Abstract
Aims
No prior study has been adequately powered to evaluate real-world safety outcomes in those receiving adjunctive ablation lesions beyond pulmonary vein isolation (PVI). We sought to evaluate characteristics and in-hospital complications among patients undergoing PVI with and without adjunctive lesions.
Methods and results
Patients in the National Cardiovascular Data Registry AFib Ablation Registry undergoing first-time atrial fibrillation (AF) ablation between 2016 and 2020 were identified and stratified into paroxysmal (PAF) and persistent AF, and separated into PVI only, PVI + cavotricuspid isthmus (CTI) ablation, and PVI + adjunctive (superior vena cava isolation, coronary sinus, vein of Marshall, atypical atrial flutter lines, other). Adjusted odds of adverse events were calculated using multivariable logistic regression. A total of 50 937 patients [PAF: 30 551 (60%), persistent AF: 20 386 (40%)] were included. Among those with PAF, there were no differences in the adjusted odds of complications between PVI + CTI or PVI + adjunctive when compared with PVI only. Among persistent AF, PVI + adjunctive was associated with a higher risk of any complication [3.0 vs. 4.5%, odds ratio (OR) 1.30, 95% confidence interval (CI) 1.07–1.58] and major complication (0.8 vs. 1.4%, OR 1.56, 95% CI 1.10–2.21), while no differences were observed in PVI + CTI compared with PVI only. Overall, there was high heterogeneity in adjunctive lesion type, and those receiving adjunctive lesions had a higher comorbidity burden.
Conclusion
Additional CTI ablation was common without an increased risk of complications. Adjunctive lesions other than CTI are commonly performed in those with more comorbidities and were associated with an increased risk of complications in persistent AF, although the current analysis is limited by high heterogeneity in adjunctive lesion set type.
Keywords: Atrial fibrillation, Ablation, Pulmonary vein isolation, Adjunctive lesions, Paroxysmal, Persistent, Outcomes, Registry, Complications
Graphical Abstract
Graphical Abstract.
What’s new?
In 50 937 patients undergoing atrial fibrillation (AF) ablation, adjunctive lesions beyond pulmonary vein isolation (PVI) were performed in nearly half of the cohort with high heterogeneity.
Among those with paroxysmal AF, there was an increased risk of prolonged hospitalization, but no difference in the risk of complications in those with adjunctive lesions when compared with PVI only.
Those with persistent AF who undergo PVI plus adjunctive lesions were at a higher risk of procedural complications than those with PVI alone, but no difference in risk was observed with PVI plus cavotricuspid isthmus ablation alone.
Introduction
Percutaneous catheter ablation for atrial fibrillation (AF) leads to a significant improvement in the quality of life, reduction in hospitalizations, decreased AF burden, and may increase the chances of survival in those with systolic heart failure.1–3 Pulmonary vein isolation (PVI) has remained the cornerstone of AF ablation based on landmark data demonstrating triggers originating from the PVs, although atrial arrhythmia recurrence rates approach 40% in paroxysmal AF (PAF) and nearly half in those with persistent AF.4,5 Several trials have evaluated various adjunctive lesions beyond PVI with the goal to improve AF-free survival with mixed results, including left atrial linear ablation, complex fractionated electrogram ablation, vein of Marshall ethanol ablation, left atrial appendage isolation, and posterior wall isolation.6–9 Although limited trial data have failed to lead to a Class I indication for a non-PVI ablation strategy per the professional society guidelines, adjunctive lesions are still commonly performed in the real world with unclear procedural risk due to underpowered studies.5,10
Using data from the National Cardiovascular Data Registry (NCDR) AFib Ablation Registry, the present study evaluated the differences among patients stratified into PAF and persistent AF undergoing PVI only, PVI plus cavotricuspid isthmus (CTI) ablation, and PVI plus adjunctive lesions in terms of patient characteristics and in-hospital complications.
Methods
Data source
The patients included in this study were enrolled the NCDR AFib Ablation Registry. Briefly, the American College of Cardiology launched the NCDR AFib Ablation Registry to assess the prevalence, demographics, management, and outcomes of patients undergoing percutaneous catheter ablation procedures to manage AF in the USA. Details have been described in the original publication of the NCDR AFib Ablation Registry.11 Briefly, the voluntary registry began collection of approximately 230 data elements from index hospitalization beginning in January 2016. A link to the full data collection forms for the index hospitalization is publicly available.12 Data are collected by sites at discharge. The NCDR Data Quality Reporting process has been designed to ensure that submissions are complete, valid, and accurate. It involves an annual audit of ∼5% of sites that are randomly selected during which submitted data are compared with source documentation and billing data as well as evaluation of sites that are outliers with regards to adverse event rates.13 Waiver of written informed consent and authorization for NCDR studies were granted by Chesapeake Research Review Incorporated. The research in this study was conducted according to the Helsinki Declaration guidelines on human research.
Study population
Between 1 January 2016, and 31 December 2020, a total of 162 hospitals submitted data on 67 970 patients undergoing AF ablation. We identified a final cohort of 50 937 patients after exclusion of patients with prior surgical or percutaneous catheter ablation (n = 15 101), left atrial appendage thrombus (n = 227), atrioventricular node ablation with pacemaker implantation (n = 429), those labelled as permanent AF (n = 101), and those with missing values on AF type or adjunctive lesions (n = 1175).
The cohort was first stratified into PAF and persistent AF and then grouped into PVI only, PVI plus CTI ablation alone, and PVI plus adjunctive lesions with or without CTI ablation. Adjunctive lesions included superior vena cava (SVC) isolation, coronary sinus ablation, ligament/vein of Marshall ablation, atypical atrial flutter (AFL) lines, and other.
Outcomes
First, in-hospital events were compared among PVI only, PVI plus CTI, and PVI plus adjunctive lesions. Periprocedural information, death, hospital stay >1 day, any complication, and major complication were collected. Major complications included death, stroke, transient ischaemic attack, cardiac arrest, cardiac surgery, vascular injury requiring intervention, access site bleeding requiring transfusion, and pericardial effusion requiring intervention.
Statistical analysis
Baseline demographic and clinic factors are presented as numbers and percentages for categorical variables. Categorical variables were compared using the χ2 test, and continuous variables were compared using the Wilcoxon rank sum test or the t-test as appropriate. Missing dichotomous variables (yes/no) were treated as no, and missing continuous variables were imputed with the overall median value. For missing categorical variables, the most common category of each variable was imputed.
Descriptive, unadjusted outcomes were summarized by the numbers and percentages of events. Then, unadjusted and adjusted multivariable logistic regressions were used to obtain odds ratios (ORs) with 95% confidence intervals (CIs) for PVI plus CTI and PVI plus adjunctive lesions vs. PVI only (reference) for in-hospital outcomes (hospital stay >1 day, any complication, and major adverse event). All tests were two-sided and P ≤ 0.05 was considered statistically significant. Variables in the multivariable model were chosen based on both clinical risk-adjusted variable selection and backward elimination. Patient characteristics that differed between ablation strategies in the univariate analysis were entered into a logistic regression model with backward selection using a P ≤ 0.05 for removal during the selection process. The final covariates in the multivariable model included age, race/ethnicity, insurance status, body mass index, chronic lung disease, obstructive sleep apnoea, cardiomyopathy, CHA2DS2-VASc score, HAS-BLED score, preprocedural creatinine, prior typical AFL, warfarin use, hospital region, and teaching hospital. Potential confounding variables were well represented and collected as part of the NCDR AFib Ablation Registry. All analyses were performed with the SAS statistical package, version 9.4 (SAS Institute Inc.).
Results
Between 1 January 2016 and 31 December 2020, a total of 50 937 patients undergoing first-time AF ablation were enrolled in the NCDR AFib Ablation Registry.
Baseline characteristics for those with paroxysmal atrial fibrillation
Of the 30 551 patients with PAF, the ablation strategies included PVI only (n = 16 374, 54.8%), PVI plus CTI only (n = 8775, 28.7%), and PVI plus adjunctive lesions with or without CTI lesion (n = 5041, 16.5%). Baseline characteristics of the PAF cohort are shown in Table 1. Those with PVI plus adjunctive lesions were older, more likely to be female, more likely to have coronary artery disease, congestive heart failure, hypertension, and diabetes than those undergoing PVI only and PVI plus CTI. Those with PVI plus adjunctive lesions also had a slightly higher prevalence of AFL than PVI only and PVI plus CTI (55.0 vs. 13.7 vs. 40.9%, P < 0.0001). Also, PVI plus adjunctive lesions had a higher prevalence of prior attempts at AF termination, specifically with direct current cardioversion. Those with PVI plus adjunctive lesions were more likely to have moderately (16.0%) and severely enlarged left atrial size (10.2%) on echocardiogram. Slightly less than half of the patients were prescribed a direct oral anti-coagulant and warfarin was prescribed in over 5% of patients prior to the procedure. Those with PVI plus adjunctive lesions were less likely to be prescribed a preprocedural anti-arrhythmic drug than those undergoing PVI only and PVI and CTI (31.9 vs. 41.1 vs. 40.5%, respectively).
