A novel measurement to evaluate supraspinatus tendinopathies with strain elastography

Abstract

Background

In addition to the traditional strain ratio (STR), a novel measurement termed hard percent of the whole tendon, provided by the device’s software and resembling shear-wave elastography, was included in the study.

Purpose

To evaluate the correlation between magnetic resonance imaging (MRI) and strain elastography findings in cases with rotator cuff tendinopathy, including tears.

Material and Methods

As a prospective study, cases suggestive of rotator cuff tendinopathy were included. After MRI evaluation by two observers, grading of the elastography examination was performed by a third radiologist. For the first measurement, the region of interest (ROI) was chosen from the gray-scale evaluation corresponding to the area showing a MRI pathological signal. Stiffness was assessed with the STR. For the second measurement, the whole visible tendon was measured as the ROI. The hard percent (unaffected areas as a percent) was measured. Correlation coefficients between MRI grade and the two strain value measurements were calculated.

Results

There were 93 cases (31 men, 62 women) with 112 involved tendons. The correlation coefficient between MRI grade and age was calculated as 0.56, between MRI grade and STR it was –0.51, and with the novel measurement the correlation coefficient was –0.53.

Conclusion

This study showed a good association of findings of tendon stiffness with those of tendinopathy on MRI. Using the hard ratio as a novel measurement, its correlation with MRI grade was as reliable as the STR. We also experienced that the benefit of elastography is a challenging issue for defining small ruptures.

Keywords

Rotator cuff tendinopathy strain elastography magnetic resonance imaging

Introduction

Rotator cuff tendinopathies are common disorders and an important cause of morbidity, including pain, restriction of mobility, and even disabilities. The tendon with the most common involvement is the supraspinatus with a rate of around 95% (1). As a cross-sectional modality, magnetic resonance imaging (MRI) has excellent soft-tissue resolution with advantages including multiplanar assessment and lack of radiation, making it the most effective and most preferred radiological method (2). In recent years, ultrasonography (USG) and elastography have also begun to be preferred modalities as the first-line method or as a complementary method (3). Elastography is mainly based on two different techniques evaluating the elasticity of the underlying tissue by measuring the degree of deformation. Strain (or compression) elastography appears to be a useful diagnostic tool, particularly for tendons (4). Compression is applied either manually or by physiologic movements such as breathing or pulsation. Comparing the strain of the target tissue with the strain of the surrounding tissue allows a semi-quantitative measurement of tissue elasticity. Because of the generation of mechanical stress by hand, this technique is highly operator-dependent. Shear-wave USG elastography may allow more objective and reproducible quantification of tissue elasticity in contrast to strain elastography, and is potentially less observer-dependent (5 –8). The aim of the present study was to evaluate the correlation between MRI and strain elastography findings in cases with rotator cuff tendinopathy, including tears.

Material and Methods

This was a prospective cross-sectional study performed over a period of nine months on consecutive patients who had a history of discomfort during activity with present or absent motion limitation (depending on the severity) and tenderness over the area of greater tuberosity, suggesting rotator cuff tendinopathy. The exclusion criteria were the presence of blunt trauma/open wounds or previous surgery. Patients had already been evaluated with MRI before the final decision on the report. For the evaluation of correlations with MRI grade, only supraspinatus tendons were included.

Informed consent was not required as all images used in the present study are anonymous. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All procedures were performed in compliance with relevant laws and institutional guidelines, and the appropriate institutional committee approved them (Okmeydanı Institute Ethical Review Board, 30 April 2019, no. 1248).

MRI examinations

MRI examinations were performed based on a standardized protocol for the evaluation of rotator cuff pathology. All patients were placed supine with the arm externally rotated, and a shoulder coil was positioned over the shoulder. Coronal proton density (PD) fat-saturated (FS), coronal T1, oblique sagittal T2 FS, and PD FS axial images were obtained using a 1.5-T superconducting magnet with high-speed gradients (Signa Excite, GE Medical Systems, Waukesha, WI, USA). A dedicated knee coil was used.

MRI grades were determined by two observers, each blinded to the results of the other.

Rotator cuff status was categorized according to the MRI tendinopathy grade as follows: grade 0 = normal; grade I = increased T2 signal with normal morphology; grade II = increased T2 signal with abnormal morphology (thickening or irregularity of the tendon); and grade III = defined tear (partial or full thickness) (9).

Real-time sono-elastography

The Sono-elx device was the Esaote Mylab Seven eHD crystaline (Via Enrico Melen, Genova, Italy), using the transducer AL2442 linear array (3–11 mHz).

