Background: Decreased right ventricular (RV) ejection fraction (EF) increases morbidity in repaired tetralogy of Fallot (rToF). Cardiac magnetic resonance (CMR) is the accepted reference standard for RV EF measurement. There are no established color Doppler tissue imaging (cDTI) values that identify decreased RV EF in rToF. We sought to assess accuracy of cDTI myocardial deformation and velocity indices in predicting CMR-derived RV EF.
Methods: Retrospective evaluation of CMR and echocardiography studies performed on the same day. cDTI-derived peak systolic velocity (pkS), isovolumic acceleration (IVA) of the tricuspid valve annulus, strain (ε) and strain rate (SR) of the RV free wall, adjusted for age, volume and pressure overload were used to compute RV EF. The probability of correctly classifying CMR measured RV EF <45% was estimated by logistic regression analysis. The accuracy of the model was tested on 10 prospective patients.
Results: Sixty matched echocardiographic and CMR studies were analyzed. None of the individual cDTI parameters could reliably identify patients with CMR-derived decreased RV EF. However, when adjusted for age in a logistic model: Y = -0.5 - 0.07 × age + 0.016 × ε + 0.46 × pkS, where Y is the logit (log odds), patients with RV EF <45% were identified with 86% sensitivity, 93.5% specificity. Prospective application of this model correctly identified all subjects with RV EF <45%.
Conclusion: The age-adjusted logistic model and not individual parameters, derived from cDTI-derived pkS, and ε can reliably identify patients with decreased RV EF <45%.
© 2013, Wiley Periodicals, Inc.