Intracardiac echocardiography (ICE) and transesophageal echocardiography (TEE) are both used to guide left atrial appendage occlusion (LAAO). Earlier comparative analyzes predate widespread adoption of 3-dimensional ICE, and ICE was often reserved for patients with comorbidities precluding TEE. With continued technological advancements and increasing adoption of ICE, contemporary real-world comparisons are warranted. We conducted a large propensity-matched analysis using the TriNetX database to compare early outcomes between ICE- and TEE-guided LAAO from 2015 through 2025. Adults undergoing LAAO were identified using procedural codes, and 1:1 propensity score matching yielded 6,357 patients in each group. The primary outcome was major adverse events (MAE), defined as a composite of all-cause mortality, stroke, myocardial infarction, bleeding, pericardial effusion, cardiac arrest, and vascular complications. Secondary outcomes included pericardiocentesis, device-related thrombus, and peri-device leak at 7 and 45 days. At 7 days, ICE guidance was associated with a lower risk of MAE compared with TEE (766 vs 937 events; HR 0.81, 95% CI 0.74 to 0.89; p <0.0001), as well as lower rates of stroke and pericardial effusion. At 45 days, ICE remained associated with lower MAE risk (1,040 vs 1,193 events; HR 0.87; p = 0.0005), stroke, cardiac arrest, and pericardial effusion. Pericardiocentesis occurred more frequently in the ICE cohort. There were no significant differences in bleeding, device-related thrombus, peri-device leak, or mortality between groups. In conclusion, ICE-guided LAAO was associated with fewer early composite adverse events compared with TEE guidance, with similar mortality and device-related complication rates. These findings support the expanding use of ICE as an effective alternative imaging modality for LAAO.
Left atrial appendage occlusion (LAAO) is a well-established alternative to long-term oral anticoagulation for stroke prevention in patients with nonvalvular atrial fibrillation who are at increased bleeding risk or unable to tolerate anticoagulant therapy. Since its introduction, LAAO has relied heavily on high-quality intraprocedural imaging to facilitate transseptal access, guide device sizing and positioning, ensure adequate seal, and enable early detection of complications. Traditionally, transesophageal echocardiography (TEE) has served as the primary imaging modality for guidance during LAAO, owing to high spatial resolution, widespread availability, and familiarity among operators.
Despite these advantages, TEE carries several practical drawbacks. TEE typically requires general anesthesia or deep sedation, mandates the presence of an anesthesiologist and a dedicated imaging physician, and carries risks such as esophageal trauma and aspiration. These factors increase procedural complexity, prolong recovery, and add to overall resource utilization. As LAAO procedural volumes continue to rise, intracardiac echocardiography (ICE) has emerged as an alternative imaging strategy that may reduce workflow burden and resource consumption. ,
Advances in ICE technology including enhanced catheter resolution and the development of 3-dimensional imaging have broadened its applicability in structural heart interventions. Nonetheless, real-world adoption has remained limited. In a large analysis from the National Cardiovascular Data Registry (NCDR) encompassing 39,759 LAAO procedures performed from 2016 through 2018, only 2,272 procedures (5.7%) were performed using ICE, and its use was more common in cases involving concomitant interventions.
Several studies and registries have compared ICE and TEE for LAAO guidance, and multiple meta-analyzes suggest that clinical outcomes are broadly similar between the 2 modalities. However, these data are constrained by small ICE sample sizes, single-center designs, low event rates, and heterogeneity among studies. , Furthermore, studies using data prior to 2021 consist predominantly of 2-dimensional ICE imaging, which as largely been supplanted by 3-dimensional ICE for LAAO guidance. Consequently, a large-scale, contemporary, real-world evaluation is needed to characterize the comparative safety and effectiveness of ICE and TEE in current practice. To address this gap, we compared early outcomes between ICE-guided and TEE-guided LAAO using a large, national real-world database.
Methods
Data source
This study used the TriNetX Linked Network, a U.S.-based subset of the TriNetX Global Collaborative Network comprised of de-identified longitudinal health records from 35 healthcare organizations. The Linked dataset includes electronic health records, payer-sourced medical and pharmacy claims, and mortality data. For every patient, insurance coverage starts and stop dates enable comprehensive and continuous longitudinal care analysis. Available data included patient demographics, diagnoses, procedures, medications, and laboratory results. The study period extended from January 1, 2015 through October 1, 2025, with 2015 marking the FDA approval of the Watchman LAAO device (Boston Scientific, Marlborough, MA). All data within the platform are fully de-identified and compliant with the Health Insurance Portability and Accountability Act. Thus, this study was considered exempt from the requirements of the Baystate Medical Center institutional review board.
Study population
A retrospective cohort study was conducted of adult patients aged 18 years and older who underwent percutaneous LAAO identified within the TriNetX Global Health Research Network. Patients were categorized into 2 cohorts based on the intraprocedural imaging modality used to guide LAAO: ICE or TEE. To ensure accuracy of cohort classification, only patients with documentation of the imaging modality at the time of the LAAO procedure were included. Individuals were excluded if they had missing imaging documentation, both ICE and TEE documented, had undergone prior left atrial appendage occlusion, or lacked sufficient clinical data for matching and outcome assessment.
Outcomes
The primary outcome was major adverse events (MAE), defined as a composite of all-cause mortality, stroke, myocardial infarction, cardiac arrest, pericardial effusion, major bleeding or hematoma, and major vascular complications. Outcomes were assessed at 7 and 45 days following the index LAAO procedure. Secondary outcomes included the individual components of the composite endpoint as well as procedural complications including cardiac tamponade, pericardiocentesis, device-related thrombus (DRT), peri-device leak (PDL), and vascular complications. Patients with a documented history of a given outcome prior to the index procedure were excluded from the analysis for that specific outcome.