Shoulder ultrasonography was performed, using a standardized protocol, by the third radiologist with four years of experience in musculoskeletal radiology. The supraspinatus and subscapularis tendons were examined with the patient sitting and the arm internally rotated and hyperextended. The transducer was placed anterior to the acromioclavicular joint and was oriented at an angle of 45° anteriorly to examine the supraspinatus tendon.

Elastography images were obtained by repeatedly compressing the shoulder with the probe. Compression–decompression cycles were performed with force and frequency of probe compression appropriately monitored by the device.

Two measurements were obtained in a single examination for those having grade 0 to 3 tendinopathy. They were the strain ratio (STR) and the hard percent area of the tendon. For the first measurement, stiffness was assessed with the STR. Region of interest (ROI) “A” was kept on the abnormal focal region on the tendon and ROI “B” was kept on the normal appearing portion of the tendon (reference area). In the evaluation of grade 0 to 2 tendinopathy, the supraspinatus tendon was completely scanned and measurements were made from the most pathological region (where the soft areas are the most dense) in the elastography examination. In patients with partial rupture, the strain rate was evaluated by placing an ROI in the partial rupture area and the spared (normal) area. After obtaining five consecutive measurements, the STR was calculated to be the arithmetic mean of those measurements. Supraspinatus tendons with complete tear were excluded from coefficient evaluation since there was no strain data obtainable.

For the second measurement, the whole visible tendon was measured as the ROI. The hard percent (unaffected areas as percent) was measured and noted. The second measurement (hard ratio) was always made from the area where the STR was taken.

Rotator cuff tendons other than supraspinatus measurements were not obtained, simply because involvement is rarely seen and requires thousands of cases for a statistical evaluation. The limited value of elastography was also taken into account for deeply seated tendons (> 3 cm). Those involvements were only diagnosed with MRI data and used in demographic calculations.

Statistical analysis

Correlation data analysis was used to calculate correlation coefficients between the two strain elastography values and MRI grade for each participant. Regression analysis was used to define the statistical significance of the difference between the two correlation coefficients. Microsoft Office 365 ProPlus Excel program was used for storage and calculations. P < 0.05 was considered significant.

Results

A total of 93 cases (31 men, 62 women; mean age = 52.9 years; age range = 24–90 years) with rotator cuff tendinopathy findings were included in the study. The total number of affected tendons was 112. The most commonly involved tendon of the rotator cuff was the supraspinatus tendon. Demographic data are summarized in Table 1.

Table 1.

Demographic data.

Cases (n = 93)		P
Sex
Female	62 (66.7)
Male	31 (33.3)
Mean age (years)	52.9
Location/side
Right	62 (55.3)
Left	50 (44.7)
Dominant hand	65 (58)	<0.01
Involved tendon (n = 112)
Supraspinatus	99 (88.4)
Subscapularis	12 (10.7)
Infraspinatus and teres minor	1 (0.9)

Values are given as n (%).

The mean STR and hard percent area of the tendons tended to decrease with increasing MRI grade. The elastography findings and the distribution of the tendons with MRI grade are shown in Table 2.

Table 2.

Association between MRI and elastography strain ratios for supraspinatus tendon.

MRI grade	Involved tendons (n = 99)	Strain ratio	Hard area (%)
0	21	0.8 ± 0.18	97.8 ± 2.3
1	31	0.54 ± 0.23	88.0 ± 9.34
2	22	0.33 ± 0.24	81.0 ± 9.88
3	18	0.15 ± 0.27	72.7 ± 13.4
4	7	NA	NA

Values are given as mean ± SD.

MRI, magnetic resonance imaging; SD, standard deviation.

There was a positive moderate correlation between age and MRI grade and a negative correlation between MRI grade and elastography findings. The difference between correlation coefficients of hard ratio–MRI grade and STR–MRI grade was not statistically significant with a “z score” of –0.2 and the corresponding probability was 0.84 (Table 3).

Table 3.

Correlation coefficients and Z score.

Parameters	Correlation coefficient (r)
1. MRI grade – age	0.557
2. MRI grade – strain ratio	–0.507
3. MRI grade – hard area	–0.528
Z score of coefficients 2 and 3	–0.2
Probability	0.84

MRI, magnetic resonance imaging.

Inter-observer concordance of the MRI results is shown in Table 4.

Table 4.

Inter-observer concordance of MRI.

MRI grading	Cases (n = 99)
The same MRI grade	80 (80.8)
1 grade difference	19 (19.2)
2 grades difference	0 (0)

Values are given as n (%).

MRI, magnetic resonance imaging.

