Comparison of extracorporeal shockwave therapy, ultrasound therapy, and corticosteroid injections for treatment of lateral epicondylitis: an umbrella review of meta-analyses

Abstract

Background

The purpose of this study was to assess the methodological quality of meta-analyses (MAs) and resolve evidence inconsistencies by quantifying overlap in primary studies, thereby providing enhanced evidence on the efficacy of extracorporeal shockwave therapy (ESWT) versus placebo, ultrasound therapy, and corticosteroid injections for lateral epicondylitis.

Methods

We conducted searches in four databases: PubMed, Embase, Cochrane Library, and Web of Science, until August 2024. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2) graded the quality and reliability of the MAs, and the quality of outcomes was graded by Grading of Recommendations Assessment, Development, and Evaluation (GRADE). Graphical Representation of Overlap for OVErviews (GROOVE) was applied to analyze overlap and classified the resulting evidence into four categories (I–IV) on the basis of evidence classification criteria.

Results

A total of nine MAs were included for analysis: five had a high AMSTAR 2 rating, three had a moderate AMSTAR 2 rating, and one had a low AMSTAR 2 rating. GROOVE analysis revealed substantial overlap, informing evidence classification. ESWT can effectively reduce the pain assessed by the visual analogue scale (VAS) compared with placebo (MD = −0.68; 95% CI −1.06, −0.3; P = 0.0004; I² = 75%). Compared with ultrasound therapy, ESWT has a significantly large reduction in the level of pain after the treatment at 1-month follow-up (MD = −1.42; 95% CI −2.14, −0.7; P = 0.0001; I² = 92%) and 3-month follow-up (MD = −1.65; 95% CI −1.81, −1.49; P < 0.00001; I² = 98%). ESWT is better than corticosteroid injection when calculating the pooled effect size of VAS (SMD = 1.13, 95% Cl 0.72, 1.55; P < 0.00001; I² = 0). ESWT also has a significant difference in the rate of 50% reduction in pain (RR = 1.38; 95% CI 1.09, 1.75; P = 0.008; I² = 41%). However, compared with placebo, it has no clinically important difference of grip strength (MD = 3.33; 95% CI 0.93, 5.73; P = 0.007; I² = 30%), and the pain score of Thomsen test (MD = −3.22; 95% CI −14.06, 7.62; P = 0.56; I² = 69%).

Conclusions

ESWT has a significant difference in reducing pain evaluation and relief of pain symptoms, and the effect is better than ultrasound therapy and corticosteroid injections.

Level of evidence I.

This protocol has been registered in the PROSPERO database

CRD42024586419

Supplementary Information

The online version contains supplementary material available at 10.1186/s10195-025-00871-w.

Introduction

Lateral epicondylitis (LE), commonly referred to as tennis elbow, is characterized by pain and tenderness localized around the common extensor origin of the elbow [1]. This condition is frequently linked to repetitive motions, particularly those involving racket sports, with 50% of tennis players experiencing at least one episode of LE during their careers [2]. However, LE is not confined to athletes; it also affects individuals across various professions and activities. Its prevalence in the general population ranges from 1–3%, accounting for approximately 7 out of every 1000 primary care consultations annually [3].

Currently, there is no universally established treatment protocol for LE. Management strategies span a spectrum from medical interventions, such as pharmacological treatments and surgical procedures, to physical therapy modalities, including therapeutic exercises, manual therapy, and other noninvasive techniques. Broadly, these approaches can be categorized into surgical and nonsurgical treatments [4]. Research indicates that nonsurgical interventions effectively resolve symptoms in approximately 90% of LE cases [5], with ultrasound therapy and corticosteroid injections being the most frequently employed methods [6]. These treatments are widely adopted due to their ability to provide short-term pain relief. However, their utility is constrained by limitations such as transient efficacy [7], potential adverse effects, and underlying treatment principles [8].

