Diagnosing acute heart failure in patients with undifferentiated dyspnea: a lung and cardiac ultrasound (LuCUS) protocol.

Abstract

The primary goal of this study was to determine accuracy for diagnosing acutely decompensated heart failure (ADHF) in the undifferentiated dyspneic emergency department (ED) patient using a lung and cardiac ultrasound (LuCUS) protocol. Secondary objectives were to determine if US findings acutely change management and if findings are more accurate than clinical gestalt. This was a prospective, observational study of adult patients presenting to the ED with undifferentiated dyspnea. The intervention consisted of a 12-view LuCUS protocol performed by experienced emergency physician sonographers. The primary objective was measured by comparing US findings to the final diagnosis independently determined by two physicians blinded to the LuCUS result. Acute treatment changes based on US findings were tracked in real time through a standardized data collection form. Data on 99 patients were analyzed; ADHF was the final diagnosis in 36%. The LuCUS protocol had sensitivity of 83% (95% confidence interval [CI] = 67% to 93%), specificity of 83% (95% CI = 70% to 91%), positive likelihood ratio of 4.8 (95% CI = 2.7 to 8.3), and negative likelihood ratio of 0.20 (95% CI = 0.09 to 0.42). Forty-seven percent of patients had changes in acute management, and 42% had changes in acute treatment. Observed agreement for the LuCUS protocol was 93% between coinvestigators. Overall, accuracy improved by 20% (83% vs. 63%, 95% CI = 8% to 31% for the difference) over clinical gestalt alone. The LuCUS protocol may accurately identify ADHF and may improve acute clinical management in dyspneic ED patients. This protocol has improved diagnostic accuracy over clinical gestalt alone.