Classifying glenohumeral synovitis: a novel intraoperative scoring system

Background: Synovitis of the shoulder is a common entity that is poorly described. This study aims to create a simple and reliable classification system for glenohumeral synovitis, which would benefit further research related to synovitis and outcomes. Methods: Twenty 30-second shoulder arthroscopy videos were distributed to 19 fellowship-trained orthopedic surgeons. The observers responded with their years in practice, fellowship type, whether synovitis affects outcomes, and whether synovitis affects plans. The surgeons then rated the videos based on the following: color of capsule (pale, pink, or red); villous projections (none, few, or extensive); capillaries in capsule (scattered or hypertrophied); and axillary recess (normal or contracted). Scores ranging from 0 to 6 were assigned. The videos were randomized and redistributed at a later date. Statistical analysis used an intraclass correlation coefficient with a mixed-effects model to calculate variability based on observer. Results: Nineteen observers completing the survey twice resulted in 760 videos being scored. There were 12 shoulder surgeons and 7 sports surgeons. Only 4 surgeons believed that synovitis did not affect outcomes, and the remaining 15 believed that it did. The intraclass correlation coefficient showed that 68% of the variation in measured scores was due to variation among patients and only 4% was due to variation among observers. There was no significant variation seen in scores due to surgeon experience, surgeon specialty, or first and second viewing. Conclusions: This interclass observer reliability shoulder synovitis study defined a system with excellent reliability among a range of surgeons with diverse training and experience. In addition, there was excellent reliability for the same surgeon between viewings. (C) 2017 Journal of Shoulder and Elbow Surgery Board of Trustees. All rights reserved.