Extracorporeal shock wave therapy (ESWT) is a noninvasive technique that employs focused acoustic waves to target specific areas of the body, promoting pain relief and tissue healing. Over the past 25–30 years, ESWT has gained recognition for its safety, ease of application, and high patient tolerance, making it a popular choice for managing various musculoskeletal disorders [9]. Nonetheless, its effectiveness in treating LE remains contentious, primarily due to the requirement for multiple treatment sessions and associated higher costs [10].

As a result, the role of ESWT in LE management continues to be a topic of debate. A rigorous assessment of the existing evidence is essential to provide healthcare professionals and patients with a clearer understanding of ESWT’s efficacy and safety profile. This review aims to evaluate the methodological quality of meta-analyses (MAs) on the subject, analyze the consistency of their findings, and offer comprehensive, accurate, and reliable evidence to guide the clinical application of ESWT in LE treatment, particularly compared with ultrasound therapy and corticosteroid injections.

Methods

An umbrella review of multiple MAs was carried out. This protocol has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42024586419) [11]. A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2) will be used to gauge the caliber of eligible systematic evaluations, and our review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards (Supplementary Material A). [12, 13].

Search methodology

We searched the following four databases until August 2024: PubMed, Embase, Web of Science, and Cochrane Library. The search was conducted using a combination of subject phrases and free words. The primary search terms that were utilized were “lateral epicondylitis,” “tennis elbow,” “extracorporeal shockwave therapy,” “meta-analysis,” etc. (Supplementary Material 1).

Eligibility criteria for studies

Included criteria: (1) patients had to be diagnosed with LE; (2) research adhered to the PRISMA guidelines [14]; (3) the MAs investigated ESWT in comparison with either a treatment-as-usual control group or a sham-control group, and if a study examined a combination of therapy modalities (e.g., ESWT combined with manual therapy), the control group was required to match one of the individual therapy components (e.g., ESWT alone or manual therapy alone); and (4) studies had to report mean, standard deviation, and number of participants at baseline and end of treatment.

Exclusion criteria: (1) nonsystematic reviews and MAs, including letters, protocols, and conference abstracts; (2) research with significant mistakes or low quality [15]; (3) evaluation methods that did not address the major outcome; and (4) participants included in the study had comorbidities and/or other joint diseases.

Discovery and processing of overlap

We must search for overlap between the original studies to lessen interference caused by data overlap if assessments of the same outcomes are found in our included MAs [16]. This is due to the potential overlap in the original data they encompass. The analysis not only provides the total counts of overlapping and non-overlapping preliminary investigations, but also graphically illustrates this overlap. We use Graphical Representation of Overlap for OVErviews (GROOVE) into three levels: low (< 5%), medium (5% to < 10%), and high (> 10%) to quantify and visualize primary study overlap across included meta-analyses (MAs). After three key steps of matrix construction [16], overlap quantification and graphic output, the overlap visualization was graded for evidence, and the specific results were shown in Supplementary Material 3.

To solve the existing overlap problem, we take the following measures: firstly, prioritize Cochrane reviews over non-Cochrane publications where overlap existed For overlapping non-Cochrane reviews: (1) when a significant degree of overlap exists among multiple non-Cochrane reviews, priority will be given to the outcomes from the meta-analysis with the highest AMSTAR 2 score. (2) If the AMSTAR 2 scores are consistent, preference will be granted to the most recently published MAs or those reporting on the largest number of randomized controlled trials (RCTs). (3) In instances of minimal or substantial overlap, findings from all relevant MAs will be considered.

Extracting data and evaluating quality

Two authors independently gather, filter, and evaluate the quality of the methodology and the data. If there is disagreement, the third author can assist in reaching a consensus through shared discussion [17]. From publications that met the eligibility requirements, data on authors, nations, the number of included RCTs, the number of patients in the ESWT and placebo groups, mean age, sex ratio, and outcomes were extracted.

The study focused on several key outcome indicators: visual analogue scale (VAS), grip strength, and pain score of Thomsen test.

