Complex regional pain syndrome (CRPS) is defined as a group of regionally painful conditions disproportionate to any inciting event [
Treatment of CRPS continues to be challenging for clinicians due to a lack of consensus and evidence-based therapies. Corticosteroids have been previously studied as a possible pharmacologic treatment for CRPS [
Of note, CRPS diagnosis by the Budapest criteria does not require specific diagnostic tests or specialized equipment [
Participants were included in this retrospective cohort study if they fulfilled the clinical Budapest criteria (see Table
Diagnosis of complex regional pain syndrome (CRPS) by clinical Budapest criteria.
Budapest criteria signs and symptoms (IASP [ | |
Continuing pain that is disproportionate to any inciting event | |
|
|
Sensory | Hyperalgesia, allodynia, altered sensation/paresthesia |
Vasomotor | Temperature asymmetry, skin colour asymmetry |
Sudomotor | Edema, sweating changes |
Motor/trophic | Decreased range of motion, weakness, trophic changes (hair, nail, skin) |
|
|
Sensory | Evidence of hyperalgesia to pinprick, allodynia to light touch, and paresthesia by different sensation upon application of same pressure to CRPS-affected and non-CRPS-affected area |
Vasomotor | Evidence of temperature asymmetry manually and skin colour asymmetry visually by simultaneous comparison between CRPS-affected and non-CRPS-affected area |
Sudomotor | Evidence of edema by observing lack of normal wrinkling, sweating changes determined by observing sweating patterns at CRPS-affected region differing from non-CRPS-affected regions |
Motor/trophic | Evidence of decreased ROM with active movement, weakness determined by decreased strength of joint compared to unaffected side, decreased or increased tendon reflexes as determined by reflex response to hammer in CRPS-affected region, trophic changes of hair or nail evaluated by comparing between CRPS-affected and non-CRPS-affected regions, skin changes by presence of shiny skin |
|
Flowchart of patients fulfilling inclusion criteria. CRPS: complex regional pain syndrome.
The typical prednisone regimen started with 60 mg followed by taper of 5 mg per day until 20 mg. Patients then remained on 15 mg for one week, 10 mg for one week, and finally 5 mg for one week. A slightly modified decreased dose was chosen in elderly, adolescent, and diabetic patients, typically starting at 40 mg before the taper, as has been described in a previous study [
The primary outcomes were the presence or absence of Budapest criteria signs and symptoms after prednisone therapy, as compared to before prednisone treatment. CRPS symptoms were gathered from patient histories. CRPS signs were assessed by one physiatrist (PW) and extracted from the clinical records by another author (AJ). Details of sign assessment by the physiatrist can be found in Table
Detailed methods of assessing CRPS signs by physiatrist.
|
Hyperalgesia was evaluated by increased pain response to pinprick |
Allodynia was determined by painful response to light touch |
Altered sensation/paresthesia was evaluated by patient noting different sensation following application of the same force on CRPS-affected and non-CRPS-affected area, usually contralateral extremity |
|
|
Temperature asymmetry was determined by simultaneous comparison between CRPS-affected and non-CRPS-affected area by clinical observation without specialized equipment |
Skin colour changes were determined by visual comparison between CRPS-affected and non-CRPS-affected area |
|
|
Edema was determined by observing lack of normal wrinkles, for example, at knuckles as well as generalized swelling |
Sudomotor signs (sweating changes or asymmetry) were determined by observing sweating patterns at CRPS-affected region differing from non-CRPS-affected regions |
|
|
Evaluation of range of motion (ROM) was determined by examining the active and passive ROM of all proximal and distal joints of the affected limb |
Weakness was determined by decreased strength compared to unaffected side |
Decreased or increased tendon reflexes were determined by reflex response to hammer in CRPS-affected region |
Increase or decrease in hair growth was evaluated by comparing between CRPS-affected and non-CRPS-affected regions and confirming with the patient that this was not due to nonnatural cause (ie., shaving only one limb) |
Increase or decrease in nail growth was determined by comparing between CRPS-affected and non-CRPS-affected regions and confirming with patient this was not due to nonnatural cause (ie., cutting nails on only one side of body) |
Skin changes were evaluated by the presence of shiny skin, brawny, or other observed asymmetries |
Restoration of limb usage was the primary goal. At final follow-up, patient chart descriptions permitted stratification of pain and range of motion into three levels rather than only present or absent. Pain was stratified into “no longer present”, “decreased pain”, or “not improved”. Any residual or minimal pain was grouped as “decreased pain”, even if insignificant in the patients’ day-to-day lives. Range of motion (ROM) was stratified into “fully restored”, “functionally restored”, or “not restored”. “Fully restored” ROM meant patients were able to actively demonstrate full ROM expected for healthy joints. “Functionally restored” ROM indicated patients were capable of using their affected limb for required activities but described or demonstrated residual joint stiffness. For instance, a patient could not fully tuck fingertips in their fist but were able to play tennis and use a pen. “Not restored” ROM meant the ROM limitations interfered considerably with many of the patient’s activities.
