Prevalence of subscapularis tears and accuracy of shoulder ultrasound in pre-operative diagnosis.

Abstract

Subscapularis tears can be difficult to diagnose and their treatment requires advanced arthroscopic skills. The objective of this study was to find the prevalence of subscapularis tears on arthroscopic examination of shoulders with rotator cuff pathology and to determine the accuracy of pre-operative ultrasound in diagnosing these tears. Ultrasound and intra-operative reports of 236 patients who underwent shoulder arthroscopy for rotator cuff pathology by the senior author at his institution were compared. Prevalence of subscapularis tear was noted and classified using Lafosse classification system. Ultrasound reports and intra-operative findings were compared to determine the accuracy, sensitivity and specificity of ultrasound in detecting subscapularis tears. The prevalence of subscapularis tears in patients needing rotator cuff repair was found to be 31.4%. A total of 6.4% of patients needing a rotator cuff repair had an isolated subscapularis tear. The sensitivity of ultrasound was 39.5% and specificity 93.1% in detection of these tears. The overall accuracy of ultrasound was 75.8%. Sensitivity of ultrasound was low (42.8%) for smaller (type 1 and 2) tears and higher (79%) for larger (types 3, 4, 5) tears. The overall positive predictive value of USS was 73.1% and negative predictive value 76.4%. The shoulder surgeon should be skilled in diagnosing and repairing subscapularis tendon tears arthroscopically and cannot completely rely on pre-operative ultrasound scans in ruling out smaller tears as its sensitivity in diagnosing smaller tears is quite low. Unsettling anterior shoulder pain with a normal ultrasound may need further arthroscopic evaluation to rule out missed subscapularis tears.