Author + information
- Received June 25, 2018
- Revision received September 18, 2018
- Accepted September 20, 2018
- Published online December 17, 2018.
- Thomas Fink, MDa,∗ (, )
- Michael Schlüter, PhDb,
- Christian-Hendrik Heeger, MDa,
- Christine Lemeš, MDa,
- Tilman Maurer, MDa,
- Bruno Reissmann, MDa,
- Laura Rottner, MDa,
- Francesco Santoro, MDa,
- Roland Richard Tilz, MDa,
- Hannes Alessandrini, MDa,
- Andreas Rillig, MDa,c,
- Shibu Mathew, MDa,
- Peter Wohlmuth, PhDb,
- Qizhi Fang, MDd,
- Randall Lee, MD, PhDd,
- Feifan Ouyang, MDa,
- Karl-Heinz Kuck, MDa and
- Andreas Metzner, MDa
- aDepartment of Cardiology, Asklepios Klinik St. Georg, Hamburg, Germany
- bAsklepios Proresearch, Hamburg, Germany
- cDepartment of Cardiology, Campus Benjamin Franklin, Charité, Universitätsmedizin Berlin, Berlin, Germany
- dDivision of Cardiology and the Cardiovascular Research Institute, University of California, San Francisco, California
- ↵∗Address for correspondence:
Dr. Thomas Fink, Department of Cardiology, Asklepios Klinik St. Georg, Lohmühlenstrasse 5, 20099 Hamburg, Germany.
Objectives This study investigated the outcome of wide-area left atrial appendage isolation (WLAAI) and subsequent LAA ligation in patients with recurrent atrial arrhythmias after pulmonary vein isolation (PVI).
Background LAA isolation and ligation may improve rhythm control and prevent LAA thrombus formation in patients with atrial fibrillation who do not respond to PVI.
Methods Patients (n = 31, mean age: 69.7 ± 7.8 years, 18 men) with arrhythmia recurrence after established PVI undergoing WLAAI with subsequent LAA ligation (LARIAT+ device) were studied. The incidence of arrhythmia recurrence, intracardiac thrombus formation, thromboembolic events, as well as changes in P-wave duration and P-wave dispersion were assessed.
Results All 31 patients underwent successful WLAAI, and successful LAA ligation was performed in 27 patients (87%). Over a median follow-up of 498 (interquartile range: 159 to 791) days, post-ligation arrhythmia recurrence was documented in 8 patients (26%). Kaplan-Meier estimate of 24-month arrhythmia-free survival after WLAAI/ligation was 69.7% (95% confidence interval: 53.9 to 90.1). Following WLAAI, LAA thrombus formation was seen in 11 patients (35.5%), but in no patient after LAA ligation. WLAAI/ligation significantly reduced P-wave duration (from 93 ± 20 ms to 72 ± 20 ms; p = 0.001) and P-wave dispersion (from 63 ± 37 ms to 38 ± 16 ms; p = 0.001).
Conclusions WLAAI and subsequent LAA ligation in PVI nonresponders led to an estimated freedom from arrhythmia recurrence in 70% of the patients at 24 months. LAA ligation successfully prevented recurrence of cardiac thrombus formation in patients with WLAAI. Significant decreases in P-wave duration and P-wave dispersion occurred with WLAAI/ligation, suggesting favorable electrical remodelling.
- arrhythmia recurrence
- atrial fibrillation
- catheter ablation
- epicardial left atrial appendage closure
- left atrial appendage isolation
The cornerstone of atrial fibrillation (AF) ablation is electrical isolation of the pulmonary veins, or pulmonary vein isolation (PVI) (1). The optimal ablation strategy in patients with persistent AF and especially in PVI nonresponders is unknown and ablation strategies in addition to PVI such as the ablation of complex fractionated electrograms or the deployment of linear lesions failed in most trials to prove superiority when compared with PVI alone (2–7). Electrical isolation of the left atrial appendage (LAA) might be a treatment option for PVI nonresponders. LAA isolation can be achieved by the creation of circular lesions around the LAA base (8,9) or by a combination of linear left atrial lesions, resulting in wide-area left atrial appendage isolation (WLAAI) (10,11). LAA isolation in addition to PVI has been shown to decrease the recurrence of AF (8).