Table 1.
Baseline characteristics of the paroxysmal atrial fibrillation cohort
| PVI only (N = 16 735) | PVI + CTI (N = 8775) | PVI + adjunctive (N = 5041) | P-value | |
|---|---|---|---|---|
| Age | 63.9 (11.0) | 64.8 (10.3) | 65.7 (10.9) | <0.0001 |
| Sex, male | 9894 (59.1%) | 5710 (65.1%) | 2935 (58.2%) | <0.0001 |
| Race | ||||
| White | 15 698 (93.8%) | 8183 (93.3%) | 4643 (92.1%) | 0.0001 |
| Black | 578 (3.5%) | 349 (4.0%) | 234 (4.6%) | 0.0003 |
| Asian | 236 (1.4%) | 107 (1.2%) | 80 (1.6%) | 0.1876 |
| Hispanic | 666 (4.0%) | 271 (3.1%) | 126 (2.5%) | <0.0001 |
| Other | 317 (1.9%) | 169 (1.9%) | 129 (2.6%) | 0.0103 |
| Insurance payer | ||||
| Private | 13 288 (79.4%) | 6922 (78.9%) | 3775 (74.9%) | <0.0001 |
| Medicare | 8020 (47.9%) | 4337 (49.4%) | 2779 (55.1%) | <0.0001 |
| Medicaid | 737 (4.4%) | 395 (4.5%) | 255 (5.1%) | 0.14 |
| State-specific plan | 271 (1.6%) | 152 (1.7%) | 116 (2.3%) | 0.005 |
| Other | 633 (3.8%) | 451 (5.1%) | 245 (4.9%) | <0.0001 |
| Patient history and risk factors | ||||
| Chronic lung disease | 1358 (8.1%) | 771 (8.8%) | 528 (10.5%) | <0.0001 |
| Coronary artery disease | 3380 (20.2%) | 1781 (20.3%) | 1192 (23.7%) | <0.0001 |
| Obstructive sleep apnoea | 4808 (28.7%) | 2601 (29.6%) | 1389 (27.6%) | 0.03 |
| Treatment | 3561 (74.9%) | 1916 (74.8%) | 1001 (73.0%) | 0.33 |
| Cardiomyopathy | 2020 (12.1%) | 1236 (14.1%) | 805 (16.0%) | <0.0001 |
| Non-ischaemic | 1094 (6.5%) | 717 (8.2%) | 409 (8.1%) | <0.0001 |
| Ischaemic | 467 (2.8%) | 294 (3.4%) | 210 (4.2%) | <0.0001 |
| Restrictive | 11 (0.1%) | 3 (0.0%) | 3 (0.1%) | 0.59 |
| Hypertrophic | 196 (1.2%) | 99 (1.1%) | 87 (1.7%) | 0.004 |
| Other | 317 (1.9%) | 169 (1.9%) | 129 (2.6%) | 0.01 |
| CHA2DS2-VASc score | 2.4 (1.6) | 2.5 (1.6) | 2.7 (1.7) | <0.0001 |
| Congestive heart failure | 2132 (12.7%) | 1255 (14.3%) | 872 (17.3%) | <0.0001 |
| NYHA Class I | 751 (4.5%) | 351 (4.0%) | 245 (4.9%) | 0.05 |
| NYHA Class II | 899 (5.4%) | 539 (6.1%) | 376 (7.5%) | <0.0001 |
| NYHA Class III | 254 (1.5%) | 182 (2.1%) | 157 (3.1%) | <0.0001 |
| NYHA Class IV | 31 (0.2%) | 24 (0.3%) | 14 (0.3%) | 0.26 |
| Left ventricular dysfunction | 864 (5.2%) | 517 (5.9%) | 362 (7.2%) | <0.0001 |
| Hypertension | 10 915 (65.2%) | 5822 (66.4%) | 3440 (68.3%) | 0.0003 |
| Diabetes | 2835 (17.0%) | 1709 (19.5%) | 1004 (19.9%) | <0.0001 |
| Stroke | 857 (5.1%) | 461 (5.3%) | 276 (5.5%) | 0.60 |
| Transient ischaemic attack | 663 (4.0%) | 358 (4.1%) | 213 (4.2%) | 0.69 |
| Thromboembolic event | 696 (4.2%) | 422 (4.8%) | 237 (4.7%) | 0.0 |
| Vascular disease | 2594 (15.5%) | 1366 (15.6%) | 922 (18.3%) | <0.0001 |
| Prior myocardial infarction | 1353 (8.1%) | 687 (7.8%) | 456 (9.0%) | 0.04 |
| Peripheral arterial disease | 491 (2.9%) | 277 (3.2%) | 181 (3.6%) | 0.06 |
| Known aortic plaque | 193 (1.2%) | 110 (1.3%) | 172 (3.4%) | <0.0001 |
| HAS-BLED score | 1.0 (0.9) | 1.1 (0.9) | 1.2 (0.9) | <0.0001 |
| Uncontrolled hypertension | 1378 (8.2%) | 771 (8.8%) | 494 (9.8%) | 0.002 |
| Abnormal renal function | 427 (2.6%) | 266 (3.0%) | 203 (4.0%) | <0.0001 |
| Abnormal liver function | 111 (0.7%) | 80 (0.9%) | 49 (1.0%) | 0.03 |
| Prior stroke | 670 (4.0%) | 357 (4.1%) | 222 (4.4%) | 0.45 |
| Prior bleeding | 606 (3.6%) | 295 (3.4%) | 185 (3.7%) | 0.51 |
| Labile INR | 122 (0.7%) | 69 (0.8%) | 59 (1.2%) | 0.009 |
| Alcohol use | 855 (5.1%) | 502 (5.7%) | 268 (5.3%) | 0.12 |
| Anti-platelet medication use | 1636 (9.8%) | 1016 (11.6%) | 727 (14.4%) | <0.0001 |
| Non-steroidal inflammatory drug use | 3416 (20.4%) | 1794 (20.5%) | 986 (19.6%) | 0.37 |
| Physical examination and labs | ||||
| Body mass index, kg/m2 | 30.8 (22.0) | 30.8 (8.4) | 30.6 (7.3) | 0.79 |
| Systolic blood pressure, mmHg | 133.4 (23.3) | 133.0 (23.5) | 133.9 (23.0) | 0.07 |
| Diastolic blood pressure, mmHg | 74.6 (13.8) | 75.0 (14.1) | 75.5 (13.3) | 0.0002 |
| Heart rate, beats/min | 68.4 (17.4) | 71.3 (20.2) | 73.1 (20.2) | <0.0001 |
| Creatinine | 1.0 (0.6) | 1.1 (0.6) | 1.1 (0.7) | <0.0001 |
| Bilirubin | 0.7 (0.9) | 0.7 (0.5) | 0.7 (1.1) | 0.28 |
| Arrhythmia history | ||||
| Symptomatic | 16 392 (98.0%) | 8569 (97.7%) | 4932 (97.8%) | 0.30 |
| Attempts at AFib termination | 8459 (50.6%) | 4467 (50.9%) | 2640 (52.4%) | 0.08 |
| Pharmacologic cardioversion | 5542 (65.5%) | 2824 (63.2%) | 1653 (62.6%) | 0.004 |
| DC cardioversion | 4543 (53.7%) | 2604 (58.3%) | 1544 (58.5%) | <0.0001 |
| Prior AFL | 2293 (13.7%) | 4819 (55.0%) | 2057 (40.9%) | <0.0001 |
| Typical | 1894 (82.6%) | 4440 (92.1%) | 1599 (77.7%) | <0.0001 |
| Pharmacologic cardioversion | 408 (2.4%) | 851 (9.7%) | 405 (8.0%) | <0.0001 |
| Direct current cardioversion | 468 (2.8%) | 836 (9.5%) | 486 (9.6%) | <0.0001 |
| Catheter ablation | 1312 (7.8%) | 421 (4.8%) | 666 (13.2%) | <0.0001 |
| Pre-procedure imaging | ||||
| Transoesophageal echocardiogram performed | 7417 (44.4%) | 4246 (48.5%) | 2193 (43.6%) | <0.0001 |
