1 INTRODUCTION

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia.1 It can lead to several health problems, such as heart failure, embolic events, and impaired quality of life. Moreover, it is associated with higher mortality rate.2-4 In case of drug-resistant symptomatic AF, catheter ablation proved to be an effective solution for rhythm control.5 However, depending on the ablation strategy and the type of AF, success rates of catheter ablation after 1 year vary considerably from 60% to 90%.5-7 Appropriate patient selection for catheter ablation is essential as neither AF recurrence nor procedural complication rates are negligible.8 However, there are no consistently confirmed predictors of AF recurrence following catheter ablation in the literature.9-18 It has been suggested that left atrial appendage (LAA) volume and function may be associated with the recurrence of AF/tachycardia in patients undergoing repeated ablation, the exact role of the LAA in the prediction of AF recurrence has not yet been resolved.19 The anatomy, including LAA volume, morphology and LAA orifice area can be accurately described using cardiac computed tomography (CT), and LAA function can be assessed by measuring LAA flow velocity using transesophageal echocardiography (TEE).

We aimed to study the role of LAA volume (LAAV), and function in the success of catheter ablation by type of AF.

2 METHODS 2.1 Study population

In our multimodality retrospective study, we included consecutive patients with symptomatic AF who underwent initial point-by-point radiofrequency catheter ablation at the Heart and Vascular Center of Semmelweis University, Budapest, Hungary between January 2014 and December 2017. All patients underwent preprocedural cardiac CT for the assessment of LA anatomy. LAA function was assessed with LAA flow velocity measured by TEE. Left ventricular ejection fraction (LVEF) was measured by transthoracic echocardiography using Biplane Simpson method. Exclusion criteria were age under 18 years, non-diagnostic CT image quality, history of prior pulmonary vein isolation or heart surgery, and missing echocardiography data on LAA flow velocity. Due to the retrospective analysis of clinically acquired data, the institutional review board waived the need for written patient informed consent. The study was performed according to the Declaration of Helsinki and Institutional Guidelines.

2.2 Cardiac CT imaging

CT examinations were performed with a 256-slice scanner (Brilliance iCT 256, Philips Healthcare, Best, The Netherlands) with prospective ECG-triggered axial acquisition mode. 100–120 kV with 200–300 mAs tube current was used depending on patient anthropometrics. Image acquisition was performed with 128 × 0.625 mm detector collimation, and 270 ms gantry rotation time. For heart rate control 50–100 mg metoprolol was given orally and 5–20 mg intravenously, if necessary. In patients with a heart rate of <80/min, mid-diastolic triggering was applied with 3%–5% padding (73%–83% of the R-R interval), and in those with ≥80/min, systolic triggering was chosen (35%–45% of the R-R interval) regardless of the presence of AF at the ECG during CT examination. In total 85–95 ml contrast material (Iomeron 400, Bracco Ltd., Milan, Italy) was injected with a flow rate of 4.5–5.5 ml/s via antecubital vein access using a four-phasic injection protocol.20 Bolus tracking in the LA was used to obtain proper scan timing. All patients received 0.8 mg sublingual nitroglycerin between the native CT and CTA examinations. CT data sets were reconstructed with 0.8 mm slice thickness and 0.4 mm increments. The LA and LAA volumes were measured using a semiautomatic software tool (EP Planning, Philips IntelliSpace Portal, Philips Healthcare, Best, The Netherlands) and if needed the borders of LA and LAA, the orifices of the pulmonary veins and the level of the mitral valve were manually adjusted.

2.3 LAA flow velocity measurement

Maximum 24 h before ablation, all patients underwent TEE examination to exclude the presence of LAA thrombus. iE33 and Epiq 7C (Philips Medical System, Andover, MA) systems equipped with S5-1 phased array and X7-2t matrix TEE transducers were used. TEE was performed during conscious sedation. The LAA was imaged from 0°, 45°, 90°, and 135° views to detect spontaneous echo contrast, sludge or thrombus. Subsequently, a sample volume was placed at the middle portion of the LAA and the peak velocity of the outflow of the LAA was measured.