Our group reported on a high incidence of LAA thrombus formation and thromboembolic events in patients with WLAAI and a significant rate of LA to LAA reconduction following initially successful LAA isolation (10,12). Based on these observations and previous reports that epicardial LAA ligation prevents thromboembolic events and could potentially lead to maintenance of sinus rhythm (SR) in AF patients (13,14), we hypothesized that combined WLAAI and LAA ligation prevents thromboembolic events in AF patients and could potentially improve arrhythmia-free survival.
In the current study, we report on the electrophysiological observations and the rhythmological outcome of a combined strategy of WLAAI and consecutive LAA ligation in a patient cohort of PVI nonresponders.
We performed a retrospective study on consecutive patients undergoing WLAAI and subsequent LAA ligation with the LARIAT device (SentreHeart, Redwood City, California) performed at the Asklepios Clinic St. Georg, Hamburg, Germany, between August 1, 2014 and September 30, 2017. The study was approved by the local ethics board. All study patients had documented symptomatic atrial arrhythmia recurrence with the indication to undergo a further ablation attempt and had documented PVI at the beginning of the WLAAI procedure. All patients gave written informed consent to the procedure, patient information was anonymized for analysis.
WLAAI and LAA ligation
Transesophageal echocardiography (TEE) was performed prior to the procedure to rule out intracardiac thrombi. Oral anticoagulation (OAC) with a vitamin K antagonist was not interrupted (international normalized ratio [INR] aim of 2.0 to 3.0). Novel oral anticoagulants were discontinued 24 h before the procedure. Procedures were performed under deep sedation using midazolam, sufentanil, and propofol. Heparin was applied targeting an activated clotting time >300 s.
Double transseptal puncture was performed under fluoroscopic guidance using a modified Brockenbrough technique and 8.5-F transseptal sheaths (SL 1; St. Jude Medical, Inc., St. Paul, Minnesota). Persistent PVI was verified via electrogram recordings from the PV with a lasso catheter. The occurrence of ectopy from the LAA was proved by activation mapping and via recording from a lasso catheter placed inside the LAA. At the beginning of the procedure, the rhythm was analyzed. If the patient was in SR at the beginning of the procedure, programmed electrical stimulation with up to 3 extra beats or burst-pacing with a minimal cycle length of 250 ms from the coronary sinus or the lasso catheter was performed to induce atrial arrhythmias.
WLAAI was conducted by creating an LA anterior line connecting the mitral annulus and the right superior PV and a posterior mitral isthmus line between the mitral annulus and the left inferior PV (Online Figure 1). If necessary, epicardial ablation of the mitral isthmus area was conducted inside the coronary sinus (CS). Endocardial surface ablation at the LA anterior wall mitral isthmus area was performed with maximum power of 30 to 35 W at the anteroseptal part of the LA and 40 W at the mitral annulus with a maximum target temperature of 43°C and a flush rate of 25 ml/min. Ablation inside the CS was limited to 20 W and a flush rate of 17 ml/min. Bidirectional block of the LA anterior line was demonstrated by prolonged conduction intervals into the LAA of at least 50 ms during SR and demonstration of early activation at sites inferior to the ablation line and late activation superior of the ablation line during SR. Block of the mitral isthmus line was proven via demonstration of double potentials along the ablation line and demonstration of a proximal to distal CS conduction during pacing from the LAA. Electric LAA isolation was defined as disappearance of all LAA potentials documented with a lasso catheter inside the LAA and demonstration of exit block during pacing or LAA ectopy.
In the case of atrial tachycardia (AT) other than perimitral flutter (e.g. focal AT, micro–re-entry AT, other macro–re–entry AT), they were mapped and ablation was applied aiming at termination of AT.
