Right and Left Ventricular Myocardial Perfusion Reserves Correlate with Right Ventricular Function and Pulmonary Hemodynamics in Patients with Pulmonary Arterial Hypertension

Purpose: To evaluate the relationships of right ventricular (RV) and left ventricular (LV) myocardial perfusion reserves with ventricular function and pulmonary hemodynamics in patients with pulmonary arterial hypertension (PAH) by using adenosine stress perfusion cardiac magnetic resonance (MR) imaging. Materials and Methods: This HIPAA-compliant study was institutional review board approved. Twenty-five patients known or suspected to have PAH underwent right heart catheterization and adenosine stress MR imaging on the same day. Sixteen matched healthy control subjects underwent cardiac MR imaging only. RV and LV perfusion values at rest and at adenosine-induced stress were calculated by using the Fermi function model. The MR imaging-derived RV and LV functional data were calculated by using dedicated software. Statistical testing included Kruskal-Wallis tests for continuous data, Spearman rank correlation tests, and multiple linear regression analyses. Results: Seventeen of the 25 patients had PAH: 11 with scleroderma-associated PAH, and six with idiopathic PAH. The remaining eight patients had scleroderma without PAH. The myocardial perfusion reserve indexes (MPRIs) in the PAH group (median RV MPRI, 1.7 [25th-75th percentile range, 1.3-2.0]; median LV MPRI, 1.8 [25th-75th percentile range, 1.6-2.1]) were significantly lower than those in the scleroderma non-PAH (median RV MPRI, 2.5 [25th-75th percentile range, 1.8-3.9] [P = .03]; median LV MPRI, 4.1 [25th-75th percentile range, 2.6-4.8] [P = .0003]) and control (median RV MPRI, 2.9 [25th-75th percentile range, 2.6-3.6] [P < .01]; median LV MPRI, 3.6 [25th-75th percentile range, 2.7-4.1] [P < .01]) groups. There were significant correlations between biventricular MPRI and both mean pulmonary arterial pressure (mPAP) (RV MPRI: rho = -0.59, Bonferroni P = .036; LV MPRI: rho = -0.79, Bonferroni P < .002) and RV stroke work index (RV MPRI: rho = -0.63, Bonferroni P = .01; LV MPRI: rho = -0.75, Bonferroni P < .002). In linear regression analysis, mPAP and RV ejection fraction were independent predictors of RV MPRI. mPAP was an independent predictor of LV MPRI. Conclusion: Biventricular vasoreactivity is significantly reduced with PAH and inversely correlated with RV workload and ejection fraction, suggesting that reduced myocardial perfusion reserve may contribute to RV dysfunction in patients with PAH. (C) RSNA, 2010