Evaluation of the glenoid track tomographic method in magnetic resonance imaging/ARTHRO-MRI

Abstract

Objective To evaluate and compare the glenoid track method in 3D-reconstructed computed tomography (3D-CT) scans with magnetic resonance imaging (MRI) and/or arthro-MRI. Methods Forty-four shoulders with clinical and radiographic diagnosis of traumatic anterior instability were assessed using 3D-CT, MRI, and/or arthro-MRI scans. Glenoid track (GT), Hill-Sachs interval (HSI), and glenoid bone loss (GBL) were determined by a radiologist using 3D-CT images, and classified as on-track/off-track. Three surgeons, blinded to the radiologist's evaluation, performed the same determinations using MRI/arthro-MRI. Descriptive analysis, variance analysis, results disagreement analysis, and receiver operating characteristic (ROC) curves were performed. Results Results from the 4 examiners were fully consistent in 61.4% of the cases. Magnetic resonance imaging/arthro-MRI diagnosed off-track injuries with 35 to 65% sensitivity and on-track injuries, with 91.67 to 95.83% specificity. Accuracy ranged from 68.1 to 79.5%. The greatest data divergence occurred for off-track injuries diagnosed by MRI/arthro-MRI. The greatest data variability referred to HSI calculation. Higher HSI and GBL values were associated with greater disagreement among examiners. Hill-Sachs interval values were lower at MRI/arthro-MRI when compared to 3D-CT. Agreement between CT and MRI/arthro-MRI for the GT method was only moderate (kappa value, 0.325-0.579). Conclusion Magnetic resonance imaging/arthro-MRI showed low accuracy and moderate agreement for the GT method; as such, it should be used with caution by surgeons.