HEMORR₂HAGES Score

The HEMORR2HAGES Score is a validated bleeding risk assessment tool designed specifically for patients with atrial fibrillation (AF) receiving anticoagulation therapy. The score was developed from a large Medicare database and uses 11 clinical risk factors to predict the annual risk of major hemorrhage. The acronym stands for Hepatic or renal disease, Ethanol abuse, Malignancy, Older age (>75), Reduced platelet count/function, Re-bleeding risk (prior bleed, weighted double), Hypertension (uncontrolled), Anemia, Genetic factors (CYP2C9 variants), Excessive fall risk, and Stroke history.

The HEMORR2HAGES Score was published by Gage et al. in 2006 and represents one of several bleeding risk scores available for AF patients on anticoagulation, alongside the HAS-BLED and ATRIA scores. What distinguishes HEMORR2HAGES is its comprehensive inclusion of risk factors, particularly genetic factors (CYP2C9 polymorphisms affecting warfarin metabolism) and the double weighting of prior bleeding history, which is the strongest predictor of future hemorrhage. The score ranges from 0 to 12 points.

Risk stratification with the HEMORR2HAGES Score helps clinicians balance the competing risks of thromboembolic stroke and hemorrhage in AF management. A score of 0-1 indicates low bleeding risk with an annual major hemorrhage rate of approximately 1.9-2.5%, strongly favoring anticoagulation if stroke risk is elevated. A score of 2-3 indicates intermediate risk (5.3-8.4% annual bleed rate), requiring careful risk-benefit analysis. A score of 4 or higher indicates high risk (>10% annual bleed rate), necessitating reassessment of anticoagulation strategy, potentially favoring direct oral anticoagulants (DOACs) over warfarin or considering left atrial appendage closure.

The individual risk factors in the HEMORR2HAGES Score reflect well-established mechanisms of bleeding risk. Hepatic disease impairs coagulation factor synthesis. Renal disease affects platelet function and drug clearance. Ethanol abuse impairs hepatic function and increases fall risk. Malignancy increases bleeding through tumor invasion, thrombocytopenia from treatment, and coagulopathy. Advanced age reduces vascular integrity and increases medication sensitivity. Reduced platelet count or function directly impairs primary hemostasis.

Prior bleeding (re-bleeding risk) receives double weight because it is the single strongest predictor of future hemorrhage. Patients with a history of major GI bleeding, intracranial hemorrhage, or other significant bleeding episodes have a markedly higher recurrence rate on anticoagulation. Uncontrolled hypertension increases the risk of intracerebral hemorrhage. Anemia may reflect occult GI blood loss and identifies patients at higher baseline bleeding risk. CYP2C9 genetic variants slow warfarin metabolism, predisposing to supratherapeutic INR levels and bleeding complications.

While the HEMORR2HAGES Score was developed primarily for warfarin-treated patients, its risk factors remain relevant in the DOAC era. Although DOACs have lower major bleeding rates than warfarin (particularly intracranial hemorrhage), the underlying patient risk factors still influence bleeding outcomes. Clinicians should use bleeding risk scores not as reasons to withhold anticoagulation (which frequently provides net clinical benefit even in higher-risk patients) but rather to identify modifiable risk factors and guide patient monitoring intensity.