Stretching-Positioned Extracorporeal Shock Wave Therapy Yields Superior Short-Term Efficacy for Post-Stroke Upper-Limb Spasticity: A Retrospective Comparative Study

Introduction

Spasticity is a common complication in the recovery phase of stroke patients. It is a motor dysfunction characterized by increased excitability of the stretch reflex, resulting in velocity-dependent hypertonia. Its mechanism originates from the excessive excitation of the descending excitatory brainstem pathways and the amplification of the stretch reflex response. Spasticity can lead to abnormal synergistic movements and inappropriate muscle activation [1,2]. In clinical practice, spasticity can cause contractures, pain, weakness, and coordination disorders. In the upper limbs, it manifests as shoulder adduction and internal rotation, elbow and wrist, and finger flexion, restricting limb freedom and severely affecting rehabilitation training outcomes and quality of life [3–5].

Early and effective rehabilitation training can suppress spasticity in stroke patients and greatly promote functional recovery. Current treatments for spasticity include stretching, oral anti-spastic drugs, local neurochemical blockade, intrathecal baclofen injection, and surgery [6–9]. Although these methods have therapeutic effects, they have their own limitations in terms of effectiveness, safety, economy, and the sustainability of therapeutic effects.

Extracorporeal shock wave therapy (ESWT) is a non-invasive modality that delivers acoustic waves to target tissues, promoting biological responses and reaching deeper structures without significant absorption by superficial tissues [10]. Since the 1980s, it has been successively applied to treat urolithiasis, periarthritis of the shoulder, tennis elbow, plantar fasciitis, and other musculoskeletal diseases [11–14]. Recent studies have reported that shock waves can reduce muscle spasticity and lower muscle tone after stroke. Dalir studied the effects of shock wave therapy on wrist flexor spasticity in 15 post-stroke patients. The results showed an improvement in the MAS grading score of spasticity, which lasted for 5 weeks [15]. Dymarek conducted shock wave therapy on the elbow, wrist, and finger flexors in 60 stroke patients and assessed them before and after treatment using the MAS spasticity grading score, surface electromyography, and infrared thermography [16]. Significant reductions in MAS spasticity scores and significant changes in surface electromyography signals and temperature distribution were observed. Radinmehr compared the effects of therapeutic ultrasound and shock wave therapy on plantar flexor spasticity in 32 stroke patients. The MAS spasticity grading score, active range of motion (aROM), passive range of motion (pROM), passive plantar flexor torque (pPFT), and timed “up and go” test (TUG) were assessed before and after treatment [17]. Significant improvements were observed in the MAS spasticity grading score, aROM, pROM, pPFT, and TUG. Although ESWT has been widely confirmed to relieve spasticity, in our clinical practice we have found that the anti-spastic effect varies greatly when ESWT is applied at different muscle lengths, and the optimal position for spastic muscles has not been reported. Although ESWT has been widely confirmed to relieve spasticity, a critical factor – treatment posture (muscle length during application) – has not been adequately examined, and there have been no published studies directly comparing stretching versus relaxation positioning. We hypothesized that a stronger anti-spasticity effect would be obtained when ESWT is performed in the stretched position [14]. Therefore, this study aimed to evaluate the short-term effects of ESWT applied to the biceps brachii in stretching versus relaxation postures on upper-limb spasticity and motor function in stroke patients. We hypothesized that a stronger anti-spasticity effect would be obtained when ESWT is performed in the stretched position.

Material and Methods

ETHICAL STATEMENT:

This retrospective cohort study was conducted in accordance with the Declaration of Helsinki and was reviewed and approved by the Institutional Review Board (IRB) of Sir Run Run Shaw Hospital, Zhejiang University School of Medicine (approval number: IRB-SRRH-ZJU/0298/2023). The IRB granted a full waiver of the requirement for informed consent for this study because the research involved no more than minimal risk to the subjects, utilized only pre-existing, anonymized data from electronic health records, and it was impracticable to obtain consent from all subjects given the retrospective nature of the study. All patient data were de-identified prior to analysis to ensure confidentiality.

