Citations

Abstract Background Clinical nutrition (CN) is becoming increasingly complex because of the rising prevalence of chronic illness, cancer, and malnutrition‐related conditions such as short bowel syndrome and refeeding syndrome. Despite its clinical significance, nutrition education among US physicians remains limited. Simultaneously, large language model (LLM)–based artificial intelligence assistants (AIAs) have emerged as tools to support complex clinical decision‐making but remain largely untested in CN. Methods This retrospective study evaluated four LLM‐based AIAs—ChatGPT (OpenAI), OpenEvidence (OpenEvidence Inc), Gemini (Google, Google DeepMind), and Copilot (Microsoft Corporation)—using five complex CN cases from our nutrition support service. Each AIA was queried with patient‐specific CN questions. Responses were blinded and reviewed by five physician CN experts using an eight‐item assessment tool evaluating clarity, relevance, evidence, and clinical utility. Results All AIAs produced clinically appropriate responses, with Gemini scoring highest in relevance (4.04) and clarity (4.16). Overall satisfaction scores ranged from 3.08 (Copilot) to 3.84 (Gemini). Citation quality and originality of insights varied and were generally limited, and no consistent differences in performance were observed across the five cases among the four AIAs. Conclusion LLM‐based AIAs can reliably replicate expert reasoning in CN. Although not yet a source of novel clinical insights, the true potential of this approach may lie in its application among physicians without specialized expertise in CN, helping to bridge existing knowledge gaps in nutrition care. Presenting full clinical cases, as shown in this study, could support AIA‐enabled e‐consultation in the future, thereby addressing gaps in CN education.

BACKGROUND: Although chronic obstructive pulmonary disease (COPD) and heart failure with preserved ejection fraction often coexist with overlapping clinical features, they are usually studied separately. The SPIROMICS HF (Subpopulations and Intermediate Outcome Measures in COPD and Heart Failure Study) is testing hypotheses that new computed tomography emphysema subtypes are associated with specific cardiovascular phenotypes (eg, cor pulmonale , cor pulmonale parvus ), common airway branch variants are associated with right heart dysfunction, and symptomatic tobacco-exposed persons with preserved spirometry have signs of increased left ventricular afterload. METHODS: SPIROMICS is a multicenter observational study of COPD with extensive pulmonary phenotyping of participants with ≥20 pack-years smoking and nonsmoking controls. COPD and COPD severity were defined by standard spirometric criteria and symptomatic tobacco-exposed persons with preserved spirometry by ≥20 pack-years, normal spirometry, and COPD Assessment Test score >10. SPIROMICS HF selected all participants in SPIROMICS visit 5 at major sites. Its comprehensive speckle-tracking echocardiography, which included physiological perturbations of leg raise and low-intensity exercise, was harmonized prospectively with the Multi-Ethnic Study of Atherosclerosis Early Heart Failure and HeartSHARE (Combining Omics, Deep Phenotyping, and Electronic Health Records for Heart Failure Subtypes and Treatment Targets) studies. The cardiopulmonary magnetic resonance imaging protocol with gadolinium administration included myocardial fibrosis sequences, pulmonary angiography, time-resolved 3-dimensional cine magnetic resonance imaging (4-dimensional flow) of venous return, and metronome-paced tachypnea to induce dynamic hyperinflation. Coronary artery calcium was assessed on computed tomography scans. The Kansas City Cardiomyopathy Questionnaire was administered. RESULTS: Of the final sample of 753 participants, 57% had COPD (15% mild, 27% moderate, and 15% severe), 18% had symptomatic tobacco-exposed persons with preserved spirometry, 16% were smoking controls, and 8% were nonsmoking controls. Reproducibility of the main measures from speckle-tracking echocardiography (intraclass correlation coefficient, 0.83–0.99), exercise echocardiography (intraclass correlation coefficient, 0.71–0.99) and magnetic resonance imaging (intraclass correlation coefficient, 0.57–0.99) were good-to-excellent, including in severe COPD. CONCLUSIONS: SPIROMICS HF aims to characterize and understand cardiopulmonary interactions in COPD and COPD-related phenotypes to inform targeted treatments for combined cardiopulmonary failure.

