Introduction
Complex regional pain syndrome (CRPS) is a disease that was discovered and described
by Silas Weir Mitchell. The American doctor described this pathology when he realized
that some patients who suffered gunshot wounds in the US secession war developed pain
and hyperalgesia in the extremities of the limbs.[1]
This syndrome has been known by several names since its discovery. Known today as
CRPS, names like post-traumatic pain syndrome, sudeck's atrophy, and reflex sympathetic
dystrophy have already been used.[2]
Complex regional pain syndrome is classified into two different groups: type 1, the
most common, is the one in which the patient does not have an identifiable nerve injury.
This type was formerly known as “reflex sympathetic dystrophy”; and CRPS type 2, previously
called “causalgia'', is the one that presents a proven associated nerve injury.[2]
[3]
Based on the diagnosis from doctors in general, the incidence of CRPS is of 26.2/100,000
people per year. As for the diagnosis made and confirmed by specialists, this number
decreased to 19.5/100,000, and with specialists who diagnose following the criteria
of the International Association for the Study of Pain (IASP) we have a rate of 16.8/100,000
inhabitants.[4]
The incidence related to gender shows a big difference, with predominance in women.
Men are 22.7% of those affected by the CRPS, while women are 77.3% of the patients.
The upper extremities are 10% more affected than the lower ones, with no preference
for laterality. Furthermore, the most common triggering factors are fractures and
sprains, with fractures leading in percentage.[4]
The pathophysiology of CRPS is not yet fully understood, with only a few hypotheses
that contribute to clarify the mechanism that causes this pathology. In the following
paragraphs, the hypotheses most commonly related to CRPS are presented.
The main mechanisms that explain the pathophysiology of CRPS are nerve injury, ischemic
reperfusion injury or oxidative stress, central sensitization, peripheral sensitization,
altered sympathetic nervous system function or sympathetic-afferent coupling, inflammatory
and immune-related factors, genetic factors, brain plasticity, and psychological factors.
Nerve Injury
Even though the pathophysiology of CRPS is not fully comprehended, it is known that
persistent inflammation in the affected limb is something common in all patients affected
by this condition. The phenomenon that explains the onset of this inflammation is
a nerve injury, which in cases of CRPS type 2 has a clear and determined origin. In
cases of CRPS type 1, however, this lesion cannot be explained, largely because the
lesion was caused in small fibers, structures that often cannot be identified in imaging
exams. About 90% of patients with CRPS do not have an identifiable nerve injury; thus,
CRPS type 2 represents only 10% of the patients affected by this syndrome.[5]
[6]
Reperfusion Injury or Oxidative Stress
The mechanism of reperfusion ischemia is also pointed out as one of the factors that
contribute to the development of CRPS type 1. Patients who experienced trauma and
had to immobilize the injured limb after surgery were more likely to develop the syndrome.[7] The mechanism of reperfusion ischemia is already well-known. The process results
in an accumulation of free radicals leading to oxidative stress. The oxygen supplied
after reperfusion generates reactive oxygen species (ROS), a situation made possible
by the lack of production of antioxidant agents by the ischemic tissue. Reactive oxygen
species promote endothelial dysfunction and DNA damage, in addition to inflammation,
which can lead to a cascade of cytokinins that will result in cell death and symptoms
related to the pathophysiology of the inflammatory process.[8]
[9] Studies that carried out ischemia experiments with consecutive reperfusion in rats
showed that the animals triggered symptoms similar to those of SDRC type 1. Hyperalgesia
and allodynia were observed even though no nerve damage was found in rodents.[10]
[11]
Central Sensitization
Sensitization of the central nervous system is also one of the hallmarks of complex
regional pain syndrome. The lesion triggers the release of neurotrophic factors and
proinflammatory substances, such as bradykinins and prostaglandins that, together,
activate phosphokinases A and C, which in turn phosphorylate specific sodium channels
of sensory neurons. This mechanism leads to peripheral sensitization of afferent nociceptors.
