Shear Wave Elastography (SWE) for Monitoring of Treatment of Tendinopathies

Shear Wave Elastography (SWE) for Monitoring of Treatment of Tendinopathies

ARTICLE IN PRESS Original Investigation Shear Wave Elastography (SWE) for Monitoring of Treatment of Tendinopathies: A Double-blinded, Longitudinal ...

521KB Sizes 0 Downloads 36 Views

ARTICLE IN PRESS

Original Investigation

Shear Wave Elastography (SWE) for Monitoring of Treatment of Tendinopathies: A Double-blinded, Longitudinal Clinical Study Timm Dirrichs, MD1, Valentin Quack, MD1, Matthias Gatz, MD, Markus Tingart, MD, Björn Rath, MD, Marcel Betsch, MD, Christiane K. Kuhl, MD, Simone Schrading, MD Rationale and Objectives: We aimed to investigate the diagnostic accuracy with which shear wave elastography (SWE) can be used to monitor response to treatment of tendinopathies, and to compare it to conventional ultrasound (US)-imaging methods (B-mode US (B-US) and power Doppler US (PD-US)). Materials and Methods: A prospective Institutional Review Board-approved longitudinal study on 35 patients with 47 symptomatic tendons (17 Achilles-, 15 patellar-, and 15 humeral-epicondylar) who underwent standardized multimodal US and standardized clinical assessment before and after 6 months of treatment (tailored stretching exercise, sport break, and local Polidocanol) was carried out. All US studies were performed by radiologists blinded to the clinical symptoms on both tendon sides to avoid biased interpretations, by B-US, PD-US, and SWE, conducted in the same order, using a high-resolution linear 15 MHz probe (Aixplorer). Orthopedic surgeons who were in turn blinded to US imaging results used established orthopedic scores (Victorian Institute of Sports Assessment questionnaire for Achilles, Victorian Institute of Sports Assessment questionnaire for patellar tendons, and Disability Arm Shoulder Hand scoring system) to rate presence, degree, and possible resolution of symptoms. We analyzed the diagnostic accuracy with which the different US imaging methods were able to detect symptomatic tendons at baseline as well as treatment effects, with orthopedic scores serving as reference standard. Results: B-US, PD-US, and SWE detected symptomatic tendons with a sensitivity of 66% (31 of 47), 72% (34 of 47), and 87.5% (41 of 47), respectively. Positive predictive value was 0.67 for B-US, 0.87 for PD-US, and 1 for SWE. After treatment, clinical scores improved in 68% (32 of 47) of tendons. Treatment effects were observable by B-US, PD-US, and SWE with a sensitivity of 3.1% (1 of 32), 28.1% (9 of 32), and 81.3% (26 of 32), respectively. B-US was false-positive in 68.8% (20 of 32), PD-US in 46.9% (15 of 32), and SWE in 12.5% (4 of 32) (SWE). Clinical scores and B-US, PD-US, and SWE findings correlated poorly (r = 0.24), moderately (r = 0.59), and strongly (r = 0.80). Conclusion: Unlike B-US or PD-US, SWE is able to depict processes associated with tendon healing and may be a useful tool to monitor treatment effects. Key Words: Shear wave elastography; sonography; tendon; tendinopathy. © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

INTRODUCTION Acad Radiol 2017; ■:■■–■■ From the Department of Diagnostic and Interventional Radiology, RWTH Aachen University Hospital, Pauwelsstr. 30 (T.D., C.K.K., S.S.); Department of Orthopedics, RWTH Aachen University Hospital, Aachen, Germany (V.Q., M.G., M.T., B.R., M.B.). Received April 27, 2017; revised August 16, 2017; accepted September 8, 2017. 1Dr. Dirrichs and Dr. Quack contributed equally to the manuscript. Address correspondence to: T.D. e-mail: [email protected] © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.acra.2017.09.011

T

endinopathies of Achilles, patellar, or epicondylar tendons, characterized by pain, swelling, or function, are some of the most common orthopedic conditions, not only in sportsmen or athletes (1) (2), but also in individuals with sedentary lifestyle (3). To date, B-mode ultrasound (B-US), power Doppler ultrasound (PD-US), and magnetic resonance imaging (MRI) are the main diagnostic tools in detecting and monitoring 1

