Short- and mid-term efficacy of extracorporeal shock wave therapy versus corticosteroid injections in the treatment of trigger finger: A non-randomized prospective comparative study

Abstract

Background and objective

Trigger finger is a common condition managed through various conservative approaches. This study aimed to compare the short- and mid-term efficacy of radial extracorporeal shock wave therapy (rESWT) versus corticosteroid injection (INF) in patients with moderate trigger finger.

Material and methods

A non-randomized prospective comparative study was conducted with 42 patients diagnosed with Quinnell grade II–III trigger finger, assigned to treatment groups (rESWT vs. INF) by sequential sampling. The following outcomes were assessed at baseline, 6 weeks, and 6 months: pain (VAS), functionality (QuickDASH and Roles & Maudsley), stiffness/severity (Quinnell), and grip strength (hand dynamometry). The primary variable was main pain improvement at 6 months. Statistical analysis included nonparametric tests (Wilcoxon, Mann–Whitney U), chi-square/Fisher tests for qualitative variables, and paired regression for temporal changes (α = 0.05).

Results

Both groups showed significant improvements in all evaluated parameters. At short-term follow-up (6 weeks), INF achieved a faster pain reduction, whereas at mid-term (6 months), rESWT demonstrated sustained benefits in pain, stiffness, and functionality, with a higher proportion of “excellent-good” outcomes on the Roles & Maudsley scale (p < .05). Grip strength increased similarly in both groups. No major adverse events were reported.

Conclusions

rESWT and INF are both effective treatments for trigger finger. INF may be preferred for rapid pain relief, while rESWT provides longer-lasting benefits and represents a useful non-invasive alternative for patients with contraindications to corticosteroids.

Resumen

Antecedentes y objetivo

El dedo en resorte es una afección frecuente para la que se emplean distintas opciones conservadoras. El objetivo fue comparar la eficacia de las ondas de choque extracorpóreas radiales (OCER) frente a la infiltración de triamcinolona (INF) en el tratamiento del dedo en resorte de moderada gravedad.

Material y métodos

Estudio comparativo prospectivo no aleatorizado con asignación por muestreo secuencial de 42 pacientes con dedo en resorte Quinnell II-III a dos grupos de tratamiento (OCER vs INF). Se evaluaron dolor (EVA), funcionalidad (DASH y Roles & Maudsley), rigidez/gravedad (Quinnell) y fuerza de prensión (dinamometría) en tres momentos: basal, 6 semanas y 6 meses. La variable principal fue la mejoría media del dolor a 6 meses. El análisis incluyó pruebas no paramétricas (Wilcoxon, U de Mann-Whitney), chi-cuadrado/Fisher para cualitativas y regresión emparejada para evolución temporal; α = 0,05.

Resultados

Ambos grupos mostraron mejorías significativas en todas las variables. A corto plazo (6 semanas), INF consiguió una reducción del dolor más rápida. A medio plazo (6 meses), OCER evidenció beneficios sostenidos en dolor, rigidez y funcionalidad, con una mayor proporción de resultados «excelente-bueno» en Roles & Maudsley (p < 0,05). La fuerza de prensión aumentó de forma comparable en ambos grupos. No se registraron eventos adversos relevantes.

Conclusiones

OCER e INF son tratamientos eficaces para el dedo en resorte. INF puede priorizarse si se busca analgesia rápida, mientras que OCER ofrece beneficios mantenidos y constituye una alternativa no invasiva útil en pacientes con contraindicaciones para corticoides.

