Brief Report
Atrial fibrillation (AF) increases the risk of embolic stroke by fivefold in high-risk patients. Percutaneous left atrial appendage occlusion (LAAO) has emerged as an established alternative for those who are unable to tolerate chronic oral anticoagulation for stroke. Patients undergoing LAAO often have multiple comorbidities, including heart failure and chronic kidney disease, that significantly influence their post-LAAO clinical trajectory, risk of adverse cardiovascular outcomes, and recovery. Frailty reflects a multidisciplinary clinical syndrome characterized by impaired physiologic reserve and functional capacity, resulting in heightened susceptibility to stressors and an increased risk of adverse outcomes, particularly in the setting of invasive procedures. Frailty leads to multiple organ system impairment (cognition, mobility, energy levels) for older patients, and the predisposition to developing frailty increases with age and multiple comorbid conditions. The prevalence of frailty varies across populations and surrogate definitions; the current estimates for patients in the United States vary between 10% and 55%. , However, whether frail patients are more vulnerable to adverse events following LAAO remains undefined. Therefore, we sought to analyze the impact of frailty on outcomes in patients who underwent LAAO using a large, nationally representative, real-world cohort of the US population.
This analysis was conducted using the United States Collaborative Network of the TriNetX (Cambridge, MA), a federated electronic health record-based longitudinal database that aggregates data from >70 currently participating healthcare organizations (HCOs) in the United States. The data covers diverse geographic regions—39% South, 22% Northeast, 16% Midwest, 13% West, and 10% unspecified. TriNetX anonymizes EHR data from a network of HCOs, mostly comprising large academic institutions with diverse inpatient and outpatient services, including both insured and noninsured patients across the US. TriNetX performs comprehensive data preprocessing to minimize missing values and harmonizes all information into a standardized clinical data model, ensuring consistency of query results across diverse data sources. Patient cohorts, diagnoses, and outcomes were defined using the International Classification of Diseases, tenth revision, clinical modification (ICD-10-CM), ICD-10 procedural coding system (ICD-10-PCS), and current procedural terminology codes. Adult patients (≥18 years) undergoing LAAO (identified using ICD-10-PCS codes—02B73ZK, 02573ZK, 02L73CK, 02L74CK, 02L73DK, 02L74DK, 02L73ZK, 02L73ZK, 02574ZK, 02B74ZK, and 02L74ZK) between 2015 and 2025 were included. Frailty was defined using the Johns Hopkins Adjusted Clinical Groups frailty-defining diagnoses, including malnutrition, dementia, severe vision impairment, decubitus ulcers, urinary incontinence, weight loss, fecal incontinence, social support needs, difficulty in walking, and falls. Patients were classified in the frail study group if ≥1 of these frailty-defining diagnoses were present ( Supplementary Tables 1 and 2 ). The Johns Hopkins ACG cluster has previously been validated to assess patients’ frailty status. , The TriNetX platform is compliant with §164.514 of the Health Insurance Portability and Accountability Act Privacy Rule; hence, this study was exempt from the Western Institutional Review Board (WIRB) approval.
The primary endpoint was all-cause mortality at 1-year follow-up, and the secondary outcomes included major adverse cardiac and cerebrovascular event (MACCE) (composite of all-cause mortality, acute myocardial infarction, and ischemic stroke), major bleeding, major vascular complications, pericardial complications, and all-cause readmission. The status of death was based on the vital status code “deceased” that TriNetX imports from the Social Security Death Index. The outcomes were examined at 1 year after LAAO ( Supplementary Table 3 ). We adhered to validated methodologies for outcome evaluation within the TriNetX platform.
Continuous variables are presented as mean ± SD and were compared using the independent-sample Student’s t-tests. Categorical variables are presented as frequencies or percentages and were compared using the Chi-square test. To mitigate the effect of probable measured confounders, we utilized 1:1 propensity-score matching (PSM), including the following types of covariates: demographics (age, sex, race), diagnoses, medications, and laboratory domains. PSM using the ‘greedy nearest-neighbor matching’ with a caliper of 0.1 pooled standard deviation (SD) of the linear propensity scores to control for differences in the cohorts ( Supplementary Tables 4 and 5 ). Covariates were considered well-balanced if standardized mean differences were <0.1.
