Europace (2017) 0, 1–8 doi:10.1093/europace/eux265
Development and validation of a risk score for predicting atrial fibrillation recurrence after a first catheter ablation procedure – ATLAS score ~o Mesquita1†*, Anto nio Miguel Ferreira1,2†, Diogo Cavaco1,2, Joa Francisco Moscoso Costa1,2, Pedro Carmo1,2, Hugo Marques3, Francisco Morgado1, ao1,2 Miguel Mendes1, and Pedro Adrag~ 1 Cardiology Department, Hospital de Santa Cruz, Av. Prof. Reinaldo dos Santos, 2790-134 Carnaxide, Lisbon, Portugal; 2Cardiology Department, Hospital da Luz, Lisbon, Portugal; and 3Radiology Department, Hospital da Luz, Lisbon, Portugal
Received 5 April 2017; editorial decision 11 July 2017; accepted 12 July 2017
Several predictors of relapse after catheter ablation of atrial fibrillation (AF) have been established, but assessing each patient’s individual risk remains challenging. Our aim was to develop and validate a score to estimate the risk of AF recurrence after the first radiofrequency pulmonary vein isolation (PVI) procedure.
................................................................................................................................................................................................... Methods Independent predictors of AF relapse were identified retrospectively in a two-centre registry of 1934 patients who underwent a first PVI procedure. Using the Cox regression hazard ratios of designated variables, a risk score was and results
developed in a random sample of 50% of the patients (development cohort) and validated in the remaining (validation cohort) half. The accuracy and discriminative power of the predictive model were assessed in both subgroups. During a follow-up of 4.2 ± 2.7 years, 522 patients (27%) relapsed. Five independent predictors of AF recurrence were identified and included in the score: age >60 years (1 point), female sex (4 points), non-paroxysmal AF (2 points), current smoking (7 points) and indexed left atrial volume (1 point for each 10 mL/m2). The score showed good discriminative power (censored c-statistic of 0.75 in both cohorts). In the development group, AF relapse rates were 8, 11, and 17%/year for low (10 points), respectively (P < 0.001). In the validation group, AF recurrence rates were 8, 11, and 18%/year, respectively (P < 0.001).
................................................................................................................................................................................................... Conclusion A simple risk score to estimate the rate of AF recurrence after ablation was developed and validated. An external assessment of its usefulness as a patient selection tool seems warranted. 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏 䊏
Atrial fibrillation • Atrial fibrillation recurrence ablation • Prognosis • Risk score
Introduction Pulmonary vein isolation (PVI) is now a well-established treatment for selected patients with atrial fibrillation (AF).1 However, despite high early success rates, AF recurrence is common during follow-up, offsetting the potential benefits of the procedure.2–4 Identifying the patients who are unlikely to remain in sinus rhythm seems a logical way to improve success rates and avoid the
Pulmonary vein isolation
unnecessary risks and costs of ineffective ablations. While several predictors of AF relapse have been identified,2,3 assessing each patient’s individual risk of recurrence remains challenging, in part due to the absence of a risk stratification tool that would condense the prognostic information conveyed by several different variables. The purpose of this study was to develop and validate a score to assess the risk of AF recurrence after a first radiofrequency PVI procedure.
* Corresponding author. E-mail address: [email protected]
The first two authors contributed equally to the study.
C The Author 2017. For permissions, please email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. V
J. Mesquita et al.
What’s new? • Development and validation of a clinical risk score to estimate the risk of atrial fibrillation (AF) recurrence after a first catheter ablation procedure. • Identification of long-term independent predictors of AF recurrence in a large dataset of patients. • Recognition of left atrial volume as the main determinant of AF recurrence after catheter ablation.