STUDY DESIGN AND PARTICIPANTS:

This retrospective comparative study analyzed data from electronic health records (EHRs) of stroke patients admitted to the Department of Rehabilitation Medicine between February 2023 and February 2025. Based on a power analysis (power=0.8, α=0.05, effect size, d=0.6 estimated from previous ESWT studies on spasticity), a sample size of 48 per group was required. Data from 96 patients who met the inclusion criteria were extracted using a standardized protocol. To minimize selection bias, a consecutive sampling approach was used. All stroke patients admitted to the department during the study period who received ESWT for upper-limb spasticity were screened for eligibility. The final 96 patients represented all consecutive cases that fully met the inclusion and exclusion criteria and had complete data records at all predefined time points, indicating that this was an unselected cohort from the specified period.

Explanation for Standardized Time Points: Although this was a retrospective analysis, the data were derived from a standardized ESWT treatment protocol for post-stroke spasticity routinely implemented in our department. This internal clinical pathway mandated that all patients receiving ESWT for upper-limb spasticity undergo scheduled assessments at predefined time points: pre-treatment (baseline), immediately after treatment, and at 6, 12, and 48 h after treatment. These assessments were part of routine clinical care to monitor therapeutic response. Therefore, the data for outcome measures at these intervals were consistently documented in the EHRs and standardized physiotherapy notes, allowing for their retrospective extraction and analysis. This standardized assessment protocol ensured the consistent availability of high-temporal-resolution data in the EHRs for retrospective analysis, despite the study’s non-prospective design. For retrospective extraction, the following time windows were allowed for each assessment: immediately after treatment (±15 min), 6 h (±1 h), 12 h (±1 h), and 48 h (±2 h). Data recorded within these windows were included to accommodate minor clinical scheduling variations.

The inclusion criteria were: (1) diagnosis of ischemic or hemorrhagic stroke confirmed by computed tomography (CT) or magnetic resonance imaging (MRI); (2) age 18–80 years; (3) stable vital signs; (4) unilateral hemiplegia with Modified Ashworth Scale (MAS) grade ≥1 in the affected biceps brachii; (5) no significant cognitive impairment (Mini-Mental State Examination score >20); (6) complete documentation of ESWT treatment parameters and outcome measures at specified time points.

Exclusion criteria were: (1) impaired consciousness, unstable medical condition, or recurrent stroke; (2) pre-existing fixed elbow joint contractures (defined as a structural limitation in elbow extension that is unresponsive to passive stretching, typically confirmed radiographically if suspected); (3) coagulation disorders, local skin infection, or ulcers at the treatment site; (4) use of oral anti-spastic drugs, neuromuscular blockade, or prior spasticity-related surgery during the study period; (5) other significant neurological, metabolic, or systemic diseases.

Patients were categorized into 2 groups based on the documented treatment protocol in their EHRs: Group A (stretching position ESWT, n=48) and Group B (relaxation position ESWT, n=48).

During the study period, the choice of ESWT application posture (stretching or relaxation) was based on the treating therapist’s assessment of patient presentation and their prevailing clinical practice, without a formal randomization protocol. This variation is a naturalistic comparison of 2 commonly used techniques within our department’s standard care. The assignment to a specific posture was consistent for each patient throughout their recorded treatment session.

As this was a retrospective study in which treatment posture was determined by clinical practice variation rather than randomization, the potential for selection bias cannot be ruled out and is acknowledged as a limitation.

INTERVENTION:

All patients received a single session of radial ESWT applied to the muscle belly of the affected biceps brachii using an HB102 pneumatic ballistic shock wave therapy device (Suzhou Hobe Medical Device Co., Ltd., Suzhou, China) with a 15-mm divergent treatment probe. The treatment parameters (an intensity of 2.5 Bar, a frequency of 8 Hz, and a total of 2000 shocks) were selected based on previous studies demonstrating efficacy and safety in post-stroke spasticity [14,16].

