Citations

Background/Objectives: Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have emerged as a promising therapeutic option for women with polycystic ovary syndrome (PCOS). Clinical trials and data demonstrate significant improvements in body weight, insulin resistance, hyperandrogenism, and, in some studies, menstrual regularity and reproductive outcomes. However, the current evidence base remains disproportionately focused on short-term metabolic and anthropometric endpoints, while outcomes most relevant to patients—including treatment satisfaction, gastrointestinal tolerability, psychological well-being, body image, fertility planning, financial burden, and long-term adherence—remain inadequately characterized. Methods: We searched PubMed and Google Scholar for English-language literature published between 2021 and 2026, using structured keyword searches. This narrative review evaluates the literature on GLP-1RA use in PCOS beyond weight-focused outcomes, examining mechanisms of action, comparative tolerability across agents, and patient-centered domains including quality of life, eating behavior, and reproductive planning. Results: The published literature reveals a substantial gap between trial-reported completion rates and treatment success, and actual persistence and patient satisfaction. This discrepancy is particularly consequential in PCOS, where treatment decisions intersect with high baseline rates of depression, anxiety, and body image disturbance, and where ovulatory benefits must be weighed against mandatory discontinuation before conception. Conclusions: We identify a systematic absence of validated patient-reported outcome measures, standardized core outcome sets, and long-term follow-up data after treatment withdrawal. This review proposes a research agenda prioritizing patient-centered endpoints as co-primary outcomes, structured assessment of acceptability and persistence, qualitative and mixed-methods approaches, and a dedicated study of fertility planning and treatment interruption.

Background: Reliable access to postoperative guidance between clinic visits remains inconsistent, and the usefulness of large language models depends on how clinical knowledge is structured and retrieved. Knowledge graph-only (KG-only), vector retrieval-augmented generation (RAG-only), and hybrid graph-retrieval (GraphRAG) architectures have not been directly compared under controlled conditions for postoperative care. Objective: It was to compare these three pipelines on a curated postoperative question-answering benchmark using automated grounding metrics and blinded clinician review. Methods: A benchmark of 250 expert-authored questions (200 in-scope, 20 emergency-intent, 30 out-of-scope) was evaluated on an identical knowledge corpus using Gemini 2.5 Flash for generation and Gemini Embedding-001 for retrieval. Three blinded clinicians rated all 750 outputs; the primary outcome was majority-consensus accuracy across all 250 questions. GraphRAG is a multi-component pipeline that also includes query expansion, a second retrieval stage, evidence fusion, and a monotonic RAG fallback, and is therefore evaluated as a complete hybrid system rather than as graph retrieval in isolation. Results: Overall accuracy was 96.8% (95% CI: 93.8–98.4) for GraphRAG, 94.8% (95% CI: 91.3–96.9) for RAG-only, and 56.0% (95% CI: 49.8–62.0) for KG-only. Both retrieval-augmented systems outperformed KG-only (adjusted p < 0.001). The GraphRAG versus RAG-only difference was 2.0 percentage points, arose from five discordant question pairs and did not reach the Bonferroni-adjusted threshold (exact McNemar p = 0.062; adjusted p = 0.188). The monotonic RAG fallback determined 44.8% of GraphRAG outputs (112 of 250 queries). GraphRAG showed higher context recall (0.851 versus 0.796; adjusted p < 0.001), whereas RAG-only showed higher context precision. Estimated hallucination risk, derived from an LLM-as-a-judge procedure using Gemini 2.5 Flash rather than independent clinical adjudication, was 0.108, 0.132, and 0.436, respectively; the GraphRAG versus RAG-only difference was not significant (adjusted p = 0.771). Inter-rater agreement was high (Fleiss’ kappa = 0.854). Conclusions: Retrieval augmentation, rather than the choice of retrieval architecture, accounted for most of the difference observed. The incremental advantage of GraphRAG over a well-implemented RAG-only system was small, was not statistically significant after correction and reflects the complete hybrid system rather than graph augmentation alone. Emergency-intent accuracy was 75.0% for both retrieval-augmented systems; the architecture is therefore suited to supervised clinical workflows with clinician review, not autonomous patient-facing triage. Prospective validation on authentic patient messages and formal component ablations are required.