2.4 Catheter ablation procedure

The indications for AF ablation procedures were in accordance with the current guidelines.1, 21 Paroxysmal AF was defined as self-terminating AF, in most cases within 48 hours. Some AF paroxysms continued up to 7 days.21 Persistent AF was defined as AF that lasts longer than 7 days.21 Conscious sedation was carried out in all cases with intravenous fentanyl, midazolam, and propofol. Basic vital parameters of the patients were monitored in all cases with non-invasive blood pressure measurements every 10 min and continuous pulse oximetry. Femoral venous access was used for all procedures. Transseptal puncture was performed routinely with fluoroscopy guidance and pressure monitoring, while intracardiac echocardiography was also utilized in difficult cases. All ablations were performed with the support of an electroanatomical mapping system (either CARTO, Biosense Webster, Inc., Diamond Bar, CA, USA; or ENSITE, St. Jude Medical, Inc., MN, USA), and the LA fast anatomical map was fused with the cardiac CT images to guide ablation (temperature-controlled mode, 43°C, 25–35 W, irrigated 4 mm tip catheter) in the majority of patients. Pulmonary vein isolation was performed in each patient. Moreover, in patients with long-standing persistent AF, additional ablation lines were drawn at the discretion of the operating physician. All patients without complications were discharged the day after the procedure.

2.5 Follow-up and definition of AF recurrence

After discharge, outpatient clinical follow-up visits were scheduled at 3, 6, and 12 months after the procedure and at least once yearly thereafter. The follow-up visits included clinical assessment of the patient and 24-hour Holter ECG monitoring. Follow-up data were registered in the structured reporting platform (Axis, Neumann Medical Ltd, Budapest, Hungary). Recurrence of AF was defined as the occurrence of atrial tachyarrhythmia that lasted for more than 30 s with or without symptoms.1, 21 AF recurrences during the first 90 days after catheter ablation were not included in order to exclude AF during this vulnerable „blanking period”, which might be only a temporary phenomenon due to the inflammation, maturation and healing of the ablated lesions.22, 23

2.6 Statistical analysis

Categorical variables are expressed as frequencies (percentages) and continuous values are expressed as mean ± SD. Normality of continuous parameters was tested with Shapiro–Wilk test. Tests for significance were conducted using Mann–Whitney-Wilcoxon or Kruskal-Wallis tests for continuous variables and Pearson's chi-square or Fisher exact tests (in case of five or less observations) for categorical variables. The event-free survival rate was estimated using Kaplan–Meier method and log-rank test was applied for the comparisons between the various groups. Cumulative event rates were calculated with event or censoring times measured from the date of ablation. For patients who did not experience AF recurrence, their time-to-event measure was censored at the last follow-up visit date.

To identify parameters associated with AF recurrence after catheter ablation, uni- and multivariate Cox proportional hazard regression model was executed. In the multivariate model, adjustment was made for age > 65 years, persistent AF, impaired eGFR (<60 ml/min/1.73 m2), body surface area-indexed LA volume (iLAV) measured by CT, LVEF<50%), sex, obesity (defined as body mass index ≥ 30 kg/m2), hypertension, dyslipidemia, diabetes (Type I and II), prior stroke/transient ischemic attack, obstructive coronary artery disease, thyroid gland diseases (hypo- and hyperthyroidism), unsuccessful preablational anti-arrhythmic drug (AAD) therapy (including sotalol, propafenone and amiodarone therapies), and LAAV. Thirty patients were randomly selected for interobserver agreement and analyzed using intraclass correlation coefficient (ICC). Relative risks were expressed as hazard ratios (HRs) with associated 95% confidence intervals (CIs). Two-tailed p values smaller than .05 were considered significant. All statistical analyses were performed in R environment (version 3.6.1). Cox proportional hazard regression analysis was done using the 'survival' package (version 3.1–8). Kaplan–Meier curve and log-rank test were performed using the 'survminer' (version 0.4.6).