DIRRICHS ET AL

tendinopathies (4,5). Typical US detectable changes, such as tendon thickening, inhomogeneous tendon structure, loss of echogenicity, and hypervascularization are known to be imaging signs associated with or indicative of tendinopathy (6–9). However, sensitivity, specificity, and inter-reader agreement of B-US or PD-US assessment are limited. This is in part because symptomatic tendons may not exhibit USdetectable morphologic changes and may not always be associated with PD-US detectable hypervascularization (10) due to false-positive findings that can be caused by subtle US abnormalities that can be found in asymptomatic individuals, especially among athletes (11). Accordingly, the role of B-US and PD-US for detection of tendon pathology remains controversial (12–16). Recent studies suggest that shear wave elastography (SWE) provides semiquantitative (color-map) and quantitative (absolute SWE values) imaging biomarkers that are useful to assess tendon integrity (8). By assessing tendon elasticity, SWE can help distinguish between asymptomatic and symptomatic tendons (17), with diseased tendons being significantly softer than healthy, asymptomatic tendons (18,19). Tendon softening, as visualized by SWE, correlates closely with the degree of clinical symptoms as assessed by clinical scoring systems such as the Victorian Institute of Sports Assessment (VISA) or Disability Arm Shoulder Hand (DASH) (18), that is, with pain and with functional deficits (20). The aim of this study was to investigate, through a longitudinal, double-blinded, randomized intraindividual comparative study, the diagnostic accuracy of SWE, compared to conventional US (B-mode and Doppler), for the assessment of treatment effects in patients undergoing treatment for tendinopathies. Because depiction of treatment effects requires the demonstration of abnormal imaging findings at baseline, that is, prior to treatment, the secondary aim was to investigate the diagnostic accuracy with which the three different US imaging methods are able to demonstrate presence of tendinopathy. MATERIALS AND METHODS Study Design and Population

A prospective, longitudinal, intraindividual clinical study including 35 symptomatic participants (mean age 43 ± 10.4 years, 20 male, 15 female) with unilateral or bilateral tendinopathies of the Achilles, patellar, or humeral radial/ulnar epicondylar tendon was carried out between December 2013 and November 2014. Every participant underwent a standardized clinical and multimodal US protocol twice: at baseline, that is, prior to therapy, and after 6 months of treatment (mean: 183 ± 7 days).

Academic Radiology, Vol ■, No ■■, ■■ 2017

tendon(s), regardless of whether the symptoms were unilateral or bilateral. Pain, diffused or localized swelling, reduced force, and reduced flexibility of the tendon were defined to be signs of tendinopathy. Participants with history of tendon rupture and tendon surgery were excluded from participation. Clinical Assessment

At the first clinical visit, every participant completed a standardized questionnaire concerning general medical history, sports activity, and current tendon pain (rest, movement, or pressure pain). At each visit, presence and degree of tendon pain was established by standardized questionnaires and standardized clinical examination by an orthopedic surgeon. Results of the questionnaire as well as results of the clinical examinations were documented in a dedicated database. The degree of Achilles and patellar tendon pain was established using the standardized VISA scoring system: VISA-A questionnaire for Achilles and VISA-P questionnaire for patellar tendons (21). The VISA-A and VISA-P scoring systems range from 0 to 100: a score of 100 means “no pain or impairment of physical activity”; a score of 0 means “maximum pain or impairment of physical activity.” The degree of epicondylar tendon pain was obtained using the standardized DASH scoring system (22). The DASH scoring system ranges from 100 to 0: a score of 100 means “highest pain”; a score of 0 means “no pain.” DASH score scale was later on mathematically inverted (100−DASH) for data analysis, to make it comparable to VISA-A and VISA-P scores. Tendinopathy Treatment

Every patient underwent a standardized tendinopathy treatment protocol according to current orthopedic guidelines (23,24) consisting of concentric stretching exercises, sports break, and topical application of polidocanol ointment (Polidocanol Thesit Gel, 3%, gepepharm). At clinical visits, self-reported compliance rates were 100% for stretching exercises and ointment application, respectively. At the first visit, participants were instructed by an orthopedic surgeon on how to perform static stretching exercises. Participants were asked to perform these exercises at least once a day (and up to three times per day). Participants with extensive sporting activity (>3 times per week) were asked to interrupt this activity throughout the study duration (6 months). Polidocanol has been proven to be a useful sclerosing agent when injected in affected tendon areas. Patients were encouraged to locally apply Thesit Gel three times a day in the first 2 weeks and whenever required in the remaining period of time. Patients were guided to apply the ointment at least 15 minutes before dressing the treated area.

