Reliability of A Vibration-Based Elastography Protocol For Assessing Achilles Tendon Stiffness Across Multiple Joint Angles In Elite Athletes

Summary

This protocol describes a standardized, portable ultrasound-based method for quantifying the functional stiffness spectrum of the Achilles tendon across multiple ankle joint angles in elite athletes, enabling reliable and reproducible assessment of tendon mechanical behavior under different loading conditions.

Abstract

The mechanical behavior of the Achilles tendon plays a critical role in athletic performance and injury risk; however, in vivo assessment of tendon stiffness remains challenging. Conventional approaches combining ultrasonography with dynamometry are expensive, laboratory-bound, and typically limited to single joint positions, while existing elastography-based techniques are often constrained by methodological assumptions or limited functional relevance.

The purpose of this study was to present and validate a standardized, portable protocol for quantifying the functional stiffness spectrum of the Achilles tendon across multiple fixed ankle joint angles. This paradigm shifts the assessment from a single static stiffness value to a continuous mechanical profile, capturing the tendon’s nonlinear response to loading. Using a force–ultrasound fusion system, mechanically induced low-frequency vibrations were applied to the tendon while ultrasound-based motion tracking was used to estimate the shear elastic modulus of superficial tendon tissue. Measurements were performed bilaterally in elite male athletes at predefined ankle joint positions ranging from relaxed and plantarflexed states to neutral and dorsiflexed positions.

The protocol demonstrated good intra-trial repeatability and excellent intra-session reproducibility across all joint angles, with coefficients of variation remaining within acceptable limits for soft-tissue elastography and intraclass correlation coefficients indicating high reliability. Achilles tendon stiffness increased non-linearly with progressive dorsiflexion, indicating angle-dependent mechanical behavior. No significant main effect of side dominance was observed across the full functional range, while sport-specific differences emerged at selected joint angles.

This protocol provides a practical and repeatable approach for characterizing Achilles tendon mechanical behavior under functionally relevant loading conditions. Its portability and standardized workflow make it suitable for laboratory, clinical, and field-based applications, offering a valuable tool for athlete monitoring, injury risk assessment, and longitudinal evaluation of tendon adaptation.

Introduction

The Achilles tendon plays a critical role in high-performance human movement by transmitting muscular forces and storing and releasing elastic energy during the stretch–shortening cycle (SSC) actions¹. Its mechanical stiffness is a key determinant of movement efficiency, influencing force transmission, elastic energy reutilization, and overall mechanical output during locomotor and explosive tasks². In elite athletes—particularly those involved in sprinting, jumping, and other SSC-dominant sports—greater Achilles tendon stiffness has been consistently associated with superior sprint speed, acceleration, running economy, jump performance, and rate of force development³. Both long-term training exposure and short-term mechanical loading have been shown to induce measurable alterations in tendon stiffness, reflecting the adaptive capacity of tendon tissue^4,5. Conversely, pathological conditions such as Achilles tendinopathy are often characterized by altered stiffness, which may impair force transmission despite preserved muscle strength⁶. The impact of tendinopathy is substantial; in elite sports, it leads to significant time loss, impaired performance, and potentially shortened careers, while in recreationally active populations, it represents a highly prevalent, recalcitrant issue that diminishes quality of life and incurs considerable healthcare costs. Accurate and reliable assessment of Achilles tendon stiffness is therefore essential for performance monitoring, load management, and injury-related evaluation in athletic populations.

Currently, the combination of ultrasonography and dynamometry is widely regarded as a reference approach for the in vivo assessment of tendon stiffness^7,8,9. While this method provides valuable insight into tendon mechanical properties under highly controlled conditions, several practical limitations restrict its broader application¹⁰. The setup is time-consuming, strongly dependent on operator expertise, and typically confined to laboratory environments. Furthermore, it represents a substantial financial barrier, often requiring significant capital investment for both the isokinetic dynamometer and premium ultrasound machinery. In addition, stiffness estimates are commonly derived under isolated or quasi-static loading conditions at a single joint configuration, which limits their applicability for routine athlete monitoring, field-based assessment, and longitudinal evaluation across training cycles. These constraints highlight the need for alternative measurement approaches that are both methodologically robust and feasible in applied sport settings.

