Premium
Quality Control and Reproducibility in M‐Mode, Two‐Dimensional, and Speckle Tracking Echocardiography Acquisition and Analysis: The CARDIA Study, Year 25 Examination Experience
Author(s) -
Armstrong Anderson C.,
Ricketts Erin P.,
Cox Christopher,
Adler Paul,
Arynchyn Alexander,
Liu Kiang,
Stengel Ellen,
Sidney Stephen,
Lewis Cora E.,
Schreiner Pamela J.,
Shikany James M.,
Keck Kimberly,
Merlo Jamie,
Gidding Samuel S.,
Lima João A. C.
Publication year - 2015
Publication title -
echocardiography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.404
H-Index - 62
eISSN - 1540-8175
pISSN - 0742-2822
DOI - 10.1111/echo.12832
Subject(s) - reproducibility , intraclass correlation , medicine , speckle tracking echocardiography , ejection fraction , cardiology , ultrasound , nuclear medicine , radiology , mathematics , heart failure , statistics
Introduction Few large studies describe quality control procedures and reproducibility findings in cardiovascular ultrasound, particularly in novel techniques such as speckle tracking echocardiography (STE). We evaluate the echocardiography assessment performance in the Coronary Artery Risk Development in Young Adults (CARDIA) study Year 25 (Y25) examination (2010–2011) and report findings from a quality control and reproducibility program conducted to assess Field Center image acquisition and reading center (RC) accuracy. Methods The CARDIA Y25 examination had 3475 echocardiograms performed in 4 US Field Centers and analyzed in a RC, assessing standard echocardiography (LA dimension, aortic root, LV mass, LV end‐diastolic volume [LVEDV], ejection fraction [LVEF]), and STE (two‐ and four‐chamber longitudinal, circumferential, and radial strains). Reproducibility was assessed using intraclass correlation coefficients (ICC), coefficients of variation (CV), and Bland–Altman plots. Results For standard echocardiography reproducibility, LV mass and LVEDV consistently had CV above 10% and aortic root below 6%. Intra‐sonographer aortic root and LV mass had the most robust values of ICC in standard echocardiography. For STE, the number of properly tracking segments was above 80% in short‐axis and four‐chamber and 58% in two‐chamber views. Longitudinal strain parameters were the most robust and radial strain showed the highest variation. Comparing Field Centers with echocardiography RC STE readings, mean differences ranged from 0.4% to 4.1% and ICC from 0.37 to 0.66, with robust results for longitudinal strains. Conclusion Echocardiography image acquisition and reading processes in the CARDIA study were highly reproducible, including robust results for STE analysis. Consistent quality control may increase the reliability of echocardiography measurements in large cohort studies.