Inclusion Criteria

Imaging Protocol

We included patients with chronic (>6 months duration) pain of the Achilles, patellar, or humeral epicondylar (radialis/ulnaris)

Every participant underwent bilateral multimodal US on a dedicated sonography unit (Aixplorer, Supersonic Imagine, Aix-

2

Academic Radiology, Vol ■, No ■■, ■■ 2017 SHEAR WAVE ELASTOGRAPHY AIDS TREATMENT MONITORING IN TENDINOPATHY

en-Provence, France). A high-resolution linear 15 MHz transducer (SuperLinear SL 15-4, Supersonic Imagine) with 256 elements and a bandwidth from 4 to 15 MHz was used in all participants. Sonographic examinations were performed according to a standardized protocol as follows: first, high-resolution B-US, then PD-US, and, finally, SWE. The same physical parameters (eg, wall filter, alias threshold) were used in every participant. To improve docking between transducer and skin, a gel standoff distance (Sonogel, Sonokit Proxon, length 100 × 100 mm, delay distance 20 mm) was used. Examinations were performed by one of three radiologists with 5, 5, and 6 years of experience in sonographic imaging and applying SWE. The performing radiologist was blinded to the patients’ clinical history and symptoms. Both sides were examined, and examinations were always conducted in the same order, beginning with the right, followed by the left side. Participants were positioned according to a standardized protocol, as described previously (16), with tendons in relaxed state each time. Tendons were systematically examined in longitudinal and cross-sectional plane using B-US, PD-US, and SWE in randomized order. Three representative B-US, PD-US, and SWE images of the distal, mid, and proximal tendon part were saved. B-Mode Ultrasound (B-US) Inhomogeneous tendon texture, hypoechogenic tendon swelling or thickening (compared to surrounding tendon areas), partial tear, tendon calcification, or fluid in the paratendon (eg, the retrocalcaneal bursa) was taken as test positive findings (25). Tendons without any of the mentioned structural abnormalities were considered “negative.” Power Doppler Ultrasound (PD-US) PD-US was done by systematically scanning the entire tendon to detect hypervascularized areas. Any neovascularization, regardless of its strength, was considered as a test-positive finding (1,7). Tendons without recognizable neovascularization were considered “negative.” Shear Wave Elastography (SWE) SWE was done by standardized measurements in each plane. Standardized size of SWE window was 1 cm2. First of all, a semiquantitative evaluation of tendon stiffness was done by visually grading the SWE color-maps (blue = soft tissue; turquoise or yellow = intermediate tissue stiffness; red = high tissue rigidity). Quantitative measurements then were obtained by placing a region of interest with a diameter of 1 mm in the most rigid area in each of the three longitudinal images. Quantitative SWE data were measured in kilopascal (kPa/Young modulus) up to a maximum of 300 kPa and in m/s (shear wave speed) up to a maximum of 10 m/s. Mean, maximum, and standard deviation of all values in kPa and m/s were evaluated. Based on cutoff values established in previous patient cohorts (8), SWE values of below 70 kPa (4.83 m/s) were considered “test positive”; values above this level were considered “test negative.”

Data Analysis

To evaluate the diagnostic accuracy of B-US, PD-US, and SWE in diagnosing tendinopathy at baseline, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the three methods were analyzed and compared. The results of the clinical scores were used as the standard of reference to identify healthy or diseased tendons. Participants with score values of 80 or greater were considered to be asymptomatic. To evaluate the diagnostic accuracy with which US imaging methods were able to depict treatment effects, changes in B-US, PD-US, and SWE findings were investigated. Clinical scores again served as standard of reference, with an improvement of clinical scores of more than 20 points considered to be a relevant change, that is, positive for treatment effects. The association between tissue stiffness (based on kPa (m/s) values) and clinical symptoms (based on clinical scores) was evaluated at baseline and after 6 months of therapy by correlating quantitative SWE values with quantitative VISA and DASH scores.

