I. Introduction
In 2019, the Guidelines for non-pharmacotherapy of cardiac arrhythmias (JCS/JHRS 2019) were published, covering areas such as treatment using cardiac electrical implantable devices (CIEDs), catheter ablation, surgical treatments, and therapies using left atrial appendage occlusion devices.1
According to the recent dramatic development of non-pharmacotherapy in the field of cardiac arrhythmias, major revisions had been made to the JCS/JHRS 2019 guidelines, owing to critical evaluation by many clinicians. Following the publication of the guidelines, however, more, important clinical evidence had been established in both Japan and abroad, and new concepts of treatment have been created. As updating and disseminating new information directly affects daily clinical practice, we decided to publish “The focused updated guidelines on non-pharmacotherapy of cardiac arrhythmias (JCS/JHRS 2021)” focusing on fields with significant progression, instead of waiting for the next revision of the entire guidelines.
Our standards for selecting the issues in the focused update guidelines were as follows: (1) therapy with a previously undetermined recommendation level that has become specified because of new evidence from both Japan and abroad, (2) newly established therapeutic concepts, and (3) important issues in which omission in the previous guidelines was inevitable because of word limits. Accordingly, the main features of these focused updated guidelines are as follows.
1. Leadless Pacemakers
Although a large international clinical trial to test the utility of leadless pacemakers had been published when the previous guidelines were announced, we did not address the recommendation because it had been suggested that the incidence of complications owing to the small body size of Japanese patients might be higher than in Western patients. However, a subanalysis of a large clinical trial based on Japanese data,2
and our subsequent clinical experience, allowed us to describe the recommendation level.
2. Conduction System Pacing
The novel anti-heart failure pacing therapy (i.e., conduction system pacing [His bundle pacing]) has appeared as a novel but different concept to cardiac resynchronization therapy (CRT). When the previous guidelines were announced, there were no guidelines or expert consensus that clarified the recommended therapeutic level; however, the new ACC/AHA/HRS guidelines in 2019 stated a recommended level,3
and the initial concept of His bundle pacing was extended to conduction system pacing including both His bundle and left bundle pacing. Therefore, based on the present consideration, we describe recommended levels of conduction system pacing.
3. Management of Implantable Cardioverter Defibrillators (ICDs) in End-of-Life Care
The recently established concept of advance care planning (ACP) has facilitated the process of discussing and making decisions with healthcare patients for their preferred goals of care if they lose their communication ability in the future. When facing the end of life, ICD therapy (especially shock therapy) is not always desired by patients or their families; therefore, the decision to deactivate the ICD therapy can be acceptable. With the overlap in the timing of publishing JCS/JHFS 2021 Statement on Palliative Care in Cardiovascular Diseases,4
we decided to describe the class recommendations based on the evidence level and decision-making process in this focused update. In association with ACP, we also mention the indications for ICDs in elderly patients.
4. Transvenous Lead Extraction
The description of several important aspects of transcutaneous lead extraction (i.e., bacteremia without lead infection, superficial infection around the subcutaneous pocket, antibiotic therapy, and CIED re-implantations after lead extraction) were excluded in the previous guidelines but are addressed in the focused update.
5. Antitachycardia Pacing (ATP) Therapy for Atrial Tachyarrhythmias
Although large, international clinical trials demonstrating the efficacy of ATP (capabilities to terminate atrial fibrillation [AF] and prevent the aggravation of AF into a persistent type) had been published in other countries when the previous guidelines were announced, we did not address this therapy because of the absence of clinical evidence from Japan and because of word limits. Taking the novel Japanese evidence of ATP for AF with CIEDs5
into account, however, we decided to include ATP in the focused update.
6. Management of AF Detected by CIEDs
The JCS/JHRS 2020 guidelines on pharmacotherapy of cardiac arrhythmias addressed the clinical significance of asymptomatic (silent) AF and advanced methods of documenting AF.6
In response to the guidelines, we prescribe appropriate management of AF detected by CIED monitoring (focusing on anticoagulant therapy especially) in the focused update.
7. Challenges in Reducing Radiation Exposure
The development of 3D mapping systems has largely contributed to identifying the precise position of catheters in the cardiac chambers without fluoroscopic guidance, thereby reducing the exposure to radiation during catheter ablation procedures. Problems due to radiation injury had been mentioned in previous guidelines; however, to promote awareness and knowledge of reducing the radiation exposure of clinicians, we focused on the use of advanced 3D mapping systems to reduce exposure dose. Besides intraoperative radiation exposure, we described unignorable radiation exposure by preprocedural computed tomography (CT) imaging.
8. Novel Evidence and Advanced Technologies for Catheter Ablation of AF
According to the CABANA study,7
a new large randomized clinical trial that tested the prognostic efficacy of catheter ablation in patients with AF, we reconsidered the class recommendations for catheter ablation of asymptomatic AF. Further, we were able to establish the class recommendations for catheter ablation in AF patients with heart failure based on even higher levels of evidence (i.e., a large randomized clinical trial and meta-analysis including several important clinical studies8). Furthermore, the efficacy of early rhythm-control therapy for AF, either with antiarrhythmic drugs or catheter ablation (pulmonary vein isolation using the cryoballoon technique), has been reported by several large randomized clinical trials.9–11
Taking these results into account, we describe the significance of early rhythm-control to prevent deterioration of AF. As for the new technologies of catheter ablation, high-power radiofrequency delivery with a short duration and the expanded use of the cryoballoon technique for persistent AF are addressed.
9. Perioperative Anticoagulation Therapy for AF Ablation
The management of anticoagulation therapy during the peri-catheter ablation period has been rewritten because various important clinical evidence has been documented in both Japan and abroad,12–18
and because a supplemental description was needed according to the contents of the Guidelines on non-pharmacotherapy of cardiac arrhythmias (JCS/JHRS 2019).6
10. Left Atrial Appendage Closure Device
Because percutaneous transcatheter therapies for left atrial appendage occlusion were not reimbursed in Japan when the previous guideline was announced, detailed description of this technology was postponed. However, sufficient clinical experience has since accumulated in both Japan and abroad, to understand the appropriate indications, safety, postoperative management, and long-term efficacy of this therapy.19
Therefore, we describe the class recommendations for this novel technology in the present focused update.
This set of guidelines recommends indications for non-pharmacotherapy of arrhythmia based on the latest findings and evidence. We first surveyed materials based on evidence from the USA and Europe, then further critically examined the levels of evidence, collected information available in Japan, and examined all material based on the experiences and opinions of members and collaborators in the Joint Working Group. This revision adds new knowledge acquired from advances in diagnostic techniques and treatment methods, or recently reported important evidence, while considering consistency with each of the previously reported guidelines published by the JCS/JHRS Joint Working Group.
The recommendation of classes and evidence levels used in this set of guidelines conform to those of the American Heart Association (AHA), American College of Cardiology (ACC), and Heart Rhythm Society (HRS) guidelines.20
The recommended class of indications for each diagnosis and treatment method is classified as I, IIa, IIb, or III, and the level of evidence is classified as level A, B, or C (Tables 1,2). The guidelines also state the grade of recommendation and level of evidence based on the “Medical Information Network Distribution Service (MINDS) Handbook for Clinical Practice Guideline Development 2007”,21
published by the MINDS Evidence-based Medicine dissemination promotion project as a guideline preparation method (Tables 3,4). However, the MINDS Grade of Recommendation and Level of Evidence should be used only as a reference, as this system regards the evidence level in a fundamentally different manner. The MINDS level of evidence (levels of evidence in literature on treatment) is a classification based on research design, and the highest level was adopted when multiple papers were considered.
Table 1.
Class of Recommendation
| Class I |
Evidence and/or general agreement that a given procedure or treatment is useful and effective |
| Class II |
Conflicting evidence and/or a divergence of opinion about the usefulness/efficacy of the given procedure
or treatment |
| Class IIa |
Weight of evidence/opinion is in favor of usefulness/efficacy |
| Class IIb |
Usefulness/efficacy is less well established by evidence/opinion |
| Class III |
Evidence or general agreement that the given procedure or treatment is not useful/effective, and
in some cases may be harmful |
Table 2.
Level of Evidence
| Level A |
Data derived from multiple randomized clinical trials or meta-analyses |
| Level B |
Data derived from a single randomized clinical trial or large-scale non-randomized studies |
| Level C |
Consensus of opinion of the experts and/or small-size clinical studies, retrospective studies, and registries |
Table 3.
MINDS Grades of Recommendation
| Grade A |
Strongly recommended and supported by strong evidence |
| Grade B |
Recommended with moderately strong supporting evidence |
| Grade C1 |
Recommended despite no strong supporting evidence |
| Grade C2 |
Not recommended because of the absence of strong supporting evidence |
| Grade D |
Not recommended as evidence indicates that the treatment is ineffective or even harmful |
The grade of recommendation is determined based on a comprehensive assessment of the level and quantity of evidence, variation of conclusion, extent of effectiveness, applicability to the clinical setting, and evidence on harms and costs. (From MINDS Treatment Guidelines Selection Committee, 2007.21)
Table 4.
MINDS Levels of Evidence (in Literature on Treatment)
| I |
Systematic review/meta-analysis of randomized controlled trials |
| II |
One or more randomized controlled trials |
| III |
Non-randomized controlled trials |
| IVa |
Analytical epidemiological studies (cohort studies) |
| IVb |
Analytical epidemiological studies (case-control studies and cross-sectional studies) |
| V |
Descriptive studies (case reports and case series) |
| VI |
Not based on patient data, or based on opinions from a specialist committee or individual specialists |
(From MINDS Treatment Guidelines Selection Committee, 2007.21)
This set of guidelines is designed to be used as a reference by doctors diagnosing and treating diseases in clinical practice, but the final decision should be made by the attending physicians after ascertaining the patient’s condition. Even when selecting a diagnosis or treatment that does not follow the guidelines, the decision of the attending physicians should be prioritized in consideration of the individual patient’s situation. In actual clinical settings, it is more important for the attending physicians to make the judgment after considering the clinical background and social situation of each patient fully while complying with the guidelines.
