Pain Practice - 2023 - Berg - The Efficacy of Oral Corticoids in Treating Complex Regional Pain Syndrome A Retrospective

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

DOI: 10.1111/papr.

13310

R E SEA RCH A RTICLE

The efficacy of oral corticoids in treating complex regional pain


syndrome: A retrospective cohort study

Corinne van den Berg MD | Frank J. P. M. Huygen MD, PhD | Jitske Tiemensma PhD

Department of Anesthesiology, Centre Abstract


for Pain Medicine, Erasmus University
Medical Centre, Rotterdam, The
Objectives: There is growing evidence supporting the role of inflammatory
Netherlands mechanisms in complex regional pain syndrome (CRPS). Corticoids, as most
effective anti-inflammatory drugs, are widely used in treating inflammation.
Correspondence
Corinne van den Berg, Department of
The aim of this study was to retrospectively assess the efficacy of oral corticoid
Anesthesiology, Centre for Pain Medicine, treatment in CRPS patients.
Erasmus University Medical Centre, Methods: Patients treated at the center of pain medicine in the Erasmus University
Postbus 2040, 3000CA Rotterdam, The
Netherlands.
Medical Centre between January 2015 and January 2020 were approached to
Email: [email protected] partake in this study. Medical records were screened for age, gender, medical
history, duration of CRPS, and CRPS severity score. Also, treatment effect, dose
and duration, pain scores (NRS), and side effects were extracted from medical
records. In addition, global perceived effect was completed in patients treated with
corticoids.
Results: Between January 2015 and January 2020, twenty-nine CRPS patients
received corticoids and met the inclusion criteria. One extreme outlier was excluded
and treatment effect was unknown for one patient. Average daily dose was 28.9 mg
(range 10–30 mg) and the mean treatment duration was 10.5 days (7–21 days).
Fourteen patients (51.9%) responded positively to treatment and thirteen (48.1%)
did not respond. Side effects were reported in five patients (17.9%).
Conclusions: Corticoid treatment was effective in more than half of the patients.
With only mild side effects reported the treatment also appears to be relatively
safe. Further research is needed to investigate the efficacy of corticoids in treating
(early) CRPS, preferably in an intervention study.

K EY WOR DS
complex regional pain syndrome, corticoids, treatment

I N T RODUC T ION Over the years, it has become clear that several mech-
anisms play a role in the development and maintenance
Complex regional pain syndrome (CRPS) is generally of CRPS. Increasing evidence supports an exaggerated
characterized by continuing pain combined with sen- inflammatory response as one of the major mechanisms.
sory, vasomotor, sudomotor, motor, and trophic symp- Studies documented increased concentrations of pro-in-
toms.1 The pain is disproportional in relation to the flammatory cytokines and neuropeptides in systemic cir-
initial trauma and the incidence varies between 5.5 and culation, cerebrospinal fluid, and in artificial skin blister
26.2 per 100.000 person years.2,3 CRPS is diagnosed fluid on the affected limb of CRPS patients.5–7 Likewise,
based on the International Association for the Study of median soluble IL-2 receptor (sIL-2R) was increased in
Pain (IASP) clinical diagnostic criteria, assessing pa- CRPS patients' serum compared to healthy blood do-
tient's symptoms and signs.4 nors, indicating increased T-cell activity in CRPS.8

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2023 The Authors. Pain Practice published by Wiley Periodicals LLC on behalf of World Institute of Pain.

394 | 
wileyonlinelibrary.com/journal/papr Pain Practice. 2024;24:394–403.
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
van den BERG et al.     | 395

