Heart failure with preserved ejection fraction (HFpEF) accounts for about half of heart failure cases and is linked to aging, obesity, diabetes, and multimorbidity, yet disease-modifying therapies remain limited. A major barrier is heterogeneity: HFpEF comprises overlapping inflammatory, fibrotic, cardiometabolic, and hemodynamic/vascular endophenotypes embedded within systemic cardiorenal and cardiohepatic cross-talk, which conventional metrics such as left ventricular ejection fraction (LVEF), natriuretic peptides (NPs), and standard imaging capture incompletely. In this narrative review, we synthesize clinical, mechanistic, and trial data to describe HFpEF endophenotypes and their multi-organ interactions; critically appraise why traditional diagnostic and enrollment strategies contributed to neutral outcomes in landmark trials; and survey emerging cardiovascular multi-omics studies. We then outline an integrative systems-biology framework that applies (i) within-layer analyses and cross-layer integration, (ii) network-based driver nomination and biomarker discovery, and (iii) target nomination to link molecular programs with circulating markers and candidate therapies. Finally, we discuss practical challenges in implementing multi-omics HFpEF research and highlight future directions such as artificial intelligence (AI)-enabled multi-omics integration, cross-organ profiling, and biomarker-guided, endotype-enriched platform trials. Collectively, these advances position HFpEF as a proving ground for precision cardiology, in which therapies are matched to molecularly defined disease programs rather than ejection-fraction cutoffs alone.

To examine whether socioeconomic differences in Medicare Part D glaucoma medication spending are more strongly associated with price exposure (spending per claim) than with utilization intensity. Ecologic observational study using state-year-level data. Medicare Part D beneficiaries receiving glaucoma medications across 50 states and the District of Columbia from 2019 through 2023 (255 state-years). We analyzed Centers for Medicare & Medicaid Services Medicare Part D Prescribers-by-Geography and Drug public use files. State-level socioeconomic context was measured using Area Deprivation Index (ADI) national rank and median household income (sensitivity analysis). Outcomes included inflation-adjusted spending per beneficiary and per claim, as well as utilization measures (claims and 30-day fills per beneficiary) and brand spending and claims shares. Multivariable linear regression models included year fixed effects and state-clustered standard errors. Spending per beneficiary and spending per claim. Median spending per beneficiary was $403.81 (interquartile range [IQR], $359.74-$444.73), and median spending per claim was $104.22 (IQR, $89.49-$118.41). Higher median household income was associated with greater spending per beneficiary ($18.18 per $10,000; 95% CI, $1.30-$35.06) and per claim ($6.99; 95% CI, $1.95-$12.03). Higher ADI was associated with lower spending per claim, although estimates were imprecise (-$3.69 per 10-point increase; 95% CI, -$7.51 to $0.13). Utilization intensity varied minimally across socioeconomic strata (claims per beneficiary, 3.8-4.0), and neither ADI nor income was consistently associated with brand spending or claims share. State-level socioeconomic context was associated with modest differences in glaucoma medication spending per claim, with minimal differences in utilization intensity. These findings suggest that socioeconomic variation in Medicare Part D glaucoma drug spending is more consistent with differences in price exposure than with differences in volume of use.

Hollow viscus perforation following spine surgery is uncommon but associated with substantial morbidity and mortality when diagnosis is delayed. Early recognition is challenging because presenting symptoms frequently overlap with expected postoperative bowel dysfunction, and imaging findings may initially be subtle. Despite its clinical significance, the literature remains fragmented across case reports and small series, and a consolidated, mechanism-based framework specific to spine surgery is lacking. We performed a single-institution case series of five patients who developed hollow viscus perforation or significant postoperative intra-abdominal pathology after spine surgery, combined with a comprehensive review of the published literature. Data were analyzed to identify common mechanisms of injury, clinical presentation patterns, diagnostic pathways, management strategies, and preventive considerations. Hollow viscus perforation after spine surgery occurred through four principal mechanisms: direct mechanical injury, functional distension related to acute colonic pseudo-obstruction physiology, delayed hardware or graft migration, and medication-associated ulceration or diverticular perforation. Early clinical findings frequently overlapped with expected postoperative bowel dysfunction, contributing to delayed recognition. Diagnosis depended on identifying deviations from expected recovery trajectories and obtaining timely computed tomography imaging. Preventive strategies included recognition of preoperative abdominal risk factors, meticulous exposure-specific surgical technique, optimization of postoperative bowel function, and judicious use of medications that impair gastrointestinal motility or tissue integrity. Hollow viscus perforation following spine surgery is an uncommon but potentially catastrophic complication arising from distinct mechanical, functional, hardware-related, and medication-associated pathways. Because early signs are often nonspecific, a low threshold for imaging and general surgical evaluation is essential. Improved preoperative risk stratification, intraoperative vigilance, and structured postoperative monitoring may reduce delays in diagnosis and improve outcomes in this high-risk population.

PURPOSE: Oncology biosimilars have been promoted to reduce Medicare Part B drug spending, yet adoption may vary across socioeconomic and geographic contexts. We quantified national trends in bevacizumab biosimilar adoption from 2019 through 2023 and examined variation by neighborhood deprivation and rurality. METHODS: We conducted a repeated cross-sectional analysis using Medicare Part B Provider and Service Public Use Files from 2019 through 2023. Provider-drug-year observations were linked to zip-code‑level Area Deprivation Index (ADI) quintiles and Rural-Urban Commuting Area classifications. Outcomes included biosimilar market share over time, variation by ADI and rurality, mean Medicare allowed amount per administration, and estimated per-administration savings. RESULTS: The analytic sample included 11,803 provider-drug-year observations, 93.0% located in urban zip codes. Biosimilar adoption increased from 0.5% in 2019 to 31.9% in 2020, exceeded 50% by 2021 (57.5%), and remained more than 60% through 2023 (61.3%). In 2023, uptake ranged from 58.1% in the least deprived quintile (ADI Q1) to 81.3% in the most deprived quintile (ADI Q5) and was lower in nonurban than urban areas (33.6% v 62.0%). Mean allowed amounts for biosimilars declined from $69.70 in US dollars (USD) in 2019 to $33.36 USD in 2023, while originator amounts remained stable. Per-administration savings reached $36.40 USD (52.2%) by 2023 and were similar across ADI quintiles. CONCLUSION: From 2019 to 2023, bevacizumab biosimilars diffused rapidly in Medicare Part B, generating an estimated $84.5 million USD in aggregate savings, with per-administration savings exceeding 50% by 2023. However, adoption remained heterogenous across socioeconomic and rural-urban contexts, suggesting that variation in realized savings is driven primarily by differential diffusion rather than systematic pricing differences. Closing these gaps will likely require coordinated interventions addressing delivery, reimbursement, and clinician-level barriers to biosimilar adoption.

Oncology knowledge spans structured and unstructured datasets of drugs, regimens, conditions, and identifiers, as well as clinician-authored narratives describing treatment sequencing across disease settings. Answering clinical questions may require accessing both databases and free-text narratives, but existing intelligent retrieval systems favor either relational facts or narrative context. We developed HemOncAgent, an artificial intelligence agent that selects among retrieval tools for the HemOnc knowledge ecosystem: HemOncKB, a curated knowledge graph for pharmacologic relationships, and HemOnc.org, a narrative resource for care pathways. Across structured and narrative benchmarks, HemOncAgent maintained high-fidelity retrieval across question types for which single-source systems showed domain-specific limitations, supporting hybrid tool-based retrieval for more reliable oncology knowledge access.