Introduction

Stenosing tenosynovitis, also known as trigger finger, is a common condition that affects the flexor tendons in the hand. It affects 2%–3% of adults, and is more prevalent in women over 50 years of age, who are 2–3 times more likely to be affected than men.^1,2 It is associated with metabolic and rheumatological diseases, renal failure, and diabetes mellitus, which increases its prevalence to 15% in these groups.^3–7

It is caused by a dysfunction of the pulley and flexor tendon system, particularly at the A1 pulley. Inflammatory thickening of this pulley generates friction and entrapment of the tendon, causing digital locking.^2,8,9 At the cellular level, fibroblastic proliferation and increased collagen production occur, as well as the involvement of inflammatory mediators such as IL-1, IL-6 and TNF-α.¹ In diabetics, the accumulation of advanced glycation end products increases the stiffness of the pulley.⁵

It commonly affects the third and fourth fingers and can be bilateral, predominantly in the dominant hand. Repetitive use and occupational factors are key determinants,^10–12 and can limit everyday activities such as writing or using electronic devices.¹³

An increase in incidence is anticipated due to population ageing and the increase in type 2 diabetes, highlighting the importance of optimising conservative treatments to improve quality of life and reduce morbidity.^14–16

This study compares the efficacy of two such treatments, triamcinolone injections (INF) and extracorporeal radial shockwave therapy (rESWT), by objectively evaluating pain, functionality, grip strength, and clinical severity.

Materials and methods

Study population

Two researchers recruited the study population between 2021 and 2022 from patients referred from primary care to the outpatient orthopaedic surgery and traumatology clinics of a tertiary hospital. Adult patients who met the clinical criteria for trigger finger (pain over the A1 pulley, painful locking, and/or nodule in the flexor tendon) and agreed to conservative treatment were included after signing an informed consent form. Only cases involving a single finger were selected, with no bilateral involvement and no previous treatment or associated conditions that compromised manual function. The cases were classified as grade II (locking on flexion, actively correctable) or grade III (locking on flexion, passively correctable) according to the Quinnell classification.

Study design

A prospective comparative study was designed to assess changes in pain (VAS), functionality (standardised DASH and Roles & Maudsley scales), grip strength (dynamometry), and disease severity (Quinnell) in INF or rESWT patients, and was approved by our hospital's ethics committee.

Following recruitment, a questionnaire was administered to collect demographic data, medical history, and symptoms. A physical examination was performed, and clinical data, VAS scores, and baseline grip strength were recorded using an analogue dynamometer. Patients were divided into two treatment groups using sequential sampling: first, the rESWT group was completed, followed by the INF group, with 21 patients per group.

The primary variable of the study was pain, measured using VAS at six months of follow-up, defined a priori as the primary efficacy outcome. Given the prospective and comparative nature of the design, after data collection, a post hoc power analysis was performed to verify the adequacy of the sample size. With a total of 42 patients (21 per group), a two-tailed significance level of α = .05, and a mean VAS reduction of 4.33 points in the rESWT group and 4.00 points in the INF group, with standard deviations of 2.0–2.5, an effect size (Cohen's d) of 1.7–1.8 was obtained, corresponding to a very large effect according to Cohen's criteria. This value was associated with a statistical power of .95 (1 − β), confirming that the sample size was adequate to detect clinically relevant differences in the primary variable of pain in the medium term.

The injections were performed by the principal investigator after informed consent was obtained. Triamcinolone acetonide (10 mg/ml) was used with 1% mepivacaine without vasoconstrictor (10 mg/ml), in a 1:1 ratio (total volume 2 ml), under strict aseptic conditions. Three injections were administered via a volar route using a supratendinous technique and a 25G needle (most common),¹⁷ at 10-day intervals.^18,19

The rESWT was administered by a researcher collaborating with the rehabilitation service after consent had been obtained. Four sessions were conducted at weekly intervals, each applying 1500 pulses at a frequency of 20 Hz and a pressure of 1.5 bar.^20,21

The initial clinical assessment included baseline dynamometry. Follow-up assessments were conducted six weeks and six months after the start of treatment.

Variables

Pain was measured using the VAS scale (0 = no pain, 10 = maximum pain). A reduction of ≥60% on the VAS was considered a positive clinical response.²² Furthermore, patients were categorised as: pain-free, experiencing mild pain, moderate pain, or severe pain.

