Abstract
Background
Left atrial appendage (LAA) closure (LAAC) could prevent stroke in patients with atrial fibrillation. However, LAAC may impair the compliance of the left atrium and result in poor prognosis. This study aimed to comparatively evaluate the prognosis of LAAC among patients with different sizes of LAA orifice.
Methods and Results
Three hundred two consecutive patients who underwent successful LAAC were included and divided into 4 groups based on LAA orifice size that was measured using transesophageal echocardiography. Clinical outcomes including thromboembolic events, major cardiocerebrovascular adverse events, and acute heart failure (AHF) were compared among 4 quartile groups and between propensity‐score matched groups of large and small LAAs. Through follow‐up of 39.6±8.4 months, survival of thromboembolic events was similar. Survival of major cardiocerebrovascular adverse events was significantly lower in the group with the largest LAA orifice (log‐rank P<0.001), including a higher incidence of AHF with New York Heart Association class III to IV (21.4%, log‐rank P=0.009). A large LAA orifice (by cutoff) could predict major cardiocerebrovascular adverse events (hazard ratio, 3.749 [95% CI, 2.074–6.779]) in most patients, except for subgroups of those aged <65 years, with paroxysmal atrial fibrillation, and/or with failed rhythm/rate control. Further compared with a propensity‐score matched small‐LAA group, the large‐LAA orifice group still presented worse survival of AHF with New York Heart Association class III to IV (log‐rank P=0.010).
Conclusions
Patients with a larger LAA orifice presented a worse prognosis after LAAC, including a higher incidence of AHF. A large LAA orifice could predict a post‐LAAC AHF event in most patients, except for young patients, patients with paroxysmal atrial fibrillation, and/or with failed rhythm/rate control.
Registration
URL: clinicaltrials.gov; Unique identifier: NCT04185142.
Keywords: heart failure, left atrial appendage, left atrial appendage closure, major adverse cardiocerebrovascular events, prognosis
Subject Categories: Atrial Fibrillation, Heart Failure
Nonstandard Abbreviations and Acronyms
- AHF
acute heart failure
- LAA
left atrial appendage
- LAAC
left atrial appendage closure
- LAVI
left atrial volume index
- MACCE
major adverse cardiocerebrovascular events
- NYHA
New York Heart Association
- PVI
pulmonary vein isolation
- TEE
transesophageal echocardiography
Clinical Perspective.
What Is New?
Based on our cohort that underwent successful left atrial appendage closure (LAAC), we observed an incidence of acute heart failure (AHF) following LAAC of 18.8%.
Compared with patients with a small left atrial appendage (LAA) orifice, although patients with larger LAA orifice had similar survival of thromboembolic events, a worse survival rate of major adverse cardiocerebrovascular events was observed, contributed by a high incidence of AHF with New York Heart Association class III to IV of 21.4%.
Survival analysis adjusting confounding factors showed that a large LAA orifice could be an independent risk factor for post‐LAAC AHF events in most patients, except for young patients, patients with paroxysmal atrial fibrillation, and/or patients with failed rhythm/rate control.
What Are the Clinical Implications?
AHF after LAAC is not an uncommon adverse event. Risk of AHF after LAAC and monitoring AHF during follow‐up should be taken into consideration for management of atrial fibrillation.
The closure of a large LAA could provide stroke prophylaxis, but could pose a risk of AHF following LAAC except for those young patients, patients with paroxysmal atrial fibrillation, and/or patients with failed rhythm/rate control.
Because LAA size and morphology could lead to a worse prognosis, LAA dimensional features should be thoroughly evaluated before LAAC.
The left atrial appendage (LAA) is a finger‐like structure originating from the embryonic left atrium (LA). As a structure rich in pectinate muscle, with a volume of ≈10 mL, 1 it serves as an essential reservoir that compensates for pressure and/or volume overload. Moreover, it has an endocrine function that regulates hemodynamics through secretion of natriuretic peptides. 2
In recent years, transcatheter LAA closure (LAAC) has been recognized as an effective method for stroke prevention. 3 However, closure of the LAA can cause an acute volume reduction of the total LA chamber leading to increased LA pressure. 4 Moreover, closure inhibits the active emptying of the LAA and interferes with its secretory function, eventually decreasing the overall compliance of the LA. 5
Concerns have been raised that LAAC might disturb hemodynamics and lead to the development of heart failure (HF). Nevertheless, the occurrence of acute HF (AHF) has not been evaluated in current, well‐designed LAAC trials, with only limited, small‐sized studies having reported that the incidence of AHF ranged from 16.0% to 23.9%, which was attributed mostly to preexisting HF. 6 , 7 , 8 More importantly, the dimensions of the LAA reflect its reservoir function, but the value of these dimensional features for predicting HF after LAAC has not yet been investigated.
In the present study, we enrolled patients who had undergone successful LAAC and stratified them according to their LAA dimensions by orifice diameter from 2‐dimensional transesophageal echocardiography (TEE). We then explored the correlation of orifice diameter and other clinical features, and further evaluated the predictive value of orifice diameter for periprocedural complications and follow‐up prognosis including thromboembolic events and development of AHF.
METHODS
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Patient Enrollment
From May 2017 to December 2019, patients with atrial fibrillation (AF) who underwent LAAC were consecutively enrolled in the study. Patients who had undergone pulmonary vein isolation combined with LAAC were from the registered CLACBAC (Combining Left Atrial Appendage Closure With Cryoballoon Ablation in the Chinese Population; number NCT04185142) study. Inclusion criteria were (1) patients diagnosed with AF and (2) patients having undergone a percutaneous LAAC procedure. Exclusion criteria were (1) patients with history of valvular heart disease, (2) patients with concomitant arrhythmia except AF who required extrapulmonary vein ablation, (3) patients who underwent other concomitant procedures except pulmonary vein isolation, (4) patients with failed occluder implantation because of any reason, (5) patients whose echocardiographic data could not be retraced, and (6) patients enrolled in other clinical trials. Detailed inclusion and exclusion criteria are described in Data S1. 9 , 10 , 11 , 12 A study flowchart is displayed in Figure 1. This study complies with the Declaration of Helsinki, and was approved by the local ethic committee of Shanghai Tenth People's Hospital (approval number SHSY‐IEC‐4.1/21/231/01). Written informed consent was given by every patient included.
Figure 1. The study flowchart.
AF indicates atrial fibrillation; CTA, computed tomography angiography; LAA, left atrial appendage; LAAC, left atrial appendage closure; PSM, propensity‐score matching; and TEE, transesophageal echocardiography.
Data Collection
For TEE measurement, 2‐dimensional TEE (GE Vivid E9 or Philips EPIQ 7C) was performed according to guidelines. 11 Under at least 3 different degrees, namely 45°, 90°, and 135°, LAA orifice diameter was measured from the circumflex artery to a point 1 to 2 cm superior within the LA ridge. Concurrently, LAA depth was measured as the length from the middle of the LAA orifice perpendicular to the inner border of the LAA. The maximum depth was acquired if LAA was lobulated. Detailed information of other data collected is described in Data S1.
Percutaneous LAAC
All patients included underwent standard cryoballoon ablation for only pulmonary vein isolation (PVI). Details of cryoballoon ablation was as previously published 13 and described in Data S1. 14 LAAC was performed under local anesthesia and as previously described. 13 Both plug occluders (WATCHMAN; Boston Scientific) and pacifier occluders (Lifetech Scientific, Shenzhen, China) were used. Under guidance of fluoroscopy and TEE, the occluder was delivered through a 14 F delivery sheath. For the plug occluder, the size was 4 to 6 mm larger than the measured diameter of the landing zone to ensure an optimal compression ratio from 10% to 20%. For the pacifier occluder, the size of the outer plate should be 2 to 3 mm larger than the measured orifice diameter to ensure complete sealing. After angiography confirmed complete sealing and tug tests confirmed stability, the occluder was deployed, and angiography and TEE were checked again to evaluate occluder positioning and possible periprocedural complications.
