Value of detecting peri‐device leak and incomplete endothelialization by cardiac CT angiography in atrial fibrillation patients post Watchman LAAC combined with radiofrequency ablation

Abstract Objectives To explore the value of detecting the peri‐device leak (PDL) and device endothelialization after left atrial appendage closure (LAAC) by cardiac computed tomography (CT) in patients with atrial fibrillation (AF), who underwent Watchman LAAC combined with radiofrequency ablation of atrial fibrillation (AFCA). Methods Patients with symptomatic drug‐refractory atrial fibrillation at high risk of stroke (CHA2DS2‐VASc Score ≥ 2), who underwent Watchman LAAC combined with AFCA in our center from March 2017 to December 2018 were enrolled. Maximum diameter of LAA orifice was determined by preoperative CCTA. A standardized view of Watchman device was obtained by postoperative CCTA multiplannar reconstruction to evaluate the PDL and device endothelialization. Results Approximately 84 patients post successful LAAC and AFCA were enrolled in this study. The satisfactory LAA occlusion rate was 100%. There was no death, bleeding, stroke, and device‐related thrombus (DRT) events. At 6‐month postprocedure, CCTA images evidenced complete endothelialization in 44 patients (no contrast enhancement in LAA); contrast enhancement in LAA and visible PDL in 33 patients; contrast enhancement in LAA but without PDL in seven patients (incomplete device endothelialization). Maximum diameter of LAA orifice could independently predict the occurrence of PDL (odds ratio, 1.31; 95% confidence interval, 1.11–1.55; p = .002), sensitivity was 69.7% and specificity was 80.4% with the cutoff value of maximum diameter of LAA orifice more than 28.2 mm on predicting PDL. Conclusions CCTA is feasible to evaluate PDL and device endothelialization after LAAC. The maximum diameter of LAA orifice derived from CT can independently predict the occurrence of post‐LAAC PDL.


| INTRODUCTION
Stroke prevention belongs to the key management among patients with atrial fibrillation (AF). Left atrial appendage (LAA) is the remnant of original left atrium (LA) in the period of embryo, and is a known major location of thrombosis in AF patients. 1 Percutaneous endovascular left atrial appendage closure (LAAC) is increasingly performed in AF patients, especially those with contraindications to long-term oral anticoagulants (OAC). This clinical practice conforms to the European Society of Cardiology guidelines with Class IIB recommendation for LAAC in AF patients with high-stroke risk and contraindications to long-term OAC. 2 However, previous studies have revealed that postoperative peri-device leak (PDL) might occur in more than 40% of cases post-LAAC. 3 Presence of PDL indicates the continued participation of LAA in the system circulation, which might still be linked with the potential risk of future stroke despite LAAC.
Recently, some researchers evidenced the presence of contrast enhancement, a sign of incomplete endothelialization of the device, post-LAAC by cardiac computed tomography (CT). 4 The determinants of PDL and device incomplete endothelialization post LAAC as well as the relationship between PDL and device incomplete endothelialization remain elusive now. In this study, we sought to evaluate the prevalence of PDL  All patients were included with the criteria as: age more than 18 years; symptomatic nonvalvular AF refractory to antiarrhythmic drugs; and with CHA 2 DS 2 -VASc Score more than or equal to 2 plus one of the following situations: (1) high bleeding risk (HAS-BLED Score ≥ 3); (2) history of stroke or systemic embolic even under OAC treatment; (3) intolerance to chronic OAC because of minor bleeding caused by anticoagulation therapy; and (4) preference for LAAC device implantation as an alternative to long-term OAC despite adequate information. The study protocol was approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. Informed consent was obtained from each patient.

| Procedure planning
All the 84 patients underwent a series of preoperative examinations including relevant laboratory tests, 12-lead electrocardiogram (ECG), transthoracic echocardiography (TTE), transesophageal echocardiography (TEE), and CCTA. CCTA and TEE images were routinely acquired to measure the maximum diameter of LAA orifice and exclude LA or LAA thrombus.

