Abstract 16281: Right Ventricular Systolic Strain Normal Ranges and Comparisons by Contemporary Vendor-Neutral Echocardiography Software: A Prospective Healthy Cohort Study
Tom Kai Ming Wang, Ankit Agrawal, Tiffany Dong, Aro Daniela Arockiam, Joseph El Dahdah, Muhammad Majid, Elio Haroun, L L Rodriguez, Richard A Grimm, Brian P Griffin, Zoran B Popovic- Physiology (medical)
- Cardiology and Cardiovascular Medicine
Background: Right ventricle longitudinal strain (global: RVGLS and free-wall: RVFWLS) by speckle-tracking echocardiography is increasingly utilized to assess right ventricle systolic function with prognostic value demonstrated in several cardiovascular conditions. Contemporary strain software have addressed past limitations to be able to analyze scans from different vendor scanners, however few studies have validated them externally. We prospectively evaluated the reference ranges and correlations of RVGLS and RVFWLS by the latest strain software in healthy patients.
Methods: We prospectively studied 100 healthy subjects undergoing echocardiography during January-April 2023 at our institution, 20 each in age groups 18-29, 30-39, 40-49, 50-59, 60+ years; half were men and half women; and half with GE and half with Philips scans. RVGLS and RVFWLS were quantified using TomTec version 51.02 (Autostrain RV), EchoPAC version 206 (AFI-RV) and VVI software for comparative analyses
Results: RVGLS and RVFWLS means (lower limits of normal LLN defined as 95 th percentile)±standard errors were -20.6%±0.3% (-16.9±0.5%) and -24.1±0.4% (-18.6±0.6%) by TomTec; -17.1±0.3% (-14.0±0.5%) and -19.6±0.4% (-15.8±0.7%) by EchoPAC; and -17.3±0.3% (-12.5±0.6%) and -19.2±0.4% (-13.6±0.8%) by VVI respectively, with significant differences between mean RVGLS and RVFWLS of TomTec and EchoPAC or TomTEC and VVI only (P<0.001). Factors with beta-coefficients and 95%CI significantly associated with RVGLS in multivariable linear regression analyses were male sex 1.35 (0.18-2.53) for TomTec; age (per 1 year) 0.06 (0.01-0.10), Philips versus GE scan 1.38 (0.17-2.58) for EchoPAC and 1.54 (0.31-2.78) for VVI. Factors with beta-coefficients and 95%CI significantly associated with RVFWLS in multivariable linear regression analyses were Philips versus GE scan 2.11 (0.67-3.56) for EchoPAC and 2.07 (0.38-3.76) for VVI, and none for TomTec.
Conclusion: RVGLS and RVFWLS measurements were feasible for all strain software on both GE and Philips scans, with means and LLN identified. Factors associated with RV strain values varied by strain software, in particular, Philips scans had less negative RVGLS and RVFWLS than GE scans when measured using EchoPAC and VVI but not TomTec.