The application of shear wave elastography with ultrasound for rotator cuff tears: a systematic review

Abstract

Background

Shear wave elastography (SWE) is an emerging ultrasound-based technology that provides a quantitative assessment of musculoskeletal tissue integrity. This systematic review investigates the use of SWE in the evaluation of rotator cuff tears.

Methods

PubMed, Embase, Web of Science, Google Scholar, and the Cochrane Library databases were searched for relevant studies from 1901 up to June 2022. Articles utilizing SWE in rotator cuff tears were selected based on inclusion and exclusion criteria. The studies included involved the assessment of shear wave velocity, tendon thickness and stiffness after healing, and fatty infiltrates evaluation using SWE. The Newcastle-Ottawa Scale was used to evaluate the risk of bias in included observational studies. Double-sided P value < .05 was considered statistically significant.

Results

Sixteen studies comprising 520 patients were included in the systematic review. SWE demonstrated that shear wave velocities in torn supraspinatus tendons were lower than in healthy supraspinatus tendons. A decrease in tendon SWE modulus elasticity was observed in tendinopathic tendons. Shear wave velocity decreased with increasing fat content and muscle atrophy. The velocity of SWE in muscle in re-tear groups was greater than in the healed group at 1 month after surgery (P < .05).

Conclusion

SWE ultrasound of the supraspinatus tendon can be a useful diagnostic tool for orthopedic surgeons that provide quantitative information on tendinopathic stiffness, velocity, fatty infiltrate, and elasticity characteristics. Decreased tendon velocity of SWE may predict recurrent rotator cuff tears and be useful in postoperative evaluations for muscle healing to plan for future management.

Rotator cuff disorders are the most common cause of shoulder pain and dysfunction.²¹ Up to 70% of all shoulder injuries are related to rotator cuff pathologies, and a majority are secondary to tears of muscle or tendons.²² Supraspinatus is the most common rotator cuff tear and comprises 70% of tears in people >80 years as it bears the majority of the shoulder-stabilizing strain.⁶ Supraspinatus tears can be categorized as partial, intrasubstance, or full thickness, and is dependent on the depth of the tear into the muscle fibers.²² The pathogenesis and mechanism of rotator cuff tears are unclear; however, trauma, advancing age, and repetitive stress are well-known risk factors.^32,34 Even when healed, 37%-40% of tendons re-tear within 16 years postoperatively and 7% require reoperation.^4,7

Conventional imaging modalities for rotator cuff pathologies include magnetic resonance imaging (MRI), MRI arthrogram, and brightness mode gray-scale ultrasound.¹ Shear wave elastography (SWE) is an emerging ultrasound-based technology that quantitatively identifies tissue stiffness. SWE generates shear waves through acoustic radiation force and measures the shear wave velocity (SWV) as it propagates through the tissue of interest, expressed in m/s. SWE furthermore compares the tissue of interest to surrounding tissues to identify tissue density and uses SWV to calculate tissue stiffness. The shear modulus (G) is defined as the ratio of stress to strain that is given as G = ρc_s², p is the tissue density and c_s is the SWV. Additionally, there is a direct relationship between G and Young’s modulus (E) which is defined as E = 3G; therefore, some studies refer the shear wave values to G or E through the unit of kPa.^8,35

Ultrasound elastography has historically been used to classify hepatic fibrosis and breast lumps where stiffness translates to clinical pathology.⁸ SWE when used on muscle, senses muscle architecture changes and allows for an insight into its biomechanical properties. A shear wave color elastogram demonstrates tendons with high velocity, those colored red on SWE, reflect healthy tendon integrity and high stiffness, whereas tendons with reduced velocity are colored blue, denoting reduced tendon integrity and stiffness (Fig. 1, A and B).^14,31

Figure 1.