Abstract The 6th Edition of the World Health Organization Classification of Digestive System Tumours has introduced two new categories of gastric neuroendocrine tumors, reflecting recent data on subtypes that do not conform to the three established subtypes: type 1 (associated with autoimmune gastritis), type 2 (associated with Zollinger-Ellison Syndrome) and type 3 (sporadic). The new type 4 category is exceptionally rare and is caused by endogenous dysfunction of parietal cell proton pumps, potentially secondary to mutation of ATP4A. Type 5 gastric NETs are caused by chronic pharmacologic inhibition of the parietal cell proton pump, typically due to proton pump inhibitor use, and are increasingly common. Given the relatively low malignant potential of the new subtypes compared to sporadic (type 3) gastric NETs, it is important that clinicians and pathologists recognize these entities, limiting radical surgical resections when unnecessary.

Background Recurrent pericarditis may be mediated through inappropriate activation of the NACHT, leucine‐rich repeat, and pyrin domain‐containing protein 3 inflammasome. Pharmaceutically manufactured cannabidiol is an oral formulation that inhibits NACHT, leucine‐rich repeat, and pyrin domain‐containing protein 3 inflammasome activation and development of pericarditis in a preclinical model. Methods This Phase 2, open‐label, multicenter trial included adult patients with symptomatic recurrent pericarditis: pericarditis chest pain ≥4 on an 11‐point numerical rating scale with an elevated C‐reactive protein (≥1 mg/dL) or at least mild pericardial inflammation on cardiac magnetic resonance imaging. At enrollment, patients were on stable doses of non‐steroidal anti‐inflammatory drugs, colchicine, and/or corticosteroids. Cannabidiol was administered during an 8‐week treatment period and an optional 18‐week extension period when background therapy was tapered and discontinued. The primary efficacy end point was change in patient‐reported pericarditis pain. Secondary end points included normalization of C‐reactive protein and percentage of patients with recurrences. Results Among 27 patients (18 women, mean age 53 years), baseline maximal numerical rating scale pain score was 5.8±1.7, and 10 patients had elevated C‐reactive protein. At Week 8, maximal numerical rating scale pain score was 2.1±1.8, and median time to numerical rating scale score ≤2 was 5 days. Among the 10 patients with baseline C‐reactive protein elevation, C‐reactive protein was normal at Week 8 in 8 patients (80%). During the extension period, 17 (71%) of 24 patients remained free of pericarditis recurrence. A serious adverse event leading to cannabidiol discontinuation occurred in 1 patient. Conclusions In symptomatic patients with recurrent pericarditis, pharmaceutically manufactured cannabidiol appears safe and associated with reductions in pericarditis pain and systemic inflammation. Registration URL: clinicaltrials.gov ; Unique Identifier: NCT05494788.

Primary enlargement of the cerebrospinal fluid (CSF)–filled brain ventricles, known as congenital cerebral ventriculomegaly (CCV), is a hallmark of congenital hydrocephalus. CCV is also enigmatically but frequently associated with autism and other neurodevelopmental disorders. To gain insight into the developmental genetic regulation of the human CSF-ventricular system, we conducted an integrated, multiomic study of about 2700 trio-based exomes from patients with primary CCV. We found that about 25% of cases were associated with rare, damaging de novo variants in mutation-intolerant genes, many of which are linked to other dominant Mendelian disorders. Thirty-five exome-wide significant CCV genes and dozens of other high-confidence CCV genes converged on pathways involved in ATP-dependent Brahma-related gene 1/Brahma-associated factor chromatin remodeling, histone H3 lysine 4 methylation, and phosphoinositide 3-kinase signaling. Knockout of selected CCV genes in mouse models supported that de novo variants in CCV genes caused ventriculomegaly by impairing both CSF dynamics and cortical cytoarchitecture through dysregulation of neuroprogenitor cell growth and maturation in the ventricular and subventricular zones. These findings indicated that genetic and epigenetic programs coordinate the “hand-in-glove” development of the CSF-ventricular system with that of the cerebral cortex and establish a genetic connection between CCV and neurodevelopmental disorders, potentially explaining why some patients with hydrocephalus continue to exhibit CCV and neurodevelopmental disorders despite CSF shunting. We suggest that combined brain imaging and whole-exome sequencing could enable early detection of, and intervention for, autism and other neurodevelopmental disorders.

