Right ventricular systolic dysfunction but not dilatation correlates with prognostically significant reductions in exercise capacity in repaired Tetralogy of Fallot

Aims The optimal timing for pulmonary valve replacement in asymptomatic patients with repaired Tetralogy of Fallot (rTOF) and pulmonary regurgitation remains uncertain but is often guided by increases in right ventricular (RV) end-diastolic volume. As cardiopulmonary exercise testing (CPET) performance is a strong prognostic indicator, we assessed which cardiovascular magnetic resonance (CMR) parameters correlate with reductions in exercise capacity to potentially improve identification of high-risk patients. Methods and results In all, 163 patients with rTOF (mean age 24.5 +/- 10.2 years) who had previously undergone CMR and standardized CPET protocols were included. The indexed right and left ventricular end-diastolic volumes (RVEDVi, LVEDVi), right and left ventricular ejection fractions (RVEF, LVEF), indexed RV stroke volume (RVSVi), and pulmonary regurgitant fraction (PRF) were quantified by CMR and correlated with CPET-determined peak oxygen consumption (VO2) or peak work. On univariable analysis, there was no significant correlation between RVEDVi and PRF with peak VO2 or peak work (% Jones-predicted). In contrast, RVEF and RVSVi had significant correlations with both peak VO2 and peak work that remained significant on multivariable analysis. For a previously established prognostic peak VO2 threshold of <27 mL/kg/min, receiver-operating characteristic curve analysis demonstrated a Harrell's c of 0.70 for RVEF (95% confidence interval 0.61-0.79) with a sensitivity of 88% for RVEF <40%. Conclusion In rTOF, CMR indices of RV systolic function are better predictors of CPET performance than RV size. An RVEF <40% may be useful to identify prognostically significant reductions in exercise capacity in patients with varying degrees of RV dilatation.