ABSTRACT Background Early catheter ablation for atrial fibrillation (AF) improves outcomes, but the optimal timing, especially in heart failure (HF) subtypes, remains unclear. Objective To evaluate the impact of ablation timing on outcomes in patients with HF with reduced ejection fraction (HFrEF) and HF with preserved ejection fraction (HFpEF). Methods We used the TriNetX database (2010–2024) to identify HF patients with AF who underwent ablation. Patients were classified as early ablation (≤ 1 year of first AF diagnosis) or delayed ablation (1–3 years), with propensity score matching (PSM) applied to balance the baseline characteristics. The primary outcome was AF recurrence, defined by repeat ablation, direct current cardioversion (DCCV), or antiarrhythmic drug (AAD) use after a 3‐month blanking period. Secondary outcomes included stroke, hospitalization, and mortality over 5 years. Results After PSM, delayed ablation was associated with higher AF recurrence in both HFrEF (HR = 1.28, p < 0.001) and HFpEF (HR = 1.21, p = 0.02), as seen in multivariate analysis. In HFrEF, delayed ablation was associated with higher incidence of re‐do ablation, AAD use, and DCCV, while in HFpEF, higher recurrence with delayed ablation was associated with increased AAD use. Both cohorts had similar stroke and mortality rates between early and delayed ablation. Conclusions Delayed ablation in HFrEF and HFpEF was linked to higher AF recurrence, driven by increased redo ablations, DCCV, AAD use in HFrEF, and higher AAD use in HFpEF. Stroke, HF hospitalization and mortality rates were similar, suggesting earlier ablation may enhance rhythm control over other outcomes.

Background The Woven EndoBridge (WEB) device represents an innovative solution for cerebral aneurysm occlusion, particularly for challenging wide-neck bifurcation aneurysms. However, factors affecting sustained occlusion remain poorly understood. We utilized machine learning to attempt to identify predictors of favorable long-term outcomes following WEB treatment. Methods In this multicenter retrospective study, we collected patient demographics, aneurysm characteristics, procedural details, and clinical outcomes. The primary endpoint was improvement in occlusion status, defined as maintained Raymond-Roy Occlusion Classification (RROC) grade 1, or improvement from grade 2 to 1, or from grade 3 to either 2 or 1 on final angiographic follow up. The dataset was split into training (75%) and validation (25%) sets. The CatBoost algorithm was selected based on performance metrics, with Shapley Additive exPlanations (SHAP) values calculated to determine feature importance. Furthermore, a multivariable binomial logistic regression model was performed to validate machine learning findings. Results Among 720 aneurysms from 36 hospitals, 84% showed improvement in occlusion at follow up. Both machine learning and multivariable logistic regression identified aneurysm height as the most consistent correlate of nonimprovement (odds ratio (OR) 0.90 per mm, p = 0.022). In the CatBoost model, the highest-ranking features by SHAP included aneurysm height, patient age, treatment acuity, ACom location, WEB-SLS device, bifurcation anatomy, aneurysm multiplicity, baseline modified Rankin Scale, access route, and partial thrombosis. Conclusions Machine-learning and regression analyses identified consistent predictors of occlusion improvement after WEB treatment, with aneurysm height most strongly linked to nonimprovement. These insights may guide patient selection and follow up. Findings require cautious interpretation and external validation in larger cohorts.

Importance Validation of prognostic tools is an essential component of their integration into clinical practice. Objective To validate 4 delirium risk stratification tools in an independent emergency department (ED) cohort. Design, Setting, and Participants This prognostic study included a retrospective cohort of patients 65 years or older who presented to an academic ED between January 1, 2021, and December 31, 2024. Data were analyzed from February 10 to August 21, 2025. Exposure Delirium screening included the Delirium Triage Screen followed by the brief Confusional Assessment Method as part of the standard of care. Four delirium risk scores were obtained, including the Kennedy rule, the Zucchelli rule, the Mayo Delirium Prediction (MDP) tools, and the Recognizing Delirium in Emergency Medicine (REDEEM) score. Main Outcomes and Measures Main outcomes included risk score performances and diagnostic test accuracy measured using the area under the receiver operating characteristics curve (AUROC), sensitivity, specificity, negative (NPV) and positive (PPV) predictive values, and positive and negative likelihood ratios (LRs) with 95% CIs. Secondary outcomes were discriminative and calibration measures. Results A total of 44 578 patients were included, of whom 1701 (3.8%) were diagnosed with delirium. The median age was 80.0 (IQR, 75.0-85.0) years, and 22 786 patients (51.1%) were female. The Kennedy rule had an AUROC of 0.777 (95% CI, 0.766-0.789). At a cutoff of 5, sensitivity was 0.55 (95% CI, 0.52-0.57); specificity, 0.85 (95% CI, 0.84-0.85); positive LR, 3.54 (95% CI, 3.37-3.72); negative LR, 0.54 (95% CI, 0.51-0.57); PPV, 0.12 (95% CI, 0.12-0.13); and NPV, 0.98 (95% CI, 0.98-0.98). The Zucchelli tool had an AUROC of 0.701 (95% CI, 0.686-0.713). At a cutoff of 5, sensitivity was 0.68 (95% CI, 0.66-0.70); specificity, 0.66 (95% CI, 0.65-0.66); positive LR, 1.99 (95% CI, 1.92-2.06); negative LR, 0.49 (95% CI, 0.45-0.52); PPV, 0.07 (95% CI, 0.07-0.08); and NPV, 0.98 (95% CI, 0.98-0.98). The MDP tool had an AUROC of 0.898 (95% CI, 0.891-0.905). At a 30% cutoff, sensitivity was 0.51 (95% CI, 0.48-0.53); specificity, 0.95 (95% CI, 0.95-0.95); positive LR, 9.69 (95% CI, 9.11-10.31); negative LR, 0.52 (95% CI, 0.50-0.55); PPV, 0.28 (95% CI, 0.26-0.29); and NPV, 0.98 (95% CI, 0.98-0.98). The REDEEM score demonstrated the highest AUROC at 0.921 (95% CI, 0.914-0.929). Using a cutoff of 11 or greater, sensitivity was 0.83 (95% CI, 0.81-0.85); specificity, 0.92 (95% CI, 0.91-0.92), negative LR, 0.18 (95% CI, 0.17-0.20); positive LR, 9.91 (95% CI, 9.54-10.29); PPV, 0.28 (95% CI, 0.27-0.29); and NPV, 0.99 (95% CI, 0.99-0.99). Conclusions and Relevance In this prognostic study comparing 4 delirium tools to improve delirium detection and identify high-risk patients, the REDEEM score and the MDP tool had better performance for delirium detection in the ED. These findings lay the groundwork for integrating validated risk stratification into ED workflows to improve early delirium detection and inform prevention strategies for high-risk older adults.