Methods Patient population and study design All consecutive patients with symptomatic drug-refractory AF undergoing percutaneous PVI in two Portuguese centres (Hospital Santa Cruz, Carnaxide, Portugal; and Hospital da Luz, Lisbon, Portugal) were included in an observational registry, which we used for this retrospective study. The first centre enrolled patients between May 2005 and November 2015, and the second centre between June 2007 and September 2015. From a pool of 2836 procedures, we excluded patients with previous AF ablations (n = 428), and those without 3D quantification of LA volume neither by computed tomography (CT) nor electroanatomical mapping (n = 352). The latter was used as an exclusion criterion, since left atrial (LA) volume has been shown to be a powerful prognostic marker in this setting.5,6 Patients who were lost to follow-up less than 6 months after the procedure were also excluded (n = 122). The final study population (n = 1934) was randomized on a 1:1 ratio into development and validation cohorts. In the development cohort, Cox regression analysis was used to identify independent predictors of clinical and/or electrocardiographic AF recurrence. The b regression coefficients of the selected variables were then used to develop a risk score whose accuracy and discriminative power were evaluated in the validation cohort. Atrial fibrillation was categorized as paroxysmal if self-terminated in _7 days or required cardioversion, or long-standing persistent if AF was maintained for more than 12 months.1 The majority of patients (n = 1332, 69%) underwent a 64-slice cardiac CT scan less than 48 h before the ablation procedure for the assessment of pulmonary vein anatomy, measurement of LA volume, exclusion of thrombi, and integration with electroanatomical mapping.6 Left atrial volume was calculated by tracing the LA borders on CT images, excluding the pulmonary veins and the left atrial appendage. In patients where cardiac CT could not be performed (mainly for logistical reasons), LA volume was R (Biosense Webster Inc., Diamond Bar, CA, estimated from CARTOV USA) electroanatomical mapping at the time of ablation. Left atrial volumes were indexed to body surface area (BSA). In patients who had LA volume estimated by different methods (2D echocardiogram, cardiac CT and/or electroanatomical mapping), a correlation analysis was performed to assess comparison between all methods.
Pulmonary vein isolation protocol Pulmonary vein isolation was guided by electroanatomical mapping, using either NavXV (St Jude MedicalV Inc, St Paul, MN, USA) or CARTOV systems. The right femoral vein was used as the preferred vascular access, through which three catheter electrodes were introduced: (i) a decapolar catheter, advanced through the coronary sinus; (ii) a variable circular mapping catheter, placed in the pulmonary veins; and (iii) an irrigated-tip ablation catheter. Left atrial access was established by trans-septal puncture. Radiofrequency ablation was performed more than 5 mm from the R
PV ostia, with continuous lesions enclosing the left and right pairs of PV. At Hospital Santa Cruz, all patients underwent conventional manually guided ablation, while at Hospital da Luz a Niobe II magnetic navigation R Inc., St. Louis, MO, USA) was used. If typical atrial system (StereotaxisV flutter was previously documented or observed at the time of the procedure, an additional cavotricuspid isthmus (CTI) ablation was performed (n = 164 patients). The treatment was considered successful if complete electrophysiological PVI was achieved. When required, an electrical cardioversion was performed at the end of the procedure. Oral anticoagulation was resumed 6 h after the ablation, maintained for 6 months and then withdrawn or continued according to CHA2DS2-VASc criteria (CHADS2 before 2009). Patients on vitamin K antagonists with subtherapeutic international normalized ratio (INR) were kept on subcutaneous enoxaparin 1 mg/kg every 12 h until an adequate INR was achieved. As a general rule, class I/III antiarrhythmic drugs were maintained in all patients for the first 3 months after the procedure and then withdrawn if there was no AF recurrence. A proton pump inhibitor was also prescribed for the first month after the ablation.
Study endpoint and patient follow-up The study endpoint was AF recurrence, defined as symptomatic or documented AF and/or other atrial arrhythmias, after a 3-month blanking period. Symptomatic AF was defined as the presence of symptoms considered to be likely due to AF episodes. Documented AF was defined by the presence of at least one episode of AF lasting more than 30 s in any electrocardiogram (ECG), 24 h Holter monitoring or event-loop recording. The follow-up protocol comprised outpatient visits with 12-lead ECG and 24 h Holter monitoring on the 1st, 3rd, 6th, and 12th months postablation, followed by yearly assessments. Patients were encouraged to contact the department if they experienced symptoms of AF recurrence. Whenever clinical records were insufficient, a structured telephonic interview was conducted. Patients who were kept on antiarrhythmic drugs after the 3rd month of follow-up were not considered as failed ablation.
Statistical analysis Normally and non-normally distributed variables were expressed as mean ± standard deviation and median, respectively. Differences between groups were assessed using independent samples t-test and Fisher’s exact test for continuous and categorical variables, respectively. The study population was split into development and validation cohorts by generating a random number for each patient and assigning odd and even numbers to each of these groups. Univariate proportional-hazards Cox regression was used to identify predictors of time to AF recurrence in the development cohort. The following variables were assessed: age, sex, type of AF (paroxysmal vs. non-paroxysmal), indexed LA volume, body mass index, hypertension, diabetes, cigarette smoking, hypercholesterolaemia, known obstructive coronary artery disease, and left ventricular systolic dysfunction (ejection fraction