AMSTAR 2 was utilized to evaluate the quality and reliability of the included MAs, while GRADE was employed to assess the level of evidence for each outcome, classifying it as high, medium, low, or very low. In addition, we categorize the resulting evidence into four categories on the basis of evidence classification criteria: I (convincing evidence), II (highly suggestive evidence), III (suggestive evidence), IV (weak evidence), and NS (not significant). Detailed criteria for evidence classification is found in Supplementary Material 2. [18].

Data integration

For data synthesis, we adopted a narrative synthesis method to combine findings from systematic reviews, while the outcomes of MAs were displayed in a tabular format.

Data was pooled using Revman 5.4.7 VAS was used for pain, while the mean difference (MD) and standardized mean difference (SMD) were used across related functional measures to maximize usage of available data. All pairwise comparisons are expressed with credible 95% confidence intervals. SMD of 0.5 was considered a clinically significant functional improvement. In addition, heterogeneity was assessed using the I² statistic with < 40% indicating low heterogeneity and > 75%, substantial heterogeneity. In the presence of low or absent heterogeneity, fixed effects models were used, and mixed effects models were used if heterogeneity was detected (² > 40%).

Results

Search results

We searched in four databases, PubMed, Embase, Web of Science, and Cochrane Library, and searched 49 MAs in total. After excluding 13 duplicate MAs, 10 MAs were eliminated by reading the title and abstract according to the article type and the criteria for inclusion and exclusion. Finally, we carefully read the full paper, again excluding 17 MAs, and finally including 9 MAs. Figure 1 depicts the literature screening procedure.

Fig. 1.

Preferred Reporting Items for Systematic reviews and Meta-analysis (PRISMA) study selection flow diagram

Study characteristics

The basic characteristics of MAs included in this study are listed in Table 1. The MAs incorporated were published between 2005 and 2023, and all relevant studies included by MAs were RCTs; 5 MAs included more than 10 RCTs [19–23], and the specific number of RCTs associated with each MA are detailed in Supplementary Material 3. In the nine MAs, five had more than 1000 patients in sample size [19, 20, 22–24], and the remaining four had less than 1000 patients [3, 21, 25, 26]. In addition, six MAs provided specific sex ratios [3, 20, 22–24, 26] and eight MAs provided the average age of patients [3, 16, 20, 22–26]; 5 MAs had a high AMSTAR 2 rating [3, 22, 23, 25, 26] while 4 had a moderate AMSTAR 2 rating [19–21, 24]. Additionally, the findings from the GROOVE analysis are presented in Supplementary Material 4. The characteristics of the reviews were summarized in Table 2.

Table 1.

Evidence classification criteria

Evidence class	Criteria
Class I	> 1000 cases (or > 20000 participants for continuous outcomes); statistical significance at P < 10−6 (random effects); no evidence of small study effects and excess significance bias; 95% prediction interval excluded null value; no large heterogeneity (I2 < 50%)
Class II	> 1000 cases (or > 20000 participants for continuous outcomes); statistical significance at P < 10−6 (random effects); largest study with 95% confidence interval excluding null value
Class III	> 1000 cases (or > 20000 participants for continuous outcomes) and statistical significance at P < 0.001
Class IV	Remaining significant associations with P < 0.05
Nonsignificant	P > 0.05

Table 2.