A waiting period of at least 3 weeks after WLAAI was maintained with the intention to re-evaluate the electrophysiological properties of the LAA before performing LAA ligation. The LARIAT+ system (SentreHeart) was used for LAA ligation. Patients with a history of cardiac surgery, thoracic radiation or pericarditis, and visible pectus excavatum were excluded from further screening to conduct LAA ligation. All patients without exclusion criteria underwent computed tomography (CT) to assess anatomical eligibility of LAA ligation. CT-based exclusion criteria for epicardial LAA ligation were an ostial LAA diameter >45 mm, bilobed or multilobed LAA with diameter >40 mm and LAA orientation behind the pulmonary artery. Electrical isolation of the LAA was reassessed via CS and LAA electrogram properties. Epicardial access was gained via a subxiphoidal puncture with a 21G micropuncture needle (15) and introduction of a microguidewire into the epicardial space. After single transseptal puncture, endocardial and epicardial magnetic tip catheters (findrWIRZ, SentreHeart) were connected at the distal anterior lobe of the LAA and the LARIAT snare was guided over the LAA and closed under fluoroscopic and echocardiographic guidance. A pigtail catheter was inserted into the pericardial space in case of late pericardial effusion and kept at least until the first day after the procedure. Procedural success and safety criteria were defined according to the Munich consensus document for LAA closure (16). A procedure was defined as successful if LAA closure without a leakage >5 mm was proven and no device- or procedure-related complication occurred. Complications were defined as “major” if the following was documented: clinically relevant pericardial effusion leading to hemodynamic instability and requiring interventional or surgical treatment or blood transfusion; bleeding leading to hemorrhagic shock, transfusion, or interventional or surgical therapy; severe pericarditis with recurrent pericardial effusions; vascular access-site complications requiring transfusion or interventional or surgical treatment; emergent surgery during procedure due to perforation or bleeding.
The epicardial catheter was taken out in absence of pericardial effusion. Low molecular-weight heparin was administered in patients on vitamin K antagonist and an INR <2.0 until a therapeutic INR of 2 to 3 was reached. Pre-existing novel oral anticoagulant therapy was reinitiated 6 h after the procedure. After LAA ligation, OAC was continued for at least 3 months after the procedure and then based on the TEE control investigations. OAC was continued in case of documented LA-LAA leakage during follow-up until repeat TEE control after 3 months. Patients were treated with proton pump inhibitors (40 mg once daily for 6 weeks), ibuprofen (600 mg 3× per day for 7 days), and colchicine (500 mg twice a day for 4 weeks) to prevent procedure-related pericarditis.
Clinical follow-up was intended after 3, 6, and 12 months including assessment of the clinical course, OAC status, TEE, and 12-lead and 72-h Holter electrocardiograms (ECG). TEE examination assessed LAA status. An LA-LAA leakage was defined as major when >5 mm in size and as minor when <5 mm in size. In case of documented LAA-LA leakage, OAC was continued until the next TEE control.
Analysis of P-wave characteristics
Measurement of P-wave duration (PWDur) and P-wave dispersion (PWDis) was conducted as described earlier (17). Measurements were repeated at 4 predefined points in time: 1) the day before the first AF ablation procedure; 2) the day before WLAAI; 3) at the post-procedural day after WLAAI/before LAA ligation; 4) at the post-procedural day after LAA ligation. P-wave measurements were conducted at standard 12-lead ECG recordings (speed: 25 mm/s, amplitude: 1 mV/cm, bandpass filter: 0.05 to 40 Hz). PWDur was defined as time difference between the P-wave onset and offset from the equipotential reference line. PWDis was defined as the difference between the maximum and the minimum of the PWDur among 12-lead ECG. All measurements were repeated 3× and the mean value was calculated manually. P-wave analysis was blinded by an independent reader.
Continuous data were summarized as mean ± SD or as median (interquartile range [IQR]), categorical data as count (percentage). Freedom from atrial arrhythmia recurrence was estimated with the Kaplan-Meier method. Survival estimates and the survival curve were shown. Differences in patient characteristic were compared with a 2-sample Student's t-test. A p value of <0.05 was considered statistically significant.
A total of 31 patients (13 female, mean age: 69.7 ± 7.8 years) underwent WLAAI with a following attempt of LAA ligation. Of these, 25 patients (80.6%) suffered from persistent AF, 1 patient (3.2%) from long-standing persistent AF, and 5 patients (16.1%) from paroxysmal AF as the underlying arrhythmia before the first ablation attempt. The median LA diameter was 48 (IQR: 46 to 54) mm. WLAAI was performed after a median of 3 (IQR: 2 to 4) previous ablation procedures. Table 1 depicts detailed patients’ baseline characteristics.
