Shear-Wave Elastography of the Achilles tendon: reliability analysis and impact of parameters modulating elasticity values

Abstract

Purpose

Shear wave elastography (SWE) has seen many advancements in Achilles tendon evaluation in recent years, yet standardization of this technique is still problematic due to the lack of knowledge regarding the optimal way to perform the examination. The purpose of this study was to evaluate the effects of ankle position, probe frequency and physical effort on the shear modulus of the Achilles tendon, but also to determine the intra and inter-observer reliability of the technique.

Methods

37 healthy volunteers were included; SWE protocol was performed by two examiners. We analyzed the shear modulus of the tendon with the ankle in neutral, maximum dorsiflexion and maximum plantar flexion using two different high frequency probes. Afterwards, the subjects performed a brief physical exercise and SWE measurements were repeated.

Results

The L18-5 probe showed the highest ICC values (ICC = 0.798, 95% CI 0.660–0.880, p < 0.001) when positioned at 2 cm from the calcaneal insertion with the ankle in a neutral state. Conversely, utilizing the same L18-5 probe at 1 cm from the insertion during maximum plantar flexion of the ankle resulted in the lowest ICC (ICC = 0.422, 95% CI 0.032–0.655, p = 0.019). Significant variations in elasticity values were noted among different ankle positions and probe types, while no significant changes in elasticity were observed post-physical exercise.

Conclusion

Ankle position and probe frequency are factors that influence elasticity values of the Achilles tendon. An ankle position between 10 and 20 degrees of plantar flexion is the most suitable for SWE evaluation. However, more research focusing on Achilles tendon SWE is essential due to the challenges encountered in standardizing this region.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40477-024-00877-w.

Introduction

Shear-wave elastography (SWE) is an imaging technique that has been extensively studied in the recent years for its potential applications in the musculoskeletal (MSK) system. Tendons were more extensively studied due to their excellent visualization through ultrasound (US) techniques [1–3]. Unlike strain elastography which utilizes manual compression applied by the examiner to determine a qualitative assessment of tissue elasticity [4], SWE allows the ultrasound transducer to both emit a pulse wave and ulteriorly use tracking waves to quantify the degree of deformation of underlying tissues [5]. The Achilles tendon is the largest tendon in the human body, but despite its capability to withstand extreme tensile forces, it is also one of the most commonly injured [6, 7].

While many studies have focused on the potential applications of SWE in Achilles tendon examination [8–10] there is a lack of data regarding the standardization of the technique. Furthermore, a recent systematic review [11] raised concerns about the high variability of SWE measurements, specifically when the Achilles tendon was involved.

It has been described that in the case of tendons, examination in the longitudinal plane is more suitable due to better propagation of shear waves along this axis, while transducer pressure, degree of contraction and insonation angle are known as potential pitfalls [12, 13]. Due to their highly anisotropic nature, probe positioning also plays a very important role in SWE evaluation of tendons [14, 15], as well as ankle positioning in the case of the Achilles tendon [1], as even small shifts in joint position may alter elasticity values [16–18]. The examination is best performed with the probe placed directly on the skin, using minimal pressure, the size of the region of interest (ROI) not influencing the elasticity values [19] Standardization of joint position by utilizing ankle fixators throughout the examination has been reported as a means of obtaining higher reproducibility [20, 21], yet complete negation of anisotropy is not always possible. SWE measurements of MSK structures are further hindered when linear probes are used at target depths deeper than 2–3 cm or in extremely superficial structures [12, 13, 22, 23]. Lastly, as is the case for all applications of SWE, variability between machines and probes has been described [13]; therefore, values are not transposable.

While SWE in the use of tendon pathology seems to be a promising tool, it is clear that further standardization of this method is necessary as accurate and reproducible measurements are essential for diagnosis. Therefore, the purpose of this study was to determine the effects of ankle position, physical exercise and probe frequency on the shear modulus, but also to determine the reliability of this method in the evaluation of the Achilles tendon.

Materials and methods

Patients

Between May and July of 2022, 37 healthy volunteers, ages between 18 and 40 years were recruited. Exclusion criteria were previous diagnosis of MSK systemic inflammatory disease, surgery of the ankle joint, history of pain or trauma, and recent use of steroid medication. Clinical and B-Mode US examination was performed to exclude potential pathology.

