Summary

Glenoid bone loss measured with a linear method is easy, reproducible and has excellent interobserver agreement.

Abstract

Introduction

Glenoid bone loss is associated with anterior shoulder instability and is considered an independent risk factor for arthroscopic capsulolabral repair failure. Therefore, quantitative analysis and accurate measurement are important to determine proper surgical treatment. The aim of the study was to assess interobserver variability of linear method for glenoid bone loss quantification.

Methods

Thirty patients with shoulder instability and CT scans were included. Images were processed in multiplanar reconstruction (MPR) to provide an en face view of the glenoid
Linear measurement applying the perfect circle method was performed. Each measurement was performed by four observers with a standardized measurement protocol. Interobserver reliability were analyzed using intraclass correlation coefficients (ICCs), 95% confidence intervals (CIs)

Results

Mean values and standard deviation (± SD) of glenoid bone loss were 16,7% ± 9,4 (range 0–36,2); 15,8 % ± 9,2 (range 0–37); 15,7 %± 8,2 (range 1,3-35,7); and 16,5 % ± 9,9 (range 1,5-38,1) for observer number 1, 2,3 and 4 respectively, with no significant differences (p=0,96). Interobserver reliability showed ICC values from 0.86 to 0.95.

Conclusions

This study showed that glenoid bone loss measured with a linear method is easy, reproducible and has excellent interobserver agreement.