Objectives: To establish the accuracy and reliability of cardiac dual-source CT (DSCT) and two-dimensional contrast-enhanced echocardiography (CE-Echo) in estimating left ventricular (LV) parameters with respect to cardiac magnetic resonance imaging (CMR) as the reference standard.
Methods: Twenty-five consecutive heart transplant recipients (20 male, mean age 62.7±10.4 years, mean time since transplantation 8.1±5.9 years) were prospectively recruited. Two blinded readers independently assessed LV ejection fraction (EF), end-diastolic volume (EDV), end-systolic volume (ESV), and stroke volume (SV) for each patient after manual tracing of the endo- and epicardial contours in DSCT, CE-Echo and CMR cine images. Student's t-test for paired samples for differences, and Bland and Altman plots and Lin's concordance-correlation coefficients (CCC) for agreement were calculated.
Results: There was no statistical difference between left ventricular parameters determined by DSCT and CMR. CE-Echo resulted in significant underestimation of left ventricular volumes (mean difference EDV: 15.94±14.19 ml and 17.1±17.06 ml, ESV: 8.5±9.3 and 7.32±9.14 ml with respect to DSCT and CMR), and overestimation of EF compared with the cross-sectional imaging modalities (3.78±8.47% and 2.14±8.35% with respect to DSCT and CMR). Concordance correlation coefficients for LV parameters using DSCT and CMR were higher (CCC≥0.75) than CCC values observed between CE-Echo and DSCT- or CMR-derived data (CCC≥0.54 and CCC≥0.49, respectively). Interobserver agreement was higher for DSCT and CMR values (CCC≥0.72 and CCC≥0.87, respectively).
Conclusion: In orthotopic heart transplantation cardiac DSCT allows accurate and reliable estimation of LV parameters compared with CMR, whereas CE-Echo seems to be insufficient to obtain precise measurements.
Copyright © 2012 Elsevier Ireland Ltd. All rights reserved.