# Echocardiography can accurately estimate pulmonary artery wedge pressure without left atrial volume information—diagnostic and prognostic performance

## TL;DR

This study shows that a new echocardiography method can estimate pulmonary artery wedge pressure as accurately as existing methods, without needing left atrial volume data.

## Contribution

A novel echocardiography-based method (ePAWP-NOLA) is introduced that estimates pulmonary artery wedge pressure without left atrial volume information.

## Key findings

- ePAWP-NOLA performed as well as ePAWP-LA in estimating pulmonary artery wedge pressure.
- ePAWP-NOLA had better diagnostic accuracy than the ASE/EACVI diastolic dysfunction algorithm.
- Both ePAWP-NOLA and ePAWP-LA were associated with adverse outcomes like death or device implantation.

## Abstract

A quantitative estimate of pulmonary artery wedge pressure (PAWP) can be obtained using echocardiography, but including left atrial (LA) volume (ePAWP-LA) in the estimation may be misleading. We aimed to derive and validate a new estimate without LA volume information (ePAWP-NOLA) and compare its performance to the ASE/EACVI algorithms for diastolic dysfunction.

ePAWP-NOLA was derived and validated in separate datasets of patients who had undergone right heart catheterization and echocardiography. The prognosis was assessed in the validation cohort and the National Echocardiography Database Australia (NEDA) using Cox regression adjusted for age, sex, and left ventricular ejection fraction (LVEF). In the derivation cohort (60 ± 15 years, 40% males, 31% with LVEF < 50%), ePAWP-NOLA was derived from mitral (E), and pulmonary vein systolic (S) and diastolic (D) Doppler velocities (n = 134, mean difference ± SD vs. PAWP: 0.0 ± 5.5 mmHg). In the validation cohort (n = 116, 51 ± 14 years, 69% males, 89% with LVEF < 50%), PAWP agreed with both ePAWP-NOLA and ePAWP-LA (difference 1.3 ± 6.1, 3.2 ± 6.3 mmHg, respectively). PAWP > 15 mmHg was accurately detected by both ePAWP-NOLA and ePAWP-LA [area under the curve: AUC (95%CI): 0.84 (0.76–0.92), 0.80 (0.72–0.88)]. AUC for the ASE/EACVI algorithm was lower [0.69 (0.61–0.77)]). ePAWP-NOLA and ePAWP-LA correlated with right ventricular afterload and were associated with death or implantation of left ventricular assist device, and with cardiovascular death in NEDA.

ePAWP-NOLA has diagnostic and prognostic performance comparable to ePAWP-LA, and improved diagnostic performance compared to the ASE/EACVI diastolic dysfunction algorithm.

Graphical Abstract