Left Atrial Expansion Index for Noninvasive Estimation of Pulmonary Capillary Wedge Pressure: A Cardiac Catheterization Validation Study

Background: Pulmonary capillary wedge pressure (PCWP) plays a pivotal role in cardiac disease diagnosis and management. Right heart catheterization (RHC) invasively provides accurate PCWP measurement, but it is impractical for widespread use in all patients. The left atrial expansion index (LAEI), measured on transthoracic echocardiography, describes the relative left atrial volume increase during the left atrial reservoir phase. The aim of this study was to validate LAEI as a noninvasive parameter for PCWP estimation. Methods: A total of 649 chronic cardiac patients (mean age, 66 ± 14 years; mean PCWP, 14 ± 7.6 mm Hg; mean left ventricular ejection fraction, 50 ± 15%) who underwent both clinically indicated RHC and transthoracic echocardiography within 24 hours were retrospectively enrolled. Patients were randomly divided into derivation (n = 509) and validation (n = 140) cohorts. PCWP was measured during RHC and defined as elevated when >12 mm Hg. Transthoracic echocardiographic parameters and LAEI were measured offline, blinded to RHC results. Results: In the derivation cohort, LAEI correlated logarithmically with PCWP, and the log-transformed LAEI (lnLAEI) correlated linearly with PCWP (r = −0.73, P < .001). lnLAEI showed an independent and additive predictive role for PCWP estimation over clinical and diastolic dysfunction (DD) parameters. The diagnostic accuracy of lnLAEI for elevated PCWP identification (area under the curve = 0.875, P < .001; optimal lnLAEI cutoff < 4.02) was higher than either the single DD parameters or their combination. In the validation cohort, lnLAEI cutoff < 4.02 showed higher accuracy than the 2016 DD algorithm (88% vs 74%) for elevated PCWP identification. Finally, the equation PCWP = 38.3 − 6.2 × lnLAEI, obtained from the derivation cohort, predicted invasively measured PCWP in the validation cohort. Conclusions: In a cohort of patients with various chronic cardiac diseases, lnLAEI performed better than DD parameters