Abstract 2436: Tricuspid E/Ea is a Powerful Predictor of Cardiac Events in Patients with Chronic Pulmonary Hypertension

Abstract

Background: Right ventricular (RV) filling pressure can be estimated using tissue Doppler imaging (TDI) from the tricuspid lateral annulus. There are sparse data on the significance of estimated RV filling pressure in predicting the prognosis of chronic pulmonary hypertension (PH). Methods: In 50 consecutive patients with chronic PH undergoing right heart catheterization, early diastolic myocardial velocity (E a ) at the tricuspid lateral annulus and early diastolic tricuspid inflow (E) were obtained from 4-chamber view using TDI. The tricuspid E/E a ratio was calculated and compared with the invasive hemodynamic variables including pulmonary arterial systolic pressure (PASP), mean right atrial pressure (RAP) and RV end-diastolic pressure (RVEDP). Cardiac events including cardiac death or rehospitalization due to RV failure were assessed at long-term follow-up (mean 14.0±1.4 months). Results: The tricuspid E/E a ratio was significantly positively correlated with RAP (r=0.80, p<0.001) and RVEDP (r=0.72, p<0.001), irrespective of RV systolic function (depressed RV systolic function, n=32, r=0.81; preserved RV systolic function, n=18, r=0.75; both p<0.001). An E/Ea ratio > 7.3 had a sensitivity of 87% and a specificity of 97% for mean RAP ≥ 10mmHg. We divided patients into group A with cardiac events (n=19) and group B without events (n=31). There was no significant difference in PASP between the groups. However, serum brain natriuretic peptide and E/Ea ratio were significantly higher in group A than in group B (349±310 vs 129±136 pg/dl, p=0.001; 7.0±3.2 vs 4.5±.9, p=0.004, respectively). Multivariate Cox’s analysis using TDI and conventional echocardiographic variables showed that the tricuspid E/E a ratio (relative risk 1.226, 95%CI 1.011–1.486, p = 0.039) and RA area index (relative risk 1.078, 95%CI 1.002–1.159, p = 0.043) were predictive for cardiac events. Kaplan-Meier analysis showed patients with E/E a > 7.3 had a significantly worse outcome than those with E/E a ≤ 7.3 (log-lank: 10.5, p = 0.0012). Conclusion: RV filling pressure can be well estimated using the tricuspid E/Ea ratio. This index and RA size are powerful predictors of cardiac events in patients with chronic PH, suggesting the prognostic importance of RV diastolic dysfunction.