Data were represented as hazard ratios (HRs) and 95% confidence intervals (CIs) on matched cohorts using Kaplan–Meier analysis using the log-rank test. The TriNetX platform calculates HRs with CIs using R’s Survival package v3.2-3. For generating HRs, TriNetX sets the parameter robust = FALSE using the R survival package. This represents a limitation of the platform, as it does not account for potential clustering of participants within healthcare organizations (HCO) or specific geographic regions—partly due to TriNetX’s data privacy requirement to restrict visibility by HCO source. All statistical analyses were conducted within the TriNetX platform. Statistical significance was set at a two-tailed p-value <0.05.
Between January 1, 2015 and October 29, 2025, a total of 36,677 adult patients undergoing LAAO were identified (frail: 8,384; nonfrail: 28,293) ( Table 1 ), following which 8,051 matched pairs were identified. Before matching, frail patients were more often older (80 years vs 78 years), female (44% vs 38%), and White (88% vs 85%), and had more comorbid diagnoses. Matched cohort analysis demonstrated that patients in the frail cohort undergoing LAAO had a significantly higher risk of all-cause mortality (HR 1.48; 95% CI 1.31 to 1.66; p <0.001), MACCE (HR 1.26; 95% CI 1.18 to 1.34; p <0.001), major bleeding (HR 1.08; 95% CI 1.02 to 1.15; p = 0.004), major vascular complications (HR 1.32; 95% CI 1.17 to 1.49; p <0.001), and all-cause readmission (HR 1.15; 95% CI 1.11 to 1.19; p <0.001). The risk of pericardial complications (HR 0.98; 95% CI 0.70 to 1.37; p = 0.92) was comparable between the 2 groups ( Figure 1 ). Among frail patients with ≥2 frailty-defining conditions and the nonfrail cohort undergoing LAAO (3,301 matched pairs), matched cohort analysis demonstrated that frail patients had significantly higher all-cause mortality (HR 1.67; 95% CI 1.52 to 1.82; p <0.001), MACCE (HR 1.45; 95% CI 1.38 to 1.51; p <0.001), major bleeding (HR 1.22; 95% CI 1.19 to 1.25; p <0.001), major vascular complications (HR 1.49; 95% CI 1.32 to 1.76; p <0.001), and all-cause readmission (HR 1.29; 95% CI 1.26 to 1.31; p <0.001).
Table 1
Baseline characteristics of the study cohort before and after propensity score matching
| Variables | Before propensity-score matching | After propensity-score matching | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Frail (n = 8,384) | Not Frail (n = 28,293) | p-value | Std. difference | Frail (n = 8,051) | Not frail (n = 8,051) | p- value | Std. difference | ||
