The Human Factor: Physicians' Unbiased Interpretation of Atrial Fibrillation in the Emergency Department.

BACKGROUND

Misinterpretation of atrial fibrillation (AF) poses significant clinical risks. While the impact of computerized ECG interpretations (CIE) on clinical decision-making is well-documented, there is limited real-world evidence on the diagnostic accuracy of emergency department (ED) clinicians in the absence of such tools. This study examines the frequency and clinical impact of AF misdiagnosis in an ED operating without CIE.

METHODS

ECGs of consecutive patients discharged with a diagnosis of AF over a five-year period were reevaluated by two cardiologists, with an electrophysiologist serving as an arbitrator. Misdiagnosed cases underwent further clinical follow-up.

RESULTS

Of 1828 ECGs initially identified as AF, 31 were excluded due to paced rhythms or indeterminate tracings, yielding a final cohort of 1797. Among these, 49 cases (2.7%) were misclassified, encompassing supraventricular tachycardia (n = 30), sinus rhythm with premature atrial contractions, Mobitz type I block, or artifact (n = 18), and multifocal atrial tachycardia (n = 1). Patients with erroneous AF diagnoses were older (mean age 76.7 vs. 72.5 years, p = 0.032) and frequently received inappropriate therapies, including rate-control agents (n = 12), antiarrhythmics (n = 8), and anticoagulants (n = 24).

CONCLUSIONS

AF misdiagnosis remains prevalent in the ED and can lead to unnecessary pharmacotherapy. The misdiagnosis rate appears comparable between CIE-assisted and physician-only interpretations.