Subscapularis tendon tear classification and diagnosis: A systemic review and meta-analysis

Background In the current study, we performed a systematic review and meta-analysis regarding the comparison of accuracy, sensitivity, and specificity of the techniques in diagnosing SSC tendon tears. Also, we performed a systematic review of the classification of SSC tendon tears. Methods English language, peer-reviewed journal publications from the first date available to March 2022 were extracted by searching PubMed and Web of Science databases. A forest plot was used to graphically show the results of pooled sensitivity, specificity, and accuracy of different diagnostic modalities. Results There were six studies on using MRI to diagnose subscapularis tendon tears, five studies on MRI, four studies on clinical examination, one on ultrasonography, and one on CT arthrography. Pooled sensitivity values for MRI, MRA, clinical examination, ultrasonography, and CT arthrography were 0.71 (CI: 0.54; 0.87), 0.83 (0.77; 0.88), 0.49 (0.31; 0.67), 0.39 (0.29; 0.51), and 0.90 (0.72–0.97), respectively. The pooled specificity values for MRI, MRA, clinical examination, ultrasonography, and CT arthrography were 0.93 (CI: 0.89; 0.96), 0.86 (0.75; 0.93), 0.89 (0.73; 0.96), 0.93 (0.88; 0.96), and 0.90 (0.69; 0.98), respectively. The pooled diagnostic accuracy values for MRI, MRA, clinical examination, ultrasonography, and CT arthrography were 0.84 (CI: 0.80; 0.88), 0.85 (0.77; 0.90), 0.76 (0.66; 0.84), 0.76 (0.70; 0.81), and 0.90 (0.78; 0.96), respectively. Conclusion According to our systematic review and meta-analysis, MR arthrography was the most accurate in diagnosing subscapularis tears. MR arthrography was the most sensitive, and MRI and ultrasonography were the most specific in detecting subscapularis tears.


Background
The rotator cuff muscle group is formed by the subscapularis (SSC) muscle, supraspinatus muscle, infraspinatus muscle, and teres minor muscle at the posterior scapular region. The SSC muscle originates from the subscapularis fossa of the scapula, inserts into the lesser tubercle of the humerus, and is innervated by the superior and inferior subscapular nerves. The SSC muscle is the largest component of the aforementioned rotator cuff; it is responsible for the elevation and internal rotation of the shoulder and has a crucial role in stabilizing the glenohumeral joint (1). As the subscapularis tendon tear is a prevalent painful condition followed by a significant loss of function, timely diagnosis and management of this condition are of crucial importance (2). Until now, several classifications have been proposed by Fox et al., Lyons, Lafosse et al., and Touissant et al. (3)(4)(5)(6). However, there is no consensus regarding the classification of this condition, which may hinder clinical evaluation, diagnosis, and management.
The most sensitive and specific clinical examinations to assess an SCC tendon tear are the lift-off test, the belly-press test, and the bearhug test. Increased external rotation compared to normal shoulder rotation and weakness in internal rotation also help in the diagnosis (7). Although these tests are essential for the diagnosis of SSC tendon tears, various imaging modalities such as magnetic resonance imaging (MRI), computed tomography (CT) scanning, magnetic resonance arthrography (MRA), and ultrasonography (US) may also be indicated (8). Since the SCC plays a crucial role in shoulder function, evaluating diagnostic modalities is of great significance. Misdiagnosed SSC tendon tears may result in unbalanced force, persistent shoulder pain, and weakness even after cuff repair (9). The gold standard diagnostic test for SSC tendon tears is arthroscopy, which helps the physician precisely evaluate the humeral and glenoid aspects of the joint space (10). As arthroscopy is an invasive, technically demanding, and expensive technique, using the aforementioned clinical assessments and imaging modalities can replace the necessity of performing an arthroscopy to diagnose SCC tendon tears. While several studies have provided evidence-based guidelines for the examination, diagnosis, and management of SSC tendon tears, no previous systematic review and meta-analysis studies have evaluated the diagnostic accuracy of MRI, MRA, CT scanning, US, and clinical assessments in the diagnosis of SSC tendon tears. Also, as mentioned earlier in the manuscript, a consensus regarding the classification of SSC tendon tears is still lacking.
In the current study, we performed a systematic review and metaanalysis regarding the comparison of accuracy, sensitivity, and specificity of these techniques in diagnosing SSC tendon tears. Also, we performed a systematic review of the classification of SSC tendon tears.