Basic information about study patients included

Study	Year	Country	No. of RCTs included	Sample size	Mean age	Sex ratio (female)	Outcome indicators	Databases searched
Rachelle Buchbinder [24]	2005	Australia	9	1009	44.7	49.95%	Overall pain, grip strength	Cochrane Library, Medline, Embase, Cinahl, Scisearch
Eli T [20]	2014	America	22	2280	47.2	35.14%	Risk ratio of overall improvement, patient-rated tennis elbow evaluation score, maximum grip strength	PubMed, CENTRAL
Christoph Weber [21]	2015	Germany	15	595	NA	NA	Pain during the last 24 h, pain during activity, pain during Thomsen test, pain during day and night, pain at isometric testing	PubMed, Embase, Cochrane Library
Chenchen Yan [3]	2019	China	5	233	44.3	57.51%	Pain reduction (VAS), grip strength, evaluation scores of elbow function	PubMed, Embase, Cochrane Library, SpringerLink
Yuan Xiong [25]	2019	China	4	237	45.0	NA	Pain reduction (VAS), grip strength (PSS, GSS)	Pubmed, Medline, Embase, Cochrane Library, SpringerLink, ClinicalTrials.gov, OVID
Chenxiao Zheng [26]	2020	China	9	715	47.3	51.33%	Main overall pain (VAS), 50% pain reduction, pain score of Thomsen test, grip strength, adverse events	PubMed, Embase, Web of Science, Cochrane Library
Seo Yeon Yoon [23]	2020	Korea	12	1166	46.9	53.43%	Pain reduction (VAS), improvement in grip strength	PubMed, Embase, CENTRAL
Gaowen Yao [22]	2020	China	13	1035	44.7	53.82%	Pain reduction (VAS), grip strength	PubMed, OVID, Embase, Cochrane Library, Web of Science
Lobat Majidi [19]	2023	Iran	45	2856	45.9	NA	Mean pain	PubMed, Scopus, Web of Sciences, CENTRAL, Research Registers of ongoing trials (ClinicalTrials.gov)

CENTRAL Cochrane Central Register of Controlled Trials, RCTs randomized controlled trials, NA not accessed

Results of umbrella review

Pain evaluation by VAS

In total, three MAs with high AMSTAR-2 score reported VAS [22, 23, 26]. We analyzed the overlap in MAs using the GROOVE tool and described the results in Supplementary Material 4. Due to the solution in the methodology, the results of Yao et al. [22] were considered to be the best evidence currently known. It showed that there is significantly lower VAS in the ESWT group (MD = −0.68, 95% CI −1.06, −0.3, P = 0.0004, I² = 75%). The results were graded as high and the evidence level was III (Table 2).

VAS (ESWT versus ultrasound therapy treatment)

Only one MA reported VAS on ESWT group versus ultrasound therapy group [3] with a high AMSTAR-2 score. The study by Yan et al. [3] showed that there is a significantly large reduction in the level of pain after the treatment at 1-month follow-up (MD = −1.42; 95% CI −2.14, −0.7; P = 0.0001; I² = 92%), while the difference in the pain relief between the treatment groups persisted at 3-month follow-up (MD = −1.65; 95% CI −1.81, −1.49; P < 0.00001; I² = 98%). The results were graded as high and the evidence level was IV (Table 2).

VAS (ESWT versus corticosteroid injection treatment)

Only one MA reported VAS on ESWT group versus corticosteroid injection group [25], with a high AMSTAR-2 score. The study by Xiong Y et al. [25] showed that ESWT is better than corticosteroid injection when calculating the pooled effect size of VAS (SMD = 1.13; 95% Cl 0.72, 1.55; P < 0.00001; I² = 0). The results were graded as high and the evidence level was IV (Table 2).

Grip strength

A total of three MAs reported grip strength [23, 24, 26], with two [23, 26] having a high AMSTAR-2 score and one [24] having a moderate AMSTAR-2 score. We analyzed the overlap in MAs using the GROOVE tool and described the results in Supplementary Material 4. Due to the solution in the methodology, the results of Yoon et al. [23] are considered to be the best evidence currently known. It showed that the grip strength after ESWT also revealed no clinically important difference (MD = 3.33; 95% CI 0.93, 5.73; P = 0.007, I² = 30%). The results were graded as moderate and the evidence level was IV (Table 2).

