282 Advantage and Validation of Vendor-Independent Software for Global and Regional Strain Analysis

Abstract

One of the main limitations for incorporating strain imaging into widespread clinical practice is inter-vendor incompatibility. This study sought to compare the variability of two-dimensional speckle-tracking derived global and regional longitudinal strain using vendor-specific software (VSS) and vendor-independent software (VIS) from images acquired by two different commercially available high-end ultrasound systems. 40 subjects underwent two sequential echocardiographic acquisitions using two different ultrasound systems (GE Vivid E9 and Philips iE33). Global longitudinal strain (GLS) and regional peak longitudinal strain were derived using VSS (EchoPAC BT 13 and QLAB version 10.3) and VIS (TomTec Image Arena version 4.6). Agreement (Blan-Altman Bias) and reproducibility of strain values between VSS and VIS were assessed using intraclass correlation coefficient (ICC). GLS values analysed by VIS on images acquired on different ultrasound systems showed no significant difference (TomTec strain analysis on GE images; -18.6±4.8 vs TomTec strain analysis on Philips images; -19.2±4.8, p=0.09). VIS GLS was comparable to VSS GLS, whilst regional strain was lower in agreement compared to GLS. There was good overall agreement using VIS for GLS (Bias: Philips images =-0.02, Bias: GE images =-0.91) and high inter- and intra-observer reproducibility (Interobserver ICC =0.92, Intraobserver ICC =0.96). VIS provides good agreement with VSS for GLS. Variability exists for regional strain between VIS and VSS. Good agreement for VIS strain values from images acquired using different ultrasound systems suggest VIS could potentially be useful for serial follow-up of GLS.