| Age | 79.8 ± 8.0 | 77.7 ± 8.7 | <0.01 | 0.251 | 79.6 ± 8.0 | 79.6 ± 7.7 | 0.57 | 0.009 | |
| Age at index | 77.1 ± 8.2 | 74.4 ± 8.9 | <0.01 | 0.319 | 77.0 ± 8.2 | 76.9 ± 7.8 | 0.8 | 0.004 | |
| Sex | Male | 4,681 (55.8) | 17,341 (61.4) | <0.01 | 0.113 | 4,525 (56.2) | 4,484 (55.7) | 0.52 | 0.010 |
| Female | 3,703 (44.2) | 10,898 (38.6) | <0.01 | 0.113 | 3,526 (43.8) | 3,567 (44.3) | 0.52 | 0.010 | |
| Race | White | 7,407 (88.3) | 24,081 (85.3) | <0.01 | 0.091 | 7,109 (88.3) | 7,087 (88) | 0.59 | 0.008 |
| African American | 474 (5.7) | 1,548 (5.5) | 0.55 | 0.007 | 447 (5.6) | 454 (5.6) | 0.81 | 0.004 | |
| Asian | 218 (2.6) | 801 (2.8) | 0.25 | 0.015 | 211 (2.6) | 213 (2.6) | 0.92 | 0.002 | |
| Comorbidities | Heart failure | 5,112 (61) | 12,547 (44.4) | <0.01 | 0.336 | 4,829 (60) | 4,835 (60.1) | 0.92 | 0.002 |
| Hypertension | 7,875 (93.9) | 22,886 (81) | <0.01 | 0.397 | 7,544 (93.7) | 7,578 (94.1) | 0.26 | 0.018 | |
| Ischemic heart diseases | 6,048 (72.1) | 16,521 (58.5) | <0.01 | 0.289 | 5,752 (71.4) | 5,727 (71.1) | 0.66 | 0.007 | |
| Acute myocardial infarction | 2,041 (24.3) | 4,205 (14.9) | <0.01 | 0.240 | 1,896 (23.5) | 1,868 (23.2) | 0.60 | 0.008 | |
| Peripheral vascular diseases | 1,942 (23.2) | 4,013 (14.2) | <0.01 | 0.231 | 1,806 (22.4) | 1,822 (22.6) | 0.76 | 0.005 | |
| Aortic aneurysm and dissection | 909 (10.8) | 2,207 (7.8) | <0.01 | 0.104 | 856 (10.6) | 861 (10.7) | 0.9 | 0.002 | |
| Cardiomyopathy | 1,990 (23.7) | 5,410 (19.2) | <0.01 | 0.112 | 1,884 (23.4) | 1,878 (23.3) | 0.91 | 0.002 | |
| Nonrheumatic mitral valve disorders | 3,594 (42.9) | 9,000 (31.9) | <0.01 | 0.229 | 3,394 (42.2) | 3,418 (42.5) | 0.70 | 0.006 | |
| Nonrheumatic aortic valve disorders | 2,895 (34.5) | 6,881 (24.4) | <0.01 | 0.224 | 2,723 (33.8) | 2,688 (33.4) | 0.56 | 0.009 | |
| Paroxysmal tachycardia | 2,418 (28.8) | 5,811 (20.6) | <0.01 | 0.192 | 2,263 (28.1) | 2,286 (28.4) | 0.69 | 0.006 | |
| Nonrheumatic tricuspid valve disorders | 2,117 (25.3) | 4,417 (15.6) | <0.01 | 0.240 | 1,967 (24.4) | 2,007 (24.9) | 0.47 | 0.012 | |
| Stroke | 2,367 (28.2) | 4,135 (14.6) | <0.01 | 0.336 | 2,143 (26.6) | 2,170 (27) | 0.63 | 0.008 | |
| Pericardial effusion | 839 (10) | 1,576 (5.6) | <0.01 | 0.166 | 756 (9.4) | 750 (9.3) | 0.87 | 0.003 | |
| Hyperlipidemias | 7,246 (86.4) | 19,768 (70) | <0.01 | 0.406 | 6,920 (86) | 6,942 (86.2) | 0.62 | 0.008 | |
| Diabetes mellitus | 4,010 (47.8) | 9,839 (34.8) | <0.01 | 0.266 | 3,787 (47) | 3,770 (46.8) | 0.79 | 0.004 | |