Discussion

The most commonly involved component of the rotator cuff was the supraspinatus tendon. Since it is the most functional, the largest, and the most observable tendon, this result must not be surprising. The involved tendons were located in the dominant extremity with a ratio slightly >50%, which indicates that the tendons that tend to be affected should be from overuse compared to the counter-side. Tendinopathy occurs due to collagen breakdown, which softens and weakens the tendon, and may finally lead to tendon tear or rupture. Aging also affects the tendons and the result is likely due to an accumulation of trauma over time (10 –12).

The inter-observer concordance was calculated to be 80.8% and all discordant cases had a difference of one grade. This result can be interpreted as quite good but not flawless and reminds us of the potential weaknesses of MRI. Of those 19 cases, the two observers challenged 13 of them for the presence of partial rupture accompanied by tendinosis. Tudisco et al. (13) also reported that strain elastography and Hatta et al. (14) that shear-wave elastography were feasible methods applicable in clinical practice in the assessment of small supraspinatus tendon tears.

The calculated correlation coefficient between age and MRI grade is interpreted as a moderate positive correlation. Since the MRI is accepted as the reference imaging modality for involved tendons, the calculation was performed with MRI data. In the present study, we also experienced that partially ruptured tendons were also the tendons with tendinopathy in their background. The completely ruptured tendons should be the same, but it was not possible to evaluate elastography data since they were absent in the region of interest. Finally, there was a female dominance in the study cohort, but it is not easy to say that tendinopathy is seen more commonly among women for the simple reason that it is common for men to avoid contacting a physician in our country.

Elastography is a modality that is highly operator-dependent, even more than USG. We tried to reduce the impact of this disadvantage. The device was constantly monitored for level of pressure on the probe, with one eye always on the quality factor level (displayed on the left side of the image). Moreover, according to Hatta et al. (15), the STR of the rotator cuff tendon can be measured with minimal influence of overlying soft tissues if its depth from the skin is < 22 mm. We must remind the readers that our measurements were performed with tissues with a depth < 2 cm in the present study.

There were two methods applied in this study. The first method was measuring the well-known STR. As a novel method, measurements of hard area percent of the visible tendon were added for all cases where the function was available with our device. This method resembles the shear-wave technique and calculates the ratio of the healthy or hard portions of the tendon, since the tendons with tendinopathy and rupture are softer.

According to the results, there was a remarkable negative correlation between STR and MRI grade. The correlation coefficient obtained with the hard percent parameter was slightly higher, but the difference was not statistically significant. Therefore, the hard ratio measurement cannot be claimed to be a better method but can be considered as an additional reliable method. When the STR values were evaluated, high standard deviation values were determined.

Vasishta et al. (16) reported that strain elastography shows a moderate to strong association with MRI findings of tendinopathy. That study has the closest subject to ours. In a similar study, an association between tendinopathy and MRI grade was also confirmed, but the researchers used a different method. To calculate the STR, a 10-mm thick sonar aid gel pad and subcutaneous fatty tissue were used (17). Demirel et al. (18) concluded that the use of strain elastography in the diagnosis of supraspinatus muscle impingement syndrome may become widespread over time in clinical practice.

In recent years, a new modality called MR elastography was described that could be useful for the diagnosis and examination of the effect of treatment, such as a tear of the supraspinatus muscle (19).

The present study has some limitations. First, despite the precautions we took, strain elastography is still a user-dependent modality. Second, MRI findings may affect interpretation when evaluating elastography. It is possible to obtain different correlation coefficients if MRI data are not known during the evaluation of elastography. It should also be noted that we did not have the chance to evaluate the diagnostic accuracy of elastography and MRI, hence its full effect on the correlation coefficient. At this point, a gold standard such as arthroscopy is required. There were very few arthroscopy results in our case pool. MRI can be accepted as the reference modality for rotator cuff tendinopathy, but also has limitations. The images can be distorted with motion artefacts and small ruptures can be missed if the slice does not pass through them. Moreover, distinguishing tendinopathy from small ruptures can be challenging, as mentioned above. In the present study, we determined the presence of small partial tears that were not detected in MRI by elastography. This study also revealed inter-observer differences.

In conclusion, the present study showed a good association between the findings of tendon stiffness measured with elastography and those of tendinopathy on MRI. Strain elastography is a rapid, easily applicable technique without a known negative effect on living tissues. Considering the MRI limitations for the diagnosis of rotator cuff tendinopathy, this technique may be defined as a single reliable method or as an addition to MRI evaluation. Since most of the controversial cases were with partial rupture evaluation, we found elastography to have a complementary role for small ruptures that could be missed or misdiagnosed as tendinopathy with MRI.

Fig. 1.

A 60-year-old woman. Right shoulder. Elastography strain ratio = 0.46, elastography hard ratio = 99.3%, MRI grade = 0. Completely normal appearance without pathological signal.

