By Michael H. Crawford, MD, Editor
An analysis of the ARTESiA trial of apixaban vs. aspirin for stroke prevention in subclinical atrial fibrillation by CHA2DS2VASc score has shown that the benefits of apixaban outweigh the risk of major bleeding at scores > 4 and the opposite is true at scores < 4.
Lopes RD, Granger CB, Wojdyla DM, et al. Apixaban versus aspirin according to CHA2DS2VASc score in subclinical atrial fibrillation: Insights from ARTESiA. J Am Coll Cardiol 2024; May 19. https://doi.org/10.1016/j.jacc...
The CHA2DS2VASc score is used to identify low-risk atrial fibrillation (AF) patients (stroke risk ≤ 1%/year) in whom oral anticoagulation (OAC) can be withheld. Subclinical AF (SAF) has been defined as short asymptomatic episodes detected by implanted electrocardiogram (ECG) monitors (loop recorders, pacemakers). SAF is associated with stroke but at a lower risk than clinical AF (CAF). The Apixaban for the Reduction of Thrombo-Embolism in Patients with Device-Detected Sub-Clinical Atrial Fibrillation (ARTESiA) trial showed that OAC with apixaban reduced stroke risk by 37% but significantly increased major bleeding.1 Thus, it would be desirable to risk stratify SAF patients to determine who needs OAC.
To this end, the ARTESiA investigators performed a subgroup analysis according to the baseline CHA2DS2VASc score divided into three groups: < 4, 4, and > 4. SAF was defined as one or more episodes of AF identified on either an implanted loop recorder or pacemaker lasting ≥ 6 minutes but not > 24 hours. Eligible AF patients were ≥ 55 years of age with a CHA2DS2VASc ≥ 3, a prior stroke, or age ≥ 75 years. Excluded were patients with CAF, another indication for OAC, uncorrected major bleeding within six months, or a creatinine clearance < 25 mL/min. Almost all the enrolled patients had a pacemaker (95%) and were randomized to apixaban 5 mg twice per day (or 2.5 mg if criteria met) or aspirin 81 mg/day. Patients were censored if they developed CAF (AF episode > 24 hours). The primary outcome was stroke or systemic embolism and stroke severity by the modified Rankin Scale score as adjudicated by an independent panel of neurologists. The primary safety endpoint was major bleeding.
The distribution of the 4,012 patients (mean age 77 years; 36% women) in the three CHA2DS2VASc score groups was 39% in the < 4 group, 34% in the 4 group, and 27% in the > 4 group. Over the median 3.5-year follow-up period in the < 4 group, the stroke rate was 0.85% per year with apixaban and 0.97% with aspirin (hazard ratio [HR], 0.85; 95% confidence interval [CI], 0.5-1.52) and the major bleeding rate was 0.4% per year with apixaban and 1.1% with aspirin (HR, 1.27; 95% CI, 0.8-2.1).
In the 4 group, the stroke rate was 0.54% per year with apixaban and 0.86% with aspirin (HR, 0.63; 95% CI, 0.32-1.27) and major bleeding was 1.2% per year with apixaban and 0.9% with aspirin (HR, 1.31; 95% CI, 0.75-2.29). In the > 4 group, the stroke rate was 0.98% per year with apixaban and 2.25% with aspirin (HR, 0.44; 95% CI, 0.25-0.77; P = 0.06) and the major bleeding rate was 2.13% with apixaban and 1.45% with aspirin (HR, 1.48; 95% CI, 0.89-2.45). Subgroup analyses showed that only gastrointestinal bleeding was statistically significantly different; it was found in 5% of those taking apixaban and in 2% of those taking aspirin (P = 0.01). In the > 4 patients, the number needed to treat to prevent one stroke with apixaban was 25 and the number needed to treat to cause one major bleed was 59. The severity of strokes increased with higher CHA2DS2VASc scores, and the rate of severe bleeds was lower in the apixaban-treated patients.
The authors concluded that in patients with SAF and a CHA2DS2VASc score > 4, the benefits of apixaban treatment for preventing stroke outweighed the risks and the opposite is true for a CHA2DS2VASc score < 4. In the one-third of SAF patients with scores of 4, patient preferences may help inform the decision.
COMMENTARY
The recent publication of the main trial results from the ARTESiA study were disappointing because the benefit of apixaban compared to aspirin seemed to be at the expense of increased major bleeding. In CAF, the CHA2DS2VASc score is recommended for determining for whom OAC treatment might be reasonable: a score of ≥ 1 in men and ≥ 2 in women. Thus, it is logical to analyze the ARTESiA data to determine whether the CHA2DS2VASc score would help decide which patients with SAF to treat with OAC. Thus, the Lopes et al paper is of interest.
The researchers showed that the benefit of apixaban compared to aspirin increased as the CHA2DS2VASc score increased and that, although apixaban increased the risk of major bleeding across the board, at a CHA2DS2VASc score > 4, the benefit exceeded the risk. Interestingly, there was no sex difference in the score at which benefit exceeded risk. Accordingly, they concluded that at scores < 4 there is a low risk of stroke, and the risk of bleeding outweighs the potential benefit. At scores > 4, the stroke risk is high and the benefit of apixaban outweighs the risk. At a score of 4, it is reasonable to consider anticoagulation.
The absolute risk of stroke in SAF in ARTESiA was low (1% to 1.5%/year) over the median 3.5-year follow-up period. A major reason for this finding is that only 9% of the patients had a history of prior stroke. Given this low overall risk of stroke, one could argue that any increase in major bleeding is not acceptable. However, the increase in major bleeding was mainly driven by gastrointestinal bleeding, which was not severe and was readily managed conservatively. Also, the severity of stroke was increased in those with scores > 4. On the other hand, the authors do not recommend apixaban for primary prevention in SAF patients with scores < 4 of either sex. A limitation of ARTESiA is that the comparator was aspirin, not placebo. This probably was because they believed it would be unethical to not provide some treatment. In addition, we do not know if the results would be the same with other OACs.
ARTESiA highlights the opportunity to prevent stroke in patients with implanted devices. Almost all their patients had devices that were or included pacemakers, which usually are monitored periodically. Whether these data would apply to those with watches or other nonimplanted devices is not clear, but the possibility exists if the AF detected can be verified. Also, in ARTESiA one-quarter of their patients went on to develop CAF, half of them within 18 months. Thus, those not put on OAC should be monitored closely.
REFERENCE
- Healey JS, Lopes RD, Granger CB, et al. Apixaban for stroke prevention in subclinical atrial fibrillation. N Engl J Med 2024;390:107-117.