| Procedure
Catheter ablation of AF (AFCA) and LAAC were performed via femoral venous access under local anesthesia, and heparin was used to achieve a target activated coagulation time of more than 250 s. LAAC procedure was under local anesthesia. The orifice diameter and depth of the LAA were measured by TEE before procedure. TEE was introduced under deep sedation to reconfirm the position of the device before release. A mean left atrial pressure above 10 mmHg was obtained before measurement.
Transeptal puncture was done by standard fluoroscopy, and the X-ray images were taken in right anterior oblique (RAO) 45°. Only WATCHMANTM 2.5 (Boston Scientific) devices were used. The device with appropriate size was chosen when the depth was allowed, generally 4-6 mm larger the maximum diameter measured by TEE before procedure. The device was then advanced into the delivery sheath and deployed by sheath retraction guided by fluoroscopy. Preliminary assessment was performed by angiography and tug test under fluoroscopy to check the device position and stability. TEE was then performed to reconfirm the position with minimal (<5 mm) to no PDL. The device was released if it was verified by the assessment of PASS criteria. The specific details were described as our previous essay. 5 Each patient received OAC therapy during follow-up.

| Follow-up
All patients were required to accept follow-up at least twice within 6 months after therapy. TEE was performed at 3 months of follow-up to observe if satisfactory occlusion (no PDL or with PDL < 5 mm) was maintained. In case of successful occlusion, OAC was discontinued and patients were recommended for dual antiplatelet therapy (DAPT) for another 3 months followed by aspirin. Otherwise, the original OAC was continued and a repeated TEE was performed within 3 months. The second follow-up was at 6 months including a 12-lead ECG, Holter and CCTA. 12-lead ECG and Holter were examined to observe the recurrence of AF, and CCTA was reapplied to evaluate postoperative PDL and device endothelialization. The average linear attenuation coefficient less than 100 Hu was defined as LAA complete occlusion. 7      with PDL whose width less than 5 mm and seven patients with transfabric leak were identified by CCTA at 6 months follow-up. The detailed TEE and CCTA results are listed in Table 2.

| Statistical analyses
At 3 months TEE, 21 patients with PDL whose width less than 5 mm were identified, among them, PDL was detected in 18 patients (85.7%) by CCTA at 6 months follow-up (Table 3). Totally, CCTA at 6 months detected 33 patients with PDL whose width less than 5 mm and seven patients with trans-fabric leak, among them, PDL was detected in 18 patients (45.0%) by TEE at 3 months follow-up, and PDL was absent in 22 patients (55.0%) at 3 months postprocedure as detected by TEE (Table 4).

| Peri-device leak
In 51 cases (60.7%) there was no PDL detectable by CCTA. A leak with a width of less than 3 mm was present in 31 cases (36.9%), and a leak width of 3-4.9 mm was present in two cases (2.4%).
None were found to have a severe leak more than 5 mm in width.
Baseline characteristics according to the presence or absence of PDL are listed in Table 5.  CCTA examination before LAAC can assist the 3D electrophysiological mapping system to construct the LA model during catheter ablation. 10 In addition, preoperative CCTA can measure the T A B L E 2 Characterization of LAA closure results by 3-month postoperative TEE and 6-month postoperative CCTA

| Maximum diameter of LAA orifice and PDL
Imprecise assessment of LAA orifice, for example, the selection of too small LAAC device size based on TEE measurement, at preoperative phase is one reason for the occurrence of PDL after the procedure.
Other factors linked to PDL are also suggested. 14 LAA used to be considered to possess enough compliance to accommodate a larger size LAAC device. However, the remodeling of LA and LAA in AF patients may result in reduced compliance of LA and LAA, particularly in response to a compressible LAAC device that is deployed at relatively low radial force. Histological studies showed that the LAA specimens of AF patients existed obvious dilatation, stretching and reduction of pectinate muscle volume. 15 In addition, most patients with chronic AF will exhibit significant thickening of endocardium and deranged thickening elastic fibers extending to the epicardium. 16 All these changes will reduce the elasticity and compliance of LAA orifice, which increases the risk of postoperative PDL due to mismatch of LAAC device and LAA orifice. For those patients with larger size of LAA orifice and higher risk of PDL, the application of Watchman device may be suboptimal and relate to incomplete LAAC, since the available maximum diameter of the Watchman LAAC device is 33 mm now. Larger Watchman LAAC device is therefore warranted in the future to cover the need of these patients.
T A B L E 5 Comparison between patients with and without peri-device leak

| Limitation
In our patient cohort, there were no patients with significant PDL ≥ 5 mm, PDL > 3 mm was found in only 2.4% (2/84) patients, which is likely to limit the strength of ROC analysis. Moreover, the clinical significance of PDL is still controversial. Long term follow-up results are controversial now on the relationship between PDL and risk of stroke and DRT. 19 Our planned upcoming large-scale and long-term clinical follow-up and prospective control study are helpful to answer this question.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.