The US and SWE examinations were performed using a SuperSonic™ MACH™ 30 machine (SuperSonic Imagine, Aix-en-Provence, France) equipped with linear L18-5 and L20-6 transducers. The US and SWE were performed by two examiners with 2, respectively 7 years of MSK ultrasound experience, using a blinded protocol.

B-mode US and SWE examination

The patient was positioned lying prone with their feet hanging off the edge of the bed. The examination protocol encompassed three distinct ankle joint positions: neutral position, maximum passive dorsiflexion, and maximum passive plantar flexion.

The protocol for B-mode examination was the following: thickness of the tendon was determined in two points at 1 and 2 cm distance from the superior part of calcaneus. Cross-sectional area (CSA) was obtained by measuring the point of maximum thickness. Length was measured for the free part of the tendon from the distal tip of the soleus muscle to the insertion on the superior edge of the calcaneus.

The SWE examination was performed after a period of at least 10 min of complete relaxation of the tendon, using the L18-5 linear probe, followed by the L20-6 hockey stick probe with the default MSK preset selected. Depth of the Q-box was set at 1 cm. After obtaining a homogenous color map, three measurements were performed, and the mean value was used for further analysis. Between each acquisition the probe was lifted from the patient’s skin. The SWE protocol was the following: (1) Examination with the ankle in neutral position, L18-5 probe—The angle of the joint placed at 10° plantar flexion, measured using a goniometer). SWE measurements were conducted with the probe having direct skin contact, without applying pressure and using a sufficient amount of gel. The measurements were taken at two distances—1 cm and 2 cm—from the superior part of the calcaneus. The Q-Box was positioned 0.5 cm below the probe footprint, and the size of the ROI was set to 2 mm. (2) Examination with the ankle in neutral position, L20-6 probe—SWE acquisitions were made at 0.2, 0.5 and 1 cm from the insertion of the tendon on the superior edge of the calcaneus with the ROI diameter set to 2 mm. (3) With the ankle in passive dorsiflexion and plantar flexion, L18-5 probe—Maximum dorsiflexion was obtained by positioning the patient’s ankle in the desired position and fastening with a strap attached to the examination bed, while maximum plantar flexion was obtained by asking the patient to place the back of the foot on a support attached to the end of the bed. Measurements were made 1 and 2 cm from the insertion with the Q-Box at 0.5 cm below the probe foot print and with the ROI set to 2 mm. (4) Examination after physical exercise, L18-5 probe—the subjects were asked to perform 15 heel raises and immediately after that the measurements in neutral position were repeated—ROI of 2 mm, Q-Box at 0.5 cm below the probe footprint at 1 cm and 2 cm from the tendon insertion.

The protocol was repeated by the two examiners in 3 consecutive days for all subjects, with the same conditions (Fig. 1). Total time of examination for both ankles of a subject was approximately 20 min.

Fig. 1.

SWE examination of the Achilles tendon at 1 cm and 2 cm from the calcaneus in the neutral position (a), maximum dorsiflexion (b) and maximum plantar flexion (c) with the L18-5 probe. Examination with the L20-6 hockey stick probe in the neutral position at 0.2, 0.5 and 1 cm from the calcaneus (d)

Statistical analysis

The distribution of the numerical variables was assessed using the Shapiro-Wilks test. All the numerical variables were non-normally distributed; thus, they were presented as median and interquartile range (IQR). The categorical variables were presented as number and percent. The Related Samples Wilcoxon Signed Rank Test was used to compare the differences between medians of the repeated measurements and Independent-Samples Mann–Whitney U test to compare the medians of the independent samples (elasticity of right and left tendons, or dominant, and non-dominant tendons). The intraclass correlation coefficient (ICC) for average measures (the average of the three measurements performed in the same session by the same examiner, first session of measures was chosen for each examiner) using the Two-way mixed effects model and consistency was calculated for the inter-observer agreement assessment. The ICC for single measures (three different measurements performed in three different days by each examiner) using the One-way random effects model was calculated to assess the intra-observer agreement. The ICC below 0.5 was categorized as poor reliability, between 0.5 and 0.75 moderate reliability, between 0.75 and 0.9 good reliability, and above 0.9 excellent reliability [24]. A p value below 0.05 was considered significant. The statistical analysis was performed using IBM SPSS Statistics v23.