After LAAC, 3‐month oral anticoagulation therapy (dabigatrian, rivaroxaban, or warfarin) was prescribed. Once follow‐up TEE or computed tomography angiography in the third month showed satisfactory sealing (residual flow <3 mm), an antithrombotic regimen was recommended with 3‐month double antiplatelet therapy and ensuing 6‐month single antiplatelet therapy. For patients who underwent ablation, class I/III antiarrhythmic drugs were prescribed for 3 months after the procedure.
Follow‐Up Schedule
Patients were required to have outpatient follow‐up in the third month, sixth month, and every year after the procedure. Additionally, TEE examination in the third month from the procedure was required. Composite primary end points included thromboembolic events and major hemorrhagic events. Thromboembolic events comprised cerebral infarction, transient ischemic attack, and perivascular thrombosis confirmed via imaging examination. Secondary end points were (1) major adverse cardiocerebrovascular events (MACCE) including ischemic or hemorrhagic stroke, AHF with New York Heart Association (NYHA) functional class grading III to IV, acute myocardial infarction, malignant arrhythmia, cardiac tamponade, and all‐cause mortality; (2) rehospitalization because of cardiovascular diseases; (3) rehospitalization because of AHF; (4) rehospitalization because of atrial tachycardia; and (5) other cardiovascular events including bradycardia requiring pacemaker implantation and AF recurrence for patients who underwent concomitant PVI.
AHF was confirmed when at least 2 of the following conditions were met 15 : (1) severe chest distress with dyspnea and/or edema of bilateral lower limbs excluding acute onset of pulmonary or renal diseases, (2) level of NT‐proBNP (N‐terminal pro‐B‐type natriuretic peptide) at admission >300 pg/mL at sinus rhythm, level of NT‐proBNP >800 pg/mL at AF rhythm, and (3) echocardiography showing left ventricular ejection fraction (LVEF) <40% and structural change (left atrial volume index [LAVI] ≥34 mL/m2 and/or left ventricular mass index ≥115 g/m2 for men or ≥95 g/m2 for women). NYHA functional class grading was defined by severity of symptoms according to guideline. 15 Specifically, grade IV was defined as when patients were unable to carry on physical activity because of discomfort and discomfort persisted at rest, and grade III was defined as markedly limited physical activity because of discomfort. Detailed definitions of other follow‐up events are described in Data S1. 16
Statistical Analysis
Patients were divided into 4 groups based on quartile (Q) of LAA orifice diameter. Continuous variables were described as mean±standard deviation if they conformed to normal distribution, or median with interquartile range if they had a skewed distribution. Categorical variables were described as numbers with percentages. The comparison of baseline characteristics was done by ANOVA and sign‐rank test for continuous variables or Cochran‐Mantel‐Haenszel test for categorical variables. A P value for trend was generated by linear, logistic, or Cox regression with quartiles of LAA orifice diameter. For continuous variables that did not conform normal distribution even after transformation, a Wilcoxon‐type test for trend was applied, and the comparison of baseline and follow‐up information was done by repeated‐measures ANOVA. The association between 2 variables was analyzed by Spearman correlation analysis.
A receiver operator characteristics analysis of LAA orifice diameter for MACCE was performed, and a cutoff value was determined by the largest Yoden index. Based on cutoff value, patients were divided into groups of large LAA and small LAA for further analysis. For survival data, a Kaplan‐Meier estimate was used with a log‐rank test to detect the difference between patients with different LAA orifice diameters. A Cox proportional hazard model was applied to further evaluate the prognostic value of LAA orifice diameter. Time‐dependent Cox model, log(−log[survival]) plot and Schoenfeld residual plot were used to validate the proportional hazards assumption. If any variable could not satisfy the assumption, a time‐dependent Cox regression analysis was performed. The multivariable models for overall population and for subgroup analysis were adjusted for age and sex.
To reduce bias of baseline characteristics between large‐LAA and small‐LAA groups, 1:1 greedy propensity‐score matching was performed. Significantly different baseline variables including age, sex, body mass index, AF type, occluder type, concomitant PVI, previous history of HF, LAVI, left ventricular mass index, ratio of mitral inflow velocity and average mitral valve velocity, and LVEF were incorporated in a multivariable logistic regression model to generate propensity scores and matched by greedy algorithm.
A 2‐sided P<0.05 was considered statistically significant. SAS 9.4 software (SAS Institute, Cary, NC) was used to conduct the analysis.
RESULTS
To reveal the correlation of LAA orifice diameter with other clinical parameters, 302 patients were included. Based on quartiles of LAA orifice diameter, the cohort was divided into 4 groups, Q1 (≤18.0 mm), Q2 (18.0–20.7 mm), Q3 (20.7–23.3 mm), and Q4 (>23.3 mm). The average ages were similar, whereas Q4 had a nonsignificant higher proportion of men (66.7%, P=0.092). Patients in Q4 had a worse condition, including a higher proportion of persistent AF (77.3%, P=0.076), greater extent of diastolic dysfunction (ratio of mitral inflow velocity and average mitral valve velocity=15.4 [11.2–17.5], P=0.046), and lower LVEF (56.4% [52.9%–62.5%], P=0.013). A higher prevalence of spontaneous echo contrast was observed in Q4 (34.7%, P=0.004). Detailed baseline information is listed in Table 1. Correlation analysis showed that LAA orifice diameter had a weak positive correlation with LAVI (r=0.157, P=0.017), left ventricular mass index (r=0.136, P=0.040), ratio of mitral inflow velocity and average mitral valve velocity (r=0.135, P=0.019), and a weak negative correlation with LVEF (r=−0.130, P=0.024) (Figure 2).
Table 1.