Historically, CRPS was treated in an all fits one codes are used to classify the reason for the visit, given
manner. Today, treatment is mostly tailored to the most diagnoses, and treatment for a patient. Specific codes are
prominent mechanism(s) present in a specific CRPS used for CRPS: G90.5, G90.6, and G90.7 as ICD-10 codes
case, so-called mechanism-based treatment.9 No causal and 150 for DBC. Based on these codes, a search was
therapy for CRPS is available. The mechanism(s) pres- conducted in electronic patient records. Subsequently,
ent may be relevant to predict responses to individual all identified patients were approached by letter, email,
treatment options. Corticoids, as most effective anti-in- and/or phone to provide permission to view their medi-
flammatory drugs, are a natural treatment option in cal records, according to the General Data Protection
cases with prominent inflammation.10 Corticoids are Regulation.15 Only adult patients (>18 years) were ap-
widely used in treating inflammatory diseases and act proached. After permission, a single researcher (CvdB)
anti-inflammatory by interfering with various processes, viewed the medical files to assess whether a patient was
causing upregulation of the anti-inflammatory genes eligible for inclusion in this retrospective study. The in-
and downregulation of the pro-inflammatory genes.11 clusion criteria involved a CRPS diagnosis, based on the
Additionally, corticoids inhibit transcription factors IASP clinical diagnostic criteria, and the patient has to be
that control synthesis of pro-inflammatory mediators. treated at our center for pain medicine. In cases of uncer-
Corticoids also inhibit phospholipase A2, causing pro- tainty, the other authors were consulted and it was jointly
duction of inflammatory mediators.12,13 determined whether the patient was eligible. Medical files
A recent review showed corticoid treatment to be suc- from eligible patients were viewed entirely from the start
cessful in treating CRPS, especially regarding pain relief of treatment and cases who were treated at our tertiary
and improvement in range of motion.14 However, the op- referral center before the age of 18 remained in the study.
timal route of administration and optimal dose are still Patients referred for a second opinion or preoperative ad-
unknown. Today, in our clinic, a standardized moderate vice were excluded. Due to the lack of medical informa-
dose regimen is conducted in patients with clinical signs tion, patients who continued treatment started in another
of inflammation and an elevated sIL-2R level. We con- hospital were also excluded. Only patients who were
ducted this retrospective study to evaluate the effective- treated with corticoids were included in the present study.
ness of our oral corticoid treatment protocol for CRPS
in our expert center.
Data collection

M ET HOD S After inclusion, a single researcher (CvdB) viewed the


medical records. The main study parameter was reported
We conducted a retrospective study of all patients with treatment effect. Patients were categorized as either a re-
CRPS who visited the center for pain medicine between sponder or nonresponder to corticoid treatment based
January 2015 and January 2020. Our clinic is a tertiary on clinician-reported treatment effects. Medical record
referral center with CRPS being one of the fields of ex- was screened for age, gender, medical history, duration
pertise. Ethical approval was obtained from the Medical of CRPS, serum sIL-2R levels, and CRPS severity score
Ethics Committee of Erasmus University Medical Centre (CSS). The CSS is a tool to quantify clinical features as-
Rotterdam (MEC-2020-0408). sociated with CRPS based on the presence/absence of
16 clinically assessed signs and symptoms (Table 1).16,17
Furthermore, treatment dose, duration of treatment,
Treatment with oral corticoids pain scores by numeric rating scale (NRS) before and
after treatment, and side effects were noted.
Currently, short-term corticoid treatment is prescribed In addition, all patients treated with corticoids and
to CRPS patients with clinical signs of inflammation from whom clinician-reported treatment effect was pres-
and elevated sIL-2R levels in our clinic. Treatment con- ent in the medical record were approached by the re-
tains 30 mg oral prednisolone per day for a duration of searcher. Patients' assessment of corticoid treatment was
at least 7 days. measured using Global Perceived Effect (GPE), taken by
telephone. GPE asks the patient to rate how much their
condition has improved or deteriorated since corticoid
Patients treatment on a 7-point Likert scale, with higher scores
indicating more severe conditions (Table 2).
Electronic patient records were searched for patients
with CRPS in the period between January 2015 and
January 2020. In the Netherlands, hospitals use both Statistical analysis
International Classification of Diseases (ICD-10) codes
and diagnosis-treat combination (in Dutch: Diagnose- Statistical analysis was performed with IBM SPSS software,
behandelcombinatie) codes, also called DBCs. Both version 28 (IBM Corporation, Armonk, NY). Descriptive
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
396 |    ORAL CORTICOIDS IN CRPS

TA BL E 1 CRPS severity score by Harden et al.17

Symptomsa Signsb

Continuing, disproportionate pain Hyperalgesia to single pinprick


Allodynia or hyperalgesia Allodynia
Temperature asymmetry Temperature asymmetry by palpation
Color asymmetry Color asymmetry
Sweating asymmetry Sweating asymmetry
Edema Asymmetric edema
Dystrophic changes Dystrophic changes
Motor abnormalities (weakness, tremor, dystonia, decreased range Motor abnormalities (tremor/myoclonus, dystonia,
of motion, and myoclonus) decreased active range of motion, and weakness)
a
Symptoms as reported by the patient and registered as absent or present.
b
Signs as observed during physical examination by the physician and registered as absent or present.

