Abstract 11648: Mitigation of Variability of 3D Echo-Derived Regional Strain Acquired by Multiple Ultrasound Systems Using Vendor Independent Analysis
Introduction: Three-Dimensional Echocardiography (3DE) based quantitative evaluation of left ventricular (LV) cardiac mechanics has been shown to lack reproducibility among ultrasound systems.
Hypothesis: 3DE-derived longitudinal strain values acquired from, and analyzed by, different vendors will demonstrate variability which is reduced by using a vendor-independent analysis package.
Methods: Fresh porcine heart phantoms were passively driven with a rotation of 5° and compaction of 10mm to simulate normal cardiac function. Stroke volume ranged from 30-70ml. Full-volume image data were acquired using three different ultrasound systems. The data were analyzed by three separate observers on vendor specific software prior to analysis with a vendor independent program.
Results: Good linear relationships were observed for each vendor specific analysis package for strain with Pearson Correlation Coefficients (r) of 0.86-0.89. Comparable linear regressions were observed for results determined by a vendor independent program (r=0.86-0.89). The normalized mean strain plots illustrate the significant variability in the data generated by vendor specific software as opposed to the values derived by the vendor independent analysis program. ANOVAs revealed p<0.001 for vendor specific analysis packages and p>0.05 for the vendor independent program.
Conclusions: Longitudinal regional strain values differ when quantified by vendor specific analysis packages; however, this variability is mitigated by use of a vendor independent quantification method.
Author Disclosures: C. Streiff: None. M. Zhu: None. E. Shimada: None. M. Ashraf: None. D.J. Sahn: None.
- © 2014 by American Heart Association, Inc.