This entire process of peripheral sensitization leads to continuous depolarization
that will lead to a blockage of magnesium ions in NMDA receptors, the main receptor
of the glutamatergic system located in the postsynaptic membrane of neurons. The blocking
of magnesium ions causes these receptors to be activated, thus leading to depolarization
of the pain pathway and amplifying its signal, finally causing central sensitization.[12]
Peripheral Sensitization
After a tissue injury to the peripheral, as to the central, nervous system provides
a sensitization that protects the body against unnecessary movements. This sensitization
in the peripheral nervous system occurs thanks to the increase in the firing rate
of the nociceptors and its response to normally painful stimuli, with an antagonist
decrease in the firing threshold for thermal and mechanical stimuli.[13]
[14]
[15] Inflammatory mediators such as proinflammatory cytokines (TNF-alpha, IL-1beta),
prostaglandin E2 (PGE[2]) , bradykinin, and nerve growth factor (NGF) increase the
sensitivity and excitability of nociceptors by enhancing the activity of pronociceptive
receptors and ion channels.[13]
Moreover, a study developed by Moy et al. in 2017 demonstrates that phosphorylation
of the 5′ cap-binding protein eIF4E by its specific mitogen-activated protein kinase
(MAPK) interacting kinases (MNKs) 1/2 is a key factor in nociceptor sensitization
and the development of chronic pain.[14] The authors, by these studies, advocate in favor of a new pain pathway of nociceptive
plasticity; thus, it is clear that the pathophysiology of CRPS is, indeed, intriguing
and multifactorial.
Altered Sympathetic Nervous System Function or Sympathetic–afferent Coupling
The medical literature also assumes that modifications in the sympathetic nervous
system contribute to CRPS, which in the past was known as reflex sympathetic dystrophy.
This information, nevertheless, is controversial, since a prospective study in patients
early after fractures shows that those with reduced sympathetic outflow after the
injury are at greatest risk of developing subsequent CRPS symptoms.[16]
Localized injuries have been shown to result in the expression of catecholamine receptors
on nociceptive fibers, so the circulating catecholamines released after the stress
and pain might directly increase the firing of nociceptors, a reflex known as sympathetic-afferent
coupling, observed in various studies that hypothesized that this mechanism plays
a role in the severity of the symptoms.[17]
[18]
[19]
[20]
[21]
[22] These, once again, demonstrate a multifactorial principle in the genesis of CRPS
with the intersections of sympathetic and peripheral sensitization theories.
A common symptom of CRPS that is explained in part by the sympathetic system is the
temperature asymmetry. Vasodilating drugs and sympathetic blockade have been cornerstones
of therapy in cold CRPS for years. However, only a limited part of these patients
improves on this kind of therapy. Research has shown a pivotal role for inflammation
in the pathophysiology of CRPS,[23] which will be discussed below.
Inflammatory and Immune-related Factors
Here we will discuss what may be the latest and most complex mechanism involving CRPS.
This concept seems to link the many other mechanisms and to participate decisively
in the elucidation of pathophysiology.
Evidence of the involvement of inflammatory mechanisms, especially in the acute phase,
is present in studies documenting raised concentrations of proinflammatory neuropeptides
and mediators, such as substance P, calcitonin gene-related peptide, bradykinin, and
cytokines interleukin-1β (IL-1β), interleukin-2 (IL-2), and interleukin-6 (IL-6),
tumor necrosis factor α (TNF- α) in the systemic circulation, cerebrospinal fluid,
and affected limbs of patients with CRPS, which can explain the symptoms of vasodilation
causing a warm red appearance in the affected area, and may increase hair growth and
sweating.[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
In humans, increased numbers of proinflammatory monocytes (CD14 + ; CD16 + ) in addition
to altered innate immune responses and mast cells have been reported in patients with
CRPS compared with healthy controls.[38]
[39]
[40]
[41] A new theory advocates that antibodies from people with CRPS may be capable of transferring
the condition to previously unaffected individuals, also supporting a role for immune
mechanisms.[42] In a work by Goebel et al. from 2011, immunoglobulin G (IgG) from patients with
CRPS when injected into mice in the absence of any injury induced motor changes, characteristic
of CRPS.[43] This autoimmune model[44] can explain the link between the autonomic nervous system and the CRPS, since the
presence of autoantibodies directed against β2 adrenergic and muscarinic type 2 receptors
were found in patients with CRPS.[45]
[46]
[47]
Furthermore, a study with mice model conducted in 2019 demonstrated that, in CRPS,
IgG is significantly increased and seems to prolong swelling and induce stable hyperalgesia.