Methods
We conducted a systematic review of the pieces of evidence for the diagnosis of subscapular tears with different diagnostic tools and studies on classification methods for subscapular tears. English language, peer-reviewed journal publications from the first date available to March 2022 were extracted by searching PubMed and Web of Science databases. The combination of the following search terms was used: subscapularis tear, diagnosis, and classification. After removing duplicate studies, the retrieved records were screened for title and abstract. The full text of eligible studies, selected from the previous step, was screened and reviewed. The data of interest were the sensitivity, specificity, and accuracy of different diagnostic tools in subscapularis tendon tears.
The eligible studies were observational studies (cross-sectional and cohort studies) on the diagnostic accuracy of imaging modalities and clinical assessments in subscapularis tendon tears. We excluded the studies that have one or more of the following criteria: (1) studies that were on rotator cuff tendons other than the subscapularis tendon (e.g., supraspinatus); (2) systematic reviews, meta-analysis studies, reviews, case reports, case series, and gray literature; (3) studies with no control group; (4) studies where the number of patients with subscapularis tendon tears was not specified; and (5) full text in any language other than English.
For meta-analysis, we also excluded studies as reference tests other than arthroscopy (e.g., MRI or MR arthrography) because arthroscopy is the gold standard for diagnosing subscapularis tendon tears.
For data analysis, we used a meta package in R statistical software (version 4.1.1). The sensitivity, specificity, and accuracy of different diagnostic tests were calculated with a 95% confidence interval (CI). Sensitivity was considered as true-positive cases divided by total patients with a subscapularis tendon tear; specificity was considered as true-negative cases divided by total cases with an intact subscapularis tendon (confirmed by arthroscopy). Accuracy was considered as true-positive and true-negative cases divided by total subjects in the study. The random-effects model was used for calculating pooled sensitivity, specificity, and accuracy. In this study, subgroup analysis was used to report pooled sensitivity, specificity, and accuracy for different diagnostic tools. The forest plot was used to graphically represent the results of calculated pooled sensitivity, specificity, and accuracy for different subgroups. The I 2 statistic was used to evaluate the heterogeneity in the included studies for each subgroup.

Results
The flow diagram of selected studies is shown in Figure 1. After reming duplicate records, 305 studies retrieved from online databases were screened for title andabstract. Thirty-two studies accomplished the inclusion criteria for full-text review. Six studies were excluded, leading to a final inclusion of 26 studies. The characteristics of the selected studies are presented in Table 1. Among them, seven studies were on the classification of the subscapularis tendon tear and 19 studies were on the diagnosis of subscapularis tendon tear. One study was excluded from meta-analysis because the reference test was MRI and not arthroscopy. Eighteen studies were included in our meta-analysis, and of 2,593 total subjects, 892 had subscapularis tendon tears.

Accuracy of different types of methods in the diagnosis of subscapularis tendon tears
There were six studies on using MRI to diagnose subscapularis tendon tears, five studies on MRI, four studies on clinical examination, one on ultrasonography, and one on CT arthrography. As shown in Figure 2

Classification of subscapularis tendon tears
Seven studies were found on the classification of subscapularis tendon tears (3-6, 29, 30). The different classification types are listed in Table 2.

