Abstract 264: Accuracy and Indication of Emergency Department Echocardiography, What Needs to Be Improved? A Single Center Experience

Abstract

Background: Focused echocardiographic (FE) assessment of left and right ventricular (LV and RV) size and function, pericardial and aortic pathology and inferior vena caval (IVC) size and respiratory variation) has become a valuable part of the assessment of patients (pts) in the emergency department (ED). However, there are no head to head comparisons on the accuracy of FE as compared with a full formal study in the echo department. We compare the accuracy of these two echo modalities in a busy university-affiliated community hospital. Methods: The electronic medical record (EMR) was used to screen patients admitted between 9/1/2017 to 12/31/2017. Only patients with FE in ED, as well as a formal study on the same admission were included. A board-certified cardiologist finalized the findings of departmental echo. A total of 1913 pts were reviewed with 222 patients having both FE and formal studies. There was no report documented for FE in 161 pts, leaving 61 patients with both studies available to analyze. Result: The main indications for FE were shortness of breath (46%), tachycardia (16%), and chest pain (15%), while syncope and hypotension comprised 4% each. As compared to a formal echo: FE accurately estimated IVC dimension in 88%. Evaluation of LV ejection fraction (EF) (normal vs abnormal) was concordant in 85% of pts with consistent underestimation of EF by >10% in the other 15% of pts by FE. Concordance in the assessment of pericardial effusion was 90% with no significant effusions missed by FE. FE assessment of the RV missed 3 cases of significant RV enlargement, 2 of which were associated with acute pulmonary emboli. Finally, the FE reported aortic root size as indeterminate in 25% of cases and missed 4 cases of enlarged aortic roots. Of particular concern was lack of documentation of FE findings in the EMR in 73%. Conclusion: In general, FE is a valuable tool which provides accurate assessment of volume status (IVC dimension) and for the most part accurately assessed EF and the presence of a pericardial effusion. However, FE may significantly underestimate EF in 15% of patients. Both LV and RV function and size may be difficult to assess based on time constraints, inadequate echo windows or technical issues with imaging. FE was unable to assess the aortic root accurately in 25% of cases. Despite the limitations of FE (acuity of medical conditions, the relative brevity of the study and modest training in echocardiography undertaken by most ED providers), FE remains a valuable tool which can be improved upon to enhance diagnostic accuracy. To improve ED FE utility and accuracy we suggest education surrounding appropriate use criteria, improved report documentation in the EMR, rotation of ED residents through the echo department for enhanced training, and rapid consultation with a dedicated echo technician in ED to aid with difficult studies. A prospective collaborative quality improvement initiative is planned.