Results

Among the subjects 26 were females (70.3%) with a median age of 26 years (25.0–30.0), median BMI of 22.8 kg/m² (20.3–24.6) and median physical activity of 2.0 (0–4.0) hours per week. Measurements in B-Mode US of the Achilles tendon are presented in Table 1.

Table 1.

GSUS characteristics of the Achilles tendons. (N = 74)

Parameter	Median (IQR)
Achilles tendon
Thickness 1 cm (mm)	4.50 (4.10–5.10)
Thickness 2 cm (mm)	4.60 (4.20–5.00)
Length (mm)	51.90 (43.40–55.00)
Area (mm2)	62.0 (55.0–71.0)

N number of examined tendons, IQR interquartile range

Inter-observer agreement

Overall agreement with the L18-5 probe was excellent ICC = 0.968, 95% CI (0.962–0.975), (p < 0.001) and good ICC = 0.801, 95% CI (0.732–0.851), (p < 0.001) with the L20-6 probe. ICC inter-observer values for the Achilles tendon varying with ankle position and probes are presented in online resource 1.

Intra-observer agreement

ICC values for intra-observer agreement utilizing the two different probes are presented in Table 2. Values for variations with ankle angle and probe are presented in online resource 2.

Table 2.

Overall intra-observer agreement for both examiners

L18-5	L20-6
Examiner 1
0.979 (0.975–0.982)	0.558 (0.485–0.627)
Examiner 2
0.977 (0.973–0.981)	0.502 (0.416–0.582)

Factors influencing tendon elasticity

L18-5 transducer, neutral position

Median values were 114.22 kPa (81.57–158.47) and 136.88 kPa (109.00–172.97) at 1 cm and 2 cm distance from the enthesis, p < 0.001.

Achilles tendon median values after exercise at 1 cm and 2 cm from the insertion were 110.67 (84.50–159.10) kPa and 137.00 (109.67–181.13) kPa, respectively, without statistically significant difference from the corresponding medians of measurements in the relaxed condition of the tendon (p = 0.848 and p = 0.641, respectively).

When comparing the median of elasticity values of the right and left tendon, values were higher for the right tendon, the differences being statistically significant [154.1 (88.0–189.6) vs 111.5 (81.6–139.1) kPa, p = 0.019, and 162.7 (132.9–205.8) vs 123.9 (104.3–149.0) kPa, p = 0.013] at 1 cm, respectively 2 cm from the tendon insertion. The intra-observer agreement between the right and left tendon measurement in neutral position was moderate for both examiners [examiner 1: ICC = 0.673, 95% CI (0.481–0.794), p < 0.001; examiner 2: ICC = 0.698, 95% CI (0.492–0.830), p < 0.001].

Out of 37 subjects, 30 subjects (81.1%) had dominant right lower limb with median elasticity values of 143.6 (81.3–178.367) and 148.667 (119.733–204.3) kPa at 1 and 2 cm, respectively, while values for the non-dominant limb were 113.3 (82.8–144.633) and 134.767 (109.0–162.767) kPa at 1 and 2 cm, respectively, the difference not being statistically significant (p = 0.282 and 0.146, respectively).

L18-5 transducer, passive dorsiflexion and plantar flexion

Median elasticity values with variations in ankle position are presented in Table 3.

Table 3.

Median values of Achilles tendon elasticity varying with ankle position

Parameter	Median (IQR)	p value*
Dorsiflexion 1 cm	1080.65 (914.17–1182.50)	< 0.001
Dorsiflexion 2 cm	1111.85 (985.93–1194.27)	< 0.001
Plantar Flexion 1 cm	71.10 (49.20–96.63)	< 0.001
Plantar Flexion 2 cm	83.65 (56.93–128.50)	< 0.001

*p value obtained by comparing the medians with the corresponding median calculated for the neutral position in the same portion of the Achilles tendon with the same probe

L20-6 transducer, neutral position.

Median elasticity values for the L20-6 transducer at 0.2, 0.5 and 1 cm from the insertion on the calcaneum were 402.12 (341.83–507.10) kPa, 299.82 kPa (230.30–393.50) kPa and 215.30 (159.50–290.77) kPa respectively, the differences between the medians obtained at these different levels being statistically significant (p < 0.001), but also when compared to the median obtained at the same level with the L18-5 probe (p < 0.001).