Ultrasound-based elastography techniques have emerged as valuable tools for the in vivo assessment of tendon mechanical properties. Among these, shear wave elastography (SWE) has been widely applied to musculoskeletal tissues; however, its use has highlighted important methodological challenges¹¹. Previous studies have demonstrated that elastography-derived stiffness measurements are highly sensitive to joint angle, probe orientation, tissue pre-compression, region-of-interest (ROI) selection, and data processing strategies, particularly in highly anisotropic structures such as tendons. To mitigate operator-induced variability, some authors have advocated for the use of custom external harnesses to secure the ultrasound probe, although this often comes at the expense of testing efficiency and rapid data acquisition. As a result, methodological standards and rigorous measurement protocols—whether utilizing freehand techniques or external stabilization—have been strongly advocated to ensure valid and reproducible stiffness assessment. These methodological considerations are not limited to SWE but are broadly relevant to elastography-based techniques that infer tissue stiffness from mechanically induced wave propagation.

In recent years, vibration-based ultrasound elastography has gained attention as a practical and field-adaptable alternative for assessing the mechanical properties of superficial musculoskeletal tissues¹². In this approach, mechanical vibrations—with frequency and amplitude parameters specifically optimized for the acoustic and structural properties of the target tissue—are externally applied to the tissue, and the resulting wave propagation is tracked using ultrasound imaging to derive stiffness-related parameters. While previous pioneering studies have successfully utilized ultrasonography coupled with an external actuator to evaluate tendon mechanics—employing a bulky mechanical shaker strapped to the limb to generate continuous sinusoidal waves^13,14—the present protocol utilizes a transient vibration approach. By employing a flexible, hand-held configuration where the mechanical excitation tip is manually co-positioned immediately adjacent to the ultrasound transducer to deliver extremely brief (300 ms) transient impulses, this system eliminates the need for complex and time-consuming external strapping setups. This advancement significantly reduces subject burden and, compared with traditional laboratory-based dynamometry–ultrasound combinations, makes vibration-based elastography systems more portable, non-invasive, and feasible for repeated measurements in applied sport settings. However, despite these advantages, existing studies have typically assessed Achilles tendon stiffness at a single joint configuration, providing only a limited snapshot of tendon mechanical behavior.

Tendon stiffness is inherently dependent on the configuration of the muscle–tendon unit, varying as a function of joint angle and muscle length. A single-angle measurement therefore fails to capture the functional variability in tendon stiffness that occurs across the ankle range of motion and during sport-specific postures. This limitation reduces the practical relevance of stiffness measurements for athletes exposed to multi-angle loading and rapid force transitions. To date, few studies have systematically quantified Achilles tendon stiffness across multiple, standardized joint angles using a reproducible elastography-based protocol¹⁵.

To address this methodological gap, we propose a Functional Stiffness Spectrum Paradigm. This approach reconceptualizes tendon stiffness not as a scalar property but as a continuous function of joint position, quantifying the tendon’s mechanical output across a physiological range of loading states. By isolating the shear elastic modulus of the free tendon across multiple angles, this method provides a tissue-specific assessment that complements traditional dynamometry of the muscle-tendon unit. The purpose of this manuscript is to present a detailed, step-by-step protocol for implementing this method, including subject positioning, joint angle standardization, probe handling, ROI selection, and data acquisition procedures. This protocol is designed to facilitate reproducible assessment of the Achilles tendon’s functional stiffness spectrum and to provide researchers and practitioners with a practical tool for investigating sport-specific tendon adaptations and functional biomechanics in elite athletes. Importantly, to provide practical guidance on the utility of this method, its applicability boundaries must be clearly defined. This approach is highly appropriate for the non-invasive, static or quasi-static profiling of local tendon mechanics—such as monitoring longitudinal adaptations, screening for side-to-side asymmetries, or tracking tendinopathy rehabilitation. However, it is not suitable for highly dynamic, continuous movement tasks where maintaining consistent acoustic coupling is unfeasible, nor is it applicable during the acute phase of full tendon ruptures where baseline tension is absent. Furthermore, practitioners should note that due to the saturation effect of shear wave propagation under extreme tissue tension, absolute measurement precision may be reduced at extreme ranges of motion (e.g., maximal dorsiflexion).

