Efficacy of stellate ganglion interventions for complex regional pain syndrome in the upper limb—A systematic review and meta-analysis

ABSTRACT

Background and Aims:

Stellate ganglion (SG) interventions for treating upper limb complex regional pain syndrome (CRPS) have been reported in studies. However, more substantial evidence is required to reach a consensus on its analgesic efficacy. To the best of our knowledge, no systematic review demonstrating the effectiveness of SG intervention for upper limb CRPS has been reported. Hence, this meta-analysis was done to ascertain the efficacy of SG intervention in managing patients with upper limb CRPS.

Methods:

A database search of PubMed, Cochrane, Embase, Scopus and Google Scholar was done for articles published between January 2001 and December 2021. Two independent reviewers extracted data from the included studies, and the studies were evaluated for any potential risk of bias, and a meta-analysis was performed.

Results:

Eight studies were included in the qualitative synthesis, four were randomised controlled trials (RCTs), and the rest were non-RCTs. Six studies were assessed quantitatively. A significant reduction in pain scores post-intervention, up to two weeks, and four to 28 weeks after the procedure was observed, although the included studies exhibited marked heterogeneity. Qualitative analysis of these studies revealed an overall improvement in disability scores and functionality as assessed by an improved range of motion. Most studies reported no complications or only short-term minor complications after the procedure.

Conclusion:

Stellate ganglion interventions improve pain and disability scores with self-limiting short-term complications and no long-term complications. However, further studies with a large sample size are required to validate this treatment modality.

INTRODUCTION

Complex regional pain syndrome (CRPS) is a debilitating condition that affects extremities, usually after an injury, with pain persisting disproportionate to the inciting event and not limited to the site of injury. The term CRPS includes earlier diagnostic terms like reflex sympathetic dystrophy, sudeck dystrophy, causalgia, algodystrophy and reflex neuromuscular dystrophy.[1,2] The diagnosis of CRPS is usually clinical. The International Association of Study of Pain (IASP) criteria, also known as the Budapest criteria, is used to diagnose CRPS.[3,4] The pathophysiology of CRPS is still unclear, and several mechanisms have been proposed. Dysregulation of the sympathetic system, immune dysfunction, endothelial damage and tissue hypoxia have been postulated as the causative factors.[5]

The management of CRPS depends on a multidisciplinary approach utilising pharmacologic, interventional and neuromodulation therapies targeting physical and psychological pain.[6-8] Stellate ganglion (SG) interventions appear to work by blocking sympathetically mediated pain in the upper limb.[5] Additional convincing evidence is still required to obtain an agreement on its analgesic effectiveness. To our knowledge, no systematic review demonstrates the efficacy of SG intervention in upper limb CRPS. Therefore, this meta-analysis was conducted to determine the effectiveness of SG intervention based on the available literature and to summarise the current body of evidence that supports its use in treating patients with upper limb CRPS.

METHODS

Study design

The PROSPERO registration number for the study protocol is CRD42022297822. This systematic review and meta-analysis were reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) standards.

Eligibility criteria

This review comprised studies that examined the function of SG intervention in treating CRPS. The published research using a well-formulated Population, Intervention, Control, Outcome, Study (PICOS) design was included [Table 1]. The meta-analysis included studies that administered SG block to adult patients with CRPS in the upper limb. These studies were compared with a control or placebo group. The primary outcome was the improvement in pain scores. The influence on quality of life, duration of pain alleviation and potential consequences were considered secondary outcomes. The meta-analysis encompassed prospective randomised controlled trials (RCTs), non-randomised controlled trials, cohort studies and retrospective studies. Animal studies, publications not in English and articles whose full text was not accessible were excluded. Case studies, editorials, scoping reviews, literature reviews and conference abstracts were excluded. Children, patients with current infections, people with coagulopathies, people with abnormal bleeding, people with cancer pain and people with psychiatric conditions were excluded. Studies were only qualified if they used appropriately described study techniques and designs and provided sufficient data.

Table 1.