Statistical Analysis

A positive imaging study (B-US, PD-US, or SWE) of a tendon with clinical symptoms as evidenced by a pathologic clinical score was taken as a “true-positive” result. A negative imaging study in the same setting was considered a “false-negative” result. A positive imaging study in patients with asymptomatic tendon was considered a “false-positive” result; a negative imaging study in that same setting was considered a “truenegative” result. An improvement of clinical symptoms that occurred after treatment was considered a treatment success. If imaging studies exhibited a change (a resolution) of preexisting findings in such patients, this was considered a “positive” correlation with regard to detection of treatment success. Lack of normalization of imaging findings in such patients was considered a “negative” correlation. Patients lacking clinically detectable improvement of their symptoms (ie, whose clinical scores remained pathologic upon follow-up) were defined as treatment failures. If imaging studies exhibited a change (a resolution) of preexisting findings in such patients, this was considered a “negative” correlation with regard to detection of treatment success. Lack of a resolution of US imaging findings in a patient with treatment failure was considered a “positive” correlation. Imaging findings regarding treatment effects were considered false positive in case a resolution of findings was seen, but the patient did not improve clinically. Continuous variables were presented as mean ± standard deviation or as median with interquartile range; categorical data are presented as numbers and percentages. The correlation between clinical VISA-A, VISA-P, and DASH scores and quantitative SWE values was evaluated using Spearman rank correlation coefficients. Using the method of 3

Academic Radiology, Vol ■, No ■■, ■■ 2017

DIRRICHS ET AL

Ultrasound Findings B-US revealed morphologic changes in 31 of 47 (66%) symptomatic tendons, PD-US revealed pathologic neovascularization in 34 of 47 (72.3%), and SWE revealed reduced tissue stiffness (<70 kPa) in 41 of 47 tendons (87%). The mean SWE value was 45 kPa (3.87 m/s) (range 25–135 kPa/0.96– 6.71 m/s) for symptomatic and 147 kPa (7 m/s) (range 63– 210 kPa/4.58–8.37 m/s) for asymptomatic tendons. This difference was statistically significant (P < .001) and was observed for all tendon groups. False-positive findings were made as follows: B-US detected morphologic changes in 17.4% (4 of 23) of asymptomatic tendons, PD-US detected neovascularization in 21.7% (5 of 23) of clinically asymptomatic tendons, and SWE exhibited below-normal tendon stiffness in none (0 of 23) of asymptomatic tendons. This yielded a specificity of 82.6% for US, 78.8% for PD, and 100% for SWE. Thus, NPV was 0.58 for US, 0.65 for PD, and 0.84 for SWE. PPV was 0.67 for US, 0.87 for PD, and 1 for SWE.

Clopper and Pearson (26), 95% confidence intervals were calculated. All reported P values are two-sided and P values <0.05 were considered to be statistically significant. Statistical analysis was performed using SPSS Statistics for Windows, Version 21.0 (IBM Corp., Armonk, NY).

RESULTS A total of 35 patients (mean age 46 ± 18 years) were included; US was performed on both sides, yielding 70 tendons for analysis. Of these, 47 were symptomatic at baseline, whereas 23 were asymptomatic. The anatomic distribution of symptomatic tendons at baseline was as follows: 17 Achilles tendons, 15 patellar tendons, and 15 humeral epicondylar tendons. Details of patients’ demographics and clinical symptom scores are shown in Table 1. The course of clinical scores and imaging findings under therapy are shown in Table 2.

Baseline (Pretreatment) Findings After 6 Months of Treatment

Clinical Scores Symptomatic tendons exhibited a mean clinical score of 40 (range 21–62); asymptomatic tendons a mean score of 98 (range 78–100).

Clinical Scores After therapy, 68.1% of tendons (32 of 47) were completely asymptomatic or showed a relevant improvement of clinical

TABLE 1. Demographics of Symptomatic Participants at Baseline (n = 47)

Age (years) Mean ± SD Median/range Sex Male Female Clinical score Baseline (Mean/SD)

All (n = 47)

Achilles Tendon (n = 17)

Patellar Tendon (n = 15)

Humeral Epicondylar Tendon (n = 15)

46 ± 18 45/20–71

50 ± 13 49/31–69

26 ± 5 30/20–40

49 ± 15 43/24–70

37 (78.7%) 10 (23.4%)

11 (64.7%) 6 (35.3%) VISA-A score 54 ± 16

15 (100%) 0 (0%) VISA-P score 38 ± 15

11 (73.3%) 4 (26.7%) DASH score* 48 ± 12

40 ± 20

DASH, Disability Arm Shoulder Hand scoring system; SD, standard deviation; VISA-A, Victorian Institute of Sports Assessment questionnaire for Achilles; VISA-P, Victorian Institute of Sports Assessment questionnaire for patellar tendons. * Initial DASH score (0–100) was mathematically inverted (100–DASH) to make it comparable with the VISA score.

