicks M. Complex regional pain
syndrome: are there distinct subtype and sequential stages of the syndrome. Pain 2002; 95: 119-
24. Kingerey, WS Davies MF, Clark JD. A substance p receptor (NK-1) antagonist can reverse vascular
and nociceptive abnormalities in a rat model of complex regional pain syndrome type II. Pain
25. Hill R. NK-1 (substance P) receptor antagonists- why are they not analgesic in humans? Trends
Pharmacol Sci 2000; 21: 244-6.
26. Mailis A, Wade J.Profi le of Caucasian women with possible genetic predisposition to refl ex
sympathetic dystrophy: a pilot study. Clin J Pain 1994; 10: 210-7.
27. Kemler MA, van de Vusse AC, van den Berg-Loonen EM, Barendse GA, van Kleef M, Weber WE.
HLA-DQ1 associated with refl ex sympathetic dystrophy.Neurology 1999; 53: 1350-1.
28. Van de Beek WJ, van Hilten JJ, Roep BO.HLA-DQ1 associated with refl ex sympathetic dystrophy.
Neurology. 2000; 55: 457-8.
29. Vaneker M, Laan van der L, Allebes WA, Goris JA. Genetic factors associated with Complex
regional Pain SyndromeI: HLA DRB and TNF alfa promotor gene polymorphism. Disability Med
2002; 2: 69-74.
30. Vusse van de AC, Goossens VJ, Kemler MA, Weber WE. Screening of patients with complex
regional pain sydrome for antecedent infections. Clin J Pain 2001; 17: 378-9.