At each visit, deidentified photographs of the CRPS-affected limb were taken. The edema, vasomotor signs, and degree of ROM were captured in each photograph. Representative photographs were chosen to document progression of recovery.
Any adverse reactions were documented at each visit.
CRPS onset was estimated as the earliest date in the patient’s records when the documented signs and symptoms fulfilled the clinical Budapest criteria. Duration of follow-up was determined by calculating duration between first and last clinical encounter. To confirm reductions of CRPS signs and symptoms were statistically significant, McNemar’s test on the difference of proportions, applicable to pretreatment and posttreatment, was performed. Kendall’s tau
Patient demographics and clinical data are provided in Table
Patient demographics and relevant variables collected for the retrospective cohort.
Variable | Duration or proportion |
---|---|
Average age | 51.5 ± 18.7 years [11–85 years] |
Female sex | 66.7% |
Inciting trigger (fracture) | 41.0% |
Inciting trigger (idiopathic) | 25.6% |
Inciting trigger (surgery) | 12.8% |
Inciting trigger (trauma) | 12.8% |
CRPS in upper extremity | 74.4% |
CRPS right side unilateral | 48.7% |
CRPS left side unilateral | 43.6% |
CRPS bilateral | 7.7% |
CRPS type I | 79.5% |
Fulfilled Budapest research criteria | 59.0% |
Mean duration of prednisone therapy | 27.2 ± 5.3 days |
Mean duration of CRPS onset to start of prednisone treatment | 80.8 ± 67.7 days |
Mean duration between initial and final visit | 116.4 ± 159.1 days |
After completing prednisone treatment, only one patient (2.6%) continued to fulfil the clinical Budapest criteria. This patient had the longest duration of symptoms prior to treatment at 324 days, significantly longer than our average of 80.8 ± 67.7 days. Symptoms and signs resolved following oral prednisone treatment per McNemar’s test (
(a, b) Prevalence of complex regional pain syndrome signs and symptoms before and after short course of prednisone treatment. There is a statistically significant decrease in all signs and symptoms categories per McNemar’s test (
Range of motion (ROM) levels at final visit after completion of prednisone treatment. Functional range of motor recovery allowed patients to perform most activities without trouble such as using a pen or playing tennis.
Range of motion stratification | Prevalence after prednisone treatment |
---|---|
Fully restored ROM | 51.3% (20/39) |
Functionally restored ROM | 43.6% (17/39) |
ROM not restored | 5.1% (2/39) |
Pain levels at final visit after completion of prednisone treatment.
Pain stratification | Prevalence after prednisone treatment |
---|---|
Pain no longer present | 48.7% (19/39) |
Decreased pain permitting functional use | 48.7% (19/39) |
Pain not improved | 2.6% (1/39) |
The degree of ROM recovery and time to treatment had a low positive correlation (
Representative photographs tracking patient response to prednisone after various inciting triggers are shown. Recovery of CRPS after carpal tunnel syndrome (Figure
Complex regional pain syndrome type II due to carpal tunnel syndrome with multiple joint ranges affected. There is progression from before prednisone treatment (left images) to two weeks of treatment (middle images) to completion of four-week course of prednisone (right images).
Complex regional pain syndrome after right humerus fracture with involvement of the shoulder, elbow, wrist, and fingers presenting with pain and decreased range of motion. Top images were before treatment, and bottom images were after 4-week course of prednisone.
Complex regional pain syndrome after elbow dislocation. Initial presentation is shown in the first column. The patient no longer met Budapest criteria at 4 weeks on prednisone, with functionally but not fully restored range of motion (second column). However, the patient continued to recover after completion of treatment (third column).