TA BL E 2 Global perceived effect. clinician-reported treatment effect was unknown in one,


To what extent have you recovered from your symptoms since resulting in an analysis of 27 patients (Figure 1). The
starting corticoid treatment? outlier had an acute flare-up of fulminant CRPS and re-
1. Very much improved ceived a total of 2202.5 mg prednisolone over 238 days.
At first, this excessive corticoid treatment seemed effec-
2. Much improved
tive, but an amputation could not be prevented.
3. A little improved
4. No change
5. A little deterioration Responder versus nonresponder
6. Much deterioration
7. Very much deterioration Of the 27 patients, 14 patients (51.9%) were responders,
showing positive treatment effects, and 13 (48.1%) were
nonresponders, showing no effect to corticoid treatment.
statistics were used to calculate frequencies of categorical Table 3 presents the demographic and clinical character-
variables and to calculate measures of central tendency and istics of both responders and nonresponders. There were
variability of continuous variables. The Shapiro–Wilk test no differences in baseline characteristics between both
was used to analyze whether continuous variables were nor- groups. Patients received an average daily corticoid dose
mally distributed. Variables with a normal distribution are of 28.9 mg (range 10–30 mg) and mean treatment dura-
reported in means and standard deviations, otherwise, me- tion was 10.5 days (range 7–21 days). Table 4 shows treat-
dians, and interquartile ranges are used. Depending on the ment specifications compared between responders and
shape of distribution, continuous variables were compared nonresponders. There were no significant differences.
between two groups using either a two-sided independent In 8 of the 14 responders, effect of corticoid treatment
t-test or a two-sided Mann–Whitney U test. Categorical was further specified in medical records; improvement of
variables were compared using the Chi-squared test. A inflammatory features (swelling, color) was reported in
two-tailed p-value below 0.05 was considered to indicate five patients, improvement in function in one, pain relief in
statistical significance for all analyses. five patients, and allodynia improved in one. Duration of
the mentioned treatment effect was not clearly described
in medical records. In five patients, described effect only
R E SU LT S persisted during corticoid treatment, and symptoms re-
turned immediately after stopping or a few days later.
Patients

One hundred fifty-three patients were treated for CRPS Global perceived effect
in our expert center between January 2015 and January
2020. All patients were approached to request permis- In total 24 patients completed the GPE. Two patients
sion to view their medical records. In total, 123 patients (1 responder and 1 nonresponder) indicated that they
gave consent of which nine patients did not meet eligi- no longer sufficiently remember the treatment. In ad-
bility criteria after viewing medical records. A total of dition, one patient (responder) could not be reached
114 patients were eligible for inclusion in the current by telephone. Resulting in a response rate of 88.9%.
study. Corticoid treatment was given to 29 (25.4%) of Based on GPE we distinguished three groups; improve-
the patients. However, one outlier was excluded and ment (GPE 1–3), no change (GPE 4), and deterioration
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
van den BERG et al.     | 397

FIGURE 1 Flowchart of patient selection.

(GPE 5–7). Comparing GPE with clinician-reported sIL-2R levels


treatment effect all responders scored 4 or less and all
nonresponders scored 4 or higher (Table 5). The GPE There were no significant differences in the distribution
score of the responder group (median 3.0, IQR 3.0–4.0) of sIL-2R levels before (p = 0.765) and after (p = 0.753)
was significantly lower than in the nonresponder group corticoid treatment between the responders and nonre-
(median 4.0, IQR 4.0–5.75) (p < 0.001), with lower scores sponders (Figure 2 and Table 4). In both groups, a dif-
indicating more improvement. ference in sIL-2R level before and after was reported,
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
398 |    ORAL CORTICOIDS IN CRPS

TA BL E 3 Demographic and clinical characteristics of participants received corticoid treatment.

Responder Nonresponder

N = 14 N = 13 Significance

Gender 1.00
Female 13 (92.9%) 12 (92.3%)
Male 1 (7.1%) 1 (7.7%)
BMI 26.99 ± 4.61 26.68 ± 4.36 0.861
Age at the start corticoid treatment, years 40.28 ± 15.01 43.07 ± 15.53 0.639
CRPS duration at start treatment, months (median, IQR) 53.28 (15.75–110.25) 42.38 (6.5–49.0) 0.369
Affected limb 0.154
Upper limb 2 (14.3%) 4 (30.8%)
Lower limb 12 (85.7%) 9 (69.2%)
Affected side 0.915
Left 8 (57.1%) 7 (53.8%)
Right 6 (42.9%) 5 (38.5%)
Both 0 1 (7.7%)
Initiating event 0.661
Trauma 11 (78.6%) 7 (53.8%)
Surgery 2 (14.3%) 5 (38.5%)
Spontaneous 1 (7.1%) 1 (7.7%)
Note: Values are presented as number (%) for categorical variables and mean  ± SD or median (IQR) for continuous variables.
Abbreviations: BMI, body mass index; IQR, interquartile range; SD, standard deviation.

TA BL E 4 Comparison of treatment specifications and side effects between responders and nonresponders.