Results of the LAA isolation procedures
A total of 31 consecutive patients with documented persistent PVI underwent WLAAI. During the procedure, 9 patients (29%) were in AF, whereas the remaining 22 patients (71.0%) were in LA AT or LA AT was induced with programmed stimulation or burst pacing (Table 2). In 14 of 31 patients (45.2%), AT or ectopy from the LAA was documented during the ablation procedure. In 29 of 31 patients (93.5%), an LA anterior line and a mitral isthmus line were created resulting in WLAAI. In 1 patient, WLAAI occurred after block of a mitral isthmus line without ablation at the anterior LA wall due to extensive anterior fibrosis. In another patient, LAA isolation occurred after block of an anterior line and additional complex fractionated electrogram ablation along the mitral isthmus without intention to block the mitral isthmus line. Finally, WLAAI was documented at the end of the procedure in all 31 patients. No major complications occurred during the WLAAI procedures. The applied ablation strategies are depicted in Table 2.
Procedural data of LAA ligation
LAA ligation was attempted after 59 (IQR: 39 to 104) days following WLAAI and was successful in 27 of 31 patients (87.1%). In 3 patients it was not successful (1 patient with previously unknown pericardial adhesions, 1 with a superior LAA lobe that was not detected during CT screening, 1 with an ostial LAA diameter >45 mm that was underestimated during CT screening). These patients were converted to endocardial LAA closure with a Watchman device (Boston Scientific Corp., Marlborough, Massachusetts). In 1 patient, the LAA ligation procedure had to be aborted due to intraprocedural right ventricular perforation after subxiphoidal puncture, which resulted in the need for emergent cardiac surgery. The patient underwent surgical LAA closure during surgical closure of the right ventricular perforation site.
Four major periprocedural complications occurred in 3 patients (9.7%) with attempted LAA ligation. One patient suffered from 2 major complications (the above-mentioned heart surgery had concomitant periprocedural stroke); in 2 patients intraprocedural cardiac tamponades with >500 ml blood effusion occurred after established pericardial access. Six patients (19.4%) had mild pericarditis (3 of these with concomitant pleural effusion, 1 with additional minor groin complication). All patients were in SR at the beginning of the LAA ligation procedure. Persistent WLAAI during LAA ligation procedure was demonstrated in 24 of 31 patients (77.4%). In the 7 patients with electrical reconduction into the LAA, the decision was made to attempt LAA ligation based on delayed electrical LAA activation and reduced LAA flow (<0.2 m/s) as demonstrated via TEE during the procedure. In 2 of the patients with electrical LA to LAA reconnection, LAA isolation was registered after successful LAA ligation via absence of electrical activation of the CS region distal to the anterior and mitral isthmus lines. Table 3 shows the data of LAA ligation procedures.
Arrhythmia recurrence after WLAAI and LAA ligation
All 31 patients underwent clinical follow-up with a median duration of 498 (IQR: 159, 791) days after WLAAI (Table 4). Arrhythmia recurrence was documented in 8 of 31 patients (25.8%). Kaplan-Meier estimates for 12- and 24-month arrhythmia-free survivals after WLAAI were 74.7% and 69.7% (95% confidence intervals [CIs]: 60.0% to 92.9% and 53.9% to 90.1%, respectively). There was no difference regarding arrhythmia-free survival for patients undergoing LAA isolation due to AF and AT (p = 0.55) (Figure 1). Analysis after stratification of patients with and without LAA ectopy estimated arrhythmia-free survival of 92.9% (95% CI: 80.3% to 100.0%) for patients with LAA ectopy and 48.3% (95% CI: 27.3% to 85.5%) for patients without LAA ectopy (p = 0.017). Arrhythmia recurrence occurred in 1 of 14 patients with LAA ectopy (7.1%) with 9 of 14 patients (64.3%) being in stable SR off antiarrhythmic drugs. In the patients group without documented LAA ectopy, arrhythmia recurrence was documented in 7 of 17 patients (41.2%) with 6 of 17 patients (35.3%) being in stable SR off antiarrhythmic drugs. Arrhythmia recurrence occurred between 3 and 6 months of follow-up in 5 patients and between 6 and 12 months of follow-up in 3 patients. Notably, persistent arrhythmia recurrence was seen in only 2 of 8 patients with arrhythmia recurrence (1 patient with persistent AF, 1 patient with AT); the remaining 6 patients suffered from paroxysmal, self-sustaining arrhythmias (5 patients with paroxysmal AF, 1 patient with paroxysmal AT). In the patient with persistent AF, a rate-control therapy was applied. The patient with persistent AT underwent repeat ablation procedure after WLAAI and ligation; focal ablation at the ostium of the LAA was conducted to terminate firing from the LAA stump. No further arrhythmia recurrence was documented until follow-up 694 days after repeat ablation.