Importance Primary brain and central nervous system cancer (collectively referred to as CNS cancer) comprises 2% of all human cancers and poses significant health and economic challenges in the United States. Objective To analyze CNS cancer burden in the US, stratified by time, location (state and division), sex, age group, and Sociodemographic Index (SDI). Design, Setting, and Participants This cross-sectional study involved a repeated analysis of Global Burden of Disease Study (GBD) 2021 data in 2024. Using data from 183 sources, CNS cancer metrics in the US were estimated across states and years. US CNS cancer metrics across all sexes and age groups were included in the GBD. Exposure CNS cancer diagnosis. Main Outcomes and Measures Overall and age-standardized estimates of the incidence, prevalence, mortality, disability-adjusted life-years (DALYs), years of life lost, and years lived with disability per 100 000 population, including 95% uncertainty intervals (UIs), and time trends. Results In 2021, for all age groups and sexes across the US, there were 31 780 incident cases (95% UI, 29971.1 to 32843.9). Age-standardized incidence, DALYs, and mortality rates per 100 000 population were 6.91 (95% UI, 6.58 to 7.12), 134.38 (95% UI, 129.83 to 137.95), and 4.1 (95% UI, 3.87 to 4.22), respectively. Despite no significant change observed in the overall incidence between 1990 and 2021, DALY and mortality rates decreased by 15.77% (95% UI, −17.75% to −13.68%) and 8.41% (95% UI, −11.09% to −6.22%), respectively. Substantial geographic variability was noted. Mississippi, Alabama, Kentucky, and Kansas (West North Central and East South Central divisions) and West Virginia faced persistently high burdens over the past 30 years. Sex differences were evident; disease burden was consistently higher in males compared with females. Age-specific estimates showed a bimodal distribution: the youngest group (&amp;lt;5 years) showed a significant decrease in incidence rate (−34.42% to −11.56%), whereas older age groups (&amp;gt;70 years) experienced increasing trends. DALYs and mortality rates were negatively correlated with SDI (ρ = −0.6860 and ρ = −0.6391; P &amp;lt; .001). Conclusions and Relevance These findings provide valuable insights into the CNS cancer burden across the US by age, sex, location, and SDI, enabling better public health status assessments, health care policy restructuring, and resource redistribution for improved care.