II. Cardiovascular Implantable Electronic Devices (CIEDs)
1. Pacemakers
1.1 Leadless Pacemaker (Table 5)
Table 5.
Recommendations and Evidence Levels for VVI Leadless Pacemaker
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
For patients with symptomatic bradycardia AF, in whom venous access should
be preserved, or with venous occlusion or stenosis, VVI leadless pacemaker
is recommended |
I |
B |
B |
III |
For non-AF bradycardiac patients with condition that precludes the use of a
transvenous pacemaker, including compromised venous access and the need
to preserve venous access, VVI leadless pacemaker should be considered |
IIa |
B |
C1 |
III |
For patients who underwent CIED extraction due to infection and completed
the antibiotic treatment, VVI leadless pacemaker may be considered |
IIb |
C |
C1 |
IVa |
AF, atrial fibrillation; CIED, cardiac implantable electronic device; COR, class of recommendation; GOR, grade of recommendation; LOE, level of evidence.
Although the leadless pacemaker is an emerging pacing technology, the clinically available mode of a leadless pacemaker in Japan, as of January 2021, is VVI; therefore, an appropriate indication should be considered. Symptomatic bradycardic atrial fibrillation (AF) is a Class I indication. Patients with atrioventricular (AV) block without AF or sinus node dysfunction might benefit from the leadless VVI pacemaker only when the risk of implantation of the atrial lead is higher than its benefit, the patient has severe frailty, is bed-ridden, or has less than 1-year survival.
An investigational device exemption (IDE) study included 36 patients from Japan.2
Albeit a smaller body status, the safety and efficacy endpoints in Japanese patients were comparable to those in patients from the rest of the world. Cardiac perforation occurred in approximately 1%, and 15% of them required surgical repair.2
Given that the risk factors for cardiac perforation include female sex, old age (≥85 years), chronic respiratory disease, body mass index <20 kg/m2, congestive heart failure, and steroid use, a typical candidate for a leadless pacemaker may be high-risk.22,23
Hence, a careful risk evaluation is important.
The leadless pacemaker is MRI conditional (1.5 and 3 T).24
There are no established data on radiation therapy, but a case study reported that up to 30 Gy radiation therapy in a patient with mediastinal malignancy with a leadless pacemaker inside the radiation field yielded no remarkable damage to the pacemaker.25
Nevertheless, it is recommended that electrical parameters should be checked prior to and after radiation therapy or MRI.24,25
1.1.1 Leadless Pacemaker in Hemodialysis Patients
In hemodialysis patients, the leadless pacemaker is often chosen due to the venous occlusion caused by previous hemodialysis catheters, to improve the patency of an arteriovenous fistula, or to minimize the infection rate. In 3 studies of the Micra transcatheter pacing system (IDE, Micra Transcatheter Pacing System Continued Access Study, and Post-Approval Registry), 201 out of 2,819 patients (7%) were under hemodialysis.26
A majority of the patients (72%) had conditions that precluded transvenous pacing, including the need to preserve venous access (79%), prior infection (20%), and venous occlusion (17%). A successful implant was achieved in 98%, and safety and electrical parameters were similar to non-hemodialysis patients. Although an infection rate of 8% has been reported with conventional pacemakers,27
no infection was reported in these patients.
1.1.2 Extraction
Worldwide experience with successful retrieval of the leadless pacemaker has been reported.28
Data from the manufacturer recorded 40 successful retrievals, and operators for 29 successful retrievals provided information. Of the 29 retrievals, 11 were during the index procedure; 18 retrievals were performed after a median of 46 days (1–95 days). Reasons included threshold increment after tether removal (n=5), loss of capture (n=3), dislodgement (n=3) for immediate retrieval, and elevated threshold at follow-up (n=11), endovascular infection (n=1), need for transvenous device (n=2), and CRT upgrade (n=1) for delayed retrieval. Average time of the retrieval and fluoroscopy was 63.11 min and 16.7 min, respectively, and no adverse events were observed.
1.1.3 Infection
a. Leadless Pacemaker Implantation Following CIED Infection
Of the 1,820 patients in the Micra post approval registry, 105 had re-implantation of leadless pacemaker within 30 days after CIED extraction.29
The extracted CIEDs were pacemakers (70.5%), CRT-P (9.5%), and ICD/CRT-D (12.4%), and 93% were complete explants. Pacing-dependent patients underwent a same-day implant more frequently, and non-dependent patients underwent the implant after a median of 7 days; 91% had IV antibiotics pre-implantation, and 42% had them post-implantation as well. During an average of 8.5 months, 2 patients died of sepsis, and 4 patients required a system upgrade, but no leadless pacemaker was explanted due to re-infection. This finding might suggest that a leadless pacemaker is reasonably safe with no recurrent device infection, which may be due to the absence of a subcutaneous pocket, a smaller surface area as compared with the transvenous lead, a greater tendency of encapsulation, and a high-flow environment in the heart cavity.
b. Bacteremia/Endocarditis Following Implantation of a Leadless Pacemaker
A total of 16 out of 720 patients (2.2%) in the IDE study developed 21 serious cases of bacteremia or endocarditis during follow-up.30
Infection occurred at a mean of 4.8 months after implantation; 13 events were caused by Gram-positive organisms and 3 by Gram-negative bacteria. Two of the three cases of endocarditis resulted in death; one patient with prosthetic aortic valve endocarditis died immediately postoperatively, the other with aortic valve endocarditis 108 days post leadless pacemaker implantation refused surgical intervention. All but 2 patients responded well to antibiotics, and no persistent bacteremia was observed after antibiotic cessation.
1.1.4. Future Pacing Mode (VDD)
MicraTM, the clinically available leadless pacemaker in Japan as of January 2021, provides accelerometer-based rate-adaptive pacing. Accelerometer signals identify 4 distinct segments of cardiac activity: isovolumic contraction and mitral/tricuspid valve closure (A1), aortic/pulmonic valve closure (A2), passive ventricular filling (A3), and atrial contraction (A4). Using the downloadable algorithm with accelerometer-based atrial sensing, the MARVEL study showed that AV synchrony improved in patients with complete AV block from 37.5% to 80% compared with VVI mode.31
An improved automated algorithm was evaluated in patients with sinus rhythm with complete AV block in the MARVEL 2 study.32
Median AV synchrony significantly improved in VDD mode compared with VVI mode (27% vs. 94%). AV synchrony ranged from 89.2% during the resting period to 69.8% while standing. The decline in AV synchrony could be due to orthostatic tachycardia or a decrease in the A4 signal because of reduced venous return.
1.2 Conduction System Pacing (Table 6)
Table 6.
Recommendations and Evidence Levels for His Bundle Pacing
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
| Indication for His bundle pacing for patients with bradycardia |
In patients with atrioventricular block who have an indication for permanent
pacing with an LVEF between 36% and 50% and are expected to require
ventricular pacing over time, it is reasonable to perform His bundle pacing |
IIa |
A |
B |
II |
In patients with atrioventricular block who have an indication for permanent
pacing with a normal LVEF, His bundle pacing may be considered |
IIb |
C |
C1 |
III |
| Indication for His bundle pacing as an alternative method for CRT |
In patients with heart failure who have an indication for CRT but not
cardioverter -defibrillator, His bundle pacing may be considered when
trans-cardiac vein pacing is ineffective or impossible for any reason |
IIb |
C |
C1 |
VI |
COR, class of recommendation; CRT, cardiac resynchronization therapy; GOR, grade of recommendation; LOE, level of evidence; LVEF, left ventricular ejection fraction.
Long-lasting right ventricular (RV) pacing produces an iatrogenic desynchronized contraction of the left ventricle (LV), and possibly induces reduced contraction and mechanical remodeling of the LV.33,34
To resolve this issue of RV pacing, several animal and clinical studies have demonstrated that artificial pacing of the His bundle instead of the RV can provide long-term physiological ventricular conduction.35–37
However, when this concept was initially proposed, permanent and stable His bundle pacing was technically difficult at that time, and therefore, this method was not widely used.
Recently, however, a newly designed pacing lead and delivery system aimed at pacing the His bundle has emerged and provided >80% success rate,38–43
and the method has been taken up very rapidly since. More recently, this concept has been further developed to enable transseptal left bundle pacing,44
and His bundle and left bundle pacing have become collectively known as “conduction system pacing”.
1.2.1 His Bundle Pacing for Patients With Bradycardia
Several clinical studies have demonstrated the efficacy of His bundle pacing for patients with bradycardia (mainly atrio-ventricular block) who have an indication for permanent pacing with a normal or moderately reduced LV ejection fraction (LVEF) (Table 7).38–43
Two randomized clinical trials compared the efficacy of RV pacing to that of His bundle pacing using a crossover design (switching from one pacing method to the other within a fixed period),38,41
and demonstrated that His bundle pacing improved LVEF. However, the results from the 2 studies were not consistent regarding the improvement in NYHA functional class and 6-min walk distance. Two clinical observational trials, in which the patients were separately assigned to His bundle pacing at a certain hospital and RV pacing at a second hospital, showed an improvement in LVEF39
and reduction in heart failure hospitalizations.42
Furthermore, a meta-analysis accumulating data from many trials revealed LVEF improvement in patients with a preoperative LVEF <50%.43
To date, however, the efficacy of His bundle pacing has not been confirmed by large randomized clinical trials; previous clinical studies were small-scale, and did not show an improvement in the mortality rate as a single primary endpoint.
Table 7.
