Treatment of Complex Regional Pain Syndrome in Children and Adolescents: A Systematic Review

Background: Complex Regional Pain Syndrome (CRPS), as known as Reex Sympathetic Dystrophy, is characterized by chronic, spontaneous and provoked pain of the distal extremities, whose severity is disproportionate to the triggering event. In growing age, CRPS affects most commonly women aged 5 to 17 years, with a maximum incidence peak around the 13th year of age. Among the orthopedics, diagnosis remains challenging, as well as, the treatment is still debated and multidisciplinary. The purpose of this systematic review is to analyze the available literature to provide an update on the latest evidence related to the treatment of CRPS in growing age. Methods: Three reviewers searched Pubmed and Web of Science databases from their date of inception to the 20th September 2019 in accordance with the Preferred Reporting Items for Systemic Reviews and Meta-Analyses guidelines.Data extraction was performed independently by 3 reviewers based on predened criteria and the methodologic quality of included studies was quantied by the the Newcastle-Ottawa Quality Assessment Scale Cohort Studies. Results:A total of n = 264 articles were found. At the end of the rst screening, following the previously described selection criteria, we selected n = 103 articles eligible for full-text reading. Ultimately, after full-text reading and a reference list check, we selected n = 6. The articles focused on physical (PT), cognitive-behavioral (CBT) and pharmacological (PhT) treatments. The combination of PT+CBT is the most ecacy and suggested, but a commonly accepted protocol has been not developed. Despite the good patients outcome, several are the adverse events recorded in PT. Conclusions: Physical therapy in association with occupational, and cognitive behavioral treatment is the recommended option in the management of pediatric CPRS. Pharmacological therapy should be reserved for refractory and selected patients. The design and development of a standard protocol are strongly suggested.


Background
First described in the 17th century as "causalgia" [1], complex regional pain syndrome (CRPS) is characterized by chronic, spontaneous and provoked pain of the distal extremities, whose severity is disproportionate to the triggering event [2]. Three different CRPS subtypes have been distinguished: Type 1, previously known as re ex sympathetic dystrophy (RSD), whose cause is not always known. Type 2, which results from nerve damage. Type 3 or not otherwise speci ed CRPS, which partly shares clinical and diagnostic aspects with the previous types [2].
CRPS type 1 affects children and adolescents aged 5 to 17 years, with a peak incidence around the 13th year of age, and it is more frequently found in women (70% of cases) [3,4]. The pathogenic mechanism is still unclear, although several hypotheses have been proposed. Genetic factors, altered microcirculation, and traumas, such as sprains, fractures and surgical procedures, contribute to pain symptoms. Anxiety, somatization, and familial and school problems could also play a role. Chronic pain, generally unilateral and limb localized, autonomic and motor dysfunction and trophic disorders are the main symptoms of CRPS type 1. There are two presentations of the syndrome: the "warm" one, with red, warm, swollen skin that usually occurs in the acute phase, and the ''cold'' one, with blue/purple, cold, sweaty skin, which is usually associated with the chronic phase [3][4][5][6]. The diagnosis of CRPS type 1 is clinical and based on the Budapest diagnostic criteria [4]. However, diagnosis remains challenging due to the lack of validated diagnostic tests and the di culty of differential diagnosis. Laboratory and imaging tests can be helpful in the event of diagnostic doubt [1]. Treatment is multidisciplinary, and it is mostly based on physical and psychological therapy and medications; only in selected subjects is treatment invasive [7]. The purpose of this systematic review was to analyse the available literature to provide an update on the latest evidence related to the treatment of CRPS type 1 in children and adolescents, highlighting the multidisciplinary approach.

Literature search strategy
A systematic review of the current literature was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [8]. On 20th September 2019, three independent authors (SM, VF and DiMF) performed a systematic review of two different medical electronic databases (PubMed and Web of Science). To achieve the maximum sensitivity of the search strategy, a search string was used ("(complex regional pain syndrome OR re ex sympathetic dystrophy OR Sudeck's atrophy) AND (pediatric) AND (treatment)").

Selection criteria
The reference lists of all retrieved articles were reviewed for further identi cation of potentially relevant studies, and the articles were assessed using the inclusion and exclusion criteria. The following inclusion criteria were used when screening titles and abstracts: a. Studies of any level of evidence.
b. Studies written in the English-language. c. Studies reporting clinical or preclinical results. d. Studies on the treatment of complex regional pain syndrome type 1.
The exclusion criteria were as follows: 1. Review articles.
3. Articles written in other languages. 4. Diagnosis or differential diagnosis of complex regional pain syndrome type 1.
Duplicates and articles on other topics, with poor scienti c methodology, or without an accessible abstract were not included in the study; abstracts, case reports, conference presentations, editorials and expert opinions were also excluded.

Data extraction and criteria appraisal
All data were extracted from article texts, tables and gures. Three investigators (SM, VF and DMF) independently reviewed each article. Discrepancies between the two reviewers were resolved by discussion and consensus. The nal results and any remaining controversy on the reviewed article were reviewed and discussed with the senior investigators (VA and CA), who served as independent reviewers and assessed study quality. Con icts about data were resolved by the senior surgeon (PV). Reference lists from the selected papers were also screened. The PRISMA owchart for the selection and screening method is provided in Fig. 1.