The dynamometric assessment was performed by the principal investigator and two collaborators. Maximum grip strength was used to assess the severity of the pathology²³ and the efficacy of the treatment.²⁴ Measurements were taken with an analogue dynamometer, with the patient's arm close to the body, their elbow at 90°, and their forearm in a neutral position. Each patient performed four maximum grips of 3–4 s, with the mean of the values recorded in kilograms as the criterion for analysis.^13,25,26

Functionality was assessed using the DASH questionnaire and the Roles & Maudsley scale, which classifies the results as: 1, excellent; 2, good; 3, acceptable; and 4, poor. Categories ‘excellent’ or ‘good’ were considered ‘success’, while ‘acceptable’ or ‘poor’ were considered ‘failure’.

‘Cure’ was defined as grade 0 on the Quinnell classification. Grades >I were considered ‘pathological locking’.

Data analysis

One of the study collaborators performed the data collection blinded and compiled the anonymised database. A researcher from the research unit conducted the statistical analysis.

Means, standard deviations, and quartiles were calculated for quantitative variables. Normality was assessed using the Shapiro–Wilk test and Fisher's exact test for small frequencies. Absolute and relative frequencies were used for qualitative variables.

McNemar's test was applied for qualitative variables at two time points, and the Mann–Whitney U test was used to compare medians between groups. The Wilcoxon test was used to compare intragroup differences between time points. Paired linear regression was applied to analyse the evolution of numerical variables.

The χ² test was used for independent categorical variables with an expected frequency >5; for lower frequencies, the Fisher's exact test was used. Contingency tables were used to visualise changes in functional categories (Roles & Maudsley).

Statistical analysis was performed using R Core Team 2024, version 4.3.3.

Results

The statistical analysis was performed on a final sample of 42 patients, who were evenly distributed across two treatment groups using sequential sampling: one group treated with rESWT therapy and the other with INF. The results were evaluated at three time points: baseline (first interview), six weeks, and six months, using the DASH questionnaire, the VAS scale for pain, the Roles & Maudsley scale, the Quinnell classification, and dynamometry to measure grip strength.

The analysis was structured into two main sections: descriptive statistics and comparative statistics. Furthermore, intra-group and intergroup changes over time were assessed, as well as the relationship between clinical variables.

General sample data

No statistically significant differences were observed between the baseline characteristics of the treatment groups (Table 1). These results confirm the initial comparability between groups, which reinforces the validity of subsequent analyses of treatment effects by minimising possible biases related to confounding variables.

Table 1.

Comparison of baseline characteristics between treatment groups.

Variable	Total (42)	rESWT group (21)	INF group (21)	p
Age (in years), mean (SD)	63.86 (10.68)	62.71 (11.23)	65 (10.21)	.47
Sex (n)
Female	26	14	12	.51
Male	16	7	9
BMI (kg/m^2), mean (SD)	27.17 (3.53)	27.24 (3.69)	27.1 (3.41)	.87
Duration of symptoms (days), mean (SD)	450.12 (379.16)	437.29 (368.75)	463.14 (392.41)	.79
Laterality, n
Right-handed	37	19	18	.65
Left-handed	5	2	3
Affected hand, n
Right	29	15	14	.74
Left	13	6	7
Affected finger, n
Thumb (1st)	4	2	2	1
Index finger (2nd)	5	2	3	.64
Middle finger (3rd)	18	10	8	.52
Ring finger (4th)	14	6	8	.54
Little finger (5th)	1	1	0	.31

Pain (VAS)

Pain intensity was assessed using the VAS scale at three time points during the study: baseline, six weeks, and six months. Comparisons were made both between treatment groups and within each group over time.

At baseline, the mean VAS score for the total sample was 5.29 (SD: 2.35), indicating a moderate level of pain. At six weeks, a significant reduction was observed, with a mean score of 2.05 (SD: 2.14), and at six months, the values decreased further, to a mean score of 1.12 (SD: 2.05), reflecting sustained improvement in both groups.