Summary of Studies of His Bundle Pacing
| Study design |
No. of
patients |
FU
(months) |
Indication
of pacing |
LVEF
(pre-HBP) |
LVEF
(post-HBP) |
LVEF
(post-RVP) |
Success
rate of
HBP |
Results |
| Occhetta et al, 2006, JACC38 |
Randomized,
crossover,
RVP vs. HBP |
18 |
12 |
AVJ ablation
for AF |
51.3% |
53.4% |
50.0% |
95.8% |
HBP improved NYHA
class, 6-min walk distance,
QOL, and hemodynamic
status |
| Sharma et al, 2015, Heart Rhythm39 |
Observational,
RVP vs. HBP |
192;
RVP 98,
HBP 94 |
24 |
AVB 62%,
SSS 38% |
56.0% |
UD |
UD |
80.0% |
HBP improved the
incidence of an admission
due to heart failure, but not
the mortality rate |
| Vijayaraman et al, 2015, JACC Clinical Electrophysiology40 |
Observational,
single arm |
100 |
19 |
AVB |
54.0% |
UD |
None |
84.0% |
Significant increase in the
pacing threshold, pacing
failure was observed in 5% |
| Kronborg et al, 2014, Europace41 |
Randomized,
crossover,
RVP vs. HBP |
38 |
12 |
AVB |
50.0% |
55.0% |
50.0% |
UD |
HBP improved LVEF, but
not NYHA class, 6 min
walk distance, and QOL |
| Abdelrahman et al, 2018, JACC42 |
Observational,
RVP vs. HBP |
765;
RVP 433,
HBP 332 |
24 |
AVB 65%,
SSS 35% |
54.5% |
UD |
UD |
91.6% |
HBP improved the
composite endpoint (total
mortality, heart failure
admission, and upgrade to
BiVP). The incidence of
heart failure admissions
was significantly reduced
as the sole endpoint |
| Zanon et al, 2018, Europace43 |
Meta-analysis
for HBP |
1,438 |
16.9 |
AVB 62.1%,
SSS 34.2% |
42.8% (31%
in a group
with a
previous
LVEF <50%) |
49.5% (42%
in a group
with previous
LVEF <50%) |
None |
84.8% |
HBP significantly improved
LVEF in patients with a
previous LVEF <50% |
AF, atrial fibrillation; AVB, atrioventricular block; AVJ, atrioventricular junction; BiVP, biventricular pacing; FU, follow-up period; HBP, His bundle pacing; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; QOL, quality of life; RVP, right ventricular pacing; SSS, sick sinus syndrome; UD, undetermined.
At present, we recommend that it is reasonable to perform His bundle pacing instead of RV pacing in patients with atrioventricular block who have an indication for permanent pacing with 36%>LVEF>50% and who are expected to require ventricular pacing over time. However, for patients with normal cardiac function, the adverse effects of RV pacing on LV performance can be expected to be insignificant; therefore, an indication for His bundle pacing should be carefully considered. Additionally, upgrading to His bundle pacing in patients with preexisting standard RV pacing and moderately reduced LVEF (36–50%) may impose an additional risk of lead extraction or lead implantation, and the clinical evidence of this challenge is not adequate. Implantation of His bundle pacing lead in patients with sick sinus syndrome but without a conduction system disturbance may cause iatrogenic injury to the conduction system; therefore, the indication for His bundle pacing should be very carefully considered.
1.2.2 Role of His Bundle Pacing in Patients With a Conduction Disturbance and Heart Failure (CRT-Compatible Patients)
a. Role of His Bundle Pacing in Patients Who Do Not Respond to CRT or in Whom LV Pacing via a Cardiac Vein Is Impossible (Table 8)45–48
Table 8.
Summary of Studies Comparing the Efficacy of His Bundle Pacing to That of CRT
| Study design |
No. of
patients |
FU
(months) |
QRS
morphology |
Pre-QRS
width
(ms) |
Post-HBP
QRS
width
(ms) |
Post-
Bi-V
QRS
width
(ms) |
Pre-
LVEF |
Post-
HBP
LVEF |
Post-
BiVP
LVEF |
HBP
success
rate |
Results |
| Barba-Pichardo et al, 2013, Europace45 |
Observational,
single arm
(unsuccessful
CRT patients
enrolled) |
16 |
31.3 |
CLBBB
(100%) |
166 |
97 |
None |
29% |
36% |
None |
75% |
Comparison between
pre- and post-HBP.
LVEF, NYHA class,
and reduction in LAD,
LVESD, and LVEDD
significantly improved |
| Lustgarten et al, 2015, Heart Rhythm46 |
Randomized,
crossover,
BiVP vs. HBP |
29 |
12 |
CLBBB
(97%) |
169 |
Non-
selective
160,
selective
131,
HBP+LVP
145 |
165 |
26% |
32% |
31% |
72% |
Comparison of the two
groups.
LVEF, NYHA class and
distance of 6-min walk
in both groups
significantly and
equally improved |
| Ajijola et al, 2017, Heart Rhythm47 |
Observational,
single arm |
21 |
12 |
CLBBB
(81%) |
180 |
129 |
None |
25% |
41% |
None |
76% |
Comparison between
pre- and post-HBP.
LVEF, NYHA class,
and reduction in
LVEDD significantly
improved |
| Sharma et al, 2018, Heart Rhythm48 |
Observational,
single arm |
Total
106 |
14 |
BBB (42%),
non-BBB
(16%),
RVP (32%) |
BBB 163,
non-BBB
103,
RVP 177 |
BBB 116,
non-BBB
108,
RVP 125 |
None |
30% |
44% |
None |
90% |
Comparison between
pre- and post-HBP.
LVEF and NYHA class
significantly improved,
but the reduction in
LVEDD did not.
As for cases with a
previous LVEF ≤35%
(n=72), LVEF improved
from 25% to 40% |
Group 1: LVP
failure (n=25),
CRT
non-responder
(n=8) |
33 |
|
BBB (64%),
non-BBB
(6%),
RVP (30%) |
BBB 161,
non-BBB
90,
RVP 175 |
BBB 115,
non-BBB
105,
RVP 115 |
None |
26% |
30% |
None |
91% |
7 of the 8 CRT non-
responder cases
responded to HBP, and
LVEF improved from
30% to 38% |
Group 2: AVB,
BBB, RVP |
73 |
|
BBB (37%),
non-BBB
(21%),
RVP (42%) |
BBB 164,
non-BBB
105,
RVP 179 |
BBB 116,
non-BBB
108,
RVP 125 |
None |
32% |
44% |
None |
89% |
|
AVB, atrio-ventricular block; BBB, bundle branch block; BiVP, bi-ventricular pacing; CLBBB, complete left bundle branch block; CRT; cardiac resynchronization therapy; FU, follow-up period; HBP, His bundle pacing; LAD, left atrial dimension; LVEDD, left ventricular end-diastolic dimension; LVESD, left ventricular end-systolic dimension; LVEF, left ventricular ejection fraction; LVP, left bundle pacing; NYHA, New York Heart Association; RVP, right ventricular pacing.
The efficacy of His bundle pacing in patients who had been proven to be non-responders to CRT was evaluated in 2 small-scale studies (Barba-Pichardo et al [n=16]45
and Sharma et al [n=8]48, which demonstrated a significant improvement in LVEF after His bundle pacing (LVEF increased from 29% to 36% and from 30% to 38%, respectively). Sharma et al evaluated His bundle pacing in 25 patients in whom CRT was impossible due to trans-venous LV pacing failure, and found an improvement in NHYA class and reduction in heart failure admissions.48
b. Role of Antecedent His Bundle Pacing in Candidates for CRT
Three different clinical studies attempted to perform His bundle pacing antecedently in candidates for CRT (Table 8).45–48
One randomized study compared the efficacy of HBP to that of CRT using a crossover study design in 29 patients, and showed a significant improvement in LVEF and NYHA functional class for whichever method that was used.46
Furthermore, 2 other single-arm observational studies demonstrated a significant improvement in LVEF and NYHA functional class.47,48
c. Indication for His Bundle Pacing in Candidates for CRT
As there are no data showing the long-term superior efficacy of His bundle pacing over CRT, and because a pacing impulse delivered at the His bundle may fail to capture the left bundle in patients with left bundle branch block, antecedent His bundle pacing without a CRT attempt is difficult to recommend. Therefore, His bundle pacing in candidates for CRT can be considered as an alternative option when (1) heart failure is deteriorating even after CRT, when there is (2) no optimal cardiac vein for LV pacing, (3) an unacceptable increase in the LV pacing threshold, (4) unavoidable phrenic nerve stimulation, (5) repeated dislodgement of the trans-venous LV lead, (6) technical limitations, or when (7) CRT is impossible due to complications or the clinical situation.
1.2.3 Role of Left Bundle Pacing
To resolve several problems of His bundle pacing, such as an increase in the pacing threshold and inability of left bundle capture in patients with complete left bundle branch block, a novel technique has emerged, aimed at pacing the left bundle directly by implanting a pacing lead at a distal site on the ventricular septum toward the LV.49–61
Although this method has been shown to be superior to His bundle pacing in reducing the pacing threshold,59
several concerns have been reported:49,61
perforation of the ventricular septum, injury to the conduction system, dislodgement of the lead, injury to the coronary artery (septal branch) and thromboembolism.51,59–61
Furthermore, because the follow-up intervals of the studies49,50,56
were ≤3 months, the long-term efficacy of left bundle pacing has not been clarified. Some investigators have assumed that left bundle branch pacing can be an alternative method to CRT; however, the significance of this method is still unknown owing to the lack of evidence from a randomized clinical trial including a large number of heart failure patients.58
2. Implantable Cardioverter-Defibrillators
2.1 Deactivating ICD in End-of-Life Care (Table 9)
Table 9.
Recommendations and Evidence Levels for Deactivation of ICDs
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
In ICD patients who are the end of life, the deactivation decision should be
made with support from a multidisciplinary team after obtaining sufficient
information about the ICD for the patient and family |
IIa |
C |
C1 |
VI |
COR, class of recommendation; GOR, Grade of Recommendation; ICD, implantable cardioverter-defibrillator; LOE, level of evidence.