Idiopathic left leg complex regional pain syndrome in adolescent female. Left images were taken before prednisone treatment, middle images were taken at two weeks of treatment, and right images were taken at two months after starting treatment. The patient made a full recovery.
The majority of patients (71.8%) reported no side effects (Table
Breakdown of the reported side effects during the course of prednisone therapy.
Side effect | Proportion |
---|---|
None | 71.8% (28/39) |
Difficulty sleeping | 12.8% (5/39) |
Headache | 7.7% (3/39) |
Weight gain | 2.6% (1/39) |
Nausea | 2.6% (1/39) |
Vomiting | 2.6% (1/39) |
Elevated blood glucose | 2.6% (1/39) |
Elevated blood pressure | 2.6% (1/39) |
Osteopenia | 2.6% (1/39) |
Anxiety | 2.6% (1/39) |
This study evaluated the effectiveness of prednisone for CRPS diagnosed using the current International Association for the Study of Pain (IASP) criteria (Budapest clinical criteria), in an outpatient community setting, using a retrospective cohort design. Following prednisone treatment, all but one patient no longer met these criteria (97.4%). Over 90% of patients reported functional improvement in range of motion. We found significant decreases following treatment in the CRPS clinical features. No factors such as age, sex, upper/lower body, injury mechanism, or short/long treatment duration predicted outcomes to prednisone treatment. This study builds on the existing CRPS literature to demonstrate how the Budapest criteria can be used in the community setting to facilitate early treatment and how a longer follow-up duration can prevent relapse. We highlight how this study has a similar patient population to large CRPS studies [
Our results are suggestive of consistent prednisone effectiveness in patients diagnosed using the Budapest criteria. We note that Bean et al. [
The pain level results (Table
Our retrospective study also focused on the clinical features of CRPS before and after treatment. The sensory, vasomotor, and sudomotor/edema categories of the Budapest criteria decreased the most. The motor/trophic group features decreased significantly though not to the same extent. This reflects the ongoing stiffness in contracted joints. Our finding aligns with a systematic review of CRPS outcomes that found that motor symptoms tend to persist, whereas sudomotor and vasomotor symptoms improve [
It has been speculated that prednisone treatment may be most effective only in early CRPS. [
Our community-based study reflects a patient population similar to other published CRPS populations. Fracture has been reported to incite CRPS in 42% to 46% of cases [
While previous studies have investigated corticosteroids as a treatment for CRPS, our study has several key differences: updated diagnostic criteria, length of follow-up, documentation of CRPS onset to treatment duration, and photographs. A retrospective study by Atalay et al. reported positive outcomes following treatment with prednisolone on 45 patients using the previous IASP CRPS criteria [
Recovery from CRPS clinical features is demonstrated in the pre- and posttreatment photographs (Figures
There is concern when prescribing a course of steroids due to contraindications and side effects. In our study, we offered a relatively short course of prednisone therapy. The majority of the patients reported no side effects with the tapering regimen (71.8%). All side effects were temporary and subsided after prednisone cessation. Only one patient withdrew from the study due to side effects (Figure
There are several limitations in interpreting our data. First, there was no quantitative measure of pain and ROM, though we note the Budapest criteria is a clinical tool that does not provide for a specific quantitative measure. Harden explained the challenge of seeking specific biometrics in CRPS. Our primary outcome was the resolution of the signs and symptoms and return to functional limb usage. As a retrospective study of clinical practice, there are no placebo controls in this study. Confounders in this study include different durations of therapy, different times of CRPS onset to initiation of prednisone therapy, varied concurrent therapies (such as vitamin D, calcium, and physiotherapy), medication adherence, and different number of follow-up visits. This study therefore can only be interpreted as being suggestive of beneficial effects of corticosteroids in early CRPS. A controlled prospective study is needed to address the design limitations of this study and account for the influence of any confounders. Such a trial could be better suited to describe the patient population most responsive to prednisone therapy.
These results suggest that in acute CRPS patients diagnosed by the Budapest criteria with multijoint involvement, prednisone may be an important therapy to promote patient recovery. Our study highlights the safe and effective use of prednisone in the routine community clinic setting without specialty services or tests.
The patient data used to support the findings of this study are restricted by the Vancouver Island Health Authority Ethics Review Board in order to protect patient privacy. Data are available from Paul Winston (
The authors report no conflicts of interest.
The authors are grateful to Victor Espinosa for his help with the statistical analysis.