Assessment of right ventricular volumes and function using cardiovascular magnetic resonance cine imaging after atrial redirection surgery for complete transposition of the great arteries

Abstract

Cardiovascular magnetic resonance (CMR) imaging is the reference standard for measurement of right ventricular (RV) volumes and function. To date, no study has compared methods of data acquisition and analysis by CMR for adults with a systemic RV. Our objective was to evaluate RV size and function using axial and short axis views in adults post atrial switch (Mustard) surgery. A total of 34 adults (20 male, mean age at CMR 32 ± 6 years) were identified at our centre. Volumes, RV end-diastolic (EDV) and end-systolic (ESV) were measured in short axis and axial orientations by two independent experienced readers, blinded to clinical and CMR data. Intra and interobserver measurements in each view were compared using Bland-Altman plots and intraclass correlation coefficients (ICC). Although mean volumes were larger in the axial as compared with the short axis view [RVEDV 247 ± 67 vs. 233 ± 54 ml (p = 0.002) and RVESV 148 ± 54 vs. 136 ± 50 ml (p = 0.001)], mean RV ejection fractions (EF) were similar [41 ± 9 % vs. 43 ± 12 % (p = 0.13)]. Bland-Altman plots demonstrated better agreement for axial measures of RVEDV and right ventricular ejection fraction (RVEF) within and between observers. Similarly, ICC values were stronger for axial as compared with short axis volumes and function-intraobserver RVEDV 0.99 (0.98-0.99) versus 0.96 (0.92-0.98) and RVEF 0.96 (0.93-0.98) versus 0.90 (0.82-0.95); interobserver RVEDV 0.97 (0.94-0.98) versus 0.90 (0.73-0.95) and RVEF 0.85 (0.53-0.94) versus 0.82 (0.67-0.90). Axially derived measurements of RV volumes and function have better agreement and reproducibility as compared with short axis values; whereas axial volumes tend to be larger, RVEF is not significantly different between the two methods.