Baseline Characteristics of Patients Undergoing LAA Closure Divided by Mean LAA Orifice Diameter
| Characteristic | Overall, n=302 | Q1 OD≤18.0, n=78 | Q2 18.0<OD≤20.7, n=83 | Q3 20.7<OD≤23.3, n=66 | Q4 OD>23.3, n=75 | P for trend |
|---|---|---|---|---|---|---|
| Age, y | 71.2±8.7 | 70.3±8.7 | 72.3±8.7 | 70.4±9.1 | 71.2±8.7 | 0.640 |
| Men, n (%) | 181 (59.9) | 38 (48.7) | 54 (65.1) | 39 (59.1) | 50 (66.7) | 0.057 |
| Body mass index, kg/m2 | 25.6±3.5 | 26.4±3.7 | 25.6±3.3 | 24.7±3.5 | 25.7±3.4 | 0.144 |
| AF type, n (%) | 0.095 | |||||
| Paroxysmal AF | 103 (34.1) | 26 (33.3) | 27 (32.5) | 20 (30.3) | 30 (40.0) | |
| Short‐term persistent AF | 69 (22.9) | 21 (26.9) | 19 (22.9) | 9 (13.6) | 20 (26.7) | |
| Long‐term persistent AF | 130 (43.1) | 31 (39.7) | 37 (44.6) | 37 (56.1) | 25 (33.3) | |
| CHA2DS2‐VASc | 4.0 [3.0–5.0] | 4.0 [3.0–5.0] | 4.0 [3.0–6.0] | 3.0 [2.0–5.0] | 4.0 [3.0–5.0] | 0.796 |
| HAS‐BLED | 2.0 [2.0–3.0] | 3.0 [2.0–3.0] | 3.0 [2.0–3.0] | 2.0 [1.3–3.0] | 2.0 [2.0–3.0] | 0.437 |
| NT‐proBNP, pg/mL | 800.1 [360.3–1419.0] | 602.6 [229.6–1389.0] | 779.4 [466.1–1563.8] | 702.1 [376.3–1222.0] | 849.6 [444.6–1686.5] | 0.025* |
| NYHA functional class, n (%) | 0.525 | |||||
| IV | 13 (4.3) | 4 (5.1) | 3 (3.6) | 4 (6.1) | 2 (2.7) | |
| III | 17 (5.6) | 4 (5.1) | 5 (6.0) | 4 (6.1) | 4 (5.3) | |
| II | 50 (16.6) | 10 (12.8) | 14 (16.9) | 14 (21.2) | 12 (16.0) | |
| I | 21 (7.0) | 4 (5.1) | 6 (7.2) | 4 (6.1) | 7 (9.3) | |
| Previous illness, n (%) | ||||||
| Stroke | 101 (33.4) | 20 (25.6) | 35 (42.2) | 22 (33.3) | 24 (32.0) | 0.676 |
| Cerebral infarction | 98 (32.5) | 19 (24.4) | 35 (42.2) | 21 (31.8) | 23 (30.7) | 0.725 |
| Intracranial hemorrhage | 10 (3.3) | 2 (2.6) | 5 (6.0) | 1 (1.5) | 2 (2.7) | 0.654 |
| Heart failure | 60 (19.9) | 12 (15.4) | 12 (14.5) | 15 (22.7) | 21 (28.0) | 0.557 |
| Hypertension | 222 (73.5) | 62 (79.5) | 60 (72.3) | 46 (69.7) | 54 (72.0) | 0.274 |
| Diabetes | 78 (25.8) | 19 (24.4) | 18 (21.7) | 15 (22.7) | 26 (34.7) | 0.074 |
| Perivascular disease | 17 (5.6) | 4 (5.1) | 5 (6.0) | 3 (4.5) | 5 (6.7) | 0.784 |
| Coronary heart disease | 93 (30.8) | 19 (24.4) | 30 (36.1) | 19 (28.8) | 25 (33.3) | 0.405 |
| Echocardiographic measurement | ||||||
| LAVI, mL/m2 | 25.3 [20.3–31.4] | 24.2 [17.4–29.7] | 25.0 [19.6–29.5] | 24.4 [20.1–31.4] | 25.3 [20.3–31.4] | 0.002* |
| LVEF, % | 58.0 [55.0–63.9] | 59.1 [55.0–64.0] | 58.0 [55.0–63.0] | 62.0 [57.8–64.5] | 56.4 [52.9–62.5] | 0.013* |
| LVEF group, n (%) | 0.052 | |||||
| >50% | 269 (89.1) | 72 (92.3) | 76 (91.6) | 59 (89.4) | 62 (82.7) | |
| 40%–50% | 13 (4.3) | 2 (2.6) | 4 (4.8) | 2 (3.0) | 5 (6.7) | |
| ≤40% | 20 (6.6) | 4 (5.1) | 3 (3.6) | 5 (7.6) | 8 (10.7) | |
| LVMI, g/m2 | 96.4 [85.1–108.8] | 94.7 [86.2–107.5] | 93.9 [81.1–101.5] | 98.5 [89.6–109.2] | 96.4 [85.1–108.8] | 0.122 |
| LV hypertrophy, n (%) | 47 (15.6) | 14 (17.9) | 8 (9.6) | 9 (13.6) | 16 (21.3) | 0.436 |
| E/e′ | 13.2 [10.4–17.1] | 11.3 [9.2–15.1] | 12.5 [10.8–18.5] | 13.2 [9.7–18.8] | 15.4 [11.2–17.5] | 0.046 * |
| E/e′group, n (%) | 0.016* | |||||
| ≥14 | 110 (36.4) | 23 (29.5) | 26 (31.3) | 22 (33.3) | 39 (52.0) | |
| 7–14 | 172 (57.0) | 50 (64.1) | 53 (63.9) | 38 (57.7) | 31 (41.3) | |
| <7 | 20 (6.6) | 5 (6.4) | 4 (4.8) | 6 (9.0) | 5 (6.7) | |
| LAA measurement | ||||||
| Mean LAA orifice diameter, mm | 20.7 [18.0–23.3] | 16.7 [15.7–17.3] | 20.0 [19.3–20.3] | 22.0 [21.7–22.7] | 26.0 [24.7–27.8] | <0.001* |
| Mean LAA depth, mm | 21.0 [18.7–23.7] | 18.3 [16.0–20.0] | 20.7 [19.2–22.0] | 22.8 [20.2–24.3] | 24.0 [21.3–26.3] | <0.001* |
| Spontaneous echo contrast, n (%) | 68 (22.5) | 8 (10.3) | 20 (24.1) | 14 (21.2) | 26 (34.7) | 0.001* |
HAS‐BLED a score to evaluate major bleeding risk for atrial fibrillation population incoporating hypertension, abnormal liver/kidney function, stroke history, bleeding history, labile international normalized ratio, elder age, and drug predisposing bleeding/alcohol abuse.
Continuous variables are presented as mean±SD or as median with interquartile range; categorical variables are presented as percentage. AF indicates atrial fibrillation; E/e′, ratio of mitral inflow velocity (E) and average mitral valve velocity (e′); LA, left atrium; LAA, left atrial appendage; LAVI, left atrial volume index; LV, left ventricle; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; NYHA, New York Heart Association; NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide; OD, orifice diameter; and Q, quartile. 1
Significant P value.
Figure 2. Correlation of left atrial appendage (LAA) orifice diameter (OD) with other clinical variables.
A, OD was strongly correlated with LAA depth. B, OD was not associated with baseline NT‐proBNP (N‐terminal pro‐B‐type natriuretic peptide) level. C through F, Displayed correlation with echocardiographic measurements, showing that OD was weakly correlated with left atrial volume index (LAVI), left ventricular mass index (LVMI), ratio of mitral inflow velocity (E) and average mitral valve velocity (e′) (E/e′), and negatively correlated with left ventricular ejection fraction (LVEF).
With regard to LAAC procedural information, a pacifier occluder was preferable for closing a large LAA (42.7%, P=0.001). Meanwhile, the incidence of periprocedural complications and postprocedural residual flow were comparable among the 4 quartile groups. Among 220 (72.8%) patients undergoing TEE examination at the third month, no difference was found in residual flow, device‐related thrombosis, or other complications. Detailed information is listed in Table 2. AF ablation parameters were similar among the groups, as displayed in Table S1.
Table 2.
LAAC Procedure Detail of Patients With Different LAA Orifice Diameter
| Detail | Overall, n=302 | Q1 OD≤18.0, n=78 | Q2 18.0<OD≤20.7, n=83 | Q3 20.7<OD≤23.3, n=66 | Q4 OD>23.3, n=75 | P for trend |
|---|---|---|---|---|---|---|
| LAAC procedure time, min | 47.8±16.2 | 48.1±15.4 | 46.9±17.3 | 49.2±14.6 | 47.4±13.8 | 0.476 |
| Concomitant PVI, n (%) | 177 (58.6) | 46 (59.0) | 51 (61.4) | 42 (63.6) | 38 (50.7) | 0.356 |
| Plug occluder, n (%) | 225 (74.5) | 65 (83.3) | 63 (75.9) | 54 (81.8) | 43 (57.3) | 0.001* |
| Disc size (pacifier), mm | 34.0 [30.0–36.0] | 30.0 [26.0–32.0] | 32.0 [30.0–34.0] | 34.0 [31.0–36.0] | 36.0 [34.0–38.0] | <0.001* |
| Compression ratio (plug), % | 20.9 [17.0–25.0] | 20.0 [17.0–25.0] | 21.0 [17.5–26.0] | 20.0 [17.0–23.2] | 20.9 [17.0–25.0] | 0.588 |
| Recapture, n (%) | 58 (19.2) | 7 (9.0) | 20 (24.0) | 14 (21.2) | 17 (22.7) | 0.061 |
| Resize, n (%) | 27 (8.9) | 7 (9.0) | 7 (8.4) | 7 (10.6) | 6 (8.0) | 0.080 |
| Postprocedural residual flow, n (%) | 14 (4.6) | 2 (2.6) | 4 (4.8) | 4 (6.1) | 4 (4.6) | 0.383 |
| Remnant LAA, n (%) | 2 (0.7) | 1 (1.3) | 0 | 1 (1.5) | 0 | 0.572 |
| Protruded occluder, n (%) | 3 (1.0) | 1 (1.3) | 0 | 2 (3.0) | 0 | 0.848 |
| Periprocedural complication, n (%) | 11 (3.6) | 3 (3.8) | 2 (2.4) | 5 (7.6) | 1 (1.3) | 0.779 |
| Pericardial effusion | 5 (1.7) | 2 (2.6) | 0 | 3 (4.5) | 0 | 0.608 |
| Cardiac tamponade | 1 (0.3) | 1 (1.3) | 0 | 0 | 0 | 0.594 |
| Occluder detached from LAA | 1 (0.3) | 0 | 0 | 1 (1.5) | 0 | 0.689 |
| Circumflex artery compressed | 1 (0.3) | 0 | 0 | 1 (1.5) | 0 | 0.689 |
| TEE follow‐up at the third month, patients (%) | 220 (72.8) | 60 (76.9) | 58 (69.9) | 47 (71.2) | 55 (73.3) | 0.684 |
| Residual flow, n (%) | 14 (6.4) | 4 (6.7) | 3 (5.2) | 3 (6.4) | 4 (7.3) | 0.841 |
| ≥3 mm, n (%) | 6 (2.7) | 1 (1.7) | 1 (1.7) | 1 (2.1) | 3 (5.5) | 0.290 |
| <3 mm, n (%) | 8 (3.6) | 3 (5.0) | 2 (3.4) | 2 (4.2) | 1 (1.8) | 0.428 |
| Device related thrombosis, n (%) | 2 (0.9) | 0 | 1 (1.7) | 0 | 1 (1.8) | 0.498 |
LAA indicates left atrial appendage; LAAC, left atrial appendage closure; OD, orifice diameter; PVI, pulmonary vein isolation; Q, quartile; and TEE, transesophageal echocardiography.
