Decision aid for early identification of acute underlying illness in emergency department patients with atrial fibrillation or flutter. Issue 3 (May 2020)

• Record Type: Journal Article
• Authors: Scheuermeyer, Frank X.; Norena, Monica; Innes, Grant; Grunau, Brian; Christenson, Jim; Grafstein, Eric; Barbic, David; Barrett, Tyler
• Is Part Of: CJEM. Volume 22:Issue 3(2020)
• Page Start: 301
• Page End: 308
• Publication Date: 2020-05
• Subjects: Atrial fibrillation, -- cardiac disease, -- arrhythmia; Emergency Treatment -- Periodicals; Emergency Medicine -- Periodicals; Emergency medical services -- Canada -- Periodicals; Medical emergencies -- Canada -- Periodicals; Emergency medical services; Medical emergencies; Canada; Periodicals; 616.02505
• DOI: 10.1017/cem.2019.454
• Languages: English
• ISSNs: 1481-8035

Abstract

ABSTRACT: Background: Emergency department (ED) patients with atrial fibrillation or flutter (AFF) with underlying occult condition such as sepsis or heart failure, and who are managed with rate or rhythm control, have poor prognoses. Such conditions may not be easy to identify early in the ED evaluation when critical treatment decisions are made. We sought to develop a simple decision aid to quickly identify undifferentiated ED AFF patients who are at high risk of acute underlying illness. Methods: We collected consecutive ED patients with electrocardiogram-proven AFF over a 1-year period and performed a chart review to ascertain demographics, comorbidities, and investigations. The primary outcome was having an acute underlying illness according to prespecified criteria. We used logistic regression to identify factors associated with the primary outcome, and developed criteria to identify those with an underlying illness at presentation. Results: Of 1, 083 consecutive undifferentiated ED AFF patients, 400 (36.9%) had an acute underlying illness; they were older with more comorbidities. Modeling demonstrated that three predictors (ambulance arrival; chief complaint of chest pain, dyspnea, or weakness; CHA2 DS2 -VASc score greater than 2) identified 93% of patients with acute underlying illness (95% confidence interval [CI], 91–96%) with 54% (95% CI, 50–58%) specificity. The decision aid missed 28 patients; (7.0%) simple blood tests and chest radiography identified all within an hour of presentation. Conclusions: In ED patients with undifferentiated AFF, this simple predictive model rapidly differentiates patients at risk of acute underlying illness, who will likely merit investigations before AFF-specific therapy.