New-onset atrial fibrillation in the medical intensive care unit: Catch me if you can.

Atrial fibrillation (AF) is often first detected in an acute setting. Prior studies have reported a wide range of estimates of the incidence of new-onset AF during acute medical illness in an intensive care unit (ICU) setting, ranging from 3% to 44%.1 New-onset AF in critically ill patients has been associatedwithworse outcomes includingmortality; however, optimal management strategies remain unclear. We read with interest the retrospective observational cohort study by Brunetti et al., involving 2234 patients with a medical ICU stay.2 They identified new-onset AF in 10.8% of patients (95%CI 9.5%– 12.2%) using ICD-10 codes that were validatedmanually through ECG and chart review. As compared to patients without AF, patients with new-onset AF had greater odds of in-hospital mortality (21.2% vs. 10.3%, odds ratio [OR] = 1.9, 95%CI 1.3–2.7). Among patients with new-onset AF who survived the index hospitalization, 18.9% (95%CI 13.6%–25.3%) had a recurrence of AF in the first year following discharge. The authors ascertained AF recurrence by manually reviewing records for 12-lead and ambulatory ECGs. We recently estimated the incidence of new-onset AF lasting at least 30 s in critically ill patients at 18.9% (95%CI 14.2%−24.3%).3 This estimate is higher than that of Brunetti’s group; differences are likely explained by different study designs. We applied a 14-day continuous ECG patch at ICU admission. In the ICU, all patients also underwent standard continuous ECG monitoring. Interestingly, the clinical team recognized only 70%of episodes detected by the patch; clinicianswere more likely to recognize longer episodes of AF. Although we do not know theminimum duration of AF that confers an increased risk, using continuous monitoring and collecting the episode duration is the most rigorous way to evaluate these episodes. The long-term management of patients with new-onset AF during an acute illness is unclear. The AF Occurring Transiently with Stress (AFOTS) follow-up study is systematically examining the long-term outcomes of patients with new-onset AF during hospitalization for intercurrent noncardiac illness or surgery.4 Participants with new-onset AF are matched to control participants for age, sex, and nature of illness. Study participants are in sinus rhythm at discharge and wear a 14-day Holter 1 and 6months after discharge. Whilewe await the results of this study, we urge readers to consider the influence of continuous ECG monitoring on the incidence of newonset AF in hospitalized patients and to re-assess the long-term need for oral anticoagulation in these patients after hospital discharge.