Significant P value.
Postprocedural medication and management are displayed in Table S2. The antithromboembotic regimen was comparable among the 4 groups. For AF rhythm and rate management, 202 (66.9%) patients had either PVI or antiarrhythmic drugs for rhythm control, 109 (36.1%) had rate control, and 38 (12.6%) patients had neither rhythm nor rate control. Of note, there was a nonsignificant higher proportion of antiarrhythmic drugs and correspondingly lower proportion of ablation for rhythm control in Q4.
Over the course of 39.6±8.4 months, 20 (6.6%) patients were lost to follow‐up. Incidence of primary follow‐up events are listed in Table 3, and detailed follow‐up results are listed in Table 4. For stroke prevention, overall stroke incidence was 1.4% per 100 patient‐years, with the highest incidence in Q4 of 3.4% per 100 patient‐years. There was no significant difference among the 4 groups (Table 3). In terms of AF management outcome, higher AF recurrence in Q4 was observed in the overall population (38.6%, P=0.052) and in patients who underwent PVI (41.7%, P=0.054) (Table 4).
Table 3.
Primary Follow‐Up Events After LAA Closure Procedure
| Event | Events (rate*) | LAA OD, quartile | LAA OD, cutoff | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Overall, n=282 | Q1, n=72 | Q2, n=79 | Q3, n=61 | Q4, n=70 | HR (95% CI) | P value | HR (95% CI) | P value | |
| MACCE | 45/850.1 (5.3%) | 10/226.5 (4.4%) | 9/247.2 (3.6%) | 5/197.8 (2.5%) | 21/178.5 (11.8%) | 1.407 (1.072, 7.847)† | 0.014† , ‡ | 3.580 (1.989, 6.445)† | <0.001† , ‡ |
| AHF, NYHA III–IV | 32/840.6 (3.8%) | 7/226.8 (3.1%) | 6/244.0 (2.5%) | 4/195.1 (2.1%) | 15/174.7 (8.6%) | 2.111 (1.444, 3.088) | <0.001‡ | 8.253 (3.687, 18.476) | <0.001‡ |
| AHF | 53/826 (6.4%) | 14/216.6 (6.5%) | 10/240.1 (4.2%) | 7/194.6 (3.6%) | 22/174.6 (12.6%) | 1.267 (0.992, 1.618) | 0.058 | 3.743 (2.158, 6.493) | <0.001‡ |
| Thromboembolic event | 29/875.5 (3.3%) | 7/231.5 (3.0%) | 6/243.7 (2.5%) | 3/204.7 (1.5%) | 13/195.6 (6.6%) | 1.289 (0.923, 1.801) | 0.136 | 3.765 (1.796, 7.894) | <0.001‡ |
| Stroke | 12/840.7 (1.4%) | 4/224.2 (1.8%) | 2/244.0 (0.8%) | 0/195.1 (0%) | 6/177.5 (3.4%) | 0.985 (0.642, 1.511) | 0.946 | 7.995 (2.332, 27.413) | <0.001‡ |
| Ischemic | 9/899.5 (1.0%) | 4/237.1 (1.7%) | 1/249.5 (0.4%) | 0/209 (0%) | 4/204 (2.0%) | 0.929 (0.501, 1.721) | 0.815 | 2.962 (0.791, 11.093)† | 0.107† |
| Hemorrhagic | 4/899.5 (0.4%) | 1/237.1 (0.4%) | 1/249.5 (0.4%) | 0/209 (0%) | 2/204 (1.0%) | 1.276 (0.525, 3.102) | 0.591 | 3.904 (0.494, 30.830) | 0.197 |
| Major hemorrhage | 5/919.1 (0.5%) | 1/245.0 (0.4%) | 1/257.5 (0.4%) | 0/208.4 (0%) | 3/208.3 (1.4%) | 1.134 (0.448, 2.871) | 0.791 | 4.885 (0.717, 32.299) | 0.105 |
AHF indicates acute heart failure; HR, hazard ratio; LAA, left atrial appendage; MACCE, major adverse cardiocerebrovascular events; NYHA, New York Heart Association; OD, orifice diameter; and Q, quartile.
Incidence rate was calculated as event per 100 patient‐years.
HR (95% CI) and P value were generated by time‐fixed Cox proportional hazard model because they satisfy the proportional hazard assumption. HR (95% CI) and P value of other variables were calculated by time‐dependent Cox proportional hazard model.
Significant P value.
Table 4.
Detailed Follow‐Up Outcome After Left Atrial Appendage Closure
| Outcome | Overall, n=282 | Q1 OD ≤18.0, n=72 | Q2 18.0<OD≤20.7, n=79 | Q3 20.7<OD≤23.3, n=61 | Q4 OD>23.3, n=70 | P for trend |
|---|---|---|---|---|---|---|
| Thromboembolic event | 29 (10.3) | 7 (9.7) | 6 (7.6) | 7 (11.5) | 9 (12.9) | 0.073 |
| Ischemic/hemorrhagic stroke/transient ischemic attack | 19 (6.7) | 4 (5.6) | 5 (6.3) | 4 (6.6) | 6 (8.6) | 0.850 |
| Pulmonary embolism | 1 (0.4) | 0 | 0 | 1 (1.6) | 0 | 1.000 |
| Perivascular thrombosis | 9 (3.2) | 3 (4.2) | 1 (1.3) | 3 (4.9) | 2 (2.9) | 0.278 |
| Major hemorrhagic event | 5 (1.8) | 1 (1.4) | 1 (1.3) | 0 | 3 (4.3) | 0.290 |
| MACCE | 45 (16.0) | 10 (13.9) | 9 (11.4) | 5 (8.2) | 21 (30.0) | 0.014* |
| All‐cause mortality | 4 (1.4) | 0 | 3 (3.8) | 0 | 1 (1.4) | 0.886 |
| Ischemic/hemorrhagic stroke | 12 (4.3) | 4 (5.6) | 2 (2.5) | 0 | 6 (8.6) | 0.423 |
| AHF, NYHA III–IV | 32 (11.3) | 7 (9.7) | 6 (7.6) | 4 (6.6) | 15 (21.4) | 0.027* |
| Cardiac tamponade | 2 (0.7) | 0 | 1 (1.3) | 1 (1.6) | 0 | 0.242 |
| Malignant arrhythmia | 0 | 0 | 0 | 0 | 0 | 1.000 |
| Acute myocardial infarction | 1 (0.4) | 1 (1.4) | 0 | 0 | 0 | 1.000 |
| Rehospitalization | ||||||
| CVD admission | 98 (34.8) | 24 (33.3) | 24 (30.4) | 16 (26.2) | 34 (48.6) | 0.005* |
| AHF admission | 53 (18.8) | 14 (19.4) | 10 (12.7) | 7 (11.5) | 22 (31.4) | 0.070 |
| AF/atrial flutter admission | 53 (18.8) | 12 (16.7) | 9 (11.4) | 10 (16.4) | 22 (31.4) | <0.001* |
| Rhythm/rate control outcome | ||||||
| AF recurrence | 76 (27.0) | 17 (23.6) | 18 (22.8) | 14 (23.0) | 27 (38.6) | 0.052 |
| AF recurrence after ablation | 42/166 (25.3) | 9/43 (20.9) | 10/48 (20.8) | 8/39 (20.5) | 15/36 (41.7) | 0.054 |
| AF recurrence after AADs | 4/23 (17.4) | 1/6 (16.7) | 1/4 (25.0) | 1/4 (25.0) | 1/9 (11.1) | 0.727 |
| Uncontrolled rate | 2/101 (2.0) | 0 | 0 | 1 (5.3) | 1 (3.2) | 0.242 |
AADs indicates antiarrhythmic drugs; AF, atrial fibrillation; AHF, acute heart failure; CVD, cardiovascular disease; MACCE, major adverse cardiocerebrovascular events; NYHA, New York Heart Association; OD, orifice diameter; and Q, quartile.