Abstract Objective This study was undertaken to examine acute and chronic electroencephalographic (EEG) and seizure characteristics in new onset refractory status epilepticus (NORSE). Methods Multicenter inpatient and follow‐up clinical and EEG data were analyzed using International League Against Epilepsy definitions and American Clinical Neurophysiology Society (ACNS) EEG terminology. Results Forty‐four patients were included from the Yale NORSE/FIRES Biorepository (median age = 29 years, 33 female, 38 cryptogenic). Nine hundred twenty inpatient EEG days were reviewed (median = 13.5 days/patient). Presenting status epilepticus (SE) types included 24 convulsive SE (CSE), five focal–motor (FMSE), and 15 nonconvulsive (NCSE). Of 39 patients with post discharge follow‐up (median = 14.7 months), 61.5% ( n = 24) had seizures and required more antiseizure medications than seizure‐free patients (median = 4 vs. 1, p < .001). Inpatient EEG captured epileptiform discharges in all and periodic discharges in 38 patients. Seizures were captured in 38 patients: 33 electrographic (11 only electrographic) and 27 electroclinical. Total seizure burden ranged 6.5–29 615.0 min (median = 50.7). Seizures fulfilling ACNS EEG criteria for SE were captured in 20 patients; two had CSE and the remainder NCSE with coma, nine of whom also had FMSE. Inpatient seizure days (median = 50.0 vs. 15.5%, p = .004) and seizure burden (median = 42.9 vs. 0 min, p = .009) were higher in the first half of monitored days. Interictal findings were equally present and did not predict postdischarge seizures, whereas electroclinical seizures (90.9% vs. 45.5%, p = .037), seizure burden (median = 116.8 vs. 32.5 min, p = .045), and proportion of seizure days (36.3% vs. 20.0%, p = .037) did. Among 12 patients with follow‐up EEGs (median = 19.6 months post onset), posterior‐dominant rhythm (PDR) returned in nine; five had periodic discharges, and three had seizure captured. Significance Seizures are most commonly convulsive upon presentation and nonconvulsive or clinically subtle throughout the inpatient course in NORSE. Inpatient seizures, but not interictal abnormalities, occur earlier in the inpatient course and presence of electroclinical seizures, seizure burden, and proportion of seizure days but not interictal findings are associated with postdischarge seizures. Although return of PDR reflects neurological recovery, seizure persistence and EEG abnormalities post‐NORSE are common.

Abstract Objective Ceribell Inc.'s point‐of‐care electroencephalographic (EEG) system and artificial intelligence‐based Automated Seizure Burden Estimator (ASBE; ClarityPro) have US Food and Drug Administration clearance for diagnosing electrographic status epilepticus (ESE). The AccuRASE study using ASBE version 6 (V6) showed high negative predictive value (NPV) but limited sensitivity and positive predictive value (PPV) at certain thresholds. Version 8 (V8) is the updated algorithm trained with additional EEG samples. Methods We tested V8 on the previously used rapid‐response EEG test dataset (not used to train V8). Sensitivity, specificity, PPV, and NPV were compared against blinded expert annotations. ESE, ESE and possible ESE (ESE/pESE), electrographic seizures (ESz), and Esz with highly epileptiform patterns (Esz/HEPs) at burden thresholds of >1%, >10%, >20%, >50%, and >90% were analyzed. Additionally, we evaluated the built‐in ESE and ESE/pESE alerts (≥90% over 5 min, ≥20% over 1 h, or ≥10 continuous minutes of seizure). Results V8 showed 100% sensitivity at lower thresholds (>10%–20%) for ESE (V6 .86), without a specificity loss (V8 .9 vs. V6 .85), and retained NPV (V8 1.0 vs. V6 .99). Sensitivity at 50% threshold was .86 for ESE (V6 .71), and specificity was .94 (V6 .91); PPV for ESE was 30% for ESE and 80% for ESE/pESE. The specificity for ESE remained high for 90% burden (V8 .96, V6 .97) but the sensitivity dropped (.29 vs. V6 .43). Sensitivity was lower for ESz/HEPs (.54 → .44 at >10% threshold), although the algorithm is not designed to detect HEPs. Built‐in alert analysis showed significantly improved sensitivity for ESE (.57 → .86); specificity and NPV remained high (>.94). Significance ASBE V8 showed meaningful improvements in sensitivity, continued excellent NPV, and improved built‐in alert accuracy for ESE and ESE/pESE. The best seizure burden threshold for ruling out ESE now is 20% (vs. 10% with V6) and that for ruling in ESE/pESE with reasonable certainty is 50% (vs. 90% with V6).