Background: Breast cancer polygenic risk scores (PRS) and traditional risk models (e.g., the Gail model [Gail]) are known to contribute largely independent information, but it is unclear how the overlap varies by ancestry, age, disease type (invasive breast cancer, DCIS), and risk threshold. Methods: In a retrospective case–control study, we evaluated risk prediction performance in 180,398 women (161,849 of European ancestry; 18,549 of Asian ancestry). Odds ratios (ORs) from logistic regression models and the area under the receiver operating characteristic curve (AUC) were estimated. Results: PRS for invasive disease showed a stronger association in younger (<50 years) women (OR = 2.51, AUC = 0.622) than in women ≥ 50 years (OR = 2.06, AUC = 0.653) of European ancestry. PRS performance in Asians was lower (OR range = 1.62–1.64, AUC = 0.551–0.600). Gail performance was modest across groups and poor in younger Asian women (OR = 0.94–0.99, AUC = 0.523–0.533). Age interactions were observed for both PRS (p < 0.001) and Gail (p < 0.001) in Europeans, whereas in Asians, age interaction was observed only for Gail (invasive: p < 0.001; DCIS: p = 0.002). PRS identified more high-risk individuals than Gail in Asian populations, especially ≥50 years, while Gail identified more in Europeans. Overlap between PRS, Gail, and family history was limited at higher thresholds. Calibration analysis, comparing empirical and model-based ROC curves, showed divergence for both PRS and Gail (p < 0.001), which indicates miscalibration. In Europeans, family history and prior biopsies drove Gail discrimination. In younger Asians, age at first live birth was influential. Conclusions: PRS adds value to risk stratification beyond traditional tools, especially in younger women and Asian ancestry populations.

Background: Post-treatment prognosis and monitoring are critical for determining the timing of salvage treatment in glioblastoma patients but has been challenging due to difficulties differentiating progression from treatment effects in conventional images. This exploratory study aimed to establish the correlation of radiomics image features from time series of amino acid tracer 18F-DOPA PET images, with outcomes, using machine learning and dimension reduction analysis. Methods: 18F-DOPA PET images were collected for a patient cohort with wild-type IDH and unmethylated MGMT who underwent dose-escalated radiation therapy. Quantitative features were derived from the high uptake region (T/N > 2.0) in pre- and post-radiation therapy follow-up 18F-DOPA PET images. A customized workflow was utilized for pre-selecting predictive features, followed by manifold learning. Machine learning algorithms were employed to establish associations between imaging features and remaining survival (RS), defined as the time between a follow-up scan and date of death. Results: The ML models exhibited 81–83% ROC_AUC in predicting RS evaluated on an independent test dataset. A RS map is proposed for monitoring tumor alterations through serial 18F-DOPA PET scans, demonstrating superior sensitivity and better correlation with survival compared to the RANO criteria. Conclusions: Our study demonstrates that ML models utilizing FU 18F-DOPA PET images have the potential to effectively predict future survival outcomes in patients with glioblastoma treated with dose-escalated radiation therapy. The capability to assess changes in tumor over time through imaging can potentially assist in patient stratification and the selection of salvage treatments, while also aiding in distinguishing treatment effects from genuine tumor progression.

Background: Atrial fibrillation (AF) significantly impacts quality of life (QoL) and is often observed in patients with transthyretin amyloid cardiomyopathy (ATTR-CM). Acoramidis achieves near-complete (≥90%) transthyretin (TTR) stabilization and is approved in the USA, the UK, Europe and Japan for the treatment of ATTR-CM in adults. In the phase 3 ATTRibute-CM study (NCT03860935), acoramidis treatment significantly reduced the time to all-cause mortality (ACM) or first cardiovascular-related hospitalization (CVH) at Month 30, and resulted in improved 6-minute walk distance (6MWD), Kansas City Cardiomyopathy Questionnaire-Overall Summary (KCCQ-OS) scores and N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels relative to placebo. The clinical efficacy of acoramidis has not yet been reported in patients with ATTR-CM and with an AF or atrial flutter (AFL) diagnosis at baseline. Research Question: How does acoramidis treatment affect ACM or first CVH, 6MWD, KCCQ-OS scores, and NT-proBNP levels at Month 30 in participants with ATTR-CM with or without AF/AFL diagnosis at baseline? Methods: The ATTRibute-CM study design has been previously published. This post hoc analysis included participants in the modified intention-to-treat population (N=611) grouped by AF/AFL diagnosis at baseline (defined as recorded AF medical history or the presence of AF or AFL on an ECG at enrollment). The rate of ACM or first CVH at Month 30 was compared between groups using a stratified Cox proportional hazards model. Changes from baseline to Month 30 in 6MWD, KCCQ-OS scores and NT-proBNP levels were summarized descriptively. Results: Overall, 62.8% (384/611) of participants had an AF/AFL diagnosis at baseline (acoramidis: 255/409, placebo: 129/202). These participants had lower mean 6MWD and KCCQ-OS scores and higher median NT-proBNP levels than those without AF/AFL diagnosis at baseline. Treatment with acoramidis reduced ACM or first CVH, slowed the decline in 6MWD and KCCQ-OS scores and blunted the rise in NT-proBNP compared with placebo, regardless of AF/AFL diagnosis at baseline ( Table ). Conclusions: Acoramidis improved clinical outcomes (ACM, CVH), functional status, QoL and NT-proBNP levels relative to placebo in participants with ATTR-CM, regardless of AF/AFL diagnosis at baseline.