A total of 12 of 31 patients (38.7%) were on antiarrhythmic drug therapy at the end of the follow-up (5 on class I and 7 on class III antiarrhythmic drugs, 4 of the patients on antiarrhythmic drugs had arrhythmia recurrence). Figure 1 shows the Kaplan-Meier curve for arrhythmia-free survival, and Table 4 shows detailed follow-up data.
Incidence of intracardiac thrombus formation after WLAAI and LAA ligation
There was no documented intracardiac thrombus formation before WLAAI. All patients were on sufficient OAC according to current guidelines (INR: 2.0 to 3.0 in case of vitamin K antagonist or novel oral anticoagulant therapy). LAA thrombus formation was documented after WLAAI in 11 of 31 patients (35.5%) occurring at a median of 32 (IQR: 2 to 50) days after WLAAI (p < 0.001 to incidence prior to WLAAI) (Figures 2 and 3A⇓⇓). LAA thrombus resolved after change of OAC in 10 patients and continue of OAC in 1 patient until conduction of the LAA ligation procedure. Detailed characteristics of patients with documented LAA thrombus are shown in Online Table 1.
Median LAA flow velocity was 0.20 (IQR: 0.15, 0.25) m/s after WLAAI (p < 0.001 to pre WLAAI) (Figure 3B, Online Table 2). LAA flow in patients with documented thrombus formation was not significantly different to LAA flow in patients without thrombus formation (median: 0.18 [IQR: 0.15 to 0.24] m/s vs. 0.20 [IQR: 0.15 to 0.24]; p = 0.37) (Online Table 2). In 9 of 11 patients (81.8%) with LAA thrombus after WLAAI, the LAA was still isolated as assessed during LAA ligation procedure (Online Table 2). Online Figure 2 provides TEE and CT images of patients with documented LAA thrombus.
Following LAA closure, despite discontinuation of OAC in 16 of 31 patients (51.6%), there was no intracardiac thrombus found in any patient and no thromboembolic event occurred (median duration of follow-up: 498 days; p < 0.001 to incidence after WLAAI) (Figures 2 and 3A). Table 4 shows the detailed OAC regime of the patients during the follow-up period.
TEE after LAA ligation was available in 30 of 31 patients (96.8%) after a median duration of 175 [IQR: 94 to 221] days (26 patients with successful LAA ligation, 3 patients with frustrane LAA ligation and successful endocardial LAA closure, and 1 patient with frustrane LAA ligation and surgical LAA closure). Complete LAA closure was demonstrated in 23 of 26 patients with TEE control after LAA ligation (88.5%). In 3 of 26 patients (11.5%), LAA ligation leakages of 5 mm were documented. The 3 patients that underwent endocardial LAA closure and the patient that underwent surgical LAA closure had complete LAA closure.
Average PWDur of all 12 leads was significantly reduced after LAA ligation as compared to PWDur before WLAAI (average PWDur before initial ablation was 118 ± 37 ms, before WLAAI 93 ± 20 ms, after WLAAI/before LAA ligation 78 ± 18 ms, and after LAA ligation 72 ± 20 ms; p = 0.001 comparing post-LAA ligation to pre-WLAAI) (Figure 4). Additionally, PWDur in leads II and III was significantly reduced after LAA ligation as compared to PWDur before WLAAI and after WLAAI/before LAA ligation (Figure 4).
