Thromboembolic stroke is one of the most feared complications of atrial fibrillation (AF). The risk of developing a stroke is increased 5-fold in patients with AF [
In the recent guidelines of the European Society of Cardiology (ESC) in collaboration with the European Association of Cardio-Thoracic Surgery (EACTS), surgical LAA occlusion or exclusion in patients undergoing (thoracoscopic) AF surgery is a Class 2b-B recommendation [
To the best of our knowledge, the WD has never been implanted during a surgical AF ablation. Since the WD has to be implanted under aspirin and heparin, this could possibly lead to an increase in bleeding risk for the surgical procedure. Therefore, this study aimed to determine if WD implantation in a hybrid AF ablation setting (i.e., combination of thoracoscopic epicardial surgical and endocardial catheter ablation) is safe and feasible and can become subject of further testing to serve as a bail-out in cases in which surgical LAA occlusion methods cannot be applied.
In this prospective nonrandomized, single center, pilot study, all consecutive patients scheduled for a hybrid AF ablation between August 2015 and October 2016 were screened. Patients are considered for a hybrid ablation if they fulfill at least one of the following criteria: (1) failure of at least one antiarrhythmic drugs (AAD) class I or III, (2) LA volume of ≥29ml/m2, (3) previous failed catheter ablation, (4) persistent or long-standing persistent AF, or (5) based on patient preference. As per protocol, 10 patients meeting the following inclusion criteria were included after informed consent for implanting the WD was obtained: at least 18 years of age; eligibility for a hybrid ablation; eligibility for at least short-term oral anticoagulation (OAC) therapy; absence of conditions requiring long-term OAC therapy, and CHA2DS2-VASc ≥ 1. Failure or impossibility of surgical LAA occlusion or expectation of being not suitable for surgical LAA occlusion was not an inclusion criterion for this study. All procedures were carried out at the Maastricht University Medical Center, Maastricht, the Netherlands, by the same surgeons (M.L.M. and B.M.) and electrophysiologist (L.P.). The usual learning curve and clinical experience with WD implantation were obtained by the implanting electrophysiologist before the present study. The study was approved by the local ethical committee and registered with ClinicalTrials.gov, number
Two to three days before the ablation procedure, OAC was discontinued. The procedure was performed in a hybrid operating room under general anesthesia with double-lumen endotracheal tube placement for selective right lung ventilation. Transesophageal echocardiography (TEE) was used to confirm absence of LAA thrombi and to guide WD implantation. One 5 mm camera port and two 5 mm workings ports were inserted in the left hemithorax. The pericardium was opened posterior to the phrenic nerve. After blunt dissection of the pericardial reflections of the superior and inferior caval vein, antral isolation of the left and right pulmonary veins (PV), each side as a pair, was performed with a bipolar radiofrequency (RF) clamp (Synergy series, Atricure). Connecting lines between both superior PVs (roof line) and inferior PVs (inferior line) were made epicardially using a unilateral bipolar RF rail device (Coolrail, Atricure), creating a so-called box lesion.
Via a femoral venous approach, a His bundle (St. Jude Medical, St. Paul, MN, USA) and coronary sinus catheter (Medtronic, Minneapolis, MN, USA) were placed under fluoroscopy, followed by full heparinization using a dose of 1000E per 10 kg. After transseptal puncture, an activated clotting time >300 seconds was maintained. Patients who were still in AF after epicardial ablation were electrically cardioverted to restore sinus rhythm. The PVs and box lesion were mapped, and exit and entrance block were checked using a circular mapping catheter (Lasso, Biosense Webster, Diamond Bar, CA, USA). The endpoint for the ablation procedure was bidirectional block in each PV and in the box. In case of incomplete lesions, endocardial touch-up ablation was performed with a 3.5-mm cooled tip RF catheter (SmartTouch, Biosense Webster). If necessary, additional right and/or left atrial lesions were made at the electrophysiologist’s discretion. At the end of the procedure, after the WD was implanted, half of the normally used dosage (2.5 mg) of protamine was intravenously administered to avoid thrombus formation on the WD. A chest tube was placed in the pleural cavity.