PICOS framework

	Determitants
Population	Adults with CRPS of upper limb
Intervention	Stellate ganglion intervention
Controls	Varies from study to study, compared to control groups or placebo group
Outcomes	Primary Objective
	Estimation of pain scores
	Secondary Objectives
	Effect on quality of life
	Duration of pain relief
	Complications
Study design	Prospective randomised and non-randomised controlled studies, cohort studies and retrospective studies

CRPS = Complex regional pain syndrome

Electronic literature search and study selection

The Cochrane Central Register of Controlled Trials (Central), Embase, Scopus, Google Scholar and PubMed databases were searched for publications published between January 2001 and December 2021. The timeframe for the search was chosen to mitigate the influence of methodological discrepancies on the overall findings; thus, the analysis is limited to a specific timeframe. This approach aims to minimise potential variations caused by different methods used in older studies. The terms ‘stellate ganglion intervention’, ‘stellate ganglion block’, ‘upper limb complex regional pain syndrome (CRPS)’ and ‘sympathetic block’ were used to conduct the database search [Supplementary Material 1]. Using various search phrase combinations, two reviewers (GP and BB) conducted an online literature search separately. The bibliographies and references of the chosen papers were manually screened. The complete text of the selected publications was examined once the abstract was examined and determined to be appropriate. A second reviewer chose the papers to be considered for the final analysis after independently evaluating and selecting them.

Data extraction

Two separate reviewers (GP and BB) examined the chosen publications. Utilising a standardised data extraction form, reviewers independently and completely extracted and summarised data from the reports. The following information was extracted from the studies: patient demographics, study characteristics (author, publication year, study design, sample size) and outcomes (pain score, effect on quality of life, analgesia duration and any complications related to the procedure) from the included studies that assessed the levels of pain at various follow-up times. When data were available at any follow-up interval, pooled analyses were performed. The means and standard deviation (SD) were taken from the studies where reported. For pain outcomes up to two weeks and four to 28 weeks after the intervention, data extraction for quantitative analysis was carried out, and the baseline mean was compared.

Quality assessment of individual studies

The Review Manager Software version 5.4.1 (2014, Cochrane Collaboration, Copenhagen, Denmark) was used to assess the risk of bias for the included RCTs. Two independent reviewers (GP and BB) evaluated random-sequence generation, allocation concealment, participant blinding, outcome assessor blinding, incomplete outcome data and selective reporting in individual studies to assess the included studies’ selection bias, performance bias, detection bias, attrition bias, reporting bias and other biases. By categorising the factors above as having a ‘high’, ‘low’ or ‘unclear’ risk of bias, the study’s internal validity was evaluated. The risk of bias (ROB) was assessed using the Cochrane ROB tool (the Cochrane Collaboration, Copenhagen, Denmark). The risk of bias tool for randomised studies (ROB 2) was used to assess the quality of RCTs and the risk of bias in non-randomised studies of interventions (ROBINS-I) was used to determine the quality of non-randomised research.

Primary outcome and secondary outcome

Change in pain score after SG Block (SGB) was the primary outcome. Missing data, namely pain scores at various time frames not mentioned in the study, were sought from the corresponding authors by emailing them. The secondary outcomes included intervention’s impact on quality of life, the length of pain alleviation and any adverse effects. Different studies used different measures to assess the impact on quality of life, so it was not feasible to carry meta-analysis of secondary outcomes.

Statistical analysis

The data were analysed using the RevMan 5.4.1 tool (Cochrane Collaboration, Copenhagen, Denmark, 2014). A random-effects model with an inverse variance strategy was employed. Standardised mean differences (SMDs) with 95% confidence intervals (CIs) were estimated for pain outcomes. SMD assists in expressing the magnitude of the intervention impact on the study’s observed variability. Additionally, it enables before-and-after comparisons that are not reliant on particular measurement units. The results are shown with 95% CI, P values and related forest plots. Egger’s test or a funnel plot-based publication bias was not performed. We only conducted a meta-analysis (quantitative synthesis) for pain outcomes because there was not enough data available for the other outcomes.

Assessment of heterogeneity

The I² statistic was used to express statistical heterogeneity between the studies. I², or the percentage of variability in the treatment estimates, is indicated by the following ranges of values: 0–40% for potentially inconsequential heterogeneity; 30–60% for moderate heterogeneity; 50–90% for significant heterogeneity; and 75–100% for substantial heterogeneity. Cochran’s Q statistics show considerable heterogeneity at a P value of 0.10.[9,10] Subgroup analysis was conducted to look for the source of heterogeneity.

