Management of AF Patients with Prior ICH – Race Specific Recommendations for High-Risk Patients

April 5, 2016

Results of a large Taiwanese registry analysis could open discussion around avoidance of oral anticoagulants in post-ICH patients.

Every clinician has faced the conundrum of how best to manage the risk of thromboembolism in high-risk atrial fibrillation (AF) patients with prior intracranial hemorrhage (ICH):  aspirin, oral anticoagulant (OAC) or neither?  Sadly, many of these clinically important patients have been systematically excluded from AF trials. New findings from a large Taiwanese study, however, suggest that  warfarin may actually be beneficial for AF patients with prior ICH and CHA2DS2-Vasc score ≥6, results that could potentially change traditional thinking about avoiding OAC for fear of bleeding.

In this very large registry study of 307,640 Taiwanese patients with AF and a CHA2DS2-Vasc score ≥2, 4.2% of participants had prior ICH (subarachnoid hemorrhage, 12.3%; intracerebral hemorrhage, 68.6%; epidural hemorrhage, 2.5%; subdural hematoma, 12.6%; unspecified bleeding, 4.0%). Over ~3 years of follow-up, the rates of ischemic stroke in these 12,917 patients varied, depending on which of the three treatment groups they fell into: antiplatelet therapy alone (5.2%), warfarin (3.4%) or no treatment at all (5.8%). The rates of ICH were also variable: antiplatelet therapy alone (5.3%), warfarin (5.9%) or no treatment at all (4.2%) (Table). 

In total, the number needed to treat (NNT) was lower than the number needed to harm (NNH) (37 vs 56) for those with a CHA2DS2-Vasc score ≥6; this benefit was not observed in those with CHA2DS2-Vasc score <6. Untreated patients with prior ICH had a 5X higher risk of ICH than those without prior ICH. Authors also concluded that aspirin or other antiplatelet therapies should be avoided following ICH because they increase the risk of intracranial bleeding by 1.1% over no treatment at all without decreasing risk for ischemic stroke.   

Both the major limitation and major strength of this study lie in its representation of a racially homogenous cohort. Asians are known to be at higher risk for ICH than non-Asians and this study may help physicians who tend to avoid OAC in such patients change their practice--a strong, clinically relevant conclusion. However, because only Asians were represented, the generalizability of these conclusions is limited.



Antiplatelet (n=3552)

Warfarin (8.9%)


No treatment (n=8215)

Ischemic Stroke









Chao TF, Liu CJ, Liao JN, et al. The use of oral anticoagulants for stroke prevention in atrial fibrillation patients with history of intracranial hemorrhage.Circulation 2016; DOI: 10.1161/CIRCULATIONAHA.115.019794.