Use of NT-proBNP to predict postoperative evolution in cardiac surgery

Abstract

Introduction: the amino-terminal pro brain-type natriuretic peptide (NT-proBNP) is a diagnostic and prognostic biomarker in heart failure. Its use as a prognosis predictor of postoperative evolution in cardiac surgery has not been established.
Objective: to determine if the value of preoperativeNTin cardiac surgery is associated with postoperative evolution parameters.
Primary objective: to evaluate its association with the length of stay in intensive care unit. Secondary objectives: to evaluate its association with the time of mechanical ventilation and inotropic agents requirements. To determine the evolution of NT-proBNP concentration after cardiopulmonary bypass.
Methods: multicentric retrospective study, endorsed by the Research Committee of the National Institute of Cardiac Surgery and the Ethic Research Committee of the Clinic Hospital. It included patients who underwent cardiac surgery between March and August 2018. NT-proBNP was measured during anesthesia induction and after cardiopulmonary bypass. A possible association of preoperative NT-proBNP with risk factors and type of procedure performed was studied. By analysing the ROC curve, the area under curve (AUC) was calculated and then, the best cut-off value of NT-proBNP to predict prolonged intensive care unit stay was determined. Intensive care unit stays, mechanical ventilation and inotropic requirements were defined as prolonged when they exceeded 2 days, 6 and 24 hours respectively.
Through the use of multivaried logistics, the predicting value of NT-proBNP was determined for each one of the aforementioned variables. A value of ? 0.05 was considered significant.
Results: a total of 155 patients were included in the study. Age, creatininemia, and left ventricular ejection fraction were 65.8±11.4 years, 1.15±1.10 mg/dl and 52.8±11.9% respectively. Female prevalence was 30.3%, arterial hypertension 77.4%, diabetes mellitus 25.2% and dyslipidemia 50.3%. In 42.6% isolated myocardial revascularization was performed, in 12.9% myocardial revascularization plus one or more valve procedures, and in 44.5% isolated valve procedures.
In all cases there was a significant reduction between preoperative (443 pg/ml, interquartile range 143-1.193) and postoperative NT-proBNP (362 pg/ml, interquartile range 138-939) (p<0.001). Age, creatininemia, left ventricular ejection fraction, functional classification IV of the New York Heart Association and dyslipidemia turned out to be predictors of preoperative NT-proBNP. Preoperative NT-proBNP was higher in patients with prolonged intensive care unit stay, mechanical ventilation and inotropic requirements. However, it turned out to be an independent predictor only for prolonged intensive care unit stay. (OR=1.62; IC95%:1.11-2.35. p=0.012). The best cut-off value for prolonged intensive care unit stay was 409 pg/ml (AUC=0.68).
Conclusion: preoperative determination of NT-proBNP is an efficient tool to predict postoperative evolution. Cardiopulmonary bypass is associated to a significant drop in that marker.