Concordantly, PWDis was significantly smaller after LAA ligation than is PWDis before LAA ligation and after WLAAI/before LAA ligation (PWDis before initial ablation was 77 ± 32 ms, before WLAAI 63 ± 37 ms, after WLAAI/before LAA ligation 51 ± 23 ms, and after LAA ligation 38 ± 16 ms; p = 0.001 comparing post-LAA ligation to pre-WLAAI and p = 0.004 comparing post-LAA ligation to post-WLAAI) (Figure 4).
We report on a highly selective patient cohort consisting of patients with symptomatic arrhythmia recurrence despite multiple catheter ablation procedures and documented PVI. The combination of WLAAI and LAA ligation resulted in a high rate of freedom from arrhythmia recurrence after 24 months of follow-up. Importantly, despite discontinuation of OAC in >50% of patients, there was no intracardiac thrombus formation or cardioembolic events after LAA ligation. Our combined WLAAI/LAA ligation approach resulted in an estimated freedom from arrhythmias in about 70% of the patients after 24 months, and a strong conversion of the incidence of persistent AF to paroxysmal AF in patients that did have a recurrence of AF after WLAAI/ligation.
Arrhythmia-free survival after WLAAI/ligation
The combination of WLAAI and ligation resulted in an acceptable long-term freedom from atrial arrhythmias of more than two-thirds of the patients in a group of patients with AF recurrence after established PVI. There is evidence of an antiarrhythmic effect of WLAAI in patients with AT (8). WLAAI/ligation might lead to mechanical LA debulking, resulting in a reduced mass of electrically active tissue in the atria that can form re-entry wave fronts and lead to sustained AF. Additionally, focal automaticity from the LAA is excluded (18).
Bordignon et al. (11) reported estimated arrhythmia recurrence in 35% of the patients after 12 months and our group documented recurrence in 36.5% after a median of 6.5 months follow-up (10,11) after PVI and WLAAI without LAA ligation. LA to LAA reconnection rates were reported between 27% and 42% after previous successful WLAAI and are believed to be a potential contributor to arrhythmia recurrence (11,12). Although comparison of the different studies is difficult due to different patient cohorts, freedom from arrhythmia recurrence was higher in our patient cohort of PVI nonresponders. The durability of WLAAI in this study was comparable to the above-mentioned 22.6% of the patients showing LAA reconnection. LAA exclusion via ligation with the LARIAT device itself can isolate the LAA and therefore help to maintain both a durable LAA isolation and stable SR in AF patients (14,19–22). We observed repeat WLAAI after LAA ligation in 2 patients with LAA reconnection. An improvement of durability of WLAAI after LAA ligation might be a key factor for the positive outcome of our patients.
Prevention of LAA thrombus recurrence with LAA ligation
We observed a significant reduction in LAA flow velocity and a high incidence of thrombus formation inside the LAA after WLAAI. This is in line with our report of a high incidence of LAA thrombus formation and cardioembolic events after LAA isolation (10). There is no data on the efficacy of LAA ligation in the prevention of intracardiac thrombus formation following WLAAI yet. In the present study, we did not observe a single event of intracardiac thrombus formation or thromboembolism after LAA ligation despite discontinuation of OAC in >50% of patients. Even in patients with prior documentation of LAA thrombus, LAA ligation prevented recurrence of thrombus formation during the follow-up period. This observation suggests that LAA ligation might be a viable option to prevent thromboembolism in patients undergoing WLAAI. Nevertheless, prospective data comparing OAC and LAA closure in patients undergoing WLAAI are missing, but intensified OAC has proven to dissolute intracardiac thrombi in previous studies (23,24). Thus, intensified OAC after documentation of intracardiac thrombi might also be a less invasive alternative to LAA closure for these patients. It also has to be considered that patients with previous LAA thrombus underwent LAA ligation under therapeutic OAC with INR >2.0, potentially increasing the risk of bleeding during epicardial puncture and LAA ligation. This circumstance is mirrored by our 2 cases with cardiac tamponade and the case of right ventricular bleeding that underwent surgical repair. Evaluation of the optimal therapy strategy should consider risks of LAA closure procedure and of bleeding events when long-term intensified OAC has to be applied.