The intergroup comparisons at each time point revealed similar initial pain levels (baseline VAS: 5.14 [SD: 2.54] in the rESWT group and 5.43 [SD: 2.2] in the INF group), with no statistically significant differences observed. At six weeks, both groups showed a marked reduction in pain: 2.52 (SD: 2.36) for rESWT and 1.57 (SD: 1.83) for INF, with no significant differences between treatments. At six months, VAS scores continued to decline, reaching .81 (SD: 1.54) in the rESWT group and 1.43 (SD: 2.46) in the INF group, again with no statistically significant differences between groups.

Intragroup analysis using paired linear regression revealed significant improvements in all comparisons between the three assessed time points, except between six weeks and six months in the INF group (Table 2 and Fig. 1).

Table 2.

Intragroup comparison of VAS over time.

Temporal comparison	Group	Mean difference (points) (95% CI)	p
Baseline vs. 6 weeks	rESWT	−2.62 (−3.66 to −1.57)	<.001
	INF	−3.86 (−4.9 to −2.81)	<.001
6 weeks vs. 6 months	rESWT	−1.71 (−2.76 to −.67)	.001
	INF	−.14 (−1.19 to .9)	.789
Baseline vs. 6 months	rESWT	−4.33 (−5.38 to −3.29)	<.001
	INF	−4 (−5.04 to −2.96)	<.001

Fig. 1.

Comparison using a box plot (VAS, group, and time).

Furthermore, the VAS scale was subdivided into four categories: no pain (score 0), mild pain (1–3), moderate pain (4–6), and severe pain (7–10). At baseline, the distribution of pain was similar between both groups, with no statistically significant differences (p > .05). In the rESWT group, two patients were pain-free, four experienced mild pain, seven had moderate pain, and eight had severe pain. In the INF group, one patient was pain-free, three had mild pain, nine had moderate pain, and eight had severe pain.

At six weeks, both treatments achieved a marked reduction in severe pain and an increase in the proportion of patients who were pain-free or had mild pain, with statistically significant differences between the groups (p < .05). In the rESWT group, seven patients were pain-free, seven had mild pain, six had moderate pain, and one had severe pain. In the INF group, the results were even more favourable in the short term: 11 patients were pain-free, seven had mild pain, and three had moderate pain, with no cases of severe pain.

At six months, in the rESWT group, 15 patients were pain-free, five had mild pain, and one had moderate pain, with no recurrence of severe pain. In the INF group, 13 patients were pain-free, four had mild pain, three had moderate pain, and one had severe pain. The χ² test indicated statistically significant differences in favour of the rESWT group in this assessment (p < .05) (Fig. 2).

Fig. 2.

Bar chart (VAS subgroups by time and treatment group).

Analysis of the number of patients achieving a ≥60% reduction on the VAS revealed differences in pain progression between the treatment groups. At six weeks, the INF group had a higher proportion of patients with clinically significant improvement, totalling 14 patients, while in the rESWT group this reduction was observed in 7 patients.

However, at six months the trend reversed, with a higher rate of improvement observed in the rESWT group, which rose to 15 patients, while the number of patients in the INF group remained stable at 14. Nevertheless, statistical analyses revealed no significant differences between the groups (p > .05).

Functionality results (DASH and Roles & Maudsley scale)

The DASH questionnaire scores were standardised (0–100) and assessed at three time points: baseline, six weeks, and six months, to evaluate functional progress within each group and differences between groups. For statistical analysis, Student's t-test was used for within-group comparisons and the Mann–Whitney U test for between-group comparisons. Furthermore, paired linear regression was applied to analyse the evolution of numerical variables by group.

Initially, the mean standardised DASH scores were similar in both groups: 33.5 (SD: 23.4) in the rESWT group and 37.5 (SD: 17.2) in the INF group. At six weeks, both therapies showed significant functional improvement, with scores falling to 18.5 (SD: 15.3) and 20.9 (SD: 20.4), respectively (p < .01 intragroup). At six months, scores continued to decrease to 15.0 (SD: 14.9) in the rESWT group and 16.0 (SD: 18.5) in the INF group. This reflects mild residual disability and sustained functional recovery, with no significant differences between groups (p > .05) (Fig. 3).

Fig. 3.

