Intraobserver and interobserver reliability of the modified Walch classification using radiographs and computed...

Area of Science:

• Orthopedic surgery
• Radiology
• Medical imaging analysis

Background:

• The Walch classification system is used for categorizing glenoid morphology in primary glenohumeral osteoarthritis.
• A modified Walch classification was recently introduced, adding categories B3 and D and refining subtypes A2 and C.

Purpose of the Study:

• To evaluate the intraobserver and interobserver reliability of the modified Walch classification system.
• To compare the agreement of the classification using plain radiographs versus computed tomography (CT).

Main Methods:

• Three fellowship-trained shoulder surgeons independently classified 100 shoulders with primary glenohumeral osteoarthritis.
• Assessments were performed using both radiographs and CT scans across four separate sessions.
• Reliability was statistically analyzed using the kappa (κ) coefficient.

Main Results:

• Substantial intraobserver agreement was found for both radiographs (average κ=0.73) and CT scans (average κ=0.73).
• Moderate interobserver agreement was observed for radiographs (average κ=0.55) and CT scans (average κ=0.52).
• The distribution of classifications included A1 (18), A2 (12), B1 (20), B2 (25), B3 (22), C (1), and D (2).

Conclusions:

• The modified Walch classification demonstrates substantial intraobserver agreement.
• Interobserver agreement for the modified Walch classification was fair, indicating potential variability between different observers.
• Automated computer-based analysis of CT scans may enhance the utility of this classification system.