Early Direct Oral Anticoagulant Initiation After Intracranial Hemorrhage Among Japanese Patients With Nonvalvular Atrial Fibrillation ― SAFE-ICH Registry 12-Month Outcomes ―

Q: What is the clinical evidence from this study?

Study design: Cohort. Population: Nonvalvular Atrial Fibrillation and Intracranial Hemorrhage (n=240). Intervention: Early direct oral anticoagulant (DOAC) (re)initiation. Primary outcome: 12-month composite of symptomatic ICH, any stroke, or all-cause death (95% CI 17.4-28.3).

Masatoshi Koga; Yasuyuki Iguchi; Ryo Itabashi; Hiroharu Kataoka; Michikazu Nakai; Kanta Tanaka; Kaori Miwa; Rei Kondo; Tatemi Todaka; Michiko Yokosawa; Kunikazu Yoshimura; Shigeo Yamashiro; Takahiro Ogishima; Yuka Terasawa; Nobukazu Komoribayashi; Nobuyuki Ohara; Taketo Hatano; Shuji Arakawa; Kazutaka Sonoda; Joji Hagii; Teruyuki Hirano; Fumio Miyashita; Hiroyuki Ohnishi; Kazumi Kimura; Toshihiro Ueda; Keiko Maezono-Kandori; Yukako Yazawa; Masanori Kinosada; Tsuyoshi Izumo; Shouhei Noshiro; Kuniyasu Wada; Shuhei Okazaki; Toshiichi Watanabe; Yasushi Takagi; Yasuhiro Ito; Masako Sugimoto; Atsushi Takita; Hiroaki Kobayashi; Masafumi Ihara; Shigeru Fujimoto; Ḱazunori Toyoda; on behalf of the SAFE-ICH Registry Group

doi:10.1253/circrep.cr-26-0249

October 5, 2026Circulation ReportsOpen Access

Early Direct Oral Anticoagulant Initiation After Intracranial Hemorrhage Among Japanese Patients With Nonvalvular Atrial Fibrillation ― SAFE-ICH Registry 12-Month Outcomes ―

View Full Paper

Ask AI

Bookmark

Share

Key result

Early DOAC reinitiation after ICH carries a ~23% 1-year major event rate, dominated by bleeding.

95% CI 17.4-28.3
n=240

Why the study?

Understanding the effects of early direct oral anticoagulant (DOAC) (re)initiation after intracranial hemorrhage (ICH) in atrial fibrillation is crucial, with prior 30-day outcomes showing acceptable ischemic and hemorrhagic risks but longer-term data lacking.

Key Points

To evaluate the 12-month ischemic and hemorrhagic outcomes after early direct oral anticoagulant (DOAC) (re)initiation within 14 days of intracranial hemorrhage in patients with nonvalvular atrial fibrillation.
Conducted a prospective, observational cohort study (SAFE-ICH Registry) across 32 hospitals in Japan.
Enrolled 240 patients with nonvalvular atrial fibrillation who (re)initiated DOAC therapy ≤14 days after intracranial hemorrhage onset (median 7 days).
Monitored outcomes through clinical assessments, neuroimaging, or telephone interviews over a 12-month follow-up period.
The 12-month composite outcome incidence of symptomatic intracranial hemorrhage, any stroke, or all-cause death was 22.5% (95% CI 17.4–28.3%; 29.1/100 person-years).
Individual 12-month secondary outcomes included symptomatic intracranial hemorrhage in 6.3% (95% CI 3.5–10.1%; 7.9/100 person-years), any stroke in 6.7% (95% CI 3.9–10.6%; 8.4/100 person-years), and all-cause death in 14.2% (95% CI 10.0–19.2%; 17.3/100 person-years).
Long-term hemorrhagic stroke rates numerically exceeded ischemic stroke rates (intraparenchymal hemorrhage: 3.8% [95% CI 1.7–7.0%] vs ischemic stroke: 2.9% [95% CI 1.2–5.9%]), with hemorrhagic stroke first emerging approximately 3 months after starting DOACs.

Study Design

Type

Cohort (n=240)

Multicenter

Yes

Structured PICO

P

Population

240 Japanese patients (mean age 79.4, 38.7% female) with nonvalvular atrial fibrillation who experienced intracranial hemorrhage and had DOAC therapy (re)initiated within 14 days, followed for 12 months.

