Continuation or discontinuation of oral anticoagulants after an increase in HAS-BLED scores and the risks of clinical outcomes in patients with atrial fibrillation. a nationwide cohort study

Abstract Background The bleeding risk profile of patients with atrial fibrillation (AF) may change over time and the increment of HAS-BLED score (a validated bleeding risk score) is perceived to result in discontinuations of oral anticoagulants (OACs). The goal of the present study was to investigate the changes of HAS-BLED scores of AF patients initially with a low bleeding risk (HAS-BLED 0–2). Second, the impact of continuation or discontinuation of OACs on clinical outcomes after patients' bleeding risk profile worsened (ie HAS-BLED increased) were also studied. Methods The present study used the Taiwan nationwide health insurance research (NHIRD) database. From year 2000 to 2015, a total of 24,990 AF patients aged >20 years with a CHA2DS2-VASc score >1 (males) or >2 (females) having a HAS-BLED score of 0–2 who were treated with OACs (warfarin or NOACs) were identified and followed up for changes of the HAS-BLED scores. The risks of clinical outcomes were compared between patients who continued or stopped OAC's once their HAS-BLED score increased to ≥3. Results Mean HAS-BLED score of study population continuously increased from 1.54 (SD 1.63) at baseline, to 3.33 (SD 1.36) at the end of follow up with increasing age and incident comorbidities (Figure A). At end of 1 year, 5,229 (20.9%) patients had an increment of their HAS-BLED score to ≥3, mainly due to newly diagnosed impaired renal function, occurrences of bleeding events and concurrent use of antiplatelet agents or NSAIDs. Among 4,777 patients who consistently had a HAS-BLED score > =3, 1,062 (22.2%) stopped their use of OACs. Patients who kept on OACs (n=3,715; 77.8%) even after their HAS-BLED score increased to ≥3 were associated with a lower risk of ischemic stroke (aHR 0.601), major bleeding (aHR 0.780), all-cause mortality (aHR 0.876) and any adverse events (aHR 0.748) (Figure B). Conclusions For patients whose HAS-BLED scores increased to ≥3, the continuation of OACs (which occurred in the majority of patients) was associated with better clinical outcomes. An increased HAS-BLED score in AF patients is not a reason to withhold OACs, but reminds physicians to correct modifiable bleeding risk factors and follow up patients more closely. The low risk of bleeding despite continuing OAC in high HAS-BLED score patients may reflect the early reviews and careful follow-up of these “high risk” patients (as recommended in guidelines). Funding Acknowledgement Type of funding source: None