Protocol

This study was approved by the Research Ethics Committee of Beijing Sport University (Approval number: 2025608H), and all procedures were conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent for study participation and publication of anonymized images.

Participant preparation

Recruitment and eligibility

Participants were recruited from national-level sports teams and included professional male athletes aged 18–26 years across multiple sport disciplines (e.g., sprinting, tennis, basketball). Participants were screened to ensure a normal body mass index (BMI)¹⁶. The dominant leg was determined by asking participants to kick a ball.

Inclusion and exclusion criteria

Participants met the following inclusion criteria: male sex, normal BMI, and national-level sporting qualification. Exclusion criteria included a history of ankle injury or surgery, neurological or systemic disease, acute musculoskeletal pain or inflammation involving the Achilles tendon or surrounding structures, and self-reported anabolic drug use.

Testing environment and pretest instructions

All measurements were conducted under standardized laboratory conditions using the same testing room and examiners for all participants. Participants were instructed to avoid high-intensity exercise for 48 h prior to testing¹⁷.

Equipment components and connections

A portable vibration-based ultrasound elastography system was used in this study. The specific commercial products and software used are detailed in the Table of Materials. The system consisted of four main components: (1) a main unit with integrated system software (version 1.0), (2) a linear-array ultrasound transducer, (3) an external excitation module, and (4) an L15 vibration head.

The linear-array transducer was a 128-element probe with a nominal central frequency of 100 Hz and an amplitude of 1 mm, designed for high-resolution imaging of superficial musculoskeletal tissues. The excitation module, together with the L15 vibration head, generated low-frequency mechanical vibrations (15 ± 2 mm), which were transmitted to the tissue to induce mechanically propagating waves. Tissue motion resulting from wave propagation was tracked by the ultrasound system, and stiffness-related parameters were derived using the system’s built-in analysis software.

The transducer was connected to the main unit by aligning the connector with the corresponding interface on the rear panel of the main unit, inserting it firmly until it locked into place with the connector buttons fully engaged and flush with the probe housing, and gently pulling on the transducer cable to confirm a secure connection. The excitation module was connected to the designated socket located on the lower left side of the main unit by aligning the locking connector, inserting it fully, and manually tightening the locking mechanism to ensure a stable mechanical and electrical connection. The system was powered on by switching on the main power supply and confirming that the system status indicator illuminated, followed by powering on the tablet interface, launching the ultrasound system software by selecting the designated application icon, and verifying that the system entered the main ultrasound operating interface with real-time B-mode imaging displayed.

Shear elastic modulus (G) acquisition

Transducer preparation and placement

A uniform layer of prewarmed coupling gel was applied to the transducer surface, and the probe was lightly placed against the measurement site with the target point aligned under the anterior side of the probe. Imaging quality was confirmed prior to acquisition, ensuring that the transducer plane was nearly perpendicular to the skin surface (>75°), the transducer-to-skin distance was approximately 5 mm, no visible air bubbles were present, and that the fascia and tendon fibers were clearly visualized.

Excitation module configuration

The elastography mode (E-mode) parameters were set to a frequency of 7.5 MHz, 4 acquisition lines, a 5 mm depth range, and a 300 ms acquisition time. The excitation module was activated, and the excitation tip was positioned 3–6 mm in front of the probe’s protrusion side, perpendicular to the probe imaging plane.