Significant P value.
When comparing the survival of MACCE (Figure 3), Q4 presented the worst survival rate among the 4 quartiles (log‐rank P overall<0.001), contributed to by a significant higher incidence of AHF with NYHA III to IV (30.0%, log‐rank P overall=0.009). In addition, survival of overall AHF, cardiovascular disease, and atrial tachycardia rehospitalization were also significantly lower in Q4, and AF recurrence was higher in Q4 (Figure 4). Using receiver operating characteristic analysis for MACCE (Figure S1), an LAA orifice diameter cutoff value of 23.6 mm was chosen (sensitivity of 46.7% and specificity of 80.2%), and patients with a large LAA (orifice diameter ≥23.6 mm) had significantly worse survival rate of MACCE, AHF with NYHA III to IV, and overall AHF events (Figure 5).
Figure 3. Survival of thromboembolic events, MACCE, AHF, and AHF with NYHA III–IV among 4 quartile groups divided by LAA orifice diameter.
A, Survival curve for thromboembolic events; no significant difference was observed among groups. B, Survival curve for MACCE showed survival rate in quartile (Q) 4 was significantly lower than other groups. C and D, Survival curves for AHF showed survival rates of overall AHF and AHF with NYHA III to IV were significantly lower in Q4. AHF indicates acute heart failure; LAA, left atrial appendage; MACCE, major adverse cerebrocardiovascular events; and NYHA, New York Heart Association. *significant P value.
Figure 4. Survival of rehospitalization because of CVD (A), atrial arrhythmia (B), and AF recurrence for patients who underwent concomitant AF ablation (C).
Shadowed area in (C) indicates 3‐month blank period. AF indicates atrial fibrillation; and CVD, cardiovascular diseases. *significant P value.
Figure 5. Survival curves of patients grouped by cutoff value of LAA orifice diameter before (left) and after PSM (right).
A and B, After PSM‐adjusted baseline characteristics, survival rates of MACCE were still significantly lower in the large‐LAA group. C and D, survival rates of overall AHF became nonsignificant after PSM. E and F, Survival rates of AHF with NYHA III to IV remained significantly lower in the large‐LAA group. AHF indicates acute heart failure; LAA, left atrial appendage; MACCE, major adverse cardiocerebrovascular events; NYHA, New York Heart Association; and PSM, propensity‐score matching. *significant P value.
To investigate the contribution of potential confounding factors to post‐LAAC AHF, we first compared survival curves of AHF NHA III to IV for patient groups divided by those factors, indicating that a previous history of HF, diabetes, occlude type, and rhythm/rate control outcome contributed significantly to AHF NHA III to IV occurrence. Results are shown in Figure S2. An age‐ and sex‐adjusted multivariable Cox model for MACCE showed that age, occulder type, previous history of diabetes, LAVI, LAA depth, and orifice diameter were predictors for post‐LAAC AHF. Notably, LAA orifice diameter showed a significant hazard ratio of 3.749 (95% CI, 2.074–6.779) (Figure 6). To confirm large LAA orifice diameter per se rather than other confounding factors that could result in higher incidence of post‐LAAC AHF, we performed subgroup analysis of LAA orifice cutoff groups, as shown in Figure 7. The results indicated that a large LAA orifice could still predict MACCE after LAAC in subgroups divided by previous history of HF and diabetes, occluder type, and procedure type (concomitant PVI), except in patients aged <65 years, patients with paroxysmal AF, and patients with failed rhythm/rate control outcome. No significant interaction was found in LAA orifice diameter with other confounding variables for MACCE occurrence.
Figure 6. Forest plot of age‐ and sex‐adjusted hazard ratio for MACCE by multivariable Cox proportional hazard model.
Red bar indicates significant hazard ratio >1. AF indicates atrial fibrillation; E/e′, ratio of mitral inflow velocity (E) and average mitral valve velocity (e′); HR, hazard ratio; LAA, left atrial appendage; LAVI, left atrial volume index; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; MACCE, major adverse cardiocerebrovascular events; proBNP, pro‐B‐type natriuretic peptide; and PVI, pulmonary vein isolation. *significant P value.
Figure 7. Subgroup analysis of LAA orifice diameter cutoff groups for MACCE by multivariable Cox proportional hazard model.
Hazard ratio of LAA orifice cutoff groups was adjusted for age and sex. Red bar indicates significant hazard ratio >1. AF indicates atrial fibrillation; DM, diabetes mellitus; HF, heart failure; HR, hazard ratio; LAA, left atrial appendage; LAAC, left atrial appendage closure; LAVI, left atrial volume index; MACCE, major adverse cardiocerebrovascular events; and PVI, pulmonary vein isolation.
Moreover, we matched patients with a small LAA to patients with a large LAA using propensity‐score matching that rendered baseline and procedure information comparable (Table 5). Further survival analysis showed that the incidence of AHF with NYHA III to IV of the large‐LAA group remained significantly higher than the matched small‐LAA group (Figure 5).
Table 5.