Risk of bias assessment
A risk of bias assessment of all selected full-text articles was performed according to the Newcastle-Ottawa Quality Assessment Scale Cohort Studies (NOS) [9]. The NOS contains eight items, categorized into three dimensions including selection, comparability, and depending on the study type, outcome (cohort studies) or exposure (case-control studies). For each item, a series of response options is provided. A star system is used to perform a semi-quantitative assessment of study quality, such that the highest quality studies are awarded a maximum of one star for each item with the exception of the item related to comparability that allows the assignment of two stars. The NOS ranges between zero and nine stars. The assessments were performed by two authors (VA and CA) independently. Any discrepancy was discussed with the senior investigator for the nal decision. All the raters agreed on the nal result of every stage of the assessment (Appendix). In the systematic review, studies classi ed with more than six stars were included.

Study Selection
From the search of PubMed and Web of Science, 264 articles were included in the review, and 240 studies were selected after duplicate exclusion. Following the inclusion and exclusion criteria, the rst screening was performed. A total of 103 papers were considered eligible. Finally, after the full-text reading, reference list check and risk of bias assessment, 6 studies were included. A PRISMA [8] owchart of the method of selection and screening is provided (Fig. 1). The main focus of the included studies was related to physical (PT), cognitive behavioural (CBT) and pharmacological (PhT) treatments. A summary of the results is provided in Table 1.

Physical Therapy and Cognitive Behavioural Treatment
Three included studies contained a combination of physical therapy and cognitive treatment. Sherry et al [10] reported the complete resolution of pain symptoms in 74.7% of the sample. Seven subjects did not have remission. One child was dysfunctional with CRPS pain, and 5 had persistent mild pain but were fully functional. The authors highlighted suicide attempts (p = 0.026), an eating disorder (p = 0.028), reporting less pain initially (p = 0.021), and scoring higher on the Brief Symptom Inventory subsets for depression and paranoid ideation (p 0.037 and 0.048, respectively) as predictors of recurrence. Lee et al. [11] divided patients into 2 different groups: PT + CBT for 3 weeks vs PT + CBT for 6 weeks. Both the cohorts showed improvement in all pain and physical functioning outcome measures with short and long follow-up, without differences in pain scores, recurrent episodes of CRPS, or participation in school or activities.
Logan et al. [12] assessed 56 patients aged 8-18 years at admission and at discharge, and every parameter evaluated had statistically signi cant improvements. Thirty-two percent of patients required an assistive device at admission, while none required one at discharge

Pharmacological Treatment
Petje et al. [13] assessed functional outcome in patients treated with an intravenous infusion of iloprost, a prostacyclin analogue. The drug was administered 6 hours per day on 3 consecutive days. Among the side effects noted were headache on the rst day of infusion in all patients and ushing, and vomiting on the second day in three patients. A decrease in systolic blood pressure of an average of 7 mm Hg (5-15 mm Hg) in the rst 30 minutes after administration was detected in almost the entire cohort. Improvement of the visual analogue scale (VAS) score was found (p < 0.05). Relapse of CPRS was experienced by two patients, the rst after 3 months. Brown et al. [14] analysed the outcome of 29 patients refractory to PT + CBT. Fourteen subjects underwent amitriptyline administration, and 15 underwent gabapentin administration. After the 6-week trial, both cohorts showed improvement in pain symptoms, sleep disturbances and functionality. No statistically signi cant differences were found (p = 0.77) between the drugs. Similar adverse events were recorded (p = 0.77). Donado and colleagues [15] investigated the use of continuous regional anaesthesia (epidural or peripheral catheter) in subjects refractory to PT + CBT. Their data showed signi cant changes between admission and discharge for pain (p < 0.0001), without signi cant changes throughout the 4-month period after admission (p > 0.05).

