Which Patients With Atrial Fibrillation Should Receive Anticoagulation?

Thomas A. Zelniker, MD; Robert P. Giugliano, MD, FACC


March 29, 2018

Editorial Collaboration

Medscape &

Atrial Fibrillation in the General Population

Atrial fibrillation (AF) is the most prevalent cardiac rhythm disorder in the elderly population and is the leading cause of stroke and systemic embolic events. Although AF is classified clinically according to its duration as paroxysmal, persistent, or permanent,[1] the risk for thromboembolism appears to be consistent among these categories.

AF usually originates in the left atrium at the pulmonary veins and can be easily detected and diagnosed on the basis of an irregular pattern with varying RR intervals on ECG. Because patients with AF may be completely asymptomatic, opportunistic screening for silent AF can be easily performed at low cost and little burden for the patient, and is therefore recommended in patients aged > 65 years by guidelines from the European Society of Cardiology (ESC)[2] and the American Heart Association Stroke Council.[3] There has been increasing interest in the development of smartphone-connected wearable devices that collect the carrier's ECG (usually using a single lead) and issue a warning notice when a suspected pathologic finding is detected. However, none of these devices is currently approved by the US Food and Drug Administration, and patients can be unsettled by false-positive results.

AF is usually not life-threatening, but it can elicit serious complications. Several studies have shown a robust association between AF and thromboembolic events.[4] AF results in a state of hypercoagulation due to stasis and turbulence in the atrial auricles and may trigger the development of clots causing thromboembolic events. However, the link between the timing of stroke and the presence of AF is complex and not well understood. In a study of patients with pacemakers and defibrillators, no temporal relationship between the episodes of AF was identified.[3] Likewise, maintaining sinus rhythm by using electric cardioversion and/or antiarrhythmic drugs does not reduce the risk for stroke. On the contrary, the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial suggested a mortality benefit in rate-controlled patients compared with rhythm control.[5]

Furthermore, AF is associated with an increased risk for heart failure, ventricular arrhythmias, and death, and it worsens the prognosis of concomitant cardiovascular diseases. However, it remains uncertain whether AF is causally linked to these conditions.

Calculating Risk

The benefit of anticoagulation for stroke prophylaxis in patients with AF is well established.[6] Several studies also proved that the efficacy of anticoagulation is superior to that of antiplatelet agents alone.[7,8] However, clinicians often face a much more challenging dilemma in identifying the optimum anticoagulation method and determining the patient's risk benefit between the risk for stroke or bleeding. In this regard, the CHA2DS2VASc score—a seven-variable score consisting of age, sex, history of heart failure, hypertension, stroke, vascular disease, and diabetes—has replaced and expanded the previously used CHADS2 score.[9]

The risk for stroke or systemic embolic events grows with increasing scores, and therefore oral anticoagulation should be considered in all patients with a score ≥ 2; patients with a score of 0 are at low risk. The European guidelines[2] also recommend oral anticoagulation in males with a score of 1, whereas the ACC/AHA/HRS guidelines[1] consider a point score of 1 as intermediate risk and antithrombotic or anticoagulant therapy may be withheld, acknowledging uncertainty in the level of evidence. Supporting the US guidelines, a recent systematic review[10] concluded that stroke rates differ substantially among tested cohorts, and indicated that annual stroke rates were < 2% in most patients with CHA2DS2VASc scores of 0-2—the presumed threshold to expect a clinical benefit from anticoagulation therapy.

The trade-off between the risks and the benefits of anticoagulation is often summarized as the net clinical outcome.[11] However, weighting these risks may be difficult, because some elderly patients may perceive suffering from stroke as a larger threat than death from bleeding.

Risk assessment of bleeding is challenging because many bleeding scores share the same variables that are used to estimate the risk for stroke or systemic embolic events. In addition, the presently available bleeding scores exhibit low discrimination, with an area under the curve < 0.70. The HAS-BLED score[4] (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly, drug/alcohol use concomitantly) is one of the most established scores, with scores of ≥ 3 indicating a high risk for bleeding.

Alternative scores are the HEMORR2HAGES[12] and ATRIA[13] scores. Of note, the mentioned bleeding scores should not be primarily used as a decision-making tool regarding in which patients to avoid anticoagulation, but to assist in the identification of modifiable risk factors to reduce the risk for major bleeding.

There is growing evidence that biomarkers may become helpful tools in the risk stratification of bleeding events. A multimarker approach using cardiac troponin I, N-terminal pro-B–type natriuretic peptide, and D-dimer levels improved risk stratification when added to the CHA2DS2VASc score.[14] In addition, the ABC bleeding score, a combined clinical and biomarker score that was derived and validated in two modern AF trials, had a higher c-statistic than the HAS-BLED score.[15]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.