Catheter ablation (CA) is commonly used to obtain rhythm control in patients with atrial fibrillation (AF), yet AF recurrence is frequent. In this study, we hypothesized that measures of left atrial (LA) structure and function modified the treatment effect of amiodarone to prevent AF recurrence, which has otherwise not proven beneficial. This was a post hoc analysis of the double-blinded AMIO-CAT trial. Patients undergoing CA were randomized to short-term amiodarone treatment or placebo. All patients underwent echocardiography to assess LA volumes and strain. The primary endpoint was AF recurrence at 6-months. Secondary endpoints were cardioversion and AF-related hospitalization during follow-up. Of 212 patients, 108 were randomized to amiodarone and 104 to placebo. 206 patients were eligible for analyses of the primary endpoint, of whom 90 (44%) developed AF recurrence. No measure of LA size or function significantly modified the effect of amiodarone for preventing AF recurrence (p for interaction >0.05 for all measures). However, abnormal LA strain (<22.7%) significantly modified the treatment effect for the secondary outcomes of cardioversions (p for interaction = 0.013) and AF-related hospitalization (p for interaction = 0.014). In patients with abnormal LA strain, amiodarone significantly reduced the risk of cardioversions (OR 0.35 [0.15 to 0.80], p = 0.013) and AF-related hospitalization (OR 0.31 [0.14 to 0.73], p = 0.007) compared to placebo. In conclusion, LA measures did not modify the treatment effect of amiodarone versus placebo for preventing AF recurrence at 6 months in patients undergoing CA. However, in patients with abnormal LA strain, amiodarone may reduce cardioversions and AF-related hospitalizations as compared to placebo.
Clinical trial registration: Clinicaltrials.gov unique identifier: NCT00826826
Graphical abstract
Created in Biorender. Biering-Sørensen, T. (2025) https://Biorender.com/o209s7V .
Atrial fibrillation (AF) is the most frequent tachyarrhythmia, with an estimated lifetime risk of approximately 30%. Catheter ablation has the potential to provide long-term rhythm control in selected patients, ,,, but recurrences are frequent, prompting repeat ablation or transition to alternative management strategies. , The use of concomitant antiarrhythmic drug (AAD) for a short-term has been a point of focus in trials as an adjuvant to reduce AF recurrence following ablation, albeit no AAD has proven effective in this regard. , The recurrence of arrhythmia following short-term oral AMIOdarone after CATheter ablation for atrial fibrillation (AMIO-CAT) trial investigated whether short-term treatment with amiodarone would reduce recurrence of AF at 6 months. Although the trial did not reveal a benefit for the primary outcome, it may be possible that amiodarone could be effective in certain subgroups. In particular, the degree of left atrial (LA) remodeling was not considered an exclusion criterion in the trial. LA remodeling perpetuates AF, and measures of LA size and function may accordingly affect the potential treatment effect of amiodarone for suppressing recurrence. With these considerations in mind, we performed a post hoc analysis to investigate whether echocardiographic measures of LA structure and function impacted the treatment effect of amiodarone for preventing AF recurrence. The hypothesis was that amiodarone would be more effective in suppressing AF recurrence in patients with abnormal LA size and function, who are more prone to recurrences, as compared to those without LA abnormalities.
Methods
Design and population
This was a post hoc analysis of the AMIO-CAT trial. The AMIO-CAT trial was an investigator-initiated, randomized, double-blind, placebo-controlled, clinical trial that randomized patients referred for catheter ablation in a 1:1 ratio to either placebo or amiodarone. Patients were included from 2 centers: Rigshospitalet, Copenhagen, Denmark from February 2009 through July 2013, and Gentofte Hospital, Hellerup, Denmark from January 2011 through July 2013. Follow-up was completed in January 2014. Treatment was initiated on the night of ablation and continued for 8 weeks. The dosing regimen is specified in the main trial paper. Inclusion criteria were: Paroxysmal or persistent AF with a clinical indication for first-time or repeat ablation. Exclusion criteria were: Current treatment with amiodarone, contraindication or side effects to amiodarone, atrial arrhythmias other than AF or atrial flutter, long-term persistent AF, LVEF <35%, and significant valve disease. Additional details regarding trial design are available on Clinicaltrials.gov (unique identifier: NCT00826826) and in the main paper.