GROUP A (STRETCHING):

The patient’s arm was positioned to achieve maximal comfortable elbow extension (target range: 160°–180° of elbow angle, measured with a standard goniometer) with concomitant shoulder extension. The endpoint was defined as the point just before the patient-reported pain ≥3 on a 0–10 numeric rating scale or the therapist detected a firm, non–velocity–dependent resistance. The position was maintained using adjustable supports during ESWT application.

GROUP B (RELAXATION):

The patient’s arm was positioned with the elbow flexed at approximately 90 degrees and the forearm resting comfortably on the abdomen or a pillow, ensuring the biceps brachii was in an anatomically relaxed and shortened state. In both groups, the treatment area was palpated, marked, and cleansed with an alcohol swab. Coupling gel was applied to ensure optimal energy transmission. All ESWT sessions were performed by the same experienced practitioner to maintain consistency in application technique and pressure. The treatment was applied systematically along the entire length of the biceps brachii muscle belly, from the proximal to the distal end, to ensure comprehensive coverage of the spastic muscle tissue.

OUTCOME MEASURES:

Data for the following outcomes were extracted from standardized physiotherapy notes and EHRs at baseline, immediately after treatment, and at 6, 12, and 48 h after treatment. All clinical assessments were performed by 2 experienced physiotherapists who were blinded to the patient’s group assignment. Blinding was maintained by having the assessors perform all evaluations in a separate room, without access to the treatment records or presence during the ESWT session. The EHR fields indicating treatment posture were hidden during data extraction for outcome scoring. Inter-rater reliability for the MAS and PROM assessments was established prior to the study commencement, with an intraclass correlation coefficient (ICC) >0.8. Surface electromyography (sEMG) was performed by a dedicated technician according to a standardized protocol. To minimize variability, all PROM measurements for a given patient were taken by the same assessor, and all MAS assessments were performed by the same trained physiotherapist.

MUSCLE SPASTICITY:

Spasticity of the elbow flexors was assessed using the Modified Ashworth Scale (MAS) The MAS score was converted to a numerical score for statistical analysis (0=0, 1=1, 1+=1.5, 2=2, 3=3, 4=4). The assessment involved the rater passively moving the patient’s elbow from full flexion to full extension at a speed of approximately 1 s for the full range. To ensure consistency, all MAS assessments for a given patient were performed by the same trained physiotherapist.

UPPER-LIMB MOTOR FUNCTION:

Motor function of the affected upper extremity was evaluated using the upper-extremity subscale of the Fugl-Meyer Assessment (FMA-UE), which includes assessments of reflex activity, volitional movement within and outside synergy, coordination, and speed. The total score ranges from 0 to 66, with higher scores indicating better motor recovery.

SURFACE ELECTROMYOGRAPHY (SEMG):

The electrical activity of the biceps brachii was quantitatively measured using a UMI-SE-I electromyograph (Shaoxing United Electronics Co., Ltd., Shaoxing, China). After skin preparation, 2 disposable Ag/AgCl electrodes were placed on the muscle belly along the longitudinal orientation of the muscle fibers, with an inter-electrode distance of 2 cm, following SENIAM recommendations. A reference electrode was placed on the ipsilateral olecranon. The sEMG signals were sampled at 1000 Hz and band-pass filtered (10–500 Hz), and a 50-Hz notch filter was applied. A single, trained physiotherapist (blinded to group assignment) performed a standardized passive elbow extension at an approximate angular velocity of 10°/s. This maneuver was repeated 3 times, with a 30-s rest between trials. The sEMG data from the middle 3-s stable period of the stretch were selected for offline analysis. The raw sEMG signals were full-wave rectified. The AEMG and RMS were calculated over the selected 3-s epoch. To account for inter-individual variability, both AEMG and RMS values from post-treatment time points were normalized to the pre-treatment (baseline) value of the same patient and are expressed as a percentage of baseline. Lower normalized AEMG and RMS values indicate reduced muscle electrical activity and excitability relative to the pre-treatment state. Normalized AEMG and RMS values are expressed as percentage of the individual’s baseline value (baseline=100%). Statistical analyses were performed on these normalized values. Trials with visible motion artifact or signal saturation were excluded prior to averaging (exclusion rate <5% of all trials).

