Citations

Objectives To explore the outcomes among patients treated with TAVI (transcatheter aortic valve implantation) in the COMPARE-TAVI 1 trial by age groups and sex. Design Cohort study. Setting Three Danish university hospitals. Participants Patients treated with TAVI in the COMPARE-TAVI 1 trial (N=1030), 60 (5.8%), 337 (32.7%), 592 (57.5%) and 41 (4.0%) were in the four age groups, < 70, 70–79, 80–89 and 90+ years while 414 (40.2%) and 616 (59.8%) were of female and male sex. Interventions Transfemoral TAVI with balloon-expandable valves (Sapien or Myval) performed between June 2020 and November 2023. Outcome measures Post hoc analyses of trial secondary outcomes that include end-point committee-adjudicated outcomes. Results In the four age groups, 1-year mortality rates were 8.3%, 5.6%, 5.6% and 7.3% while it was 5.8% in both sexes. Across the four age groups, 1-year stroke rates were 6.7%, 1.8%, 3.2% and 4.9%, and 5.1% among females and 4.7% among males. In the four age groups, 1-year new pacemaker rates were 7.5%, 15%, 19% and 13% while it was 15% among females and 18% among males. Females had smaller annuli than males and their gradients were higher and the effective orifice area lower after TAVI but their frequency of prosthesis-patient mismatch was similar at 1 year. Conclusion While comorbidity levels varied between groups, clinical outcomes were similar across age groups and between sexes. Trial registration number NCT04443023 .

BACKGROUND: Intraventricular hemorrhage and subarachnoid hemorrhage are associated with neuroinflammation and increased morbidity and mortality. Rapid clearance of blood from the cerebrospinal fluid space may mitigate neuroinflammatory cascades and improve clinical outcomes. This systematic review and meta-analysis aimed to compare the clinical effectiveness of active cerebrospinal fluid exchange, including neuroendoscopic lavage and IRRAflow systems, versus external ventricular drainage in patients with intraventricular hemorrhage or subarachnoid hemorrhage. METHODS: The study followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and thoroughly assessed databases, including PubMed, Embase, Scopus, Cochrane, and Web of Science, up to September 18, 2025. Statistical analysis, both fixed-effect and random-effect model analyses, was performed by the Comprehensive Meta-Analysis software. In addition, the Cochran Q test was conducted to determine the statistical heterogeneity. In terms of low heterogeneity, the fixed-effect model is reported. RESULTS: This review included 7 studies with a total sample size of 834. Our analysis showed that active cerebrospinal fluid exchange was associated with significantly higher hematoma clearance (odds ratio, 2.72 [95% CI, 1.22–6.08]; P =0.015), better functional outcomes (odds ratio, 1.66 [95% CI, 1.12–2.48]; P =0.012), lower shunt dependency (odds ratio, 0.22 [95% CI, 0.10–0.50]; P <0.001), and reduced infection rates (odds ratio, 0.22 [95% CI, 0.12–0.42]; P <0.001). Subgroup analysis showed that neuroendoscopic lavage was most effective for hematoma clearance, while IRRAflow reduced infection rates. Also, younger patients had better neurological recovery. CONCLUSIONS: Active cerebrospinal fluid exchange techniques demonstrated superior outcomes over external ventricular drainage in managing intraventricular hemorrhage and subarachnoid hemorrhage, particularly in clot resolution and neurological recovery. Large-scale trials are warranted.

ABSTRACT Ischemic events after contemporary percutaneous coronary intervention (PCI) are uncommon but carry high morbidity/mortality. Identifying patients at highest risk is critical to guide dual antiplatelet therapy (DAPT) intensity while minimizing bleeding. Current machine learning (ML) model‐developed risk scores do not incorporate pharmacogenetic data. To develop and externally validate ML models that integrate clinical, demographic, and CYP2C19 genetic information to predict 1‐year ischemic outcomes post‐PCI to refine clinical decision making. We analyzed 8317 patients from TAILOR‐PCI trial ( n = 4572) and the Precision PCI registry ( n = 3745). The outcome was a composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis at 1 year. Boruta feature selection identified 11 predictors. Multiple ML algorithms were trained in TAILOR‐PCI and validated in Precision PCI using cross‐validation and synthetic minority oversampling (SMOTE). Model performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The best external performance was achieved with a support vector machine (SVM, polynomial kernel) (AUC 0.667; sensitivity 0.871; specificity 0.282), while XGBoost provided a more balanced profile (AUC 0.619; sensitivity 0.442; specificity 0.688). Variable importance for the SVM polynomial model demonstrated that all 11 included Boruta feature selected predictors had relatively high importance (> 75). ML models trained on large clinical trial and real‐world registry datasets can help identify the small subset of patients at high ischemic risk after PCI. This distinction is of relevance because ischemic events are rare, and most patients may safely de‐escalate DAPT to reduce bleeding risk while maintaining ischemic protection through a multimodal approach to risk stratification. Trial Registration: TAILOR‐PCI URL: https://clinicaltrials.gov/ct2/show/NCT01742117

