By Michael H. Crawford, MD, Editor

SYNOPSIS: An analysis of the utility of implanted loop recorders to detect subclinical atrial fibrillation in high-risk individuals showed that among common arrhythmia-compatible symptoms, only palpitation was predictive of discovering episodes of atrial fibrillation.

SOURCE: Reiffel JA, Verma A, Kowey PR, et al. Relation of antecedent symptoms to the likelihood of detecting subclinical atrial fibrillation with inserted cardiac monitors. Am J Cardiol 2021;145:64-68.

In the 2017 REVEAL AF study, the authors showed that in a population enriched by selecting subjects at risk for subclinical atrial fibrillation (SAF), 40% would experience one or more episodes of SAF lasting for at least six minutes. This, after 30 months of monitoring by a subcutaneously implanted loop ECG recorder.1

Entry into REVEAL AF required a CHA2DS2-VASc score ≥ 3 or a score of 2 with one or more of the following risk factors for SAF: coronary artery disease (CAD), impaired renal function, sleep apnea, or chronic obstructive lung disease (COPD). Excluded were patients with a history of AF, those on anticoagulant therapy, and those with a stroke or transient ischemic attack (TIA) within the last 12 months. REVEAL AF showed older age, body mass index, and a particular genetic profile were independently predictive of discovering SAF.

This report by Reiffel et al (who authored REVEAL AF) tested the hypothesis that symptoms would predict the occurrence of SAF in the REVEAL AF population. This substudy included 346 subjects (mean age, 71 years; 50% men) who were followed for a mean of 22 months. Comorbid conditions were common: two-thirds had diabetes or CAD, 40% reduced renal function, one-third remote stroke/TIA, one-quarter sleep apnea, and 20% heart failure or COPD. Arrhythmia-compatible symptoms were ascertained for the three months before enrollment in REVEAL AF, including palpitation, fatigue, chest discomfort, increased heart rate, syncope, dizziness, and dyspnea. Only 39 subjects showed no symptoms; most showed more than one.

SAF was detected in 6.5% of subjects at 30 days. By six months, > 20% had SAF. Few showed SAF in the time frame of the usual external monitoring devices (< 10 days). In a multivariate analysis, the only symptom associated with the detection of SAF was palpitation (HR, 1.6; 95% CI, 1.1-2.3; P = 0.01). Interestingly, palpitation was more common in those without SAF (61%) vs. those with SAF (39%), and symptoms in general were more common in those without SAF. The authors concluded that although symptoms are more common in subjects without SAF, palpitation is associated with an increased incidence of SAF, whereas other common arrhythmia-compatible symptoms are not.


AF often is discovered because of symptoms, adverse outcomes (e.g., stroke), or it is detected on routine ECG or implanted device monitoring for other reasons. The latter is so-called subclinical AF. Experience with such monitoring has shown most patients with SAF are symptomatic, but the symptoms often are not related to the episodes of AF. Other experience suggests symptoms are more common with paroxysmal AF than chronic AF. Patients with paroxysmal AF more often experience palpitation and chest discomfort, whereas those with chronic AF more often report dyspnea and fatigue. Brief episodes of AF usually are asymptomatic. This may be one reason why six minutes of AF was the criterion for SAF in REVEAL AF, since it more likely would be symptomatic and important to the patient.

Prolonged monitoring with implanted loop recorders is a potential diagnostic tool for uncovering SAF in high-risk individuals. They can be prophylactically treated to prevent bad outcomes. The issue is when to perform such monitoring. Recommending a loop recorder to everyone older than age 70 years is not going to be feasible. Perhaps looking at an enriched population using the REVEAL AF entry criteria would be. In this regard, the Reiffel et al substudy adds another possible selection criterion: palpitation. This makes sense since one would predict more AF would be discovered in such patients compared to those without. However, REVEAL AF also showed palpitation was more common in those without SAF, even though it was the only symptom predictive of SAF in a multivariate analysis. Thus, even the presence of palpitation may not be selective enough to warrant prolonged invasive monitoring.

In addition to its post-hoc, retrospective design, there were other limitations to this substudy. Symptom collection happened before the monitoring period only. It is unclear whether the symptoms were associated with the actual SAF events. Palpitation could be the result of other events, such as re-entrant supraventricular tachycardia or even ventricular tachycardia. Many patients reported more than one symptom, yet the authors did not conduct a combination symptoms analysis. Also, they did not perform an analysis of the added value of palpitation to other clinical risk factors for AF. Consequently, at this point, it is certainly reasonable to conduct short-term external monitoring in patients with frequent arrhythmia-compatible symptoms and perhaps discuss an implantable loop recorder in those with less frequent symptoms. However, moving to implanting loop recorders in patients with only AF risk factors is an unproven way to reduce the morbidity and mortality of AF. This substudy of REVEAL AF does not establish that the additional presence of palpitation with risk factors for AF tips the balance toward implanting a loop recorder. 


  1. Reiffel JA, Verma A, Kowey PR, et al. Incidence of previously undiagnosed atrial fibrillation using insertable cardiac monitors in a high-risk population: The REVEAL AF study. JAMA Cardiol 2017;2:1120-1127.