Discussion
To our knowledge, this is the first systematic review and metaanalysis comparing the accuracy of different methods in diagnosing subscapularis tears. Our systematic review and meta-analysis showed that MR arthrography and CT arthrography were the most accuracte in diagnosing subscapularis tears, with accuracy values of 85% and 90%, respectively. However, there was only one study on Flow diagram of the selection process. were the most specific in detecting subscapularis tears, with a specificity of 93% for both of them. However, ultrasonography has much lower sensitivity, resulting in lower accuracy than MRI. Clinical assessment and imaging studies are the two main methods for diagnosing subscapularis tendon tears (9). Generally,   Forest plot of specificity of different diagnostic modalities.  ultrasonography has a more diagnostic accuracy for bicep tendon tears (93%) and supraspinatus tendon tears (83%) compared to subscapularis tendon tears (76%). However, ultrasonography is highly specific in detecting subscapularis tendon tears (93%) compared to other imaging modalities. Thus, a positive result could be considered for subsequent definitive diagnosis and management Yoo and Rhee's subscapularis tendon tear classification

IV
Up to the second facet tear: the first and second facets are exposed with much more medial retraction of the tendon, which is approximately a two-thirds tear of the entire subscapularis superior-inferior length (the entire tendinous portion) Frontiers in Surgery procedures such as arthroscopy, but a negative result needs more diagnostic tests to approve. Further studies on assessing the diagnostic accuracy of ultrasonography for subscapularis tendon tears are required to make more definitive conclusions. MRI and MR arthrography are more reliable compared to ultrasonography or clinical assessment for the diagnosis of subscapularis tendon tears (33). A meta-analysis conducted by Malavolta et al. on the efficacy of MRI and MR arthrography in the diagnosis of subscapularis tendon tears showed that the pooled sensitivity and specificity of MRI and MR arthrography in the diagnosis of the subscapularis tendon tears were 68% and 90%, respectively. However, this study did not conduct a separate meta-analysis for MRI and MR arthrography (33). Our study is in line with this meta-analysis as the sensitivity of MRI and MR arthrography is lower than their specificity, which means that MRI and MR arthrography have lower false-positive cases than false-negative cases. Previous studies show that the diagnostic value of MRI is higher in a complete tear of the subscapularis tendon (Type 4 by the Lafosse classification) (16) and the tear of the other rotator cuff tendons (34). The relatively lower sensitivity of MRI and MR arthrography for subscapularis tendon tears than that for other rotator cuff tendons could be explained by its three-dimensional footprint topography of the humeral head, which is explained by Yoo et al. (30), or by some noninsertional types of subscapularis tears (35). $$$$The accuracy also increases with the higher expertise of the reviewer (36). On the other hand, the diagnostic accuracy of MRI and MR arthrography is not affected by the time elapsed from injury to perform the imaging study (16).
Considering the accuracy of each diagnostic tool is important to obtain good clinical and functional outcomes in the treatment of subscapularis tendon tears (37, 38), although treatment of elderly patients is often limited surgically due to the bad quality of the tissue. Hence, a shoulder replacement could be indicated (39).
Classifications of Lafosse ( Figures 5A-E), Fox, Lyons, Martetschlager, and Toussaint are based on the insertion site lesions and according to anatomic data and arthroscopic lesion-related findings (3-6, 29). Yoo et al. described a classification based on a three-dimensional anatomic footprint (30). Dierckman's classification was based on noninsertional tendinopathy of the subscapularis (40). None of the classifications included interstitial tears of the subscapularis tendon described by Saremi et al. (Figure 5F) (35).
The main strengths of this study are as follows: (1) this is the first systematic review and meta-analysis comparing the accuracy of different methods in diagnosing subscapularis tears; (2) this study compared not only the different imaging modalities but also the accuracy of imaging modalities with clinical examinations; and (3) we conducted a meta-analysis of the sensitivity and specificity in addition to the accuracy of different diagnostic modalities.
The limitation of this study is that we did not compare the specific clinical tests, and we also did not compare complete and partial tears of the subscapularis tendon tear. However, adding another level of subgroups could have affected the possibility of conducting a meta-analysis.

Conclusion
According to our systematic review and meta-analysis, MR arthrography was the most accurate in diagnosing subscapularis tears, MR arthrography was the most sensitive, and MRI and ultrasonography were the most specific in detecting subscapularis tears. Further studies on assessing the diagnostic accuracy of ultrasonography and CT arthrography for subscapularis tendon tears are required to make more definitive conclusions.