Left Atrial Appendage Occlusion in Patients With Non-Valvular Atrial Fibrillation and Cerebral Amyloid Angiopathy

Insights From the LOGIC (Left atrial appendage Occlusion in patients with Gastrointestinal or IntraCranial bleeding) International Multicenter Registry

© 2024 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of the Journal of Invasive Cardiology or HMP Global, their employees, and affiliates. 

J INVASIVE CARDIOL 2024. doi:10.25270/jic/24.00239. Epub November 19, 2024.

Objectives. Oral anticoagulation therapy (OAC) is often contraindicated in patients with atrial fibrillation (AF) and cerebral amyloid angiopathy (CAA) because of the high hemorrhagic risk. Left atrial appendage occlusion (LAAO) can prevent thromboembolic events while avoiding long-term anticoagulation.  However, a short period of antithrombotic therapy (AT) is still recommended after LAAO, and, therefore, it is unclear whether patients with CAA can be candidates for LAAO. The aim of the study was to investigate the safety and efficacy of LAAO in patients with CAA and AF.

Methods. In this sub-study of the LOGIC (Left atrial appendage Occlusion in patients with Gastrointestinal or IntraCranial bleeding) registry, the authors considered only patients with previous intracranial (IC) bleeding, and patients with CAA were compared with patients who did not have CAA. Outcomes of interest were death from any causes and cardiovascular death, ischemic stroke, transient ischemic attack and systemic embolization, and any bleeding and major bleeding at 12 months.

Results. The analysis included 270 patients, 49 (18%) of whom had CAA. Patients with CAA were more frequently discharged without AT after LAAO compared with patients who did not have CAA (36.7% vs 6.8%, P < .001), and this was confirmed at the 1-year follow-up (30.4% vs 14.1%, P = .001). There were no significant differences in all-cause or cardiovascular mortality, or ischemic or hemorrhagic endpoints at 1 and 12 months.

Conclusions. LAAO seems to be safe and effective in reducing both ischemic and hemorrhagic risk in patients with AF and CAA. Although patients with CAA are more likely to be discharged without AT after LAAO, there are no significant differences in ischemic and hemorrhagic outcomes compared with patients with a history of IC bleeding from other causes.

Cerebral amyloid angiopathy (CAA) is an age-related disease that involves the pathological deposition of β-amyloid protein in leptomeningeal and cortical cerebral vessels. These vascular alterations increase the risk of intracranial bleeding, particularly when oral anticoagulation (OAC) is prescribed.¹ With the progressive increase in the aging population, the concomitant diagnosis of non-valvular atrial fibrillation (NVAF) and CAA is becoming a common clinical scenario. Patients with a concomitant diagnosis of cerebral amyloid angiopathy (CAA) and atrial fibrillation (AF) often present a clinical dilemma because of the coexistence of hemorrhagic and ischemic risks.

Percutaneous left atrial appendage occlusion (LAAO) is a safe and effective alternative for reducing the risk of ischemic stroke in patients with non-valvular AF and contraindications to long-term OAC due to hemorrhagic risk.² Therefore, LAAO could be an attractive option for patients with CAA and AF. Notwithstanding, LAAO devices typically require, at minimum, short-term antithrombotic therapy after the procedure. The use of antithrombotic agents following LAAO may increase the risk of intracranial (IC) bleeding in CAA patients, particularly those with previous lobar intracranial hemorrhage.³ Clear evidence-based recommendations for managing these clinical scenarios are lacking, and therefore the therapeutic strategy is commonly decided on a case-by-case basis. In some cases, a short period of single low-dose antiplatelet therapy is prescribed after LAAO, whereas in other patients, any AT is avoided. Data on outcomes in this specific clinical scenario are limited to a few small observational studies.^4-6

The aim of this LOGIC (Left atrial appendage Occlusion in patients with Gastrointestinal or IntraCranial bleeding) registry sub-study was to investigate the safety and efficacy of LAAO in patients with NVAF and CAA.

Study population

The patient selection and methods of the LOGIC registry have been previously reported.⁷ In brief, the study enrolled consecutive patients with NVAF who were referred to 7 participating centers in Europe and the USA between January 2010 and January 2021 and underwent LAAO because of a history of gastrointestinal (GI) or IC bleeding. Data were collected prospectively at each center, entered into a dedicated database, and analyzed retrospectively. Variables collected included demographics, baseline data, indications for LAAO, CHA2DS2-VASc score, HAS-BLED score, antithrombotic medications at discharge and follow-up, procedural characteristics, peri-procedural adverse events, and clinical and imaging follow-up. Patients without events and with a follow-up of less than 12 months were excluded from the final analysis. The clinical indication for LAAO was determined through a multidisciplinary approach at each participating center.

