The Changes of Pennation Angle in the Rotator Cuff Muscles with Torn Tendons

INTRODUCTION: Rotator cuff tear is one of the most common disorders of the shoulder joint among middle aged or elder population. When a tear occurs, the corresponding muscle loses its function due to a loss of tendon continuity and the changes of muscular structure. The pennation angle, which was defined as an angle between the muscle fiber with its inserting intramuscular tendon, is one of the four architectural parameters of pennate muscles. It is reported that the pennation angle increased after sectioning the infraspinatus tendon in a sheep model . However, there has been no report on the pennation angle changes in the human rotator cuff muscles with torn tendons. The purpose of this study was to determine the changes of pennation angle that occurred after rotator cuff tears in human cadaver shoulders. We further hoped to clarify the relationship between the pennation angle of the rotator cuff muscles and the size of tendon tears. METHODS: In a group of 61 cadaveric shoulders from the Department of Anatomy and Anthropology, Tohoku University School of Medicine, we found 10 shoulders (in 6 cadavers) with various types of rotator cuff tears. These 10 shoulders formed the tear group. Another ageand gendermatched group of 10 shoulders with intact rotator cuff tendons was selected as the control group. The shoulders in both groups were free of other disorders of the shoulder. After removing the superficial soft tissue, the status of the rotator cuff was carefully examined. There were 3 shoulders with partial thickness tear and 7 shoulders with full thickness tear. The tear width (anterior-posterior dimension) and length (medial-lateral dimension) of full-thickness rotator cuff tears were measured using a digital caliper. The length was defined as a distance from the most retracted tear edge (all were in the supraspinatus tendon) to the lateral border of the footprint, whereas the width of the tear was defined as the length of the footprint exposed due to tendon rupture. Because of the curvature of the lateral aspect of the proximal humerus, the exposed lengths of the bony structures (footprint on the lesser tuberosity, the bicipital groove, the superior facet, and the middle facet of the greater tuberosity) were measured separately, and then summed. Then the muscles were removed from the scapula. The muscle fibers were carefully removed slice by slice (1.5 2 mm in thickness) in the plane parallel to the surface of the muscle till we reached the slice which contained the intramuscular tendon. This slice was named the central slice. Photographs were taken parallel to the plane of these central slices with use of a specialized photographing device (SL MPS-II, Sugiura Lab., INC, Tokyo, Japan). On the image of each central slice, 10 muscle fibers were selected with 5 fibers on each side of the intramuscular tendon. These 10 muscle fibers were selected from 10 different portions of the muscle so that they were likely to represent the entire muscle fibers on the central slice. By using an image software (Photoshop version 6.0, Adobe Systems Corporation, San Jose, CA, USA), we drew lines along the selected muscle fibers and along the central axis of the intramuscular tendon. An angle measuring software (MB-Ruler 3.4, MB Software Solutions, LLC, Baltimore, USA) was then used to measure the pennation angle, which was formed by the line along the muscle fiber and the line along the central axis of the intramuscular tendon. The values in each central slice were averaged. In the subscapularis muscle, there were several intramuscular tendons. We separately measured the pennation angle in each subdivision of the muscle and averaged the data. The humeral head diameter was also measured with use of a digital caliper. The original tear width and length (mm) were normalized by dividing the values by the humeral head diameter (mm). The normalized area was also calculated by multiplying the normalized width and the normalized length of the tear. The data were analyzed using SPSS (version 13.0, Chicago, IL, USA). Independent t-test was used to compare the pennation angles of the four rotator cuff muscles between the two groups. Pearson’s correlation coefficient was used to clarify the relationship between the size (normalized width, length, and area) of tear and the pennation angle in the supraspinatus and infraspinatus muscles. The level of significance was set at p < 0 .05. RESULTS: 1. The difference of the mean pennation angles between the control group and the tear group In the control group the mean pennation angles of the subscapularis, the supraspinatus, the infraspinatus and the teres minor were 9.6°±1.7°, 10.2°±2.3°, 12.5° ±2.1° and 6.0°±1.1° respectively, while in the tear group they were 13.3°±5.9°, 18.6°±10.0°, 23.9° ±10.4° and 6.5°±1.7° respectively. The differences were significant in the supraspinatus muscle and the infraspinatus muscle. 2. The correlation of tear size with the mean pennation angles in the supraspinatus and infraspinatus muscles In the 7 specimens with full thickness rotator cuff tears, there was a strong positive correlation between the normalized length of the tear and the pennation angle of the supraspinatus muscle (r=0.854, P=0.014) (Figure 1). There was also a moderate positive correlation between the normalized area of the tear and the pennation angle of the infraspinatus muscle (r=0.759, P=0.048) (Figure 2). DISCUSSION: This is the first study focusing on the changes of pennation angle of human rotator cuff muscles after rotator cuff tears. We observed that the tear group had larger pennation angles in the supraspinatus and infraspinatus muscles as compared to the control group. Moreover, the mean pennation angle was positively correlated with the tear length in the supraspinatus and the tear area in the infraspinatus. The supraspinatus is the most commonly affected muscle in cuff-tear shoulders, and thus, the tear length might have directly affected the pennation angle of the supraspinatus, whereas the infraspinatus is less commonly affected than the supraspinatus, which may have resulted in less effect on the correlation results. Another explanation would be as follows. The most retracted edge of full-thickness rotator cuff tear was always located in the supraspinatus muscle in the current series, thus the tear length was directly related to the retraction distance of the supraspinatus muscle. Based on these results, we assumed that the length of the rotator cuff tear can be used as a reliable predictor for the pennation angle changes of the supraspinatus muscle. While both the length and the width of the tear should be considered when predicting the structural changes of the infraspinatus muscle. CONCLUSIONS: The pennation angles of the supraspinatus muscle and the infraspinatus muscle increased significantly in shoulders with rotator cuff tears. The larger the tear size, the greater the pennation angle in these two muscles. REFERENCE: 1. Meyer DC, et al. J Orthop Res. 2004 Sep;22(5):1004-1007