E-mode imaging and depth adjustment

The ultrasound system was switched to E-mode, and the reference line was positioned such that the acquisition depth range began just below the superficial tendon fascia. The region of interest (ROI) was adjusted to cover the tendon thickness while strictly avoiding the skin, subcutaneous tissue, and Kager’s fat pad.

Data acquisition and quality control

Continuous measurement was initiated by clicking the On button, and the system automatically calculated the shear modulus (G), providing mean ± SD values of valid data. Participant and operator posture were maintained constant during acquisition to obtain at least 10 valid continuous data points. Data acquisition was stopped by pressing the Freeze function once sufficient data points were collected. The dataset was reviewed for outliers, and abnormal data points were removed using the system’s editing function.

Measurements were repeated at least three times at each ankle angle. A measurement was considered valid only if the standard deviation (SD) of the continuous data points was less than 10% of the mean, in accordance with the device’s internal validity requirements; otherwise, the measurement was discarded and repeated. B-mode images and mechanical imaging maps were saved for documentation (Figure 1).

Figure 1. Schematic representation of the experimental setup and functional stiffness spectrum acquisition protocol. (A) Experimental setup. (B) Specific measurement zones on the Achilles tendon. (C) Ankle joint angles in the experimental sequence. Abbreviations: PF = plantarflexion, DF = dorsiflexion. Please click here to view a larger version of this figure.

Data acquisition procedure

Subject registration and anatomical localization

Participant demographic and athletic information were recorded upon arrival. Participants were instructed to remove their shoes and socks and lie prone on the examination couch with their ankles fully extended over the edge by approximately 5 cm. The superior apex of the calcaneal tuberosity was located via palpation, and a point 5 cm proximal to this landmark was marked using a skin marker to define the initial measurement site. The marked site was verified using ultrasound imaging in the longitudinal view.

Baseline measurement

The initial stiffness acquisition was performed at the baseline state (no-boot relaxed state) following the procedures described above.

Multi-angle measurement (functional stiffness spectrum)

Measurements were conducted sequentially on both Achilles tendons under the following conditions: relaxed, 0° (neutral), 20° plantarflexion (PF), 40° PF, 20° dorsiflexion (DF), and 40° DF. A randomized testing order was intentionally avoided, as testing an extreme dorsiflexion position prior to plantarflexion positions would induce tissue hysteresis and pre-conditioning, artificially altering baseline mechanics and affecting subsequent measurements.

Figure 2. Representative interface of the system during data acquisition. The central panel displays a longitudinal B-mode ultrasound image of the Achilles tendon, showing clear, parallel fiber alignment. The yellow panel on the right displays real-time quantification of the shear elastic modulus (G). The system automatically calculates the mean value (20.46 kPa in this example) and standard deviation (0.37 kPa) from the list of valid measurements shown below. This readout demonstrates high measurement stability with a low standard deviation (SD < 10% of the mean), satisfying the protocol's quality control criteria. Please click here to view a larger version of this figure.

Boot installation and angle setting

The participant's foot was placed into the adjustable ankle testing boot, ensuring the heel rested completely flush against the posterior heel cup of the boot base. The forefoot, midfoot, and lower leg were secured using the attached hook-and-loop straps to prevent heel lift or lateral shifting during testing. The bilateral locking knobs on the boot’s hinge mechanism were loosened, and the ankle was manually guided to the target angle by aligning structural markers with the goniometric scale. The locking knobs were then firmly tightened to secure the ankle joint at the target angle. Ultrasound measurement was performed immediately after locking the angle to prevent viscoelastic tendon relaxation.

Post-procedure

Participants were instructed to remove the ankle boot, and all instruments and ultrasound probes were cleaned and sanitized.

