Objectives

Cardiovascular magnetic resonance (CMR) imaging is routinely performed for assessing right ventricular (RV) systolic but not diastolic function. We aimed to investigate CMR-based assessment of RV diastolic function in pediatric patients with repaired tetralogy of Fallot (rTOF), compared to transthoracic echocardiography (TTE) measurements.

Methods

A total of 130 consecutive pediatric patients with rTOF who underwent clinically indicated CMR and same-day TTE were included. Forty-three controls were recruited. Phase-contrast images were used to measure trans-tricuspid valve flow velocities during early (E) and late diastolic (A) phases (measured in cm/s). Feature tracking of the tricuspid annulus was performed to derive early (e′) and late diastolic (a′) myocardial velocities (measured in cm/s). RV diastolic function was evaluated by E/A ratio, E/e′ ratio, and E-wave deceleration time (measured in milliseconds). Regression analyses were utilized to identify potential variables associated with RV diastolic dysfunction (DD). The performance of CMR-derived parameters in diagnosing RV DD was assessed using receiver-operating characteristic analyses.

Results

Good agreement was found between CMR and TTE measurements (ICC 0.70–0.89). Patients with RV DD (n= 67) showed significantly different CMR-derived parameters including E and e′ velocities, and E/A and E/e′ ratio, compared to patients without DD (n= 63)(all p< 0.05). CMR-derived E and e′ velocities and E/e′ ratio were independent predictors of RV DD. E/e′ of 5.8 demonstrated the highest discrimination of RV DD (AUC 0.76, sensitivity 70%, specificity 86%).

Conclusions

CMR-derived parameters showed good agreement with TTE parameters in determining RV DD. CMR-derived E/e′ was proved to be the most effective in identifying RV DD.