ORBIT Bleeding Risk Score in Atrial Fibrillation
The ORBIT score is a 5-factor bleeding risk assessment tool (Older age ≥75 years, Reduced hemoglobin/anemia, Bleeding history, Insufficient kidney function, Treatment with antiplatelet) that predicts major bleeding in anticoagulated AF patients, but current evidence shows it does NOT perform better than the widely-used HAS-BLED score and should NOT replace it in clinical practice. 1, 2, 3
What is the ORBIT Score?
The ORBIT bleeding risk score was developed in 2015 from the ORBIT-AF registry and includes five clinical variables 1:
- Older age (≥75 years) = 1 point
- Reduced hemoglobin/hematocrit or history of anemia = 2 points
- Bleeding history = 2 points
- Insufficient kidney function (GFR <60 mL/min) = 1 point
- Treatment with antiplatelet agents = 1 point
Risk stratification:
- Low risk: 0-2 points
- Medium risk: 3 points
- High risk: ≥4 points
The score was initially promoted as having superior predictive ability compared to HAS-BLED (C-index 0.67 vs 0.63 in the derivation study) 1. However, this claim has not held up in subsequent validation studies.
How Does ORBIT Guide Anticoagulation Decisions?
It doesn't—and it shouldn't be used to withhold anticoagulation. Current guidelines explicitly state that bleeding risk scores should NOT be used to decide whether to start or stop oral anticoagulation 4. This is a critical point that applies to ALL bleeding risk scores, including both ORBIT and HAS-BLED.
The Evidence Against Using ORBIT to Guide Anticoagulation:
Nearly all AF patients with high bleeding risk also have high stroke risk, making anticoagulation still beneficial despite bleeding concerns 5. In the ORBIT-AF registry analysis, 98.5% of patients with high ORBIT bleeding scores (≥4) also had high stroke risk (CHA2DS2-VASc ≥2) 5. This demonstrates that bleeding risk and stroke risk are concordant—patients at highest bleeding risk are also at highest stroke risk and therefore derive the greatest absolute benefit from anticoagulation.
The proper role of bleeding risk assessment is to:
- Identify and modify reversible bleeding risk factors (uncontrolled hypertension, concurrent antiplatelet therapy, alcohol excess) 4
- Inform shared decision-making discussions about absolute risks 6
- Guide selection of specific anticoagulant agents (e.g., DOACs over warfarin for lower intracranial bleeding risk) 6, 4
- Prompt closer monitoring and follow-up for high-risk patients 4
- Consider alternative stroke prevention strategies like left atrial appendage occlusion in select cases 5
ORBIT vs HAS-BLED: The Comparative Evidence
Multiple meta-analyses and validation studies show ORBIT does not outperform HAS-BLED:
The 2022 meta-analysis by Xu et al. included 17 studies and found no statistical difference in predictive ability between ORBIT and HAS-BLED for major bleeding (C-index 0.63 vs 0.61, respectively) 2. This finding was consistent across both vitamin K antagonist and DOAC-treated patients 2.
Critical limitations of the ORBIT score:
- Poor calibration: ORBIT consistently under-predicts bleeding risk, particularly in low and intermediate risk categories 7, 3
- Excessive low-risk classification: ORBIT categorizes 88% of patients as "low risk" compared to only 33% with HAS-BLED, potentially creating false reassurance 7
- Worse performance with DOACs: In the ESC-EHRA EORP-AF registry of NOAC-treated patients, ORBIT showed poorer calibration than HAS-BLED and provided no reclassification improvement 3
The most recent 2025 study from the Murcia AF Project III cohort (3,259 patients) found all bleeding scores performed modestly (C-indexes <0.7), with ORBIT (0.664) only marginally better than HAS-BLED (0.651) for major bleeding, but HAS-BLED demonstrated superior calibration 8.
Current Guideline Recommendations
The 2024 ESC Guidelines explicitly state: "Use of bleeding risk scores to decide on starting or withdrawing oral anticoagulation is not recommended in patients with AF to avoid under-use of anticoagulation" (Class III recommendation) 4.
The 2023 ACC/AHA/ACCP/HRS Guidelines recommend: CHA2DS2-VASc score for stroke risk assessment (Class I) 9, with anticoagulation decisions based on stroke risk, not bleeding risk. Assessment and management of modifiable bleeding risk factors is recommended as part of shared decision-making (Class I) 4.
For stroke risk assessment, use CHA2DS2-VASc 6, 10:
- Score ≥2 in men or ≥3 in women: Anticoagulation recommended
- Score 1 in men or 2 in women: Consider anticoagulation based on individual factors
- Score 0 in men or 1 in women: No anticoagulation needed
For anticoagulant selection, DOACs are preferred over warfarin (except with mechanical valves or moderate-to-severe mitral stenosis) 6, 4, 6.
Common Pitfalls to Avoid
Never withhold anticoagulation based solely on a high bleeding risk score—this leads to preventable strokes 4
Don't assume ORBIT is "better" because it's newer—validation studies consistently show it's not superior to HAS-BLED 2, 3, 8
Address modifiable bleeding risk factors rather than avoiding anticoagulation: discontinue unnecessary antiplatelet therapy, control hypertension, treat anemia, reduce alcohol intake 4
Remember that high bleeding risk patients are also high stroke risk patients—they need anticoagulation most 5
Use bleeding risk assessment to enhance safety, not to deny therapy—consider DOAC selection, closer monitoring, or LAA occlusion in appropriate cases 4, 5