Comparison using a box plot (DASH, group, and time).

Intragroup comparisons revealed significant differences between baseline and six weeks, and between baseline and six months, but not between six weeks and six months (Table 3).

Table 3.

Intragroup comparison of DASH scores over time.

Temporal comparison	Group	Mean difference (points) (95% CI)	p
Baseline vs. 6 weeks	rESWT	−15.0 (−22.4 to −7.6)	.002
	INF	−16.6 (−25.1 to −8.1)	.002
6 weeks vs. 6 months	rESWT	−3.5 (−9.9 to +2.9)	.331
	INF	−4.9 (−12.6 to +2.8)	.167
Baseline vs. 6 months	rESWT	−18.5 (−26.1 to −10.9)	.001
	INF	−21.5 (−29.9 to −13.1)	<.001

No statistically significant differences were found between the groups at any of the assessed time points (Table 4).

Table 4.

Inter-group comparison of DASH scores over time.

Temporal comparison (rESWT vs. INF)	Mean difference (points) (95% CI)	p
Baseline	−4.0 (−14.6 to 6.6)	.5
6 weeks	−2.4 (−12.3 to 7.5)	.55
6 months	−1.0 (−11.2 to 9.2)	.97

Functional progress was analysed using the Roles & Maudsley scale, with participants in both groups being assessed at six weeks and six months. To facilitate analysis, the categories ‘excellent’ and ‘good’ were grouped as ‘excellent-good’, and ‘acceptable’ and ‘poor’ as ‘acceptable-poor’.

At six weeks, both groups had 17 patients in the’excellent-good’ category and four in the ‘acceptable-poor’ category.

At six months, the rESWT group had shown marked improvement, with 20 patients in the ‘excellent-good’ category and one in the ‘acceptable-poor’ category. The INF group's distribution remained the same as at six weeks (17 ‘excellent-good’ and four ‘acceptable-poor’). An intergroup comparison using a χ² test showed statistically significant differences in favour of the rESWT group at both time points (p < .05).

The contingency tables reflected the changes between categories. In the rESWT group, the four patients initially classified as ‘acceptable-poor’ moved to ‘excellent-good’. In contrast, there was less movement between categories in the INF (Fig. 4).

Fig. 4.

Contingency table (Roles & Maudsley scale, group, and time). rESWT group (left): most patients remained in the higher categories (‘excellent’ and ‘good’), with notable improvements from lower categories. INF group (right): although several patients remained in the ‘excellent’ category, there was a greater shift towards lower categories, such as ‘acceptable’ and ‘poor’.

Grip strength results (dynamometry)

The evolution of mean grip strength (kg) in the affected hand was assessed. Measurements were taken at baseline, at six weeks, and at six months. No statistically significant differences between groups were found at any of these time points (p > .05) (Table 5).

Table 5.

Inter-group comparison of dynamometry over time.

Temporal comparison (ESWT vs. INF)	Mean difference (kg) (95% CI)	p
Baseline	1.78 (−2.36 to 5.92)	.412
6 weeks	1.36 (−2.91 to 5.63)	.529
6 months	1.62 (−2.74 to 6.00)	.448

Intragroup analysis showed a significant improvement in mean strength from baseline to six weeks and six months in both groups. However, there were no significant differences between six weeks and six months (Table 6).

Table 6.

Intra-group comparison of dynamometry over time.

Temporal comparison	Group	Mean difference (kg) (95% CI)	p
Baseline vs. 6 weeks	rESWT	2.84 (1.35 to 4.29)	<.001
	INF	3.26 (1.89 to 4.63)	<.001
6 weeks vs. 6 months	rESWT	1.22 (−.24 to 2.7)	.1
	INF	.96 (−.52 to 2.43)	.199
Baseline vs. 6 months	rESWT	4.06 (2.59 to 5.52)	<.001
	INF	4.22 (2.76 to 5.68)	<.001

Quinnell classification: stiffness and severity of the condition

The rESWT group showed a significant improvement in the proportion of patients with at least a one-grade reduction in the Quinnell classification. At six weeks, 17 patients achieved this improvement, increasing to 21 patients at six months (p < .05). The INF group also showed improvement: 19 patients improved at six weeks and 18 at six months, with statistical significance maintained (p < .05).