Responder Nonresponder

N = 14 N = 13 Significance

Duration of treatment, days 10.86 ± 3.23 10.08 ± 1.44 0.432


Daily dose, mg 27.69 ± 5.99 30 ± 0 0.178
Total dose, mg 275.96 ± 59.16 304.61 ± 38.43 0.078
sIL-2R before treatment, pg/mL 4432.92 ± 1681.84 4641.36 ± 1681.84 0.765
sIL-2R after treatment, pg/mL 3887.27 ± 1531.54 3625.00 ± 2057.78 0.753
Δ sIL-2R 997.00 (−397.00; 3031.00) 220.50 (−142.50; 1948.00)
Side effects 0.260
Yes 4 (28.6%) 1 (7.7%)
No 8 (57.1%) 7 (53.8%)
Unknown 2 (14.3%) 5 (38.5%)
Note: Values are presented as number (%) for categorical variables and mean  ±  SD or median (IQR) for continuous variables.
Abbreviations: IQR, interquartile range; mg, milligram; pg/mL, picograms per milliliter; SD, standard deviation; sIL-2R, soluble IL-2 receptor; Δ, delta.

but the difference showed to be both positive and nega- at the time of visit and 24 h before the visit were compa-
tive. Meaning in some patients sIL-2R level after treat- rable between both groups. Two symptoms (i.e., subjec-
ment was higher than before treatment, and in others, tive symptoms reported by patients) were significantly
levels fell. higher in the nonresponder group than in the responder
group: affected side colder (p = 0.035) and nail changes
(p = 0.029). Affected side warmer was significantly higher
CRPS severity score in the responder group (p = 0.037). In signs (i.e., objective
signs observed by the physician) no significant differ-
Table 6 shows the proportion of symptoms and signs in ences between both groups were found. In addition, the
responders and nonresponders according to the CSS at mean CRPS severity score was comparable between both
first outpatient visit in our center. Median pain scores groups.
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
van den BERG et al.     | 399

TA BL E 5 Global perceived effect spread across responders and nonresponders.

Global perceived effecta

Improvement (1–3) No change (4) Deterioration (5–7)

Clinician-reported treatment effect


Responder (N = 12) 7 5 0
Nonresponder (N = 12) 0 7 5
a
GPE was completed by 24 patients (12 responders and 12 nonresponders).

FIGURE 2 Boxplot of the mean sIL-2R levels before and after corticoid treatment between the responders and nonresponders.

Side effects Inflammation plays a major role in the pathophys-


iology of CRPS.6,18 Therefore, corticoid treatment ap-
Side effects were reported in five patients, four respond- pears to be a key drug in treating CRPS because of the
ers and one non-responder. The nature of the side effects anti-inflammatory and immunosuppressive effects.
was not well documented in all patients, but vomiting, However, the corticoid treatment policy is variable
dizziness, and occurrence of a wound were reported in according to the current CRPS guidelines. The Dutch
four out of five. In sixteen patients no side effects were guidelines, last updated in 2014, advise restraint due to
described in medical records. possible side effects.19 The German guidelines reported
the ideal experience with a relatively high dose of glu-
cocorticoids, starting with 100 mg daily and tapering
DI SC US SION A N D CONC LUSION S off by 25 mg each for 4 days.20 In the second edition of
the guidelines in the United Kingdom, corticoid treat-
As described in this retrospective study, corticoid treat- ment is not mentioned.21 On the contrary, the fifth edi-
ment was prescribed to 29 CRPS patients treated in our tion of practical diagnostic and treatment guidelines
expert center between January 2015 and 2020. Patients for CRPS in the United States notes that a short course
received a moderate daily dose, average 28.9 mg/day, and of steroids may be indicated in early CRPS with prom-
were treated for 21 days or less. Corticoid treatment seemed inent inflammation. Longer courses are unproven,
to be effective in more than half of the CRPS patients when and there are numerous serious contraindications to
looking at clinician-reported treatment effect and GPE. chronic steroid use.22
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
400 |    ORAL CORTICOIDS IN CRPS

TA BL E 6 CRPS severity score: presence of symptoms and signs of CRPS in each group.