Whether to deactivate an implantable cardioverter defibrillator (ICD) that was implanted and has been maintained before the patient had reached the end of life remains controversial. Shock therapy not only causes pain to the patient but also distresses the family members who provide care for the patient, thereby contradicting the purpose of palliative care in terminally ill patients. A retrospective observational study of 49 ICD patients who were considered to be near the end of life notwithstanding any cardiovascular disease reported that 42.9% experienced ICD activation within the year before death. Moreover, ICDs were not deactivated in 24.5% of patients, even after an end-of-life diagnosis was made, and only one-third of ICDs were deactivated.62
A subanalysis of data from the MADIT-II study, which reported the ICD therapy status in 98 terminally ill patients, revealed that of 47 patients without ICD deactivation who did not establish Do Not Attempt Resuscitation (DNAR) orders, 6 (13%) experienced ICD shock therapy within 1 week before death, and 9 (19%) experienced ICD shock therapy within 24 h of death.63
A Swedish observational study reported that 32 of 65 ICD patients with DNAR orders had deactivated ICDs, and 10 of 33 patients without deactivation experienced ICD shock therapy within 24 h of death.64
Among 51 ICD patients who were diagnosed with end stage heart failure at the Tokyo Women’s Medical University Hospital between 2010 and 2018, 12 of 39 patients with DNAR orders had deactivated ICDs. In addition, 21 patients (41%) experienced shock therapy within 3 months before death, 14 patients (27%) experienced shock therapy within 1 month before death, and 12 patients (24%) experienced electrical storms (including antitachycardia pacing) within 1 month before death.65
Thus, the worsening condition of terminally ill patients might contribute to the development of ventricular arrhythmias requiring ICD therapy. To avoid painful shock from the viewpoint of palliative care, cardiologists and healthcare professionals should discuss deactivation of ICDs in patients with end stage disease with the patients/family members.
Regarding the deactivation of ICDs, discussion is required before confirming the will of patients and families during the planning of end-of-life care based on advance care planning (ACP). This process needs to consider the ethics, cognitive ability of patient, etc., and judgment should be made not only by cardiologists and nurses but also based on consultation with palliative care and medical teams (including psychiatrists and psychologists); sufficient informed consent, depending on the individual situation, should be provided (Figure 1).4
The preparation of ACP directives at the end of life plays an important role with respect to the patient’s will. Of course, the content of such directives can be subsequently changed.
The HRS Expert Consensus Statement states that communication on ICD deactivation is essential; communication is an ongoing process that starts when informed consent is obtained prior to ICD implantation and continues as the patient’s condition and treatment goals change with disease progression.66
However, a previous report found that only 4% of patients discussed ICD deactivation with their physician in the clinical setting.67
According to the survey results of ICD patients with malignancies whose prognoses were relatively easy to predict, 35.3% of patients with stage IV cancer underwent ICD deactivation.68
Thus, there are several possible reasons for disconnecting between the recommendations and actual clinical situations. Medical staff do not actively discuss end-of-life topics and tend to postpone certain issues, such as the deactivation of ICDs, until just before death.69
A survey in the USA reported that some clinicians thought that a DNAR order did not mean that ICD should be deactivated, and one-quarter of clinicians thought that ICD deactivation was ethically considered suicide assistance.67
Besides deactivation, not replacing the device is another option for treatment withdrawal. In addition, ICD can be deactivated while maintaining bradycardia pacing or biventricular pacing to prevent the worsening of heart failure symptoms and impairment of quality of life. These decisions should be determined following the steps described above.
In ICD patients who are at the end of life notwithstanding any cardiovascular disease, the deactivation decision should be made closely with clinical psychologists, psychiatrists, and palliative care teams based on full evaluations of the psychological and physical stresses caused by ICD shock therapy while maintaining good communication with the patient’s family.
3. Transvenous Lead Extraction
3.1 Indication for Lead Extraction Due to Definite Device Infection
Early removal of the total device system (the device and all leads) is a Class I indication for a definite CIED infection, such as device pocket infection or lead infection with vegetation, regardless of whether there are any systemic infection symptoms or bacteremia.
3.2 Indication for Lead Extraction for Bacteremia Without a Definite Device Infection
Early removal of the total device system is not always a Class I indication for bacteremia without a definite device-related infection. First-line management should be investigation of the infection focus, removal of all easily extracted lines, such as a central venous catheter and temporary pacing wire, and administration of antibiotics based on susceptibility testing results for identified bacteria. If the infection focus is unclear and the clinical course is poor, an indication for lead extraction should be considered based on the pathogenic bacteria, the patient’s status, and the risk related to lead extraction.70
The following recommendations for lead extraction are based on specific pathogenic bacteria.
3.2.1 Staphylococcus Aureus, Coagulase-Negative Staphylococci, Propionibacterium spp., and Candida spp.
Early removal of the total device system is recommended for patients with an implanted CIED device and an occult blood stream infection caused by
Staphylococcus aureus, coagulase-negative staphylococci (CNS),
Propionibacterium
spp., or
Candida
spp.
Staphylococcus aureus
is an especially virulent bacterium that can cause tissue-destructive effects and severe clinical symptoms.71
CNS, such as
S. epidermidis, are weakly pathogenic bacteria; however, the incidence of device pocket infections due to CNS is high.72
Propionibacterium
spp. are Gram-positive anaerobic rod bacteria that produce a biofilm similar to
S. aureus
and CNS, resulting in antibiotic refractoriness in cases where the device remains in the patient.73
Candida
spp. may cause a refractory blood stream infection, especially in compromised hosts, due to biofilm production.74
3.2.2 Alpha-/Beta-Hemolytic Streptococcus spp. and Enterococcus spp.
Early lead extraction is recommended for bacteremia caused by alpha-/beta-hemolytic
Streptococcus
spp. and
Enterococcus
spp. Lead extraction should also be attempted if the bacteremia is recurrent or refractory to antibiotic therapy. The risk of a concomitant infection with the device is considered low.75
3.2.3 Gram-Negative Bacteria and Streptococcus Pneumoniae
Lead extraction is recommended if the bacteremia from Gram-negative bacteria or
S. pneumoniae
is recurrent or refractory to antibiotic therapy. The risk of a concomitant infection with the device is considered low, and antibiotic therapy only may be curative.75
3.3 Strategy for Superficial Device Pocket Infections
A superficial pocket infection does not reach the device, is caused by surface skin closure sutures, and occurs within 30 days of device implantation. Oral antibiotics effective against
Staphylococcus aureus
should be administered and local lesion care to the superficial infection site is mandatory for at least 7–10 days until pocket redness and systemic symptoms, such as fever and inflammatory reactions, disappear. Superficial lesions should be treated in the same way as device infections if the lesions are refractory or associated with apparent infectious symptoms.
3.4 Management After Lead Extraction for Device Infection and Implantation of a New Device
Antibiotic therapy based on the results of sensitivity testing is mandatory after the removal of an entire CIED system (including the leads). A bacterial examination of a swab or tissue culture from the surgical pocket,76
the tip or entirety of the extracted lead,70
and the removed device77
should be performed on more than 2 sets of blood cultures before administering antibiotics.
The antibiotic administration period is at least 2 weeks without a blood stream infection for pocket infections, at least 2 weeks without lead or valvular vegetations for blood stream infections (4 weeks is recommended for
Staphylococcus aureus
infections), at least 2–4 weeks for lead vegetations, and at least 4–6 weeks for valvular vegetations.70
Implantation of a new device after CIED system removal should be carefully evaluated, and discontinuation of the device may be considered if there is no (or a lesser) indication for implantation. When implantation of a new device is indicated, the side opposite the original site is recommended. A device without the need for transvenous leads, such as a subcutaneous ICD or leadless pacemaker, may be useful. A new device should be implanted after confirmation of negative blood cultures for 72 h after the lead extraction; however, the timing can be delayed based on the patient’s systemic symptoms and the bacterial source, such as an infective abscess. The use of a wearable cardioverter defibrillator can be considered until implantation for high-risk patients with life-threatening arrhythmias.
4. Role of Cardiac Implantable Electrical Devices in the Termination and Prevention of Atrial Arrhythmias
Among patients with a CIED, it is important to prevent the occurrence and persistence of atrial arrhythmias (ATAs) including atrial fibrillation (AF). Previous studies have shown that patients with paroxysmal AF have significantly lower incidences of stroke and all-cause mortality compared with patients with persistent AF.78–80
Botto et al reported that combining data on AF duration with that on scores of thromboembolic risk (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and prior stroke, or transient ischemic attack [CHADS2]) could clearly discriminate the risk of thromboembolic events in patients with a dual-chamber rate-responsive pacemaker.81
In a large series of patients with an ICD or a cardiac resynchronization therapy device with defibrillator (CRT-D), Vergara et al indicated that atrial high-rate episodes occurred during the 48 h preceding lethal ventricular tachyarrhythmias (VAs), which were detected by the shock device system in ≈20% of the patients with VAs.82
Moreover, Nakajima et al reported that some patients developed heart failure (HF) after transient AF attacks, even in CRT responders, mainly due to the loss of percent of biventricular pacing.83
Because the appearance and persistence of ATAs are associated with several risks of stroke, VAs, and HF, therapeutic intervention should be executed in such patients with a CIED and ATAs.