TABLE 2. Imaging Findings in Symptomatic Tendons at Baseline and After 6 Months of Therapy (n = 47)

Clinical score SWE score (kPa) Symptomatic tendons Positive B-mode US Positive PD-US Positive SWE

Baseline

6 Months Following Treatment

Measureable Improvement

P Value

40 ± 20 42 ± 15 47/47 (100%) 31/47 (66%) 34/47 (72.3%) 41/47 (87%)

98 ± 17 106 ± 27 15/47 (31.9%) 29/47 (61.7%) 25/47 (53.2%) 6/47 (12.8%)

+245% +152.4% +68.1% +4.3% +26.5% +85.4%

<.01 <.01 <.01 >.05 >.05 <.01

B-mode US, B-mode ultrasound; PD-US, power Doppler ultrasound; SWE, shear wave elastography.

4

Academic Radiology, Vol ■, No ■■, ■■ 2017 SHEAR WAVE ELASTOGRAPHY AIDS TREATMENT MONITORING IN TENDINOPATHY

kPa-value (up to 120)

120

SWE-Value

Clinical Score

VISA-score-value (up to 100)

106.00 98.00

100

80

60

40

42.00

40.00

20

0

Baseline

Baseline

After 6 months

After 6 months

Figure 1. Tendon recovery: course of clinical scores and SWE values under therapy. Graph shows the development of clinical scores and SWE values from baseline to the end of surveillance after 6 months of tendinopathy treatment. Synchronous to the increase of clinical scores (which is equal to symptom decrease), SWE values rose significantly, which demonstrates the close correlation between SWE findings and clinical symptoms.SWE, shear wave elastography; VISA, Victorian Institute of Sports Assessment.

TABLE 3. Sensitivity/Specificity of Different Modalities Prior to and 6 Months After Initiating Tendinopathy Treatment Baseline Modality B-mode Power Doppler SWE

6 Months Following Treatment

SENS

SPEC

PPV

NPV

SENS

SPEC

PPV

NPV

0.66 0.72 0.87

0.83 0.79 1

0.67 0.87 1

0.58 0.65 0.83

0.05 0.47 0.81

0.31 0.53 0.88

0.3 0.32 0.75

0.13 0.32 0.57

NPV, negative predictive value; PPV, positive predictive value; SENS, sensitivity; SPEC, specificity; SWE, shear wave elastography.

scores as defined above. In 29.8% (14 of 47) of tendons, no change was observed. A deterioration of clinical symptoms was apparent in 2.1% of tendons (1 of 47). B-Mode Ultrasound A resolution of structural B-US findings was observed in 3.1% of tendons (1 of 32) that had improved clinically. This yielded sensitivity for treatment effects of 2.1% (1 of 47) for all symptomatic tendons, or 4.8% (1 of 21) for symptomatic tendons with positive baseline B-US. Power Doppler Ultrasound A resolution of neovascularization at PD-US was observed in 28.1% of tendons (9 of 32) that had improved clinically. This yielded sensitivity for treatment effects of 19.1% (9 of 47) for all symptomatic tendons, and 47.4% (9 of 19) for symptomatic tendons with positive baseline PD-US. Shear Wave Elastography A back-to-normal tendon stiffness was observed by SWE in 81.3% of tendons (26 of 32) that had improved clinically. This yielded sensitivity for treatment effects of 55.3% (26 of 47) for all symptomatic tendons, and 81.3% (26 of 32) for symptomatic tendons with positive baseline SWE. In those who improved clinically, mean SWE values rose from 42 kPa (3.74 m/s) to 106 kPa (5.94 m/s) (Fig 1).

In patients who improved after treatment, B-US failed to depict this improvement, that is, its findings were still abnormal, in 62.5% of asymptomatic or clinically improved tendons (20 of 32). PD-US failed to depict this improvement in 46.9% (15 of 32), and SWE in 12.5% (4 of 32) of tendons that became asymptomatic or clinically improved after treatment. This yielded a specificity of 31.3% for US, 53.1% for PD, and 87.5% for SWE. Thus, NPV was 0.13 for US, 0.32 for PD, and 0.57 for SWE. PPV was 0.3 for US, 0.32 for PD, and 0.75 for SWE. Values of sensitivity, specificity, NPV, and PPV at baseline and under therapy are shown in Table 3.