P-wave changes after WLAAI/ligation
PWDis is a marker of inhomogeneous and anisotropic intra-atrial conduction (23–25). Increased PWDur and PWDis is associate with underlying AF and is a predictive marker for the occurrence of paroxysmal AF (25–27). Increased PWDis can predict the onset of AF after coronary bypass surgery and myocardial infarction (26,27). P-wave duration was markedly decreased after circumferential PVI and the attenuation of PWDur could be associated with freedom from arrhythmia recurrence (28,29). Electrical activity from LAA ectopy can be visible in the surface ECG (30). Potential mechanisms of increased PWDis in AF are heterogeneous electrical conduction inside the atria due to fibrosis and delayed LAA activation (25–27,31).
Kawamura et al. (17) demonstrated a decrease in PWDur and dispersion after LAA ligation using the LARIAT device. We report for the first time that WLAAI significantly reduces PWDur and PWDis.
We speculate that this effect is most likely attributed to exclusion of the LAA, which is activated after propagation of electricity from the sinus node, resulting in shorter PWDur and PWDis (31). Additionally, a reduction of the mass of atrial myocardium, which is activated during spreading of the electrical activity, mirroring debulking of atrial tissue due to electrical and/or mechanical exclusion of the LAA, might be an underlying mechanism. As discussed, WLAAI can result in recovery of LA to LAA conduction in a substantial number of patients (12). Additional LAA ligation most likely results in permanent LAA isolation in a higher number of patients, as reflected by changed P-wave characteristics.
Comparison of circumferential and WLAAI
Di Biase et al. (8,18) previously described a different approach to achieve LAA isolation applying circumferential ablation lesions at the LAA base. This approach resulted in improved freedom from arrhythmia recurrence in the multicenter, randomized BELIEF (Effect of Empirical Left Atrial Appendage Isolation on Long-Term Procedure Outcome in Patients With Persistent or Long-Standing Persistent Atrial Fibrillation Undergoing Catheter Ablation) trial (18). Contrary to the findings after WLAAI, increased incidences of LAA thrombus formation and thromboembolism were not reported for circumferential LAA isolation (8,18). Both techniques differ substantially from each other. The most important difference with regard to thrombogenicity might be that WLAAI isolates not only the LAA, but also a large area around the LAA base, potentially resulting in decreased LAA and LA function and increased LAA thrombus formation. The BELIEF trial demonstrated decreased LAA function after circumferential LAA isolation as well (18), so reasons for the differences regarding thrombus formation between these approaches need further studies.
When should WLAAI/ligation be conducted?
All patients in our study suffered from therapy-resistant AF after PVI and can therefore be categorized as “PVI nonresponders.” The optimal ablation strategy for this patient category is not known (5–8). Initial reports suggest the merits of LAA isolation in PVI nonresponders (8,10,11), but the risk of LAA thrombus formation has been a legitimate concern (10). LAA ligation after WLAAI is feasible and has potential beneficial effects of preventing LAA thrombus formation and additive effects on electrical remodeling. WLAAI with subsequent LAA ligation is a viable treatment option in these PVI non-responding patients only after careful consideration of the individual symptoms and risks. Prospective randomized studies are needed to validate our initial findings.
This is a retrospective study with its typical limitations on a limited number of patients. However, the study reflects the experience in a high-volume electrophysiology center of therapy refractory patients, and this is the first report on patients undergoing WLAAI and subsequent LAA ligation.
The combination of WLAAI and LAA ligation in PVI nonresponders led to freedom from arrhythmia recurrence in about two-thirds of the patients after 24 months. LAA ligation successfully prevented recurrence of LAA thrombus formation and thromboembolic events. Significant decreases in PWDur and PWDis occurred with WLAAI/ligation, suggesting favorable electrical remodeling. Future studies are required to determine whether LAA ligation alone can facilitate freedom of AF in PVI refractory patients.
COMPETENCY IN MEDICAL KNOWLEDGE: WLAAI and ligation may improve rhythm control and prevent strokes in PVI nonresponders. Conduction of WLAAI and LAA ligation are complex interventional procedures, which is reflected by a relevant incidence of periprocedural complications. The approach can only be applied by experienced operators who underwent extensive training in interventional electrophysiology. Additionally, infrastructural requirements for centers performing WLAAI and ligation include the availability of a well-functioning complication management and cardiac surgery.
