Very Early Recurrence Predicts Late Recurrence In Patients After Atrial Fibrillation Catheter Ablation

BackgroundAtrial fibrillation is one of the most common arrhythmias and catheter ablation has been shown to be an effective treatment for patients with symptomatic atrial fibrillation.Early recurrences after atrial fibrillation ablation within 3 months are common and are associated with late recurrences.Invasive monitoring methods and frequent clinic visits are needed to evaluate the exact incidence and characteristics of early recurrences.However,Invasive monitoring methods are not widely available.Most of early recurrences occur within 7 days after ablation and the relationship between very early recurrences within 7 days and late recurrences remains unclear.ObjectiveTo investigate the relationship between very early recurrence and late recurrence after atrial fibrillation catheter ablation.MethodsData of patients with atrial fibrillation for first catheter ablation from March 1st 2016 to August 30th 2018 in were analyzed retrospectively.Very early recurrence(VER)was evaluated by continuous noninvasive electrocardiography monitoring for 7 days after ablation.All patients were followed up in atrial fibrillation clinic at an interval of every 3 months.The risk factors of VER were analyzed by univariate and multivariate Logistic regression analysis and the association between VER and late recurrence(LR)was analyzed by univariate and multivariate Cox regression analysis.ResultsEighty-eight patients with paroxysmal or persistent atrial fibrillation(paroxysmal 65.90%,female 39.80%,mean age 62.86±9.69 years)were included.Twenty-seven patients(30.7%)experienced VER.VER occurred 5.41±2.02 days after ablation and 62.96%of VER occurred in multiple days.The median atrial fibrillation burden of VER was 9.16%.Left ventricular ejection fraction was lower in VER patients but without statistical significance(64.11±6.90%vs 67.44±8.06%,P=0.066).Multivariate Logistic regression analysis shown that VER risk was lower in patients with coronary artery heart disease(OR 0.113,95%CI 0.014-0.927,P=0.042).After a mean follow-up of 539.4±211.7 days,17 patients(19.3%)experienced LR.Eight(29.63%,8/27)and 9(14.75%,9/61)patients experienced LR among patients with VER and without VER respectively.Multivariate Cox regression analysis revealed that duration of atrial fibrillation(HR 1.009,95%CI 1.001-1.016;P=0.025),diabetes mellitus(HR 3.471,95%CI 1.243-9.690;P=0.018)and VER was independent predictor of LR(HR 4.884,95%CI 1.664-14.337;P=0.004)and the similar associations were shown in low burden VER(HR 6.470,95%CI 1.745-23.981;P=0.005),high burden VER(3.858,95%CI 1.045-14.249;P=0.043),early VER(HR 10.038,95%CI 2.273-44.3332,P=0.002)and late VER(HR 3.674,95%CI 1.089-12.394,P=0.036).VER predicted LR during follow-up with a sensitivity of 47.06%and a specificity of 73.24%ConclusionVER after catheter ablation was associated with LR regardless of the burden and the period of recurrent atrial fibrillation.