| Left atrial size | ||||
| Normal | 1719 (50.1%) | 883 (46.4%) | 551 (46.2%) | 0.26 |
| Mildly enlarged | 979 (28.5%) | 611 (32.1%) | 329 (27.6%) | 0.002 |
| Moderately enlarged | 499 (14.5%) | 263 (13.8%) | 191 (16.0%) | 0.011 |
| Severely enlarged | 234 (6.8%) | 148 (7.8%) | 122 (10.2%) | <0.0001 |
| Computed tomography performed prior | 8084 (90.6%) | 3989 (91.0%) | 2228 (90.8%) | 0.78 |
| Magnetic resonance imaging performed prior | 8909 (53.2%) | 4371 (49.8%) | 2445 (48.5%) | <0.0001 |
| Pre-procedure medications | ||||
| Anti-thrombotic therapy | ||||
| Warfarin | 995 (5.9%) | 563 (6.4%) | 378 (7.5%) | 0.0004 |
| Direct oral anti-coagulant | 7789 (46.5%) | 4086 (46.6%) | 2325 (46.1%) | 0.86 |
| Aspirin | 3653 (21.8%) | 2055 (23.4%) | 1221 (24.2%) | 0.0003 |
| Clopidogrel | 491 (2.9%) | 271 (3.1%) | 209 (4.1%) | <0.0001 |
| Prasugrel | 33 (0.2%) | 20 (0.2%) | 12 (0.2%) | 0.80 |
| Ticagrelor | 49 (0.3%) | 28 (0.3%) | 14 (0.3%) | 0.90 |
| Rate-control therapy | ||||
| Beta-blocker | 8381 (50.1%) | 4417 (50.3%) | 2460 (48.8%) | 0.19 |
| Digoxin | 367 (2.2%) | 249 (2.8%) | 157 (3.1%) | 0.0001 |
| Anti-arrhythmic therapy | 3732 (41.1%) | 2517 (40.5%) | 1623 (31.9%) | <0.0001 |
| Amiodarone | 1790 (25.9%) | 1123 (31.9%) | 501 (32.6%) | <0.0001 |
| Dofetilide | 683 (9.9%) | 202 (5.7%) | 98 (6.4%) | <0.0001 |
| Dronedarone | 622 (9.0%) | 266 (7.6%) | 152 (9.9%) | 0.004 |
| Flecainide | 1889 (27.4%) | 1093 (31.1%) | 402 (26.2%) | <0.0001 |
| Propafenone | 538 (7.8%) | 283 (8.0%) | 116 (7.6%) | 0.003 |
| Sotalol | 1426 (20.7%) | 566 (16.1%) | 284 (18.5%) | <0.0001 |
| Procedure information | ||||
| General anaesthesia | 16 081 (96.1%) | 8615 (98.2%) | 4759 (94.4%) | <0.0001 |
| Double transseptal | 5126 (30.6%) | 3406 (38.8%) | 2265 (44.9%) | <0.0001 |
| All veins present able to be isolated by PVI | 15 833 (94.6%) | 8426 (96.0%) | 4792 (95.1%) | <0.0001 |
| Assessed with circumferential vein catheter | 14 337 (90.8%) | 7693 (91.5%) | 4165 (87.1%) | <0.0001 |
| Isolation confirmation | ||||
| Entrance block | 2408 (14.4%) | 967 (11.0%) | 612 (12.1%) | <0.0001 |
| Exit block | 1264 (7.6%) | 504 (5.7%) | 198 (3.9%) | <0.0001 |
| Bidirectional block | 11 036 (65.9%) | 6322 (72.0%) | 3487 (69.2%) | <0.0001 |
| Atrial arrhythmia present during procedure | 1515 (9.1%) | 4395 (50.3%) | 2769 (55.2%) | <0.0001 |
| Cardioversion performed during procedure | 2540 (15.2%) | 1543 (17.6%) | 1371 (27.3%) | <0.0001 |
| Radiation dose | ||||
| Hospital characteristics | ||||
| Hospital region | ||||
| Northeast | 2130 (12.7%) | 753 (8.6%) | 369 (7.3%) | <0.0001 |
| West | 2839 (17.0%) | 1634 (18.6%) | 1034 (20.5%) | <0.0001 |
| Midwest | 4978 (29.7%) | 2097 (23.9%) | 1106 (21.9%) | <0.0001 |
| South | 6788 (40.6%) | 4291 (48.9%) | 2532 (50.2%) | <0.0001 |
| Location | ||||
| Rural | 840 (5.0%) | 599 (6.8%) | 372 (7.4%) | <0.0001 |
| Suburban | 5059 (30.2%) | 2294 (26.1%) | 1521 (30.2%) | <0.0001 |
| Urban | 10 836 (64.8%) | 5882 (67.0%) | 3148 (62.4%) | <0.0001 |
| Hospital type | ||||
| Government | 505 (3.0%) | 456 (5.2%) | 26 (0.5%) | <0.0001 |
| Private | 13 510 (80.7%) | 7068 (80.5%) | 4169 (82.7%) | 0.003 |
| University | 2720 (16.3%) | 1251 (14.3%) | 846 (16.8%) | <0.0001 |
| Teaching | 9853 (58.9%) | 4342 (49.5%) | 2629 (52.2%) | <0.0001 |
| Patient beds | 561.0 (266.5) | 548.5 (269.9) | 523.0 (243.4) | <0.0001 |
| Annual volume | 1856.7 (978.8) | 1823.3 (923.5) | 1834.8 (1020.6) | 0.03 |
CTI, cavotricuspid isthmus; PVI, pulmonary vein isolation; AFL, atrial flutter; INR, international normalized ratio; NYHA, New York Heart Association.
Baseline characteristics for those with persistent atrial fibrillation
Of the 20 386 patients with persistent AF, the ablation strategies included PVI only (n = 9076, 44.5%), PVI plus CTI only (n = 6222, 30.5%), and PVI plus adjunctive lesions (n = 5088, 25.0%). Baseline characteristics of the persistent AF cohort are shown in Table 2. Those with PVI plus adjunctive lesions in the persistent AF cohort were also older and had a higher prevalence of coronary artery disease, congestive heart failure, and hypertension. Those with PVI plus CTI reported a higher prevalence of prior history of AFL when compared with PVI only and PVI plus adjunctive lesions (45.8 vs. 13.4 vs. 30.6%, P < 0.0001). Prior attempts at AF termination occurred in ∼80% of each group with slightly higher prevalence in the PVI only group. Approximately 18% of patients had a severely enlarged left atrium with no differences across the groups. Also, approximately a half of the patients were prescribed a direct oral anti-coagulation while warfarin was prescribed in ∼6% prior to the procedure. Those with PVI plus adjunctive lesions were less likely to be prescribed a preprocedural anti-arrhythmic drug than PVI only and PVI and CTI (30.5 vs. 41.2 vs. 40.1%, respectively).
Table 2.