Discussion

In our study, the highest inter-observer reliability was obtained using the L18-5 probe with the ankle in a neutral position at 2 cm from the insertion on the superior edge of the calcaneus, while the lowest was in plantar flexion also at 2 cm from the insertion. Intra-observer agreement was moderate to good with the L20-6 probe, the lowest reliability with this probe being obtained for measurements near the bone proximity. A small but statistically significant difference in elasticity was found between measurements made at 1 cm and 2 cm from the insertion, but also between measurements made with the ankle in different degrees of plantar flexion and dorsiflexion, while physical exercise did not significantly influence these values.

Inter and intra-operator reliability reported in literature [15, 25, 26] was similar to that obtained in our study, yet it is important to note that even though the elasticity values were measured with the ankle in a neutral position, similar to our protocol, measurements were made in different regions of the tendon. The most similar protocol to that which we used was described in Aubry et al. [27] where four different ankle positions were used with both flexion and extension, yet reliability in this study was much lower compared to ours. While their study holds significance in Achilles tendon SWE, it’s important to recognize that advancements in SWE technology since its publication may have considerably improved the technique's applicability to superficial and rigid structures like tendons.

When analyzed separately, the lowest ICC obtained by us with the L18-5 probe was in passive plantar flexion. This is most likely explained by the fact that this position allows for the accentuation of the tendon’s natural curvatures, consequently increasing anisotropy and resulting in more inhomogeneous color maps. Similar challenges were encountered in our previous study [19] on the patellar tendon, where areas of anisotropy led to comparable difficulties. In the case of the L20-6 probe the lowest ICC was obtained at 0.2 cm from the tendon’s insertion; this is not at all surprising as measurements made at such a small distance from a boney surface have been described [28] to be prone to the so called “bone proximity” artifact. This occurs whenever tissues appear next to extremely dense structures, such as bones, where homogenous propagation of shear waves is prevented and can also lead to an increase in elasticity values.

In our study, while some elasticity values for the Achilles tendon aligned with those reported in previous literature [10], a notable divergence emerged from other studies conducted under similar conditions and equipment setups [8, 20, 27, 29–32]. Comparing outcomes reported in units such as meters per second (m/s) proved inappropriate for direct comparison.

The discordance in findings lacks a clear elucidation. Factors encompassing demographic variations or localized measurements within distinct tendon segments offer inadequate explanations for these discrepancies. Schneebeli et al.’s hypothesis [30] posited that maintaining the ankle at 0 degrees may sustain stress on the Achilles tendon, potentially rendering this position less optimal for accurately assessing tendon elasticity. This perspective finds support in the suggestion that an ankle positioned between 15° and 25° plantar flexion could alleviate Achilles tendon tension [33]. Notably, our study revealed significantly reduced elasticity values during maximum plantar flexion in comparison to the neutral position.

An intriguing consideration arises regarding the influence of ankle positioning on SWE assessments of the Achilles tendon. While the 0-degree ankle joint position may not achieve complete tendon relaxation, potentially limiting the accuracy of SWE measurements due to persistent tension, an alternative, such as ankle plantar flexion, might introduce anisotropic effects.

The potential for plantar flexion to induce anisotropy within the Achilles tendon poses a noteworthy concern for SWE examinations. Alterations in tendon alignment or orientation as a result of plantar flexion could lead to varying wave propagation characteristics, potentially complicating the interpretation and reliability of SWE data.

However, considering our findings and acknowledging this theoretical backdrop, an optimal compromise emerges. It is conceivable that a slight degree of plantar flexion could achieve adequate relaxation of the tendon, eliminating excessive tension while minimizing the potential introduction of considerable anisotropic effects. This fine balance could potentially offer a more favorable condition for SWE acquisitions, providing an equilibrium between tension relief and minimized anisotropy, thereby optimizing the accuracy and reliability of Achilles tendon assessments using SWE.