Abstract Body surface gastric mapping (BSGM) enables non-invasive measurement of gastric electrophysiology, but the signals are approximately 100 times weaker than cardiac potentials and overlap spectrally with motion artifacts, necessitating labor-intensive manual review that limits clinical scalability. We present an uncertainty-aware deep learning framework combining a signal reconstruction network with a parallel uncertainty estimation network to automate artifact correction in high-resolution BSGM. Models were trained on 2,398 multihour, 64-channel recordings from 27 international clinical sites using weak supervision, a physiology-aware loss function, and uncertainty-gated quality control. In an independent cohort of 127 patients, the system achieved relative reductions of 39% in signal reconstruction error, 9% in total data removed, and 23% in amplitude–movement correlation compared with the industry-standard Wiener filter. Improved signal fidelity altered automated clinical phenotyping in 7% of patients by recovering previously obscured gastric rhythms. Uncertainty-aware deep learning enables reliable automated artifact correction in body-surface gastric mapping, improving signal fidelity and enabling scalable clinical interpretation. The system is FDA-cleared (510(k) K252504) and deployed in clinical practice, demonstrating that data-driven artifact correction can meet regulatory requirements for medical devices and reduce dependence on specialist manual review.

Abstract Background Diagnosis and risk stratification in rare hematologic malignancies such as myeloproliferative neoplasms (MPNs) – polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF) – require expert review of longitudinal, heterogeneous clinical records. This process is cognitively demanding, inconsistently applied, and difficult to scale beyond tertiary centers. No automated phenotyping workflow currently exists for hematologic malignancies. Methods A HIPAA-compliant large language model (LLM) framework for phenotyping MPN was developed to integrate (i) rule-based retrieval of bone marrow biopsy reports, clinical notes, and structured laboratory results from the electronic health record (EHR); (ii) zero-shot extraction of diagnostic and prognostic variables from unstructured text using GPT-4 Turbo; (iii) a clinician-informed source-prioritization algorithm to reconcile conflicting multi-source data; (iv) WHO/ICC-criteria–based diagnostic classification; and (v) NCCN-based risk stratification using the conventional risk model for PV, IPSET-thrombosis for ET, and DIPSS, DIPSS-plus, and MIPSS70/MIPSS70+ v2 for MF. Patients were identified via MPN-related ICD-9/10 codes; cases met 2017 WHO criteria or had a hematologist-documented diagnosis, and controls did not. The cohort was split into a prompt-development set (n = 60) and a held-out test set (n = 450; 75 cases and 75 controls per disease). Ground truth was established by independent dual-clinician chart review with consensus adjudication. LLM performance was evaluated against the ground truth: variable-level extraction using accuracy, F1 score, and Cohen’s kappa; patient-level diagnostic classification using sensitivity, specificity, and Cohen’s kappa; and prognostic risk stratification (among confirmed cases) using accuracy, weighted F1 score, and quadratic-weighted Cohen’s kappa. Wilson 95% confidence intervals (CIs) were used for proportions and bootstrap 95% CIs with 500 resamples for F1 scores. Results The held-out test set included 450 patients (PV: 150; ET: 150; MF: 150) with pathology reports and structured laboratory results, and 172 patients (PV: 52; ET: 55; MF: 65) with clinical notes. From pathology reports, overall variable extraction accuracy and F1 score were 99% (95% CI, 98–100) and 1.00 (0.99–1.00) for PV, 100% (99–100) and 0.99 (0.96–1.00) for ET, and 100% (99–100) and 0.99 (0.97–1.00) for MF. From clinical notes, overall accuracy and F1 score were 96% (91–100) and 0.94 (0.85–1.00) for PV, 100% (100–100) and 1.00 (1.00–1.00) for ET, and 100% (99–100) and 0.98 (0.95–1.00) for MF. Diagnostic sensitivity was 100% (95% CI, 95.1–100.0) for PV, ET, and MF; specificity was 98.7% (92.8–99.8) for PV and 100% (95.1–100.0) for both ET and MF, with Cohen’s kappa of 0.99 for PV and 1.00 for ET and MF. Risk stratification accuracy was 100% with weighted F1 score of 1.00 and quadratic-weighted Cohen’s kappa of 1.00 across all three diseases. A pre-specified source-ablation analysis showed that pathology reports alone were sufficient for diagnosis (sensitivity 98.7% for PV, 100% for ET, 96.0% for MF; specificity 100% across all three subtypes) but inadequate for prognostication (accuracy 69.3% for PV, 93.3% for ET, 77.3% for MF). Adding clinical notes to pathology reports recovered full prognostic accuracy of 100% across all three diseases. Conclusions This first-in-class automated framework achieved expert-level performance for MPN diagnosis and risk stratification from real-world EHR data, establishing a foundation for scalable, standardized phenotyping in rare hematologic malignancies. Prospective, multi-site validation is warranted before clinical deployment.