All participants provided oral and written informed consent. The trial was approved by the Danish Regional Ethics Committee, the Danish Medicine Agency, and the Danish Protection Agency.
Ablation
All patients received oral anticoagulation for at least 4 weeks before ablation and 3 months after ablation.
Three-dimensional mapping systems with either NavX or CARTO systems were used to guide radiofrequency ablation (RFA) to obtain pulmonary vein isolation, verified by either three-dimensional remapping or circular mapping catheters. Ablation for additional potential substrates was left to the discretion of the electrophysiologist. Patients with documented atrial flutter also underwent cavotricuspid isthmus block.
Monitoring and outcome
All patients participated in 4 study visits (prerandomization, at 1 month, at 3 months, and at 6 months). All study visits included a 12-lead ECG. In addition, the patients underwent 3-day Holter monitoring at 6 to 8 weeks after ablation and at 6 months for screening of AF recurrence (lasting >60 seconds). In case of symptoms, patients were instructed to seek medical attention for arrhythmia documentation and treatment if clinically indicated. Patients who were unable to undergo arrhythmia documentation despite symptoms underwent additional monitoring by either Holter or event recording.
The primary outcome was AF recurrence at 6 months of follow-up, not including the 3-month blanking period. Secondary outcomes included 1) cardioversions within 6 months, 2) and AF-related hospitalizations within 6 months, both of which included the blanking period.
Echocardiography
All patients underwent a transthoracic echocardiogram within 3 months prior to ablation. The echocardiographic substudy has previously been described. Patients included from Righospitalet underwent echocardiography with Philips iE 33 ultrasound machines, whereas patients included from Gentofte Hospital underwent echocardiography with GE Vivid 7 ultrasound machines. Conventional echocardiographic analyses were performed with Xcelera quantification software (Philips Healthcare, the Netherlands) for patients included from Rigshospitalet, and with EchoPAC BT 112 (GE Healthcare, Horten, Norway) for patients included from Gentofte Hospital. Left ventricular (LV) and LA strain measurements for all patients were performed with vendor-independent software (EchoInsight v. 2.2.0.x, Epsilon Imaging).
Conventional measurements were performed according to current guidelines. These included measurement of LV dimensions from the parasternal long-axis view, which were used to calculate the LV mass, subsequently indexed to body surface area. LA dimensions (LAd) were also measured in the parasternal long-axis view, perpendicular to the aortic sinus, at LV end-systole and end-diastole, corresponding to maximal and minimal LAd, respectively. LV ejection fraction was measured with the Simpson’s biplane method. Maximal and minimal LA volumes (LAVmax and LAVmin) were measured with the biplane area-length method at LV end-systole and end-diastole, respectively. These were indexed to body surface area. The total LA emptying fraction (LAEF) was calculated as (LAVmax-LAVmin)/LAVmax, and the LA expansion index (LAEI) was calculated as (LAVmax-LAVmin)/LAVmin.
LV speckle tracking was performed to acquire the global longitudinal strain after manual contouring the LV endocardial border in the 3 apical projections. LA speckle tracking was performed to acquire the LA reservoir strain after manual contouring of the LA endocardial surface in the apical 4-chamber and 2-chamber views. Analyses were performed with R-wave triggering. Of note, LA strain was feasible in 91% of patients (193/212). Abnormal LA measures were defined according to median values (cut-offs: maximal LAd: 4.4 cm, minimal LAd: 3.8 cm, LAVmax: 33.9 ml/m 2, LAVmin: 21.6 ml/m 2, LAEF: 32.9%, LAEI: 49.1%, LA reservoir strain: 22.7%).