Grip strength (ESWT versus ultrasound therapy treatment)

Only one MA reported grip strength (ESWT group versus ultrasound therapy group) [3], with a high AMSTAR-2 score. The study by Yan et al. [3] showed that the ESWT group had a better recovery of grip strength compared with the ultrasound therapy group (MD = 0.75; 95% CI 0.22, 1.28; P = 0.006; I² = 66%) at 1 month after the treatment. Meanwhile, the difference in comparison between ESWT and ultrasound therapy group at 6 months after therapy revealed the same outcome (MD = 2.15; 95% CI 1.68, 2.63; P = 0.0001; I² = 44%). The results were graded as high and the evidence level was IV (Table 2).

Pain score of Thomsen test

A total of two MAs reported the pain score of Thomsen test [24, 26], with one [26] having a high AMSTAR-2 score and the other [24] having a moderate AMSTAR-2 score. We analyzed the overlap in MAs using the GROOVE tool and described the results in Supplementary Material 4. The results of the study by Zheng et al. [26] were more reliable by combining the methodology and time-bound nature of the studies included. In the study by Zheng et al. [26], there was no significant difference between ESWT and control in decreasing the pain score of Thomsen test (MD = −3.22; 95% CI −14.06, 7.62; P = 0.56; I² = 69%). The results were rated as high and the level of evidence was IV (Table 2).

A 50% improvement in overall pain

A total of two MAs reported 50% improvement in overall pain [24, 26], with one having a high AMSTAR-2 score [26] and the other having a moderate AMSTAR-2 score [24]. We analyzed the overlap in MAs using the GROOVE tool and described the results in Supplementary Material 4. The results of the study by Zheng et al. [26] were more reliable by combining the methodology. There was a significant difference between ESWT and control in the rate of 50% reduction in pain (RR = 1.38; 95% CI 1.09, 1.75; P = 0.008; I² = 41%). The results were rated as high and the level of evidence was IV (Table 2).

Evaluation scores of elbow function

Only one MA reported evaluation scores of elbow function [20], with a high AMSTAR-2 score. The study by Yan et al. [3] showed that there were no significant differences in the function scores between the treatment groups at 3-month follow-up (SMD = 1.49; 95% CI −0.44, 3.42; P = 0.13; I² = 95%), indicating that ESWT and ultrasound therapy have similar effects on functional improvement. The results were graded as high and the evidence level was IV (Table 2, 3).

Table 3.

Summary of characteristics of the meta-analysis

Study	Year	Type of metric (summary effect)	MA model	Effect	95% CI	Number of included studies	I2	GRADE	Evidence class
Results of meta-analysis of pain evaluation by VAS
Seo Yeon Yoon [23]	2020	MD	Random	−0.68	−1.17 to −0.19	12	71	High	IV
Gaowen Yao [22]	2020	MD	Random	−0.68	−1.06 to −0.3	14	75	High	III
Chenxiao Zheng [26]	2020	MD	Random	−4.23	−8.78 to 0.32	4	60	High	NS
Results of meta-analysis of pain evaluation by VAS (ESWT group versus US group)
Chenchen Yan [3]	2019	MD	Random	−0.08	−0.26 to 0.11	5	0	High	NS
Results of meta-analysis of pain evaluation by VAS (ESWT group versus CS group)
Yuan Xiong [25]	2019	SMD	Random	1.13	0.72 to 1.55	3	0	High	IV
Results of meta-analysis of grip strength
Seo Yeon Yoon [23]	2020	MD	Fixed	3.33	0.93 to 5.73	6	30	High	IV
Chenxiao Zheng [26]	2020	MD	Random	3.52	2.43 to 4.6	3	0	High	IV
Rachelle Buchbinder [24]	2005	SMD	Fixed	0.05	−0.13 to 0.24	3	59	Moderate	NS
Results of meta-analysis of grip strength (ESWT group versus US group)
Chenchen Yan a(1-month follow-up) [3]	2019	MD	Fixed	0.75	0.22 to 1.28	2	66	High	IV
Chenchen Yan b(6-month follow-up) [3]	2019	MD	Fixed	2.15	1.68 to 2.63	2	44	High	IV
Results of meta-analysis of pain score of Thomsen test
Chenxiao Zheng [26]	2020	MD	Random	−3.22	−14.06 to 7.62	3	69	High	NS
Rachelle Buchbinder [24]	2005	MD	Random	−9.04	−19.37 to 1.28	3	73	Moderate	NS
Results of meta-analysis of 50% improvement in overall pain
Chenxiao Zheng [26]	2020	RR	Fixed	1.38	1.09 to 1.75	4	41	High	NS
Rachelle Buchbinder [24]	2005	RR	Fixed	1.02	0.55 to 1.90	1	NA	Moderate	NS
Results of meta-analysis of evaluation scores of elbow function
Chenchen Yan [3]	2019	SMD	Random	1.49	−0.44 to 3.42	3	95	High	NS