Evidence quality as per GRADE Pro

The GRADE Pro Guideline Development Tool (Software, McMaster University and Evidence Prime, 2022, available from ‘www.gradepro.org’) was used to evaluate the quality of the reviewed evidence by considering factors like trial design, ROB, directness of outcomes, heterogeneity, precision within results, bias due to publishing, estimate effect, dosage relationship with response and confounders. The GRADE thus obtained may represent high, moderate, low or very low-quality evidence.[11]

RESULTS

Search results and Study selection

A database search was performed in January 2022, which resulted in 2348 studies. Of these, 1199 were duplicates, and the remaining 1149 records were finally screened. The eight articles[12-19] were included in the qualitative synthesis as outlined in the PRISMA flowchart [Figure 1]. Of these, six articles were included in the quantitative synthesis.[14-19]

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram

Study characteristics

The present review included different study designs, RCTs[12-15] and non-randomised trials (non-RCT prospective and retrospective studies).[16-18]

CRPS has been defined according to IASP or Budapest criteria across all studies. Studies used the Visual Analogue Scale (VAS), Numerical Rating Scale (NRS) and Numerical Pain Intensity Scale (NPIS) to evaluate pain intensity. The sample size ranged from 10[12] to 161[18] cases, and the follow-up period ranged from 1 week[12,13,17] to 28 weeks[15] across studies. The mean age of included patients ranged from 25[13] to 61 years,[19] and the mean pain score ranged from 4.4[19] to 9.94.[17] Table 2 shows details about demographic data and study characteristics.

Table 2.

Patient and study characteristics

Authors (year)	Study design	Number of patients	Age in years	CRPS defined according to IASP/Budapest	Intervention	Drug administered/Lesions made
Jadon 2016[17]	Prospective Non randomised	24	55±13.9	Budapest	Fluoroscopy guided Maximum three blocks at weekly intervals to achieve >50% pain relief	6 ml of 1% Lidocaine + 40 mg Methyl Prednisolone (Depot Preparation)
Matthias Schürmann 2001[16]	Prospective non-randomised	33	55	IASP	Blind	15 ml of 0.5% Bupivacaine+1% Prilocaine
Rashmi Datta 2017[18]	Prospective Nonrandomised	161	48.3±30.3	Budapest	Ultrasound blocks repeated after three weeks depending on the relief of symptoms	5 ml of 0.3% of Bupivacaine+4 mg of Dexamethasone
Ruben Alenakian 2020[19]	Retrospective	61	55.6±12.4	Budapest	Ultrasound-guided variable number of blocks until the achievement of >30% pain relief or no further relief	5 ml of 0.5% Lidocaine
Seung Doo Yoo 2012[14]	Prospective Randomised	44	Blind: 59.1±4.5, Ultrasound: 61.3±5.6	IASP	Ultrasound-guided and blind two blocks at intervals of 1 week	0.5% Lidocaine (10 ml) & 0.5% Lidocaine (5 ml)
William E. Ackerman 2006[15]	Prospective Randomised	25	37±1.24	IASP	Fluoroscopic-guided three blocks at a weekly interval	7 ml of 1% Lidocaine & 7 ml of 1% Lidocaine + Clonidine 1 ug/kg
MSA Nascimento 2010[13]	Prospective Randomised	44	25-50	IASP	Fluoroscopy-guided four blocks at a weekly interval	Group 1:70 mg of 1% Lidocaine Group II: 70 mg of 1% lidocaine plus 30 ug of clonidine
Farnad Imani 2016[12]	Prospective Randomised	14	Not mentioned	Not mentioned	Ultrasound and fluoroscopy	5 ml of 0.25% Bupivacaine and 1 ml of Triamcinolone 40 mg
Authors (year)	End point of successful block		Primary outcome		Conclusion	Complications
Jadon 2016[17]	Temperature rise (2°C)		VAS and DASH pre-procedure and one week after the procedure		SGB reduces pain and improves functional ability of the limb in patients with CRPS. The duration of the disease serves as an important factor in the success of treatment, and SGB should be considered as early as possible once other treatment modalities fail	No complication
Matthias Schürmann 2001[16]	Temperature rise (1.5°C)		VAS pre-procedure and immediately		Proof of sympathetically maintained pain based on pain relief after SG blockade is not conclusive.	No comments about complications
Rashmi Datta 2017[18]	Temperature rise (2°C)		NPRS and DASH pre-procedure, one hour post-procedure and 1 and 3 weeks post-procedure.		Serial SGBs attained an average reduction in pain by 3 NPRS. SGBs should be administered early during CRPS, and pharmacotherapy can be reduced subsequently.	Contralateral Horner syndrome and small pneumothorax were reported
Ruben Alenakian 2020[19]	Temperature rise (1.7±1.6°C)		NRS pre-procedure and within 28 days after the last blocks Pre-procedure and 20 min after the procedure		SGBs are safe and effective in reducing sympathetically maintained pain in patients with CRPS.	Haematoma, hoarseness and dysphagia
Seung Doo Yoo 2012[14]	Horner’s Syndrome		VAS Hand swelling Pre-procedure, two weeks and four weeks		Ultrasound-guided SGB decreased VAS values compared to the classical blind approach technique in Post-stroke CRPS patients. Ultrasound-guided SGB can reduce the volume of anaesthetic required	No adverse effects in the ultrasound group Two patients from the blind group developed a haematoma
William E. Ackerman 2006[15]	Temperature rise (2°C)		NPIS pre and post-block and two weeks after the last block and monthly for six months pre-procedure		Duration of symptoms greater than 16 weeks before the initial SGB.	Not mentioned
MSA Nascimento 2010[13]	Temperature rise (2.2°C)		VAS pre- and post-procedure after each block and one week after the last block		Progressive and significant reduction in pain scores and a significant increase in the duration of analgesia were observed in all groups.	Drowsiness, dry mouth
Farnad Imani 2016[12]	Not mentioned		Weekly pain attacks VAS pre- and post-procedure Pain Disability Index		Compared with fluoroscopic guidance, ultrasound is a safe and effective method with lower complications and improved patient disability scores.	Dysrhythmia, nasal congestion and hoarseness in the ultrasound-guided group.

