HEMORR₂HAGES Score for Major Bleeding Risk

Consider using the HEMORR₂HAGES score to stratify patients’ risk for bleeding after anticoagulation for atrial fibrillation in conjunction with situation specific risks/benefits.

The HEMORR₂HAGES score was developed to quantify the risk of hemorrhage in patients with atrial fibrillation on anticoagulation and to aid in the management of antithrombotic therapy.

• Combined hemorrhage risk factors from three previously existing clinical prediction rules to form the HEMORR₂HAGES mnemonic.
• Validated the score by retrospectively applying it to a cohort of patients from the National Registry of Atrial Fibrillation database.
• Scoring system consists of eleven criteria, one of which (history of prior bleeding) is worth 2 points, while the other ten are worth 1 point each.
• For the basis of comparison to other bleeding risk schemes, patients with a score of 0 or 1 were classified as low-risk, 2 or 3 intermediate-risk, ≥4 high-risk.
• Study found that the number of bleeds per 100 patient-years of warfarin increased as their HEMORR₂HAGES score increased.
• For patients on warfarin, study found that HEMORR₂HAGES was more discriminatory compared to the other schemes.

Points to keep in mind:

• Degree of hemorrhage was not examined in the study; all study patients with ICD-9 codes for active hemorrhage in any location were included.
• The genetic risk factor was included in their score, but not actually tested in their cohort of patients.
• There were variable definitions of hepatic or renal disease, thrombocytopenia depending on the state in which the patient lived.
• All data was obtained retrospectively, via chart review and ICD-9 codes. Events occurring outside of the hospital or coded incorrectly were not included.
• Anemia, malignancy and uncontrolled hypertension were included in a patient’s score if they were entered in as ICD-9 diagnoses, regardless if they were clinically important or not (e.g. distant history of cancer which has been in complete remission would still count as malignancy by ICD-9).
• Excessive fall risk was determine by chart review for “high risk of falling, dementia, Parkinson's disease, or psychiatric disease.”
• Study also examined ASA usage, but does not present specific data on the risk of hemorrhage on ASA with different HEMORR₂HAGES scores.
• There was no comparison to more recently derived bleeding risk scores such as ATRIA or HAS-BLED.

A subsequent systemic review and meta-analysis comparing the performance of HAS-BLED, ATRIA and HEMORR₂HAGES recommended HAS-BLED for the assessment of atrial fibrillation patients’ major bleeding risk. The analysis found that HEMORR₂HAGES had a higher diagnostic accuracy, but considered it difficult to use due to its complexity.

Combines components from several other previously validated bleeding risk scores into one.