Background: Three-dimensional echocardiography (3DE) evaluation of left ventricular (LV) volume and function in pediatrics compares favorably with cardiac magnetic resonance imaging. The aim of this study was to establish from a multicenter, normal pediatric z-score values of 3DE left ventricular volumes and function. Methods: Six hundred and ninety-eight healthy children (ages 0 to 18 years) were recruited from five centers. LV 3DE was acquired from the 4-chamber view. A vendor independent software analyzed end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), and ejection fraction (EF) using semi-automated quantification. Body surface area (BSA) based z-scores were generated. Intraobserver and interobserver variability were calculated using intraclass correlation (ICC) and repeatability coefficient (RC). Results: Z-scores were generated for ESV, EDV, and SV. The ICC for intraobserver variability for EDV, ESV, and SV were 0.99, 0.99, and 0.99 respectively. The ICC for interobserver variability for EDV, ESV, and SV were 0.98, 0.94, and 0.98 respectively. The RC for intraobserver and interobserver variability for LV EF was 4.39% (95% CI: 3.01, 5.59) and interobserver was 7.08% (95%CI: 5.51, 8.42). Conclusions: We report pediatric Z-scores for normal LV volumes using the semi-automated method from five centers, enhancing its generalizability. 3DE evaluation of LV volumes and EF in pediatric patients is highly reproducible.

Background: Three-dimensional echocardiography (3DE) evaluation of left ventricular (LV) volume and function in pediatrics compares favorably with cardiac magnetic resonance imaging. A multicenter trial with automated and semi-automated LV quantification allows for generation of normative data in large pediatric patients. The aims of this study were to evaluate the feasibility and reproducibility of measuring three-dimensional echocardiography (3DE) volumes and function in pediatric patients in a multicenter trial; to determine if automated software (without contouring edits) will improve the reproducibility in volume and function analysis; and thus establish normal z score values in this unique population. Methods: Six hundred and ninety-eight healthy children (ages 0 to 18 years) were recruited from 5 centers. Left ventricular (LV) 3DE was acquired from the 4-chamber view. A vendor independent software analyzed end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), and ejection fraction (EF) using automated and semi-automated quantification. Feasibility and reproducibility were assessed. Body surface area (BSA) based z-scores were generated. Results: Feasibility was 79% (523/658). Reproducibility was good between centers using the semi-automated quantification. Reproducibility was decreased using the automated quantification. Therefore, Z-scores were generated for ESV, EDV, and SV using the semi-automated method. Conclusions: 3DE can reliably evaluate LV volumes and EF in pediatric patients at different centers. We report pediatric Z-scores for normal LV volumes using the semi-automated method. Further optimization of technology will be necessary for reliable use of fully automated quantification by 3DE in children.