Dedicated Training Program for Shoulder Sonography

Objectives The purpose of this study was to define and report on the effect of a comprehensive musculoskeletal sonography training program to improve accuracy (sensitivity and specificity) for the diagnosis of rotator cuff tears in relatively inexperienced operators. Methods Before the training program was implemented, radiologists (n = 12) had a mean of 2 years (range, <1–12 years) of experience performing and interpreting musculoskeletal sonography. Pre‐ and post‐training shoulder sonographic results were compared to surgical reports or, in their absence, to shoulder magnetic resonance imaging or computed tomographic arthrographic results if within 2 months of the sonographic examination. A total of 82 patients were included in the pre‐training group (January 2010–December 2011), and 50 patients were included in the post‐training group (January 2012–June 2013). The accuracy, sensitivity, specificity, and positive and negative predictive values were determined for the presence or absence of supraspinatus and infraspinatus tendon tears. Results After implementation of the training program, the sensitivity of sonography for detecting full‐thickness rotator cuff tears increased by 14%, and the sensitivity for detecting partial‐thickness rotator cuff tears increased by 3%. Conclusions Quality improvement programs and acquisition standardization along with ongoing, focused case conferences for the entire care team increased the sensitivity of shoulder sonography for diagnosing both full‐ and partial‐thickness rotator cuff tears, independent of the years of operator experience.