PT  - JOURNAL ARTICLE
AU  - Geurink, Kyle
AU  - Holmes, DaJuanicia
AU  - Ezekowitz, Michael D
AU  - Pieper, Karen
AU  - Fonarow, Gregg
AU  - Kowey, Peter R
AU  - Reiffel, James A
AU  - Singer, Daniel E
AU  - Freeman, James
AU  - Gersh, Bernard J
AU  - Mahaffey, Kenneth W
AU  - Hylek, Elaine M
AU  - Naccarelli, Gerald
AU  - Piccini, Jonathan P
AU  - Peterson, Eric D
AU  - Pokorney, Sean D
TI  - Patterns of oral anticoagulation use with cardioversion in clinical practice
AID  - 10.1136/heartjnl-2019-316315
DP  - 2021 Apr 01
TA  - Heart
PG  - 642--649
VI  - 107
IP  - 8
4099  - http://heart.bmj.com/content/107/8/642.short
4100  - http://heart.bmj.com/content/107/8/642.full
SO  - Heart2021 Apr 01; 107
AB  - Background Cardioversion is common among patients with atrial fibrillation (AF). We hypothesised that novel oral anticoagulants (NOAC) used in clinical practice resulted in similar rates of stroke compared with vitamin K antagonists (VKA) for cardioversion.Methods Using the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation II, patients with AF who had a cardioversion, follow-up data and an AF diagnosis within 6 months of enrolment were identified retrospectively. Clinical outcomes were compared for patients receiving a NOAC or VKA for 1 year following cardioversion.Results Among 13 004 patients with AF, 2260 (17%) underwent cardioversion. 1613 met the inclusion criteria for this analysis. At the time of cardioversion, 283 (17.5%) were receiving a VKA and 1330 (82.5%) a NOAC. A transoesophageal echocardiogram (TOE) was performed in 403 (25%) cardioversions. The incidence of stroke/transient ischaemic attack (TIA) at 30 days was the same for patients having (3.04 per 100 patient-years) or not having (3.04 per 100 patient-years) a TOE (p=0.99). There were no differences in the incidence of death (HR 1.19, 95% CI 0.62 to 2.28, p=0.61), cardiovascular hospitalisation (HR 1.02, 95% CI 0.76 to 1.35, p=0.91), stroke/TIA (HR 1.18, 95% CI 0.30 to 4.74, p=0.81) or bleeding-related hospitalisation (HR 1.29, 95% CI 0.66 to 2.52, p=0.45) at 1 year for patients treated with either a NOAC or VKA.Conclusions Cardioversion was a low-risk procedure for patients treated with NOAC, and there were statistically similar rates of stroke/TIA 30 days after cardioversion as for patients treated with VKA. There were no statically significant differences in death, stroke/TIA or major bleeding at 1 year among patients treated with NOAC compared with VKA after cardioversion.