Percutaneous Left Atrial Appendage Occlusion for Atrial Fibrillation in Cardiac Amyloidosis

ABSTRACT

Background

Cardiac amyloidosis (CA) is frequently complicated by atrial fibrillation (AF), yet outcomes after left atrial appendage occlusion (LAAO) in this population remain poorly defined.

Methods

We conducted a retrospective TriNetX study of adults with AF undergoing LAAO, comparing patients with and without CA after 1:1 propensity matching. Outcomes were assessed using Kaplan–Meier analyses and Cox regression.

Results

Among 532 matched pairs, mortality and major adverse cardiovascular events were similar between groups, whereas major bleeding was higher in CA (HR 1.90).

Conclusions

LAAO yields comparable ischemic outcomes in CA, though bleeding risk is increased.

Left atrial appendage occlusion‐related outcomes in patients with amyloid cardiomyopathy.

1. Introduction

Cardiac amyloidosis (CA) is associated with significant morbidity and mortality and is commonly associated with atrial fibrillation (AF) [1]. In the general population with nonvalvular AF and increased bleeding risk, left atrial appendage occlusion (LAAO) has been shown to be a reasonable alternative to oral anticoagulation (AC) [2, 3], however, LAAO remains understudied in patients with CA.

2. Methods

We performed a retrospective study utilizing the TriNetX Network, including health records from 106 healthcare institutions. Eligible participants were adults aged 18 years or older diagnosed with AF (ICD‐10 code I48) who underwent percutaneous LAAO (CPT 33340) between January 1, 2010, and December 31, 2024. Two study groups were established based on the presence or absence of CA (ICD‐10 code E85.4). Individuals with a prosthetic heart valve (Z95.2) or rheumatic mitral stenosis (I05.0) were excluded. The date of the LAAO procedure was designated as the index date.

Follow‐up began 30 days after the index procedure. Primary endpoint was all‐cause mortality, defined by death records. Secondary endpoints included ischemic stroke, acute myocardial infarction, all‐cause hospitalization (any inpatient encounter), major adverse cardiovascular events (MACE; composite of myocardial infarction, cardiac arrest, and ischemic stroke), acute heart failure events (hospitalization for heart failure or administration of intravenous diuretics), and major bleeding (intracranial or major gastrointestinal bleeding codes).

Propensity score matching (1:1 greedy nearest‐neighbor) was conducted to balance baseline characteristics between CA and non‐CA groups. Matching variables included demographics, comorbid conditions, baseline medications, and available laboratory parameters. Event‐free survival was analyzed using Kaplan–Meier estimates, with comparisons made via log‐rank tests. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using Cox proportional hazards regression. Statistical significance was defined as a two‐sided p < 0.05. Institutional review board approval and informed consent were not required, as the analysis utilized anonymized, publicly available data in accordance with the Common Rule.

3. Results

A total of 532 patients in each cohort were included. Mean age was 80.1 ± 7.5 years and 78.7 ± 7.5 years, respectively, with ~30% women. Most participants were White (~78%), and a small proportion were Hispanic (~3.0%). Prevalence of hypertension (~92%), diabetes mellitus (~37%), chronic kidney disease (~40%), and heart failure (~61%) was similar between groups. Anticoagulant therapy (i.e., apixaban, edoxaban, rivaroxaban, dabigatran, or warfarin) post LAAO was used in nearly all patients (95.9% vs. 95.3%). BMI (~27–28 kg/m²) and hemoglobin A1c (6.0%–6.1%) were similar between cohorts. Mean follow‐up time was 1.90 years in the CA cohort and 2.24 years in the non‐CA cohort (Table 1).

TABLE 1.

Baseline characteristics. Pre‐ and post‐propensity score matching baseline characteristics of the included cohorts.