Quantitative Tendon Assessment by SWE

At baseline, symptomatic tendons exhibited a mean SWE value of 3.74 m/s with a median of 3.6 m/s, range 1.73 m/s– 7.7 m/s. After 6 months of therapy, mean SWE values rose to 5.9 m/s, with a median of 5.5 m/s, range 3.3 m/s–9 m/s (Fig 1; Table 2). Images of typical tendon findings are shown in Figure 2. Among the 32 patients with an overall improvement of clinical symptoms, mean SWE values increased by 4.6 m/s, range 3.7 m/s–5.9 m/s after 6 months. 5

Academic Radiology, Vol ■, No ■■, ■■ 2017

DIRRICHS ET AL

(a)

(c)

(b)

(d)

Figure 2. Sonographic images of a symptomatic Achilles tendon and its changes under therapy. Tendon recovery: Panels (a) to (d) show sonographic images of the right symptomatic Achilles tendon of a 35-year-old male. The patient had relevant clinical impairment, with a VISA-A score of 48. B-mode and power Doppler (b) revealed no visible morphologic or perfusion changes, even though the patient was symptomatic. Shear wave elastography in contrast showed a blue color chart and a significantly reduced tendon stiffness of 48 kPa (a). After 6 months of therapy, SWE values increased significantly to 151 kPa (c); the patient then was nearly asymptomatic with a VISA-Ascore of 92 (d). No relevant changes were detectable in B-mode and Power-Doppler. SWE, shear wave elastography; VISA-A, Victorian Institute of Sports Assessment questionnaire for Achilles. (Color version of figure is available online.)

Clinical scores and changes at B-US correlated poorly (r = 0.24), correlated moderately with PD-US findings (r = 0.59), and correlated strongly with SWE findings (r = 0.80). DISCUSSION In this study on 35 patients with 47 symptomatic tendons, we found that SWE is not only able to identify symptomatic tendons with high sensitivity (87.5%) and specificity (100%), but also is able to help depict response to treatment with high accuracy that is significantly higher than that offered by B-US or PD-US. Tendinopathy is one of the most common diseases among athletes or recreational exercisers, but it occurs in inactive people, too. The etiology of tendinopathy is still not fully understood; it is likely multifactorial, but tendon overuse or pathologic load combined with reduced resting periods (10,27) seems to be the most important factor. For tendinopathies of the Achilles tendon, compression of the tendon due to calcaneal impingement (28) might be another contributing factor. Independent of the individual cause, the final common pathway of tendinopathies appears to be recurrent microdamage (3). Microdamage leads to decreased collagen fiber organization, increased numbers of endothelial cells and fibroblasts, accompanied by inflammatory changes (23), and is clinically associated with tendon pain, tendon weakening, and finally, tendon rupture. Neovascularization seems to play an important role during this degenerative process (3,7,27). If tendon degeneration is diagnosed early, it has been shown that protective measures can help avoid progression and complications such as tendon rupture (29). Therefore, early diagnosis 6

of tendinopathy, that is, diagnosis long before chronic damage or rupture occurs, seems of particular interest. Treatment of tendinopathies still remains difficult: Most available conservative and surgical treatments have no scientific evidence. Only a few treatments have been investigated in randomized controlled trials and have shown to have an effect, for example, eccentric exercises, extracorporeal shock wave treatment, sclerosing injections (Polidocanol), or nonsteroidal anti-inflammatory drugs (30). Besides limited treatment options, there are no established treatment monitoring instruments, which aggravates detection of early treatment effects, possible complications, or adverse drug reactions. US imaging (B-mode and Power Doppler) still remains the cornerstone of the workup of patients with suspected tendinopathies (12–14) or for monitoring treatment. This is in spite of the ongoing debate regarding its actual clinical usefulness: It is well established that US offers only a limited sensitivity; that is, it is associated with a high rate of falsenegative findings in symptomatic individuals (10). Moreover, US offers only a modest specificity such that it is associated with a high number of false-positive findings, especially in athletes (11). In professional athletes (31), that is, a cohort where early diagnosis of tendon pathology would be of particular interest, the likelihood of false-positive (abnormal) US findings is even higher than in sedentary individuals. All these aspects have led to the fact that the actual role of tendon US in the workup of patients with tendinopathies is controversial. Recent reviews (31,32) cast doubt on the correlation between visually ascertainable morphologic changes in US and clinical symptoms altogether, recommending to “take a step back”

