Inter- and intraobserver reproducibility of Strain and 2D Shear Wave Elastography - a phantom study.

Abstract

To analyse the intra- and interobserver variability of two elastographic methods of quantification available on thesame machine, the technical factors that may influence variability as well as the intra- and interobserver variability for thesame indices between two different ultrasound machines in an in vitro experimental setting. Threedifferent types of silicone experimental devices (ED) were conceived for the purpose of this study. Two observers performedrepeated measurements on two ultrasound machines. Strain elastography, with strain ratio determination between the ED wasperformed on both machines. Shear wave ratio was also assessed. The data obtained were used to calculate intra- and interobservervariability. Reproducibility was assessed in relation to the size of the elastographic region of interest (ROI) and tothe difference in stiffness between the ED, through the value of the ICCs (Intraclass Correlation Coefficient). Strainratio had high inter- and intraobserver reproducibility, regardless of the machine used, on a large number of determinations.The choice of a small ROI diameter (5 mm) over a large ROI diameter (15mm) increased reproducibility (ICC = 0.87 vs 0.78,p=0.000). It is observed that, by Shear Wave Elastography, only when analysing structures with a large difference in hardness,significance is obtained in terms of interobserver reproducibility (ICC = 0.75, p=0.000). On a large number ofdeterminations, both techniques are inter- and intraobserver reproducible. It is preferable to opt for a smaller ROI diameter inorder to increase interobserver reproducibility. SWE Ratio provides significant reproducibility only when analysing structureswith large difference in hardness.