Discussion

In recent years, numerous studies [8, 27–30] have explored the use of ESWT for the treatment of LE. However, the efficacy of ESWT in managing LE remains a subject of debate. To better elucidate the therapeutic effects of ESWT on LE, we compared ESWT with placebo groups, ultrasound therapy groups, and corticosteroid injection groups across various metrics, including VAS scores, grip strength, and Thomsen test pain scores. Our findings indicate that ESWT significantly reduces pain assessment scores and alleviates pain symptoms in patients with LE and that ESWT has no significant effect on grip strength improvement. In addition, another study [3], under the condition of long-term follow-up period, found that both ESWT and ultrasound therapy have a better response effect on grip strength, as well as a greater effect on functional improvement of elbow joint, suggesting that the role of ESWT in improving grip strength is limited or needs further research. Analysis demonstrated that the improvement effect of ESWT on Thomsen test pain scores and elbow function scores lacked clinical significance.

The significant pain relief shown by ESWT (VAS improvement) may be the main driver of functional improvement, as pain is the main inhibitor of activity. Reducing pain during grasping directly promotes greater strength generation and endurance. As for the functional outcomes, we analyzed that the improvement in grip strength between ESWT and placebo might not have reached the standard of clinical difference, but compared with ultrasound therapy, ESWT might have greater potential functional advantages in the long term and support ESWT as a feasible option for restoring hand function in patients with LE. As for the elbow function score, due to the wide confidence interval and high heterogeneity of this study, it highlights relatively large uncertainty. ESWT and placebo may have certain efficacy in restoring elbow function, but the variability of the included studies cannot be ignored. Even factors other than treatment may affect the functional outcome, and the conclusion needs to be evaluated carefully.

ESWT and ultrasound therapy share several similarities and differences in their effectiveness for alleviating pain and enhancing functional recovery in patients with LE. Both ESWT and ultrasound therapy have demonstrated significant efficacy in the treatment of LE, alleviating pain by enhancing blood circulation to the affected region. Evidence [5] indicates that both modalities improve pain and functional outcomes in patients, with Yalvaç et al. [31] concluding that ESWT is comparable to ultrasound therapy in managing LE. As a viable alternative intervention, ESWT exhibits similar efficacy to ultrasound therapy. However, earlier studies highlight notable differences between ESWT and ultrasound therapy in terms of pain reduction and grip strength recovery. For instance, Thiele et al. [32] emphasize that for chronic cases, a follow-up period exceeding 3 months—preferably 1 year—is essential to observe the pronounced benefits of ESWT, a finding consistent with our results. This aligns with the primary objective of this study. Regarding short-term outcomes, Özmen et al. [33] report that both ultrasound therapy and ESWT effectively reduce pain and enhance functional prognosis. Chen L. [34] corroborates these findings, further advocating for ESWT due to its safety and noninvasive nature. Nevertheless, in terms of functional improvement, Schroeder et al. [35] note that while ESWT is a safe and effective option for treating peripheral musculoskeletal disorders in athletes, yielding statistically significant short-term pain relief and grip strength enhancement, it does not significantly outperform other treatments in functional performance. This suggests that although ESWT may not offer superior functional benefits, its safety profile and sustained efficacy justify its use as a preferred alternative.

