Anticoagulation therapy in elderly patients with atrial fibrillation: results from the Registry of Atrial Fibrillation To Investigate the Implementation of New Guidelines (RAFTING).

Background: Patients with atrial fibrillation aged 75 years or older have a CHA2DS2VASc score that dictates oral anticoagulants. We recorded physicians' anticoagulation attitudes in elderly patients with atrial fibrillation and assessed the impact of stroke and bleeding risk.

Methods: Atrial Fibrillation To Investigate the Implementation of New Guidelines , a countrywide prospective registry performed in Greece during 2010, a period when only vitamin-K antagonists (VKA) were available, enrolled 1127 patients with atrial fibrillation diagnosis during Emergency Departments visit in 31 representative hospitals; 807 patients had known atrial fibrillation and of those, 342 aged 75 years or older. We recorded preadmission anticoagulation treatment and associated it with clinical characteristics and stroke/bleeding risk.

Results: Patients on VKA (n = 207; 61%) were younger (81 ± 4 vs. 83 ± 5; P < 0.001) but no other significant differences were noticed, including mean CHA2DS2VASc (high: 2-4, very high: >4) or modified HASBLED (low: 0-2, high: >2) scores. VKA were prescribed in 65% of patients with very high CHA2DS2VASc score as compared with 55% of those with high score (P = 0.065). VKA were used equally in low or high-modified HASBLED score (61% vs. 59%; P = 0.78). The interaction between CHA2DS2VASc and HASBLED was significant (P < 0.001) in patients on VKA; in patients with low HASBLED, VKA use was similar in high versus very high CHA2DS2VASc score (58 vs. 64%), whereas in patients with high HASBLED, VKA use tended to be higher in very high versus high CHA2DS2VASc score (66 vs. 43%).

Conclusion: In this countrywide atrial fibrillation registry, 61% of elderly patients received VKA, a decision driven mainly by stroke risk. VKA use was not higher in patients with low bleeding risk.