Abstract Background Sleep disorders are increasingly recognized as important risk factors for cardiovascular disease (CVD), yet their contribution to cardiovascular mortality remains insufficiently examined at the population level. This study aimed to evaluate long-term trends in CVD mortality where sleep disorders were identified as contributing causes of death. Methods We analyzed U.S. national mortality data from 1999 to 2020 among individuals aged 15 to 84 years. Cases were selected where CVD (ICD-10: I11.0, I11.9, I20-I25, I48, I49.0, I49.9, I50.0, I50.1, I50.9) was the underlying cause and sleep disorders (ICD-10: G47.0-G47.3) were listed as contributing causes. Joinpoint regression was employed to assess temporal changes in age-adjusted mortality rates (AARs), with standard errors provided in the input data and model selection performed using Weighted BIC. APCs were estimated using the Empirical Quantile Method with 5001 resamples. Results A total of 118,440 deaths were included across 22 annual time points. Joinpoint regression identified two significant inflection points (2005 and 2018). From 1999 to 2005, AARs increased sharply (APC = 14.47%; 95% CI: 11.56-19.54; p = 0.0016), followed by a slower rise from 2005 to 2018 (APC = 6.62%; 95% CI: 4.23-10.83; p &lt; 0.000001), and a renewed acceleration from 2018 to 2020 (APC = 11.54%; 95% CI: 7.01-14.12; p &lt; 0.000001). Males had more than twice the AAR compared to females (3.00 vs. 1.31 per 100,000). Black individuals had the highest burden by race (AAR = 2.55), and geographic variation ranged widely from 0.70 in New York to 3.38 in Wyoming. Conclusions Over two decades, cardiovascular mortality with contributory sleep disorders has risen significantly, with widening demographic and geographic disparities. These findings reinforce that sleep disorders are not merely comorbidities but pivotal, modifiable drivers of cardiovascular morbidity and mortality. Untreated sleep conditions such as insomnia, obstructive sleep apnea, and circadian disruptions exacerbate autonomic dysfunction, inflammation, hypertension, and metabolic dysregulation. Addressing sleep health must become a core component of cardiovascular risk reduction strategies, both in clinical practice and public health policy. Failure to integrate sleep disorder screening, diagnosis, and management into CVD care pathways risks perpetuating a preventable and growing burden of disease. This abstract is funded by: None

Background: Agentic artificial intelligence (AI) systems, characterized by autonomous goal-directed behavior, multi-step reasoning, task decomposition, and tool use, are increasingly proposed for healthcare applications. However, their autonomy raises concerns regarding transparency, accountability, and human oversight. While explainable AI (XAI) has been widely studied in traditional predictive models, less is known about how explainability is implemented within agentic architectures. Objective: To map the emerging literature on explainable agentic AI (XAAI) in healthcare and characterize the types, scope, and forms of explainability used in these systems. Methods: A scoping review was conducted following PRISMA-ScR guidelines. PubMed, Embase, IEEE Xplore, and ACM Digital Library were searched through November 2025. Eligible studies described healthcare-related agentic AI systems incorporating explicit explainability mechanisms. Data were extracted on system architecture, explainability type (intrinsic, post hoc, hybrid), explanation scope (local, global), explanation form, and reported clinical outcomes. Results: Nine studies met the inclusion criteria. All systems demonstrated core agentic features, including autonomy, task decomposition, and tool integration, often within multi-agent frameworks. Explainability was predominantly intrinsic and workflow-native, typically delivered through textual reasoning traces and example-based grounding in retrieved clinical evidence. Feature-based and global explanations were comparatively rare and largely confined to hybrid architectures. Across domains including radiology, neurology, psychiatry, and biomedical research, XAAI systems were reported to improve performance and interpretability relative to baseline models in the included studies. However, these findings were derived from heterogeneous, predominantly experimental or retrospective studies, and structured human-in-the-loop oversight was infrequently described. Conclusions: Current XAAI systems appear to emphasize process transparency and evidence grounding rather than mechanistic model-level attribution. The available evidence remains limited and heterogeneous, and findings should be interpreted as early trends rather than established characteristics. Further progress will require standardized evaluation frameworks, clearer reporting of oversight mechanisms, and validation in real-world clinical settings to support safe and trustworthy integration of agentic AI into healthcare practice.