(A) Basic physics of SWE, in step 1 the shear waves are generated using acoustic radiation force, in step 2, the fast plane wave excitation is used to tract displacement and velocity as shear waves propagate and the tissue displacement is calculated using an algorithm. In step 3, the tissue displacements are used to calculate the shear wave velocity (C_s) and shear modulus. (B) Relationship between shear wave velocity and shear modulus, the tissue density equates to of water (1 g/cm³). SWE, shear wave elastography; ARI, autoregulation index; SW, shear wave; LW, longitudinal wave; RX, constant receive; US, ultrasound.

Despite its ability to give insight into muscle health and viability, the role of SWE in rotator cuff tears is currently limited. SWE has typically been used to diagnose rotator cuff tears but may also be repurposed as a convenient and cost-effective alternative to MRI in the postoperative setting and help identify patients with poorer healing and at a higher risk of re-tears. Therefore, this systematic review summarizes the application of SWE for rotator cuff tears, and its potential future use in rotator cuff repair treatment.

Methods

Literature search

This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines and was listed retrospectively on the PROSPERO International Prospective Register of Systematic Review (CRD42021300155). Medline (via PubMed), Embase, Web of Science, and the Cochrane CENTRAL databases were searched for relevant articles from infinity to June 2022. The search keywords in different combinations included: “shear wave elastography,” “SWE,” “sonoelastography,” “ultrasound,” “tendon,” “rotator cuff,” “rotator cuff tears,” “rotator cuff repair,” and “supraspinatus.” The titles and abstracts of eligible studies were manually screened by 2 authors (I.S. and G.B.) based on relevance. Any disagreements on eligibility of studies between the two authors were resolved by discussion with a third reviewer (S.A.). Additionally, the reference lists of identified articles were checked manually, and full texts of the studies were assessed for eligibility.

Eligibility criteria

There are currently no randomized controlled trials of SWE use in rotator cuff tears. Therefore, observational studies (case control, case cohort, cross-sectional, case series, retrospective, or prospective studies) were included that assessed SWE for the imaging of the supraspinatus tendon. All cadaveric studies, animal models, reviews, conferences, abstracts, non-English articles, and studies not reporting on rotator cuff tears were excluded from the systematic review.

Study selection and data extraction

The title/abstract of studies identified during the search were imported into Endnote X10 for preliminary screening and removal of duplications. Screening for eligible studies was conducted and followed by a full-text screening by 2 independent reviewers (I.S. and G.B.), and any disparity in selecting eligible articles or assessing findings was resolved through consultation of a third author (A.S.).

Requisite data were extracted by three independent authors (I.S., G.B., and S.A.) into a data extraction form. The extracted data included the following items: first author, year of population, study design, sample size, tendon location, and key findings for each included study. The outcomes included the assessment of SWV, tendon thickness and stiffness after healing, and fat amount evaluation using SWE.

Risk of bias assessment

The Newcastle-Ottawa scale assessed the risk of bias within the observational studies.³⁰ Using this scale, each included study was evaluated based on three essential domains: A) selection of the study subjects (such as included patients and their representative population), and the exposure (rotator cuff tears) was ascertained based on predefined criteria; B) comparability of groups on demographic characteristics and important potential confounders (achieved by adequate control of secondary risk factors); and C) ascertainment of the prespecified outcome (exposure/treatment) based on SWE assessment with sufficient follow-up duration. Due to the absence of a control group, components relating to the selection of control groups (question number 2) were removed from the Newcastle-Ottawa Scale. As such, the total score was out of 8 rather than 9. Each article was evaluated by 2 independent reviewers (I.S. and G.B.) and any disagreements were resolved by discussion with a third reviewer (S.A.).

Results

Search strategy results

The literature search yielded 461 records. After title/abstract screening, 32 were retrieved and screened for eligibility. After full-text screening, 16 studies were included in the systematic review. The study selection process is shown in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis flow diagram, Figure 2.

Figure 2.

PRISMA flow of study selection. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis.