E

Exposure

Direct oral anticoagulant (DOAC) therapy (re)initiated ≤14 days after intracranial hemorrhage onset.

O

Outcome

12-month composite outcome of symptomatic ICH, any stroke, or all-cause deathcomposite

Early DOAC (re)initiation within 14 days of intracranial hemorrhage in Japanese patients with nonvalvular atrial fibrillation is associated with a 22.5% composite event rate at 12 months, with a notable long-term increase in hemorrhagic stroke risk relative to ischemic stroke.

Limitations

Conducted exclusively in Japan, which may limit generalizability to other populations
DOAC dosing tends to be lower in Japan than in other countries
Low proportion of patients with lobar and cerebral amyloid angiopathy-related hemorrhage
No randomized comparator was used (single-arm registry)
Selection bias may exist because DOAC therapy initiation was decided by the treating physician
Long-term blood pressure control and DOAC adherence were not evaluated
Did not report on the use of DOAC reversal agents
Conducted exclusively in Japan, limiting generalizability
DOAC dosing tends to be lower in Japan
Low proportion of lobar and cerebral amyloid angiopathy-related hemorrhage
No randomized comparator was used
Selection bias due to DOAC initiation timing being decided by the treating physician
Long-term blood pressure and DOAC adherence were not evaluated

Implication

Early DOAC resumption after ICH warrants caution in NVAF given excess hemorrhage; leaves optimal timing unresolved pending RCTs.

Abstract

Background: Understanding the effects of early direct oral anticoagulant (DOAC) (re)initiation after intracranial hemorrhage (ICH) in atrial fibrillation is crucial. Previous 30-day outcomes in the SAFE-ICH Registry showed acceptable ischemic/hemorrhagic risks; therefore, we evaluated the 12-month outcomes. Methods and Results: The SAFE-ICH Registry, a prospective, observational cohort study conducted at 32 hospitals in Japan, included 240 patients with nonvalvular atrial fibrillation (NVAF) who experienced ICH and in whom DOAC therapy was (re)initiated ≤14 days after ICH. Short-term (≤30 days) and long-term (≤12 months) ischemic/hemorrhagic outcomes were investigated. Data were collected through clinical assessments, imaging or telephone interviews over 12 months. The 12-month composite outcome incidence (symptomatic ICH, any stroke, all-cause death) was 22.5% (29.1/100 person-years). The secondary outcome incidence was 6.3% (7.9/100 person-years) for symptomatic ICH, 6.7% (8.4/100 person-years) for any stroke (ischemic: 2.9% [3.6/100 person-years]; hemorrhagic: 3.8% [4.6/100 person-years]), and 14.2% (17.3/100 person-years) for all-cause death. The major bleeding incidence was 6.7% (8.4/100 person-years). Hemorrhagic stroke first appeared approximately 3 months after DOAC (re)initiation, and by month 10, its rate was comparable to that of ischemic stroke. Most deaths were due to aspiration pneumonia. Conclusions: Early DOAC (re)initiation after ICH in NVAF was associated with a numerically higher long-term hemorrhagic than ischemic stroke risk. These risks warrant careful consideration when determining the optimal DOAC duration and timing after ICH.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Koga et al. (2026) conducted a cohort in Nonvalvular Atrial Fibrillation and Intracranial Hemorrhage (n=240). Early direct oral anticoagulant (DOAC) (re)initiation was evaluated on 12-month composite of symptomatic ICH, any stroke, or all-cause death (95% CI 17.4-28.3). Early DOAC (re)initiation within 14 days after intracranial hemorrhage in NVAF patients resulted in a 12-month composite event incidence of 22.5%, with hemorrhagic events exceeding ischemic events.

synapsesocial.com/papers/6ac5a19c164a9ca62de5b34d https://doi.org/10.1253/circrep.cr-26-0249

Topics

Atrial fibrillation

Persistent AF management

Anticoagulation in AF

View Full Paper

Ask AI

Bookmark

Share

Early Direct Oral Anticoagulant Initiation After Intracranial Hemorrhage Among Japanese Patients With Nonvalvular Atrial Fibrillation ― SAFE-ICH Registry 12-Month Outcomes ―

Key Points

Study Design

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Topics

Early Direct Oral Anticoagulant Initiation After Intracranial Hemorrhage Among Japanese Patients With Nonvalvular Atrial Fibrillation ― SAFE-ICH Registry 12-Month Outcomes ―

Key Points

Study Design

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Topics

Keep reading