cRPS=Complex regional pain syndrome; SGB=Stellate ganglion block; NRS=Numerical Rating Scale; VAS=Visual Analogue Scale; NPIS=Numerical Pain Intensity Scale; NPRS=Numerical Pain Rating Scale; DASH=Disabilities of arm shoulder and hand; IASP=International Association of Study of Pain

Risk of bias assessment

For randomised studies,[12-15] the ROB-2 tool was used, which indicated that included studies had a high risk of bias except for one study by Imani et al.[12] [Figure 2]. All RCTs[12-15] showed a high risk of bias resulting from problems in blinding participants and assessors. The study by Nascimento et al.[13] and Yoo et al.[14] showed a high risk of bias due to allocation concealment. Studies by Nascimento et al.[13] and Ackerman et al.[15] showed an increased risk of bias owing to problems in random sequence generation. Studies by Yoo et al.[14] and Ackerman et al.[15] showed a high risk of blinding of outcome assessment bias.[15]

Figure 2.

Risk of bias assessment for randomised controlled trials

The ROBINS-I tool was utilised for non-randomised studies [Figure 3].[16,18] It was concluded that the included studies were of moderate to high quality [Figure 3]. Studies by Jadon et al.[17] and Schürmann et al.[16] showed an increased risk of bias due to confounding and outcome measurement. The study done by Datta et al.[18] showed a high risk of bias in selecting research participants due to missing data. The study done by Aleanakian et al.[19] showed a high risk of bias due to confounding, selection of participants in the study, deviation from the intended intervention and that resulting due to missing data.

Figure 3.

Risk of bias assessment for non randomised controlled trials

RESULTS OF INDIVIDUAL STUDIES

Primary outcome

Comparison of pain scores

Change in pain score after the procedure is the primary outcome in most studies. We used a random effect model to pool results and obtain a weighted average and 95% CI. Due to considerable heterogeneity among the included studies, the random effect yielded a more comprehensive estimate of the confidence interval than the fixed effect.

A. Pooled results of all included studies[14-19] [Figure 4]

Figure 4.

