Fetal Echocardiographic Z Score Pilot Project: Study Design and Impact of Gestational Age and Variable Type on Reproducibility of Measurements Within and Across Investigators

• DOI: 10.1016/j.echo.2023.05.010
• Publication Date: 2023-06-10T15:07:27Z
• Authors: Anita J. Moon-Grady; Hyejung Lee; Leo Lopez; Oluwatosin Fatusin; Lindsay R. Freud; Whitnee Hogan; Anita Krishnan; Carol McFarland; L. LuAnn Minich; Shaine A. Morris; Nelangi Pinto; Angela P. Presson; Theresa A. Tacy; Mary T. Donofrio
• Keywords: Concordance correlation coefficient; Repeatability; Interclass correlation; Ductus arteriosus

Fetal echocardiography is widely available, but normative data are not robust. In this pilot study, the authors evaluated (1) feasibility of prespecified measurements in a normal fetal echocardiogram to inform study design and (2) measurement variability assign thresholds clinical significance guide analyses larger echocardiographic Z score initiatives.Images from predefined gestational age groups (16-20, >20-24, >24-28, >28-32 weeks) were retrospectively analyzed. expert raters attended online group training then independently analyzed 73 studies (18 per group) fully crossed 53 variables; each observer repeated measures for 12 fetuses. Kruskal-Wallis tests used compare across centers groups. Coefficients variation (CoVs) calculated at subject level as ratio SD mean. Intraclass correlation coefficients show inter- intrarater reliabilities. Cohen's d > 0.8 was define clinically important differences. Measurements plotted against age, biparietal diameter, femur length.Expert completed set mean 23 ± 9 min/fetus. Missingness ranged 0% 29%. CoVs similar all variables (P < .05) except ductus arteriosus velocity left ventricular ejection time, which both higher older age. >15% right systolic diastolic widths despite fair good repeatability (intraclass coefficient 0.5); ductal velocities two-dimensional measures, short-axis dimensions, isovolumic times had high interobserver excellent intraobserver agreement 0.6). did improve when ratios (e.g., tricuspid/mitral annulus) instead linear measurements. Overall, 27 acceptable repeatability, while 14 excessive between readers agreement.There considerable quantification practice that may affect multicenter studies, be feasible standard normalization. As missingness substantial, prospective will needed. Data aid calculation sample sizes distinguishing significant statistically effects.