The Comparison Between Groups With Large and Small LAA Before and After Propensity‐Score Matching
| Large LAA, OD ≥23.6 mm, n=69 | Small LAA, OD <23.6 mm, n=233 | P value | Small LAAmatched, OD<23.6 mm, n=69 | P value | |
|---|---|---|---|---|---|
| Age, y | 71.8±8.3 | 71±8.8 | 0.516 | 70.9±8.9 | 0.540 |
| Men, n (%) | 47 (68.1) | 134 (54.9) | 0.114 | 48 (69.6) | 0.854 |
| Body mass index, kg/m2 | 25.7±3.2 | 25.6±3.6 | 0.843 | 25.6±3.0 | 0.918 |
| Persistent AF, n (%) | 55 (79.7) | 155 (63.5) | 0.006* | 58 (84.1) | 0.507 |
| CHA2DS2‐VASc | 4.0 [3.0, 5.0] | 4.0 [3.0, 5.0] | 0.425 | 4.0 [3.0, 5.0] | 0.462 |
| HAS‐BLED | 2.0 [2.0, 3.0] | 2.0 [2.0, 3.0] | 0.959 | 2.0 [2.0, 3.0] | 0.873 |
| NT‐proBNP, pg/mL | 935.1 [462.5–1769.0] | 722.3 [332.6–1389.3] | 0.020* | 1009.5 [457.4–1570.3] | 0.880 |
| Previous illness, n (%) | |||||
| Stroke | 22 (31.9) | 79 (32.4) | 0.755 | 25 (36.2) | 0.590 |
| Cerebral infarction | 21 (30.4) | 77 (31.6) | 0.684 | 23 (33.3) | 0.715 |
| Intracranial hemorrhage | 1 (1.4) | 9 (3.7) | 0.325 | 2 (2.9) | 0.556 |
| Heart failure | 19 (19.9) | 22 (27.5) | 0.069 | 41 (16.8) | 0.576 |
| Hypertension | 51 (73.5) | 171 (73.9) | 0.931 | 56 (70.1) | 0.308 |
| Diabetes | 28 (25.8) | 50 (40.6) | 0.001* | 28 (20.5) | 1.000 |
| Perivascular disease | 4 (5.6) | 13 (5.8) | 0.945 | 3 (5.3) | 0.698 |
| Coronary heart disease | 23 (30.8) | 70 (33.3) | 0.603 | 27 (28.7) | 0.479 |
| Echocardiographic measurement | |||||
| LAVI, mL/m2 | 27.0 [22.9–33.9] | 24.5 [19.0–29.8] | 0.007* | 27.1 [24.4–32.7] | 0.645 |
| LVEF, % | 56.0 [51.8–62.5] | 59.2 [55.0–63.8] | 0.003* | 58.0 [52.5–62.4] | 0.432 |
| LVEF group, n (%) | 0.056 | 0.759 | |||
| >50% | 56 (81.2) | 213 (91.4) | 57 (82.6) | ||
| 40%–50% | 5 (7.2) | 8 (3.4) | 3 (4.3) | ||
| ≤40% | 8 (11.6) | 12 (5.2) | 9 (13.0) | ||
| LVMI, g/m2 | 101.7 [88.6–119.0] | 95.0 [84.7–105.5] | 0.021* | 100.7 [90.7–111.0] | 0.608 |
| LV hypertrophy, n (%) | 16 (23.2) | 31 (13.3) | 0.047* | 13 (18.8) | 0.531 |
| E/e′ | 15.0 [11.0–17.0] | 12.0 [10.0–17.0] | 0.008* | 13.0 [10.0–18.0] | 0.287 |
| E/e′ group, n (%) | 0.003* | 0.055 | |||
| ≥14 | 37 (53.6) | 73 (31.3) | 25 (36.2) | ||
| 7–14, n (%) | 28 (40.6) | 144 (61.8) | 42 (60.9) | ||
| <7, n (%) | 4 (5.8) | 16 (6.9) | 2 (2.9) | ||
| LAA measurement | |||||
| Mean LAA orifice diameter, mm | 26.3 [25.0–28.0] | 20.0 [17.3–21.5] | <0.001* | 20.0 [17.3–21.3] | <0.001* |
| Mean LAA depth, mm | 24.0 [21.3–26.7] | 20.3 [18.0–22.7] | <0.001* | 21.0 [19.0–23.0] | <0.001* |
| Spontaneous echo contrast, n (%) | 25 (36.2) | 43 (18.5) | 0.002* | 22 (31.9) | 0.590 |
| Treatment, n (%) | |||||
| Rhythm/rate control | 61 (88.4) | 203 (87.1) | 0.778 | 57 (82.6) | 0.333 |
| Rhythm control | 42 (60.9) | 160 (68.7) | 0.227 | 39 (56.5) | 0.604 |
| Concomitant PVI | 33 (47.8) | 144 (59.0) | 0.038* | 33 (47.8) | 1.000 |
| I/III AADs | 9 (13.0) | 16 (6.9) | 0.102 | 6 (8.7) | 0.412 |
| Rate control | 34 (49.3) | 75 (32.2) | 0.009* | 29 (42.0) | 0.393 |
| Plug occluder | 38 (55.1) | 187 (80.3) | <0.001* | 42 (60.9) | 0.490 |
| Rhythm/rate control outcome | |||||
| AF recurrence | 25/64 (39.1) | 51/218 (23.4) | 0.013* | 19/63 (30.2) | 0.292 |
| AF recurrence after ablation | 13/31 (41.9) | 29/135 (21.5) | 0.018* | 10/30 (33.3) | 0.488 |
| AF recurrence after AADs | 1/8 (34.8) | 3/15 (20.0) | 1.000 | 1/5 (20.0) | 1.000 |
| Uncontrolled rate | 1/31 (3.2) | 1/70 (1.4) | 1.000 | 1/27 (3.7) | 1.000 |
HAS‐BLED a score to evaluate major bleeding risk for atrial fibrillation population incoporating hypertension, abnormal liver/kidney function, stroke history, bleeding history, labile international normalized ratio, elder age, and drug predisposing bleeding/alcohol abuse.
AADs indicates antiarrhythmic drugs; AF, atrial fibrillation; E/e′, ratio of mitral inflow velocity (E) and average mitral valve velocity (e′); LAA, left atrial appendage; LAVI, left atrial volume index; LV, left ventricle; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide; OD, orifice diameter; and PVI, pulmonary vein isolation. 2
Significant P value.
In addition, comparing the change of NT‐proBNP level and echocardiographic indexes during follow‐up between matched groups, only the level of NT‐proBNP at the third month of the large‐LAA group was significantly higher than baseline level, whereas the change of LAVI and LVEF were nonsignificant (Figure 8).
Figure 8. Change of NT‐proBNP and echocardiographic measurements of matched groups during follow‐up.
A, Level of NT‐proBNP was increased at the third month and fell back to baseline level, whereas only the increased level of NT‐proBNP of the large‐LAA group at the third month was statistically significant. Neither the change of LAVI (B) nor LVEF (C) showed significance at the third month and 1 year. Asterisk (*) indicates P<0.05. FU indicates follow‐up; LAA indicates left atrial appendage; LAVI, left atrial volume index; LVEF, left ventricular ejection fraction; and NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide.
DISCUSSION
In the present study, we explored the correlation of LAA orifice diameter with other clinical variables, and we comparatively evaluated the outcomes of patients with different LAA sizes after LAAC. Our major findings are as follows: (1) Patients with a larger LAA orifice had a worse condition. (2) Patients with a larger LAA orifice presented worse survival for MACCE, resulting from a higher incidence of AHF, although no difference in periprocedural complication and thromboembolic events. (3) A comparison of propensity‐score matched groups showed that a large LAA orifice could be an independent risk factor for post‐LAAC AHF events.
Although successive clinical trials like PREVAIL (Evaluation of the WATCHMAN LAA Closure Device in Patients With Atrial Fibrillation Versus Long Term Warfarin Therapy), 17 EWOLUTION (Registry on WATCHMAN Outcomes in Real‐Life Utilization), 18 and PRAGUE‐17 (Left Atrial Appendage Closure versus Novel Anticoagulation Agents in Atrial Fibrillation) 19 have provided promising evidence toward the safety and efficacy of LAAC for preventing thromboembolic events and reducing the risk of bleeding, studies reported that some patients presented poor prognosis after LAAC. In 2019, Schneider et al reported an onset of acute decompensated HF after LAAC in a 79‐year‐old woman, stressing their concern of developing HF after LAAC. 20 More recently, 3 small‐sized studies reported significant occurrence of post‐LAAC HF ranging from 16% (16/106) to 23.9% (16/67). 6 , 7 , 8 In line with previous findings, we also observed an incidence of post‐LAAC HF of 17.5%, including 10.6% with NYHA class III to IV. It should be noted that HF after LAAC is not uncommon and warrants further investigation.