Responder Nonresponder

N = 14 N = 13 Significance

NRS at the time of visit 7.0 (6.0–7.5) 7.3 (7.0–8.0) 0.156


NRS 24 h before visit 7.5 (5.8–9.5) 8.0 (6.7–8.2) 0.825
Symptoms
Continuing pain 14 (100) 13 (100) NAa
Allodynia and/or hyperalgesia 14 (100) 13 (100) NAa
Allodynia 11 (78.6) 13 (100) 0.077
Hyperalgesia 12 (85.7) 12 (92.3) 0.586
Temperature asymmetry 14 (100) 13 (100) NAa
Affected side warmer 6 (42.9) 1 (7.7) 0.037
Affected side colder 4 (28.6) 9 (69.2) 0.035
Variable; warm/cold 4 (28.6) 3 (23.1) 0.745
Color asymmetry 13 (92.9) 13 (100) 0.326
Red 5 (35.7) 7 (53.8) 0.343
Blue 7 (50.0) 5 (38.5) 0.547
Other color 1 (7.1) 1 (7.7) 0.326
Sweating asymmetry 9 (64.3) 7 (53.8) 0.581
Edema 13 (92.9) 10 (76.9) 0.244
Dystrophic changes 10 (71.4) 11 (84.6) 0.410
Nails 5 (35.7) 10 (76.9) 0.029
Hair 8 (57.1) 6 (46.2) 0.535
Skin 2 (14.3) 3 (23.1) 0.451
Motor abnormalities 14 (100) 12 (92.3) 0.290
Weakness 14 (100) 12 (92.3) 0.290
Tremor 7 (50.0) 2 (15.4) 0.057
Dystonia 3 (21.4) 7 (53.8) 0.081
Decreased ROM 13 (92.9) 10 (76.9) 0.244
Myoclonus 5 (35.7) 1 (7.7) 0.080
Signs
Hyperalgesia to pinprick 4 (28.6) 3 (23.1) 0.745
Allodyniab 11 (78.6) 12 (92.3) 0.315
Temperature asymmetry on palpation 9 (64.3) 7 (53.8) 0.581
Affected side cooler 6 (42.9) 7 (53.8) 0.568
Affected side warmer 3 (21.4) 0 0.077
Color asymmetry 8 (57.1) 10 (76.9) 0.276
Red 3 (21.4) 4 (30.8) 0.580
Blue or pale 4 (28.6) 6 (46.2) 0.345
Mottled 1 (7.1) 0 0.326
Scar 0 0 NAa
Sweating asymmetry 2 (14.3) 1 (7.7) 0.511
Increased on affected side 2 (14.3) 1 (7.7) 0.586
Decreased on affected size 0 0 NAa
Asymmetric edema 8 (57.1) 3 (23.1) 0.072
Dystrophic changes 3 (21.4) 5 (38.5) 0.333
Nails 1 (7.1) 2 (15.4) 0.747
Hair 1 (7.1) 2 (15.4) 0.747
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
van den BERG et al.     | 401

TA BL E 6 (Continued)

Responder Nonresponder

N = 14 N = 13 Significance

Skin 2 (14.3) 3 (23.1) 0.780


Motor abnormalities affected side 13 (92.9) 12 (92.3) 0.957
Tremor or myoclonus 1 (7.1) 0 0.326
Dystonia 7 (50.0) 3 (23.1) 0.148
Decreased ROM 12 (85.7) 10 (76.9) 0.557
c
Weakness 13 (92.9) 12 (92.3) 0.957
CRPS severity score 11.3 ± 1.69 11.1 ± 1.34 0.734
Note: Values are presented as number (%) for categorical variables and mean  ± SD or median (IQR) for continuous variables.
Abbreviations: IQR, interquartile range; NA, not applicable; ROM, range of motion; SD, standard deviation.
a
No p-value is available because the variable is 100% present or absent in both groups.
b
Allodynia is normally calssified into light touch, deep joint pressure, vibration, cold, and warm). Due to missing of this specification in medical records, the
specification of allodynia is not described in this table.
c
Weakness is normally rated in the severity score as 1/5: flicker of movement, 2/5: movement with gravity, 3/5: movement against gravity, 4/5: weak. However, this
ratio was not mentioned specifically in medical records and is therefore not described here.