Academic Radiology, Vol ■, No ■■, ■■ 2017 SHEAR WAVE ELASTOGRAPHY AIDS TREATMENT MONITORING IN TENDINOPATHY

and to critically relate US findings with clinical symptoms. Another issue that limits the clinical usefulness of B-US as well as PD-US, especially regarding demonstrating treatment effects, is the fact that it provides qualitative or semiquantitative information only. Over the last 5 years, several studies have revealed that SWE is a reliable tool to differentiate between diseased and healthy tendons, as diseased tendons are significantly softer than healthy ones (12,17–19,33). However, SWE not only helps in differentiating between “healthy” and “diseased” tendons, it also adds relevant diagnostic information suitable to quantitatively rate the degree of tendon impairment (16). This quantitative information, additional to that of conventional B-US and PD-US, and its close correlation to clinical symptoms, makes SWE a reliable tool to visualize clinically relevant tendon damage. Our study provides further evidence to the fact that SWE is superior to conventional B-US and PD-US in diagnosing tendinopathy. A sensitivity of 66% (B-US) and 72% (PDUS), compared to 87% of that of SWE, and especially the high specificity (100% at baseline, 87% after treatment), emphasizes the ability of SWE to diagnose and monitor tendinopathy and its treatment. A combination of the three different US modalities (B-US, PD-US, and SWE) in a onestop-shop-examination might enhance the detection rate of tendon pathology and visualization of tendon healing. Besides the clinically important fact that the implemented specific orthopedic treatment led to symptom decrease (as revealed by clinical scores), SWE was able to detect associated tendon changes that were mostly “invisible” to B-US and PDUS. Moreover, beyond the mere categorization of a tendon as “diseased” or “healthy,” SWE delivers quantitative information about the degree of tendon impairment. This makes SWE a reliable tool to monitor treatment effects. The main limitation of our study is the fact that we included tendons of different locations (Achilles, patella, and epicondylus). However, the fact that SWE is able to depict tendinopathy and tendon healing independently from anatomic regions in Achilles, patellar, and humeral epicondylar tendons emphasizes its validity. We did not establish interreader agreement; however, as we involved as many as three different radiologists with variable levels of expertise, our results will be fairly representative for clinical practice. Last, we did not investigate the utility of tendon MRI for demonstration of healing effects; another study is currently underway in our department to compare the accuracy of SWE with that of MRI.

CONCLUSION In this prospective longitudinal study of patients suffering from Achilles, patellar, or humeral epicondylar tendinopathy, SWE proved to be the method of choice for providing imaging correlates for symptomatic tendons and for monitoring treatment. SWE will depict tendon healing better and earlier than B-US and PD-US. SWE provides an objective, quantitative assess-

