Diagnosing acute heart failure in the emergency department

Ferguson, Craig James (2014) Diagnosing acute heart failure in the emergency department. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (9MB) | Preview

Abstract

Background: Acute, decompensated heart failure is a serious and common presentation in patients attending Emergency Departments. Diagnosis of this condition in this environment can be challenging. Main Objective:  To assess the diagnostic utility of variables related to the diagnosis of decompensated heart failure in the Emergency Department setting.  To create a Clinical Decision Rule to facilitate the diagnosis of decompensated heart failure in the Emergency Department setting. Methods:  A systematic review and meta-analysis of literature related to the diagnosis of acute heart failure in the Emergency Department was performed to assess and rank the diagnostic utility of all potential predictor variables. Bivariate meta-analysis was performed where appropriate to provide summary statistics for variable utility.  A diagnostic study performed in a single, urban Emergency Department to allow multivariable analysis of the data to derive a Clinical Decision Rule. Logistic regression, Random Forest analysis and CART analysis were used in the analysis. Results: 44 papers were included in the systematic review providing data on 41 potential diagnostic variables. A history of heart failure, clinical opinion of heart failure, natriuretic peptide levels and the findings of cardiomegaly, pulmonary oedema or cephalisation of vessels on the chest x-ray had the greatest diagnostic utility.  105 patients were recruited in the diagnostic study and 62 potential variables were assessed against a reference standard of two cardiologists opinions. Cardiothoracic ratio, natriuretic peptide levels, the presence of pulmonary oedema, clinical impression of heart failure and urea levels were found to be the best performing variables.  Two or more of these variables, using selected cut-off values for continuous variables, had a specificity of nearly 100% for definite or possible heart failure. Different cut-off values could be used with these variables to provide high sensitivity for this condition. Conclusion: The clinical and investigative findings with the greatest diagnostic utility for the acutely dyspnoeic patient with heart failure have been identified and can be used to rule in heart failure.