Forest plot for pain outcome (a) Pooled data of all the studies; (b) Up to 2 weeks post-procedure; (c) 4 weeks to 28 weeks post-procedure. CI=Confidence interval; SD= Standard deviation, SMD=Standard mean difference; SGB=Stellate ganglion block

• Six included studies were assessed for the overall effect of the change in pain score after SGB.[14-19] The pooled result of all studies indicated a decrease in pain scores after SGB as compared to baseline score with SMD 2.33 95% CI [1.24,3.42], P < 0.0001. The result showed a significant difference in all studies with SMD [95% CI] 2.33 [1.24,3.42] (P < 0.001) and I² = 96% compared to baseline pain scores. The included studies were heterogeneous (I² of 96%, P < 0.001).[14-19] Sensitivity analysis was done but did not result in any significant decrease in heterogeneity.

B. Pooled results of studies up to two weeks[14,16-18] [Figure 4]

• Four studies were assessed for change in pain score after SGB[14,16-18] Compared to the baseline, it revealed a considerable reduction in pain scores with SMD [95% CI] 2.33 [1.24,3.42] (P < 0.001). The studies were heterogeneous (P < 0.001, I² = 97%). A sensitivity analysis was done, removing the study by Jadon et al.[17] resulted in I² = 57%.

C. Pooled results of studies at four to twenty-eight weeks[14,15,19] [Figure 4]

• The change in pain score after SGB showed a significant decrease with SMD [95% CI] 2.10 [0.69,3.51] (P = 0.004) compared to baseline pain scores in three studies.[14,15,19] The included studies were heterogeneous P < 0.001 and I² = 95%. A sensitivity analysis was done, and removing the study by Aleanakian et al.[19] yielded I² = 0%.

Secondary outcome

Jadon et al.,[17], in their study, evaluated the DASH (disabilities of arm, shoulder, and hand) score immediately post SG block and found it to be significantly decreased from 93 to 25 (P < 0.001). Datta et al.,[18] in their study, saw a decrease in the mean DASH score from 53 to 10.4, which was significant P = 0.005.

Pain Disability Index (PDI) was significantly decreased in two studies by Imani et al.[12] and Ackerman et al.[15] (P < 0.05). Imani et al.[12] compared PDI in ultrasound and fluoroscopic groups and found it to be reduced more in the USG group at six-month intervals, but the difference between groups was not significant P = 0.05. The study by Ackerman et al.[15] indicated a decrease in Beck’s depressive inventory score post-procedure.

Duration of pain relief

There was a wide variation in the duration of pain relief. Most of the studies showed pain relief from 2 weeks to 4 weeks. Two studies by Imani et al.[12] and Ackerman et al.[15] demonstrated a meaningful reduction of pain scores up to 6 months.

Complications

Complications ranged from haematoma, dysphagia, contralateral Horner’s syndrome, pneumothorax and dysrhythmia. However, two studies reported no complications.[14,17] The study by Datta et al.[18] reported a complication rate as high as 11.1%, whereas, for another study by Aleanakian et al.,[19] it ranged from 0.3% to 3.1%. No long-term complications were documented in any of the studies included in the analysis.

Subgroup analysis

Subgroup analysis was carried out to identify the causes of heterogeneity related to study design (RCTs vs. non-RCT) and time to follow-up (up to two weeks vs. four weeks to 28 weeks) [Figure 5]. The results are presented in Table 3. Subgroup analysis comparing RCTs vs. non-RCTs yielded SMD [95% CI] 1.53 [0.69,2.37] P < 0.001 with an I² = 0% for subgroup difference. Time to follow-up, when compared, they yielded SMD [95% CI] 1.52 [0.82,2.22] P < 0.001 and I² = 0% for subgroup difference. The subgroup analysis results revealed that neither the study design nor the follow-up were reasons for heterogeneity.

Figure 5.

Subgroup analysis (a) Subgroup analysis of RCT vs. Non-RCT; (b) Subgroup analysis 2 weeks vs 4 weeks follow-up. SD=Standard deviation; CI=Confidence interval, SMD=Standard mean difference; SGB=Stellate ganglion block; RCT= Randomised controlled trial; Non-RCT= Non-randomised controlled trial

Table 3.

Subgroup Analysis for ascertaining heterogeneity

Subgroup	Number of studies	Standard Mean difference 95%CI	I 2	P
Study Design
RCT	2	1.40 [1.02,1.78]	0%	0.45
Non-RCT	4	1.92 [0.64, 3.21]
Follow-up Period
2 weeks	2	1.40 [1.02,1.78]	0%	0.75
4 weeks	2	1.68 [0.01,3.34]

RCT=Randomised controlled trial; CI=Confidence interval

GRADE analysis using GRADE Pro GDT

The GRADE Pro GDT recommendation for the primary objective was ‘Very Low’ evidence quality. There were serious issues with the ROB of included RCTs.[14,15] There were inconsistencies in the assessment of non-RCT studies.[16-19] Figure 6 displays the outcomes of evidence quality according to GRADE Pro’s guiding principles.