Baseline characteristics of the persistent atrial fibrillation cohort
| PVI only (N = 9076) | PVI + CTI (N = 6222) | PVI + adjunctive (N = 5088) | P-value | |
|---|---|---|---|---|
| Age | 65.7 (9.9) | 66.3 (9.6) | 67.3 (9.5) | <0.0001 |
| 19.0–100.0 | 21.0–98.0 | 20.0–90.0 | ||
| Sex, male | 6399 (70.5%) | 4481 (72.0%) | 3467 (68.1%) | <0.0001 |
| Race | ||||
| White | 8652 (95.3%) | 5848 (94.0%) | 4773 (93.8%) | <0.0001 |
| Black | 243 (2.7%) | 225 (3.6%) | 156 (3.1%) | 0.004 |
| Asian | 88 (1.0%) | 63 (1.0%) | 55 (1.1%) | 0.82 |
| Hispanic | 286 (3.2%) | 185 (3.0%) | 119 (2.3%) | 0.02 |
| Other | 497 (5.5%) | 339 (5.4%) | 245 (4.8%) | 0.20 |
| Insurance payer | ||||
| Private | 7148 (78.8%) | 4814 (77.4%) | 3812 (74.9%) | <0.0001 |
| Medicare | 4808 (53.0%) | 3414 (54.9%) | 2988 (58.7%) | <0.0001 |
| Medicaid | 409 (4.5%) | 309 (5.0%) | 232 (4.6%) | 0.38 |
| State-specific plan | 128 (1.4%) | 88 (1.4%) | 110 (2.2%) | 0.001 |
| Other | 340 (3.7%) | 329 (5.3%) | 296 (5.8%) | <0.0001 |
| Patient history and risk factors | ||||
| Chronic lung disease | 957 (10.5%) | 692 (11.1%) | 571 (11.2%) | 0.36 |
| Coronary artery disease | 2110 (23.3%) | 1530 (24.6%) | 1307 (25.7%) | 0.004 |
| Obstructive sleep apnoea | 3369 (37.1%) | 2137 (34.3%) | 1743 (34.3%) | 0.0002 |
| Treatment | 2561 (76.9%) | 1620 (76.6%) | 1295 (74.9%) | 0.25 |
| Cardiomyopathy | 2609 (28.8%) | 1851 (29.8%) | 1448 (28.5%) | 0.26 |
| Non-ischaemic | 1668 (18.4%) | 1178 (18.9%) | 897 (17.6%) | 0.20 |
| Ischaemic | 457 (5.0%) | 347 (5.6%) | 282 (5.5%) | 0.25 |
| Restrictive | 4 (0.0%) | 5 (0.1%) | 7 (0.1%) | 0.16 |
| Hypertrophic | 122 (1.3%) | 65 (1.0%) | 82 (1.6%) | 0.03 |
| Other | 497 (5.5%) | 339 (5.4%) | 245 (4.8%) | 0.20 |
| CHA2DS2-VASc score | 2.7 (1.6) | 2.8 (1.6) | 3.0 (1.6) | <0.0001 |
| Congestive heart failure | 2475 (27.3%) | 1765 (28.4%) | 1569 (30.8%) | <0.0001 |
| NYHA Class I | 696 (7.7%) | 417 (6.7%) | 364 (7.2%) | 0.07 |
| NYHA Class II | 1134 (12.5%) | 747 (12.0%) | 702 (13.8%) | 0.01 |
| NYHA Class III | 387 (4.3%) | 358 (5.8%) | 313 (6.2%) | <0.0001 |
| NYHA Class IV | 27 (0.3%) | 24 (0.4%) | 24 (0.5%) | 0.25 |
| Left ventricular dysfunction | 1230 (13.6%) | 941 (15.1%) | 770 (15.1%) | 0.006 |
| Hypertension | 6778 (74.7%) | 4579 (73.6%) | 3866 (76.0%) | 0.01 |
| Diabetes | 1978 (21.8%) | 1491 (24.0%) | 1205 (23.7%) | 0.003 |
| Stroke | 512 (5.6%) | 389 (6.3%) | 330 (6.5%) | 0.09 |
| Transient ischaemic attack | 344 (3.8%) | 268 (4.3%) | 216 (4.2%) | 0.21 |
| Thromboembolic event | 465 (5.1%) | 385 (6.2%) | 249 (4.9%) | 0.003 |
| Vascular disease | 1552 (17.1%) | 1236 (19.9%) | 936 (18.4%) | <0.0001 |
| Prior myocardial infarction | 774 (8.5%) | 593 (9.5%) | 469 (9.2%) | 0.09 |
| Peripheral arterial disease | 313 (3.4%) | 238 (3.8%) | 214 (4.2%) | 0.07 |
| Known aortic plaque | 102 (1.1%) | 76 (1.2%) | 93 (1.8%) | 0.0014 |
| HAS-BLED score | 1.1 (0.9) | 1.2 (0.9) | 1.2 (0.9) | <0.0001 |
| Uncontrolled hypertension | 737 (8.1%) | 523 (8.4%) | 495 (9.7%) | 0.004 |
| Abnormal renal function | 332 (3.7%) | 271 (4.4%) | 251 (4.9%) | 0.001 |
| Abnormal liver function | 63 (0.7%) | 49 (0.8%) | 48 (0.9%) | 0.27 |
| Prior stroke | 416 (4.6%) | 308 (5.0%) | 258 (5.1%) | 0.36 |
| Prior bleeding | 349 (3.8%) | 254 (4.1%) | 217 (4.3%) | 0.46 |
| Labile INR | 101 (1.1%) | 62 (1.0%) | 68 (1.3%) | 0.23 |
| Alcohol use | 637 (7.0%) | 415 (6.7%) | 344 (6.8%) | 0.68 |
| Anti-platelet medication use | 790 (8.7%) | 671 (10.8%) | 656 (12.9%) | <0.0001 |
| Non-steroidal inflammatory drug use | 1825 (20.1%) | 1284 (20.7%) | 978 (19.2%) | 0.17 |
| Physical examination and labs | ||||
| Body mass index, mg/m2 | 32.5 (10.0) | 32.0 (8.2) | 32.2 (10.6) | 0.004 |
| Systolic blood pressure, mmHg | 131.2 (22.9) | 131.1 (23.4) | 131.2 (23.0) | 0.96 |
| Diastolic blood pressure, mmHg | 78.3 (15.3) | 77.4 (15.0) | 78.3 (15.1) | 0.0008 |
| Heart rate, beats/min | 77.3 (21.0) | 79.1 (23.0) | 81.2 (22.4) | <0.0001 |
| Body mass index, mg/m2 | 32.5 (10.0) | 32.0 (8.2) | 32.2 (10.6) | 0.004 |
| Creatinine | 1.1 (0.5) | 1.1 (0.5) | 1.1 (0.6) | 0.005 |
| Bilirubin | 0.8 (1.0) | 0.8 (0.8) | 0.9 (1.5) | 0.09 |
| Arrhythmia history | ||||
| Symptomatic | 8839 (97.4%) | 6062 (97.4%) | 4972 (97.7%) | 0.46 |
| Attempts at AFib termination | 7422 (81.8%) | 4889 (78.6%) | 4104 (80.7%) | <0.0001 |
| Pharmacologic cardioversion | 3207 (43.2%) | 2070 (42.4%) | 1814 (44.2%) | 0.21 |
| Direct current cardioversion | 6701 (90.3%) | 4398 (90.0%) | 3666 (89.3%) | 0.2 |
| Prior AFL | 1212 (13.4%) | 2847 (45.8%) | 1555 (30.6%) | <0.0001 |
| Typical | 895 (73.8%) | 2529 (88.8%) | 1119 (72.0%) | <0.0001 |
| Pharmacologic cardioversion | 228 (2.5%) | 458 (7.4%) | 320 (6.3%) | <0.0001 |
| Direct current cardioversion | 342 (3.8%) | 817 (13.1%) | 461 (9.1%) | <0.0001 |
| Catheter ablation | 584 (6.4%) | 176 (2.8%) | 327 (6.4%) | <0.0001 |
| Pre-procedure imaging | ||||
| Transoesophageal echocardiogram performed | 5377 (59.3%) | 3682 (59.3%) | 3026 (59.6%) | 0.96 |
| Left atrial size | ||||
| Normal | 508 (25.1%) | 342 (23.6%) | 252 (22.7%) | 0.25 |
| Mildly enlarged | 615 (30.4%) | 435 (30.0%) | 364 (32.8%) | 0.68 |
| Moderately enlarged | 519 (25.7%) | 416 (28.7%) | 285 (25.7%) | 0.02 |
| Severely enlarged | 360 (17.8%) | 258 (17.8%) | 209 (18.8%) | 0.83 |
| Computed tomography performed prior | 4047 (90.4%) | 2575 (85.8%) | 2028 (86.5%) | <0.0001 |
| Magnetic resonance imaging performed prior | 4463 (49.2%) | 2985 (48.0%) | 2325 (45.7%) | 0.0004 |
| Pre-procedure medications | ||||
| Anti-thrombotic therapy | ||||
| Warfarin | 727 (8.0%) | 555 (8.9%) | 529 (10.4%) | <0.0001 |
| Direct oral anti-coagulant | 4455 (49.1%) | 3239 (52.1%) | 2521 (49.5%) | 0.001 |
| Aspirin | 1822 (20.1%) | 1309 (21.0%) | 1011 (19.9%) | 0.23 |
| Rate-control therapy | ||||
| Beta-blocker | 5427 (59.8%) | 3630 (58.3%) | 2945 (57.9%) | 0.05 |
| Digoxin | 417 (4.6%) | 296 (4.8%) | 274 (5.4%) | 0.1 |
| Anti-arrhythmic therapy | 6898 (41.2%) | 3515 (40.1%) | 1536 (30.5%) | <0.0001 |
| Amiodarone | 1846 (26.8%) | 1391 (39.6%) | 851 (55.4%) | <0.0001 |
| Dofetilide | 430 (6.2%) | 257 (7.3%) | 132 (8.6%) | <0.0001 |
| Dronedarone | 221 (3.2%) | 122 (3.5%) | 123 (8.0%) | 0.12 |
| Flecainide | 459 (6.7%) | 378 (10.7%) | 198 (12.9%) | <0.0001 |
| Propafenone | 170 (2.5%) | 93 (2.6%) | 60 (3.9%) | 0.005 |
| Sotalol | 635 (9.2%) | 310 (8.8%) | 264 (17.2%) | <0.0001 |
| Procedure information | ||||
| General | 8835 (97.3%) | 6071 (97.6%) | 5013 (98.5%) | <0.0001 |
| Double transseptal | 5126 (30.6%) | 3406 (38.8%) | 2265 (44.9%) | <0.0001 |
| All veins present able to be isolated by PVI | 4871 (95.7%) | 8529 (94.0%) | 5883 (94.6%) | <0.0001 |
| Assessed with circumferential vein catheter | 7793 (91.4%) | 5325 (90.7%) | 4420 (90.8%) | 0.26 |
| Isolation confirmation | ||||
| Entrance block | 1310 (14.4%) | 714 (11.5%) | 732 (14.4%) | <0.0001 |
| Exit block | 654 (7.2%) | 260 (4.2%) | 190 (3.7%) | <0.0001 |
| Bidirectional block | 5992 (66.0%) | 4526 (72.7%) | 3472 (68.2%) | <0.0001 |
| Atrial arrhythmia present during procedure | 1049 (11.6%) | 3330 (53.7%) | 2540 (50.0%) | <0.0001 |
| Cardioversion performed during procedure | 4912 (54.1%) | 2906 (46.7%) | 3141 (61.7%) | <0.0001 |
| Hospital characteristics | ||||
| Hospital region | ||||
| Northeast | 1147 (12.6%) | 501 (8.1%) | 422 (8.3%) | <0.0001 |
| West | 1334 (14.7%) | 801 (12.9%) | 999 (19.6%) | <0.0001 |
| Midwest | 3199 (35.2%) | 1765 (28.4%) | 1357 (26.7%) | <0.0001 |
| South | 3396 (37.4%) | 3155 (50.7%) | 2310 (45.4%) | <0.0001 |
| Location | ||||
| Rural | 529 (5.8%) | 331 (5.3%) | 322 (6.3%) | 0.07 |
| Suburban | 2806 (30.9%) | 1706 (27.4%) | 1895 (37.2%) | <0.0001 |
| Urban | 5741 (63.3%) | 4185 (67.3%) | 2871 (56.4%) | <0.0001 |
| Hospital type | ||||
| Government | 324 (3.6%) | 365 (5.9%) | 25 (0.5%) | <0.0001 |
| Private | 7407 (81.6%) | 4905 (78.8%) | 4260 (83.7%) | <0.0001 |
| University | 1345 (14.8%) | 952 (15.3%) | 803 (15.8%) | 0.3003 |
| Teaching | 5527 (60.9%) | 3097 (49.8%) | 2823 (55.5%) | <0.0001 |
| Patient beds | 559.4 (265.8) | 566.2 (272.9) | 516.3 (246.9) | <0.0001 |
| Annual volume | 1842.0 (957.2) | 1835.1 (874.6) | 1769.2 (989.0) | <.0001 |
CTI, cavotricuspid isthmus; PVI, pulmonary vein isolation; AFL, atrial flutter; INR, international normalized ratio; NYHA, New York Heart Association.
