Two-Dimensional Strain is more Precise than Conventional Measures of Left Ventricular Systolic Function in Pediatric Patients.

ABSTRACT

Ejection fraction (EF) and fractional shortening (FS) are standard methods of quantifying left ventricular (LV) systolic function. 2D global longitudinal strain (2D GLS) is a well-established, but underutilized method for LV function quantification. The aim of this study was to assess precision of GLS compared to EF & FS in pediatrics. Echocardiograms were prospectively analyzed by 2 blinded observers. FS, EF, and GLS were calculated following standard methods. Bland-Altman was applied to assess agreement. Intraclass correlation coefficient (ICC) was used to measure reliability. Coefficient of variation was used to demonstrate relative variability between methods. 103 pediatric echos were evaluated for inter-observer reproducibility, and 15 patients for intra-observer reproducibility. GLS had higher inter-observer agreement and reliability (bias 7%, 95% LOA - 3.4 to + 3.5, ICC 0.86 CI 0.80-0.90) compared to EF (bias 27%, 95% LOA - 18.9 to + 19.5; ICC 0.25 CI 0.07-0.43) and FS (bias 12%, 95% LOA - 11.9 to + 12.2; ICC 0.53 CI 0.38-0.66). GLS also had higher intra-observer agreement (bias 4%, 95% LOA - 3.6 to + 3.7; ICC 0.87 CI 0.66-0.96) compared to EF (bias 11%, 95% LOA - 14.9 to + 15.1; ICC 0.26 CI - 0.28-0.67) and FS (bias 12%, 95% LOA - 12.2 to + 12.5; ICC 0.38 CI - 0.15-0.74). GLS is a more precise method for quantifying LV function in pediatrics, with lower variability compared to EF and FS. GLS provides a more reliable evaluation of LV systolic function and should be utilized more widely in pediatrics.