Prior to the procedure, all patients underwent imaging assessments using transesophageal echocardiography (TOE) and/or cardiac multidetector computed tomography (MDCT) to exclude LAA thrombus and confirm the technical feasibility of the procedure through device sizing. The choice of postprocedural therapy was determined based on the physician's preference.

For the purposes of this sub-study, only patients with a history of IC hemorrhage were taken into consideration. These patients were then divided into 2 groups, depending on whether they had received a diagnosis of CAA.

Cerebral amyloid angiopathy assessment

Patients included in the LOGIC registry for history of IC bleeding were reviewed by the participating physicians at each center and assigned to either the CAA group or the non-CAA group based on current diagnostic criteria for CAA.⁸ Probable CAA was defined as the presence of 2 or more strictly lobar hemorrhagic lesions or 1 lobar hemorrhagic lesion and at least 1 white matter lesion. Possible CAA was defined as the presence of a single lobar hemorrhagic or white matter lesion.

Left atrial appendage occlusion procedure and follow-up

Procedures were performed under general anesthesia, conscious sedation, or local anesthesia, according to recommended practices reported in international consensus documents.^9-10 LAAO was guided by fluoroscopy and TOE or intracardiac echocardiography (ICE) according to institutional protocols and operator preference. An ACT of greater than 250 seconds was achieved by administration of unfractionated heparin. All patients were monitored overnight in the hospital and usually discharged 1 or 2 days after the procedure, depending on institutional policy. Adverse events were considered periprocedural if they occurred before hospital discharge or within 7 days of the LAAO procedure. Periprocedural bleeding was not included in the 12-month bleeding outcome. The choice and duration of post-procedural AT was tailored to each patient's hemorrhagic and thromboembolic risk profile and acute procedural success and was left to physician preference. Post-procedural imaging (TOE and/or MDCT) to confirm proper device implantation and to rule out residual device-related leak or thrombosis was performed according to each center's protocol or when clinically indicated. Patient data were collected in accordance with institutional ethics guidelines and the study was approved by the institutional ethics committee.  Written informed consent was obtained from all patients before LAAO.

Study outcomes

Outcomes of interest including the following: death from any causes and cardiovascular death, ischemic stroke, transient ischemic attack and systemic embolization, and any bleeding and major bleeding at 12 months, excluding periprocedural events. Major bleeding events were defined as type 3 or greater on the Bleeding Academic Research Consortium (BARC) scale.¹¹ Other outcomes were spontaneous myocardial infarction, procedural success, and periprocedural complications.  Only the first event at follow-up was considered for both bleeding and stroke. Definitions and clinical endpoints were defined according to the Munich consensus document.¹² Technical success was defined as successful device deployment and implantation; procedural success was defined as technical success without major procedure-related complications (major pericardial effusion/tamponade, stroke, systemic embolism, major bleeding, and device embolization requiring surgical removal). Pericardial effusion was considered clinically relevant if pericardiocentesis or surgery was required.

Statistical analysis

The Kolmogorov-Smirnov test was used before analysis to test data for normality. Categorical variables were expressed as numbers and percentages, and continuous variables were expressed as mean ± SD or median and IQR, as appropriate. The unpaired 2-tailed t test or nonparametric Mann-Whitney U test was used for comparisons of continuous variables, and the chi-squared test was used for categorical variables. Technical and procedural success rates were calculated as percentages of the total number of patients. Patient-years were calculated as the product of the number of patients and the mean years of follow-up. Rates of stroke/bleeding events were calculated as the number of events per 100 patient-years. All P-values were 2-sided, and a P-value of less than 0.05 was considered significant. Statistical analysis was performed using SPSS version 27 (IBM).

Out of the 270 patients with a history of IC bleeding included in the LOGIC registry, 49 (18%) were diagnosed with CAA. Table 1 reports the baseline clinical characteristics. The CAA group had a lower prevalence of ischemic heart disease compared with the non-CAA group. However, there were no significant differences between the 2 populations in terms of the remaining clinical characteristics. Of the 49 CAA patients, 15 (30.6%) met the Boston Criteria 2.0 for a possible diagnosis of CAA, while the remaining 34 (69.4%) met the criteria for a probable diagnosis.

