Abstract:
Introduction
The use of novel oral anticoagulants (NOAC) in patients undergoing catheter ablation of atrial fibrillation (AF) is as effective and safe as standard Warfarin therapy. Phenprocoumon is a vitamin K antagonist (VKA) with a long elimination half-life (T 1⁄2 110-130h) and this could further reduce the risk of thromboembolism due to more uniform plasma levels. On the other hand, this could lead to bleeding that is more difficult to control. Comparative studies between the vitamin K antagonist Phenprocoumon and the NOAC are lacking. In addition, the question arises whether uninterrupted therapy with NOAC is safe and effective. The aim of the study is to compare NOAC (Dabigatran, Apixaban and Rivaroxaban) with Phenprocoumon as well as to compare uninterrupted oral anticoagulation with interrupted anticoagulation.
Material and methods
We performed a retrospective observational study. Between January 2011 and May 2017, 2219 catheter ablations of atrial fibrillation were performed in 1735 consecutive patients with an average mean age of 63.3 years; 1488 (67%) procedures were performed in men. 929 procedures (42%) were performed under oral anticoagulation with Phenprocoumon, 697 (31%) under Dabigatran, 399 (18%) under Rivaroxaban and 194 (9%) under Apixaban. Immediately after the market launch of NOAC, catheter ablations of atrial fibrillation were performed only after discontinuation of NOAC (n 679, 30%), periprocedural anticoagulation was performed with a low molecular weight heparin. All patients were observed under in hospital conditions for at least 48 hours after catheter ablation and a neurological examination was performed daily. In the case of neurological deficits, whether transient or persistent, a neurologist was consulted and cerebral imaging was performed.
Results
A CHA2DS2-VASC score >2 was seen more frequently in patients with Phenprocoumon than in patients with NOAC (58% versus 42%, p<0.001). Also,
60
patients with renal failure were treated more frequently with a VKA than with a NOAC (12% vs 7% respectively, p<0.01).
During the in hospital stay following catheter ablation, 37 (1.6%) thromboembolic events occurred, including 23 transient ischaemic events.
The overall risk of thromboembolic events was significantly higher in patients treated with Phenprocoumon than in patients treated with NOAC (21 versus 16 cases, 2.2% versus 1.2%, p=0.04). In particular, stroke was more frequent in patients treated with Phenprocoumon as compared to patients treated with NOAC (1% vs. 0.3% respectively, p=0.02). Uninterrupted oral anticoagulation was associated with lower rates of thromboembolic events in the NOAC group compared to the Phenprocoumon group (0.5% vs. 2.6% respectively, p=0.05). Bleeding complications (n=285, 12.8%) after ablation occurred without difference in patients treated with Phenprocoumon (n 122, 13%) and NOAC (n 163, 12.6%). There was no difference for both groups when considering the rates for minor bleedings, non-major but clinically relevant bleedings as well as major bleedings. In summary, it came to very few major bleedings with 6% of total bleeding events. Most of the bleedings were minor ones (n 206, 72%). An intraprocedural cardioversion significantly increased the thromboembolic risk (OR 1.2, p=0.012).
Age and HAS-BLED score did not measurably have an impact on thromboembolic risk. When anticoagulation was interrupted, the risk for thromboembolic complications increased significantly (OR 2.1, p=0.013). The interruption of oral anticoagulation led to an increase of peri-procedural bleeding events in both groups with OR 2.4 (Phenprocoumon 9% versus 21%, NOAC 7% versus 18%, p=0.07). When a NOAC was interrupted the risk for bleedings increased with an OR of 2.3 (p=0.001).
Conclusions
In this observational study of consecutive patients undergoing catheter ablation of atrial fibrillation, oral anticoagulation with one of the NOACs reduced the risk of thromboembolic events compared to Phenprocoumon.
Despite the better protection against thromboembolic complications, bleeding events related to catheter ablation occurred with the same frequency when using an oral anticoagulation therapy with Phenprocoumon and with the NOACs, respectively. Uninterrupted oral anticoagulation was preferable to further reduce the risk of bleeding and /or thromboembolic events.