Discussion
CRPS is a common disorder not completely understood, with no data available regarding the incidence of paediatric CRPS [5] because the diagnosis is uncertain and underestimated. Clinical evaluation, including a neurologic exam, combined with a thorough history collection is mandatory [4] to rule out other possible reasons for chronic pain, such as orthopaedic, neurological and rheumatologic conditions [5]. Early diagnosis is as important as or more than treatment; in fact, a longer disease course and sequelae [2] are associated with late identi cation. Unfortunately, no speci c diagnostic tools have been developed for children and adolescents, so the adult criteria are used [5]. Orthopaedists have a key role in the recognition of the disease due to very little evidence, no common consensus among the physicians, and a lack of guidelines [6]. As reported by Berde and Lebel [16], often the choice of treatment may vary according to the experience and resources of the clinician. Several treatments have been described [2], including acupuncture, transcranial magnetic stimulation, and invasive procedures, but the e cacy has been proven for the combination of physical and cognitive behavioural therapy and some pharmacological treatments only. The most established treatment is a programme of physical rehabilitation and cognitive behavioural therapy [15]. The goal is to restore normal function, increase the joint motion range, load tolerance and strength, and concurrently assist the child in accepting and managing the pain [2]. Different protocols were illustrated in selected studies, highlighting the absence of a standard treatment protocol. Sherry and colleagues [10] suggested aerobic training and progressive resistive exercises, in addition to hydrotherapy, desensitization with towel rubbing, hand massage, textured fabric rubs, and contrast baths (2 °C and 38 °C). During the patient's hospitalization, 5-6 hours of daily exercise therapy and 45 minutes to 3 hours of home exercise programmes (HEPs) were performed. In the Logan et al. trial [12], the patients underwent open-chain and closed-chain activities and an individualized HEP, and each child's functional goals, such as playing a speci c sport, were incorporated into the physical schedule. In addition, this multidisciplinary rehabilitation approach addresses the entire pain experience, incorporating desensitization, exposure to feared activities, skills for coping with pain, and changes to social responses to pain. Lee et al. [11] designed a protocol including transcutaneous electrical nerve stimulation, progressive weight bearing, tactile desensitization, massage, contrast baths and an HEP. Six weekly sessions of individual CBT incorporating pain management strategies, including relaxation training, deep breathing exercises, biofeedback, and guided imagery, were also included. Patient compliance, nurse care and parent treatment programmes [17] are mandatory to promote successful remission from pain and restoration of functional ability. Despite a rigorous rehabilitation programme, Sherry et al. described their patients as a motivated and eager to please sample [10]. Lee et al. [11] recorded compliance varying between 78 and 82%. No adverse events were recorded in the three studies, but the remission rate varied between 79% and 100% [10][11][12]. Several authors have investigated the role of PT + CBT in paediatric CRPS type 1, especially the brain and neurological changes and treatment action on the central nervous system. Frot et al. found evidence of emotional integration of pain in CRPS patients [18]. Lebel et al. [19] concluded that some changes in the brain persist, especially in the amygdala and basal ganglia, even after symptomatic recovery [20]. Diers et al. [21] demonstrated that behavioural extinction training reduces the emotional involvement in processing painful stimuli and induces a shift to a more sensory-discriminative way of pain processing post-treatment. Kregel et al. [22] emphasize that conservative treatments for patients with chronic musculoskeletal pain may induce both functional and morphological changes in predominantly prefrontal brain regions. For these reasons, non-invasive treatments are often recommended, even in recurrent forms [2,4]. On the other hand, the literature presents evidence of good outcomes after intravenous infusion of drugs and regional nerve blockades. Three pharmacological trials were selected in the study, and different molecules were investigated. Petje et al. [13] assessed the outcome of iloprost intravenous infusion, an analogue of prostacyclin, which induces transitory complete sympathicolysis and avoids the anxiety associated with a lumbar sympathetic blockade. Despite the good rate of response, relevant adverse reactions such as headache, ushing, vomiting, and a decrease in systolic blood pressure were recorded in all cohorts; consequently, the same authors do not suggest iloprost as primary therapy.
Other drugs proposed for treating neuropathic pain were gabapentin and amitriptyline. The rst avoids the release of neurotransmitters acting on voltage-gated calcium channels [23], and the latter e neuropathic reinforces the serotonin transporter [24]. Brown et al. compared the two molecules in a refractory PT + CBT schedule. The series revealed that both drugs are effective in reducing pain scores and improving sleep, without signi cant differences. However, ventricular conduction abnormalities were noted in the gabapentin group, while amitriptyline was linked to QT prolongation, torsade de pointes and sudden cardiac death. For these reasons, the use of both drugs in selected patients and with proper monitoring may be considered. Several invasive options have been proposed, including the use of continuous regional anaesthesia with epidural or peripheral catheters, which demonstrate a reduction in pain score and improvement in function score at short-and long-term follow-up. On the other hand, in the Donado et al.
series [15], 39% of the sample did not clinically improve the pain symptoms, and 43% had no functional advantages. Nevertheless, the authors suggested the treatment in addition to an active PT + CBT protocol. All the studies included in the systematic review emphasized the utility of PT + CBT, even when additional approaches were undertaken. The comparison of management with versus without rehabilitation was considered ethically unacceptable [2]; however, the outcome can be not related to a single treatment, and the results have been in uenced by conservative treatment even in pharmacologic protocol studies. Future research directions should focus on the identi cation of disease onset mechanisms and the development of more de ned, proper and easy-to-use diagnostic tools. The design of high-quality, prospective, large- cohort, long-term follow-up studies is strongly encouraged, as is design of a speci c assessment score.
The heterogeneity of the scores utilized in the objective clinical assessment of the patients, the absence of standard protocols, and the lack of randomized, blinded prospective trials are the main limits in the comparison of study results.

Conclusion
In conclusion, conservative treatment (physical, occupational, and cognitive behavioural therapy) is the recommended option in the management of paediatric CPRS. Pharmacological therapy should be reserved for refractory and selected patients. The design and development of standard protocols are strongly suggested.

Abbreviations
Complex regional pain syndrome = CRPS; re ex sympathetic dystrophy = RSD; Preferred Reporting Items for Systematic Reviews and Meta-Analyses = PRISMA; Newcastle-Ottawa Quality Assessment Scale Cohort