Table 2 presents the procedural features. The duration of the procedure was slightly but significantly longer in the CAA group (61.1 ± 28.5 min vs 63.5 ± 18.3 min; P = .008). No differences were found regarding procedural complications between the 2 groups (7.2% vs 4.1%; P = .22). In the CAA group, 2 patients (4.1%) experienced periprocedural major bleeding. One patient developed an IC hemorrhage 12 hours after starting dual antiplatelet therapy, while the second patient experienced a severe drop in hemoglobin levels, requiring a blood transfusion. The most frequently implanted devices were the Watchman (Boston Scientific) and the Amulet (Abbott), and there were no differences between the 2 groups in terms of device choice.

At discharge, the general population and non-CAA group showed significant heterogeneity in terms of AT (Table 3). Patients diagnosed with CAA were discharged without any AT in 36.7% of cases (vs 6.8% in the non-CAA group; P < .001). Only 1 patient in the CAA group was discharged with anticoagulant therapy in addition to single antiplatelet therapy (SAPT). At 12 months, most patients in both groups were taking SAPT (69.6% in the CAA group and 71.4% in the non-CAA group; P = .09). However, there were about twice as many patients with CAA as those without CAA who were not taking any AT (30.4% vs 14.1%; P = .01).

At 12 months, there were no significant differences in overall mortality (6.8% vs 4.1%; P = .48) or cardiovascular mortality (2.3% vs 2.0%; P = .92) (Table 4). Ischemic stroke rates were comparable between the 2 groups (1.3% vs 2.0%; P = .72). Additionally, no patients with CAA experienced TIA or any systemic embolism. Hemorrhagic outcomes were similar between the 2 groups, with no substantial differences in either any or severe bleeding (Figure). Device-related thrombosis (DRT) was a rare occurrence—although it was only ruled out in about half of the patients—mostly due to the COVID-19 pandemic, which limited follow-up diagnostics.

The major findings of the present study are as follows:

1. The LAAO procedure is as safe and effective in patients with possible or probable CAA as in patients with previous IC bleeding from other causes.
2. Patients with CAA usually receive a more cautious antithrombotic pharmacologic therapeutic strategy than patients without CAA after LAAO because of concerns about hemorrhagic complications.
3. Although an AT-free strategy is more frequently used in patients with CAA after LAAO, there is no increase in ischemic events at follow-up up to 12 months.

LAAO is a widely used strategy to reduce the risk of embolization in patients with atrial fibrillation and contraindication to anticoagulation because of the concomitant high risk of bleeding. In early validation studies of the procedure, early post-implantation antithrombotic therapy with anticoagulants and a single antiplatelet agent was administered to avoid device-related thrombotic complications.¹³ Later, post-implantation therapy with dual antiplatelet therapy was tested and validated.¹⁴ These approaches raised questions about the feasibility of such a therapy, even for a short period of time, in patients at very high risk of bleeding. In real-world clinical practice, it is not uncommon for a recommended post-implant antithrombotic therapy to be abandoned in favor of off-label milder antithrombotic strategies, such as single antiplatelet therapy, or even for antithrombotic therapy to be avoided altogether.^15-17 Recently, a low-dose direct oral anticoagulant (DOAC) approach after LAAO has been tested and shown to be safe and effective in reducing bleeding events without increasing ischemic risk.^18-19 The optimal antithrombotic regimen after LAAO in high-risk bleeding patients remains controversial. Post-implantation therapy is often left to the discretion of the operator and evaluated on a case-by-case basis, taking into account the risk-benefit ratio and balancing ischemic and hemorrhagic risk.

Intracranial hemorrhage is one of the most serious complications of oral anticoagulant therapy. LAAO has been shown to be safe and effective in patients with a history of intracranial hemorrhage and a contraindication to resuming anticoagulation.²⁰ However, among patients with a history of intracranial hemorrhage, those with CAA represent a population at very high risk of bleeding. Therefore, there is often concern about initiating any type of low-dose antithrombotic therapy. Patients with AF and CAA have a narrow pharmacological therapeutic margin. Therefore, LAAO may be a viable alternative despite concerns related to post-implantation AT.¹

Data on LAAO in patients with CAA are limited and come from small observational case series. A multicenter observational study of 26 patients with CAA who underwent LAAO showed that post-implantation therapy was highly variable in both intensity and duration, ranging from full anticoagulation therapy (with DOAC or vitamin K antagonist) to single antiplatelet therapy, and, in 1 case, no therapy was given.⁶ A single-center observational study enrolled 39 patients with possible or probable CAA treated with LAAO. The majority of patients were discharged on monotherapy with antiplatelet therapy for at least 1 month. The study found an incidence of ischemic stroke of 7.7 (95% CI, 1.9-20.9) per 100 person-years (expected rate 8.0; 95% CI, 7.0-9.1) and an incidence of intracranial hemorrhage of 4.4 (95% CI, 0.7-14.7) per 100 person-years (expected rate 7.0; 95% CI, 6.0-8.1).⁵ Both studies concluded that LAAO is feasible in patients with CAA and AF to reduce ischemic and hemorrhagic risk.