ELBOW JOINT FUNCTION:

The functional status of the elbow joint was assessed using the Mayo Elbow Performance Score (MEPS). This scale evaluates 4 domains: pain (45 points), arc of motion (20 points), stability (10 points), and the ability to perform 5 daily activities (25 points), with a maximum total score of 100. A higher score indicates better elbow function.

PASSIVE RANGE OF MOTION (PROM):

The passive range of motion for elbow extension (in degrees) was measured using a standard transparent plastic goniometer (Baseline^®). The patient was positioned supine with the shoulder in a neutral position. The axis of the goniometer was aligned with the lateral epicondyle of the humerus, the fixed arm was aligned with the midline of the humerus (pointing towards the acromion), and the moving arm was aligned with the midline of the radius (pointing towards the radial styloid process). The rater passively extended the patient’s elbow to the first point of resistance (the point at which a noticeable increase in resistance was felt by the therapist), and the angle was recorded. The recorded angle represents the elbow extension angle, where higher values (closer to 180°) indicate greater extension. Thus, an increase in the reported PROM value denotes improved extension (ie, a reduction in flexion contracture). This measurement reflects soft tissue extensibility under a standardized force rather than the anatomical limit of the joint. All PROM measurements were taken by the same assessor to minimize inter-rater variability.

It is important to distinguish that the MAS assesses velocity-dependent resistance during passive movement (reflecting a neurogenic component of spasticity), while the PROM measurement in this study captured end-range limitation primarily influenced by soft tissue stiffness and viscoelastic properties.

STATISTICAL ANALYSIS:

All statistical analyses were performed using SPSS software (version 23.0, IBM Corp., Armonk, NY, USA). Continuous data are presented as mean±standard deviation (x±s). The normality of data distribution for all variables was assessed using the Shapiro-Wilk test. For inter-group comparisons of baseline demographic and clinical characteristics, independent samples t-tests were used for continuous variables, and chi-square (χ²) tests were used for categorical data.

The primary analysis focused on the changes in outcome measures over time between the 2 groups. For intra- and inter-group comparisons across the 5 time points (baseline, immediately post-treatment, 6 h, 12 h, and 48 h), a 2-way repeated-measures analysis of variance (ANOVA) was employed. Given the ordinal nature of the MAS, non-parametric tests (Friedman test for within-group, Mann-Whitney U test for between-group comparisons at each time point) were also performed as supplementary analyses. The results from both parametric and non-parametric approaches were consistent; parametric results are presented for consistency with other continuous outcomes. This model included ‘Treatment Group’ (Stretching vs Relaxation) as the between-subjects factor and ‘Time’ as the within-subjects factor. The assumption of sphericity was checked using Mauchly’s test. If the assumption was violated, Greenhouse-Geisser correction was applied to adjust the degrees of freedom.

If a significant interaction (Group×Time) effect was found, simple main effects analyses were conducted with Bonferroni adjustment for the 10 pairwise comparisons (5 time points×2 groups). Mauchly’s test indicated that the assumption of sphericity was violated for MAS (W=0.62, P<0.001) and FMA-UE (W=0.71, P =0.008); therefore, the Greenhouse-Geisser correction (ɛ=0.75–0.88) was applied to adjust the degrees of freedom for those outcomes. This involved analyzing between-group differences at each of the 5 time points and analyzing within-group changes from baseline to each subsequent time point within each treatment group.

The primary analysis was a 2-way repeated-measures ANOVA. For any outcome measures with missing data at the predefined follow-up times. Missing data constituted 4.2% of all data points, distributed as follows: MAS n=3, FMA-UE n=4, PROM n=2, all at the 48-h assessment. These were assumed to be missing at random (MAR). In the mixed-effects sensitivity model, an unstructured covariance structure was specified for the repeated time factor to best accommodate the observed variance-covariance pattern. A mixed-effects model was employed as a sensitivity analysis to confirm the robustness of the ANOVA results. The mixed model included fixed effects for Group, Time, and their interaction, and a random intercept for subject. This approach utilizes all available data without resorting to listwise deletion, thus preserving statistical power.