According to the standard hospital protocol, the day before the procedure, 300 mg aspirin was administered intravenously. After completion of the epicardial and endocardial ablation, the WD was implanted in the same operating room as the ablation procedure. Heparin was not rebolused before the implantation. As previously described in detail [
Low-molecular-weight heparin in combination with aspirin 80 mg was started 6 hours after the procedure. On the second or third postoperative day, the same OAC regime as used by the patient before the procedure was restarted. In case of use of direct OAC, heparin was stopped. In case of use of coumarine derivatives, heparin was stopped when an adequate INR level was reached. Patients were treated with OAC and aspirin for at least 45 days. If the TEE after 45 days revealed complete closure of the LAA, or a residual peridevice flow ≤5 mm in width (measured using color Doppler), and no thrombus on the device, OAC was discontinued and replaced with clopidogrel 75 mg once daily. Six months after implantation, TEE was repeated. In case of complete closure of the LAA, or a peridevice flow ≤5 mm, clopidogrel was stopped and only aspirin continued indefinitely. In case of thrombus formation on the device or a peridevice flow >5 mm, anticoagulation was continued.
According to the standard hospital protocol, rhythm was determined at 3, 6, and 12 months after the ablation using at least an ECG, but preferably at least a 24-hour Holter examination. In case of symptoms, additional ECGs and/or Holter examinations were performed. Recurrence was defined, according to the guidelines, as any supraventricular arrhythmia (AF, atrial flutter and atrial tachycardia) lasting >30 seconds [
The Primary Safety Endpoint comprised, in accordance with previous WD publications [
The Primary Feasibility Endpoint was device delivery success. This was defined as successful delivery and release of the WD into the LAA, including successful retrieval and redeployment if necessary, and a residual peridevice flow ≤5 mm in width.
Data were prospectively entered into a database. Data were analyzed using SPSS 24.0 (SPSS Inc., Chicago, IL, USA). Continuous variables with normal distribution were presented as mean ± standard deviation (SD); nonnormal variables were reported as median and interquartile range [IR] and categorical variables as frequencies with percentages.
Of the 38 patients who have been screened, 19 patients met the inclusion criteria and were informed about this study. The main reason for exclusion was a CHA2DS2-VASc of 0. Other reasons were contraindication for OAC due to recurrent bleedings, reoperation, kidney dysfunction, and being unable to be present at follow-up. Eleven patients gave informed consent.
The baseline characteristics are summarized in Table
Patient characteristics.
General characteristics | No (%), Mean ± SD or Median [IR] |
---|---|
Male | 5 (50%) |
| |
Age (years) | 65 ± 4 |
| |
Body Mass Index (kg/m2) | 29.5 ± 6 |
| |
CHA2DS2-VASc | 2 [2-3] |
| |
AF characteristics | No (%) or Median [IR] |
| |
Paroxysmal AF | 2 (20) |
| |
Persistent AF | 7 (70) |
| |
Long-standing persistent AF | 1 (10) |
| |
Atrial Flutter | 3 (30) |
| |
AF duration (months) | 33 [21-94] |
| |
Echocardiography | No (%) or Mean ± SD |
| |
Left ventricular ejection fraction (%) | 56 ± 8 |
| |
LA size (mm), LA volume (mL), LA volume index | 47 ± 6, 99 ± 22, 50 ± 13 |
| |
RA volume (mL) | 75 ± 31 |
| |
Moderate to severe valvular disease | 0 |
SD = Standard Deviation, IR= interquartile range, AF = Atrial Fibrillation, OAC = Oral Anticoagulation, LA = Left Atrium, RA = right atrium.
The Primary Feasibility Endpoint was met in 10/11 procedures. In three patients, the LAA consisted of a single lobe, in 6 patients of 2 lobes, and in one patient of multiple lobes. One patient had to be excluded before the implantation procedure was started because the LAA was insufficiently visible on TEE, so the measurements necessary for the implantation could not be performed. All other 10 devices could be delivered successfully and the residual peridevice flow was ≤5 mm in width in all patients at 45 days and 6 months after implantation [Figure
No major complications (as defined in previous WD publications [
After 6 weeks none of the TEEs showed thrombus formation or migration of the device. The peridevice flow was ≤5 mm in all cases, with a mean of 1.2±1.5 [range 0-4]. OAC was stopped and clopidogrel started in all patients except one (tachycardiomyopathy).
