Ultrasound of the symptomatic shoulder arthroplasty: Spectrum and prevalence of periarticular soft tissue pathology.

Abstract

To describe our experience using ultrasound (US) to evaluate postoperative complications in the presence of in situ shoulder arthroplasty. Review of patients who underwent US evaluation following total shoulder arthroplasty (TSA), reverse total shoulder arthroplasty (RTSA) or hemiarthroplasty from 2007 to 2020. All studies were reviewed independently by two musculoskeletal radiologists to assess for joint effusion, periarticular collection, and characterization of associated rotator cuff tears. Tendon tears were assessed with respect to (1) thickness: low grade (<50% thickness), high grade (>50% thickness), full thickness; (2) morphology (focal vs. diffuse) and location (insertion vs. critical zone). Inter-reader agreements were determined using Cohen's kappa test. Ninety-seven studies were performed in 72 patients following TSA, RTSA, or hemiarthroplasty. Thirty-seven exams were solely for diagnostic purposes, and 59 were for guiding joint or periarticular collection aspiration. Twenty-eight studies assessed the cuff tendons post TSA. The mean time between surgery and US examination was 29.2 months. Complete or high-grade tears were identified in 8/28 (28.6%) diagnostic exams. The most commonly torn tendon among TSA patients was the subscapularis, with 13/28 (46.4%) demonstrating at least partial tearing. Inter-reader agreement was excellent for presence of effusion (k=0.79, p < .001) and periarticular collection (k=0.87, p < .001), and excellent agreement for presence of subscapularis tear (k=0.78, p < .001), with fair agreement for assessment of supraspinatus (k=0.66, p < .001) and infraspinatus (k=0.60, p < .001) tears. The most commonly torn tendon following anatomic TSA identified by US was the subscapularis, which was torn or deficient in 46.4% of cases. The majority of studies were performed for the guidance of percutaneous aspiration.