All patients treated with corticoids in this study re- hypothalamic–pituitary–adrenal axis (HPA axis). HPA-
ceived a moderate dose (between 7.5 and 40 mg/day) of axis impairments are present in inflammatory and auto-
oral prednisolone. Treatment was shown to be effective in immune diseases such as rheumatoid arthritis, Crohn's
51.9% of the patients. Patients in this study already suffer disease, multiple sclerosis, and asthma.28,29 All diseases
from CRPS for 1–58 months at the first visit to our expert associated with increased inflammatory activity—just
center, and the CRPS duration at the start of corticoid like CRPS. In addition to stress, both physical and psy-
treatment is between 6.5 and 110 months. Therefore, it is chological, the HPA axis can also be activated by in-
likely that the longer duration of CRPS in our patients flammatory mediators.30 Activation of the HPA axis
plays an important role in the shown treatment effective- causes endogenous cortisol production, which has an
ness. Especially in the acute stage of CRPS, the classical inhibitory effect on the HPA axis, also known as nega-
signs of peripheral inflammation are described: pain, tive feedback. Subsequently, activated negative feedback
increased temperature, swelling, redness, and loss of can ensure suppressed inflammation. Only one study re-
function.23,24 In this phase, corticoid treatment is likely garding the functioning of the HPA axis in CRPS exists.
to be effective. However, at least in the majority of pa- Park and Ahn showed that a relatively high frequency of
tients with longer-existing CRPS the acute inflammation spontaneous pain attacks was associated with a reduced
extinguished and there is residual damage. In this stage cortisol awakening response and flattened slope of the
with residual damage, which may be both peripheral and diurnal cortisol decline.31 Reduced cortisol levels indi-
central, the anti-inflammatory effect of given corticoid cate dysfunction of the HPA axis, which also can influ-
treatment will be minimal or even absent. Barbalinardo ence the effect of corticoid treatment.
et al. indeed showed limited efficacy in CRPS patients In addition to the possible dysfunction of the HPA
with a duration of more than 3 months.25 axis, forms of glucocorticoid resistance could also be re-
In addition to CRPS duration, warm and cold CRPS lated. The response to corticoids is not only determined
subtypes may also play a role.26 Looking at CSS we found by the concentration of corticoids but also by differences
significant differences in some symptoms related to in- in individual glucocorticoid sensitivity. In the general
flammation. The affected side was significantly warmer population, it is estimated that around 30% of people
in responders, while the affected side was significantly are nonresponders or “glucocorticoid resistant.”32 This
colder in nonresponders. This may indicate warm CRPS glucocorticoid sensitivity causes a variety of clinical re-
in the responder group and cold CRPS in the nonre- sponses and therefore may also influence the shown effi-
sponder group. However, in other symptoms related to cacy of corticoid treatment in this study.
inflammation: color asymmetry, sweating asymmetry, We found variable sIL-2R values with both positive
and edema no differences were found. Furthermore, the and negative differences in sIL-2R levels before and after
signs related to inflammation were comparable between corticoid treatment. Meaning in some patients sIL-2R
both groups. In addition, it is known that inflammation level after treatment was higher than before treatment,
still plays a role in cold CRPS, with present levels of and in others, levels fell. In addition to a possible role
pro-inflammatory cytokines.27 for glucocorticoid resistance, this variation is probably
The increased inflammatory activity in because the sIL-2R was not determined immediately be-
CRPS could also be related to dysfunction of the fore and after treatment. For example, sIL-2R before was
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
402 |    ORAL CORTICOIDS IN CRPS

