By Michael H. Crawford, MD, Editor
SYNOPSIS: A study of apixaban vs. no anticoagulation in patients following an anticoagulant for atrial fibrillation-related intracerebral hemorrhage exhibited a high risk of stroke and vascular death, regardless of whether the patients were treated subsequently with apixaban.
SOURCE: Schreuder FHBM, van Nieuwenhuizen KM, Hofmeijer J, et al. Apixaban versus no anticoagulation after anticoagulation-associated intracerebral hemorrhage in patients with atrial fibrillation in the Netherlands (APACHE-AF): A randomized, open-label, phase 2 trial. Lancet Neurol 2021;20:907-916.
Whether to restart anticoagulants after an anticoagulant-related intracerebral hemorrhage (ICH) in patients with atrial fibrillation is a current clinical dilemma. Schreuder et al conceived the apixaban versus antiplatelet drugs or no antithrombotic drugs after anticoagulation-associated ICH in patients with atrial fibrillation (APACHE-AF) trial to provide pilot data (Phase II) to inform a larger Phase III randomized, clinical trial. Their aim was to test the hypothesis that apixaban treatment in patients with AF who had survived an anticoagulant-associated ICH was the best long-term alternative.
APACHE-AF was a prospective, randomized, open-label trial with a blinded endpoints assessment conducted in 16 hospitals in the Netherlands. Patients with a spontaneous ICH in the prior seven to 90 days during treatment with an anticoagulant for AF and a CHA2DS2-VASc score of 2 or more were included. The authors excluded patients with other reasons for anticoagulation or who presented with other potential contraindications to continuing anticoagulation (e.g., pregnancy).
Patients were randomized to apixaban at standard doses or no anticoagulation. They were stratified for intention to start an antiplatelet drug in the no anticoagulation group, age (≤ age 75 years vs. > age 75 years) and location of the ICH (lobar vs. non-lobar). The primary outcome was non-fatal stroke or vascular death. Secondary outcomes included the type of stroke, bleeding, other cardiovascular events, vascular death, and all-cause mortality. Outcomes were assessed by an adjudication committee blinded to treatment allocation. Between 2015 and 2020, 101 patients were recruited. The median age was 78 years, 54% were men, 99% were white, and the median CHA2DS2-VASc score was 4. Median follow-up was 1.9 years (range, one to three years). There were nine crossovers in the apixaban group and 11 in the no anticoagulation group.
The primary outcome occurred in 26% of the apixaban group and 24% of the no anticoagulant group (adjusted HR, 1.05; 95% CI, 0.48-2.31; P = 0.90). ICH occurred in 8% of the apixaban group, all of whom were taking apixaban when the event occurred, and one patient in the no anticoagulant group, who was taking rivaroxaban because of an intercurrent pulmonary embolism. Major bleeding also was more common on apixaban (12% vs. 6%). Ischemic stroke occurred in six patients in both groups, and three of each group crossed over to the other treatment. Major vascular events were similar in both groups (28% vs. 31%). The authors concluded there was a high risk of subsequent stroke or vascular death in patients with atrial fibrillation following an ICH during treatment with anticoagulants, regardless of whether they were treated with apixaban.
COMMENTARY
Determining the next step after an ICH in patients on anticoagulants with AF is a contentious issue. After hearing about lawsuits against cardiologists on this issue, my current practice is to consult my neurology colleagues. Often, this turns into a round robin, as they will consult me for my opinion about this risk of vascular events if anticoagulants are withheld. Thus, this paper (albeit a pilot study) was of interest. Schreuder et al found a high rate of stroke or vascular death of about 12% per annum, regardless of whether anticoagulation was given. This is not entirely surprising since these were elderly patients (median age = 78 years) with CHA2DS2-VASc scores of ≥ 2 (median = 4). However, the results of prior observational studies suggested resuming anticoagulants in such patients was beneficial overall. Of course, observational studies can be biased by indication confounding. For example, patients with a high risk of vascular events but a low risk of recurrent ICH could be treated preferentially with anticoagulants. Also, most observational studies used warfarin. The APACHE-AF authors employed apixaban because of its low bleeding risk in AF populations vs. warfarin, but still saw a higher risk of major bleeding vs. those treated with no antithrombic medications or only antiplatelet drugs. Because of crossovers in both groups, the APACHE-AF investigators conducted an on-treatment analysis, which showed 13% fewer strokes or vascular deaths in those taking apixaban. Therefore, the authors concluded their study did not inform clinical practice as it does not support or deny a benefit of apixaban in this clinical situation.
The strengths of the APACHE-AF study included the exclusion of subdural hematoma patients, which does not always happen in observational studies. Also, the up to 90-day enrollment period resulted in a more stable and homogeneous population where anticoagulation resumption was a reasonable option. In addition, there was a median follow-up of almost two years, and outcomes were adjudicated blindly. The major weakness of the study was the small cohort, resulting in tiny subgroups. This makes identifying subgroups that may benefit from anticoagulation challenging. Further, since this was an open-label trial, there may have been observation biases in outcome assessment. Finally, blood pressure control could have been more aggressive. Average blood pressure was ≥ 140/80 mmHg in both groups. There is a study in progress examining tight blood pressure control in ICH patients (TRIDENT).1
There are seven other randomized, controlled trials on this issue in progress. The authors of only one other have reported results — the SoSTART study, which included 203 patients and revealed anticoagulants were noninferior to not starting them, but not superior, either.2 If, like SoSTART, none of the other trials help clarify this clinical dilemma, there is hope that a meta-analysis of them will provide some clarity.
REFERENCES
- ClinicalTrials.gov. Triple therapy prevention of recurrent intracerebral Disease EveNts Trial (TRIDENT) cognitive sub-study (TRIDENT COG).
- SoSTART Collaboration. Effects of oral anticoagulation for atrial fibrillation after spontaneous intracranial haemorrhage in the UK: A randomised, open-label, assessor-masked, pilot-phase, non-inferiority trial. Lancet Neurol 2021;20:842-853.
A study of apixaban vs. no anticoagulation in patients following an anticoagulant for atrial fibrillation-related intracerebral hemorrhage exhibited a high risk of stroke and vascular death, regardless of whether the patients were treated subsequently with apixaban.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content