The author also says that CRPS IgG-injected mice displayed sustained microglia and
astrocyte activation in the dorsal horn of the spinal cord and pain-related brain
regions, indicating a link with central sensitization. Even more interesting, genetic
deletion of interleukin-1 (IL-1) using IL-1αβ knockout mice and perioperative IL-1
receptor type 1 blockade with the drug anakinra prevented these changes, showing one
other possible correlation between various theories such as inflammatory, autoimmune,
and genetics hypothesis to the CRPS pathophysiology.[48]
Although well explained, the autoimmune theory is not yet totally elucidated. One
parallel, randomized, placebo-controlled, multicenter trial tried to confirm the efficacy
of low-dose IVIg compared with placebo in reducing pain during 6 weeks in adult patients
who had CRPS from 1 to 5 years but did not succeed, emphasizing once again the multifactorial
genesis of CRPS.[49]
Brain Plasticity
Neuroimaging testing suggests that several brain changes are associated with CRPS.[50]
[51]
[52]
[53]
[54] Although recent studies point out that the change in the brain is associated with
the onset of the syndrome, it is not known for sure if this is what occurs or if it
is the syndrome that causes the change in the brain.[50] However, the longer you develop the syndrome, the greater the change in neuroplasticity.[53]
In CRPS, there is an asymmetry between the corresponding brain regions. A reduction
(structure or function) occurs on the side of the affected brain or there is an increase
on the side of the unaffected brain.[50]
The main brain areas where this difference in asymmetry occurs are endogenous pain
inhibitory pathways (opioid-mediated), primary and secondary somatosensory cortices,
the primary motor cortex, the insula, and the cingulate cortex.[50]
[51]
[52]
[53]
[54]
Genetic Factors
There is a lack of studies of genetic factors in CRPS. However, some family-based
studies reported a potential genetic predisposition.[51] The studies have identified polymorphisms at the genes encoding α1a adrenoceptors
and the human leukocyte antigen (HLA) system (HLA-DQ8, HLA-B62), tumor necrosis factor-alpha
(TNFα) gene and the angiotensin-converting enzyme gene.[50]
[51] Due to the lack of studies on the genetic factors involved in CRPS, the pathophysiological
mechanism related to genetics is still not completely understood.[52]
Psychological Factors
It is not known for certain whether psychological factors are directly related to
the development of CRPS, but it is suspected that anxiety and depression are related
to the syndrome.[50]
[51] It is suspected that psychological factors alone do not contribute to the onset
of the syndrome; however, psychological factors linked to tissue damage may imply
an increase in the severity of the syndrome, which might represent an increase in
the intensity of the pain.[50]
[54] It is also suspected that psychological factors, especially before invasive procedures,
provide an increase in the chance of developing CRPS.[50]
The main signs and symptoms found in patients with CRPS are spontaneous pain; hypoesthesia
and hyperpathia; edema; skin blood flow abnormality; color change; abnormal sudomotor
activity; tissue atrophy; and involuntary movements.[4]
[55]
The diagnosis of CRPS is clinical, without imaging indicators or precise serum markers
that indicate this pathology. The Budapest criteria is the one adopted to perform
the diagnosis.[56]
The objective of the present systematic review is to show the results of the most
recent unconventional treatments that add quality of life and improve the prognoses
of CRPS patients.