Research into the utility of CIED functions for rhythm control to prevent and terminate ATAs has been conducted for some time.84
There are several device-based approaches for the prevention of ATAs, including multisite atrial pacing and pacing algorithms to increase the frequency of atrial pacing. However, their effectiveness in preventing and terminating ATAs are not sufficient for clinical use.85,86
The SAVE PACe trial showed that pacemaker implantation and the use of new device-based algorithms of dual-chamber minimal ventricular pacing (MVP) could significantly suppress progression to persistent AF in patients with sick sinus syndrome, intact atrioventricular conduction, and a normal QRS interval (7.9% [MVP group] vs. 12.7% [control group], hazard ratio [HR]: 0.60, 95% confidence interval [CI]: 0.41–0.88; P=0.009).87
However, a recent meta-analysis on the efficacy of MVP revealed that a low burden of percent of ventricular pacing did not significantly suppress AF progression.88
Atrial antitachycardia pacing (A-ATP) has been investigated for its efficacy in terminating ATAs since the early first decade of the 2000s. A-ATP is likely to be effective for atrial flutter and atrial tachycardias, however, it may sometimes progress to AF after using A-ATP. The ATTEST trial was a randomized clinical study that evaluated the efficacy of preventive pacing and the first generation of A-ATP in patients with a pacemaker.89
The 324 patients who had ACC/AHA/NASPE Class I or II indication for a standard dual-chamber pacemaker90
and a history of ATAs were randomized into first generation of A-ATP therapies ON or OFF using the ATA prevention algorithm. In each group, the burden and frequency of ATAs were accessed during the 3-month study period and the median ATA burden was 4.2 h/month in the ON group vs. 1.1 h/month in the OFF group (P=0.20). In addition, the median ATA frequency was 1.3 episodes/month (ON) vs. 1.2 episodes/month (OFF) (P=0.65). The preventive pacing and first generation of A-ATP did not significantly reduce either the ATA burden or frequency.89
Subsequently, the second generation of A-ATP (ReactiveATPTM), which can be delivered at the onset of arrhythmia as well as during its dynamic changes towards slower or more organized rhythms, was developed. The MINERVA trial was a randomized clinical study to evaluate the efficacy of combination therapy of atrial preventive pacing, second-generation A-ATP, and MVP in patients with a pacemaker.91
A total of 1,166 patients who had ACC/AHA/NASPE Class I or II indication for a standard dual-chamber pacemaker90
and a history of ATAs were randomly assigned to (1) not using any algorithm (control DDDR group), (2) using atrial preventive pacing, second-generation A-ATP, and MVP (DDDRP + MVP group), and (3) using only MVP (MVP group). During a median follow-up of 34 months, the 2-year incidence of a composite clinical endpoint including all-cause death, cardiovascular hospitalization, or permanent AF was significantly lower in the DDDRP + MVP group as compared with the control DDDR group (HR: 0.74, 95% CI: 0.55–0.99, P=0.04), while no significant difference was observed between the DDDRP + MVP and MVP groups (HR: 0.93, 95% CI: 0.68–1.26, P=0.63 vs. the MVP group). As for secondary endpoint of progression to permanent AF, the DDDRP + MVP group showed a significantly lower risk than the control DDDR group (HR: 0.39, 95% CI: 0.21–0.75, P=0.004) and the MVP group (HR: 0.49, 95% CI: 0.25–0.95, P=0.034). Padeletti et al analyzed the MINERVA trial data and highlighted that the generalized estimation of equation-adjusted efficacy of the second generation of A-ATP was 44.4% (95% CI: 41.3–47.6%) in the DDDRP + MVP group.92
In addition, the risk of progression to permanent AF was significantly reduced in patients with high efficacy (>44.4%) of the second generation of A-ATP (HR: 0.32, 95% CI: 0.13–0.78, P=0.012).
High efficacy of A-ATP in patients with congenital heart disease accompanying ATAs has been reported.93,94
Kramer et al revealed that implantation of an atrial antitachycardia pacing device was associated with reduced direct current cardioversion burden, and the overall success rate of A-ATP was 69% in a cohort of 91 patients with congenital heart disease and ATAs.95
Although there are few studies assessing the efficacy of A-ATP in patients with an ICD or a CRT device, Crossley et al analyzed the data obtained from a large device database of a remote monitoring system (Medtronic CareLinkTM) and reported the effectiveness of the second generation of A-ATP.96
Among 1,062 patients with an ICD, the use of the second generation of A-ATP was associated with reduced risk of ≥7 days of ATAs as compared with no use, even after controlling for other covariates (HR: 0.58, 95% CI: 0.43–0.77). As for the patients with CRT-D, the use of the second generation of A-ATP significantly reduced the risk of risk of ≥7 days of ATAs compared with no use (HR: 0.73; 95% CI: 0.57–0.93). In addition, Ueda et al showed the efficacy of the second generation of A-ATP in CRT patients: 23 of 44 patients (52%) received a total of 2,862 ATP deliveries during a mean follow-up of 832 days, and the median success rate of A-ATP was 23.6% (interquartile range: 12.5–50.0%) in the A-ATP group.5
Along the way, there have been several studies reporting the efficacy of the second generation of A-ATP; however, the effectiveness differs according to each individual and the results may vary in each study. Subanalysis of the MINERVA trial indicated that a cycle length of ATAs ≤210 ms or relatively irregular AA intervals of the ATAs was associated with low efficacy of the second generation of A-ATP.92
Moreover, the efficacy of the second generation of A-ATP is much lower in patients with a CRT device than in those with a pacemaker.5
Further prospective studies including a larger number of patients are needed to evaluate how to use the second generation of A-ATP and when to terminate its function.
5. Device-Detected Atrial Fibrillation (AF) (Table 10)
Table 10.
Recommendations and Evidence Levels for the Management of Device-Detected AHREs
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
In patients with device-detected AHREs, further evaluation to document
clinically relevant AF is recommended |
I |
B |
A |
IVa |
In patients with device-detected AHREs, anticoagulation therapy is reasonable
in patients with CHADS2 score ≥1 by taking the efficacy and safety into
consideration on an individual basis |
IIa |
B |
C1 |
IVa |
AF, atrial fibrillation; AHRE, atrial high-rate episode; COR, class of recommendation; GOR, Grade of Recommendation; LOE, level of evidence.
CIEDs can record intracardiac electrograms, thereby allowing continuous monitoring of cardiac atrial/ventricular arrhythmias. As remote monitoring has become a popular and guideline-recommended standard ambulatory medical care tool for most CIED patients,1
the opportunity to detect subclinical cardiac arrhythmias (especially subclinical atrial fibrillation [AF]) has been increasing.
5.1 Prevalence of CIED-Detected AF
A number of studies have recorded the incidence of CIED-detected AF. In the ASSERT trial of 2,580 patients receiving a pacemaker or ICD, subclinical AF was documented in 261 patients (10.1%) within 3 months after implantation. In another pooled analysis of 3 prospective studies (TRENDS, PANORAM, and SOS AF), which included 6,580 patients with implanted CIEDs without a history of AF, around one-third had a daily AF burden of ≥5/min and >20% of them had a daily AF burden ≥23/h during a mean follow-up of 2.4 years.97
In another study using a home monitoring database, a total of 3,004,763 transmissions from 11,624 CIED patients were analyzed; subclinical AF was detected in ≈60% of the patients with a pacemaker and in ≈10% of those with an ICD,98
thereby indicating that subclinical AF is commonly detected regardless of CIED type. Furthermore, other studies report that the remote monitoring system can detect 95% of asymptomatic AF,99
and that the time to the first diagnosis of AF is earlier than with the traditional monitoring system.100
These data demonstrate that subclinical AF is detected earlier and more frequently in CIED patients, especially with combined use of a remote monitoring system.
5.2 Clinical Significance of CIED-Detected AF
It has been reported in previous studies that newly detected AF in CIED patients is associated with heart failure admission, appropriate ICD discharge, and death,101
and also closely related to the development of stroke/systemic embolic events (Table 11).102–106
Accordingly, early detection and management of CIED-detected AF, especially the indication of anticoagulation therapy to prevent stroke/systemic embolic events, are the most important for physicians.
Table 11.
Studies of Device-Detected Atrial Fibrillation
| Author, Year |
Study |
Contents |
Results |
Hearley et al,
2012104 |
ASSERT |
n=2,580 (pacemaker 2,451/ICD 129)
Follow-up the patients for 3 months to detect AHREs
(mean 2.5 years)
AHREs definition: Atrial rate >190 bpm for >6 min |
AHREs associated with ischemic stroke or
systemic embolism |
Shanmugan et al,
2012105 |
Home monitor
CRT |
n=560 heart failure patients with CRT
Follow-up patients after introducing home monitoring
AHREs definition: Atrial rate >180 bpm for at least 1%
or total of 14 min/day |
AHREs of 3.8 h over 24 h associated with
thromboembolic events |
Boriani et al,
2014102 |
SOS AF
project |
n=10,016 (pacemaker/ICD/CRT)
Pooled analysis from 5 prospective studies
AHREs definition: Atrial rate >175 bpm, lasting ≥20 s |
Among the thresholds of AF burden, 1 h
was associated with the highest hazard
ratio for ischemic stroke |
Glotzer et al,
2009103 |
TRENDS trial |
n=3,045 (pacemaker/ICD/CRT)
Annualized thromboembolic event rates according to
AHREs burden subsets: zero, low (<5.5 h), and high
(≥5.5 h)
AHREs definition: Atrial rate >175 bpm, lasting ≥20 s |
AHREs burden ≥5.5 h associated with
thromboembolic events |
Van Gelder et al,
2017106 |
ASSERT
subanalysis |
n=2,455 (patients in whom the longest episode was
≤6 min were excluded from the analysis, n=125)
AHREs definition: Atrial rate >190 bpm for >6 min |
AHREs >24 h associated with an increased
risk of ischemic stroke or systemic
embolism |
AF, atrial fibrillation; AHRE, atrial high-rate episode; CRT, cardiac resynchronization therapy; ICD, implantable cardioverter-defibrillator.
5.3 Diagnostic Accuracy of CIED-Detected AF
Although CIED-detected AFs are recorded as atrial high-rate episodes (AHREs), AHREs and AF are not strictly the same. Thus, careful interpretation is needed if the AHREs are silent. As the cause of false positives, the oversensing of far-field R-wave, lead noise, and other atrial arrhythmias can be considered. As the cause of false negatives, undersensing during the blanking period can be considered; however, diagnostic accuracy has been improved in the recent CIEDs, because the sensing of the atrial electrogram during the blanking period is currently enabled.107,108
In an analysis of a duration-based study of CIED-detected AHREs, true atrial tachycardia/AF was 82.7% and false positives were 17.3% for AHREs lasting <6 min; however, false positives dropped to 6.8%, 3.3%, and 1.8% when the threshold duration was increased to 30 min, 6 h, and 24 h, respectively.109
5.4 Anticoagulation Therapy for CIED-Detected AF
A number of studies report that CIED-detected AF is closely associated with an increased risk of stroke, but the cutoff value of AF duration is still unclear (Table 11).102–106
Based on the initial ASSERT study,104
anticoagulation was recommended for patients with >5–6 min of device-detected AF in the ESC guideline 2016;110
this recommendation was deleted after a subanalysis that showed that stroke/systemic emboli occurred only in patients with >24 h of device-detected AF (Figure 2).106
Presently, there is no recommendation regarding the duration of CIED-detected AF in the AHA/ACC/HRS guideline focus update 2019111
or the ESC guideline 2020.112
However, the positive result from a recent prospective study (TACTIC-AF),113
which allowed intermittent anticoagulation guided by CIED-detected AF duration (restart anticoagulation in patients with AF lasting ≥6 min or with a total AF burden <6 h/day), and the summarized data from previous studies,111,114
clearly show there is definitely a positive relationship between the duration of CIED-detected AF and embolic events. As the majority of evidence for the indication of anticoagulation for stroke/systemic emboli is surface ECG-detected AF, further evaluation is needed for CIED-detected AF. Prospective studies on the feasibility of anticoagulation in patients with CIED-detected AHREs are currently underway (NOAH-AFNET 6115
and ARISTESiA116
trials); the findings from these trials will be helpful for us. At present, the indication of anticoagulation for device-detected AF should be considered individually based on the following: duration and burden of CIED-detected AF, patient’s risk factors for stroke/systemic emboli, etc. (Table 10).