Statistics
Continuous variables that exhibited Gaussian-distribution are presented as mean ± standard deviation, whereas those not showing Gaussian-distribution are presented as median with interquartile ranges (25% to 75%). Categorical variables are presented as total numbers and proportions. Comparisons were made according to randomization and according to LA remodeling (defined as LAVmax >34 ml/m 2). Comparisons were made with the Student’s t -test, Wilcoxon Rank-Sum test, and the Chi 2 as appropriate.
For outcome-based analyses, logistic regression was applied to investigate the association between echocardiographic variables and the endpoints.
Treatment effect for the outcomes were assessed by testing whether LA measures modified the association between randomization and outcome in a logistic regression model.
Restricted cubic spline curves were created to visualize the treatment effect of randomization according to LA measures on a continuous scale. This allowed for testing and visualization of any nonlinear associations, with the number of knots chosen based on the model that yielded the lowest Akaike information criterion.
All analyses were performed according to the intention to treat principle. For the primary endpoint, 206 patients underwent complete follow-up and were eligible for analyses. All patients were included in the analyses of the secondary endpoints.
All statistical analyses were performed with STATA/SE 15.1 (StataCorp LP, Texas, USA). A p-value <0.05 was considered significant in two-tailed tests.
Results
Of the 212 patients who underwent randomization, 104 were randomized to placebo and 108 were randomized to amiodarone treatment. As previously described, the groups were well-balanced in terms of clinical characteristics, LVEF, maximal LA diameter, and LAVmax. By extension, no differences in extended LA measures were noted across randomization groups ( Table 1 ). Of the 212 patients, 106 (50%) had an enlarged LA and 106 (50%) had normal-sized LA. Clinical and echocardiographic characteristics stratified by LA dilatation is shown in Table 2 . Briefly, patients with LA dilatation less frequently had a history of stroke or transient ischemic attack and fewer had undergone previous ablation. By echocardiography, those with LA dilatation had larger LV size and mass, and abnormalities in all LA measures, albeit not statistically significant differences in LAEF and LAEI.
Table 1
Baseline echocardiographic measures according to randomization
|
Placebo
n: 104 |
Amiodarone
n: 108 |
p-value | |
|---|---|---|---|
| Left ventricular internal diameter, cm | 4.9 ± 0.6 | 4.9 ± 0.6 | 0.51 |
| Left ventricular mass index g/m 2 | 84 (71, 101) | 77 (67, 90) | 0.017 |
| Left ventricular ejection fraction, % | 50 ± 8 | 50 ± 8 | 0.99 |
| Global longitudinal strain, % | 15.6 ± 3.4 | 16.2 ± 3.5 | 0.24 |
| Maximal left atrial dimension, cm | 4.4 ± 0.7 | 4.4 ± 0.7 | 0.95 |
| Minimal left atrial dimension, cm | 3.8 ± 0.7 | 3.8 ± 0.7 | 0.84 |
| Maximal left atrial volume, ml/m 2 | 33.8 (26.9, 41.1) | 34.1 (26.6, 42.6) | 0.71 |
