Matt Olson, DO | EMIM PGY3
The Article:
The Idea:
To determine if using Lung ultrasound (LUS), in addition to clinical assessment, for patients with acute dyspnea can aid in the diagnosis of acute decompensated heart failure (ADHF).
The Study:
This was a randomized control trial that was conducted in two emergency departments from January 2014 to March 2015 on patients with acute dyspnea. 518 patients were enrolled in the study and randomized into one of two arms, the lung ultrasound arm and the CXR/Nt-proBNP arm. Every patient started with an initial clinical evaluation including history and physical exam. After the clinical exam the physician was asked to indicate a presumptive cause of the dyspnea. Afterwards the patients were separated into their arm (lung ultrasound vs. CXR/Nt-proBNP). Lung ultrasound was performed at bedside by the clinician with a curvilinear probe in eight zones looking for 3 or more B-lines to indicate a positive exam. After the lung ultrasound or CXR/Nt-proBNP a new presumptive diagnosis was recorded by the provider. Afterwards a CXR and Nt-proBNP were conducted on the lung ultrasound group. After hospital discharge two expert intensivists/emergency physicians (blinded to LUS results) reviewed the medical record and determined if the patient fit the diagnosis/definition of acute decompensated heart failure (based on European Society of Cardiology).
The Findings:
The study found that the accuracy of clinical evaluation alone in the identification of acute decompensated heart failure was not significantly different in the groups. It also showed that combining CXR/Nt-proBNP with clinical evaluation was not statistically significant for sensitivity or specificity however the combined group with lung ultrasound was statistically significant. The lung ultrasound group also was showed to have a higher net reclassification improvement verses CXR/Nt-proBNP (8.9% vs. 4.5% for ADHF and non ADHF respectively). The median time needed to formulate the diagnostic hypothesis (measured from when the first diagnostic hypothesis was recorded to when the integrated diagnosis taking test results into account) was 104.5min in the CXR/Nt-proBNP group vs. 5min in the lung ultrasound group. The lung ultrasound arm was shown to be more sensitive (84.7 vs. 81.0), more specific (91.0 vs. 88.8) and had a higher positive predictive value (89.7 vs. 81.8) in the clinical evaluation along with the integrated evaluation.
The Takeaway:
Overall this study demonstrated that in adult patients presenting to the emergency department with acute dyspnea, a diagnostic protocol based on integration of lung ultrasound and clinical assessment is more accurate than currently recommended CXR/NT-proBNP. Lung ultrasound can increase sensitivity and specificity of diagnosing acute decompensated heart failure. It can also improve time to diagnosis of acute decompensated heart failure when compared to only using CXR/NT-proBNP.