BACKGROUND: Atrial fibrillation (AF) is the most common serious heart rhythm disorder and is associated with a 4 to 5-fold increase in the risk of ischemic stroke. AF is often detected for the first time while a patient is hospitalized for an acute medical illness or after surgery. In these settings, AF can be transient and frequently asymptomatic; detection often occurs during prolonged periods of continuous ECG monitoring in an intensive care unit (ICU).

Atrial Fibrillation Occurring Transiently with Stress (AFOTS) describes the manifestation of AF in the acute care setting as the only evidence of AF, for which the incidence and appropriate management are uncertain. AFOTS may be directly and uniquely due to a reversible precipitant and thus unlikely to recur after this precipitant is removed, thereby having minimal impact on the patient's long-term prognosis. Alternatively, AFOTS may be the first detection of a chronic condition that is already present but undiagnosed - facilitated by inpatient continuous ECG monitoring.

Previous published studies have provided a wide range of estimates for the incidence of AFOTS in the ICU population (3-44%). Differences in estimates may be explained by the methods used to detect AFOTS in these studies.

A precise estimate of the incidence of AFOTS is required in order to understand its long-term significance. The present study is designed to generate an accurate estimate of the incidence of AFOTS.

PRIMARY OBJECTIVE: To determine the incidence of AF, lasting 30 seconds or more, in hospitalized patients in the ICU over the age of 65 and without a prior history of AF. This will be achieved with a high-sensitivity, 14-day continuous ECG monitor.

IMPORTANCE: AFOTS may be common, and may respond to established therapies for stroke prevention. However, varying methodologies in existing studies have resulted in wide variations around its incidence. The systematic protocol employed in this study will generate a precise and accurate estimate of the incidence of AFOTS.

There is uncertainty around the incidence, recurrence and management of AFOTS. The results of this study will be integrated with those from the other studies in the AFOTS research program. Together, they will inform the monitoring of hospitalized patients for AFOTS, the outpatient rhythm monitoring for recurrences of AF and ultimately stroke prevention.

STUDY DESIGN: This is a prospective descriptive epidemiologic study. It is a component study of the AFOTS Research Program.

POPULATION: Consecutive participants aged 65 years and older without a history of AF, who are admitted to the ICU at Hamilton General or Juravinski hospital for treatment of medical illness or for recovery from noncardiac surgery.

OUTCOMES: The primary outcome will be the proportion of patients with AF lasting 30 seconds or more, as detected by the patch monitor. Secondary outcomes will include: the proportion of patients who have AF documented by the clinical team; the proportion of patients with AF lasting 5 minutes or more, 1 hour or more, 6 hours or more and 24 hours or more; the burden of AF, defined as time spent in AF per 24 hours and the proportion of AF episodes that occur with an average heart rate of 1-40 bpm, 41-60 bpm, 61-80 bpm, 81-100bpm, 101-120 bpm, and 121-140 bpm and > 140 bpm.

Inclusion Criteria:

• Adults, admitted to the ICU, without a history of AF who are either:

  1. 65 years of age or older
  2. 50 years of age or older and have at least one CHADS2 Risk Factor

Exclusion Criteria:

• Patients who are palliative or not expected to survive for at least 12 hours.
• Patients for whom it is judged that placement of the ECG monitor would preclude administration of necessary care.
• Patients with known adhesive allergies.
• Patients with primary cardiovascular diagnoses (i.e. confirmed myocardial infarction, heart failure, arrhythmia) or cardiac surgery.

Consecutive patients admitted to the medical or surgical ICU