Data processing and statistical analysis

Data aggregation

For each measurement trial, the internal SD of the data points was verified to be <10% of the mean. The inter-trial coefficient of variation (CV) across the three valid trials was calculated for each measurement angle and was required to be <30%; otherwise, the dataset was discarded and re-measured. The overall mean of the three successful trials was calculated and used for subsequent analyses.

Statistical modeling

The intraclass correlation coefficient (ICC) was calculated to evaluate measurement reproducibility. The effects of variables on Achilles tendon stiffness were analyzed using a Generalized Mixed Models (GLMM). Achilles tendon stiffness (G) was specified as the dependent variable, with ankle joint angle, sport type, and dominant leg as fixed factors. Subject ID was included as a random effect to account for repeated measures. Post-hoc analyses with Bonferroni correction were conducted.

Data visualization

Processed data were exported and visualized using line graphs for stiffness spectrum analysis and bar charts for group comparisons.

Results

Before interpreting the statistical outcomes, it was critical to define the criteria for a successful versus failed implementation of this protocol. Participant demographic characteristics are presented in Table 1. A successful measurement was visually characterized by a high-quality B-mode image displaying a clear, continuous tendon fibrillar structure parallel to the skin surface, coupled with a stable, homogeneous elastography color map within the predefined Region of Interest (ROI) (as shown in Figure 2). Quantitatively, success was achieved when the continuous data points within a single capture yielded a coefficient of variation (CV) of <30%. Conversely, a failed implementation was typically indicated by poor acoustic coupling (resulting in dark voids or signal dropouts in the elastography map), motion artifacts, or excessive operator-induced probe pressure, which artificially stiffened the superficial tissue. Any trial exhibiting a CV ≥ 30%, or displaying discontinuous elastography filling, constituted a technical failure and necessitated immediate probe repositioning and re-testing.

	Basketball	Volleyball	Football	Tennis	Sprint	Long-distance running	Kruskal-Wallis p
Age (year)	22.2±2.29	20.6±1.59	21.1±2.42	20.9±2.63	21.7±3	21.1±2.2	0.639
Height (m)	1.87±0.09	1.87±0.05	1.77±0.05	1.81±0.05	1.76±0.05	1.75±0.05	<0.001
Weight (kg)	81.8±9.91	76±8.71	69.1±6.45	70.6±4.39	70.6±5	65.9±5.86	<0.001
BMI	23.4±1.48	21.7±1.84	22.1±1.69	21.5±1.34	22.8±1.34	21.5±1.67	0.033
Training Frequency (times per week)	4.75±2.67	4.89±1.54	5.1±1.10	4.82±1.40	5.89±1.83	6±0.87	0.125
Sport age (year)	9.75±4.25	6.22±2.91	10.8±2.62	9.73±4.13	3.44±1.74	5.5±3.89	<0.001

Table 1: Demographic characteristics of athletes.

Intratrial reliability and precision

The internal precision of the protocol was assessed by calculating the coefficient of variation (CV) for Achilles tendon shear elastic modulus (G) across all measurement conditions (6 joint angles × 2 limbs × N participants). Average CV values ranged from 14.0% to 25.2% across different joint angles (Table 2). Notably, the measurement variability exhibited an angle-dependent pattern: CV values remained lower and highly stable during resting and plantarflexion (PF) states, but systematically increased as the ankle was positioned into extreme dorsiflexion (DF).

Furthermore, intra-session reproducibility across consecutive measurement trials was evaluated using the intraclass correlation coefficient (ICC). The results demonstrated good to excellent relative reliability across all assessed joint angles. Specifically, ICC (2,1) values ranged from 0.871 to 0.974 (Table 2), with the highest reliability observed in the relaxed state (ICC = 0.974, 95% CI: 0.943–0.990) and the lowest, yet still robust, reliability at the neutral 0° position (ICC = 0.871, 95% CI: 0.751–0.939). Together with the CV data, these findings confirmed the overall biomechanical robustness and stability of the multi-angle measurement protocol.

