Comparative Analysis of 2- Versus 3-Dimensional Sonography of the Supraspinatus Tendon

Abstract

The purpose of this study was to determine the reliability in the analysis of images acquired using a dedicated 3-dimensional (3D) ultrasound transducer and conventional 2-dimensional (2D) images in the detection and characterization of supraspinatus tendon tears. Images of the supraspinatus tendon in 42 patients who had undergone 2D and 3D sonography were classified by two readers independently as no tear, full-thickness tear, or partial-thickness tear. When present, the tear size and location were recorded. Inter- and intra-rater reliability for the two data sets were calculated. There was substantial agreement between the two readers in the classification of supraspinatus tendon tears on 2D images (κ = 0.79) but only moderate agreement on 3D images (κ = 0.48). There was moderate to substantial agreement in the classification of tears between the 2D and 3D images for both readers (reader 1, κ = 0.64; reader 2, κ= 0.54). The most common cause of a discordant result was the interpretation of a small hypoechoic region at the footprint as a partial tear on 3D images, compared to a normal appearance on 2D images. Use of a dedicated 3D ultrasound transducer has poorer inter- and intra-rater reliability for the assessment of supraspinatus tendon tears compared to 2D sonography because of the interpretation of small partial-thickness tears at the footprint on 3D images, a common region affected by anisotropy.