Estimation of increased pulmonary wedge pressure by an algorithm based on noninvasively measured pulmonary diastolic pressure in cardiac patients independent of left ventricular ejection fraction

AIM: Pulmonary artery diastolic pressure (PADP) correlates closely with pulmonary wedge pressure (PAWP); therefore, we sought to evaluate whether an algorithm based on PADP assessment by the Doppler pulmonary regurgitation (PR) end-diastolic gradient (PRG) may aid in estimating increased PAWP in cardiac patients with reduced or preserved left ventricular (LV) ejection fraction (EF). METHODS AND RESULTS: Right heart catheterization, with estimation of PAWP, right atrial pressure (RAP), PADP, and Doppler echocardiography, was carried out in 183 patients with coronary artery disease (n = 63), dilated cardiomyopathy (n = 52), or aortic stenosis (n = 68). One-hundred and seventeen patients had LV EF <50%. We measured the pressure gradients across the tricuspid and pulmonary valves from tricuspid regurgitation (TRV) and PR velocities. Doppler-estimated PADP (e-PADP) was obtained by adding the estimated RAP to PRG. An algorithm based on e-PADP to predict PAWP, that included TRV, left atrial volume index, and mitral E/A, was developed and validated in derivation (n = 90) and validation (n = 93) subgroups. Both invasive PADP (r = .92, P < .001) and e-PADP (r = .72, P < .001) correlated closely with PAWP, and e-PADP predicted PAWP (AUC: 0.85, CI: 0.79-0.91) with a 94% positive predictive value (PPV) and a 55% negative predictive value (NPV), after exclusion of five patients with precapillary pulmonary hypertension. The e-PADP-based algorithm predicted PAWP with higher accuracy (PPV = 94%; NPV = 67%; accuracy = 85%; kappa: 0.65, P < .001) than the ASE-EACVI 2016 recommendations (PPV = 97%; NPV = 47%; accuracy = 68% undetermined = 18.9%; kappa: 0.15, P < .001). CONCLUSIONS: An algorithm based on noninvasively e-PADP can accurately predict increased PAWP in patients with cardiac disease and reduced or preserved LV EF.