Accuracy and diagnostic performance of doppler echocardiography to estimate mean pulmonary artery pressure in heart failure

Background: Multiple echocardiographic algorithms have been proposed to estimate mean pulmonary artery pressure (PAPM) and assess pulmonary hypertension (PH) likelihood. We assessed the accuracy of four echocardiographic approaches to estimate PAPM in heart failure (HF) patients undergoing near-simultaneous right heart catheterization (RHC), and compared diagnostic performance to identify PH with recommendation-advised tricuspid regurgitation peak velocity (TRVmax). Methods: We employed four validated echocardiographic algorithms incorporating tricuspid regurgitation peak or mean gradient, pulmonary regurgitation peak gradient, and right ventricular outflow tract acceleration time to estimate PAPM. Echocardiographic estimates of right atrial pressure were incorporated in all algorithms but one. Association and agreement with invasive PAPM were assessed. Diagnostic performance of all algorithms to identify PH was evaluated and compared with the recommended TRVmax cut-off. Results: In 112 HF patients, all echocardiographic algorithms demonstrated reasonable association (r =.41–.65; p < 0.001) and good agreement with invasive PAPM, with relatively lower mean bias and higher precision observed in algorithms that incorporated tricuspid regurgitation peak or mean gradient. All methods demonstrated strong ability to identify PH (AUC =.70–.80; p < 0.001) but did not outperform TRVmax (AUC =.84; p < 0.001). Echocardiographic estimates of right atrial pressure were falsely elevated in 30% of patients. Conclusions: Echocardiographic estimates demonstrate reasonable association with invasive PAPM and strong ability to identify PH in HF. However, none of the algorithms outperformed recommendation-advised TRVmax. The incremental value of echocardiographic estimates of right atrial pressure may need to be re-evaluated