Results
[Table 1] aimed to analyze how quantitatively the level of pain decreases in a given treatment
against CRPS.
Table 1
Description of Clinical Trials studies for CRPS treatment
|
REFERENCE
|
NUMBER OF PATIENTS
(EFFECTIVE/CONTROL)
|
CRPS
TYPE
|
TREATMENT
|
FOLLOW-UP
(MONTHS)
|
PAIN RELIEF TREATMENT COMPARED WITH BASELINE
|
CONTROL PAIN RELIEF COMPARED WITH BASELINE (%)
|
CLASSIFICATION
|
|
[57]
|
54 (36/18)
|
1
|
spinal cord stimulation
|
60
|
25.37
|
14.28
|
fair
|
|
[58]
|
22 (22/0)
|
1
|
stellate ganglion blockade
|
0.5
|
87.5
|
−
|
excellent
|
|
[59]
|
13 (7/6)
|
1
|
low-dose intravenous immunoglobulin
|
0.83
|
25
|
12.5
|
fair
|
|
[60]
|
67 (33/34)
|
1
|
ct-guided radiofrequency neurolysis
|
24
|
67.6
|
21.2
|
good
|
|
[61]
|
42 (42/0)
|
1
|
intrathecal baclofen
|
12
|
22.4
|
−
|
bad
|
|
[62]
|
14 (14/0)
|
1
|
spinal cord stimulation
|
6
|
71.8
|
−
|
good
|
|
[63]
|
29 (15/14)
|
1
|
Ketamine
|
3
|
71.4
|
22.6
|
good
|
|
[64]
|
56 (29/27)
|
1
|
intravenous magnesium
|
3
|
14.8
|
10
|
bad
|
|
[65]
|
13 (6/7)
|
1
|
infliximabe
|
1.5
|
38
|
0
|
fair
|
|
[66]
|
74 (40/34)
|
1
|
amino-bisphosphonate neridronate
|
1.3
|
65.6
|
32.1
|
good
|
|
[67]
|
22 (12/10)
|
1
|
intramuscular magnesium sulphate
|
0.75
|
8.21
|
−
|
bad
|
|
[68]
|
29 (15/14)
|
1
|
thoracic sympathetic block
|
12
|
36
|
5.19
|
fair
|
|
[69]
|
51 (51/0)
|
1
|
spinal cord stimulation of the dorsal root ganglion
|
12
|
66.3
|
56.7
|
good
|
|
[70]
|
28 (14/14)
|
1
|
mirror therapy in stroke patients
|
6
|
50
|
−
|
good
|
|
[71]
|
53 (27/26)
|
1
|
pain exposure physical therapy
|
9
|
28.6
|
30.1
|
fair
|
|
[72]
|
29 (29/0)
|
1
|
oral corticosteroids
|
1.5
|
13.33
|
−
|
bad
|
|
[73]
|
52 (C.O.T.)
|
1
|
prednisolone
|
2
|
57.2
|
−
|
good
|
|
[74]
|
30 (15/15)
|
1
|
transcutaneous electrical nerve stimulation
|
UNK
|
70
|
−
|
excellent
|
|
[75]
|
33 (18/15)
|
1
|
exposure in vivo
|
6
|
49
|
47.5
|
fair
|
|
[76]
|
105 (55/50)
|
1 and 2
|
dorsal root ganglion stimulation
|
12
|
81.39
|
67.16
|
excellent
|
|
[77]
|
28 (C.O.T.)