In the intergroup analysis, no significant differences were found in the rate of improvement between groups, at either six weeks or six months (p > .05).

Cure, defined as Quinnell grade 0, showed significant increases in both groups. In the rESWT group, the number of cured patients rose from four (at six weeks) to 15 (at six months) (p < .05). In the INF group, it increased from three to 12 patients (p < .05). However, the differences between groups were not statistically significant at any time point (p > .05).

Finally, in the intragroup analysis, the rESWT group showed a significant increase in patients without locking (Quinnell ≤1), from 13 to 19 between six weeks and six months (p < .05). In the INF group, this proportion rose from 15 to 16 patients, without reaching statistical significance (p > .05). The intergroup analysis revealed no significant differences at either time point (p > .05).

Discussion

Both extracorporeal shockwave therapy (rESWT) and corticosteroid injections (INF) have demonstrated superiority over placebo in controlled studies. In the case of rESWT, randomised double-blind clinical trials have demonstrated a significant reduction in pain and sustained functional improvement in patients with trigger finger treated with high-energy shock waves, with benefits maintained at six months.^21,27 Similarly, corticosteroid injections have been shown to be more effective than placebo,²⁸ providing positive results in reducing digital locking and relieving pain from the first few weeks. These findings reinforce the role of both therapies as valid, evidence-based options for the conservative management of trigger finger.

Our study builds upon the research initiated by Yildirim et al.,²⁹ and is one of the few to directly compare the effectiveness of rESWT with corticosteroid injections. Its main advantages include an updated therapeutic protocol, involving three serial injections of triamcinolone and mepivacaine versus a single dose of betamethasone and lidocaine, and a more intensive rESWT treatment (four sessions of 1500 pulses at 20 Hz), which is in line with recent evidence.^20,21 Furthermore, by including patients with Quinnell grades II and III, the study more accurately reflects the typical clinical presentation seen in practice. Multidimensional assessment using DASH, VAS, Roles & Maudsley, Quinnell scores, and dynamometry provides a more comprehensive view of functional progress.

The results of this study confirm that both rESWT and INF are effective conservative therapies for moderate-to-severe trigger finger, achieving significant reductions in pain and improving function and grip strength, as well as reducing the Quinnell classification. These results are consistent with previous studies on the efficacy of these treatments in tendon pathologies.^4,21,23,30 While no statistically significant differences were observed between the groups, clinically relevant differences were identified: faster improvement with INF and more sustained effects with rESWT. This suggests that the two therapies have different mechanisms of action and response times.

Pain (VAS)

Both treatments resulted in a significant reduction in pain at all time points assessed. Corticosteroid injections provided faster relief in the first few weeks, while shockwave therapy showed progressive and sustained improvement over the medium term. While the differences between the groups were not significant, the temporal progression suggests distinct mechanisms: the immediate analgesic effect of corticosteroids is due to their potent anti-inflammatory action,^3,6 while the delayed improvement associated with shockwave therapy has been linked to neurogenic, angiogenic, and regenerative processes.^20,21,23 These results confirm the efficacy of both strategies, with complementary response patterns depending on the clinical objective: rapid relief (INF) or a sustained effect (rESWT).

Functionality (DASH and Roles & Maudsley)

Both treatments were associated with significant functional improvement, as evidenced by a reduction in DASH scores and progression on the Roles & Maudsley scale.²⁷ While the differences between the groups were not statistically significant, the six-month follow-up revealed a trend towards better functional maintenance with rESWT, with a higher proportion patients in the ‘excellent’ or ‘good’ categories. This finding suggests that shock waves may promote a more stable functional recovery, consistent with studies describing their prolonged effect on tendon reorganisation and tissue quality.

