Reproducibility of Cardiac Magnetic Resonance Imaging (CMRI)-Derived Right Ventricular Parameters in Repaired Tetralogy of Fallot (ToF).

BACKGROUND

Quantification of right ventricular (RV) volumes is challenging owing to variable reproducibility and is especially so in congenital heart disease. Cardiac magnetic resonance (CMR) has the ability to more comprehensively survey the entire right ventricle and is currently considered the gold standard.

AIMS

We aimed to determine the inter-observer reproducibility of CMR-derived RV volumes generated by two independent and experienced (SCMR Level III) observers in Tetralogy of Fallot (ToF) patients with varying degrees of RV dilatation.

METHODS

We performed a retrospective analysis of 120 consecutive patients with repaired ToF who underwent CMR. Two blinded observers calculated RV volumes in each oblique short axis slice independently. Bland-Altman analysis and inter-observer correlation coefficients (ICC) were assessed.

RESULTS

The coefficients of variation for RV parameters were: 2.9%, 8% and 3.4% for right ventricular end diastolic volume (RVEDV), right ventricular end systolic volume (RVESV) and right ventricular ejection fraction (RVEF) respectively. For RVEDV the interobserver correlation was 0.992 demonstrating excellent volumetric correlation between observers. The mean difference between the observers for right ventricular end diastolic volume index (RVEDVi) was 2.5ml/m (95% limits of agreement -7.3 to 12.2ml/m). For patients with mild-moderate RV dilatation (RVEDVi <150ml/m) the mean difference of RVEDVi was 1.8ml/m (95% limits of agreement -5.7 to 9.3ml/m). For patients with severe RV dilatation (RVEDVi≥150ml/m) the mean difference was -3.4ml/m (95% limits of agreement -8.6 to 15.4ml/m).

CONCLUSIONS

In patients with repaired ToF and variable degrees of RV dilatation, CMR assessment of RV volumes and function has high inter-observer reproducibility. This allows for optimal timing of pulmonary valve replacement, based on progression of RV dilatation over time.