|
1 and 2
|
spinal cord stimulation
|
2.5
|
45.24
|
12.37
|
fair
|
|
[49]
|
108 (56/52)
|
1 and 2
|
low-dose intravenous immunoglobulin treatment
|
1.5
|
6.76
|
3.5
|
bad
|
|
[78]
|
22 (11/11)
|
1
|
transcranial direct current stimulation
|
1.5
|
16.23
|
8.77
|
bad
|
|
[79]
|
9 (4/5)
|
1
|
mycophenolate treatment
|
5.5
|
45.45
|
−
|
fair
|
|
[80]
|
15 (C.O.T.)
|
UNK
|
paravertebral block performed at the t2 level
|
UNK
|
50.06
|
−
|
good
|
|
[81]
|
52 (26/26)
|
1
|
bioptron light therapy combined with conventional therapy
|
0.5
|
93.65
|
80.1
|
excellent
|
|
[82]
|
30 (15/15)
|
1
|
fluidotherapy combined with conventional therapy
|
0.75
|
50
|
37.5
|
good
|
|
[83]
|
12 (12/0)
|
1
|
short term glucocorticoid
|
0.2
|
50
|
−
|
good
|
|
[84]
|
24(12/12)
|
1 and 2
|
selective l4 dorsal root ganglion stimulation
|
3
|
37.72
|
−
|
fair
|
Abbreviations: C.O.T., crossed over treatment; CT, computed tomography; UNK, uninformed.
For that, we analyzed some initial quantitative mathematical quantities (average of
pain, the median of pain, maximum pain) given by the articles in the baseline. We
adopted these mathematical numbers as the initial average of the table at 100%.
Later, in the analysis of the article, with the quantitative number of the pain level
given by the articles after the treatment (final average) of the last follow-up, we
were able to arrive at how much the pain decreased in that particular treatment. This
value was calculated by subtracting the final average multiplied by 100%, followed
by dividing it by the initial average, from 100%. It is worth mentioning that this
analysis was performed in the effective therapy group (100% of the articles in the
table) and the control when it presented quantitative data.
Furthermore, the number of patients in the table does not necessarily reflect the
initial number, but rather the final present number described in each article. According
to the quantitative data obtained in pain relief, we classified the treatment as bad,
fair, good or excellent.
For the treatment to be considered bad it should reach a maximum of 24.9%, fair should
be between 25 and 49.9%. To be evaluated as good, it should be between 50 and 74.9%,
and for excellent, the range was from 75 to 100%.
Of the 29 articles that completed our selection criteria, 20.7% (6/29) of the performances
of the treatments were bad, 31% (9/29) were fair, 34.5% (10/29) were good, and 13.8%
(4/29) were excellent. The average painful reduction of all articles was 46.37% 7
months of follow-up (average).
Among these, surgical treatments (10/29) showed an average pain reduction of 56.9%
and follow-up of 14.67 months; in contrast, conservative treatments (19/29) had a
reduction of 40.82% and a follow-up of 3.16 months.
Moreover, the table shows an average of 38.48 patients in an average follow-up of
7 months. It is worth mentioning that the ratio between the average number of patients
in the experimental group and in the control group is 1.52 (616/405, excluding crossed-over
treatment).
Discussion
According to the data in [Table 1], surgical treatments (10/29) showed an average pain reduction of 56.9% and a follow-up
of 14.67 months. In contrast, conservative treatments (19/29) had a reduction of 40.82%
and a follow-up of 3.16 months. Thus, we conclude that invasive treatments are more
effective in combating one of the symptoms of CRPS: pain.
Among the surgical treatments, according to [Table 1], in descending order, the procedures that most reduce pain are: stellate ganglion
blockade, dorsal root ganglion stimulation, spinal cord stimulation,[62] computed tomography-guided radiofrequency neurolysis, spinal cord stimulation of
the dorsal root ganglion, paravertebral block performed at the t2 level, spinal cord
stimulation,[77] selective l4 dorsal root ganglion stimulation, thoracic sympathetic block, and spinal
cord stimulation.
However, the most efficient treatment for pain reduction was a nonsurgical treatment,
bioptron light therapy combined with conventional therapy.