	Relax	PF 40	PF 20	0	DF 20	DF 40
Average CV	0.16	0.14	0.16	0.25	0.24	0.25
ICC(2,1)	0.974	0.962	0.925	0.871	0.957	0.965
95% CI for ICC	[0.943, 0.990]	[0.930, 0.980]	[0.847, 0.967]	[0.751, 0.939]	[0.927, 0.976]	[0.933, 0.983]

Table 2: Measurement reliability (intraclass correlation coefficient) and internal precision (coefficient of variation) of Achilles tendon shear elastic modulus across distinct joint angles.

Functional stiffness of the Achilles tendon

Achilles tendon stiffness (G) was quantified across six ankle joint angles for both dominant and non-dominant limbs.Generalized Mixed Models (GLMM) results for fixed effects are summarized in Table 3. Achilles tendon stiffness across the functional range of motion was successfully quantified. As expected, tendon stiffness increased non-linearly from plantarflexion (slack) to dorsiflexion (tension) in all participants (see Figure 3).

GLMM revealed a significant main effect of joint angle (p < 0.001), whereas side (dominant vs. non-dominant) and sport type showed no main effects. The Angle × Sport interaction was significant (p = 0.049), indicating sport-specific stiffness differences at certain ankle angles. To substantiate these differences, post-hoc simple effects analyses were conducted. The differences were most pronounced at 20° plantarflexion (PF20), where both Basketball (203 ± 187 kPa; p = 0.046, Cohen’s d = 0.58) and long-distance running athletes (188 ± 138 kPa; p = 0.048, Cohen’s d = 0.62) exhibited significantly higher tendon stiffness compared to Tennis athletes (122 ± 62 kPa). Furthermore, at the neutral position (0°), Basketball athletes (1033 ± 912 kPa) maintained significantly higher stiffness than Tennis athletes (574 ± 382 kPa; p = 0.008, Cohen’s d = 0.66). Conversely, at 40° dorsiflexion (DF40), no significant differences were observed among sports, suggesting convergence of mechanical properties under maximal tendon load.

Factor	X²	df	p
Angle	8964.919	5	< .001
Side (Dominant/Non-dominant)	0.47	1	0.493
Sport	4.423	5	0.49
Angle × Side	1.715	5	0.887
Side × Sport	10.182	5	0.07
Angle × Sport	37.788	25	0.049
Angle × Side × Sport	26.065	25	0.404

Table 3: Fixed effects test results of the Generalized Mixed Models (GLMM).

Figure 3. Functional stiffness spectrum of the Achilles tendon across different ankle joint angles. Data are presented as mean ± SD. The X-axis represents the ankle joint position, ranging from slack states (relaxed, plantarflexion [PF]) to tensioned states (neutral 0°, dorsiflexion [DF]). The Y-axis represents the shear elastic modulus (stiffness) plotted on a log10 scale. Shear modulus increased non-linearly with increasing dorsiflexion. No significant main effect of side dominance or Angle × Side interaction was found (p > 0.05), indicating overall functional symmetry between dominant and non-dominant tendons across the tested range. Asterisks (*) indicate a significant difference (p < 0.05) compared to the reference group (Tennis) based on GLMM parameter estimates. Please click here to view a larger version of this figure.

Supplemental Figure S1. Achilles tendon shear modulus across joint angles comparing the left and right sides. Data are presented as mean ± SD. The X-axis represents the ankle joint angle, ranging from slack positions (Relax, Plantarflexion) to tensioned positions (Neutral 0°, Dorsiflexion/Extension). The Y-axis represents the shear modulus (stiffness) plotted on a log10 scale. Shear modulus increased non-linearly with increasing dorsiflexion. A significant main effect was observed only for joint angle, while no significant main effects were found for side or sport. Furthermore, a significant Angle × Side interaction was detected, whereas all other interaction effects remained non-significant. * indicates a significant difference (p < 0.05) between the left and right sides in Neutral 0° based on GLMM parameter estimates. Abbreviations: PF = plantarflexion; DF = dorsiflexion.Please click here to download this file.