Characteristics of included studies

A total of 16 studies comprising 520 patients were included in the current systematic review.^{1,3,5,8,10,16,17,19,23,}26, 27, 28, 29^,31,33 All included studies were observational studies, from which four were validity studies, 3 prospective, 2 case control, 1 retrospective, 1 cross-sectional, and 1 case series. Summary of included studies with key findings of each study was reported in Table I. Details for risk of bias assessment are shown in Table II.

Table I.

Characteristics of included studies.

Study	Study design	Sample size	Tear location	Key findings
Chua et al 202112	Prospective	21 patients	Supraspinatus tendon	Sutures migrating to the middle of the tendon during the postoperative healing process is a normal phenomenon observed on ultrasonography.
Deng et al 20213	Validity	80 patients	Supraspinatus tendon	SWE with SWV might identify degree of supraspinatus tendon tear and improve the value of ultrasonography.
Lawrence et al 202132	Descriptive	22 patients	Supraspinatus tendon	Utility of ultrasound SWE in this population (ie, patients with a small to medium supraspinatus rotator cuff tear) before surgical rotator cuff repair remains unclear.
Nocera et al 20218	Prospective cohort	12 patients	Supraspinatus tendon	Significant correlation occurred between TMR and SWE at 6 mo or with power Doppler at any time point.
Sakaki et al 202121	Case control	26 participants	Supraspinatus muscle	Anterior superficial region in patients with rotator cuff tear was mainly responsible for reduced active stiffness.
Sakaki et al 202128	Case series	8 patients	Supraspinatus muscle	Active stiffness of the anterior superficial region may improve 6 mo rather than 3 mo postoperatively because of the different stages of muscle force, structural repair tendon strength, and remodeling.
Hackett et al 202014	Validity	20 patients	Supraspinatus tendon	Excellent intrarater trial agreement, with an intraclass correlation coefficient = 0.96. In the interrater testing, the mean shear wave velocity in normal tendons was 9.90 ± 0.07 m/s (= 294 kPa), with intraclass correlation coefficient = 0.45.
Lin et al 202023	Retrospective	88 patients	Supraspinatus muscle and tendon	SWE can detect biomechanical differences within the supraspinatus muscle that are not morphologically evident on gray-scale ultrasound.
Itoigawa et al 202036	Prospective cohort	60 patients	Supraspinatus muscle and tendon	The SWE value of the muscle in the retear group was greater than in the healed group at 1 mo after surgery (P < .05).
Itoigawa et al 201826	Descriptive	38 patients	Supraspinatus tendon	The highest correlation with stiffness of the supraspinatus musculotendinous unit was with the SWE modulus of the posterior deep muscle.
Baumer et al 201824	Case control	30 tendons	Supraspinatus muscle	Tendon stiffness positively associated with age under passive and active conditions; softer tendon resulted from muscle activation.
Gilbert et al 201730	Cross-sectional	42 patients	Supraspinatus muscle	SWE may be a sufficient tool in detecting and estimating the amount of fatty degeneration in the supraspinatus muscle in real time.
Hou et al 201731	Observational	53 tendons	Supraspinatus muscle	Decrease in tendon stiffness in the proximal tendon in symptomatic patients; no difference seen in the distal tendon.
Krepkin et al 201725	Validity	9 tendons	Supraspinatus muscle	Negative correlation between T2 MRI and tendon stiffness.
Rosskopf et al 201629	Validity	8 tendons	Supraspinous muscle	Excellent reliability (interclass CC: 0.89; intraclass CC: 0.7-0.8); stiffer tendon in controls than in patients.
Itoigawa et al. 201527	Descriptive	3 patients	Supraspinatus tendon	SWE combined with B-mode ultrasound imaging could be a feasible method for quantifying the local stiffness of the rotator cuff muscles.

SWE, shear wave elastography; SWV, shear wave velocity; TMR, tendon-to-deltoid muscle; MRI, magnetic resonance imaging; B-mode, brightness mode; CC, correlation coefficient.

Table II.

Risk of bias of included studies calculated with the Newcastle-Ottawa Scale.