Propensity score matching
To minimize confounding and balance baseline characteristics between groups, 1:1 propensity score matching was performed between patients undergoing ICE-guided and TEE-guided LAAO. Propensity scores were estimated using logistic regression based on baseline demographic and clinical variables, including age, sex, race, body mass index, atrial fibrillation subtype, congestive heart failure, hypertension, diabetes, chronic kidney disease, coronary artery disease, prior stroke, prior myocardial infarction, baseline antithrombotic therapy, and cardiovascular medication use. Patients were matched using a greedy nearest-neighbor algorithm without replacement and a caliper width of 0.1 pooled standard deviations of the logit of the propensity score. Covariate balance after matching was assessed using standardized mean differences, with values <0.1 considered indicative of adequate balance.
Statistical analysis
Baseline characteristics were summarized as means with standard deviations for continuous variables and as frequencies with percentages for categorical variables. Comparisons between matched cohorts were performed using Student’s t tests for continuous variables and the chi-square tests for categorical variables, as appropriate. Clinical outcomes were evaluated using Kaplan–Meier survival analysis, with group differences assessed via the log-rank test. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs). A 2-sided p-value <0.05 was considered statistically significant. All analyzes were performed within the TriNetX analytics platform.
Results
Baseline characteristics
A total of 12,714 matched patients were included, with 6,357 undergoing ICE-guided LAAO and 6,357 undergoing TEE-guided LAAO. Baseline demographic and clinical characteristics were well balanced between groups following propensity score matching ( Table 1 ). The mean age was similar between cohorts (ICE 75.9 ± 7.9 vs TEE 76.0 ± 7.5 years; p = 0.52), and the proportion of female patients was comparable (40.3% vs 40.0%; p = 0.76). Comorbidities including hypertension (72.9% vs 72.8%), diabetes (30.5% vs 30.9%), chronic kidney disease (26.0% vs 26.4%), coronary artery disease (49.9% vs 49.7%), and prior stroke (11.8% vs 11.9%) were nearly identical between groups. Atrial fibrillation subtype distribution did not differ significantly. Medication use—including oral anticoagulants, antiplatelet therapy, beta-blockers, calcium-channel blockers, ACE inhibitors, and ARBs—was likewise similar. BMI was slightly lower in the ICE cohort (29.3 ± 6.2 vs 30.1 ± 6.5; p <0.0001), though all other measured covariates achieved excellent balance (standardized mean difference <0.1).
Table 1
Baseline characteristics of ICE- versus TEE-guided LAAO
| ICE (6357 patients) | TEE (6357 patients) | p-value | |
|---|---|---|---|
| Age | 75.9 ± 7.87 | 76 ± 7.47 | 0.5204 |
| Female | 2559 (40.3%) | 2542 (40%) | 0.7584 |
| BMI | 29.3 ± 6.16 | 30.1 ± 6.45 | <0.0001 |
|
Race
White Black Hispanic Asian |
|||
| 5664 (89.1%) | 5612 (88.3%) | 0.1454 | |
| 260 (4.1%) | 256 (4.03%) | 0.8573 | |
| 178 (2.8%) | 152 (2.39%) | 0.1470 | |
| 111 (1.746%) | 137 (2.155%) | 0.0954 | |
| CHF | 1300 (20.45%) | 1295 (20.37%) | 0.9124 |
| Systolic heart failure | 985 (15.49%) | 986 (15.51%) | 0.9805 |
| Diastolic heart failure | 1399 (22%) | 1387 (21.82%) | 0.7970 |
| Hypertension | 4636 (72.93%) | 4625 (72.75%) | 0.8264 |
| Diabetes | 1940 (30.52%) | 1965 (30.91%) | 0.6308 |
| Hx of Stroke | 749 (11.78%) | 755 (11.87%) | 0.8691 |
| Hx of AMI | 511 (8.04%) | 518 (8.148%) | 0.8199 |
| CKD | 1654 (26.02%) | 1679 (26.41%) | 0.6142 |
|
A.Fib
Paroxysmal Persistent Longstanding Permanent |
|||
| 4762 (74.91%) | 4706 (74.03%) | 0.2547 | |
| 2481 (39.03%) | 2463 (38.75%) | 0.7433 | |
| 608 (9.56%) | 591 (9.29%) | 0.6059 | |
| 597 (9.39%) | 612 (9.63%) | 0.6502 | |
| CAD | 3172 (49.89%) | 3158 (49.68%) | 0.8039 |
|
Antithrombotic therapy
Warfarin Eliquis Aspirin Ticagrelor Clopidogrel Rivaroxaban |
|||
| 548 (8.62%) | 537 (8.44%) | 0.7270 | |
| 3136 (49.33%) | 3103 (48.812%) | 0.5583 | |
| 2183 (34.34%) | 2230 (35.08%) | 0.3812 | |
| 50 (0.787%) | 48 (0.755%) | 0.8393 | |
| 1060 (16.67%) | 1084 (17.052%) | 0.5697 | |
| 984 (15.479%) | 975 (15.337%) | 0.8250 | |
| Beta-blockers | 4113 (64.7%) | 4083 (64.23%) | 0.5783 |
| Calcium channel blockers | 2319 (36.47%) | 2301 (36.196%) | 0.7400 |
| ARB | 1807 (28.43%) | 1770 (27.84%) | 0.4655 |
| ACEi | 1140 (17.93%) | 1142 (17.96%) | 0.9631 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
