Use of B-type natriuretic peptides to detect the existence and severity of diastolic dysfunction in non-cardiac critically ill patients: a pilot study

Abstract

Background: This observational study was designed to evaluate and compare the usefulness of BNP and NT-proBNP concentrations to detect the existence and severity of left ventricular diastolic dysfunction (LVDD) in critically ill patients.

Methods: A prospective, observational study in a university hospital. The sample included 86 consecutive adult patients. Based on echocardiography data, LVDD were classified into normal, impaired relaxation, pseudonormal or restrictive patterns. Patients were classified according to whether filling pressures were elevated or non-elevated in the echocardiography. Sampling for natriuretic peptides was performed immediately before echocardiography.

Results: Fifty patients showed LVDD. The most frequently observed pattern was impaired relaxation (N.=35), followed by the restrictive (N.=9) and the pseudonormal (N.=6) patterns. BNP concentrations in restrictive and pseudonormal patterns were higher than in normal and impaired relaxation patterns, while NT-proBNP only showed differences between normal and pseudonormal or restrictive patterns. Cut-off values using ROC curve analyses to detect LVDD were 125 ng/L for BNP and 390 ng/L NT-proBNP. BNP and NT-proBNP concentrations were higher in the 15 patients with restrictive and pseudonormal patterns, suggesting elevated filling pressures. Cut-off values using ROC curve analyses to detect echocardiography signs of elevated filling pressures were 254 ng/L for BNP and 968 for NT-proBNP. Both natriopeptides performed in a similar way to detect LVDD and elevated filling pressures.

Conclusion: Both BNP and NT-proBNP are useful screening tools to detect the presence of advanced degrees of LVDD, and especially to rule out elevated filling pressures.