Grip strength (dynamometry)

Both treatments achieved a significant improvement in grip strength at six months, with no significant differences between groups. This indicates comparable efficacy in biomechanical recovery. This improvement appears to be related to a reduction in mechanical restriction and pain,^14,24,25 which are key factors in hand function recovery. The absence of intergroup differences contrasts with some studies suggesting a greater impact of rESWT on strength recovery,^13,25,26 possibly due to individual factors influencing the therapeutic response, such as initial severity or fibrosis. Conversely, the effect of corticosteroids may be less long-lasting in cases of chronic fibrosis.^6,16

Severity and healing (Quinnell classification)

Both treatments showed substantial improvement in the Quinnell classification, with a high proportion of patients free from functional locking at the end of follow-up. Although the differences between groups did not reach statistical significance, the rESWT group showed a trend towards a higher rate of complete cure, which reinforces its possible sustained beneficial effect in the medium term.^20,21,27 These results are consistent with studies linking rESWT to tendon repair processes and a reduction in fibrosis, favouring a more stable functional recovery.

Limitations

This study has several limitations. The small sample size may have limited the detection of significant differences in some analyses. Although baseline comparability between groups was confirmed, sequential sampling may have introduced bias in the allocation. Further studies with larger sample sizes, longer follow-up periods, and randomised designs are needed to confirm these findings and investigate the biological mechanisms underlying rESWT in greater depth.

Conclusion

This study demonstrates that both radial extracorporeal shockwave therapy (rESWT) and corticosteroid injections (INF) are effective treatments for the conservative management of moderate-to-severe trigger finger. Both therapies offered significant improvements in pain, function, and grip strength, with high cure rates and favourable clinical outcomes at the six-month follow-up.

INF showed a faster onset of pain relief, while rESWT demonstrated a more sustained response, with a higher proportion of patients being pain-free and having optimal function at the mid-term follow-up. Given its safety and non-invasive nature, rESWT is a particularly suitable alternative in patients with contraindications to corticosteroids or who prioritise conservative options.

This study adds value by directly comparing both interventions using a more comprehensive therapeutic and evaluative protocol than previous studies, including medium-term follow-up and validated functional scales. Against a backdrop of rising incidence of the condition due to an ageing population and metabolic diseases, it will be key to optimise its management through individualised treatment selection, future research into clinical subgroups, therapeutic combinations, cost-effectiveness analyses, and outcomes with longer-term follow-up.

Level of evidence

Level of evidence: II.

Authors’ contributions

All authors contributed substantially to: (1) the conception, study design or data collection and analysis; (2) the drafting or critical revision of the manuscript; and (3) the final approval of the submitted version.

Adrián Hermenegildo Suárez Cabañas: conceptualisation, research, drafting of the manuscript and data analysis.

José Manuel García Castellano: supervision and critical review of the manuscript.

Margarita Ramírez Sánchez: research and data collection.

Jesús María González Martín: statistical analysis.

Alberto Álvarez Jiménez: research and data collection.

Milán Santana Bañolas: research and data collection.

Ethical considerations

The study was approved by the Committee for Ethics in Medical Research at the Dr Negrín University Hospital of Gran Canaria (CEIm HUGCDN) under reference number 2020-190-1. It was conducted in accordance with the ethical principles of the Declaration of Helsinki and current legislation.

Informed consent

All participants were informed of the study's objectives and procedures and signed a written informed consent form prior to their inclusion in the study. The study does not contain any patient-identifying data.

Declaration of generative AI and AI-assisted technologies in the writing process

Generative artificial intelligence (ChatGPT, OpenAI) was used to a limited extent for grammatical corrections and language refinement when drafting the manuscript. The authors produced all scientific content, analysis, and conclusions exclusively.

Funding

No funding was received to undertake this study.

Conflict of interests

The authors have no conflicts of interests to declare.

Acknowledgements

The authors would like to express their sincere gratitude to Dr Vicente Vera for his valuable help in recruiting patients for this study. His dedication and commitment to research were fundamental to completing this work. Sadly, Dr Vera passed away during the course of the study. We honour his memory and his significant contribution to the medical and scientific community.