Procedural information
In the PAF cohort, general anaesthesia was used in over 94%. A double transseptal technique was used more in the PVI plus adjunctive cohort than PVI only and PVI plus CTI (44.9 vs. 30.6 vs. 38.8%). Pulmonary vein isolation was confirmed with bidirectional block in nearly 70%, and a circumferential vein catheter was used in nearly 90% of cases. Direct current cardioversion during the procedure was more common in the PVI plus adjunctive lesion cohort, occurring in 27.3% compared with 15.2% in PVI only and 17.6% in PVI plus CTI.
In the persistent AF cohort, general anaesthesia was used in over 94%. A double transseptal technique was used more in the PVI plus adjunctive cohort than PVI only and PVI plus CTI (44.9 vs. 30.6 vs. 38.8%). Pulmonary vein isolation was confirmed with bidirectional block in nearly 70%, and a circumferential vein catheter was used in nearly 90% of cases. Direct current cardioversion was more common in the PVI plus adjunctive lesion cohort, occurring in 27.3% compared with 15.2% in PVI only and 17.6% in PVI plus CTI.
Adjunctive lesion strategies
The lesion strategies used in those with PVI plus adjunctive lesions in both PAF and persistent AF cohorts are shown in Table 3. When compared with those with persistent AF, those with PAF were more likely to receive SVC isolation (5.7 vs. 2.8%, P < 0.0001), atypical AFL lines (12.5 vs. 8.1%, P < 0.0001), and less likely to receive multiple adjunctive lesions.
Table 3.
Descriptive analysis of adjunctive lesion strategies in the paroxysmal and persistent atrial fibrillation cohorts
| Paroxysmal (N = 5041) | Persistent (N = 5088) | P-value | |
|---|---|---|---|
| Superior vena cava isolation | 287 (5.7%) | 140 (2.8%) | <0.0001 |
| Coronary sinus | 126 (2.5%) | 128 (2.5%) | 0.96 |
| Ligament/vein of Marshall | 53 (1.1%) | 38 (0.7%) | 0.10 |
| Other | 1695 (33.6%) | 1785 (35.1%) | 0.12 |
| Atypical AFL lines | 631 (12.5%) | 411 (8.1%) | <0.0001 |
| Multiple lesions, including CTI | 1846 (36.6%) | 2083 (40.9%) | <0.0001 |
| Multiple lesions, non-CTI | 403 (8.0%) | 503 (9.9%) | 0.0009 |
CTI, cavotricuspid isthmus; AFL, atrial flutter.
Outcomes
The unadjusted rates of in-hospital adverse events in the PAF cohort are presented in Table 4. In-hospital death was rare across the groups. Hospital stay >1 day was more frequent in PVI plus adjunctive lesions than PVI only and PVI plus CTI (14.1 vs. 8.2 vs. 10.1%, P < 0.0001). The rates of any complication were higher in the PVI plus adjunctive lesion cohort with 2.9%, when compared with 2.3% in PVI only and 2.1% in PVI plus CTI cohorts (P = 0.008). Major complications occurred in 0.8% of PVI only, 0.7% of PVI plus CTI, and 1.1% of PVI plus adjunctive lesions with no statistically significant difference across the groups.
Table 4.
Unadjusted prevalence of adverse events among the paroxysmal atrial fibrillation cohort
| PVI only (N = 16 735) | PVI + CTI (N = 8775) | PVI + adj (N = 5041) | P-value | |
|---|---|---|---|---|
| In-hospital death | 5 (0.0%) | 3 (0.0%) | 6 (0.1%) | 0.03 |
| Hospitalization (>1 vs. ≤1 day) | 1366 (8.2%) | 885 (10.1%) | 711 (14.1%) | <0.0001 |
| Any complication | 379 (2.3%) | 183 (2.1%) | 146 (2.9%) | 0.0076 |
| Major complication | 133 (0.8%) | 64 (0.7%) | 53 (1.1%) | 0.11 |
| Specific complication | ||||
| Bradycardia adverse events | 45 (0.3%) | 28 (0.3%) | 24 (0.5%) | 0.07 |
| Cardiac arrest | 15 (0.1%) | 3 (0.0%) | 5 (0.1%) | 0.25 |
| Myocardial infarction | 8 (0.0%) | 2 (0.0%) | 4 (0.1%) | 0.32 |
| Air embolism | 12 (0.1%) | 2 (0.0%) | 4 (0.1%) | 0.25 |
| LA thrombus | 5 (0.0%) | 2 (0.0%) | 0 (0.0%) | 0.47 |
| Cardiac thromboembolic event | 3 (0.0%) | 0 (0.0%) | 2 (0.0%) | 0.21 |
| TIA/stroke | 18 (0.1%) | 10 (0.1%) | 15 (0.3%) | 0.005 |
| Arterial thrombosis | 10 (0.1%) | 1 (0.0%) | 4 (0.1%) | 0.14 |
| Deep-vein thrombosis | 11 (0.1%) | 1 (0.0%) | 6 (0.1%) | 0.04 |
| Respiratory failure | 40 (0.2%) | 20 (0.2%) | 20 (0.4%) | 0.12 |
| Phrenic nerve damage | 40 (0.2%) | 17 (0.2%) | 0 (0.0%) | 0.002 |
| Pulmonary embolism | 7 (0.0%) | 1 (0.0%) | 7 (0.1%) | 0.004 |
| Pulmonary vein damage/dissection | 10 (0.1%) | 2 (0.0%) | 4 (0.1%) | 0.31 |
| Pneumonia | 21 (0.1%) | 3 (0.0%) | 8 (0.2%) | 0.04 |
| Sepsis | 5 (0.0%) | 1 (0.0%) | 0 (0.0%) | 0.34 |
| Acute renal failure | 14 (0.1%) | 11 (0.1%) | 12 (0.2%) | 0.02 |
| GU bleeding | 7 (0.0%) | 1 (0.0%) | 3 (0.1%) | 0.30 |
| Heart failure | 32 (0.2%) | 24 (0.3%) | 22 (0.4%) | 0.01 |
| Pericardial effusion resulting in cardiac tamponade | 42 (0.3%) | 24 (0.3%) | 19 (0.4%) | 0.33 |
| Pericardial effusion requiring cardiac surgery | 73 (0.4%) | 41 (0.5%) | 25 (0.5%) | 0.84 |
| Cardiac surgery | 17 (0.1%) | 6 (0.1%) | 6 (0.1%) | 0.60 |
| Haemorrhage (non-access site) | 15 (0.1%) | 11 (0.1%) | 5 (0.1%) | 0.69 |
| Haematoma at access site | 56 (0.3%) | 20 (0.2%) | 20 (0.4%) | 0.18 |
| Bleeding requiring transfusion (access site) | 27 (0.2%) | 8 (0.1%) | 9 (0.2%) | 0.29 |
| AV fistula requiring intervention | 9 (0.1%) | 2 (0.0%) | 6 (0.1%) | 0.07 |
| Pseudoaneurysm requiring intervention | 28 (0.2%) | 9 (0.1%) | 11 (0.2%) | 0.23 |
| Vascular injury | 18 (0.1%) | 6 (0.1%) | 6 (0.1%) | 0.56 |
CTI, cavotricuspid isthmus; PVI, pulmonary vein isolation; AV, arteriovenous; GU, genitourinary; LA, left atrium; TIA, transient ischemic attack.