| Thyroid disorders | 3,164 (37.7) | 7,038 (24.9) | <0.01 | 0.279 | 2,975 (37) | 2,983 (37.1) | 0.9 | 0.002 | |
| Osteoarthritis | 4,927 (58.8) | 10,571 (37.4) | <0.01 | 0.437 | 4,634 (57.6) | 4,684 (58.2) | 0.43 | 0.013 | |
| Sleep apnea | 3,376 (40.3) | 8,811 (31.2) | <0.01 | 0.190 | 3,204 (39.8) | 3,140 (39) | 0.30 | 0.016 | |
| Chronic kidney disease | 3,755 (44.8) | 8,243 (29.2) | <0.01 | 0.327 | 3,517 (43.7) | 3,492 (43.4) | 0.69 | 0.006 | |
| Liver diseases | 1,857 (22.1) | 4,015 (14.2) | <0.01 | 0.207 | 1,735 (21.6) | 1,719 (21.4) | 0.76 | 0.005 | |
| Gastro-esophageal reflux disease | 4,603 (54.9) | 10,383 (36.8) | <0.01 | 0.370 | 4,337 (53.9) | 4,361 (54.2) | 0.70 | 0.006 | |
| Gastritis and duodenitis | 1,612 (19.2) | 3,458 (12.2) | <0.01 | 0.193 | 1,513 (18.8) | 1,495 (18.6) | 0.72 | 0.006 | |
| Chronic lower respiratory diseases | 3,778 (45.1) | 8,714 (30.9) | <0.01 | 0.296 | 3,547 (44.1) | 3,570 (44.3) | 0.72 | 0.006 | |
| Iron deficiency anemia | 3,004 (35.8) | 6,681 (23.7) | <0.01 | 0.269 | 2,818 (35) | 2,825 (35.1) | 0.91 | 0.002 | |
| Neoplasms | 4,828 (57.6) | 11,685 (41.4) | <0.01 | 0.328 | 4,571 (56.8) | 4,581 (56.9) | 0.87 | 0.003 | |
| Nicotine dependence | 1,335 (15.9) | 3,370 (11.9) | <0.01 | 0.115 | 1,262 (15.7) | 1,236 (15.4) | 0.57 | 0.009 | |
| Alcohol related disorders | 857 (10.2) | 1,615 (5.7) | <0.01 | 0.167 | 790 (9.8) | 794 (9.9) | 0.92 | 0.002 | |
| Prior procedures | Percutaneous coronary intervention | 820 (9.8) | 2,161 (7.7) | <0.01 | 0.075 | 787 (9.8) | 786 (9.8) | 0.98 | <0.001 |
| Pacemaker or implantable cardioverter-defibrillator | 1,361 (16.2) | 3,456 (12.2) | <0.01 | 0.115 | 1,295 (16.1) | 1,338 (16.6) | 0.36 | 0.014 | |
| Coronary artery bypass grafting | 196 (2.3) | 395 (1.4) | <0.01 | 0.069 | 186 (2.3) | 182 (2.3) | 0.83 | 0.003 | |
| Cardiac valve surgery | 543 (6.5) | 1,314 (4.7) | <0.01 | 0.080 | 519 (6.4) | 528 (6.6) | 0.77 | 0.005 | |
| Medications | Beta blockers | 7,548 (90) | 22,311 (79) | <0.01 | 0.308 | 7,219 (89.7) | 7,221 (89.7) | 0.96 | 0.001 |
| Antiarrhythmics | 7,450 (88.9) | 21,899 (77.5) | <0.01 | 0.306 | 7,121 (88.4) | 7,133 (88.6) | 0.77 | 0.005 | |
| Antilipemic agents | 6,906 (82.4) | 19,593 (69.4) | <0.01 | 0.307 | 6,590 (81.9) | 6,565 (81.5) | 0.61 | 0.008 | |
| Diuretics | 6,388 (76.2) | 17,720 (62.8) | <0.01 | 0.295 | 6,085 (75.6) | 6,051 (75.2) | 0.53 | 0.010 | |
| Calcium channel blockers | 5,850 (69.8) | 16,101 (57) | <0.01 | 0.267 | 5,570 (69.2) | 5,559 (69) | 0.85 | 0.003 | |