In recent years, the application of ESWT in the management of tendinopathies, including LE, has gained significant traction. ESWT is thought to facilitate tissue repair and regeneration in chronic tendinopathy through various mechanisms, such as neovascularization, cellular proliferation, and antiinflammatory effects [19]. Similarly, corticosteroid injections are recognized as an effective intervention for LE, primarily by reducing synovial blood flow, inhibiting leukocyte activity, suppressing cytokine and protease production, and modulating collagen synthesis, thereby exerting antiinflammatory and analgesic effects [36]. Cakar et al. [37] revealed that corticosteroid injections showed the most significant improvement in assessment of tennis elbow and grip strength measurements in the early stage. However, benefits decreased significantly after 30 days. Therefore, they concluded that the combined injection of autologous blood and corticosteroids (AB + CS) is the optimal treatment plan.

Comparative studies on the efficacy of corticosteroid injections and ESWT in treating chronic LE have demonstrated that both approaches are effective and safe. Regarding pain relief, Liu et al. [38] recently confirmed that both ESWT and injection therapy significantly alleviate LE-related pain, with ESWT showing superior outcomes in short- to medium-term grip strength recovery. In terms of functional recovery, research indicates that the therapeutic effects of LE treatments are time-dependent. Zhang L. et al. [17] observed that ESWT yields better short-term (1-month) results, whereas corticosteroid injections are more effective in the long term (3–6 months). Both treatments exhibit similar rates of mild adverse events, which holds clinical relevance for determining treatment duration and aligns with the conclusions drawn in this study. In light of these findings, Chen Q. et al. [39] recommend prioritizing ESWT, as most LE symptoms tend to resolve spontaneously within 12 months without intervention.

While ESWT demonstrated clear benefits for pain reduction, its impact on functional recovery showed greater complexity: As for grip strength, there is a paradox to discuss. Although ESWT showed statistically significant grip strength improvement versus placebo [23] (MD = 3.33, 95% CI 0.93, 5.73; P = 0.007), this fell below the established MCID threshold of 14% improvement for lateral epicondylitis [6]. Crucially, long-term data revealed divergent patterns: at 6-month follow-up, ESWT significantly outperformed ultrasound therapy in grip recovery [3] (MD = 2.15; 95% CI 1.68, 2.63; P = 0.0001), suggesting functional benefits may require extended observation windows to manifest. This delayed effect aligns with tendinopathy healing mechanisms where structural remodeling lags behind symptomatic relief [37].

As for Thomsen test limitations, no significant difference was observed in Thomsen test pain scores between ESWT and controls [26] (MD = −3.22; 95% CI −14.06, 7.62; P = 0.56). We posit that this stems from the test’s inherent limitations: as a provocation test measuring pain during resisted wrist extension, it primarily assesses nociception rather than functional capacity. Its binary output (painful or not) lacks sensitivity to capture graded functional improvements [29].

As for elbow function discrepancy, pooled analysis showed no difference in elbow function scores between ESWT and ultrasound therapy at 3 months [3] (SMD = 1.49; 95% CI −0.44, 3.42; P = 0.13). Notably, the extreme heterogeneity (I² = 95%) indicates fundamental differences in assessment tools across studies from disease-specific (e.g., Patient-Rated Tennis Elbow Evaluation) to generic instruments. This methodological variability obscures true treatment effects on function.