In the PAF cohort, when compared with PVI only, those with PVI plus CTI had a higher odds of hospital stay >1 day in both unadjusted and multivariable adjusted analyses (adjusted OR 1.12; 95% CI 1.01–1.25; P < 0.0001), as shown in Figure 1. There were no differences in the risk of any complication or major complication. When compared with PVI only, those receiving PVI plus adjunctive lesion were at a significantly higher risk of hospital stay >1 day in both unadjusted and adjusted analyses (adjusted OR 1.45; 95% CI 1.30–1.62; P < 0.0001). No differences were observed in any or major complications after adjustment in PVI plus adjunctive compared with PVI only patients.
Figure 1.
Unadjusted and adjusted outcomes of in-hospital adverse events for those undergoing PVI plus CTI ablation only or adjunctive lesions when compared with PVI only (reference) patients with paroxysmal AF (A) and persistent AF (B). AF, atrial fibrillation; CTI, cavotricuspid isthmus; PVI, pulmonary vein isolation.
The unadjusted rates of in-hospital adverse events in the persistent AF cohort are presented in Table 5. In-hospital death was rare across the groups. Hospital stay >1 day occurred in 16.2% of patients in the PVI plus adjunctive lesions, compared with 13.1% in PVI only and 13.5% in PVI plus CTI (P < 0.0001). Any complication was more frequent in the PVI plus adjunctive lesion cohort with 4.5%, when compared with 3.0% in PVI only and 3.2% in PVI plus CTI (P = 0.008). The rates of major complications were more common in PVI plus adjunctive lesions (1.4%) when compared with PVI only (0.8%) and PVI plus CTI (1.4 vs. 0.8 vs. 1.1%, P = 0.0008). Specific complications, including stroke/transient ischaemic attack, acute renal failure, and heart failure, were statistically higher in PVI plus adjunctive when compared with the other groups.
Table 5.
Unadjusted prevalence of adverse events among the persistent atrial fibrillation cohort
| PVI only (N = 9076) | PVI + CTI (N = 6222) | PVI + adj (N = 5088) | P-value | |
|---|---|---|---|---|
| In-hospital death | 3 (0.0%) | 3 (0.0%) | 7 (0.1%) | 0.05 |
| Hospitalization (>1 vs. ≤1 day) | 1191 (13.1%) | 843 (13.5%) | 825 (16.2%) | <0.0001 |
| Any complication | 268 (3.0%) | 196 (3.2%) | 227 (4.5%) | <0.0001 |
| Major complication | 74 (0.8%) | 51 (0.8%) | 72 (1.4%) | 0.001 |
| Specific complication | ||||
| Bradycardia adverse events | 48 (0.5%) | 40 (0.6%) | 40 (0.8%) | 0.17 |
| Cardiac arrest | 8 (0.1%) | 3 (0.0%) | 9 (0.2%) | 0.09 |
| Myocardial infarction | 1 (0.0%) | 2 (0.0%) | 2 (0.0%) | 0.53 |
| Air embolism | 5 (0.1%) | 2 (0.0%) | 6 (0.1%) | 0.18 |
| Left atrial thrombus | 3 (0.0%) | 1 (0.0%) | 0 (0.0%) | 0.39 |
| Cardiac thromboembolic event | 1 (0.0%) | 1 (0.0%) | 1 (0.0%) | 0.92 |
| TIA/stroke | 15 (0.2%) | 11 (0.2%) | 23 (0.5%) | 0.002 |
| Arterial thrombosis | 4 (0.0%) | 2 (0.0%) | 4 (0.1%) | 0.52 |
| Deep-vein thrombosis | 5 (0.1%) | 4 (0.1%) | 5 (0.1%) | 0.63 |
| Respiratory failure | 35 (0.4%) | 24 (0.4%) | 31 (0.6%) | 0.11 |
| Phrenic nerve damage | 21 (0.2%) | 7 (0.1%) | 6 (0.1%) | 0.13 |
| Pulmonary embolism | 3 (0.0%) | 4 (0.1%) | 4 (0.1%) | 0.49 |
| Pulmonary vein damage/dissection | 2 (0.0%) | 2 (0.0%) | 4 (0.1%) | 0.25 |
| Pneumonia | 10 (0.1%) | 9 (0.1%) | 14 (0.3%) | 0.06 |
| Sepsis | 3 (0.0%) | 2 (0.0%) | 0 (0.0%) | 0.44 |
| Acute renal failure | 17 (0.2%) | 14 (0.2%) | 21 (0.4%) | 0.03 |
| GU bleeding | 3 (0.0%) | 4 (0.1%) | 4 (0.1%) | 0.49 |
| Heart failure | 52 (0.6%) | 46 (0.7%) | 57 (1.1%) | 0.002 |
| Pericardial effusion resulting in cardiac tamponade | 21 (0.2%) | 15 (0.2%) | 18 (0.4%) | 0.36 |
| Pericardial effusion requiring cardiac surgery | 37 (0.4%) | 25 (0.4%) | 28 (0.6%) | 0.40 |
| Cardiac surgery | 7 (0.1%) | 9 (0.1%) | 8 (0.2%) | 0.32 |
| Haemorrhage (non-access site) | 6 (0.1%) | 9 (0.1%) | 10 (0.2%) | 0.09 |
| Haematoma at access site | 28 (0.3%) | 29 (0.5%) | 29 (0.6%) | 0.06 |
| Bleeding requiring transfusion (access site) | 10 (0.1%) | 12 (0.2%) | 13 (0.3%) | 0.12 |
| AV fistula requiring intervention | 3 (0.0%) | 3 (0.0%) | 6 (0.1%) | 0.12 |
| Pseudoaneurysm requiring intervention | 16 (0.2%) | 12 (0.2%) | 7 (0.1%) | 0.77 |
| Vascular injury | 9 (0.1%) | 5 (0.1%) | 10 (0.2%) | 0.16 |
Major complications included death, stroke, TIA, cardiac arrest, cardiac surgery, vascular injury, access site bleeding, and pericardial effusion.
CTI, cavotricuspid isthmus; PVI, pulmonary vein isolation; AV, arteriovenous; GU, genitourinary; TIA, transient ischemic attack.
In the multivariable adjusted analysis of the persistent AF cohort, there were no differences in the risk of hospital stay >1 day, any complication, or major complication in PVI plus CTI when compared with PVI only, as shown in Figure 1. However, when compared with PVI only, PVI plus adjunctive had a higher risk of any complication (OR 1.30; 95% CI 1.07–1.58; P = 0.008) and major complication (OR 1.56; 95% CI 1.10–2.21; P = 0.014).
Discussion
In this analysis of the largest AF ablation registry including elective outpatient procedures worldwide including 50 937 patients from 2016 to 2020, we observed several important in-hospital findings in patients undergoing PVI with or without adjunctive lesions during first-time ablation. First, 40% of the catheter ablations are being performed in those with persistent AF, and nearly half of the study cohort underwent additional lesions beyond PVI including CTI ablation and other adjunctive lesions. Second, adjunctive lesions were more common in those with persistent AF and a high burden of comorbidities. Third, there was high heterogeneity in the types of adjunctive lesion sets used. Fourth, among those with PAF, those with additional CTI or adjunctive lesions were more likely to experience prolonged hospitalization than PVI only, while there was no increased risk of complications. Lastly, among those with persistent AF, those with PVI plus adjunctive lesions, but not PVI plus CTI alone, were more likely to experience any complications and major complications.
For over two decades, percutaneous catheter ablation for AF has been shown to be more effective in reducing AF burden than anti-arrhythmic drug therapy in several randomized trials.14,15 Additionally, ablation has also been shown to reduce hospitalizations, improve quality of life, and may improve survival in patients with systolic heart failure.1–3 Given the superior efficacy, catheter ablation has been strongly endorsed by professional society guidelines, yet approximately one-third of patients with PAF undergoing ablation will have recurrence by 1 year and nearly half of those with persistent AF will have recurrence by 1 year.5,6,16 Additional ablation strategies targeting arrhythmogenic areas beyond PVI to improve AF-free survival have been evaluated in multiple studies with varying results, although these studies have lacked the power to comprehensively and accurately evaluate procedural complication risk.