3 RESULTS 3.1 Patient characteristics

A total of 561 patients were included in the current analysis. Mean age was 61.9 ± 10.2 years and 34.9% of the patients were female. Recurrence of AF was reported in 40.8% of the patients (34.6% in patients with paroxysmal and 53.5% in those with persistent AF). Median recurrence-free time was 22.7 (9.3–43.1) months (21.8 [9.4–43.2] months in paroxysmal and 23.6 [9.0–42.6] months in persistent AF. An excellent interobserver agreement was obtained for both the iLAV (ICC = 0.99), and LAAV (ICC = 0.90) measurements. Correlation among LA and LAA parameters are reported in Figure S1. The proportion of individuals aged >65 years (40.7% vs. 49.3%; p = .046), female gender (30% vs. 41.9%; p = .005), persistent AF (25.9% vs. 43.2%; p < .001), and LVEF <50% (6.9% vs. 21.0%; p < .001) were significantly higher in patients with AF recurrence. Moreover, patients with AF recurrence had significantly higher iLAV (54.4 ± 19.3 ml/m2 vs. 61.8 ± 23.9 ml/m2; p < .001), LAAV (7.6 ± 3.2 ml vs. 8.8 ± 5.2 ml; p = .002) and LAA orifice area (387.6 ± 140.5 mm2 vs. 454.4 ± 167.7 mm2; p < .001). Anthropometric data, cardiovascular comorbidities, AAD therapy and imaging parameters are summarized in Table 1. Medications and procedural times are reported in Table S1.

TABLE 1. Patient characteristics All patients (n = 561) No AF recurrence (n = 332) AF recurrence (n = 229) p Anthropometric data and comorbidities Age > 65 years, n (%) 248 (44.2) 135 (40.7) 113 (49.3) .046 Female, n (%) 561 (34.9) 100 (30.1) 96 (41.9) .005 Persistent AF, n (%) 185 (33.0) 86 (25.9) 99 (43.2) <.001 Obesity, n (%) 187 (33.3) 112 (33.7) 75 (32.8) 0.856 Hypertension, n (%) 411 (73.3) 238 (71.7) 173 (75.5) 0.333 Hyperlipidemia, n (%) 143 (25.5) 86 (25.9) 57 (24.9) 0.844 Diabetes, n (%) 82 (14.6) 50 (15.1) 32 (14.0) 0.808 Obstructive CAD, n (%) 51 (9.1) 28 (8.4) 23 (10.0) 0.552 Stroke/TIA, n (%) 43 (7.7) 27 (8.1) 16 (7.0) 0.747 Thyroid gland disease, n (%) 56 (10.0) 36 (10.8) 20 (8.7) 0.475 eGFR<60 ml/min/1.73 m2 138 (24.6) 83 (25.0) 55 (24.0) 0.842 Imaging parameters LVEF<50%, n (%) 71 (12.7) 23 (6.9) 48 (21.0) <.001 iLAV (ml/m2) 57.4 ± 21.6 54.4 ± 19.3 61.8 ± 23.9 <.001 LAAV (ml) 8.1 ± 4.2 7.6 ± 3.2 8.8 ± 5.2 .002 LAA orifice area (mm2) 414.9 ± 155.6 387.6 ± 140.5 454.4 ± 167.7 <.001 LAA flow velocity (cm/s) 34.1 ± 13.0 34.1 ± 13.2 34.2 ± 12.9 0.965 Abbreviations: AAD, anti-arrhythmic drug; AF, atrial fibrillation; CAD, coronary artery disease; eGFR, estimated glomerular filtration rate; iLAV, body surface area-indexed left atrial volume; LAA, left atrial appendage; LAAV, left atrial volume; LVEF, left ventricular ejection fraction; TIA, transient ischemic attack.

We also examined the differences of the clinical and imaging parameters between patients with paroxysmal and persistent AF. Those patients with persistent AF had significantly higher proportion of age > 65 years (41.0% vs. 50.8%; p = .030), hypertension (67% vs. 85.9%; p < .001) and LVEF <50% (6.6% vs. 24.9%; p < .001). Regarding the CT parameters, we measured significantly higher iLAV (51.0 ± 15.9 ml/m2 vs. 70.4 ± 25.6 ml/m2; p < .001), LAAV (7.4 ± 3.0 ml vs. 9.5 ± 5.6 ml; p = .002), LAA orifice area (385.2 ± 132.8 mm2 vs. 475.2 ± 179.7 mm2; p < .001) and lower LAA flow velocity (35.3 ± 13.4 cm/s vs. 31.7 ± 12.0 cm/s; p < .001). Detailed data on the clinical and imaging parameters by AF type can be seen in Table 2 and Figure 1.