Characteristic	Pre‐PSM amyloid (n = 535)	Pre‐PSM no amyloid (n = 26 222)	Pre, p	Pre std diff	Post‐PSM amyloid (n = 532)	Post‐PSM no amyloid (n = 532)	Post, p	Post std diff
Demographics
Age, mean (SD)	80.4 (6.9)	78.7 (7.8)	< 0.001	0.224	80.3 (6.8)	80.1 (7.5)	0.580	0.034
White, n (%)	420 (78.5)	22 683 (86.5)	< 0.001	0.212	420 (78.9)	414 (77.8)	0.655	0.027
Female, n (%)	161 (30.1)	10 297 (39.3)	< 0.001	0.194	161 (30.3)	151 (28.4)	0.501	0.041
Hispanic or Latino, n (%)	15 (2.8)	736 (2.8)	0.997	< 0.001	15 (2.8)	16 (3.0)	0.855	0.011
Black or African American, n (%)	50 (9.3)	1231 (4.7)	< 0.001	0.183	48 (9.0)	57 (10.7)	0.355	0.057
Comorbid problems
Hypertensive diseases, n (%)	497 (92.9)	23 143 (88.3)	0.001	0.159	494 (92.9)	488 (91.7)	0.490	0.042
Thyroid disorders, n (%)	181 (33.8)	7858 (30.0)	0.054	0.083	180 (33.8)	178 (33.5)	0.897	0.008
Z55–Z65 SDoH codes, n (%)	32 (6.0)	1067 (4.1)	0.027	0.088	32 (6.0)	27 (5.1)	0.503	0.041
Overweight/obesity, n (%)	203 (37.9)	10 433 (39.8)	0.388	0.038	202 (38.0)	210 (39.5)	0.615	0.031
Alcohol‐related disorders, n (%)	34 (6.4)	1805 (6.9)	0.632	0.021	34 (6.4)	46 (8.6)	0.163	0.086
Lipid disorders, n (%)	457 (85.4)	20 536 (78.3)	< 0.001	0.185	454 (85.3)	455 (85.5)	0.931	0.005
Sleep apnea, n (%)	222 (41.5)	9609 (36.6)	0.021	0.100	220 (41.4)	215 (40.4)	0.755	0.019
Cerebral infarction, n (%)	214 (40.0)	4888 (18.6)	< 0.001	0.483	211 (39.7)	211 (39.7)	1.000	< 0.001
Tobacco use, n (%)	23 (4.3)	1150 (4.4)	0.923	0.004	23 (4.3)	24 (4.5)	0.881	0.009
Neoplasms, n (%)	316 (59.1)	13 194 (50.3)	< 0.001	0.176	316 (59.4)	322 (60.5)	0.707	0.023
Chronic ischemic heart disease, n (%)	324 (60.6)	15 295 (58.3)	0.300	0.045	324 (60.9)	317 (59.6)	0.661	0.027
Chronic lower respiratory diseases, n (%)	198 (37.0)	9519 (36.3)	0.736	0.015	196 (36.8)	188 (35.3)	0.610	0.031
Diabetes mellitus, n (%)	210 (39.3)	10 071 (38.4)	0.691	0.017	208 (39.1)	194 (36.5)	0.376	0.054
Chronic kidney disease, n (%)	222 (41.5)	8681 (33.1)	< 0.001	0.174	220 (41.4)	209 (39.3)	0.492	0.042
Heart failure, n (%)	329 (61.5)	12 355 (47.1)	< 0.001	0.292	326 (61.3)	333 (62.6)	0.659	0.027
Acute myocardial infarction, n (%)	124 (23.2)	4442 (16.9)	< 0.001	0.156	121 (22.7)	111 (20.9)	0.458	0.046
TIA, n (%)	94 (17.6)	2938 (11.2)	< 0.001	0.182	94 (17.7)	78 (14.7)	0.183	0.082
Peripheral artery disease, n (%)	290 (54.2)	11 457 (43.7)	< 0.001	0.211	289 (54.3)	290 (54.5)	0.951	0.004
GI hemorrhage, n (%)	88 (16.4)	5877 (22.4)	0.001	0.151	87 (16.4)	90 (16.9)	0.805	0.015
Coronary therapeutic procedures, n (%)	43 (8.0)	2224 (8.5)	0.715	0.016	43 (8.1)	47 (8.8)	0.659	0.027
Pharmacotherapies
Beta blockers, n (%)	497 (92.9)	21 790 (83.1)	< 0.001	0.305	494 (92.9)	499 (93.8)	0.539	0.038
Calcium channel blockers, n (%)	366 (68.4)	15 782 (60.2)	< 0.001	0.172	363 (68.2)	353 (66.4)	0.513	0.040
Statins, n (%)	442 (82.6)	18 840 (71.8)	< 0.001	0.259	439 (82.5)	445 (83.6)	0.624	0.030
ARBs, n (%)	232 (43.4)	9976 (38.0)	0.012	0.108	230 (43.2)	239 (44.9)	0.578	0.034
Spironolactone, n (%)	145 (27.1)	4305 (16.4)	< 0.001	0.261	142 (26.7)	136 (25.6)	0.675	0.026
Aspirin, n (%)	451 (84.3)	18 098 (69.0)	< 0.001	0.367	448 (84.2)	449 (84.4)	0.933	0.005
Anticoagulants, n (%)	513 (95.9)	23 217 (88.5)	< 0.001	0.277	510 (95.9)	507 (95.3)	0.654	0.027
Sacubitril, n (%)	44 (8.2)	1232 (4.7)	< 0.001	0.144	43 (8.1)	47 (8.8)	0.659	0.027
Loop diuretics, n (%)	347 (64.9)	13 835 (52.8)	< 0.001	0.248	344 (64.7)	348 (65.4)	0.797	0.016
ACE inhibitors, n (%)	261 (48.8)	10 562 (40.3)	< 0.001	0.172	259 (48.7)	244 (45.9)	0.357	0.056
Insulin, n (%)	222 (41.5)	9276 (35.4)	0.003	0.126	220 (41.4)	215 (40.4)	0.755	0.019
Metformin, n (%)	107 (20.0)	5059 (19.3)	0.682	0.018	107 (20.1)	90 (16.9)	0.180	0.082
NSAIDs, n (%)	106 (19.8)	5120 (19.5)	0.868	0.007	105 (19.7)	90 (16.9)	0.235	0.073
Key clinical factors and laboratory data
BMI, mean (SD)	27.7 (5.5)	29.8 (6.5)	< 0.001	0.355	27.7 (5.5)	28.9 (5.9)	0.002	0.217
Hemoglobin A1c, mean (SD)	6.0 (1.0)	6.2 (1.3)	0.070	0.104	6.0 (1.0)	6.1 (1.4)	0.367	0.066
eGFR, mean (SD)	62.3 (26.2)	63.3 (23.7)	0.364	0.039	62.3 (26.2)	60.3 (23.7)	0.195	0.081
Hemoglobin, mean (SD)	12.6 (2.1)	12.7 (2.1)	0.269	0.050	12.6 (2.1)	12.8 (2.1)	0.206	0.080
LVEF, mean (SD)	56.3 (10.7)	55.5 (13.4)	0.496	0.065	56.1 (10.7)	54.0 (14.2)	0.165	0.170
INR, mean (SD)	1.3 (0.4)	1.3 (0.5)	0.008	0.132	1.3 (0.4)	1.4 (0.5)	0.018	0.154
Platelets, mean (SD)	207.8 (72.5)	219.0 (75.3)	0.001	0.151	208.0 (72.5)	213.5 (82.3)	0.265	0.070