Study ID	Selection			Comparability		Outcomes
	Representativeness of exposed cohort	Ascertainment of exposure	Outcome of interest not present at study start	Comparability of groups on 2ry risk factors	Control for confounders	Assessment of outcomes	Appropriate follow-up (length)	Adequacy of follow-up (loss)
Chua et al 202112	∗	∗	∗	∗	∗	∗	∗	∗
Deng et al 202113	∗	∗	∗	∗	∗	∗	-	∗
Lawrence et al 202132	∗	∗	∗	∗	∗	∗	-	-
Nocera et al 20218	∗	∗	-	∗	∗	∗	∗	-
Sakaki et al 202121	∗	∗	∗	∗	∗	∗	-	-
Sakaki et al 202128	∗	∗	∗	∗	∗	∗	∗	∗
Hackett et al 202014	∗	∗	-	∗	∗	∗	-	-
Lin et al 202023	∗	∗	∗	∗	-	∗	-	-
Itoigawa et al 202041	∗	∗	-	∗	-	∗	∗	∗
Itoigawa et al 201826	∗	∗	∗	-	∗	∗	∗	∗
Baumer et al 201824	∗	∗	-	-	∗	∗	-	NR
Gilbert et al 201730	∗	∗	-	-	∗	∗	∗	-
Hou et al 201731	∗	∗	∗	-	-	∗	∗	∗
Krepkin et al 201725	∗	∗	∗	∗	NR	∗	NR	-
Rosskopf et al 201629	∗	-	∗	∗	∗	∗	∗	NR
Itoigawa et al, 201527	∗	∗	∗	-	∗	-	∗	-

NR, no risk, ∗, high risk, -, medium risk.

Outcomes

Shear wave velocity

Baumer et al investigated the relationship between pathology and age on SWV under both passive (seated with arm at rest) and active conditions (seating and lifting their forearm and abducting their shoulder). Using SWE, Baumer et al showed that patients with rotator cuff tears had a lower mean SWV in their muscle and tendon under active conditions (P = .02) when compared to healthy participants. No difference was detected under passive conditions (P = .78). Baumer et al additionally showed that SWV was lower in the supraspinatus of older patients in both passive and active conditions (P = .049 and .039 respectively). Baumer et al also demonstrated SWV of the supraspinatus muscle and tendon had high repeatability, with intra- and interuser intraclass correlation coefficient values of greater than 0.87 and 0.73, respectively.² Day-to-day repeatability demonstrated intraclass correlation coefficient values greater than 0.33 for passive muscle, 0.48 for passive tendon, 0.65 for active muscle, and 0.94 for active tendon.² Furthermore, Krepkin et al found that SWV was negatively correlated with both T2∗ weighted values measured on MRI and tear size of the degenerated supraspinatus tendon.¹⁵ Deng et al reported that SWV of 80 supraspinatus tendon tears was 4.59 ± 1.00  m/s, and lower than uninjured supraspinatus tendon 6.68 ± 1.05  m/s, P  =  .01. SWV of supraspinatus tendon in tendinopathy (where the tendons are extensively thickened with diffuse non-uniform hypoechogenicity in ultrasound), partial tear, and full-thickness tear groups respectively were 5.66 ± 0.97, 4.66 ± 1.00, and 3.78 ± 0.55  m/s, all lower than that of the contralateral uninjured supraspinatus tendon, (P ≤ .05).⁵

Tendon thickness and stiffness after healing

Chua et al utilized SWE to assess supraspinatus tendon stiffness and elasticity up to 24 weeks postrotator cuff repair. The study observed an increased tendon modulus elasticity between 8 days and 24 weeks, from 154 ± 75 kPa to 209 ± 96 kPa, P = .05.³ Itoigawa et al reported that increasing supraspinatus stiffness measured by axial stretching along the direction of the muscle line of action correlated with SWE values of the posterior deep muscle of the supraspinatus (R = 0.69).¹² The mean SWE stiffnesses in the anterior deep, anterior superficial, posterior deep, and posterior superficial muscle regions of the supraspinatus muscle in vivo were 40.0 ± 12.4, 34.0 ± 9.9, 32.7 ± 12.7, 39.1 ± 15.7 kPa, respectively.¹³ Igotowa et al also found SWE stiffness values in the muscle in the re-tear group were greater than the healed group at 1 month postoperation (18.6 ± 10.6 vs. 12.0 ± 3.5, respectively, P = .03).