| Angiotensin II inhibitor | 3,665 (43.7) | 10,141 (35.9) | <0.01 | 0.160 | 3,495 (43.4) | 3,504 (43.5) | 0.89 | 0.002 | |
| Antianginals | 4,158 (49.6) | 11,242 (39.8) | <0.01 | 0.198 | 3,939 (48.9) | 3,865 (48) | 0.24 | 0.018 | |
| ACE inhibitors | 3,863 (46.1) | 10,228 (36.2) | <0.01 | 0.201 | 3,676 (45.7) | 3,655 (45.4) | 0.74 | 0.005 | |
| Midodrine | 723 (8.6) | 832 (2.9) | <0.01 | 0.245 | 604 (7.5) | 556 (6.9) | 0.14 | 0.023 | |
| Milrinone | 133 (1.6) | 319 (1.1) | <0.01 | 0.039 | 128 (1.6) | 117 (1.5) | 0.48 | 0.011 | |
| Sacubitril | 579 (6.9) | 1,547 (5.5) | <0.01 | 0.059 | 553 (6.9) | 551 (6.8) | 0.95 | 0.001 | |
| Analgesics | 7,970 (95.1) | 24,221 (85.8) | <0.01 | 0.320 | 7,637 (94.9) | 7,636 (94.8) | 0.97 | 0.001 | |
| Antidepressants | 4,603 (54.9) | 9,324 (33) | <0.01 | 0.452 | 4,298 (53.4) | 4,261 (52.9) | 0.56 | 0.009 | |
| Sedatives | 6,978 (83.2) | 19,976 (70.7) | <0.01 | 0.300 | 6,659 (82.7) | 6,648 (82.6) | 0.82 | 0.004 | |
| Anticoagulants | 7,863 (93.8) | 23,697 (83.9) | <0.01 | 0.318 | 7,530 (93.5) | 7,530 (93.5) | 1 | <0.001 | |
| Platelet aggregation inhibitors | 6,506 (77.6) | 18,062 (64) | <0.01 | 0.303 | 6,201 (77) | 6,195 (76.9) | 0.91 | 0.002 | |
| Aspirin | 6,300 (75.1) | 17,205 (60.9) | <0.01 | 0.308 | 6,000 (74.5) | 5,996 (74.5) | 0.94 | 0.001 | |
| Clopidogrel | 2,912 (34.7) | 7,402 (26.2) | <0.01 | 0.186 | 2,753 (34.2) | 2,725 (33.8) | 0.64 | 0.007 | |
| Antihemorrhagics | 1,990 (23.7) | 4,210 (14.9) | <0.01 | 0.225 | 1,870 (23.2) | 1,867 (23.2) | 0.96 | 0.001 | |
| Heparin antagonists | 1,620 (19.3) | 5,396 (19.1) | 0.66 | 0.005 | 1,571 (19.5) | 1,573 (19.5) | 0.97 | 0.001 | |
| Antacids | 7,136 (85.1) | 19,736 (69.9) | <0.01 | 0.371 | 6,810 (84.6) | 6,832 (84.9) | 0.63 | 0.008 | |
| Antiulcer agents | 3,859 (46) | 9,335 (33.1) | <0.01 | 0.268 | 3,642 (45.2) | 3,611 (44.9) | 0.62 | 0.008 | |
| Glucocorticoids | 6,799 (81.1) | 18,331 (64.9) | <0.01 | 0.371 | 6,473 (80.4) | 6,474 (80.4) | 0.98 | <0.001 | |
| Insulin | 3,895 (46.5) | 8,718 (30.9) | <0.01 | 0.324 | 3,664 (45.5) | 3,620 (45) | 0.49 | 0.011 | |
| Metformin | 1,866 (22.3) | 4,719 (16.7) | <0.01 | 0.140 | 1,773 (22) | 1,718 (21.3) | 0.29 | 0.017 | |
| Empagliflozin | 812 (9.7) | 1,930 (6.8) | <0.01 | 0.104 | 767 (9.5) | 733 (9.1) | 0.36 | 0.015 | |
| Dapagliflozin | 499 (6) | 1,025 (3.6) | <0.01 | 0.109 | 459 (5.7) | 456 (5.7) | 0.92 | 0.002 | |
| Antineoplastics | 1,255 (15) | 2,827 (10) | <0.01 | 0.150 | 1,168 (14.5) | 1,157 (14.4) | 0.81 | 0.004 | |