| Minimal left atrial volume, ml/m 2 | 21.6 (16.9, 29.9) | 21.7 (15.8, 30.4) | 0.65 |
| Left atrium emptying fraction, % | 33.5 (21.7, 46.1) | 32.6 (22.5, 42.8) | 0.98 |
| Left atrial expansion index, % | 50.4 (27.7, 85.6) | 48.5 (29.1, 74.9) | 0.98 |
| Left atrial reservoir strain, % | 22.7 (18.1, 29.6) | 22.9 (15.5, 28.0) | 0.29 |
Table 2
Baseline characteristics according to left atrial remodeling
|
Maximal left atrial volume <34 ml/m
2
n: 106 |
Maximal left atrial volume ≥34 ml/m
2
n: 106 |
p-value | |
|---|---|---|---|
| Age, years | 61 (54, 67) | 62 (54, 66) | 0.94 |
| Female sex, n (%) | 18 (17) | 18 (17) | 1.00 |
| Body mass index, kg/m 2 | 26.2 ± 3.8 | 27.3 ± 4.6 | 0.054 |
| Atrial fibrillation duration, months | 60 (24, 120) | 57 (24, 120) | 0.88 |
| Randomization to amiodarone, n (%) | 53 (50) | 55 (52) | 0.78 |
| Previous antiarrhythmic drugs | 1 (1, 2) | 1 (1, 2) | 0.88 |
| Previous atrial fibrillation ablation, n (%) | 38 (36) | 23 (22) | 0.023 |
| Persistent atrial fibrillation, n (%) | 46 (43) | 59 (56) | 0.07 |
| Hypertension, n (%) | 48 (45) | 36 (34) | 0.09 |
| Coronary artery disease, n (%) | 9 (9) | 5 (5) | 0.27 |
| Previous transient ischemic attack or stroke, n (%) | 10 (10) | 3 (3) | 0.045 |
| Diabetes mellitus, n (%) | 8 (8) | 10 (9) | 0.62 |
| History of typical atrial flutter, n (%) | 12 (11) | 17 (16) | 0.32 |
| Sleep apnoea, n (%) | 2 (2) | 2 (2) | 1.00 |
| Pacemaker, n (%) | 3 (3) | 3 (3) | 1.00 |
| Implantable cardioverter-defibrillator, n (%) | 0 (0) | 1 (1) | 0.32 |
New York Heart Association class, n (%)
|
79 (75) 27 (26) |
72 (68) 34 (32) |
0.29 |
CHA
2
DS
2
-VASc score
|
39 (37) 26 (25) 23 (22) 7 (7) 5 (5) 4 (4) 1 (1) 1 (1) |
47 (44) 23 (22) 24 (23) 11 (11) 1 (1) 0 (0) 0 (0) 0 (0) |
0.16 |
| Statins, n (%) | 34 (32) | 30 (28) | 0.55 |
| Angiotensin converting enzyme inhibitor or angiotensin receptor blocker, n (%) | 41 (39) | 34 (32) | 0.31 |
Antiarrhythmic drugs at inclusion, n (%)
|
59 (56) 14 (13) 14 (13) 23 (22) 1 (1) 13 (12) |
68 (64) 12 (11) 14 (13) 25 (24) 2 (2) 13 (12) |
0.21 0.68 1.00 0.74 0.50 1.00 |
| Echocardiography | |||
| Left ventricular internal diameter, cm | 4.8 ± 0.5 | 5.0 ± 0.5 | 0.001 |
| Left ventricular mass index, g/m 2 | 77 (66, 87) | 87 (72, 102) | <0.001 |
| Left ventricular ejection fraction, % | 50 ± 8 | 51 ± 8 | 0.76 |
| Global longitudinal strain, % | 16.3 ± 3.3 | 15.6 ± 3.6 | 0.16 |
| Maximal left atrial diameter, cm | 4.1 ± 0.6 | 4.7 ± 0.6 | <0.001 |
| Minimal left atrial diameter, cm | 3.5 ± 0.6 | 4.1 ± 0.7 | <0.001 |
| Maximal left atrial volume, ml/m 2 | 26.9 (23.0, 30.4) | 41.3 (37.1, 47.7) | <0.001 |
| Minimal left atrial volume, ml/m 2 | 16.8 (13.7, 20.9) | 30.0 (23.7, 34.7) | <0.001 |
| Left atrium emptying fraction, % | 35.8 (24.6, 45.5) | 30.3 (20.7, 42.8) | 0.055 |
| Left atrial expansion index, % | 55.7 (32.6, 83.5) | 43.5 (26.1, 74.9) | 0.055 |
| Left atrial reservoir strain, % | 25.3 (19.3, 30.4) | 19.8 (14.6, 27.4) | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
