Incidence and Case-Fatality of Hemorrhagic Stroke In Patients With Atrial Fibrillation: A Nationwıde Registry-Based Cohort Study

Paula Tiili, Mika Lehto, Olli Halminen, Alex Luojus, Elis Kouki, Jussi Niiranen, Haukka Jari, Mika Linna, Pirjo Mustonen, Juhani Airaksinen, Janne Kinnunen, Jukka Putaala

Keywords: Atrial Fibrillation, hemorrhagic stroke, incidence, case fatality, registry-based

Background:
Oral anticoagulants, while highly efficient in thrombosis prophylaxis, also impose an increased risk of bleeding. There are limited population-based data on the epidemiology of hemorrhagic stroke in patients with atrial fibrillation (AF). We aimed to describe hemorrhagic stroke incidence and case-fatality among patients with AF in Finland.

Research questions:
• What is the incidence of hemorrhagic stroke in patients with AF?
• What is the case-fatality in hemorrhagic stroke in patients with atrial AF?
• Has there been a change in the incidence or case-fatality in the recent decade?

Method:
The Finnish AntiCoagulation in Atrial Fibrillation (FinACAF) is a nationwide retrospective registry-linkage database of patients diagnosed with AF between 2004-2018 in Finland. To assess temporal trends, we compared two time periods, 2009-2013 and 2014-2018, with a minimum of 1-year follow-up. Patients with a prior diagnosis of any intracranial hemorrhage during a minimum look-back time of 5 years were excluded. Crude incidence rates and 30-day case-fatality were calculated for subarachnoid hemorrhage (SAH) and intracerebral hemorrhage (ICH).

Results:
In years 2009-2013, 85,000 patients (49% female) contributed a total of 219,506 person-years. Crude incidence rate for SAH was 0.33 (95% confidence interval 0.26-0.42) and for ICH 2.28 (2.08-2.49) per 1,000 person-years, respectively. The 30-day case-fatality was 50.7% (40-64%) and 51% (47-56%), respectively.

In years 2014-2018,90,569 patients contributed altogether 238,631 person-years. Crude incidence rate for SAH was 0.38 (0.30-0.46) and for ICH 2.69 (2.48-2.90) per 1,000 person-years, respectively. The 30-day case-fatality was 34% (26-46%) for SAH and 45% (41-49%) for ICH.

Conclusions:
In this population-based nationwide registry-based study, we observed an increasing incidence of both SAH and ICH in patients with AF during the last decade. However, 30-day mortality from hemorrhagic stroke decreased, particularly that of SAH. Further research on factors associated with the risk of hemorrhagic stroke in this population is pending.

Points for discussion:
Collaboration between researchers at primary care and other levels of care. Experiences? Benefits? Difficulties?