In the present study, CTI ablation alone was commonly performed in addition to PVI, including 28.7% of the PAF cohort and 30.5% of the persistent AF cohort. Atrial fibrillation and AFL often coexist and are closely interrelated.17 Cavotricuspid isthmus ablation combined with PVI in patients with a history of AFL has been shown to reduce recurrence of atrial arrhythmia.18,19 Although a safe, efficacious, and durable procedure, prophylactic CTI ablation has not been shown to improve freedom from atrial arrhythmia recurrence.20,21 Therefore, a Class I indication by professional society guidelines exists for concomitant CTI ablation only in those with previously documented or inducible AFL.5 Over half of the PAF cohort and 46% of the persistent AF cohort had prior documented AFL; however, inducible AFL at the time of the ablation was not captured in the registry. Despite the low procedural risk of CTI ablation with no significant differences across the groups, we found prolonged hospitalization occurred more often with the addition of CTI ablation in those with PAF. While the registry does not capture the reason for prolonged hospitalization, those with PAF undergoing additional CTI ablation may have required ongoing arrhythmia management, including initiation and monitoring of anti-arrhythmic drug therapy, or management of competing comorbidities. As same-day discharge following AF ablation becomes more common, further efforts are warranted to understand the reasons for the prolonged hospitalization in those with PAF requiring additional ablation beyond PVI.22
The major finding of the present study was the increased risk of complications observed in those with persistent AF undergoing ablation with adjunctive lesion sets. While PVI remains the cornerstone for AF ablation, recurrence of atrial arrhythmia remains common, often requiring repeat procedures, and further atrial substrate modification, particularly with persistent AF.23 Strategies including linear ablation, targeting complex fractionated atrial electrogram ablation, or magnetic resonance imaging–guided fibrosis ablation have not demonstrated superiority to PVI in clinical trials.6,24,25 Although several adjunctive strategies including vein of Marshall ethanol infusion, ablation of non-pulmonary vein (PV) triggers, left atrial appendage isolation, and posterior wall isolation have shown promise in improving AF-free survival in those with persistent AF, no clear consensus exists for guidance of further ablation strategies beyond PVI, leading to debate and uncertainty.7,8,26,27 As shown in the present study, there was significant heterogeneity in the type of adjunctive lesions performed in both PAF and persistent AF and over a third received multiple lesions. Similarly, in a study of the Get With The Guidelines-Atrial Fibrillation Registry which included 3139 patients from 2016 to 2018, the investigators demonstrated a high use of adjunctive lesions in first-time ablations, including left atrial linear ablations in over a third of patients and posterior wall isolation in nearly a quarter of those with persistent AF, although the rates of procedural complications in those receiving adjunctive lesions were not reported.10
While we observed a higher risk in those with persistent AF undergoing adjunctive lesions, this analysis cannot establish a causal relationship between performing adjunctive ablations and complication risk. The current study is not equipped to determine the underlying cause of complications. Due to the significant heterogeneity in adjunctive lesions performed, often with multiple lesions performed, individual adjunctive lesion outcome analysis was not performed. Furthermore, several factors that may impact the risk relationship remain unknown. The reason for ablation strategy was unavailable in the registry and may be influenced by several factors, including an atrial arrhythmia requiring additional ablation observed before or during the procedure, provider skill and preference for additional lesion type, anatomical factors, or a pre-determined empiric strategy. Furthermore, data on anti-coagulation strategy implementation were limited, including rates of uninterrupted anti-coagulation and the timing of anti-coagulation initiation. While we adjusted for a comprehensive set of comorbidities, those who receive adjunctive may be at inherently higher risk and residual confounding cannot be entirely ruled out. The patients with adjunctive lesions were older and had a higher burden of coronary artery disease, heart failure, hypertension, and diabetes. Indeed, when complications were considered individually, the rates of stroke, acute renal failure, and heart failure were higher in those undergoing adjunctive lesions, although the absolute numbers were small. It is worth noting that phrenic nerve injury was more common in those with PVI only. While it has been shown that phrenic nerve injury is more common with cryoballoon ablation, the current data collection form does not distinguish the ablation modality.28 Other major complications that may be directly related to vascular access or catheter manipulation, such as pericardial effusion or major bleeding, were not significantly different across the groups. The findings call awareness to optimizing and management of volume status and blood pressure pre-, intra-, and post-procedurally to mitigate risk.
Other factors such as patient selection, ongoing development of safe technology, maintaining adequate volume, and familiarization with newer ablation strategy techniques prior to implementation may also lead to improvement in catheter ablation outcomes in this high-risk cohort. Standardization of ablation strategies beyond PVI to adequately study outcomes in large, randomized trials is also warranted; however, generalizability remains a challenge due to the spectrum of AF burden and the still unknown mechanisms underlying AF initiation and perpetuation. Ultimately, approaches aimed at prevention and delaying progression of AF with lifestyle modification and a paradigm shift to early catheter ablation prior to anti-arrhythmic drug failure and worsening comorbidities may prove instrumental in reducing procedural complications.29–31
Limitations
First, the NCDR AFib Ablation Registry is observational data; therefore, causal inferences cannot be made. Although the registry has a large sample size that allows for generalizable data reflecting real-world practice trends and safety data, these findings do not suggest the avoidance of adjunctive lesions during AF ablation. Second, the registry is limited to the index hospitalization. Late complications, such as pulmonary vein stenosis or atrioesophageal fistula, were not captured, nor were long-term atrial arrhythmia recurrence rates. Third, there were several adjunctive lesions likely commonly performed that were not identified on the data collection form, such as posterior wall isolation and left atrial appendage isolation. Lastly, despite adjustment for an extensive list of potential confounders, there may be unmeasured confounders that can influence the risk relationship.
Conclusions
In the largest nationwide cohort of 50 937 patients undergoing first-time AF ablation, those with PAF who underwent PVI plus CTI or adjunctive lesions experienced more prolonged hospitalizations compared with those treated with PVI only, but adjunctive lesion sets otherwise had no impact on adverse outcomes. In patients with persistent AF, those who underwent PVI plus adjunctive lesions were more likely to experience in-hospital complications when compared with those who underwent PVI only, while there was no difference in those who underwent PVI plus CTI. Further strategies are warranted to mitigate the risk of complications in those with persistent AF, including focusing on upstream AF management and further trials to replicate findings, familiarize techniques, and standardize adjunctive lesions.
Contributor Information
Douglas Darden, Kansas City Heart Rhythm Institute, 5100 W 110th St, Suite 200, Overland Park, KS, USA.
Omar Aldaas, Division of Cardiology, Department of Medicine, University of California, San Diego, La Jolla, CA 92037, USA.
Chengan Du, Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT, USA; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT, USA.
Muhammad Bilal Munir, Division of Cardiology, Department of Medicine, University of California Davis, Sacramento, CA, USA.
Gregory K Feld, Division of Cardiology, Department of Medicine, University of California, San Diego, La Jolla, CA 92037, USA.
Naga Venkata K Pothineni, Kansas City Heart Rhythm Institute, 5100 W 110th St, Suite 200, Overland Park, KS, USA.
Rakesh Gopinathannair, Kansas City Heart Rhythm Institute, 5100 W 110th St, Suite 200, Overland Park, KS, USA.
Dhanunjaya Lakkireddy, Kansas City Heart Rhythm Institute, 5100 W 110th St, Suite 200, Overland Park, KS, USA.
Jeptha P Curtis, Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT, USA; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT, USA.
James V Freeman, Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT, USA; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT, USA.
Joseph G Akar, Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT, USA; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT, USA.
Jonathan C Hsu, Division of Cardiology, Department of Medicine, University of California, San Diego, La Jolla, CA 92037, USA.
Funding
This research was supported by the American College of Cardiology Foundation’s National Cardiovascular Data Registry (NCDR).
Data Availability
The data underlying this article were provided by NCDR under funding and permission. Data will be shared on request with the corresponding author with the permission of NCDR.
