Impact of Stroke Severity and Vascular Risk Factors on Early Versus Late Anticoagulation in Patients With Stroke and Atrial Fibrillation

Article information

J Stroke. 2025;27(2):284-288

Publication date (electronic) : 2025 May 31

doi : https://doi.org/10.5853/jos.2024.05477

Masatoshi Koga¹, Mattia Branca², Daniel Strbian³, Takeshi Yoshimoto⁴, Kanta Tanaka¹, Sohei Yoshimura¹, Yusuke Yakushiji⁵, Shigeru Fujimoto⁶, Adhiyaman Vedamurthy⁷, Manju Krishnan⁸, Marjaana Tiainen³, Jochen Vehoff⁹, Gerli Sibolt³, Kosuke Matsuzono⁶, Caterina Kulyk¹⁰, Silja Räty³, Peter Slade⁸, Alexander Salerno¹¹, Dimitri Hemelsoet¹², Thomas Horvath¹³, Takenobu Kunieda⁵, Makoto Nakajima¹⁴, Hisanao Akiyama¹⁵, Yasuyuki Iguchi¹⁶, Manabu Inoue¹, Masafumi Ihara⁴, Kazunori Toyoda¹, David Seiffge¹³, Martina Goeldlin¹³, Jesse Dawson^17,*, Urs Fischer^,13,*

¹Department of Cerebrovascular Medicine, National Cerebral and Cardiovascular Center, Suita, Japan

²CTU Bern, University of Bern, Bern, Switzerland

³Department of Neurology, Helsinki University Hospital and University of Helsinki, Helsinki, Finland

⁴Department of Neurology, National Cerebral and Cardiovascular Center, Suita, Japan

⁵Department of Neurology, Kansai Medical University, Hirakata, Japan

⁶Division of Neurology, Department of Medicine, Jichi Medical University, Shimotsuke, Japan

⁷Glan Clwyd Hospital, Betsi Cadwaladr University Local Health Board, Rhyl, UK

⁸Stroke Unit, Morriston Hospital, Swansea Bay University Local Health Board, Swansea, UK

⁹Department of Neurology, Cantonal Hospital St. Gallen, St. Gallen, Switzerland

¹⁰Department of Neurology, Kepler University Hospital, Johannes Kepler University, Linz, Austria

¹¹Department of Neurology, University Hospital Lausanne, University of Lausanne, Lausanne, Switzerland

¹²Department of Neurology, Ghent University Hospital, Ghent, Belgium

¹³Department of Neurology, Inselspital, University Hospital Bern and University of Bern, Bern, Switzerland

¹⁴Department of Neurology, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan

¹⁵Department of Neurology, St. Marianna University School of Medicine, Kawasaki, Japan

¹⁶Department of Neurology, Jikei University School of Medicine, Tokyo, Japan

¹⁷School of Cardiovascular and Metabolic Health, University of Glasgow, Glasgow, UK

Correspondence: Urs Fischer Department of Neurology, University Hospital Bern, University of Bern, Rosenbühlgasse 25, CH-3010 Bern, Switzerland Tel: +41-31-632-73-05 E-mail: Urs.Fischer@insel.ch

*These authors contributed equally as last author.

Received 2024 December 13; Revised 2025 March 6; Accepted 2025 April 3.

Dear Sir:

Patients with acute ischemic stroke and atrial fibrillation (AF) have a high risk of recurrent stroke [1]. In addition, the initiation of secondary prevention with anticoagulants increases the risk of bleeding events [1,2]. This risk may be particularly high in patients who have experienced more severe strokes and those with vascular risk factors, such as diabetes, hypertension, or previous stroke [3,4]. Although early initiation of direct oral anticoagulants (DOACs) is non-inferior to delayed initiation in preventing recurrent stroke, bleeding events, and death in patients with AF-related ischemic stroke [5-7], whether this holds true for patients with severe stroke and additional vascular risk factors remains unclear.

Herein, we explored whether clinical stroke severity and vascular risk factors affect the incidence of ischemic and bleeding events in patients with ischemic stroke and AF. Additionally, we assessed whether stroke severity and vascular risk factors modify the treatment effect of early versus late DOAC initiation. This study comprised a post hoc analysis of data from the Early versus Late Initiation of Direct Oral Anticoagulants in Post-ischemic Stroke Patients With Atrial Fibrillation (ELAN) trial (ClinicalTrials. gov identifier: NCT03148457), which enrolled randomized participants with acute ischemic stroke and AF to early (within 48 h of a minor/moderate stroke or 6–7 days following a major stroke) versus late (day 3–4 following a minor, day 6–7 following a moderate, or day 12–14 following a major stroke) DOAC initiation [5]. The study protocol and information on data collection have been published previously [8]. Details of participants, randomization methods, vascular risk factors, outcomes, and statistical analysis are described in the Supplementary Methods. The primary outcome was a composite of recurrent ischemic stroke, systemic embolism, major extracranial bleeding, symptomatic intracranial hemorrhage (sICH), or vascular death within 30 days of randomization (binary). Secondary outcomes included the same composite outcome within 90 days (binary) and bleeding events (composite of major extracranial bleeding and sICH). The ethics board of each participating hospital approved this study. Prior to enrollment, local investigators sought consent from each participant, next-of-kin/legal representative, or an independent physician, in accordance with local regulations.