Recent studies [35] have highlighted the influence of ESWT energy flux density on pain relief and functional recovery in LE treatment. ESWT can be categorized into low energy (< 0.12 mJ/mm²) and high energy (≥ 0.12 mJ/mm²) on the basis of energy flux density, both of which are widely utilized for various tendinopathies. Li et al. [40]noted that pain reduction and treatment success rates correlate with energy intensity levels, with high-energy ESWT demonstrating the highest likelihood of being the optimal treatment within 3 months, followed by corticosteroid injections and low-energy ESWT. Conversely, Santilli et al. [41] reported that low-energy ESWT outperforms high-energy ESWT in managing chronic LE, offering more sustained pain relief and functional improvement. Nevertheless, research into the differential effects of ESWT energy levels remains ongoing. Future studies should emphasize larger sample sizes and randomized controlled trials to validate these findings and clarify the mechanisms underlying the variations in efficacy across different energy levels. For functional outcome indicators, it is necessary to clarify the relationship between pain relief and these indicators and determine the degree of influence of different functional outcome indicators. This will provide more solid evidence for clinical symptom-oriented treatments. In addition, ESWT protocols should be standardized to reduce the impact of deviations caused by it on the credibility of the research conclusions. In addition, the latest injection treatment methods, such as platelet-rich plasma (PRP) injection, have revealed through research [42] that there is currently only limited consensus on this therapy. The high-level user experience has not led to a convergence of opinions on the technical components of the PRP formulation and delivery. Therefore, further in-depth research on PRP should be conducted and compared with ESWT to clearly identify the advantages of different therapies.

The review has several limitations: (1) there was a risk of bias due to the inconsistent quality of the included MAs, mainly because most studies had sample sizes smaller than 1000 and demonstrated considerable heterogeneity; (2) several included MAs [19–21, 24] with incomplete data, preventing a comprehensive and unified analysis of the findings; (3) the review was limited to English-language literature, which may introduce selection bias; (4) the selection of RCTs may have some bias due to the different follow-up times; and (5) potential publication biases may distort our conclusions, including duplicate inclusion of pivotal RCTs (e.g., [24]) that may overrepresent early evidence and missing grey literature that could mask short-term functional limitations.

Conclusions

ESWT has a significant difference in reducing pain evaluation and relief of pain symptoms, and the effect is better than ultrasound therapy and corticosteroid injections.

Abbreviation

ESWT

Extracorporeal shockwave therapy

GRADE

Grading of recommendations assessment, development, and evaluation

GROOVE

Graphical representation of overlap for OVErviews

LE

Lateral epicondylitis

RCTs

Randomized controlled trials

VAS

Visual analogue scale

PRP

Platelet-rich plasma

Author contributions

P. Zhu: conception, design, data collection, processing, and manuscript initial drafting. P. Tang: data collection, processing, and manuscript initial drafting. Z. Deng: manuscript initial drafting. Y. Li: project funding acquisition and integrity oversight. All authors contributed to the interpretation of the data, provided critical revisions to the manuscript for intellectual content, and ultimately approved the final version.

Funding

This work was supported by National Key R&D Program of China (2023YFC3603400), Hunan Provincial Science Fund for Distinguished Young Scholars (2024JJ2089), National Key R&D Program of China (2019YFA0111900), National Natural Science Foundation of China (no. 882072506, 92268115, 82272611), National Clinical Research Center for Geriatric Disorders (Xiangya Hospital, grant nos. 2021KFJJ02 and 2021LNJJ05), National Clinical Research Center for Orthopedics, Sports Medicine and Rehabilitation (2021-NCRC-CXJJ-PY-40), Science and Technology Innovation Program of Hunan Province (no. 2021JJ31105), and the Key Discipline of Traditional Chinese Medicine of Zhejiang Province-Clinical Integration of Chinese and Western Medicine (TCM Orthopedics and Traumatology) of The First Affiliated Hospital of Wenzhou Medical University (no. 2024-XK-50).

Data availability

All data relevant to the study are included in the article or are available as supplementary files. There is no patient-identifiable data available.

Declarations

Patient involvement

Not applicable.

Competing interests

Financial Disclosure and Conflict of Interest. The authors affirm that they have no financial affiliation (including research funding) or involvement with any commercial organization that has a direct financial interest in any matter included in this manuscript, except as disclosed and cited in the manuscript. Any other conflict of interest (i.e., personal associations or involvement as a director, officer, or expert witness) is also disclosed and cited in the manuscript.