Discussion

This study presented a standardized protocol for quantifying the functional stiffness spectrum of the Achilles tendon in elite male athletes using a portable force–ultrasound fusion device. Unlike conventional anatomical imaging, which offers limited functional insight, this method utilized vibration-based ultrasound elastography to non-invasively map the mechanical properties of the tendon across a physiological range of ankle joint angles. The total testing duration was approximately 10–20 min per subject, and the automated extraction of elastic modulus values made this protocol a practical solution for longitudinal monitoring in both laboratory and field-based sports settings. However, as with any multi-angle assessment, the inherent viscoelastic properties of the Achilles tendon—specifically susceptibility to creep, hysteresis, and stress relaxation—must be carefully managed. While the entire session lasted 10–20 min, this included setup, anatomical landmarking, and boot installation. The actual time spent at each joint angle was brief (typically under 1 min). Furthermore, the mechanical vibration applied was transient (300 ms per acquisition window) rather than continuous, minimizing the risk of accumulated mechanical fatigue. To mitigate stress relaxation, the protocol mandated that data acquisition occurred immediately upon locking the ankle joint to capture the instantaneous stiffness before viscoelastic creep could alter tissue mechanics. Nonetheless, future implementations involving more extensive repeated loading should remain cognizant of these time-dependent properties.

To ensure reproducibility of the stiffness spectrum, strict adherence to specific acquisition details was required. First, the application of a sufficient layer of acoustic gel was critical to prevent air-interface reverberation artifacts, which can degrade shear wave signal quality. Second, the timing of measurement was a decisive factor. Due to the viscoelastic nature of the tendon¹⁸, stress relaxation occurred immediately after the ankle was locked into a new position. Therefore, the protocol required acquisition to commence immediately upon angle fixation to capture the instantaneous stiffness response rather than the relaxed state.

A detailed analysis of intra-trial repeatability revealed a pattern of variability dependent on joint angle. Lower coefficients of variation (CVs, ~14–16%) were observed in plantarflexed positions (slack state), whereas higher CVs (~24–25%) were observed in dorsiflexed positions (tensioned state). This trend likely reflected the technical challenges associated with quantifying tissue mechanics at the upper limits of stiffness. In the tensioned state (dorsiflexion), tendon stiffness increased non-linearly, causing shear waves to propagate at high velocities¹⁹. This may approach the detection limits of the portable device. Consequently, higher CV values in dorsiflexed positions reflected the complex acoustic properties of highly tensed anisotropic tissue rather than methodological unreliability or operator error. Recognizing this inherent variability was important for establishing ecologically valid quality control thresholds. Enforcing a stricter threshold (e.g., CV < 20%) across all angles would necessitate excessive re-testing in extreme stretch positions, potentially introducing physiological artifacts such as viscoelastic creep and stress relaxation. Therefore, a CV threshold of < 30% was considered a pragmatic compromise for multi-angle in vivo testing. Nonetheless, operators were required to maintain probe stability when assessing the tendon in high-tension positions to minimize additional variability.

For researchers and clinicians, elevated CV values (>20%) in extreme dorsiflexion indicate that absolute stiffness values at these angles should be interpreted with caution. This suggests that the method is most suitable for tracking intra-individual longitudinal changes rather than relying solely on cross-sectional inter-individual comparisons at maximal tension. To further reduce variability, future protocol refinements may include the use of external stabilization approaches (e.g., customized supports) to standardize probe pressure and orientation. However, any stabilization strategy must allow rapid angle adjustment to maintain the balance between mechanical stability and minimizing viscoelastic creep.