To our knowledge, our study analyzed the largest reported population with CAA and AF undergoing LAAO. We demonstrated that LAAO is safe and effective in these patients. In addition, we found that a light approach to post-procedural antithrombotic therapy, consisting mainly of SAPT or even the absence of post-procedural AT, does not increase the risk of ischemic events.

Limitations

Limitations of the LOGIC registry have been previously reported.⁷ The sample size of this sub-study for the CAA group is relatively small. Therefore, it would be desirable to conduct a larger study to validate and confirm the safety and efficacy endpoints, including the incidence of bleeding complications and stroke. However, we present one of the most consistent datasets to date on patients with CAA treated with LAAO. In approximately half of the patients, DRT could not be excluded, mainly because of logistical difficulties related to the pandemic COVID-19, which reduced the statistical power of this particular endpoint. Finally, in the control group, the location of the previous hemorrhage was not specified.

LAAO seems to be safe and effective in reducing both ischemic and hemorrhagic risk in patients with AF and CAA. Although patients with CAA are more likely to be discharged without AT after LAAO, there are no significant differences in ischemic and hemorrhagic outcomes compared with patients with a history of IC bleeding from other causes. Randomized clinical trials are needed to determine the best post-procedural antithrombotic regimen to balance the post-procedural ischemic and hemorrhagic risks.

Federico Ronco, MD¹; Gianpiero D’Amico, MD, PhD¹; Samuele Meneghin, MD¹; Domenico G. Della Rocca, MD, PhD²; Patrizio Mazzone, MD³; Stefano Bordignon, MD⁴; Gavino Casu, MD⁵; Pierluigi Merella, MD⁵; Francesco Giannini, MD⁶; Sergio Berti, MD⁷; Giuseppe D’Angelo, MD³; Maria Rita Romeo, ENG⁷; Marco Barbierato, MD¹; Andrea Natale, MD^2,8,9,10; Sakis Themistoclakis, MD¹; Francesco Gallo, MD¹; the LOGIC investigators

From the ¹Department of Cardiology, Ospedale dell'Angelo, AULSS3 Serenissima, Mestre, Venezia, Italy; ²Texas Cardiac Arrhythmia Institute, St. David's Medical Center, Austin, Texas; ³Department of Cardiac Electrophysiology and Arrhythmology, IRCCS San Raffaele Hospital and Vita-Salute University, Milan, Italy; ⁴Cardioangiologisches Centrum Bethanien - Markus Krankenhaus, Frankfurt, Germany; ⁵Clinical and Interventional Cardiology, Sassari, Italy; ⁶IRCCS Ospedale Galeazzi Sant’Ambrogio, Milan, Italy; ⁷Fondazione Toscana G. Monasterio, Ospedale Del Cuore, Massa, Italy; ⁸Interventional Electrophysiology, Scripps Clinic, La Jolla, California; ⁹Department of Cardiology, MetroHealth Medical Center, Case Western Reserve University School of Medicine, Cleveland, Ohio; ¹⁰Department of Biomedicine and Prevention, Division of Cardiology, University of Tor Vergata, Rome, Italy.

Disclosures: Dr Ronco has received speaker honoraria from Boston Scientific and Abbott. Dr Mazzone is a proctor for Abbott and Boston Scientific. Dr. Casu is a proctor for Boston Scientific. Dr Berti is a proctor for Abbott, Boston Scientific, Edwards Lifesciences, and Johnson & Johnson.

Dr Natale has received speaker honoraria from Boston Scientific, Biosense Webster, Abbott, Biotronik, and Medtronic; and is a consultant for Biosense Webster and Abbott. Dr Gallo has received speaker honoraria from Boston Scientific. The remaining authors report no financial relationships or conflicts of interest regarding the content herein.

Address for correspondence: Francesco Gallo, MD, Department of Cardiology, Ospedale dell’Angelo, Via Paccagnella, 11, Mestre, Italy. Email: galfra87@gmail.com; X: @GalloFrance87