References
- 1. Packer DL, Mark DB, Robb RA, Monahan KH, Bahnson TD, Poole JEet al. Effect of catheter ablation vs antiarrhythmic drug therapy on mortality, stroke, bleeding, and cardiac arrest among patients with atrial fibrillation: the CABANA randomized clinical trial. JAMA 2019;321:1261–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Mark DB, Anstrom KJ, Sheng S, Piccini JP, Baloch KN, Monahan KHet al. Effect of catheter ablation vs medical therapy on quality of life among patients with atrial fibrillation: the CABANA randomized clinical trial. JAMA 2019;321:1275–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Marrouche NF, Brachmann J, Andresen D, Siebels J, Boersma L, Jordaens Let al. Catheter ablation for atrial fibrillation with heart failure. N Engl J Med 2018;378:417–27. [DOI] [PubMed] [Google Scholar]
- 4. Haïssaguerre M, Jaïs P, Shah DC, Takahashi A, Hocini M, Quiniou Get al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339:659–66. [DOI] [PubMed] [Google Scholar]
- 5. Calkins H, Hindricks G, Cappato R, Kim YH, Saad EB, Aguinaga Let al. 2017 HRS/EHRA/ECAS/APHRS/SOLAECE expert consensus statement on catheter and surgical ablation of atrial fibrillation: executive summary. Heart Rhythm 2017;14:e445–94. [DOI] [PubMed] [Google Scholar]
- 6. Verma A, Jiang CY, Betts TR, Chen J, Deisenhofer I, Mantovan Ret al. Approaches to catheter ablation for persistent atrial fibrillation. N Engl J Med 2015;372:1812–22. [DOI] [PubMed] [Google Scholar]
- 7. Valderrábano M, Peterson LE, Swarup V, Schurmann PA, Makkar A, Doshi RNet al. Effect of catheter ablation with vein of Marshall ethanol infusion vs catheter ablation alone on persistent atrial fibrillation: the VENUS randomized clinical trial. JAMA 2020;324:1620–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Di Biase L, Burkhardt JD, Mohanty P, Mohanty S, Sanchez JE, Trivedi Cet al. Left atrial appendage isolation in patients with longstanding persistent AF undergoing catheter ablation: BELIEF trial. J Am Coll Cardiol 2016;68:1929–40. [DOI] [PubMed] [Google Scholar]
- 9. Kim JS, Shin SY, Na JO, Choi CU, Kim SH, Kim JWet al. Does isolation of the left atrial posterior wall improve clinical outcomes after radiofrequency catheter ablation for persistent atrial fibrillation?: a prospective randomized clinical trial. Int J Cardiol 2015;181:277–83. [DOI] [PubMed] [Google Scholar]
- 10. Loring Z, Holmes DN, Matsouaka RA, Curtis AB, Day JD, Desai Net al. Procedural patterns and safety of atrial fibrillation ablation: findings from Get With The Guidelines-Atrial Fibrillation. Circ Arrhythm Electrophysiol 2020;13:e007944. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Hsu JC, Darden D, Du C, Marine JE, Nichols S, Marcus GMet al. Initial findings from the national cardiovascular data registry of atrial fibrillation ablation procedures. J Am Coll Cardiol 2023;81:867–878. 10.1016/j.jacc.2022.11.060 [DOI] [PubMed] [Google Scholar]
- 12. Registry . NA. Data collection form v1.0.https://cvquality.acc.org/docs/default-source/ncdr/Data-Collection/afib_v1_datacollectionform.pdf?sfvrsn=d37b8dbf_0 (22 April 2022, date last accessed).
- 13. Messenger JC, Ho KKL, Young CH, Slattery LE, Draoui JC, Curtis JPet al. The National Cardiovascular Data Registry (NCDR) data quality brief: the NCDR data quality program in 2012. J Am Coll Cardiol 2012;60:1484–8. [DOI] [PubMed] [Google Scholar]
- 14. Wilber DJ, Pappone C, Neuzil P, De Paola A, Marchlinski F, Natale Aet al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA 2010;303:333–40. [DOI] [PubMed] [Google Scholar]
- 15. Pappone C, Augello G, Sala S, Gugliotta F, Vicedomini G, Gulletta Set al. A randomized trial of circumferential pulmonary vein ablation versus antiarrhythmic drug therapy in paroxysmal atrial fibrillation: the APAF study. J Am Coll Cardiol 2006;48:2340–7. [DOI] [PubMed] [Google Scholar]
- 16. Poole JE, Bahnson TD, Monahan KH, Johnson G, Rostami H, Silverstein APet al. Recurrence of atrial fibrillation after catheter ablation or antiarrhythmic drug therapy in the CABANA trial. J Am Coll Cardiol 2020;75:3105–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Waldo AL, Feld GK. Inter-relationships of atrial fibrillation and atrial flutter mechanisms and clinical implications. J Am Coll Cardiol 2008;51:779–86. [DOI] [PubMed] [Google Scholar]
- 18. Wazni O, Marrouche NF, Martin DO, Gillinov AM, Saliba W, Saad Eet al. Randomized study comparing combined pulmonary vein-left atrial junction disconnection and cavotricuspid isthmus ablation versus pulmonary vein-left atrial junction disconnection alone in patients presenting with typical atrial flutter and atrial fibrillation. Circulation 2003;108:2479–83. [DOI] [PubMed] [Google Scholar]
- 19. Mohanty S, Mohanty P, Di Biase L, Bai R, Santangeli P, Casella Met al. Results from a single-blind, randomized study comparing the impact of different ablation approaches on long-term procedure outcome in coexistent atrial fibrillation and flutter (APPROVAL). Circulation 2013;127:1853–60. [DOI] [PubMed] [Google Scholar]
- 20. Kim SH, Oh YS, Choi Y, Hwang Y, Kim JY, Kim TSet al. Long-term efficacy of prophylactic cavotricuspid isthmus ablation during atrial fibrillation ablation in patients without typical atrial flutter: a prospective, multicentre, randomized trial. Korean Circ J 2021;51:58–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Pontoppidan J, Nielsen JC, Poulsen SH, Jensen HK, Walfridsson H, Pedersen AKet al. Prophylactic cavotricuspid isthmus block during atrial fibrillation ablation in patients without atrial flutter: a randomised controlled trial. Heart 2009;95:994–9. [DOI] [PubMed] [Google Scholar]
- 22. Deyell MW, Leather RA, Macle L, Forman J, Khairy P, Zhang Ret al. Efficacy and safety of same-day discharge for atrial fibrillation ablation. JACC Clin Electrophysiol 2020;6:609–19. [DOI] [PubMed] [Google Scholar]
- 23. Ganesan AN, Shipp NJ, Brooks AG, Kuklik P, Lau DH, Lim HSet al. Long-term outcomes of catheter ablation of atrial fibrillation: a systematic review and meta-analysis. J Am Heart Assoc 2013;2:e004549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Marrouche NF, Wazni O, McGann C, Greene T, Dean JM, Dagher Let al. Effect of MRI-guided fibrosis ablation vs conventional catheter ablation on atrial arrhythmia recurrence in patients with persistent atrial fibrillation: the DECAAF II randomized clinical trial. JAMA 2022;327:2296–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Inoue K, Hikoso S, Masuda M, Furukawa Y, Hirata A, Egami Yet al. Pulmonary vein isolation alone vs. more extensive ablation with defragmentation and linear ablation of persistent atrial fibrillation: the EARNEST-PVI trial. Europace 2021;23:565–74. [DOI] [PubMed] [Google Scholar]
- 26. Della Rocca DG, Mohanty S, Mohanty P, Trivedi C, Gianni C, Al-Ahmad Aet al. Long-term outcomes of catheter ablation in patients with longstanding persistent atrial fibrillation lasting less than 2 years. J Cardiovasc Electrophysiol 2018;29:1607–15. [DOI] [PubMed] [Google Scholar]
- 27. DeLurgio DB, Crossen KJ, Gill J, Blauth C, Oza SR, Magnano ARet al. Hybrid convergent procedure for the treatment of persistent and long-standing persistent atrial fibrillation: results of CONVERGE clinical trial. Circ Arrhythm Electrophysiol 2020;13:e009288. [DOI] [PubMed] [Google Scholar]
- 28. Kuck KH, Brugada J, Fürnkranz A, Metzner A, Ouyang F, Chun KRet al. Cryoballoon or radiofrequency ablation for paroxysmal atrial fibrillation. N Engl J Med 2016;374:2235–45. [DOI] [PubMed] [Google Scholar]
- 29. Fitzgerald JL, Middeldorp ME, Gallagher C, Sanders P. Lifestyle modification and atrial fibrillation: critical care for successful ablation. J Clin Med 2022;11:2660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Andrade JG, Wells GA, Deyell MW, Bennett M, Essebag V, Champagne Jet al. Cryoablation or drug therapy for initial treatment of atrial fibrillation. N Engl J Med 2021;384:305–15. [DOI] [PubMed] [Google Scholar]
- 31. Wazni OM, Dandamudi G, Sood N, Hoyt R, Tyler J, Durrani Set al. Cryoballoon ablation as initial therapy for atrial fibrillation. N Engl J Med 2021;384:316–24. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data underlying this article were provided by NCDR under funding and permission. Data will be shared on request with the corresponding author with the permission of NCDR.