OBJECTIVES:
To assess the incidence, risk factors, and outcomes of atrial fibrillation (AF) in the ICU and to describe current practice in the management of AF.
DESIGN:
Multicenter, prospective, inception cohort study.
SETTING:
Forty-four ICUs in 12 countries in four geographical regions.
SUBJECTS:
Adult, acutely admitted ICU patients without a history of persistent/permanent AF or recent cardiac surgery were enrolled; inception periods were from October 2020 to June 2021.
INTERVENTIONS:
None.
MEASUREMENTS AND MAIN RESULTS:
We included 1,423 ICU patients and analyzed 1,415 (99.4%), among whom 221 patients had 539 episodes of AF. Most (59%) episodes were diagnosed with continuous electrocardiogram monitoring. The incidence of AF was 15.6% (95% CI, 13.8–17.6), of which newly developed AF was 13.3% (11.5–15.1). A history of arterial hypertension, paroxysmal AF, sepsis, or high disease severity at ICU admission was associated with AF. Used interventions to manage AF were fluid bolus 19% (95% CI 16–23), magnesium 16% (13–20), potassium 15% (12–19), amiodarone 51% (47–55), beta-1 selective blockers 34% (30–38), calcium channel blockers 4% (2–6), digoxin 16% (12–19), and direct current cardioversion in 4% (2–6). Patients with AF had more ischemic, thromboembolic (13.6% vs 7.9%), and severe bleeding events (5.9% vs 2.1%), and higher mortality (41.2% vs 25.2%) than those without AF. The adjusted cause-specific hazard ratio for 90-day mortality by AF was 1.38 (95% CI, 0.95–1.99).
CONCLUSIONS:
In ICU patients, AF occurred in one of six and was associated with different conditions. AF was associated with worse outcomes while not statistically significantly associated with 90-day mortality in the adjusted analyses. We observed variations in the diagnostic and management strategies for AF.