Send to

Choose Destination
Thromb Haemost. 2018 Apr;118(4):768-777. doi: 10.1055/s-0038-1636534. Epub 2018 Mar 6.

Incident Risk Factors and Major Bleeding in Patients with Atrial Fibrillation Treated with Oral Anticoagulants: A Comparison of Baseline, Follow-up and Delta HAS-BLED Scores with an Approach Focused on Modifiable Bleeding Risk Factors.

Chao TF1,2, Lip GYH2,3, Lin YJ1,2, Chang SL1,2, Lo LW1,2, Hu YF1,2, Tuan TC1,2, Liao JN1,2, Chung FP1,2, Chen TJ4, Chen SA1,2.

Author information

Division of Cardiology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan.
Institute of Clinical Medicine and Cardiovascular Research Center, National Yang-Ming University, Taipei, Taiwan.
Institute of Cardiovascular Sciences, University of Birmingham, Birmingham, United Kingdom.
Department of Family Medicine, Taipei Veterans General Hospital, Taipei, Taiwan.



When assessing bleeding risk in patients with atrial fibrillation (AF), risk stratification is often based on the baseline risks. We aimed to investigate changes in bleeding risk factors and alterations in the HAS-BLED score in AF patients. We hypothesized that a follow-up HAS-BLED score and the 'delta HAS-BLED score' (reflecting the change in score between baseline and follow-up) would be more predictive of major bleeding, when compared with baseline HAS-BLED score.


A total of 19,566 AF patients receiving warfarin and baseline HAS-BLED score ≤2 were studied. After a follow-up of 93,783 person-years, 3,032 major bleeds were observed. The accuracies of baseline, follow-up, and delta HAS-BLED scores as well as cumulative numbers of baseline modifiable bleeding risk factors, in predicting subsequent major bleeding, were analysed and compared. The mean baseline HAS-BLED score was 1.43 which increased to 2.45 with a mean 'delta HAS-BLED score' of 1.03. The HAS-BLED score remained unchanged in 38.2% of patients. Of those patients experiencing major bleeding, 76.6% had a 'delta HAS-BLED' score ≥1, compared with only 59.0% in patients without major bleeding (p < 0.001). For prediction of major bleeding, AUC was significantly higher for the follow-up HAS-BLED (0.63) or delta HAS-BLED (0.62) scores, compared with baseline HAS-BLED score (0.54). The number of baseline modifiable risk factors was non-significantly predictive of major bleeding (AUC = 0.49).


In this 'real-world' nationwide AF cohort, follow-up HAS-BLED or 'delta HAS-BLED score' was more predictive of major bleeding compared with baseline HAS-BLED or the simple determination of 'modifiable bleeding risk factors'. Bleeding risk in AF is a dynamic process and use of the HAS-BLED score should be to 'flag up' patients potentially at risk for more regular review and follow-up, and to address the modifiable bleeding risk factors during follow-up visits.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Georg Thieme Verlag Stuttgart, New York
Loading ...
Support Center