Measurements taken at two regions of the tendon revealed higher values in the area nearer to the insertion. The observed variation may stem from heightened mechanical stress at the tendon-bone attachment alongside the relatively limited stretching capacity in this region compared to the muscle–tendon junction [34]. This correlation aligns with the well-established pattern of tendinopathies predominantly occurring closer to the tendon insertion [35, 36].

When measurements were made with the ankle in maximum dorsiflexion rapid saturation of the color box was obtained which led to the device reaching its upper limit of detection (1200 kPa), with elasticity values obtained being much higher with a statistically significant difference as compared to the neutral position. This is a natural response of the tendon as higher strain rates lead to alignment of collagen fibrils in the direction of load with the tendon becoming stiffer and less deformable, thus facilitating transmission of high loads [37].

There was a small but statistically significant difference between elasticity values of the right and left tendon, with the former being slightly higher, while comparison of dominant and non-dominant limbs did not yield a statistically significant difference in elasticity. Bohm et al. [38] investigated differences in mechanical properties between dominant and non-dominant limbs. They observed higher Achilles tendon stiffness on the dominant side, while other studies [10, 39] found no difference. Comparing these results is challenging due to heterogeneous study populations that included both athletic and non-athletic participants. Moreover, some of these studies compared left and right sides, while others dominant and non-dominant limbs.

In our study, an important observation was that during consecutive SWE measurements, the patient's right limb was notably positioned farthest from the examiner. This unique scenario posed challenges for the examiner during prolonged examinations, potentially influencing the observed minor disparities. The increased distance between the examiner and the right limb might have contributed to subtle variations, suggesting that examiner fatigue due to the prolonged and strenuous position could have affected the quality of the obtained images. This observation is significant, given the sensitivity of SWE measurements to factors such as probe angle and pressure.

In our previous study [19], we also encountered similar challenges when using the L20-6 hockey probe, which yielded markedly higher elasticity values compared to the L18-5 probe. Regarding SWE measurements employing high-frequency hockey stick probes, there is limited information available, particularly in Musculoskeletal (MSK) applications, utilizing such devices. A study [40] comparing various high-frequency linear probes in an in vivo setting revealed significant differences in elasticity values, indicating that higher frequencies led to increased shear modulus. These values may potentially be affected by the higher attenuation experienced by the transducer’s higher frequency push pulse. Further investigations using high-frequency probes are warranted, as they might be more suitable for assessing structures at shallower depths than most MSK applications. There was no statistically significant difference between measurements made after physical effort and with the tendon in a relaxed state. Data from literature [41] suggests that short term adaptation of tendons to load is a transient decrease in stiffness, while long term adaptation to chronic exercise can result in an increase in stiffness [42]. This was not the case in our study, however this could be caused by the fact that the exercise included in our protocol may not have been of sufficient intensity as to elicit a change in stiffness, and heel raises may not be as strenuous for the Achilles tendon as hopping or jumping [43].

Limitations of this study are related to the low number of subjects, heterogeneity of the study population regarding physical activity, both active and inactive individuals were included, but also the limited protocol regarding physical exercise. Regarding the SWE examination, a limiting factor was the lack of measurements in the free portion of the tendon and the less extensive measurements made with the hockey stick probe. Nevertheless, it was demonstrated that small changes in ankle position can significantly influence SWE values and that obtaining a standard position with no load on the Achilles tendon can be extremely challenging. Although physical exercise did not significantly alter elasticity values, it would still be advised to perform the examination after a short period of relaxation.

Conclusions

While SWE of the Achilles tendon is a reliable and reproducible technique, it is possible that its clinical applications may be limited due to the difficulty of proper standardization. Ankle position, region of measurement and the use of different ultrasound probes are all factors that can ultimately influence elasticity values. An ankle position between 10 and 20 degrees is most optimal for elasticity measurements, yet further studies are required for Achilles tendon SWE, as this region seems to be particularly problematic in terms of standardization.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Michael-Andrei Pelea, Oana Serban, Maria Badarinza and Roxana Rosca. The first draft of the manuscript was written by Michael Andrei Pelea, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

No funding was received for conducting this study.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose.

Ethical approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the local Ethics Committee (nr.100/03.05.2022).

Consent to participate

Written informed consent was obtained from all individual participants included in the study.

Consent to publish

The authors affirm that human research participants provided written informed consent for publication of the images in Figure(s) 1a, 1b, 1c and 1d.