Prognostic value of different echocardiographic indices reflecting right ventriculo-arterial coupling in a large cohort of patients with various cardiac diseases

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Introduction – Non-invasive parameters used to assess right ventricular (RV) function, i.e. tricuspid annular plane systolic excursion (TAPSE), RV fractional area change (FAC), RV ejection fraction (RVEF), and RV free-wall longitudinal strain (RVFWLS) have shown their prognostic implications. However, since they are extremely load dependent, they do not provide an accurate representation of the RV intrinsic performance. On the other end, invasive indices of RV-arterial coupling (RVAC) derived from pressure-volume loops are not routinely performed, rising the urgency for more feasible, and reliable non-invasive estimates of RVAC. Purpose – To: i. evaluate the prognostic value of echocardiography-derived RVAC surrogates: RVEF/systolic pulmonary artery pressure (sPAP), RVFWLS/sPAP, TAPSE/sPAP, FAC/sPAP, and RV stroke volume/end-systolic volume (SV/ESV), ii. identify the cut-off values associated to all-cause mortality; and iii. compare their prognostic value with that of classical parameters of RV function. Methods – We prospectively enrolled 366 patients with various cardiac diseases, undergoing clinically-indicated comprehensive two- and three-dimensional echocardiography. Results – During a mean follow-up of 7.6 ± 1 years, 80 (21.9%) patients died. At univariable Cox regression, most of the echocardiographic parameters were related to all-cause mortality. The echocardiographic parameters with significance at univariable analysis (p < 0.01) were included in a multivariable regression model. Left ventricular ejection fraction (LVEF), RVEF, TAPSE, RVEF/sPAP and RVFWLS/sPAP remained independently associated to all-cause mortality (p < 0.05 for all). Subsequently, they were tested in receiving operator characteristics (ROC) curves. At ROC analysis, RVEF/sPAP (area under the curve, AUC = 0.807, p < 0.001) and RVFWLS/sPAP (AUC = 0.743, p < 0.001) showed the greatest predictive value (p < 0.001 between them). However, all RV parameters significantly improved their prognostic values after indexing for sPAP (p < 0.01 for all). The best cut-offs to predict the outcome were 1.5 for RVEF/sPAP (specificity 71%, sensitivity 83%) and 0.67 for RVFWLS/sPAP (specificity 72%, sensitivity 68%). At Kaplan-Meier analysis, patients with reduced RVAC (less than the predefined cut-offs) had significantly lower probability of survival (p < 0.001 for all). Conclusion – RVAC surrogates provide incremental prognostic value compared to standard RV functional measurements. RVEF/sPAP, with a cut-off value of 1.5, was the best parameter for risk stratification, and was independently related to all-cause mortality. Abstract Figure. Prognostic value of RVAC surrogates Abstract Figure. Kaplan-Meier curves survival probability