P127: Use of nt-probnp biomarker amongst cardiologists and emergency physicians to diagnose acute heart failure in the undifferentiated dyspneic emergency patient

Abstract

Introduction: Diagnosing the undifferentiated dyspneic emergency department (ED) patient remains a challenge for clinicians; in order to rule in or out acute heart failure (AHF) natriuretic peptide biomarker testing has evolved and is recommended by cardiology international guidelines to be utilized in these presentations. However there is equipoise in the emergency community for its use, largely due to perceived modest test specificity. We sought to analyze this apparent clinical dichotomy as part of a multicenter trial of undifferentiated dyspneic ED patients. Methods: Patients with dyspnea presenting between October 2010 and October 2013 to one of four ED sites -2 Canadian, 1 American, 1 New Zealand- were assessed by certified staff emergency physicians (EPs) and their chest Xray reviewed. Those patients with undifferentiated dyspnea with a potential for AHF (ie further investigated or treated for AHF but investigated and/or treated for another cause) were consented and enrolled. Two of the sites (American, New Zealand) had NT-proBNP assay ordered as a standard of care for these patients; the other 2 sites did not. At the end of Emergency care, the EP recorded the primary diagnosis of the dypnea-either “AHF” or “Not AHF.” Blinded adjudication was carried out by 2 cardiologists after reviewing sequential records: first, with index ED records but no NT-proBNP result; second, with the NT-proBNP result and lastly, with follow up 60 day records (deemed the gold standard diagnosis). EP accuracy between NT-proBNP and no NT-proBNP sites and NT-proBNP accuracy using standard cutpoints were calculated, as were the number of adjudicated cases influenced by exposure to NT-proBNP. Results: 197 patients were enrolled, 107 at NTproBNP sites and 90 at the other 2 sites. EP accuracy was 76% for either site. NT-proBNP used as a binary test with recommended age-stratified cutoffs had 80% accuracy, applied to 70% of patients (30% remained in “gray zone”).Cardiology adjudicators reversed 16% of initial diagnoses upon exposure to NT-proBNP result, ultimately diagnosing 41% of patients with AHF. Conclusion: This study supports the clinical equipoise amongst emergency physicians compared to cardiologists for the use of NT-proBNP in diagnosing acute heart failure in the undifferentiated dyspneic Emergency patient.