For all tests, a 2-tailed P value of less than 0.05 was considered statistically significant.

Results

BASELINE CHARACTERISTICS:

The baseline demographic and clinical characteristics of the 2 groups are summarized in Table 1. There were no statistically significant differences between Group A and Group B in terms of sex distribution, age, affected side, disease duration, or body mass index (P>0.05), indicating that the groups were comparable at baseline. Baseline clinical characteristics, including spasticity severity, motor function, joint mobility, time since stroke, and stroke type, were also comparable between groups (Table 1).

EFFECTS ON SPASTICITY AND MUSCLE ACTIVITY:

MAS scores are presented as median with interquartile range (IQR) throughout the text, tables, and figures, consistent with its ordinal nature. The conversion to a numerical scale (0–4) was used solely for the repeated–measures ANOVA; non–parametric follow–up tests were performed on the original ordinal scores.

A significant Group×Time interaction was observed for MAS scores (P<0.001), normalized AEMG (P<0.001), and normalized RMS (P=0.002).

Post hoc analyses with non-parametric tests revealed that both groups showed significant reduction in spasticity immediately after treatment and at 6 h compared to baseline (P<0.05). While no significant within-group change was noted at 12 h for either group, Group B demonstrated significantly higher spasticity than Group A at this time point (median [IQR]: 2 [1.5–2] vs 1+ [1–2], P=0.003). At 48 h, both groups regressed, but the rebound in spasticity was significantly smaller in Group A than in Group B (P=0.022).

The normalized AEMG and RMS values showed a similar pattern, with Group A demonstrating significantly greater reduction in muscle electrical activity at 12 h after treatment (Table 2). The between-group differences represented large effect sizes (Cohen’s d >1.4, P<0.001 for both) (Figure 1A–1C).

EFFECTS ON MOTOR FUNCTION AND RANGE OF MOTION:

A significant Group×Time interaction was also observed for FMA-UE, MEPS, and PROM (all P<0.001). As detailed in Table 3, improvements in these outcomes were consistently greater in Group A than in Group B at all post-treatment time points (P<0.05), with medium to large effect sizes. For example, the improvement in PROM at 12 h was 2.57° greater in Group A (95% CI: 1.92 to 3.22; P<0.001; Cohen’s d=1.75) (Figure 1D–1F).

EFFECT SIZE ESTIMATES:

The observed large effect sizes (eg, Cohen’s d >1.4) should be interpreted with caution, as they may partly reflect low within-group variability (indicated by small standard deviations) rather than large absolute changes. Therefore, these effect sizes do not necessarily equate to clinical meaningfulness.

Despite this caveat, the between-group differences, as indicated by medium to large effect sizes at key time points (Tables 2, 3), consistently favored the stretching position. For spasticity (MAS), effect sizes were large at 12 h (d=−1.32) and 48 h (d=−0.98). Improvements in motor function (FMA-UE) and joint mobility (PROM) also showed large effects (d range: 0.68–1.75).

Complete longitudinal data for all outcome measures at all assessed time points (baseline, immediately after treatment, 6 h, 12 h, and 48 h) are provided in Table 4.

Discussion

The principal finding of this retrospective study is that applying ESWT to the spastic biceps brachii in a stretched position was associated with a greater reduction in hypertonia and improvement in motor function over the 48-h observation period, compared to the relaxed position. This suggests that the biomechanical state of the muscle during ESWT application may be a crucial determinant of its immediate anti-spastic efficacy.

