Comparison between 2D and 3D computed tomography evaluation of glenoid bone defect in unilateral anterior gleno-humeral instability

• Published in: Radiologia medica Vol. 117; no. 1; pp. 102 - 111
• Main Authors: Magarelli, N., Milano, G., Baudi, P., Santagada, D. A., Righi, P., Spina, V., Leone, A., Amelia, R., Fabbriciani, C., Bonomo, L.
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01.02.2012
• Subjects: Adolescent; Adult; Chi-Square Distribution; Confidence Intervals; Diagnostic Radiology; Female; Humans; Humerus - diagnostic imaging; Imaging; Imaging, Three-Dimensional; Interventional Radiology; Joint Instability - diagnostic imaging; Male; Medicine; Medicine & Public Health; Middle Aged; Musculoskeletal Radiology; Neuroradiology; Radiographic Image Interpretation, Computer-Assisted; Radiologia Muscoloscheletrica; Radiology; Shoulder Joint - diagnostic imaging; Tomography, X-Ray Computed - methods; Ultrasound; Glenoid; Computed tomography; Shoulder; Instabilità; Tomografia computerizzata; Spalla; Glena; Instability; Bone defect; Difetto osseo
• ISSN: 0033-8362; 1826-6983
• DOI: 10.1007/s11547-011-0712-7

Purpose This study evaluated the agreement between 2D and 3D computed tomography (CT) measurements in identifying the size and type of glenoid-bone defect in anterior glenohumeral instability. Materials and methods One hundred patients affected by unilateral anterior glenohumeral instability underwent a CT of both shoulders. Images were processed with both 2D [multiplanar reconstruction (MPR)] and 3D [volumerendering (VR)] methods. The area of the missing glenoid was calculated in comparison with the healthy glenoid and expressed as a percentage. Agreement between the two measurements was assessed according to the Bland-Altman method; a 5% mean difference was considered as clinically relevant. Results Analysis of agreement between MPR and VR measurements of the percentage of missing glenoid showed a mean difference equal to 0.62%±1.96%. Percent agreement between the two measurements in detecting the presence of bone defect was 97% ( p <0.0001). Percent agreement between the two measurements in discriminating the type of bone defect was 97% ( p <0.0001). Conclusions Agreement between 2D (MPR) and 3D (VR) CT measurements to identify the size and type of glenoid-bone defect in anterior glenohumeral instability was so high that the two measurements can be considered interchangeable.