Poor anticoagulation relates to extended access times for cardioversion and is associated with long-term major cardiac and cerebrovascular events

Background Patients undergoing elective electrical cardioversion (ECV) for atrial fibrillation have a temporarily increased risk of thromboembolism. Current guidelines recommend adequate anticoagulation for ≥3 consecutive weeks precardioversion, i.e. consecutive INR values 2.0–3.0 in patients with vitamin K antagonists (VKA). We aimed to evaluate the occurrence and impact of subtherapeutic INRs precardioversion and to study factors associated with these unwanted fluctuations. Methods We recruited 346 consecutive patients undergoing elective ECV in the Maastricht University Medical Centre between 2008 and 2013. Predictors of subtherapeutic INR values were identified and incorporated into a logistic regression model. Results A subtherapeutic INR precardioversion occurred in 55.2% of patients. The only statistically significant predictor was VKA-naivety (Odds Ratio (OR) 4.78, 95% Confidence Interval (CI) 2.67–8.58, p<0.001). In patients with ≥1 subtherapeutic INR precardioversion, time from referral until cardioversion was 91.1±42.8days, compared to 41.7±26.6days (p<0.001) in patients without subtherapeutic INRs.No thromboembolic events occurred <30days after the ECV. Independent predictors for the combined endpoint of cardiovascular death, ischemic stroke and the need of blood transfusion (n=30, median follow-up of 374days) were coronary artery disease in the history (OR 3.35, 95%CI 1.54–7.25, p=0.002) and subtherapeutic INR precardioversion (OR 3.64, 95%CI 1.43–9.24, p=0.007). Conclusions The use of VKA often results in subtherapeutic INRs precardioversion and is associated with a significant delay until cardioversion, especially in patients with recent initiation of VKA therapy. Furthermore, subtherapeutic INR levels prior to ECV are associated with the combined endpoint of cardiovascular death, ischemic stroke and the need of blood transfusion.