Interobserver agreement in the classification of rotator cuff tears

Area of Science:

• Orthopaedic Surgery
• Musculoskeletal Imaging
• Biomechanical Analysis

Background:

• Six rotator cuff tear classification systems exist to guide understanding and treatment.
• These systems aim to standardize the description of rotator cuff pathology.

Purpose of the Study:

• To evaluate the interobserver variability of existing rotator cuff tear classification systems.
• To determine which classification method offers the highest interobserver agreement.

Main Methods:

• A literature search identified six rotator cuff tear classification systems.
• Twelve fellowship-trained orthopaedic surgeons reviewed 30 arthroscopic videos of rotator cuff tears.
• Tears were classified using the six systems, and interobserver variation was assessed via kappa analysis.

Main Results:

• High agreement was found for differentiating full-thickness from partial-thickness tears (kappa = 0.85).
• Agreement was high for the side of involvement in partial-thickness tears (kappa = 0.85) but poor for depth (kappa = 0.19).
• The best agreement for full-thickness tears was based on topography/retraction in the frontal plane (kappa = 0.54).

Conclusions:

• Most rotator cuff tear classification systems exhibit low interobserver agreement among experienced surgeons.
• Distinguishing tear types and side of involvement in partial tears are exceptions.
• Describing full-thickness tears by frontal plane topography (degree of retraction) is recommended for future research.