Importance Seizure-induced brain injury is central to the treatment urgency of new-onset refractory status epilepticus (NORSE). Identifying biomarkers that reflect ongoing neuronal damage could inform therapeutic timing and improve outcomes. Objective To quantify acute brain injury in patients with cryptogenic NORSE (cNORSE), etiology-defined status epilepticus (eSE), and chronic epilepsy. Design, Setting, and Participants This was an international cross-sectional study conducted between 2013 and 2025. Patients were enrolled at 36 hospitals in the US, 2 in Canada, and 1 in Italy, France, and Belgium. Patients with cNORSE and eSE for which biological samples were obtained during ongoing seizure activity were enrolled in the study. Comparison groups without status epilepticus comprised individuals with chronic epilepsy and healthy participants. None were excluded. Exposures Neurofilament light chain (NfL) and S100-beta (S100B) protein concentrations in serum and cerebrospinal fluid (CSF). Main Outcomes and Measures Degree of neuronal and glial damage, indexed by NfL and S100B levels, and their association with short-term functional outcomes. Results A total of 78 patients with cNORSE (mean [95% CI] age, 37 [30-41] years; 44 female [56%]) and 2 independent cohorts of 211 patients (mean [95% CI] age, 69 [66-71] years; 128 female [61%]) and 73 patients (mean [95% CI] age, 56 [45-65] years; 39 male [53%]) with eSE were included. NfL concentrations were markedly elevated in cNORSE—approximately 10-fold higher in CSF and 4-fold higher in serum—compared with the eSE cohorts (CSF: median [IQR], 6408 [1503-22 963] pg/mL compared with 694 [219-2389] pg/mL; serum: median [IQR], 231 [99-855] pg/mL compared with 55 [20-135] pg/mL; P &amp;lt;.001). Serum NfL levels were nearly 20-fold higher in cNORSE than in the cohort with epilepsy and in healthy controls (median [IQR], 11 [7-19 ] and 7 [5-14 ] pg/mL, respectively). Serum and CSF NfL levels were strongly correlated (Spearman ρ = 0.75; P &amp;lt; .001) and rose sharply between week 1 (median [IQR], 101 [51-137] pg/mL), week 2 (median [IQR], 197 [117-324] pg/mL), and week 3 (median [IQR], 598 [163-1000] pg/mL) after onset ( P &amp;lt; .001). In contrast, S100B concentrations did not differ between groups and showed no consistent temporal pattern. NfL discriminated cNORSE from eSE (area under the receiver operating characteristic curve [AUROC], 0.79; 95% CI, 0.68-0.90) and from cohorts without status epilepticus (AUROC, 0.99; 95% CI, 0.78-1.00). Higher serum NfL was independently associated with poor functional outcome at discharge (Glasgow Outcome Scale extended score, 1-4; odds ratio, 1.01; 95% CI, 1.00-1.03; P = .03). Conclusions and Relevance Results of this cross-sectional study suggest that acute neuroaxonal injury, as reflected by elevated NfL levels, was substantially greater in cNORSE than in the cohorts with eSE and in controls without status epilepticus. The rapid early rise in NfL highlights a narrow therapeutic window, emphasizing the need for prompt, effective, and potentially neuroprotective interventions in cNORSE.