At follow-up at 3 months, 5 patients received a 24-hour Holter and 3 patients a 7-day Holter. In 2 patients only ECGs were made. No recurrences were documented. Three patients were using AADs. In one patient, OAC was restarted due to side effects of clopidogrel.
After 6 months, none of the TEEs showed thrombus formation or device migration and the peridevice flow was ≤5 mm in all cases, with a mean of 1.1±1.4 [0-3]. Six patients showed complete closure of the LAA. In all patients a Holter was performed: 6 24-hour Holters, 1 48-hour Holter, and 2 7-day Holters; and in one patient a Reveal was implanted. OAC was restarted and aspirin stopped in a patient with hyperthyroidism and AF recurrences, for which a redo ablation procedure was considered necessitating the use of OAC. In all other patients, OAC or clopidogrel was stopped. While one other patient suffered recurrences of atrial tachycardia, the other 8 patients maintained sinus rhythm, of which 3 were with the use of AAD.
A 24-hour Holter after 12 months was performed in 4 patients, a 48-hour Holter in one, a 7-day Holter in 2. One patient was monitored via the reveal, in one patient a pacemaker was implanted, and in one patient only ECGs were made. Eight patients maintained sinus rhythm, all without the use of AAD. During the entire follow-up, no stroke occurred.
This study is the first to show that it is feasible and safe to implant a WD during a hybrid AF ablation. The device was implanted successfully in all patients without major complications. TEE after 6 months’ follow-up showed either persistent closure of the LAA or a peridevice flow of ≤5 mm.
In accordance with the ESC/EACTS guidelines [
During surgical treatment of AF, it is common practice to exclude the LAA epicardially. Due to high occurrence of unsuccessful closure with ligation, (stapler) resection and suture closure, the AtriClip was developed. Its safety and efficacy, in occluding the LAA and reducing stroke, have been reported [
In the PROTECT AF trial, safety, defined as 7-day procedure- and device-related complications, was a concern (8.7%). However, in the PREVAIL trial and EWOLUTION registry, the procedural safety had significantly improved (4.2% and 2.8%, respectively). Procedural and device-related strokes decreased from 1.1% in PROTECT AF to 0.4% in PREVAIL (p = 0.007) and 0.1% in EWOLUTION, while successful implantation increased from 88% in PROTECT AF to 98.5% in EWOLUTION. This is commonly seen in new interventional procedures and mainly depends on operator experience. The current result of 100% implantation success in this study is therefore explained by the experience of the implanting physician (L.P.). One of the other concerns about the WD is the acceptance of a Doppler peridevice flow up to 5 mm in width as a marker for successful implantation, because this might imply a source of thrombus formation. However, looking at published stroke rates applying this marker, this does not seem to be thrombogenic. A possible reason could be the small ostium which gives a high flow velocity and therefore less thrombogenic risk. Finally, an additional problem when implanting the WD during a hybrid AF ablation can occur. The bleeding risk might be increased since the surgical procedure is performed under aspirin, and the administered heparin for the endocardial part of the ablation is not fully antagonized at the end of the procedure. In our series, we report 1 patient needing transfusion, compared to <5% in procedures without WD implantation. Other complications are not increased compared to hybrid ablation using an AtriClip [
Due to the pilot nature of the study, only a limited number of patients were included. More data are needed to be able to draw definitive conclusions.
Implantation of the endocardial Watchman device seems to be feasible and safe in the setting of a hybrid AF ablation and could be an alternative to epicardial exclusion and occlusion methods in surgical AF ablation procedures. Larger studies are required to confirm these findings.
The data used to support the findings of this study are restricted by the ethical board MUMC/AZM in order to protect patient privacy.
To support the research presented herein Laurent Pison reports a grant from Boston Scientific to the Cardiovascular Research Institute Maastricht (CARIM) at the University of Maastricht. Dr. La Meir reports a relationship with AtriCure and Medtronic.