TABLE 2. Clinical characteristics by AF type Paroxysmal AF (n = 376) Persistent AF (n = 185) p Age > 65 years, n (%) 154 (41.0) 94 (50.8) .030 Female, n (%) 136 (36.2) 60 (32.4) 0.398 Obesity, n (%) 132 (35.1) 55 (29.7) 0.217 Hypertension, n (%) 252 (67.0) 159 (85.9) <.001 Hyperlipidemia, n (%) 86 (22.9) 57 (30.8) .050 Diabetes, n (%) 49 (13.0) 33 (17.8) 0.162 Obstructive CAD, n (%) 34 (9.0) 17 (9.2) 1.000 Stroke/TIA, n (%) 32 (8.5) 11 (5.9) 0.316 Thyroid gland disease, n (%) 46 (12.2) 10 (5.4) .011 eGFR<60 ml/min/1.73 m2 100 (26.6) 38 (20.5) 0.144 Pre-ablation AAD therapy, n (%) 196 (52.1) 87 (47.0) 0.281 LVEF<50%, n (%) 25 (6.6) 46 (24.9) <.001 Abbreviations: AAD, anti-arrhythmic drug; AF, atrial fibrillation; CAD, coronary artery disease; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; TIA, transient ischemic attack.

Comparison of LA and LAA parameters between patients with and without AF recurrence stratified by AF type. AF, atrial fibrillation; iLAV, body surface area-indexed left atrial volume; LAA, left atrial appendage

3.2 Predictors of AF recurrence

Significantly higher iLAV and LAAV values were measured in patients with persistent AF recurrences, and larger LAA orifice area values were measured both in paroxysmal and persistent recurrences, as reported in Figure 1. To explore the associations between the various examined parameters and AF recurrence, Cox proportional hazards regression analyses were performed, as stratified by AF type. In the univariate analysis, female sex (HR = 1.43; 95% CI = 1.01–2.02; p = .043) was significantly associated with AF recurrence in patients with paroxysmal AF, while in those with persistent AF LVEF <50% (HR = 2.07; 95% CI = 1.36–3.14; p < .001), iLAV (HR = 1.01; 95% CI = 1.00–1.02; p = .027), LAAV (HR = 1.07; 95% CI = 1.03–1.10; p < .001) and LAA orifice area (HR = 1.02; 95% CI = 1.00–1.03 per 10 mm2; p = .005) showed an association with AF recurrence.

After adjustment LVEF <50% (HR = 2.17; 95% CI = 1.38–3.43; p < .001) and LAAV (HR = 1.06; 95% CI = 1.01–1.12; p = .029) remained a significant predictor of AF recurrence in patients with persistent AF, while in paroxysmal AF no independent predictors could be identified in the multivariate analysis. Detailed results of the uni- and multivariate Cox regression analyses are reported in Table 3. Kaplan–Meier curves of AF recurrence-free survival in persistent AF stratified by LVEF and LAAV can be seen in Figure 2, Figures S2 and S3.

