GARFIELD-AF risk score for mortality, stroke, and bleeding within 2 years in patients with atrial fibrillation

Area of Science:

• Cardiology
• Clinical Epidemiology
• Health Services Research

Background:

• Atrial fibrillation (AF) management requires accurate risk stratification for stroke and bleeding.
• Existing risk tools like CHA2DS2-VASc and HAS-BLED have limitations in predicting outcomes.
• The Global Anticoagulant Registry in the FIELD-Atrial Fibrillation (GARFIELD-AF) integrated risk tool was developed to improve risk prediction.

Purpose of the Study:

• To evaluate the predictive performance of the GARFIELD-AF integrated risk tool for mortality, non-haemorrhagic stroke/systemic embolism, and major bleeding.
• To compare the GARFIELD-AF risk tool's performance against established risk predictors (CHA2DS2-VASc and HAS-BLED).
• To assess the tool's utility in guiding oral anticoagulation therapy decisions.

Main Methods:

• Analysis of data from 52,080 patients in the GARFIELD-AF registry with up to 2 years of follow-up.
• Development of Cox proportional hazards models using least absolute shrinkage and selection operator (LASSO) methods.
• Internal and external validation of models using ORBIT-AF and Danish nationwide registries.
• Comparison of the GARFIELD-AF risk tool with CHA2DS2-VASc and HAS-BLED for predicting key adverse events.

Main Results:

• The GARFIELD-AF risk tool demonstrated superior prediction of all-cause mortality compared to CHA2DS2-VASc across all cohorts.
• The GARFIELD-AF score outperformed CHA2DS2-VASc for non-haemorrhagic stroke prediction.
• The GARFIELD-AF tool was superior to HAS-BLED in predicting major bleeding, particularly in internal validation and the Danish AF cohort.
• The tool showed strong discriminatory value in very low- to low-risk patient groups (defined by CHA2DS2-VASc scores).
• The GARFIELD-AF tool incorporates oral anticoagulation (OAC) therapy status, enabling comparison of outcomes with different treatment strategies (no OAC, non-vitamin K antagonist oral anticoagulants, VKAs).

Conclusions:

• The GARFIELD-AF integrated risk tool offers improved prediction of mortality and non-haemorrhagic stroke compared to CHA2DS2-VASc.
• The GARFIELD-AF tool demonstrates better prediction of major bleeding than HAS-BLED, especially in lower-risk AF populations.
• This risk tool provides valuable insights for clinical decision-making regarding anticoagulation therapy in AF patients.