The incidence of ischemic stroke was similar between the CA and non‐CA cohorts (24.4% vs. 22.4%; HR 1.225, 95% CI 0.955–1.571) (Figure 1). There were no differences in acute myocardial infarction (13.5 vs. 13.5%; HR 1.140, 95% CI 0.822–1.581) or all‐cause hospitalizations between the two groups (50.2% vs. 54.1%; HR 1.058, 95% CI 0.896–1.250). Moreover, there was no difference observed in the events of acute heart failure (16.5% vs. 18.3%; HR 1.117, 95% CI 0.749–1.668), cardiac arrest (3% vs. 2.8%; HR 1.235, 95% CI 0.610–2.501), or all‐cause mortality (19.2% vs. 17.5%; HR 1.316, 95% CI 0.993–1.744). The risk of MACEs was similar between the two groups (35.3% vs. 35.2%; HR 1.146, 95% CI 0.936–1.404). However, there was a significant difference in the incidence of major bleeding, which was higher in the CA cohort compared to the non‐CA group (33.6% vs. 21.6%; HR 1.899, 95% CI 1.502–2.402).

FIGURE 1.

Outcomes. (A) Kaplan–Meier plot revealing ischemic stroke‐free survival probabilities and (B) forest plot of all hazard ratios regarding LAAO in patients with and without cardiac amyloidosis.

4. Discussion

Patients with CA who underwent LAAO had comparable rates of mortality and MACE to patients without known CA, though major bleeding was significantly higher. Patients with CA and atrial fibrillation have high risks of thromboembolic potential, warranting anticoagulation per AHA/ACC/HFSA guidelines (Class 2A) [4]. These competing risks make LAAO an important yet challenging therapeutic consideration.

In a US analysis of 220 patients with CA and 54 680 patients without CA, bleeding and stroke‐related admissions were similar, though CA was linked to higher peri‐procedural neurological events [5]. The CAMYLAAC study (40 ATTR‐CA vs. 1119 patients without CA) likewise found higher long‐term mortality in CA but no excess stroke or bleeding [1]. Our study extends these findings by including the largest CA cohort to date, revealing observed higher bleeding risk which may reflect amyloid‐related vascular fragility, small‐vessel involvement, or challenges with anticoagulation management [6]. Similarly, although major bleeding was more frequent in the CA cohort, all‐cause hospitalization rates were similar between groups, likely reflecting that bleeding accounts for only a subset of hospitalizations, and other causes of admission predominate.

Limitations of our study include lack of granularity regarding amyloidosis subtype and procedural details such as peri‐device leak or imaging follow‐up. CA was identified using the ICD‐10 code E85.4, which likely underestimates the true prevalence of disease and may lead to misclassification and potentially attenuate observed differences between cohorts. Similarly, ischemic and bleeding events were queried using diagnosis codes without formal clinical adjudication. Such non‐differential misclassification of outcomes may underestimate true differences in event rates. Lastly, the TriNetX database does not capture information regarding dosing, adherence, or duration of antithrombotics after LAAO. Therefore, we cannot determine whether the higher incidence of major bleeding in cardiac amyloidosis is driven primarily by intrinsic disease‐related vascular fragility, by differences in antithrombotic management, or both.

5. Conclusions

Patients with CA undergoing LAAO demonstrated comparable mortality and MACE rates to those without CA but a higher incidence of major bleeding. These findings highlight the need for prospective investigations to better define the safety profile of LAAO in CA.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Almakadma A. H., Ibrahim R., Pham H. N., et al., “Percutaneous Left Atrial Appendage Occlusion for Atrial Fibrillation in Cardiac Amyloidosis,” Journal of Arrhythmia 42, no. 1 (2026): e70274, 10.1002/joa3.70274.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.