III. Catheter Ablation
1. Reducing Radiation Exposure
The 3-dimensional electroanatomical mapping systems and force-sensing catheters have contributed to reducing radiation exposure during catheter ablation. Catheter positioning, anatomic information, and contact with the myocardium can be monitored without fluoroscopy. In a prospective, multicenter, randomized controlled trial that included 262 patients who underwent electrophysiologic studies for SVT, patients were randomized to receive a minimally fluoroscopic approach (MFA) using the EnSiteTM
NavXTM
navigation system or the conventional approach. Significant reductions in patients’ radiation dose, total fluoroscopy time, and operator radiation dose were documented in the MFA group, without compromising efficacy or safety. Additionally, a 96% decline in the risks of cancer incidence and mortality was estimated.117
The use of fluoroscopic imaging integrated with a 3D electroanatomical mapping system, instead of the conventional electroanatomical mapping system, during AF ablation was shown to further reduce radiation exposure, without complicating the workflow or compromising acute/mid-term efficacy and safety.118
Fluoroscopy time was reduced by 84%, and radiation dose was reduced by 73%. Most of the fluoroscopic guidance was used for femoral vein/artery puncture, placement of catheters and transseptal puncture, reconstruction of the left atrium and pulmonary vein angiography, and mapping; basically, near-zero fluoroscopy was used for the ablation process itself. The radiation exposure trend in a single center over the 7 years between 2010 and 2016 has been reported.119
Procedures included catheter ablation and device implantation. Fluoroscopic time, dose–area product, and effective dose decreased for ablation, but not for CRT device implantation. Recognizing the importance of radiation reduction in addition to technological advancements is crucial. Several studies have reported on zero or near-zero fluoroscopy ablation.120–122
A meta-analysis of 10 studies involving 2,261 patients who underwent zero or near-zero fluoroscopy ablation showed significant reductions in fluoroscopy time, ablation time, and radiation dose in comparison with the conventional approach, and showed similar procedure duration, success rate, complication rate, and recurrence rate.123
Pulmonary vein isolation by cryoballoon ablation (CBA) has been as effective and safe as conventional radiofrequency catheter ablation (RFCA) since the launch of the 2nd-generation cryoballoon catheter. In addition, the procedure time of CBA is significantly shorter than that of RFCA.124,125
Recently, the safety and efficacy of CBA have been shown for both paroxysmal and persistent atrial fibrillation (AF),126,127
and its indication has been expanded (insurance reimbursement from November 2020 in Japan). However, the fluoroscopy time in CBA is longer than that in RFCA,124,125
and extended radiation exposure is a concern. The mean fluoroscopy time in the FIRE and ICE study124
was 16.6±17.8 min in the RFCA group and 21.7±13.9 min in the CBA group (P<0.001). In the CIRCA-DOSE study,125
the median fluoroscopy time in the RFCA group was 5.2 min, whereas in the CBA group (CRYO-4 and CRYO-2) it was 17.2 min and 19.0 min, respectively (both P<0.001). It is possible to shorten the fluoroscopy time in RFCA by improving the accuracy of the 3D navigation system. However, fluoroscopy is necessary for manipulation of the catheter and sheath to isolate the pulmonary vein and for the confirmation of pulmonary vein occlusion in CBA. Therefore, the exposure dose in CBA tends to be higher than in RFCA. To evaluate the degree of pulmonary vein occlusion with a cryoballoon, 2 methods that significantly reduce radiation exposure without compromising efficacy and safety have been proposed. One method confirms leakage using an intracardiac echo Doppler,128,129
and the other predicts the degree of occlusion from pressure changes and pressure waveforms by measuring the balloon tip pressure.130–132
Although the benefits of bonus-freeze have not been evident since the launch of the 2nd-generation cryoballoon, it has been reported that bonus freezing prolongs both procedure time and fluoroscopy time, and increases complications.133
The exposure dose is significantly reduced by changing from the conventional 2-time bonus-freeze method to a 1-time method (single freeze). In addition, there is a study that found the 3D mapping system effectively reduces radiation exposure;134
however, there are also drawbacks such as increased costs.
A significant proportion of the overall exposure to ionizing radiation in patients undergoing AF ablation is contributed by preprocedural CT. It has been reported that the radiation dose of preprocedural CT (9.4±5.8 mSv) is drastically greater than that of the ablation procedure (2±2 mSv), amounting to 82% of the total radiation dose.135
In addition to fluoroscopic time, the projection angle, position, collimation, frame rate, pulse rate, magnification, use of cine, and body status also affect the radiation dose. Knowledge of the practical use of fluoroscopy is crucial. It has been shown that fluoroscopic time and radiation dose during AF ablation significantly reduced following a lecture on radiation reduction.136
Additionally, using a radiation safety time-out, a novel checklist using the concept of “as low as reasonably achievable (ALARA)”, setting a low frame rate, adjusting the patient table height as close as safely possible to the intensifier, proper collimation, use of shields, dosimeter, and protectors have been shown to reduce radiation exposure.137
2. Atrial Fibrillation (AF)
2.1 Indications (Table 12)
Table 12.
Recommendations and Evidence Levels for Catheter Ablation for Atrial Fibrillation in Heart Failure
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
AF catheter ablation may be reasonable in selected patients with symptomatic
AF and HFrEF to potentially lower mortality rate and reduce hospitalization
for HF |
IIa |
B |
B |
II |
AF, atrial fibrillation; COR, class of recommendation; GOR, grade of recommendation; HF, heart failure; HFrEF, heart failure with reduced ejection fraction; LOE, level of evidence.
2.1.1 Prognosis After AF Ablation
a. CABANA Trial
The recent catheter ablation vs. anti-arrhythmic drug treatment for atrial fibrillation (CABANA) trial7
demonstrated that AF ablation was not superior to drug therapy for the primary composite endpoint (death, disabling stroke, serious bleeding, or cardiac arrest) in the intention-to-treat analysis; however, suppression of the secondary composite endpoint (death from any cause or cardiovascular hospitalization) was observed in the AF ablation group. Furthermore, AF ablation was superior to drug therapy in the primary composite endpoint (P=0.046) in the per-protocol analysis. The quality of life (QOL) analysis also reported that AF ablation was significantly superior to drug therapy in improving QOL score.138
Although CABANA was the first, large-scale randomized controlled trial (RCT) to reveal the influence of AF ablation on the prognosis of AF patients, it failed to yield new evidence of an improvement in the long-term prognosis.
b. EAST-AFNET 49
This international RCT trial was the first to report that early rhythm-control therapy was associated with a lower risk of adverse cardiovascular outcomes than the usual care in patients with early AF. Patients with early AF (diagnosed ≤1 year before enrollment) and cardiovascular conditions were randomly assigned to receive either early rhythm-control (antiarrhythmic drug treatment of atrial AF ablation) or usual care (mainly rate-control treatment). The trial was stopped for efficacy at the 3rd interim analysis after a median follow-up of 5.1 years per patient, because of results that a first-primary-outcome event occurred less often in patients who were assigned to receive early rhythm-control (3.9 per 100 person-years) than usual care (5.0 per 100 person-years) (P=0.005). Although catheter ablation was performed in a relatively small number of patients (19.4% in the early rhythm-control group, 7.0% in usual care group), this study showed for the first time the effect of early rhythm control treatment in improving the outcome of AF patients.
2.1.2 AF Catheter Ablation in Patients With HF
A meta-analysis of six RCTs comparing the efficacy rates between pharmacotherapy (rate or rhythm control therapy) and catheter ablation in a total of 775 HF patients with reduced cardiac function has been published.139–144
AF ablation significantly reduced all-cause mortality (9.0% vs. 17.6%) and HF hospitalizations (16.4% vs. 27.6%) when compared with drug therapy.8
Furthermore, ablation improved left ventricular ejection fraction (LVEF: mean difference, 6.95%) and QOL.
This focused update determined that the role of catheter ablation therapy in AF patients with HF is an established option to improve prognosis, and recommend that “AF catheter ablation may be reasonable in selected patients with symptomatic AF and HF with reduced left ventricular ejection fraction (HFrEF) to potentially lower mortality rate and reduce hospitalization for HF” as a Class IIa recommendation.
In the latest ESC guideline 2020,112
AF catheter ablation is included as a Class I recommendation to restore LV function when AF-mediated tachycardia-induced cardiomyopathy (ventricular dysfunction secondary to rapid and/or asynchronous/irregular myocardial contraction, partially or completely reversed after treatment of the causative arrhythmia) is highly suspected. This set of guidelines also determined that catheter ablation therapy in AF patients with tachycardia-induced cardiomyopathy is a more positive option that can be expected to improve prognosis.