Figure 6.

Grade analysis using GRADE Pro Guideline Development Tool (GDT). CI=Confidence interval; SMD=Standardised mean differences; RCT=Randomised controlled trial; Non-RCT=Non-randomised controlled trial a: heterogenous results; b: allocation of participants and blinding was unclear

DISCUSSION

The finding of this review is that there is a significant reduction in pain scores post-intervention, up to 2 weeks, and 4 to 28 weeks after the procedure. However, the studies were heterogeneous, and the quality of evidence could have been higher.[14-19] Qualitative analysis of these studies revealed an overall improvement in disability scores and functionality as assessed by an improved range of motion.[14-19] However, the lack of data limited quantitative analysis of the secondary outcomes.

Different studies have used a varying number of blocks, with three studies utilising a fixed number of blocks at weekly intervals.[13-15] In contrast, two studies continued giving blocks until there was greater than 50%[17] and 30%[19] pain relief, respectively. Only two studies utilised a single block.[12,16] The endpoint for a successful block in most studies was a temperature rise of 1.5–2°C varying from study to study. No consensus could be drawn regarding the optimal number of blocks for CRPS.

CRPS can be a debilitating condition affecting every aspect of the patient’s life, including psychological well-being. Pain and disability are directly or indirectly related to the patient developing psychological problems like depression.[8] Ackerman et al.[15] showed improvement in Beck’s depression inventory score post-SG block. No other study assessed improvement in the psychological aspects of the patients.

Complications associated with this intervention are usually short-term. In their retrospective study, Imani et al.,[12] reported dysrhythmia and nasal congestion in the fluoroscopy group. Yoo et al.[14] reported no complications with ultrasound-guided intervention compared to fluoroscopy. Datta et al.[18] reported a complication rate of up to 11.1%. However, all the complications were short-term and transient. No long-term complications were seen in any of the studies. Image-guided blocks like fluoroscopy or ultrasound are always preferred over blind techniques for better localisation and preventing damage to surrounding vital structures.

To identify the sources of heterogeneity, subgroup analysis was carried out with respect to study design (RCTs vs. non-RCT) and time to follow-up (up to 2 weeks vs four weeks to 28 weeks). However, it was found that both the study design and time to follow-up were not the factors associated with heterogeneity. Overall, GRADE Pro recommended a ‘Very Low’ quality of evidence for both RCTs[14,15] and non-RCTs[16-19] because of the ROB and inconsistencies in assessment, respectively. This suggests that there is a strong likelihood that subsequent research will significantly affect our data and change the estimate of effect.

This review has several limitations. Most of the studies included were observational or non-randomised trials, and the included RCTs had a high ROB due to their poor design. Another drawback of the review is the significant heterogeneity. The included studies only offer a little verifiable evidence because of their inconsistent outcome measures, inconsistent time lengths and inconsistent study designs. Despite its limitations, this systematic review provides an updated literature summary on SG intervention for CRPS. Even though the method seemed efficient, secure and well-tolerated, the quality of the evidence that is currently accessible is still subpar. It is the need of the hour to conduct more research, ideally large-scale RCTs, to clarify its effectiveness in treating this condition.

CONCLUSION

This review concludes that low-quality evidence exists to have a positive outcome regarding pain and disability scores after SG intervention. However, there was considerable heterogeneity among studies, and the level of evidence recommended by GRADE analysis is very low. There is a scope for further studies like well-designed RCTs with large sample sizes to validate this treatment modality.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

SUPPLEMENTARY MATERIAL 1

A. PubMed

#1. (complex regional pain syndrome) AND (stellate ganglion) ((“complex regional pain syndromes”[MeSH Terms] OR (“complex”[All Fields] AND “regional”[All Fields] AND “pain”[All Fields] AND “syndromes”[All Fields]) OR “complex regional pain syndromes”[All Fields] OR (“complex”[All Fields] AND “regional”[All Fields] AND “pain”[All Fields] AND “syndrome”[All Fields]) OR “complex regional pain syndrome”[All Fields]) AND (“stellate ganglion”[MeSH Terms] OR (“stellate”[All Fields] AND “ganglion”[All Fields]) OR “stellate ganglion”[All Fields])) AND (2001/1/1:2021/12/31[pdat])

