Non-invasive determination by cardiovascular magnetic resonance of right ventricular-vascular coupling in children and adolescents with pulmonary hypertension

Abstract

Background: Pediatric pulmonary hypertension (PH) remains a disease with high morbidity and mortality in children. Understanding ventricular-vascular coupling, a measure of how well matched the ventricular and vascular function are, may elucidate pathway leading to right heart failure. Ventricular vascular coupling ratio (VVCR), comprised of effective elastance (Ea, index of arterial load) and right ventricular maximal end-systolic elastance (Ees, index of contractility), is conventionally determined by catheterization. Here, we apply a non-invasive approach to determining VVCR in pediatric subjects with PH.

Methods: This retrospective study included PH subjects who had a cardiovascular magnetic resonance (CMR) study within 14 days of cardiac catheterization. PH was defined as mean pulmonary artery pressure (mPAP) ≥ 25 mmHg on prior or current catheterization. A non-invasive measure of VVCR was derived from CMR-only (VVCRm) and compared to VVCR estimated by catheterization-derived single beat estimation (VVCRs). Indexed pulmonary vascular resistance (PVRi) and pulmonary vascular reactivity were determined during the catheterization procedure. Pearson correlation coefficients were calculated between PVRi and VVCRm. Receiver operating characteristic (ROC) curve analysis determined the diagnostic value of VVCRm in predicting vascular reactivity.

Results: Seventeen subjects (3 months-23 years; mean 11.3 ± 7.4 years) were identified between January 2009-August 2013 for inclusion with equal gender distributions. Mean mPAP was 35 mmHg ± 15 and PVRi was 8.5 Woods unit x m2 ± 7.8. VVCRm (range 0.43-2.82) increased with increasing severity as defined by PVRi (p < 0.001), and was highly correlated with PVRi (r = 0.92, 95 % CI 0.79-0.97, p < 0.0001). Regression of VVCRm and PVRi demonstrated differing lines when separated by reactivity. VVCRm was significantly correlated with VVCRs (r = 0.79, CI 0.48-0.99, p <0.0001). ROC curve analysis showed high accuracy of VVCRm in determining vascular reactivity (VVCR = 0.85 had a sensitivity of 100 % and a specificity of 80 %) with an area under the curve of 0.89 (p = 0.008).

Conclusion: Measurement of VVCRm in pediatrics is feasible. Pulmonary vascular non-reactivity may be contribute to ventricular-vascular decoupling in severe PH. Therapeutic intervention to maintain a low vascular afterload in reactive patients may preserve right ventricular functional reserve and delay the onset of RV-PA decoupling. Use of VVCRm may have significant prognostic implication.

Fig. 1

This diagram shows the single beat method by depicting how E_es and E_a are determined using the maximal pressure (P_max) as the peak of the sinusoid of a ventricular pressure tracing, end-systolic point (P_es), and (volume at end-diastole) V_ed

Fig. 2

Regression of ventricular vascular coupling ratio derived by MRI (VVCRm) and pulmonary vascular resistance indexed by reactivity. The shaded areas represent the 95 % confidence interval for each regression line. The lines depict different trajectories based on reactivity, which approached, but not reach statistical significance (p > 0.05)

Fig. 3

Scatterplot showing a positive linear relationship between ventricular vascular coupling ratio obtained by MRI compared to that obtained by single beat method

Fig. 4

Receiver operating characteristic curve. Receiver operating characteristic curve demonstrates an optimal threshold ventricular-vascular coupling ratio (VVCR) of 0.85. Using this criterion, VVCR is associated with a sensitivity of 100 % and a specificity of 80 % in determining pulmonary reactivity