Better stroke prevention in patients with device-detected atrial fibrillation and vascular disease
Junior comment:
Device-detected atrial fibrillation (DDAF) refers to episodic atrial arrhythmias that resemble AF but are often characterized by short duration and rarity; moreover, DDAF is found in older patients with cardiovascular disease, without electrocardiogram (ECG) - documented atrial fibrillation (AF) (1-2). The recently combined analysis of the NOAH-AFNET 6 and ARTESiA trials evaluated the pooled results of these two clinical trials that compared and evaluated the safety and efficacy of direct oral anticoagulants (DOAC) versus aspirin or placebo in patients with DDAF and stroke risk factors (3). The need of this combined analysis was highlighted and based on the, at least, contrary results of these trials. On the one hand NOAH-AFNET 6 was early terminated due to increase in bleeding events and no efficacy benefit in the edoxaban arm (death or non-fatal stroke)(4) while in ARTESiA apixaban was associated with lower risk of stroke or thromboembolic events but also increased risk of bleeding events, compared to aspirin (5).
he question raised was whether vascular disease prompts more aggressive anticoagulant treatment; towards this aim they divided both trial’s population on the basis of the presence or absence of vascular disease. It is of great importance to note that almost half of both trials’ population had vascular disease, directly associated indication of aspirin treatment at enrollment and moreover, at both studies vascular disease was indeed associated with increased CHA2DS2-VASc score (4.3 1.3 in NOAH-AFNET 6 and 4.5 1.2 in ARTESiA). Regarding efficacy, in both trials, the primary outcome occurred less frequently in patients without vascular disease. As regards safety endpoints, it is important to appreciate that in NOAH-AFNET 6, although major bleeding events occurred more often in the edoxaban arm, at the presence of vascular disease, the difference between edoxaban and aspirin arm was less definitive. At the absence of vascular disease bleeding events were more pronounced. Surprisingly, in ARTESiA there was no significant difference between arms regarding major bleeding events.
Reading between the lines, there are certain points needed to be highlighted. Firstly, although these two trials shared similar design and outcomes characteristics, there was an important difference of the definition in the time duration of DDAF (in NOAH-AFNET 6, patients with DDAF longer at least 6 min were enrolled, while in ARTESiA patients with DDAF lasting 6 min to 23:59 were enrolled). Interpreting the above, this could reflect different AF burden in each population (6). As addressed by the authors, the indication of DOACs, as well as the benefit of treatment is directly associated with the presence of AF, while in patients without AF, DOAC does not improve either stroke prevention or bleeding compared to aspirin. The problem raised by DDAF is the absence of ECG-documented AF and DDAF patients linger between the forementioned populations, which is even more highlighted on the basis of low AF burden (7) Secondly, the increase of major bleeding events was lower in patients with vascular disease, mainly due to the fact that aspirin was indicated in this population and the comparator treatment differed.
The above findings could suggest that anticoagulation treatment, although associated with better outcomes in patients with, compared to patients without vascular disease should be carefully addressed and guided by a shared-decision-making plan between the patients and the physicians.
Senior comment:
In the last ESC guidelines on atrial fibrillation, oral anticoagulant treatment (OAC) may be indicated (class IIb, level of evidence B) in device detected atrial fibrillation (DDAF) in high-risk patients (8). The recommendation is based on 2 recent published trials including patients with DDAF defined as arrythmia for at least 6 minutes. The NOAH-AFNET 6 study randomized patients ≥ 65 years with one additional risk factor, to edoxaban or no OAC. Patients randomized to no OAC received aspirin if indicated (4). The ARTESiA trial randomized patients ≥ 55 years and with a CHA2DS2-VASc score of at least 3, to apixaban or aspirin, open-label aspirin on top of study treatment was permitted (5). Both studies applied DOAC doses approved for stroke prevention in AF. The incidence of stroke was low in both studies.
In the present paper Schnabel et al. performed subgroup analyses of NOAH-AFNET 6 and ARTESiA in patients with and without vascular disease (3). Almost half of the randomized patients had vascular disease (56.0% in NOAH-AFNET 6 and 45.9% in ARTESiA). Patients with vascular disease also had a higher prevalence of heart failure and diabetes mellitus compared to patients without.
The primary outcome (stroke, systemic embolism, myocardial infarction, pulmonary embolism or cardiovascular disease) occurred in 5.0%/ patient-year (NOAH-AFNET 6) and 4.4%/patient-year (ARTESiA) in patients with vascular disease randomized to no OAC. The primary outcome occurred less frequently in patients without vascular disease. The meta-analysis of the effects of DOAC versus aspirin/placebo or aspirin showed an incidence rate ratio (IRR) in the vascular disease group for the primary outcome of 0.75 (95% C.I. 0.61-0.92) as compared to 1.01 (95% C.I. 0.78-1.30) in patients without vascular disease. The risk of major bleeding increased by DOAC in both patients with (IRR 1.55 (95% C.I. 1.10-2.20) and without (IRR 1.93 (95% C.I. 0.72-5.20) vascular disease.
The present study highlights that we need to define the DDAF patients who benefit most from OAC treatment. Patients with vascular disease are a vulnerable group and the disease is often accompanied by heart and renal failure which further increase the risk of thromboembolic events. A secondary analysis from the ARTESiA trial previously showed an increasing risk of stroke by CHA2DS2-VASc score (9), emphasizing that there may be other important comorbidities beyond vascular disease to include in risk stratification. In contrast to the patients with vascular disease, few thromboembolic events were found in patients without vascular disease and these patients had less benefit from OAC treatment.
A personalized approach to selecting DDAF patients with low and high risk of thromboembolic events will likely improve the risk-benefit ratio of OAC treatment. Additional studies with a focus on different risk factors and their contribution to the risk om thromboembolic events in DDAF is needed.
Our mission: To reduce the burden of cardiovascular disease.