The mechanism of extracorporeal shock wave therapy (ESWT) for muscle spasticity is still under investigation. Some studies suggest that ESWT is responsible for inducing the synthesis of nitric oxide (NO), which can further increase neovascularization in muscles and tendons, thereby improving muscle stiffness [18]. Nitric oxide (NO) is also involved in the formation of new neuromuscular junctions [19,20]. Other studies suggest that ESWT reduces the excitability of α-motor neurons by applying continuous or intermittent pressure to tendons. While recent reviews have summarized the progress of ESWT for upper-limb spasticity [14], the specific impact of treatment posture has been inadequately examined. Our study directly addresses this gap by demonstrating that the biomechanical state of the muscle during application may be a critical moderator of efficacy. There are also studies indicating that ESWT reduces acetylcholine receptors at the neuromuscular junction, causing temporary dysfunction in neuromuscular transmission and exerting an anti-spastic effect [21–23]. In addition, non-neural effects of ESWT on muscle spasticity have been reported. ESWT can trigger a biological response that alternates between metabolic and proliferative processes, affecting muscle fibrosis and rheological properties, improving the stiffness of connective tissues, and reducing spasticity [24,25].

Regarding the mechanism of stretching in relieving muscle spasticity, current research suggests that it mainly involves increasing muscle length to restore the extensibility of soft tissues and reduce muscle tone. Its mechanism is primarily to enhance the viscoelastic components of the muscle-tendon unit, increasing the extensibility of soft tissues through causing viscoelastic deformation and structural adaptation in muscles, tendons, skin, and nerve tissues [26,27]. Notably, Mahieu et al found that after 6 weeks of vibration stretching, the passive resistance torque at the same joint angle did not change, but the tendon stiffness unexpectedly decreased. Since the passive resistance torque remained stable, the researchers speculated that the decrease in tendon stiffness might be compensated for by an increase in muscle stiffness. They further proposed that stretching might promote the redistribution of polysaccharide-water complexes within the collagen framework of tendons, thereby reducing stiffness [28,29].

The observed efficacy pattern – rapid improvement after treatment, peaking at around 6 h, and then gradually declining by 48 h – offers clues to the underlying mechanism. The delayed peak effect suggests that the biological response to ESWT, such as the modulation of neurotransmitter release or the initiation of anti-inflammatory and nitric oxide-mediated processes, may require time to reach its maximum. The transient nature of the benefits, consistent with some studies on radial ESWT for spasticity, aligns with the hypothesis of a temporary neuromodulatory effect on hyperexcitable reflex arcs, rather than a permanent structural change. This temporal profile supports the notion that the primary mechanism of action for a single session of ESWT in this context is likely neurogenic modulation, the effects of which are known to be time limited. Our findings underscore the potential of stretching-positioned ESWT as a modality for producing significant short-term spasticity reduction, which could be leveraged to enhance the efficacy of subsequent physical therapy sessions. The optimal frequency for repeated ESWT sessions to sustain these benefits constitutes an important avenue for future research.

Stretching is one of the most widely used techniques in physical therapy for spasticity [30,31]. Kruse et al conducted a study on 18 patients with spastic cerebral palsy and found that static stretching increased the length of the muscle-tendon unit of the gastrocnemius muscle, showing certain effects on ankle joint range of motion and maximum dorsiflexion angle [32]. Konrad’s research also showed that stretching exercises increased tissue compliance, improved joint range of motion, and reduced passive resistance torque and tendon stiffness [29]. Additionally, some studies have found that combining ESWT with stretching training for muscle spasticity significantly reduces muscle tone. Kim et al conducted a study on 29 healthy adults with tight hamstrings, in which participants received ESWT followed by hamstring stretching exercises [33]. Compared with the control group that only received ESWT, the combination significantly improved hamstring tightness. Taheri conducted ESWT on 28 patients with plantar flexor spasticity, combined with anti-spastic drugs and stretching exercises. Compared with the control group receiving anti-spastic drugs and stretching exercises alone, the treatment group showed significant improvements in MAS, pain, and ROM [34]. Based on these findings, this study applied ESWT to the biceps brachii in a stretching position to enhance the viscoelastic components of the muscle-tendon unit, causing viscoelastic deformation and structural adaptation in muscles, tendons, skin, and nerve tissues to increase soft tissue extensibility and more effectively reduce muscle tone.

The novelty of this study lies in its direct comparison of ESWT efficacy under 2 distinct biomechanical conditions – stretch versus relaxation – a factor that has received limited attention in the literature. Our exploratory findings posit that treatment posture is not merely a procedural detail but is a potential moderator of ESWT outcomes. While definitive clinical recommendations are premature, these results suggest that optimizing patient positioning during ESWT could maximize its immediate benefits, potentially facilitating subsequent rehabilitation sessions by temporarily reducing spasticity and improving compliance.