ment of tendon integrity and might be useful to guide treatment or develop new treatment approaches. REFERENCES 1. Maffulli N, Sharma P, Luscombe KL. Achilles tendinopathy: aetiology and management. J R Soc Med 2004; 97:472–476. 2. Cassel M, Baur H, Hirschmüller A, et al. Prevalence of Achilles and patellar tendinopathy and their association to intratendinous changes in adolescent athletes. Scand J Med Sci Sports 2015; 25:310–318. 3. Ames PR, Longo UG, Denaro V, et al. Achilles tendon problems: not just an orthopaedic issue. Disabil Rehabil 2008; 30:1646–1650. 4. Mahoney JM. Imaging techniques and indications. Clin Podiatr Med Surg 2017; 34:115–128. 5. Gaulke R, Krettek C. Tendinopathies of the foot and ankle: evidence for the origin, diagnostics and therapy. Unfallchirurg 2017; 120:205–213. 6. de Jonge S, van den Berg C, de Vos RJ, et al. Incidence of midportion Achilles tendinopathy in the general population. Br J Sports Med 2011; 45:1026–1028. 7. Ohberg L, Alfredson H. Ultrasound guided sclerosis of neovessels in painful chronic Achilles tendinosis: pilot study of a new treatment. Br J Sports Med 2002; 36:173–175. 8. Pingel J, Lu Y, Starborg T, et al. 3-D ultrastructure and collagen composition of healthy and overloaded human tendon: evidence of tenocyte and matrix buckling. J Anat 2014; 224:548–555. 9. Notarnicola A, Maccagnano G, Di Leo M, et al. Overload and neovascularization of Achilles tendons in young artistic and rhythmic gymnasts compared with controls: an observational study. Musculoskelet Surg 2014; 98:115–120. 10. Visnes H, Tegnander A, Bahr R. Ultrasound characteristics of the patellar and quadriceps tendons among young elite athletes. Scand J Med Sci Sports 2015; 25:205–215. 11. Comin J, Cook JL, Malliaras P, et al. The prevalence and clinical significance of sonographic tendon abnormalities in asymptomatic ballet dancers: a 24-month longitudinal study. Br J Sports Med 2013; 47:89– 92. 12. Klauser AS, Miyamoto H, Tamegger M, et al. Achilles tendon assessed with sonoelastography: histologic agreement. Radiology 2013; 267:837– 842. 13. Lehtinen A, Peltokallio P, Taavitsainen M. Sonography of Achilles tendon correlated to operative findings. Ann Chir Gynaecol 1994; 83:322–327. 14. Khan KM, Forster BB, Robinson J, et al. Are ultrasound and magnetic resonance imaging of value in assessment of Achilles tendon disorders? A two year prospective study. Br J Sports Med 2003; 37:149– 153. 15. Paavola M, Paakkala T, Kannus P, et al. Ultrasonography in the differential diagnosis of Achilles tendon injuries and related disorders: a comparison between pre-operative ultrasonography and surgical findings. Acta Radiol 1998; 39:612–619. 16. Kayser R, Mahlfeld K, Heyde CE. Partial rupture of the proximal Achilles tendon: a differential diagnostic problem in ultrasound imaging. Br J Sports Med 2005; 39:838–842. 17. De Zordo T, Fink C, Feuchtner GM, et al. Real-time sonoelastography findings in healthy Achilles tendons. AJR Am J Roentgenol 2009; 193:134– 138. 18. Dirrichs T, Quack V, Gatz M, et al. Shear wave elastography (SWE) for the evaluation of patients with tendinopathies. Acad Radiol 2016; 23:1204– 1213. 19. Aubry S, Nueffer JP, Tanter M, et al. Viscoelasticity in Achilles tendinopathy: quantitative assessment by using real-time shear-wave elastography. Radiology 2015; 274:821–829. 20. Ooi CC, Richards PJ, Maffulli N, et al. A soft patellar tendon on ultrasound elastography is associated with pain and functional deficit in volleyball players. J Sci Med Sport 2016; 19:373–378. 21. Vestergård Iversen J, Bartels EM, Langberg H. The Victorian Institute of Sports Assessment—Achilles Questionnaire (VISA-A)—a reliable tool for measuring Achilles tendinopathy. Int J Sports Phys Ther 2012; 7:76– 84. 22. Germann G, Harth A, Wind G, et al. Standardisation and validation of the German version 2.0 of the Disability of Arm, Shoulder, Hand (DASH) questionnaire. Unfallchirurg 2003; 106:13–19. 23. Andres BM, Murrell GA. Treatment of tendinopathy: what works, what does not, and what is on the horizon. Clin Orthop Relat Res 2008; 466:1539–1554.

7

DIRRICHS ET AL

24. Mafi N, Lorentzon R, Alfredson H. Superior short-term results with eccentric calf muscle training compared to concentric training in a randomized prospective multicenter study on patients with chronic Achilles tendinosis. Knee Surg Sports Traumatol Arthrosc 2001; 9:42–47. 25. Dong Q, Fessell DP. Achilles tendon ultrasound technique. AJR Am J Roentgenol 2009; 193:W173. 26. Clopper C, Pearson ES. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 1934; 26:404–413. 27. Glazebrook MA, Wright JR, Jr, Langman M, et al. Histological analysis of Achilles tendons in an overuse rat model. J Orthop Res 2008; 26:840– 846. 28. Bah I, Kwak ST, Chimenti RL, et al. Mechanical changes in the Achilles tendon due to insertional Achilles tendinopathy. J Mech Behav Biomed Mater 2016; 53:320–328.

8

Academic Radiology, Vol ■, No ■■, ■■ 2017

29. Aubry R. Shear wave elastography in the assessment of tendon stiffness and for monitoring patients undergoing treatment. Diagn Imaging Eur 2015; 31:54–57. 30. Zwiers R, Wiegerinck JI, van Dijk CN. Treatment of midportion Achilles tendinopathy: an evidence-based overview. Knee Surg Sports Traumatol Arthrosc 2016; 24:2103–2111. 31. Ryan M, Bisset L, Newsham-West R. Should we care about tendon structure? The disconnect between structure and symptoms in tendinopathy. J Orthop Sports Phys Ther 2015; 45:823–825. 32. Docking SI, Ooi CC, Connell D. Tendinopathy: is imaging telling us the entire story? J Orthop Sports Phys Ther 2015; 45:842–852. 33. Chen XM, Cui LG, He P, et al. Shear wave elastographic characterization of normal and torn Achilles tendons: a pilot study. J Ultrasound Med 2013; 32:449–455.