measured 117–0 days before start of treatment. In four the patients. Future research should examine the effi-
patients sIL-2R level was measured during treatment cacy of oral corticoids in a controlled study. To estab-
and not afterward. Where the sIL-2R level of another lish potential predictors of treatment response, it would
patient was determined 99 days after ending treatment. be helpful to focus on mechanism-based treatment and
It is also suggested that the effect of corticoid adminis- thus focus on CRPS patients with clinically prominent
tration on circulating sIL-2R levels depends on the state inflammation.
of immune activation and the duration of the corticoid
exposure.33 Therefore, we postulate dosage of corticoid AU T HOR C ON T R I BU T ION S
treatment probably also plays a role. Corinne van den Berg drafted the manuscript. Jitske
Corticoid treatment can cause side effects, which are Tiemensma and Frank J. P. M. Huygen co-authored this
often dose and time dependent.12,34 In this study, side ef- manuscript. All authors critically edited, read, and ap-
fects were described in only five patients (17.9%). Despite proved the final manuscript.
the not well-documented nature of these side effects the
reported effects showed to be mild: vomiting, dizziness, F U N DI NG I N F OR M AT ION
and occurrence of a wound. The results of this study thus No funding was received to assist with the preparation
showed treatment with oral corticoids appears to be rel- of this manuscript.
atively safe.
This study has several limitations. First, the retro- C ON F L IC T OF I N T E R E ST STAT E M E N T
spective design and the associated reliance on what Frank J. P. M. Huygen reports personal fees from Abbott;
was reported in the medical records. Therefore, we en- grants and personal fees from Saluda; and personal fees
countered missing data. Due to a lack of clear specifi- from Boston Scientific, Grunenthal, and Pfizer outside
cation of clinician-reported treatment effects in medical the submitted work. Corinne van den Berg and Jitske
records, we were unable to indicate which effect (pain Tiemensma report no conflicts of interest.
relief, improvement of inflammatory features, and im-
proved function) was shown in the responder group. For DATA AVA I L A B I L I T Y STAT E M E N T
example, NRS before and after corticoid treatment was Research data are not shared.
only reported in two responders. Before treatment, both
patients reported pain NRS 8 and both showed a de- ORC I D
crease in NRS: 1 and 3 points decrease. Also, the results Corinne van den Berg https://orcid.
of GPE could be biased by patients' memory. The more org/0000-0003-0017-7204
so, given the relatively the long period between treatment
and completion of the questionnaire. In addition, GPE
R EF ER ENCE S
ratings also depend on current status.35 However, when
1. Harden RN, Bruehl S, Stanton-Hicks M, Wilson PR. Proposed
comparing clinician-reported treatment effect with GPE new diagnostic criteria for complex regional pain syndrome.
we found all responders scored 4 or less and all nonre- Pain Med. 2007;8:326–31.
sponders scored 4 or higher. Meaning all responders 2. Sandroni P, Benrud-Larson LM, McClelland RL, Low PA.
showed no change or improvement, while all nonre- Complex regional pain syndrome type I: incidence and prev-
sponders showed no change or deterioration. So patients alence in Olmsted county, a population-based study. Pain.
2003;103:199–207.
did remember whether there was an effect or not despite 3. de Mos M, de Bruijn AGJ, Huygen FJPM, Dieleman JP, Stricker
long period between treatment and completion of GPE. CBH, Sturkenboom MCJM. The incidence of complex regional
Furthermore, the small sample size is a limitation. Only pain syndrome: a population-based study. Pain. 2007;129:12–20.
29 patients received corticoid treatment and were in- 4. Harden NR, Bruehl S, Perez RSGM, Birklein F, Marinus J,
cluded, of whom we could only analyze 27 patients. This Maihofner C, et al. Validation of proposed diagnostic criteria
(the “Budapest Criteria”) for complex regional pain syndrome.
small number makes it difficult to draw firm conclusions
Pain. 2010;150:268–74.
about the efficacy of corticoids in treatment of CRPS. 5. Huygen FJ, De Bruijn AG, De Bruin MT, Groeneweg JG, Klein
Additionally, patients included in this study were treated J, Zijlstra FJ. Evidence for local inflammation in complex re-
in a tertiary referral hospital. Most patients in our clinic gional pain syndrome type 1. Mediators Inflamm. 2002;11:47–51.
already suffered from CRPS for several years, and they 6. Bruehl S, Warner DS. An update on the pathophysiology of com-
plex regional pain syndrome. Anesthesiology. 2010;113:713–25.
are often treatment-resistant to other treatments. As
7. Alexander GM, Peterlin BL, Perreault MJ, Grothusen JR,
mentioned above, the CRPS duration at the first visit to Schwartzman RJ. Changes in plasma cytokines and their sol-
our expert center and also at the start of corticoid treat- uble receptors in complex regional pain syndrome. J Pain.
ment was longer existing. In addition to the already de- 2012;13:10–20.
scribed possible influence on the treatment effect, this 8. Bharwani KD, Dirckx M, Stronks DL, Dik WA, Schreurs MWJ,
Huygen F. Elevated plasma levels of sIL-2R in complex regional
also ensures that our findings might not be generalizable
pain syndrome: a pathogenic role for T-lymphocytes? Mediators
to newly diagnosed CRPS patients. Inflamm. 2017;2017:2764261.
In conclusion, treatment with oral corticoids appears 9. Mangnus TJP, Bharwani KD, Dirckx M, Huygen F. From a
to be effective and relatively safe in more than half of symptom-based to a mechanism-based pharmacotherapeutic
15332500, 2024, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/papr.13310 by Instituto Aragones De Ciencias, Wiley Online Library on [03/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
van den BERG et al.     | 403