2.2 Methods
2.2.1 High-Power Short-Duration (HPSD) Method
HPSD ablation (45–70 W for 5–10 s) for the treatment of AF is emerging as an alternative to conventional ablation (20–40 W for 20–40 s).145,146
During radiofrequency (RF) ablation, tissue heating occurs via 2 mechanisms of resistive (direct) and conductive (indirect) heating. Resistive heating is an active process and rapidly begins and ends with the initiation and cessation of RF energy application. Conductive heating is a passive and time-dependent process as heat is transferred away from the ablation lesion core. Compared with conventional ablation, HPSD has been shown to create a wider lesion without obvious increase of lesion depth,147,148
which may be a preferred RF application method for atrial thin muscles.149
Although the efficiency and safety of HPSD method have been shown in a nonrandomized study,150,151
there has been no RCT comparing HPSD and conventional methods.
Recently, the efficiency and safety of the very HPSD (vHPSD) method using a novel contact force-sensing catheter has been reported from Western countries. The RF energy is applied with 90 W for 4 s, using a temperature-control mode (surface temperature: 65–70℃), which results in substantially shorter procedural and fluoroscopic times as compared with the standard method, with similar efficiency and safety.152
2.2.2 Balloon Ablation as a First-Choice Therapy for AF
Recently, two RCTs evaluated the availability of cryoballoon ablation as a first-choice treatment for paroxysmal AF. In the STOP AF first trial,10
203 patients were randomly assigned to receive treatment with antiarrhythmic drugs (class I or III agents) or pulmonary vein isolation (PVI) with a cryoballoon. After a follow-up period of 12 months, treatment success (freedom from initial failure of the procedure or atrial arrhythmia recurrence) was more evident in the ablation group than in the drug group (74.6% vs. 45.0%, P<0.001). Although 2 primary safety endpoint events (pericardial effusion and myocardial infarction) occurred, serious procedure-related adverse events were rare.
In the EARLY-AF trial,11
303 patients with symptomatic, paroxysmal, untreated AF were randomly assigned to undergo cryoballoon ablation or to receive antiarrhythmic drug therapy for initial rhythm control. All patients received an implantable cardiac monitoring device to detect atrial tachyarrhythmia. At the 1-year follow-up, recurrence of atrial tachyarrhythmia was less frequent in the ablation group than in the drug-treatment group (42.9% vs. 67.8%, P<0.001). The occurrence of serious adverse events was similar in both groups (3.2% in ablation group vs. 4.0% in drug-treatment group).
With documentation of the efficacy and safety of cryoballoon ablation as the initial treatment for paroxysmal AF in these 2 RCT trials, the popularization of first-choice PVI treatment for paroxysmal AF is anticipated to accelerate.
2.2.3 Balloon Ablation for Persistent AF
The indication of catheter ablation using balloon devices has been limited to PVI for paroxysmal AF in Japan, because PVI was thought to be efficient in treating only early-stage (paroxysmal) AF but not efficient enough for treating more advanced (persistent) AF. No ablation lesion set has been demonstrated to improve outcomes of patients with persistent AF beyond that of PVI alone in a recent RCT.153
The STOP persistent AF trial126
is a prospective, multicenter, single-arm, FDA-regulated trial designed to evaluate the safety and efficacy of PVI-only cryoballoon ablation for drug-refractory persistent AF (continuous episode <6 months) in 165 subjects (including 15 Japanese patients). Among them, 54.8% achieved the primary efficacy endpoint (12 months’ freedom from ≥30 s of atrial tachyarrhythmias), and 86.8% were free from repeat ablation. The primary safety event rate was 0.6%. Significant improvements in QOL and AF-related symptoms were also observed. Based on these results, the indication of cryoballoon ablation was expanded to persistent AF in both USA and Japan in 2020. Nevertheless, it is important to note that the approved procedure is a “PVI-only” approach for drug-refractory persistent AF, and the efficacy and safety of cryoballoon application other than PVI have not been confirmed and is off-label for the time being.
2.3 Perioperative Anticoagulant Therapy
Thromboembolism is one of the most serious complications of AF ablation, and appropriate pre-, intra-, and post-operative anticoagulant therapy are extremely crucial in minimizing its risk. Based on new knowledge of the anticoagulants used in preoperative management, the description and part of the recommendation table have been revised (Table 13).
Table 13.
Recommendations and Evidence Levels for Anticoagulation Strategies Pre-, Intra-, and Post-Ablation of Atrial Fibrillation
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
For patients with persistent AF or those with high risk of embolism (CHADS2 score ≥2), systemic
anticoagulation with warfarin or a DOAC is reasonable for ≥3 weeks prior to AF ablation |
IIa |
C |
C1 |
VI |
For patients who have been therapeutically anticoagulated with warfarin or dabigatran, performance of
the ablation procedure without interruption of warfarin or dabigatran is recommended |
I |
A |
A |
I |
For patients who have been therapeutically anticoagulated with rivaroxaban, apixaban, or edoxaban,
performance of the ablation procedure without interruption of rivaroxaban, apixaban, or edoxaban is
reasonable |
IIa |
B |
B |
II |
For patients who have been therapeutically anticoagulated with a DOAC prior to AF ablation, it is
reasonable to interrupt 1 or 2 dose(s) of DOAC prior to AF ablation with its re-initiation post-ablation |
IIa |
B |
B |
II |
Heparin should be administered immediately following femoral venous puncture or transseptal
puncture during AF ablation procedures, and adjusted to achieve and maintain an ACT ≥300 s |
I |
B |
B |
III |
Systemic anticoagulation with warfarin or a DOAC is recommended at least 3 months post AF
ablation, regardless of the apparent success or failure of the AF ablation procedure |
IIa |
C |
C1 |
VI |
For patients with a high risk for embolism (CHADS2 score ≥2), continuation of systemic anticoagulation
with warfarin or a DOAC should be considered even after 3 months of AF ablation, considering AF
recurrence during the follow-up period |
IIa |
C |
C1 |
VI |
ACT, activated clotting time; AF, atrial fibrillation; COR, class of recommendation; DOAC, direct oral anticoagulant; GOR, Grade of Recommendation; LOE, level of evidence.
2.3.1 Interruption and Continuation of Preoperative Anticoagulants
The preoperative anticoagulant, warfarin, used to be discontinued, and ablation was performed after heparinization; however, cerebral embolisms often occurred after restarting warfarin postoperatively. Therefore, ablation with continuous administration of warfarin was attempted. As compared with the group in which warfarin was discontinued preoperatively and bridged to high-dose or low-dose heparin, Wazni et al reported significantly less hemorrhagic and blood deficiency complications in the continuous warfarin group.154
Direct oral anticoagulants (DOACs) have been available in Japan since 2011, and 4 types, namely, dabigatran, rivaroxaban, apixaban, and edoxaban, are currently being used. Because the half-life of DOACs is short (≈5–17 h) and the time to reach the peak level is short, they are frequently used as an alternative anticoagulant to warfarin because of their efficacy and safety.
Therefore, the possibility of continuous administration of DOACs as with warfarin during the perioperative period of ablation has been discussed. With a short half-life and a fast peak level after oral administration, it is thought that a temporary suspension may be advantageous to avoid hemorrhagic complications such as cardiac tamponade, puncture site bleeding, and the onset of thromboembolism in the high-risk group. However, there has been concern about the onset of ischemic complications due to emboli during a temporary drug suspension. The aforementioned speculation and concern were issues that needed to be verified.
2.3.2 AF Ablation With Continuous Preoperative DOAC Administration
The comparison of continuous administration of warfarin and continuous administration of a DOAC during the perioperative period of ablation has been reported in 4 DOAC RCTs. The results for the 4 types of DOACs are compared and shown in
Figure 3.155–158
In the RE-CIRCUIT study155
using dabigatran, there was no significant difference in ischemic complications between the dabigatran continuation and warfarin continuation groups. In terms of safety, however, the warfarin continuation group experienced cardiac tamponade, inguinal hematomas, etc., and major bleeding events were significantly higher (1.6% vs. 6.9%). Although there was a large racial difference in bleeding complications, no significant difference was observed between the 2 groups among Japanese patients who participated in the study (1.6% in the dabigatran group vs. 2.2% in the warfarin group).
In the VENTURE-AF study156
of rivaroxaban, a comparison between the rivaroxaban continuation and warfarin continuation groups showed no significant differences in efficacy or safety, and ablation while continuing rivaroxaban was also allowed. This study was characterized by a small number of patients (n=124) and an extremely small number of events (0.0% vs. 2.4%) in both groups as compared with other studies.
In the AXAFA AFNET 5 study157
using apixaban, a comparison between the apixaban continuation and warfarin continuation groups exhibited no significant differences in efficacy or safety. Furthermore, the detection rate of asymptomatic microinfarcts using brain MRI was similar. The characteristics of this study were that the event occurrence rate was slightly high in both groups (4.0% vs. 4.7%), but the average CHA2DS2-VASc score of the target patient group was also considered to be a factor.
In the ELIMINATE-AF study158
using edoxaban, as with rivaroxaban and apixaban there were no significant differences regarding the efficacy or safety between the edoxaban continuation and warfarin continuation groups. Hemorrhagic complications were slightly higher in the edoxaban continuation group but there was no significant difference (2.4% vs. 1.7%).
2.3.3 AF Ablation With Interruption of Preoperative DOACs
On the other hand, there have been several studies, including RCTs, reporting on AF ablation with perioperative interruption of DOACs.12–15
In the ABRIDGE-J study12
conducted in Japan using dabigatran, the short-term interrupted dabigatran group (n=220) and continuous warfarin group (n=222) showed no significant difference in terms of ischemic complications. In terms of safety, however, as was observed in the RE-CIRCUIT study, the ocurrence of cardiac tamponade and femoral hematoma complications was significantly higher in the warfarin continuation group than in the minimally-interrupted dabigatran group (1.4% vs. 5.0%).
Furthermore, a study by Nakamura et al,13
as an RCT conducted in Japan, compared the occurrence of embolisms and hemorrhagic complications in the continuous and minimally-interrupted groups for the 4 DOACs used in Japan; for all DOACs, there was no significant difference between the 2 groups. Furthermore, MRI revealed no significant difference in the occurrence of asymptomatic microcerebral embolisms between the 2 groups.
