Abstract
Purpose
To assess reproducibility of 2D echocardiography parameters of right ventricular systolic function, such as tricuspid annular plane systolic excursion, right ventricular systolic wave velocity, fractional area change, and 3D right ventricular ejection fraction, through measurements performed by operators with different levels of experience, in patients with acute myocardial infarction.
Methods
Measurements were performed offline, independently, by three echocardiographers with different levels of experience: Reader 1 – advanced (5 years of training in 2DE, 3 years in 3DE); Reader 2 – intermediate (3 years of training in 2DE, 1 year in 3DE); Reader 3 – beginner (1 year of training in 2DE, 3 months in 3DE). Interobserver variability and agreement between readers were compared using Bland-Altman plots, as bias and limits of agreement (LOA), Pearson correlations and intraclass correlation coefficients.
Results
63 patients (52 males, 56.8±10.3 years) were analysed. All measurements showed excellent interobserver variability and agreement. Bias values were low, and LOA intervals were narrow, across all assessed parameters. Generally, bias values were lower between the advanced and intermediate readers with the exception of FAC: R1 vs R2 – bias 0.36, LOA −4.9;5.62 (r=0.96, p<0.001); R1 vs R3 – bias 0.09, LOA −6.4;6.6 (r=0.94, p<0.001); R2 vs R3 – bias −0.27, LOA −8.1;7.5 (r=0.91, p<0.001). Pearson correlation coefficients were excellent (>0.88) with significant p-values across all parameters (p<0.001). ICC were also excellent (>0.967)
Conclusion
2DE and 3DE parameters of right ventricular (RV) systolic function are highly reproducible, independent of operator experience, in patients presenting with acute myocardial infarction.