Currently, the role of LAA dimensions in predicting a post‐LAAC HF event has not yet been assessed. As mentioned, occlusion of the LAA would impair the total reservoir capacity of LA and increase the volume load of the remaining LA chamber, 4 whereas the outcome of such volume reduction depends on both the contribution of LAA as a volume reservoir before occlusion and the adaptive capacity of the remaining LA and left ventricle. On one hand, an LAA with larger volumes receives more blood, and by occluding large a LAA, the heart is rendered more vulnerable to excessive volume load. Although a large LAA could be inborn, an LAA could also become maladaptively enlarged under concurrent cardiac abnormalities such as persistent AF, 21 which might further worsen the outcomes of LAAC. On the other hand, for patients with a larger LAA, we also found worse baseline conditions. Greater LA and left ventricular remodeling would further weaken the adaptive capacity in response to volume overload. As the results show in our study, a large LAA orifice per se could be an independent risk factor for post‐LAAC, AHF, NYHA III to IV event in most patients, providing clinical support that LAAC might impair the LA and cause worsened outcome. However, we found that LAA orifice diameter could not predict MACCE in subgroups of those aged <65 years with paroxysmal AF and failed rhythm/rate control outcome before AHF. Because the development of post‐LAAC AHF is multifactorial, we think such results are reasonable because (1) younger patients and patients with paroxysmal AF had better atrial function, and volume reduction by LAAC might be well compensated; (2) failed rhythm/rate control adds insult to injury that uncontrolled rhythm/rate becomes a dominant contributor for post‐LAAC AHF. Limited to the sample size and design of this study, we were unable to further confirm the contribution of LAA orifice for AHF in those subgroups. Future studies are warranted to investigate the role of LAA dimensions in those subgroups. Collectively, patients with larger LAA dimensions are under a high risk of developing AHF after LAAC, and the decision to perform LAAC should be taken with caution.
Notably, for patients with a large LAA who are eligible for LAAC, their situation presents a dilemma, namely that a large LAA poses a high risk of thromboembolism 22 but also a higher risk of developing HF after LAAC. Previous studies have found that a bulged LAA not only correlates with reduced blood flow velocity, but also promotes platelet activation, 23 eventually resulting in a hypercoagulable state. In the present study, we also observed a higher prevalence of spontaneous echo contrast and a higher incidence of stroke after LAAC in patients with a large LAA. Therefore, patients with a large LAA are under a higher risk for stroke, and more intense stroke prevention management should be implemented. However, for patients with a large LAA, not only the benefit of stroke prevention by LAAC, but the development of post‐LAAC AHF should be considered. The decision of LAAC in patients with high risk of HF (large LAA, prior HF, lower LVEF) should be taken with caution. If LAAC is decided for those patients, more intense monitoring of HF should be as important as stroke prevention for follow‐up management. As the LAA exerts a secretory function, closure of the LAA could alter the expression pattern of brain type natriuretic peptide, 24 and routine measurement of brain type natriuretic peptide might not reflect the patient's real condition. Results from PRAGUE‐17 trial showed no difference in several HF biomarkers including both brain and atrial‐type natriuretic peptide. 25 Further investigation is required into the valuable biomarkers for monitoring following LAAC HF events. Moreover, AF ablation for rhythm control should be carefully considered for patients with a large LAA. Large LAA volume poses a higher risk for recurrence after AF ablation. 26 Similarly, we found that patients with a large LAA had a higher recurrence after concomitant cryoballoon ablation. Although the association between ablation and recurrence rates has not yet been elucidated, these higher recurrence rates should be considered when making the decision.
Limitations
The results of this study should be interpreted with caution. First, the sample size of our study is limited. Hence, the incidence rate of the 4 quartile groups might be biased, and only age and sex were adjusted considering overfitting in the multivariable Cox model. The occurrence of AHF is multifactorial, and ablation strategy, rhythm/rate control outcome, and occlude choice could confound the results. The present study included patients who underwent only PVI, and efficacy might be compromised for patients with long‐term persistent AF. The predictive value of LAA orifice diameter was not significant in patients with failed rhythm/rate control. In addition, a different occluder type is also an important confounding factor, because a pacifier device (LAmbre) could possibly cause more volume reduction compared with a plug device. Though we attempted to minimize confounding by subgroup analysis and propensity‐score matching, our study showed no significant interference with our primary conclusion, and we are unable to verify whether factors other than LAA orifice, different ablation strategy, rhythm/rate control outcome, and occluder type could lead to a worse atrial function and subsequently worse prognosis. Future well‐designed randomized studies are needed to address these important issues.
Second, our results were based on a comparison of 1:1 matched small‐LAA groups to large‐LAA groups with comparable baseline characteristics, whereas future well‐designed trials are needed to provide stronger evidence. In addition, because this studied cohort was not designed to monitor the progression of HF, follow‐up echocardiography and HF biomarker examination were not routinely performed. Therefore, follow‐up change of echocardiographic and NT‐proBNP measurement could be biased. Orifice diameter cannot comprehensively reflect the dimensional feature of LAA, because the LAA varies in shape and depth. Moreover, LAA dimension depends on volume load, because the LAA orifice and depth would enlarge following fluid infusion. 27 Further studies using imaging modalities like computed tomography and magnetic resonance imaging are warranted to precisely investigate the importance of LAA dimensional features. Last, we lack a control group without LAAC, which could further strengthen the conclusion of the study. Future well‐designed controlled studies with comprehensive echocardiographic evaluation are needed to illustrate the impact of LAAC on atrial function and prognosis.
CONCLUSIONS
Patients with a larger LAA orifice had worse conditions, and presented worse prognoses after the LAAC procedure, primarily arising from a high incidence of AHF. A large LAA orifice could be an independent risk factor for developing post‐LAAC AHF in most patients, except for patients aged <65 years, patients with paroxysmal AF, and/or patients with failed rhythm/rate control.
Sources of Funding
This study was supported by Shanghai Hospital Development Center (grant number SHDC2020CR1040B).
Disclosures
None.
Supporting information
Data S1
Tables S1–S2
Figures S1–S2
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.122.026309
For Sources of Funding and Disclosures, see page 17.
Contributor Information
Yawei Xu, Email: xuyawei@tongji.edu.cn.
Dongdong Zhao, Email: zhaodd@tongji.edu.cn.