Fig. 2.

A 39-year-old woman. Right shoulder. Elastography strain ratio = 0.54, elastography hard ratio = 97%, MRI grade = 1. Minor pathological signal of the distal tendon. Red arrows indicate tendinopathies, yellow arrows indicate collections.

Fig. 3.

A 49-year-old woman. Left shoulder. Elastography strain ratio = 0.37, elastography hard ratio = 87.3%, MRI grade = 2. Diffuse increased calibration and signal of the tendon. Red arrows indicate tendinopathies.

Fig. 4.

A 25-year-old man. Left shoulder. Elastography strain ratio = 0.06, elastography hard ratio = 86%, MRI grade = 3. Partial tear. Note that it mimics tendinopathy on MRI images. Red arrows indicate tendinopathies.

Fig. 5.

A 69-year-old woman with a complete tear on the right shoulder.

Fig. 6.

The data chart shows trendlines of correlation coefficients of hard ratio, strain ratio, and age with MRI grades.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Deniz Özel

References

Huegel

Williams

Soslowsky

LJ.

Rotator cuff biology and biomechanics: a review of normal and pathological conditions. Curr Rheumatol Rep 2015; 17:476.

Kvalvaag

Anvar

Karlberg

, et al. Shoulder MRI features with clinical correlations in subacromial pain syndrome: a cross-sectional and prognostic study. BMC Musculoskelet Disord 2017; 181:469.

Schmalzl

Fenwick

Boehm

, et al. The application of ultrasound elastography in the shoulder. J Shoulder Elbow Surg 2017; 26:2236–2246.

Fusini

Langella

Busilacchi

, et al. Real-time sonoelastography: principles and clinical applications in tendon disorders. A systematic review. Muscles Ligaments Tendons J 2018; 7:467–477.

Prado-Costa

Rebelo

Monteiro-Barroso

, et al. Ultrasound elastography: compression elastography and shear-wave elastography in the assessment of tendon injury. Insights Imaging 2018; 9:791–814.

Hatta

Giambini

Uehara

, et al. Quantitative assessment of rotator cuff muscle elasticity: Reliability and feasibility of shear wave elastography. J Biomech 2015; 48:3853–3858.

Hou

Merkle

Babb

, et al. Shear wave ultrasound elastographic evaluation of the rotator cuff tendon. J Ultrasound Med 2017; 36:95–106.

Baumer

Davis

Dischler

, et al. Shear wave elastography of the supraspinatus muscle and tendon: repeatability and preliminary findings. J Biomech 2017; 53:201–204.

Rotator cuff MRI grading. Available at: https://radiopaedia.org/articles/rotator-cuff-tear-grading.

10.

Sambandam

Khanna

Gul

, et al. Rotator cuff tears: an evidence based approach. World J Orthop 2015; 6:902–918.

11.

Baumer

Dischler

Davis

, et al. Effects of age and pathology on shear wave speed of the human rotator cuff. J Orthop Res 2018; 36:282–288.

12.

Dischler

Baumer

Finkelstein

, et al. Association Between Years of Competition and Shoulder Function in Collegiate Swimmers. Sports Health 2018; 10:113–118.

13.

Tudisco

Bisicchia

Stefanini

, et al. Tendon quality in small unilateral supraspinatus tendon tears. Real-time sonoelastography correlates with clinical findings. Knee Surg Sports Traumatol Arthrosc 2015; 23:393–398.

14.

Hatta

Giambini

Itoigawa

, et al. Quantifying extensibility of rotator cuff muscle with tendon rupture using shear wave elastography: a cadaveric study. J Biomech 2017; 61:131–136.

15.

Hatta

Yamamoto

Sano

, et al. In vivo measurement of rotator cuff tendon strain with ultrasound elastography. J Ultrasound Med 2014; 33:1641–1646.

16.

Vasishta

Kelkar

Joshi

, et al. The value of sonoelastography in the diagnosis of supraspinatus tendinopathy-a comparison study. Br J Radiol 2019; 92:20180951.

17.

Lee

Joo

Kim

, et al. Real-time sonoelastography in the diagnosis of rotator cuff tendinopathy. J Shoulder Elbow Surg 2016; 25:723–729.

18.

Demirel

Baykara

Koca

, et al. Ultrasound elastography-based assessment of the elasticity of the supraspinatus muscle in impingement syndrome: does elastography has any diagnostic value? Clin Rheumatol 2018; 37:1661–1665.

19.

Ito

Numano

Mizuhara

, et al. A new technique for MR elastography of the supraspinatus muscle: a gradient-echo type multi-echo sequence. Magn Reson Imaging 2016; 34:1181–1188.