#2. ((crps) AND (stellate ganglion)) AND (upper limb) (“crps”[All Fields] AND (“stellate ganglion”[MeSH Terms] OR (“stellate”[All Fields] AND “ganglion”[All Fields]) OR “stellate ganglion”[All Fields]) AND (“upper extremity”[MeSH Terms] OR (“upper”[All Fields] AND “extremity”[All Fields]) OR “upper extremity”[All Fields] OR (“upper”[All Fields] AND “limb”[All Fields]) OR “upper limb”[All Fields])) AND (2001/1/1:2021/12/31[pdat])

#3. ((crps complex regional pain syndromes[MeSH Terms]) AND (ganglia, stellate[MeSH Terms])) AND (upper limb[MeSH Terms]) (“complex regional pain syndromes”[MeSH Terms] AND “stellate ganglion”[MeSH Terms] AND “upper extremity”[MeSH Terms]) AND (2001/1/1:2021/12/31[pdat])

#4. (reflex sympathetic dystrophy[MeSH Terms]) AND (ganglia, stellate[MeSH Terms]) (“reflex sympathetic dystrophy”[MeSH Terms] AND “stellate ganglion”[MeSH Terms]) AND (2001/1/1:2021/12/31[pdat])

#5. (crps) AND (stellate ganglion) (“crps”[All Fields] AND (“stellate ganglion”[MeSH Terms] OR (“stellate”[All Fields] AND “ganglion”[All Fields]) OR “stellate ganglion”[All Fields])) AND (2001/1/1:2021/12/31[pdat])

#6. (complex regional pain syndromes[MeSH Terms]) AND (ganglia, stellate[MeSH Terms]) (“complex regional pain syndromes”[MeSH Terms] AND “stellate ganglion”[MeSH Terms]) AND (2001/1/1:2021/12/31[pdat])

B. Google

“upper extremity” or “ upper limb” AND “complex regional pain syndrome” or “CRPS” AND “stellate ganglion block “ or “stellate ganglion intervention”.

C. Scopus

#1 (ALL (reflex AND sympathetic AND dystrophy) AND ALL (stellate AND ganglion)) AND PUBYEAR > 2000 AND PUBYEAR < 2022 AND PUBYEAR > 2000 AND PUBYEAR < 2022

#2 (ALL (complex AND regional AND pain AND syndrome) AND ALL (stellate AND ganglion)) AND PUBYEAR > 2000 AND PUBYEAR < 2022 AND PUBYEAR > 2000 AND PUBYEAR < 2022

#3 (TITLE-ABS KEY (complex AND regional AND pain AND syndrome) AND TITLE-ABS KEY (stellate AND ganglion)) AND PUBYEAR > 2000 AND PUBYEAR < 2022 AND PUBYEAR >2000 AND PUBYEAR < 2022

#4 (TITLE-ABS-KEY (crps) AND TITLE-ABS-KEY (stellate AND ganglion)) AND PUBYEAR > 2000 AND PUBYEAR < 2022 AND PUBYEAR >2000 AND PUBYEAR < 2022

D. Cochrane

ID Search Hits

#1 (crps) AND (“stellate ganglion block”) with Cochrane Library publication date Between Jan 2001 and Dec 2021 22

#2 (“complex regional pain syndrome”):ti, ab, kw AND (“stellate ganglion block”):ti, ab, kw withCochrane Library publication date Between Jan 2001 and Dec 2021 26

#3 (“reflex sympathetic dystrophy”) AND (“stellate ganglion block”) with Cochrane Library publication date Between Jan 2001 and Dec 2021 11

#4 (crps):ti, ab, kw AND (“stellate ganglion block”):ti, ab, kw with Cochrane Library publication date Between Jan 2001 and Dec 2021 19

E. Embase

(‘Crps’:ti, ab AND ‘stellate ganglion block’:ti, ab) (‘complex regional pain syndrome’:ti, ab AND ‘stellate ganglion block’:ti, ab) (‘reflex sympathetic dystrophy’:ti ab AND ‘stellate ganglion block’:ti, ab)