The measurement of muscle stiffness can also be an effective tool in evaluating rotator cuff tendinopathy. Leong et al demonstrated increased upper trapezius stiffness assessed by SWE was associated with rotator cuff tendinopathy in a study of 43 volleyball players during active (30° and 60° shoulder abduction) and passive (0° shoulder abduction) tasks.¹⁸ Moreover, a G of 12.2 kPa yielded a sensitivity of 0.73 and specificity of 0.86 (area under curve = 0.817; P = .001) for identifying athletes at risk of rotator cuff tendinopathy. Hackett et al¹⁰ found high reliability of SWE ultrasound to differentiate between normal and tendinopathic supraspinatus tendons (n = 20). Sakaki et al showed that the stiffness of the affected anterior superficial region of the supraspinatus muscle 12 months postoperatively was significantly higher than that measured preoperatively and 3 months postoperatively (P < .05).^26,27

Fat amount

Using SWE, Rosskopf et al observed a negative relationship between increasing fat content in the tendon with SWV values, and supraspinatus muscle atrophy and fat content peaked in severe supraspinatus tears.²⁵ Gilbert et al showed a good correlation between supraspinatus fatty infiltrates identified through MRI-spectroscopy and SWV values (P = .82).⁹ Hou et al stated that SWE may be useful in the preoperative setting for patient selection and surgical planning, as it can reflect tendon quality or postoperative failure rates.¹¹ Lawrence et al evaluated 39 patients with calcifying tendinopathy using SWE and found these values coincided with a hyperechoic pattern that predicted symptom relief following fine needle aspiration, thereby assisting management. Despite this, estimated G was not significantly associated with anterior/posterior tear size (P = .09), tear retraction (P > .2), occupation ratio (P > .11), or fatty infiltration (P > .30) under any testing condition, and contradicts findings reported earlier.¹⁶

Discussion

This is the first systematic review to corroborate literature on the use of SWE in the evaluation of rotator cuff pathologies. The study found SWE to be a valid and reliable tool for the evaluation of supraspinatus pathology. SWE can quantify rotator cuff tendon stiffness, an important marker for the mechanical integrity of tendon tissue, using different components such as SWV, G, and fat infiltration. Moreover, the range of studies demonstrated SWE to have value as a predictive measure for retears in the postoperative setting. SWE also correlated well with MRI staging for supraspinatus integrity. As SWE presents itself as a reliable, cost-effective, and clinically valuable imaging modality, its application to evaluating the pre-and postoperative use in rotator cuff tear patients should be entertained further by future studies.

The use of SWE in evaluating tendon integrity was an important outcome of the assessment in this study. As highlighted by Deng et al, supraspinatus tendinopathy and tears were associated with decreased SWV compared to contralateral healthy supraspinatus tendon (P < .05),⁵ and the ability to accurately evaluate SWV using SWE can be integral in screening patients at higher risk of supraspinatus tears. Deng et al also outlined the value of the E and its role in quantifying decreased stiffness of the supraspinatus tendon which was shown to be associated with tendinopathy.⁵ Similarly, Hackett et al reported SWE’s high reliability in differentiating between normal and tendinopathic supraspinatus tendons. Hackett et al observed a high interoperator reliability with SWV, inferring its utility and accuracy would be desirable in real-world settings.¹⁰ Therefore, SWE reliably evaluates the viscoelastic properties of the rotator cuff tendon, which can allow for timely surgical and rehabilitation planning where MRI is not available or appropriate in the clinical setting.

