2015 ISAKOS Biennial Congress ePoster #1003
Predicting Subscapularis Tendon Tears on Magnetic Resonance Imaging
Mai Katakura, MD, Kisarazu, Chiba JAPAN
Hideya Yoshimura, MD, PhD, Kawaguchi, Saitama JAPAN
Tomoyuki Mochizuki, MD, PhD, Tokyo JAPAN
Akimoto Nimura, MD, PhD, Tokyo JAPAN
Takashi Ogiuchi, MD, Kawaguchi, Saitama JAPAN
Masaya Hayashi, MD, PhD, Kawaguchi, Saitama JAPAN
Masayuki Shimaya, MD, PhD, Kawaguchi, Saitama JAPAN
Daisuke Hatsushika, MD, PhD, Kawaguchi, Saitama JAPAN
Mari Uomizu, MD, Bunkyo Ku, Tokyo JAPAN
Kawaguchi Kogyo General Hospital, Saitama, JAPAN
FDA Status Not Applicable
Summary: Preoperative MRI scans of the shoulder on the axial and oblique sagittal views reliably predict subscapularis tendon tears.
The most proximal portion of the subscapularis tendon supports the long head of biceps (LHB), indicating that repair of this part is important for the stability of LHB. However, several authors previously reported that preoperative Magnetic Resonance Imaging (MRI) scans of the shoulder do not reliably predict subscapularis tendon tears. The purpose of this study was to evaluate the diagnostic accuracy of MRI assessment of the subscapularis tendon tears by retrospectively comparing preoperative MRI findings with the results determined by arthroscopic evaluations of the same shoulders.
Consecutive sixty-four patients who received primary arthroscopic rotator cuff repairs at one institution were participated in this retrospective review. MRI findings of the most proximal portion of the subscapularis tendon were evaluated on the axial and oblique-sagittal images, and were classified into five grades by one experienced shoulder surgeon. In T2-weighted axial and oblique-axial view, slices around the upper margin of the lesser tuberosity, which exactly is the anatomical site of the superior-most insertion of the subscapularis tendon, was carefully detected and used for assessment. The grades include grade0: equal low intensity tendon with normal thickness, grade1: partial change of intensity, grade2: insufficient thickness, grade3: partial high intensity, and grade4: high intensity around the lesser tuberosity. MRI scores were obtained by adding grades of axial view and oblique-sagittal view. Then MRI scores and arthroscopic evaluations of the subscapularis tendon tears of each shoulder were compared.
Result:According to the arthroscopic findings, 28 patients(44%) had complete subscapularis tendon tears and 16 patients(25%) had superficial subscapularis tendon tears. 23 out of 28 patients with complete subscapularis tendon tears had MRI score of 4 or more and 34 out of 36 patients without complete subscapularis tendon tears had MRI score less than 4. So when we evaluate MRI score of 4 or more as having subscapularis tendon tear, the sensitivity was 82% and specificity was 94%. 38 out of 44 patients with superficial or complete subscapularis tendon tears had MRI score of 2 or more and 15 out of 20 patients without superficial or complete subscapularis tendon tears had MRI score less than 2, which means the sensitivity was 86% and specificity was 75%
Some reports said preoperative MRI scans of the shoulder do not reliably predict subscapularis tendon tears. However, we found good diagnostic accuracy of preoperative MRI. The most proximal portion of the subscapularis tendon was reported to be located on the upper margin of the lesser tuberosity, so we carefully chose the appropriate slice of MRI based on this anatomical findings and this may be one of the reasons why we got good diagnostic accuracy.
Conclusion:Preoperative MRI scans of the shoulder on the axial and oblique sagittal views reliably predict subscapularis tendon tears.