Primary outcome events occurred in 29 (2.9%) and 41 (4.1%) participants in the early and late DOAC initiation groups, respectively. Supplementary Table 1 presents the baseline characteristics of all participants, as well as those randomized to early or late DOAC initiation groups. Vascular risk factors were well balanced between the groups. In univariate analysis, geographical region, history of ischemic events, vascular disease, chronic heart failure (New York Heart Association Classification), creatinine clearance, CHA₂DS₂-VASc score, pre-stroke modified Rankin Scale (mRS) score, and severe stroke (defined as a National Institute of Health Stroke Scale score of 10 or more) were associated with the primary outcome (at 30 days) or composite outcome at 90 days (Supplementary Table 2). Chronic heart failure and severe stroke were associated with both recurrent stroke and systemic embolism within 30 and 90 days. Severe stroke was associated with the risk of bleeding and major extracranial bleeding events. Full details of the univariate associations with all secondary outcomes are provided in Supplementary Tables 3 and 4.

Table 1 and Supplementary Table 5 present the results of multivariable analysis of the primary and secondary outcomes. No multivariable model was constructed for bleeding events within 30 days or for sICH within 30 or 90 days because of the absence of any univariate associations. In the adjusted model, chronic heart failure (adjusted odds ratio [OR], 2.49; 95% confidence interval [CI], 1.27–4.90) and severe stroke (adjusted OR, 3.33; 95% CI, 2.02–5.49) were associated with the primary outcome. In addition, early DOAC initiation, a diastolic blood pressure (BP) ≥79 mm Hg, and creatinine clearance ≥70 mL/min were associated with the composite outcome within 90 days. No significant risk factors were associated with bleeding events within 90 days. Severe stroke was associated with major extracranial bleeding within 30, but not 90, days. Chronic heart failure and severe stroke were associated with both recurrent ischemic stroke within 30 and that within 90 days. Diastolic BP was associated with recurrent ischemic stroke within 90 days. Age, sex, vascular disease, diabetes, creatinine clearance, pre-stroke mRS score, and severe stroke were negatively associated with an mRS of 0–2 and positively associated with death within 90 days.

Table 1.

Multivariable analysis of the effect of potential risk factors on primary outcome (at 30 days) and main secondary outcome (composite outcome at 90 days)

No evidence of any treatment interaction was observed between the risk factors and the impact of the randomized treatment on either of the composite outcomes at 30 or 90 days, except for body weight (Figures 1 and 2). The primary outcome (within 30 days) and composite outcome within 90 days were lower with early DOAC initiation compared with late DOAC initiation in participants with body weight <75 kg, and comparable in those with body weight ≥75 kg. The OR of bleeding events at 90 days for each dichotomized risk factor subgroup, as well as the interaction, are presented in forest plots in Supplementary Figure 2.

Figure 1.

Forest plot showing the OR with 95% CI for the comparison of early versus late DOAC initiation for the composite outcome at 30 days by dichotomized/ binary risk factors, and the interaction between the timing of DOAC initiation and risk factors. DOAC, direct oral anticoagulant; mRS, modified Rankin Scale; MI, myocardial infarction; BP, blood pressure; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; CI, confidence interval.

Figure 2.

Forest plot showing the OR with 95% CI comparing early versus late DOAC initiation for the composite outcome at 90 days by dichotomized/binary risk factors, and the interaction between the timing of DOAC initiation and risk factors. DOAC, direct oral anticoagulant; mRS, modified Rankin Scale; MI, myocardial infarction; BP, blood pressure; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; CI, confidence interval.

The main findings were as follows: chronic heart failure and severe stroke were associated with the composite outcome of recurrent ischemic stroke, systemic embolism, major extracranial bleeding, sICH, and vascular death at 30 and 90 days. Neither stroke severity nor vascular risk factors were associated with sICH, but severe stroke (National Institutes of Health Stroke Scale [NIHSS] score ≥10) was associated with major extracranial bleeding within 30 days. Severe stroke is also known to be associated with gastrointestinal bleeding [9]. Except for body weight, we saw no clear evidence of any interaction between stroke severity or vascular risk factors and the treatment effect of early versus late initiation of DOAC in people with AF-related ischemic stroke. However, owing to the low number of events, there was substantial imprecision in our estimates, as evidenced by the wide 95% CIs; further, we did not adjust for multiple comparisons, introducing a risk of both Type 1 and 2 errors. Further, there were a low number of bleeding events in our study, meaning that clinically important relationships could have been missed. In addition, this was a post hoc analysis; therefore, the results should be interpreted with caution, and external validation is needed. Nevertheless, several of these findings warrant further investigation.

Notably, we found that individuals with chronic heart failure and severe stroke have an increased risk of a composite of ischemic and bleeding events, and death within 90 days. Except for body weight, there appears to be no evidence of effect modification by stroke severity or vascular risk factors on ischemic events, bleeding events, or death following early DOAC initiation after AF-related ischemic stroke. Therefore, the findings of the ELAN trial are applicable to a wide range of patients with AF-related stroke. However, careful monitoring of major extracranial bleeding within 30 days is necessary in patients with severe stroke, regardless of the timing of DOAC initiation.