Semi-automated speckle-tracking for quantitative right ventricular assessment in children and adolescents

Background: Assessment of right ventricular size and function is an important part of the clinical cardiac evaluation; however, these quantitative measures are challenging by echocardiography. Automated software could be useful in place of manual measurements and qualitative assessment. This study evaluates a semi-automated software by comparing automated to manual measures in normal children. Methods: Neonates to adolescents with normal echocardiograms were prospectively enrolled. Measurements were performed using manual techniques and semi-automated software (EchoInsight (R), Epsilon Imaging, Ann Arbor, Michigan, United States of America). Right ventricular measurements included end-diastolic and end-systolic area, fractional area change, chamber dimensions, and tricuspid annular plane systolic excursion. Agreement between manual and semi-automated measures was compared. Results: Echocardiograms for 233 patients were included in the analysis. Intra- and inter-observer reliabilities for semi-automated measures were good with intraclass correlation coefficients all over 0.9 and 0.85, respectively. There was very strong correlation between manual and semi-automated methods for areas and dimensions (r = 0.93-0.99) and low bias (1.4-10.8%). For functional measures, tricuspid annular plane systolic excursion measures correlated well (r = 0.84), but fractional area change did not (r = 0.50). Both demonstrated significant bias (33.5-43.0%). The semi-automated method consistently underestimated fractional area change with a mean of 26.6% versus a manual mean of 36.1%. Conclusions: The semi-automated software is capable of generating quantitative right ventricular measures in children with good reliability. The software demonstrates very good correlation and low bias when compared to manual methods for right ventricular areas and dimensions. There is a significant difference between manual and semi-automated techniques for the functional measures.