In summary, applying ESWT to the biceps brachii in a stretching position was associated with reduced muscle tone and AEMG/RMS values, as well as improved upper-limb motor function, elbow joint function, and PROM in this cohort.

While statistically significant differences were observed between groups, the absolute changes in key outcomes (eg, ~2.5° improvement in PROM) were modest. To date, no established minimal clinically important difference (MCID) has been validated for MAS or PROM in post-stroke elbow spasticity. Therefore, the observed differences should be interpreted as short-term physiological or biomechanical modulations rather than definitive clinical superiority. Future studies should anchor observed changes to patient-reported outcomes or functional milestones to better gauge clinical relevance.

The observed baseline PROM of approximately 86° may appear inconsistent with the exclusion of elbow contractures. However, this limitation in PROM likely reflects soft tissue stiffness and velocity-independent resistance rather than fixed structural contractures. In this cohort, the PROM restriction was due to non-neural factors such as muscle fibrosis and altered viscoelasticity, which are reversible with stretching. This distinction is important for interpreting the clinical profile of post-stroke spasticity.

This study observed that some post-stroke patients presented with low MAS scores (approximately 1.0) yet severely limited PROM (eg, extension only reaching 90°). This apparent paradox can be explained by the pathophysiology of spastic paresis. MAS primarily assesses velocity-dependent hyperreflexia (classical “spasticity”), while severe PROM limitation is more likely reflective of non-neurogenic factors like soft tissue contracture due to chronic disuse [35]. Gracies clearly demonstrated that relative immobilization of paretic limbs, particularly in shortened positions, leads to muscle adaptive shortening, sarcomere loss, and increased intramuscular connective tissue and fat, ultimately causing muscle and joint contractures [35]. Multiple studies support the dissociation between spasticity severity (eg, MAS scores) and range of motion limitation. Malhotra et al emphasized that resistance assessed clinically often conflates non-reflexive soft tissue stiffness [36]. Pandyan et al also noted that spasticity should not be viewed as a pure motor disorder, but rather a manifestation involving both sensorimotor control disruption and soft tissue changes [37]. Thus, the low MAS scores in our cohort suggest minimal velocity-dependent hyperreflexia, whereas the significant PROM limitation indicates substantial soft tissue contracture secondary to disuse. This finding highlights the clinical importance of distinguishing neurogenic spasticity from non-neurogenic contracture, as their management strategies differ substantially [35,37].

This study has several limitations. First, its retrospective nature introduces potential for selection bias, and unmeasured confounding factors (such as stroke severity or precise time since onset) may exist despite comparable baseline demographics. Second, while assessors of most outcomes were blinded, the nature of the stretching intervention made complete blinding of treating therapists impossible, which might introduce performance bias. Third, this study was designed to investigate only the immediate and short-term (up to 48 h) effects of a single ESWT session. Fourth, the detailed assessment schedule (eg, 6 h, 12 h after treatment), while part of our institutional protocol, is not typical of routine clinical practice. This may limit the generalizability of our findings to settings without such intensive monitoring. Consequently, it does not provide information on the long-term sustainability of the benefits or the effects of repeated treatment sessions. The optimal frequency for repeated ESWT sessions to sustain these short-term benefits is an important avenue for future research. Determining whether daily, weekly, or bi-weekly applications are most effective could significantly enhance the clinical utility of this approach.

Conclusions

This exploratory retrospective study suggests that applying ESWT to the biceps brachii in a stretching position may provide a greater short-term reduction of upper-limb spasticity and numerical improvement in motor function over 48 h compared to the relaxation position in stroke patients. These findings indicate that treatment posture can influence the immediate effects of ESWT, but the clinical significance remains uncertain. Prospective studies are needed to confirm these results and to investigate whether incorporating muscle stretching during ESWT can lead to clinically meaningful improvements in post-stroke spasticity.