The decision to continue or suspend the drug is left to the standards of the facility, discretion of the operator, and situation surrounding the case (risk of thromboembolisms and bleeding); however, fine adjustments such as the timing of the final preoperative administration and the time of ablation (morning or afternoon) may be considered.
In an integrated analysis of the RE-CIRCUIT study155
with the continuous administration of dabigatran and the ABRIDGE-J study12
with minimally interruption of dabigatran,159
1.9% of the major bleeding events occurred within 8 h of the final oral administration of dabigatran, 0% between 8 and 24 h, and 3.5% beyond 24 h from the start of the ablation procedure. Major bleeding events were significantly lower in the group between 8 and <24 h.
2.3.4 Reversal Agents for Various Anticoagulants
The continuation of anticoagulants is desirable, considering the suppression of thrombus formation during the perioperative period of ablation, but the presence of an anticoagulant reversal agent is important when bleeding complications such as cardiac tamponade occur during the course of procedure. Reversal agents, such as protamine for heparin and vitamin K for warfarin, are also being developed for DOACs.
For dabigatran, the specific reversal agent, idarucizumab, is currently used in clinical practice during bleeding complications. In a multicenter study in Japan, Okishige et al16
reported 21 cases of perioperative cardiac tamponade due to the continuous administration of dabigatran on the morning of the ablation procedure. Hemostasis was confirmed an average of 2.1 h after the administration of idarucizumab in 16 of the 21 patients, and only 1 patient required a surgical repair.
Furthermore, based on the RE-CIRCUIT study155
and the availability of a reversal agent, studies on switching from other anticoagulants to dabigatran during the perioperative period of ablation have been conducted,17,18
and switching has been shown to have an inhibitory effect on cerebral embolisms, including those that are asymptomatic.
Reversal agents (andexanet alfa) for factor Xa inhibitors (rivaroxaban, apixaban, edoxaban) have also been developed and their effectiveness has been reported160
(not approved in Japan yet). Currently, multicenter, prospective, open-label studies are being conducted in Japan, North America, and Europe.
IV. Left Atrial Appendage Closure Device
1. Randomized Trial of the WATCHMANTM Device
In Western countries, several left atrial appendage closure (LAAC) devices have been launched. The efficacy and safety of the WATCHMANTM
device against long-term oral warfarin have been evaluated in 2 randomized controlled trials (RCTs): PROTECT AF161
and PREVAIL.162
In a combined analysis of these two RCTs,163
there were no significant differences in the primary efficacy endpoints (stroke, systemic embolism, and cardiovascular death/death with an unknown cause) and the primary safety endpoints (serious bleeding, and procedure-related complications) between the WATCHMANTM
device and long-term oral warfarin groups. Although there was no significant difference in the incidence of cerebral infarction events between groups, the incidences of hemorrhagic stroke (hazard ratio [HR]: 0.2, 95% confidence interval [CI]: 0.07–0.56, P=0.0022), disabling/fatal stroke (HR: 0.45, 95% CI: 0.21–0.94, P=0.034), and cardiovascular death/death with an unknown cause (HR: 0.59, 95%: CI: 0.37–0.94, P=0.027) were significantly lower in the WATCHMANTM
group. The results indicated that the LAAC device could serve as an alternative to oral anticoagulant (OAC) for non-valvular atrial fibrillation (NVAF) patients at high risk for bleeding events. In the SALUTE study (n=42) verifying the efficacy and safety of the WATCHMANTM
device in Japanese NVAF patients at risk for cerebral infarction (CHA2DS2-VASc score ≥2), the procedure success rate and complication rate were similar to those reported in previous studies.164
During a 2-year follow-up, there were 3 ischemic events (7.1%) without any sequelae.19
Successful LAAC, defined as a peri-device leak ≤5 mm, was achieved in all patients, but an asymptomatic device thrombus was detected in 2 patients (4.8%).19
In a combined analysis of 2 RCTs and 2 large registries, the incidence of device thrombus was 3.7%,165
similar to that of the SALUTE trial.
2. Considerations in Japanese Patients
During the SALUTE trial, all patients received a ≥27 mm WATCHMANTM
device; a ≥27 mm device was used in <40% of the enrolled patients in the PROTECT-AF and PREVAIL trials. In the WASP registry enrolling patients from Asia and the Pan-Pacific region, significantly larger devices were implanted when compared with Western countries;166
therefore, race may affect LAA size. A larger LAA orifice has been reported as a predictor of device thrombus,165
and worsening of residual peri-device leak was significantly frequent in longstanding persistent AF patients with a WATCHMANTM
device ≥27 mm.167
Hence, further evaluation of the incidence of device thrombus or residual peri-device leak in Japanese patients who receive larger devices is required, with larger sample sizes.
3. LAAC as an Alternative Option to DOAC
Non-inferiority of the LAAC to direct oral anticoagulants (DOACs) was demonstrated in the left atrial appendage closure vs. novel anticoagulation agents in atrial fibrillation (PRAGUE-17) trial.168
Patients with NVAF requiring OAC, with a history of bleeding requiring intervention or hospitalization, a cardioembolic event while taking an OAC, and/or CHA2DS2-VASc ≥3 and HAS-BLED ≥2, were enrolled and randomized to receive LAAC or DOAC. The implanted devices were either AmuletTM
(61.3%), WatchmanTM
(35.9%), or Watchman-FLXTM
(2.8%), and most patients received dual anti-platelet therapy (DAPT). During a median follow-up of 19.9 months, the annual rates of the primary outcomes (stroke, transient ischemic attack, systemic embolism, cardiovascular death, major or non-major clinically relevant bleeding, or procedure-/device-related complications) were 10.99% with LAAC and 13.42% with DOAC (HR: 0.84; 95% CI: 0.53–1.31; P=0.44). Among patients at high risk for stroke and bleeding, LAAC was non-inferior to DOAC in preventing composite cardiovascular events.
4. Peri-Procedural Complications
Several critical peri-procedural complications, such as cardiac tamponade, thromboembolism, peri-device leak, and device embolism, have been reported. Procedure-related complications were especially observed during the pivotal PROTECT-AF trial, but these adverse events declined with increasing operator experience.169
Although peri-procedural complications requiring surgical intervention are rare, the rate of cardiac tamponade was 1.2% in the post FDA approval USA experience,170
and 31% of the cases required surgical intervention. Embolization of a large sized device (≥27 mm) into the left atrium or left ventricle can cause hemodynamic instability, thereby prompting urgent surgical device retrieval.171
Therefore, it is important to evaluate the morphology of left atrial appendage before procedure, and perform LAAC with a surgical back-up.
5. Indications
In Japan, LAAC is approved for NVAF patients indicated for long-term OAC with a high risk of bleeding (HAS-BLED ≥3, a history of BARC type 3 bleeding, requiring DAPT ≥1 year, multiple episodes of falls requiring interventions, or a history of cerebral amyloid angiopathy).172
Patients with a history of a cardioembolic event while taking an OAC or poor compliance with OAC therapy are not eligible for LAAC. In contrast to OACs, the effectiveness and safety of LAAC in patients with chronic renal failure regardless of severity have been reported.173,174
However, a significant association between acute kidney disease (an absolute or a relative increase in serum creatinine of >0.3 mg/dL or ≥50% of baseline value) and higher mortality was reported, thus requiring preventive strategies, especially in patients with chronic kidney disease.175
In the ESC and AHA/ACC/HRS guidelines,110,176
percutaneous LAAC is classified as a Class IIb indication for stroke prevention in patients at an increased risk of stroke and with contraindications to long-term OAC therapy. Previous studies have shown the superiority of LAAC to long-term warfarin,163
and non-inferiority to DOACs;168
thus, LAAC could be considered in NVAF patients and those with an increased risk of stroke and contraindications to long-term OAC (Class IIb,
Table 14). The procedural success rate and safety in the SALUTE trial were similar to those in the studies from Western countries, but the sample size was small and appeared to have different characteristics (i.e., larger device size). A registry with a larger sample size is required to establish the evidence for LAAC in Japanese patients.
Table 14.
Recommendations and Evidence Levels for Left Atrial Appendage Closure Device
| |
COR |
LOE |
GOR
(MINDS) |
LOE
(MINDS) |
Percutaneous left atrial appendage closure may be considered in patients
with non-valvular AF who are at an increased risk of stroke and have
contraindications to long-term anticoagulation |
IIb |
B |
B |
II |
AF, atrial fibrillation; COR, class of recommendation; GOR, Grade of Recommendation; LOE, level of evidence.
An animal study demonstrated that the anti-thrombotic effect of the LAAC device appears at 90 days after endothelialization of the device.177
Nevertheless, a combined analysis of 2 RCTs and 2 registries revealed that 3.7% of patients suffered from device thrombosis, and a majority of the thrombi was found after cessation of warfarin at 45 days.165
In an observational registry, an antithrombotic regimen using DOAC and antiplatelet therapy demonstrated similar efficacy to that of warfarin.178
In patients ineligible for OAC due to high bleeding risk, an antithrombotic regimen using DAPT was investigated. In the ASAP trial, an antithrombotic regimen using DAPT was used in patients contraindicated for OAC, and device thrombus was observed in 4.0% of them.179
In the EWOLUTION registry, DAPT was used in 62% of the patients, and device thrombus was found in 4.0%.180
There is no RCT comparing antithrombotic regimens using warfarin with antiplatelet therapy vs. DAPT, but combined analysis of 6 large studies using propensity-matched analysis showed that device thrombosis was significantly more frequent in patients receiving DAPT.181
Therefore, antithrombotic therapy using OACs should be considered for patients who are eligible, but in patients with high bleeding risk and those contraindicated for using OACs, an antithrombotic regimen should be considered individually, as well as surveillance for device thrombus.
In patients with device thrombus, there was an approximately 4-fold increase in the incidence of thromboembolism, though resolution of device thrombus could be achieved in a majority of cases under OAC.165,182
However, recurrence of device thrombus was reported in 34.8% of patients after discontinuation of OACs.183
Therefore, an antithrombotic regimen after device thrombus resolution should be considered individually, taking into account the bleeding risk.