TRANSLATIONAL OUTLOOK: The results of this study need further evaluation in prospective and randomized trials to outline the status of WLAAI for the treatment of AF. More data are needed to estimate the benefit of LAA ligation in terms of improved rhythm control and the prevention of thrombus formation and thromboembolism. Additionally, further improvements have to be made to improve the safety of the LAA ligation procedure.
Drs. Fink and Heeger have received travel grants from SentreHeart. Dr. Tilz has received research grants from Medtronic and Biotronik; travel grants from Biosense Webster, Medtronic, Abbott, SentreHeart, and Daiichi Sankyo; Speakers Bureau/proctor honoraria from Biosense Webster, Medtronic, Abbott, SentreHeart, and Daiichi Sankyo; and consults for Biosense Webster and Biotronik. Dr. Rillig has received travel grants from Biosense, Hansen Medical, EP Solutions, Medtronic, and St. Jude Medical; lecture fees from St. Jude Medical, Medtronic, and Boehringer Ingelheim; and participated in the Boston Scientific EP fellowship. Dr. Mathew has received speaker honoraria and travel grants from Medtronic. Dr. Lee consults for and has equity in SentreHeart, Inc. and Apama, Inc. (Boston Scientific, Inc.). Dr. Kuck has received research grants and personal fees from St. Jude Medical, Medtronic, Biosense Webster, Boston Scientific, Abbott, and Edwards. Dr. Metzner has received speaker honoraria and travel grants from Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Current affiliation for Drs. Heeger and Tilz: Department of Cardiology, Angiology, and Intensive Care Medicine, University Heart Centre Luebeck, University Hospital Schleswig-Holstein, Luebeck, Germany.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- atrial tachycardia
- confidence interval
- coronary sinus
- computed tomography
- international normalized ratio
- interquartile range
- left atrial
- left atrial appendage
- oral anticoagulation
- pulmonary vein
- pulmonary vein isolation
- P-wave dispersion
- P-wave duration
- sinus rhythm
- transesophageal echocardiography
- wide-area left atrial appendage isolation
- Received June 25, 2018.
- Revision received September 18, 2018.
- Accepted September 20, 2018.
- 2018 American College of Cardiology Foundation
- Calkins H.,
- Hindricks G.,
- Cappato R.,
- et al.
- Tilz R.R.,
- Rillig A.,
- Thum A.M.,
- et al.
- Schreiber D.,
- Rostock T.,
- Fröhlich M.,
- et al.
- Fink T.,
- Schlüter M.,
- Heeger C.H.,
- et al.
- Vogler J.,
- Willems S.,
- Sultan A.,
- et al.
- Di Biase L.,
- Burkhardt J.D.,
- Mohanty P.,
- et al.
- Yorgun H.,
- Ugur C.,
- Kocyigit D.,
- Çöteli C.,
- Evranos B.,
- Aytemir K.
- Rillig A.,
- Tilz R.R.,
- Lin T.,
- et al.
- Bordignon S.,
- Perrotta L.,
- Dugo D.,
- et al.
- Reissmann B.,
- Rillig A.,
- Wissner E.,
- et al.
- Sievert H.,
- Rasekh A.,
- Bartus K.,
- et al.
- Badhwar N.,
- Mittal S.,
- Rasekh A.,
- et al.
- Gunda S.,
- Reddy M.,
- Pillarisetti J.,
- et al.
- Kawamura M.,
- Scheinman M.M.,
- Lee R.J.,
- Badhwar N.
- Di Biase L.,
- Burkhardt J.D.,
- Mohanty P.,
- et al.
- Badhwar N.,
- Lakkireddy D.,
- Kawamura M.,
- et al.
- Lakkireddy D.,
- Mahankali A.,
- Kanmanthareddy A.,
- et al.
- Seidl K.,
- Rameken M.,
- Drögemüller A.,
- et al.
- Frenkel D.,
- D'Amato S.A.,
- Al-Kazaz M.,
- et al.
- Lin T.,
- Rillig A.,
- Metzner A.,
- et al.