Intraobserver and interobserver reliability of the Copeland-Levy classification for arthroscopic evaluation of subacromial impingement.

Abstract

Defining a simple and reliable classification for acromial and bursal impingement lesions is necessary to standardize terminology, to improve communication, and to allow better evaluation of the proper treatment of impingement lesions and rotator cuff tears. The purpose of this study was to assess orthopedic surgeons' intraobserver and interobserver reliability of the Copeland-Levy classification. Six fellowship-trained orthopedic surgeons reviewed shoulder arthroscopy videos of 69 consecutive patients who underwent shoulder arthroscopy for rotator cuff tear repair or subacromial decompression. The surgeons were asked to classify impingement lesions according to the Copeland-Levy classification. One month afterward, the surgeons were requested to repeat the evaluation of the same impingement lesions. Intraobserver reliability was calculated using Cohen's weighted κ. Interobserver reliability was calculated using Kendall's W. Overall intraobserver reliability for acromial and bursal lesions was κ = 0.86 (95% confidence interval, 0.82-0.9) and κ = 0.97 (95% confidence interval, 0.95-0.98), respectively. Interobserver reliability for acromial and bursal lesions was W = 0.87 and W = 0.92, respectively. Intraobserver and interobserver reliability of the Copeland-Levy classification among senior orthopedic surgeons is excellent. Hence, we suggest the Copeland-Levy classification be used to standardize terminology of the subacromial impingement lesion.