Moreover, this study found SWE to have good utility postoperatively in evaluating tissue integrity and predicting retears. Igatowa et al demonstrated that contrary to supraspinatus tendons preoperatively, the presence of higher-than-average SWV values 1-month post rotator cuff repair inferred the recurrence of further rotator cuff tears. Although the association was significant (P = .03), no predictive modeling was done by the authors to confirm this. Furthermore, the increase/decrease in fatty infiltration is a significant risk factor for retear. Conventionally, fatty infiltrates are assessed by MRI;²⁰ however, Gilbert et al demonstrated SWV correlated well with fatty infiltrates identified on MRI-spectroscopy,¹¹ further exemplifying the use of SWE in the postoperative setting. SWE offers a cheap, accessible option for monitoring of supraspinatus healing throughout the postoperative period. It can also assist in the prediction of retears and guide rehabilitation programs.

SWE is a good tool for the assessment of supraspinatus tissue integrity. The preliminary data also reports SWE may be more sensitive than MRI or gray-scale B-ultrasound for the identifications of subclinical muscle and tendon injuries and could be useful for early diagnosis, assessment in rehabilitation, and postoperatively for the risk of re-tears. Despite these findings, doubts concerning its efficacy as a diagnostic tool are not yet confounded as currently, no randomized control trials are available evaluating its real-world efficacy. Furthermore, most musculoskeletal changes are already obvious on conventional ultrasound with color Doppler imaging, and some SWE changes may not be clinically relevant. As such, it would be important to undertake a more systematic and structured approach to the investigation of SWE. We recommend the standardization of SWE for soft tissue applications based on the manufacturers' suggestions and consensus between users. Employing parameters such as the size of the elastogram, the use of adaptors/pads/gel, and scoring systems will allow for generalizable results between studies and make SWE more reliable. Better technical training by industry providers will also allow for the development of optimized protocols dedicated to musculoskeletal applications.

The indications for SWE also need to be considered, patients with the symptomatic but nonultrasound-evident disease and patients at high risk of rotator cuff tears would be good candidates for SWE, as it is more sensitive than conventional imaging in identifying early clinical changes. Additionally, as SWE is an emerging technology, normative data in healthy controls are lacking for comparison. Therefore, further studies should comprise of large populations of varied ages, ethnicity, and levels of activity with long-term follow-up. Relevant data should be compared with histology, conventional imaging (ultrasound and MRI), and biomechanical and clinical data.

Strengths and limitations

The current study highlights the potential role of SWE in management of rotator cuff tears. However, some limitations should be acknowledged, the literature on this topic is currently limited to small observational studies thus a definitive clinical recommendation through a meta-analysis was not possible. From the results of studies, SWE presents as an ideal instrument however only level 2-3 evidence is currently published, and the exact methodologies of studies is unclear. This likely injects some residual effect of bias, of which would only be resolved via the addition of randomized controlled trials and more large prospective cohort studies. The included studies also have several weaknesses such as poor study design, small sample size, the inclusion of different ethnicity and equal sex, and subjective heterogeneity between the studies were noted which can affect the reproducibility and comparison of results. Some methods and approaches to SWE were also not adequately reported, including methods for quantification, reporting of artifacts, and technique by different users, which limit the replicability by other studies. Lastly, although SWE allows for objective quantification and is reproducible by technicians, many commercial ultrasound devices do not yet provide SWE as an add-on feature.²⁴

Conclusion

SWE ultrasonography provides unique insight into muscle and tendon integrity by quantifying speed (velocity) and elasticity. The current literature suggests this may be of value to understanding the integrity of tendons and their mechanism of healing however multicenter controlled trials with long-term follow-up and a comparison to gold-standard imaging modalities are needed.

Disclaimers:

Funding: No funding was disclosed by the authors.

Conflicts of interest: George A.C. Murrell reports that he has served on the Editorial or governing board of Journal of Shoulder and Elbow Surgery; has served on the Editorial or governing board of Shoulder and Elbow; has served as a paid consultant for Smith & Nephew; and has received research support from Smith & Nephew. All the other authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.