TABLE 3. Associates of AF recurrence in patients with paroxysmal AF Paroxysmal AF Persistent AF Unadjusted analysis Adjusted analysis Unadjusted analysis Adjusted analysis HR 95% CI p HR 95% CI p HR 95% CI p HR 95% CI p Age > 65 years, n (%) 1.01 0.71–1.42 0.967 1.02 0.69–1.51 0.916 1.08 0.72–1.60 0.722 0.90 0.55–1.47 0.684 Female, n (%) 1.43 1.01–2.02 .043 1.42 0.96–2.11 .078 1.33 0.89–2.00 0.163 1.35 0.82–2.22 0.233 Obesity, n (%) 1.08 0.75–1.54 0.685 1.09 0.75–1.59 0.651 1.07 0.69–1.66 0.764 1.06 0.64–1.77 0.811 Hypertension, n (%) 0.96 0.66–1.39 0.816 0.91 0.60–1.37 0.648 1.03 0.59–1.78 0.921 1.13 0.59–2.13 0.717 Hyperlipidemia, n (%) 0.82 0.54–1.24 0.338 0.86 0.56–1.32 0.481 0.87 0.57–1.35 0.541 0.82 0.50–1.37 0.452 Diabetes, n (%) 0.77 0.45–1.33 0.353 0.79 0.45–1.39 0.481 0.92 0.54–1.55 0.745 1.10 0.61–1.97 0.752 Obstructive CAD, n (%) 1.08 0.62–1.89 0.781 1.28 0.70–2.33 0.426 0.98 0.49–1.95 0.961 1.64 0.71–3.75 0.246 Stroke/TIA, n (%) 0.88 0.47–1.62 0.673 0.96 0.50–1.82 0.890 0.77 0.31–1.89 0.567 0.43 0.16–1.15 .091 Thyroid gland disease, n (%) 0.94 0.55–1.58 0.806 0.85 0.48–1.49 0.565 0.87 0.32–2.37 0.787 0.64 0.22–1.89 0.418 eGFR<60 ml/min/1.73 m2 1.38 0.93–2.06 0.111 1.30 0.85–2.00 0.229 0.89 0.55–1.41 0.610 0.87 0.52–1.46 0.596 Pre-ablation AAD therapy, n (%) 0.96 0.68–1.36 0.836 1.00 0.70–1.43 0.983 0.97 0.65–1.44 0.868 0.89 0.57–1.39 0.598 LVEF<50%, n (%) 1.67 0.96–2.91 .069 1.42 0.80–2.52 0.232 2.07 1.36–3.14 <.001 2.17 1.38–3.43 <.001 iLAV (ml/m2) 1.01 0.99–1.02 .098 1.01 0.99–1.02 0.330 1.01 1.00–1.02 .027 1.00 0.99–1.01 0.549 LAAV (ml) 1.03 0.97–1.09 0.325 1.00 0.93–1.06 0.889 1.07 1.03–1.10 <.001 1.06 1.01–1.12 .029 LAA orifice area (mm2), per 10 mm2 1.01 1.00–1.03 .034 1.00 1.00–1.00 0.226 1.02 1.00–1.03 0.005 1.00 1.00–1.01 0.717 LAA flow velocity (cm/s) 1.00 0.99–1.03 0.352 1.01 1.00–1.02 .079 1.00 0.98–1.02 0.919 1.00 0.98–1.02 0.812 Abbreviations: AAD, anti-arrhythmic drug; AF, atrial fibrillation; CAD, coronary artery disease; eGFR, estimated glomerular filtration rate; iLAV, body surface area-indexed left atrial volume; LAA, left atrial appendage; LAAV, left atrial volume; LVEF, left ventricular ejection fraction; TIA, transient ischemic attack. Statistically significant p-values were marked bold.

Adjusted AF recurrence-free survival according to LVEF and LAAV in patients with persistent AF. The lines represent the marginal means of the estimated AF recurrence-free survival. Marginal means were estimated from the Cox proportional hazard regression model adjusted for age > 65 years, impaired eGFR, iLAV, sex, obesity, hypertension, dyslipidemia, diabetes, prior stroke/TIA, obstructive CAD, thyroid gland diseases and unsuccessful preablational AAD therapy. Median of LAAV was used as cut-off value. AF, atrial fibrillation; LAAV, left atrial appendage volume; LVEF, left ventricular ejection fraction

4 DISCUSSION

We demonstrated that beyond impaired LVEF, a larger LAAV is an independent predictor of AF recurrence after catheter ablation in patients with persistent AF. Interestingly, this association was not present in patients with paroxysmal AF.

AF is a complex disease with incompletely understood mechanisms. Although significant progress has been made in the last two decades, the efficacy of ablation therapy remains suboptimal, particularly in persistent AF. One-year success rate of catheter ablation varies between 60% and 90%.5-7 Previous studies have shown that the majority of AF recurrence occurred in the first 2 years after catheter ablation.24 So far, persistent AF, LA enlargement, hypertension, diabetes mellitus, aging, obesity, heart failure, chronic renal insufficiency and preprocedural amiodarone failure have been reported as independent predictors of AF recurrence.1, 21, 24-27 However, the data are controversial and the conclusions of previous studies are inconsistent. Several studies aimed to investigate the role of different scoring systems in the prediction of rhythm outcomes after AF ablation. While the HATCH score was found to have no value in the prediction of AF recurrence after catheter ablation, R2CHADS2 and CHA2DS2-VASc scores were associated with rhythm outcomes.28, 29 Since APPLE score proved to be superior to the CHA2DS2-VASc score for the prediction of rhythm outcome after catheter ablation, we incorporated its factors into our multivariable models.30 Due to inconsistent definition of recurrence, estimation of the AF ablation success is challenging.