treatment in complex regional pain syndrome. Drugs. 25. Barbalinardo S, Loer SA, Goebel A, Perez RS. The treatment
2022;82:511–31. of longstanding complex regional pain syndrome with oral ste-
10. Rhen T, Cidlowski JA. Antiinflammatory action of gluco- roids. Pain Med. 2016;17:337–43.
corticoids–new mechanisms for old drugs. N Engl J Med. 26. Bruehl S, Maihöfner C, Stanton-Hicks M, Perez RS, Vatine JJ,
2005;353:1711–23. Brunner F, et al. Complex regional pain syndrome: evidence for
11. Becker DE. Basic and clinical pharmacology of glucocorticoste- warm and cold subtypes in a large prospective clinical sample.
roids. Anesth Prog. 2013;60:25–31; quiz 32. Pain. 2016;157:1674–81.
12. Ericson-Neilsen W, Kaye AD. Steroids: pharmacology, compli- 27. Dirckx M, Stronks DL, van Bodegraven-Hof EA, Wesseldijk F,
cations, and practice delivery issues. Ochsner J. 2014;14:203–7. Groeneweg JG, Huygen FJ. Inflammation in cold complex re-
13. Williams DM. Clinical pharmacology of corticosteroids. Respir gional pain syndrome. Acta Anaesthesiol Scand. 2015;59:733–9.
Care. 2018;63:655–70. 28. Silverman MN, Sternberg EM. Glucocorticoid regulation of
14. van den Berg C, de Bree PN, Huygen F, Tiemensma J. inflammation and its functional correlates: from HPA axis
Glucocorticoid treatment in patients with complex regional pain to glucocorticoid receptor dysfunction. Ann N Y Acad Sci.
syndrome: a systematic review. Eur J Pain. 2022;26:2009–35. 2012;1261:55–63.
15. EU General Data Protection Regulation (GDPR). Regulation 29. Barnes PJ. Mechanisms and resistance in glucocorticoid control
(EU) 2016/679 of the European Parliament and of the Council of inflammation. J Steroid Biochem Mol Biol. 2010;120:76–85.
of 27 April 2016 on the protection of natural persons with re- 30. Cain DW, Cidlowski JA. Immune regulation by glucocorticoids.
gard to the processing of personal data and on the free movement Nat Rev Immunol. 2017;17:233–47.
of such data, and repealing Directive 95/46/EC (General Data 31. Park JY, Ahn RS. Hypothalamic-pituitary-adrenal axis func-
Protection Regulation), OJ 2016 L 119/1. tion in patients with complex regional pain syndrome type 1.
16. Harden NR, Bruehl S, Perez R, Birklein F, Marinus J, Maihofner Psychoneuroendocrinology. 2012;37:1557–68.
C, et al. Development of a severity score for CRPS. Pain. 32. Quax RA, Manenschijn L, Koper JW, Hazes JM, Lamberts SW,
2010;151:870–6. van Rossum EF, et al. Glucocorticoid sensitivity in health and
17. Harden RN, Maihofner C, Abousaad E, Vatine JJ, Kirsling A, disease. Nat Rev Endocrinol. 2013;9:670–86.
Perez R, et al. A prospective, multisite, international validation 33. Sauer J, Rupprecht M, Arzt E, Stalla GK, Rupprecht R.
of the complex regional pain syndrome severity score. Pain. Glucocorticoids modulate soluble interleukin-2 receptor levels
2017;158:1430–6. in vivo depending on the state of immune activation and the
18. Parkitny L, McAuley JH, Di Pietro F, Stanton TR, O'Connell duration of glucocorticoid exposure. Immunopharmacology.
NE, Marinus J, et al. Inflammation in complex regional pain 1993;25:269–76.
syndrome. A systematic review and meta-analysis. Neurology. 34. Huscher D, Thiele K, Gromnica-Ihle E, Hein G, Demary W,
2013;80:106–17. Dreher R, et al. Dose-related patterns of glucocorticoid-induced
19. Perez RS, Zollinger PE, Dijkstra PU, Thomassen-Hilgersom IL, side effects. Ann Rheum Dis. 2009;68:1119–24.
Zuurmond WW, Rosenbrand KC, et al. Evidence based guide- 35. Kamper SJ, Ostelo RWJG, Knol DL, Maher CG, de Vet HCW,
lines for complex regional pain syndrome type 1. BMC Neurol. Hancock MJ. Global perceived effect scales provided reliable
2010;10:20. assessments of health transition in people with musculoskeletal
20. Birklein F, Schlereth T. Aktuelles zur Therapie des komplex-re- disorders, but ratings are strongly influenced by current status. J
gionalen Schmerzsyndroms. Nervenarzt. 2013;84:1436–44. Clin Epidemiol. 2010;63:760–766.e761.
21. Turner-Stokes L, Goebel A, Guideline Development Group.
Complex regional pain syndrome in adults: concise guidance.
Clin Med (Lond). 2011;11:596–600.
22. Harden RN, McCabe CS, Goebel A, Massey M, Suvar T, Grieve
How to cite this article: van den Berg C, Huygen
S, et al. Complex regional pain syndrome: practical diagnostic
and treatment guidelines, 5th edition. Pain Med. 2022;23:S1–S53. FJPM, Tiemensma J. The efficacy of oral corticoids
23. Birklein F, Schlereth T. Complex regional pain syndrome-sig- in treating complex regional pain syndrome: A
nificant progress in understanding. Pain. 2015;156(Suppl retrospective cohort study. Pain Pract. 2024;24:394–
1):S94–S103. 403. https://doi.org/10.1111/papr.13310
24. Bharwani KD, Dirckx M, Huygen FJPM. Complex regional pain
syndrome: diagnosis and treatment. BJA Educ. 2017;17:262–8.

You might also like