Anomalous aortic origin of a coronary artery (AAOCA) is a rare congenital anomaly affecting around 0.4–0.8% of the population, but limited mortality data exist. We characterized trends and disparities in AAOCA-related mortality in the United States from 1999 to 2023. We queried the Centers for Disease Control and Prevention Wide-Ranging Online Data for Epidemiologic Research multiple-cause-of-death database for AAOCA-related deaths among all age groups. Age-adjusted mortality rates (AAMR) per 1,000,000 population with 95% confidence interval, stratified by year, sex, race, regions, and urbanization status, were abstracted. Annual percent changes were computed by Joinpoint regression. From 1999 to 2023, there were 6665 AAOCA-related deaths. Overall AAMR fell insignificantly from 1.1 in 1999 to 0.9 by 2001, rose to 1.1 in 2008, declined significantly to 0.5 by 2020, and then increased to 0.8 by 2023. Male AAMRs exceeded females (1999: 1.3 vs 0.9; 2023: 1.2 vs 0.4). Racially, from 1999–2020 to 2021–2023, non-Hispanic (NH) Blacks exhibited the highest AAMR (1.8 vs 1.5), trailed by NH Whites (0.8 vs 0.6) and Hispanics (0.5 vs 0.6). Regionally, AAMRs from 1999 to 2020 were 0.9 in the Northeast, Midwest, and South, and 0.8 in the West. Moreover, both urban and rural AAMRs were 0.9. Among age groups, <44 years mortality declined, with the steepest drop in 15–44 years, while mortality for >45 years plateaued. Although AAOCA mortality declined overall, especially in children and young adults, it remains high among men, NH Blacks, and Midwest residents. Its recent rebound underscores the need for early detection, risk stratification, and surveillance for timely interventions.

ABSTRACT Background Shape‐sensing robotic‐assisted bronchoscopy (ssRAB) is a navigation platform for biopsy of indeterminate pulmonary lesions. Large‐scale, real‐world evidence confirming the safety profile of ssRAB compared to transthoracic needle biopsy (TTNB) is needed. Study Design and Methods A retrospective cohort study was performed using the PINC AI healthcare database among patients who underwent ssRAB or TTNB lung lesion biopsy at participating hospitals between April 2019 and March 2023. Outcomes were rates of pneumothorax and pneumothorax requiring chest‐tube intervention within 3 days, and rates of in‐hospital bleeding or all‐cause death. Quasi‐binomial logistic regression analysis was performed after one‐to‐five propensity score matching (PSM) accounting for patient‐ and hospital‐related characteristics. Results A total of 119 424 patients (5121 ssRAB, 114 303 TTNB) were identified with 4554 ssRAB and 14 319 TTNB patients after PSM. Relative to ssRAB, TTNB had significantly higher risk of pneumothorax (18.4% vs. 2.6%, OR = 7.10, p < 0.001) and pneumothorax requiring chest‐tube (10.8% vs. 1.4%, OR = 7.62, p < 0.001). TTNB was associated with a higher risk for bleeding (1.5% vs. 0.6%, OR = 2.20, p < 0.001) and all‐cause death (0.48% vs. 0.15%, OR = 2.47, p = 0.023); however, rates for both outcomes were relatively low. Interpretation In this large‐scale, real‐world database analysis with diverse patient populations, physician experience, and health care settings, ssRAB demonstrated a better safety profile compared to TTNB. Superior safety combined with a potentially comparable performance profile and known advantages of bronchoscopy, including concurrent staging, support ssRAB as an optimal choice for non‐surgical biopsies for suspicious pulmonary lesions.