The proposed Functional Stiffness Spectrum offers a methodological advancement over traditional isokinetic dynamometry. While dynamometry is considered a reference method for assessing the global mechanical properties of the muscle–tendon unit, it cannot isolate the local stiffness of the free tendon from muscular contributions. By directly assessing the free Achilles tendon, this protocol provides a localized, tissue-specific measurement. This capability may be useful for detecting localized changes in tendon stiffness among asymptomatic individuals undergoing targeted loading programs⁴. Furthermore, in pathological or tendinopathic populations, localized structural changes may alter stiffness before global muscle–tendon unit deficits become apparent²⁰. This method therefore enables detection of localized mechanical alterations that may not be captured by global testing approaches.

By quantifying the non-linear increase in stiffness from plantarflexion to dorsiflexion, this method captured tendon mechanical behavior under functionally relevant loading conditions. The angle–stiffness relationship observed in Figure 3 did not conform to a simple quadratic model, reflecting the non-linear physiological behavior of tendon tissue across a wide range of motion. The remarkable exponential increase in stiffness between PF20° and 0° corresponds to the classical ‘toe region,’ where crimped collagen fibers are rapidly straightened. It is important to note that the visually flattening appearance of the curve at higher dorsiflexion angles is influenced by the log10 scale used for data visualization. In absolute terms, stiffness continues to increase substantially, reflecting progressive strain-stiffening under high mechanical tension. These characteristics highlight the complex, non-linear mechanical behavior of tendon tissue across a wide physiological range. The increase in stiffness between PF20° and 0° corresponded to the initial straightening of collagen fibers, while continued increases at higher dorsiflexion angles reflected progressive strain-stiffening under tension. These findings support the use of a multi-angle assessment rather than single-point estimation.

Regarding statistical outcomes, GLMM confirmed a significant main effect of joint angle, supporting the sensitivity of the protocol to changes in mechanical loading. No main effects or interactions were observed for limb dominance, suggesting functional symmetry in Achilles tendon stiffness across joint angles²¹. This is consistent with biomechanical requirements for balanced force transmission and energy storage during locomotion²². However, supplementary analyses based on anatomical laterality (left vs. right) indicated side-specific differences in certain conditions, suggesting that functional symmetry may be maintained despite underlying structural asymmetries²³.

Several limitations should be considered. First, the study was limited to young elite male athletes, and future research should evaluate broader populations, including females, older adults, and symptomatic individuals. Second, measurement precision decreased in positions of maximal tension due to the physical limits of shear wave propagation. However, this did not reduce reliability to an unacceptable level, as averaging three trials produced high reproducibility (ICC > 0.87). Third, the protocol employed a static multi-angle approach rather than continuous dynamic measurement, and therefore does not replicate high-velocity loading conditions. Additionally, measurements were performed under passive conditions and did not account for the effects of active muscle contraction. Finally, this method characterizes local shear-elastic behavior under transverse vibration and should not be interpreted as a direct measure of longitudinal tensile stiffness.

In conclusion, when multiple-trial averaging (minimum three repetitions) and real-time quality control (CV < 30%) were applied, this standardized multi-angle protocol provided a reliable and practical tool for assessing Achilles tendon mechanics. By capturing tendon responses across a range of loading states, it enabled monitoring of bilateral symmetry and training adaptations. This method may support athlete monitoring and early identification of mechanical changes associated with tendinopathy when used for longitudinal assessment.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Boots	Ober	AO-36	use as suggested in protocol
Coupling Gel	Jinya Technology	TM-100	use as suggested in protocol
Excel	Microsoft	https://www.microsoft.com/microsoft-365/excel	Used by authors to data arrangement
Jamovi	The jamovi project	https://www.jamovi.org/	Used by authors to statistical analysis
Portable Ultrasound  Device	XiJian Technology	T5C1B101WT	use as suggested in protocol
Prism	Graphpad	N/A; https://www.graphpad.com	Used by authors to visualization
SPSS	IBM	https://www.ibm.com/products/spss-statistics	Used by authors to statistical analysis