References
- 1. E Gul E, Boles U, Haseeb S, Flood J, Bansal A, Glover B, Redfearn D, Simpson C, Abdollah H, Baranchuk A, et al. Left atrial appendage characteristics in patients with persistent atrial fibrillation undergoing catheter ablation (LAAPAF Study). J Atr Fibrillation. 2017;9:1526. doi: 10.4022/jafib.1526 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Beigel R, Wunderlich NC, Ho SY, Arsanjani R, Siegel RJ. The left atrial appendage: anatomy, function, and noninvasive evaluation. JACC Cardiovasc Imaging. 2014;7:1251–1265. doi: 10.1016/j.jcmg.2014.08.009 [DOI] [PubMed] [Google Scholar]
- 3. Osmancik P, Herman D, Neuzil P, Hala P, Taborsky M, Kala P, Poloczek M, Stasek J, Haman L, Branny M, et al. 4‐year outcomes after left atrial appendage closure versus nonwarfarin oral anticoagulation for atrial fibrillation. J Am Coll Cardiol. 2022;79:1–14. doi: 10.1016/j.jacc.2021.10.023 [DOI] [PubMed] [Google Scholar]
- 4. Patti G, Rapacciuolo A, Segreti A, Ussia GP, Goffredo C, Di Sciascio G. Percutaneous left atrial appendage closure: acute effects on left atrial pressure in humans. JACC Cardiovasc Interv. 2019;12:1089–1091. doi: 10.1016/j.jcin.2018.10.021 [DOI] [PubMed] [Google Scholar]
- 5. Luani B, Groscheck T, Genz C, Tanev I, Rauwolf T, Herold J, Medunjanin S, Schmeisser A, Braun‐Dullaeus RC. Left atrial enlargement and clinical considerations in patients with or without a residual interatrial shunt after closure of the left atrial appendage with the watchman‐device. BMC Cardiovasc Disord. 2017;17:294. doi: 10.1186/s12872-017-0728-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kim DY, Kim MJ, Seo J, Cho I, Shim CY, Hong GR, Kim JS, Ha JW. Predictors of subsequent heart failure after left atrial appendage closure. Circ J. 2022;86:1129–1136. doi: 10.1253/circj.CJ-21-0642 [DOI] [PubMed] [Google Scholar]
- 7. Granier M, Laugaudin G, Massin F, Cade S, Winum PF, Freitag C, Pasquie JL. Occurrence of incomplete endothelialization causing residual permeability after left atrial appendage closure. J Invasive Cardiol. 2018;30:245–250. [PubMed] [Google Scholar]
- 8. Sharma E, Apostolidou E, Sheikh W, Parulkar A, Ahmed MB, Lima FV, McCauley BD, Kennedy K, Chu AF. Hemodynamic effects of left atrial appendage occlusion. J Interv Card Electrophysiol. 2021. Epub ahead of print. doi: 10.1007/s10840-021-01006-x [DOI] [PubMed] [Google Scholar]
- 9. Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, Castella M, Diener HC, Heidbuchel H, Hendriks J, et al. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur Heart J. 2016;37:2893–2962. doi: 10.1093/eurheartj/ehw210 [DOI] [PubMed] [Google Scholar]
- 10. Ren Z, Zhang J, Wang S, Jia P, Li X, Zhang J, Guo R, Li H, Li S, Yang H, et al. Two‐year outcome from combining cryoballoon ablation and left atrial appendage closure: CLACBAC study. Front Cardiovasc Med. 2020;7:610537. doi: 10.3389/fcvm.2020.610537 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Hahn RT, Abraham T, Adams MS, Bruce CJ, Glas KE, Lang RM, Reeves ST, Shanewise JS, Siu SC, Stewart W, et al. Guidelines for performing a comprehensive transesophageal echocardiographic examination: recommendations from the American Society of Echocardiography and the Society of Cardiovascular Anesthesiologists. J Am Soc Echocardiogr. 2013;26:921–964. doi: 10.1016/j.echo.2013.07.009 [DOI] [PubMed] [Google Scholar]
- 12. Lang RM, Badano LP, Mor‐Avi V, Afilalo J, Armstrong A, Ernande L, Flachskampf FA, Foster E, Goldstein SA, Kuznetsova T, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2015;28:1–39.e14. doi: 10.1016/j.echo.2014.10.003 [DOI] [PubMed] [Google Scholar]
- 13. Ren Z, Jia P, Wang S, Li S, Li H, Guo R, Zhang J, Zhang J, Yang H, Meng W, et al. Acute left atrial ridge lesion after cryoballoon ablation: how does this affect left atrial appendage closure combined procedure? J Cardiovasc Electrophysiol. 2020;31:2865–2873. doi: 10.1111/jce.14718 [DOI] [PubMed] [Google Scholar]
- 14. Zhang J, Ren Z, Wang S, Zhang J, Yang H, Zheng Y, Meng W, Zhao D, Xu Y. Efficacy and safety of cryoballoon ablation for chinese patients over 75 years old: a comparison with a younger cohort. J Cardiovasc Electrophysiol. 2019;30:2734–2742. doi: 10.1111/jce.14220 [DOI] [PubMed] [Google Scholar]
- 15. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, Falk V, Gonzalez‐Juanatey JR, Harjola VP, Jankowska EA, et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail. 2016;18:891–975. doi: 10.1002/ejhf.592 [DOI] [PubMed] [Google Scholar]
- 16. Schulman S, Angeras U, Bergqvist D, Eriksson B, Lassen MR, Fisher W; Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis . Definition of major bleeding in clinical investigations of antihemostatic medicinal products in surgical patients. J Thromb Haemost. 2010;8:202–204. doi: 10.1111/j.1538-7836.2009.03678.x [DOI] [PubMed] [Google Scholar]
- 17. Reddy VY, Doshi SK, Kar S, Gibson DN, Price MJ, Huber K, Horton RP, Buchbinder M, Neuzil P, Gordon NT, et al. 5‐year outcomes after left atrial appendage closure: from the PREVAIL and PROTECT AF trials. J Am Coll Cardiol. 2017;70:2964–2975. doi: 10.1016/j.jacc.2017.10.021 [DOI] [PubMed] [Google Scholar]
- 18. Boersma LV, Ince H, Kische S, Pokushalov E, Schmitz T, Schmidt B, Gori T, Meincke F, Protopopov AV, Betts T, et al. Efficacy and safety of left atrial appendage closure with WATCHMAN in patients with or without contraindication to oral anticoagulation: 1‐year follow‐up outcome data of the EWOLUTION trial. Heart Rhythm. 2017;14:1302–1308. doi: 10.1016/j.hrthm.2017.05.038 [DOI] [PubMed] [Google Scholar]
- 19. Osmancik P, Herman D, Neuzil P, Hala P, Taborsky M, Kala P, Poloczek M, Stasek J, Haman L, Branny M, et al. Left atrial appendage closure versus direct oral anticoagulants in high‐risk patients with atrial fibrillation. J Am Coll Cardiol. 2020;75:3122–3135. doi: 10.1016/j.jacc.2020.04.067 [DOI] [PubMed] [Google Scholar]
- 20. Schneider B, Nazarenus D, Stollberger C. A 79‐year‐old woman with atrial fibrillation and new onset of heart failure. ESC Heart Fail. 2019;6:570–574. doi: 10.1002/ehf2.12435 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Park HC, Shin J, Ban JE, Choi JI, Park SW, Kim YH. Left atrial appendage: morphology and function in patients with paroxysmal and persistent atrial fibrillation. Int J Cardiovasc Imaging. 2013;29:935–944. doi: 10.1007/s10554-012-0161-y [DOI] [PubMed] [Google Scholar]
- 22. Jeong WK, Choi JH, Son JP, Lee S, Lee MJ, Choe YH, Bang OY. Volume and morphology of left atrial appendage as determinants of stroke subtype in patients with atrial fibrillation. Heart Rhythm. 2016;13:820–827. doi: 10.1016/j.hrthm.2015.12.026 [DOI] [PubMed] [Google Scholar]
- 23. Kosiuk J, Uhe T, Stegmann C, Ueberham L, Bertagnolli L, Dagres N, Dinov B, Mussigbrodt A, Richter S, Paetsch I, et al. Morphological determinators of platelet activation status in patients with atrial fibrillation. Int J Cardiol. 2019;279:90–95. doi: 10.1016/j.ijcard.2018.11.096 [DOI] [PubMed] [Google Scholar]
- 24. Luani B, Rauwolf T, Groscheck T, Tanev I, Herold J, Isermann B, Schmeisser A, Braun‐Dullaeus RC. Serial assessment of natriuretic peptides in patients undergoing interventional closure of the left atrial appendage. Heart Lung Circ. 2018;27:828–834. doi: 10.1016/j.hlc.2017.07.001 [DOI] [PubMed] [Google Scholar]
- 25. Herman D, Osmancik P, Neuzil P, Hala P, Lekesova V, Benesova K, Hozman M, Jarkovsky J, Novackova M, Widimsky P, et al. The effect of left atrial appendage closure on heart failure biomarkers: a PRAGUE‐17 trial subanalysis. J Cardiovasc Electrophysiol. 2021;32:2645–2654. doi: 10.1111/jce.15206 [DOI] [PubMed] [Google Scholar]
- 26. Pinto Teixeira P, Martins Oliveira M, Ramos R, Rio P, Silva Cunha P, Delgado AS, Pimenta R, Cruz FR. Left atrial appendage volume as a new predictor of atrial fibrillation recurrence after catheter ablation. J Interv Card Electrophysiol. 2017;49:165–171. doi: 10.1007/s10840-017-0256-4 [DOI] [PubMed] [Google Scholar]
- 27. Spencer RJ, DeJong P, Fahmy P, Lempereur M, Tsang MYC, Gin KG, Lee PK, Nair P, Tsang TSM, Jue J, et al. Changes in left atrial appendage dimensions following volume loading during percutaneous left atrial appendage closure. JACC Cardiovasc Interv. 2015;8:1935–1941. doi: 10.1016/j.jcin.2015.07.035 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1
Tables S1–S2
Figures S1–S2