| Hemoglobin, g/dl | 12.2 ± 2.1 | 12.7 ± 2.1 | <0.01 | 0.252 | 12.2 ± 2.1 | 12.4 ± 2.1 | <0.01 | 0.103 | |
| Hematocrit, % | 37.6 ± 6.1 | 38.9 ± 6.3 | <0.01 | 0.213 | 37.6 ± 6.1 | 38.3 ± 6.0 | <0.01 | 0.115 | |
| Prothrombin time, sec | 15.1 ± 6.1 | 15.1 ± 6.2 | 0.96 | 0.001 | 15.1 ± 6.0 | 15.2 ± 6.6 | 0.69 | 0.007 | |
| Activated partial thromboplastin time, sec | 34.9 ± 13.5 | 35.0 ± 13.7 | 0.67 | 0.006 | 35.0 ± 13.6 | 35.3 ± 14.9 | 0.22 | 0.021 | |
| International normalized ratio | 1.3 ± 0.5 | 1.3 ± 0.5 | 0.72 | 0.005 | 1.3 ± 0.5 | 1.3 ± 0.5 | 0.4 | 0.014 | |
| Cholesterol, mg/dl | 143.9 ± 40.4 | 148.8 ± 41.3 | <0.01 | 0.12 | 144.2 ± 40.3 | 146.6 ± 40.2 | <0.01 | 0.062 | |
| Low-density lipoprotein cholesterol, mg/dl | 73.1 ± 31.6 | 77.8 ± 33.4 | <0.01 | 0.145 | 73.4 ± 31.5 | 75.2 ± 32.2 | <0.01 | 0.055 | |
| High-density lipoprotein cholesterol, mg/dl | 45.0 ± 20.2 | 45.5 ± 18.5 | 0.05 | 0.028 | 45.0 ± 20.1 | 45.4 ± 19.3 | 0.21 | 0.023 | |
| Triglyceride, mg/dl | 117.0 ± 70.7 | 122.0 ± 85.4 | <0.01 | 0.063 | 117.2 ± 71.2 | 121.2 ± 73.3 | <0.01 | 0.056 | |
| Troponin I, ng/ml | 0.4 ± 3.1 | 0.8 ± 6.6 | <0.01 | 0.086 | 0.4 ± 3.1 | 0.7 ± 5.0 | <0.01 | 0.074 | |
| NT-proBNP, pg/ml | 3490.1 ± 7283.3 | 2771.6 ± 5888.5 | <0.01 | 0.108 | 3414.7 ± 7226.1 | 3018.7 ± 6259.0 | 0.04 | 0.059 | |
| Hemoglobin A1c, % | 6.1 ± 1.1 | 6.1 ± 1.1 | 0.35 | 0.014 | 6.1 ± 1.1 | 6.1 ± 1.1 | 0.11 | 0.029 | |
| C-reactive protein, mg/ml | 31.8 ± 53.2 | 29.4 ± 50.9 | 0.03 | 0.048 | 31.5 ± 53.3 | 30.9 ± 52.4 | 0.66 | 0.012 | |
| Erythrocyte sedimentation rate, mm/h | 28.6 ± 26.1 | 26.6 ± 24.7 | <0.01 | 0.079 | 28.5 ± 26.2 | 28.1 ± 25.0 | 0.53 | 0.016 | |
| Respiratory rate/min | 17.0 ± 2.6 | 17.0 ± 2.7 | 0.96 | 0.001 | 17.0 ± 2.6 | 17.0 ± 2.7 | 0.78 | 0.005 | |
| Heart rate/min | 72.6 ± 15.0 | 72.0 ± 15.1 | <0.01 | 0.038 | 72.5 ± 15.0 | 72.2 ± 14.9 | 0.22 | 0.021 | |
| Oxygen saturation, % | 86.4 ± 20.6 | 90.5 ± 16.9 | <0.01 | 0.219 | 86.6 ± 20.4 | 88.3 ± 19.2 | <0.01 | 0.085 | |
| Blood pressure, Systolic, mm Hg | 129.0 ± 21.5 | 130.2 ± 20.7 | <0.01 | 0.057 | 129.1 ± 21.5 | 130.3 ± 20.8 | <0.01 | 0.061 | |
| Blood pressure, Diastolic, mm Hg | 71.5 ± 12.9 | 72.3 ± 12.6 | <0.01 | 0.065 | 71.5 ± 13.0 | 71.2 ± 12.1 | 0.13 | 0.026 | |
| Body mass index, kg/m 2 | 28.9 ± 6.4 | 30.2 ± 6.5 | <0.01 | 0.205 | 28.9 ± 6.4 | 29.9 ± 6.4 | <0.01 | 0.162 | |
| Left ventricular ejection fraction (%) | 55.5 ± 13.5 | 55.0 ± 13.3 | 0.14 | 0.034 | 55.4 ± 13.5 | 55.6 ± 12.9 | 0.62 | 0.014 | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
