Type B aortic dissection (TBAD) is a life-threatening cardiovascular emergency that requires timely recognition. This study aimed to develop and internally evaluate an explainable machine learning model for identifying existing TBAD using routinely available clinical history and admission laboratory data.

This single-center retrospective case-control study included 1,640 participants, comprising 854 patients with CTA-confirmed TBAD and 786 hospitalized controls who underwent whole-aorta CTA and were confirmed not to have aortic dissection. Demographic characteristics, medical history, and admission laboratory test results were collected as 38 initial candidate features. Least absolute shrinkage and selection operator (LASSO) regression was applied to identify key discriminative variables. Machine-learning models based on eight algorithms were then developed and compared: support vector machine (SVM), gradient boosting machine (GBM), neural network, extreme gradient boosting (XGBoost), k-nearest neighbors (KNN), adaptive boosting (AdaBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost). Hyperparameters were tuned in the training set using grid search and repeated 10-fold cross-validation with five repeats. Model discrimination was evaluated in the held-out internal test set using the area under the receiver operating characteristic curve (AUC). The final representative model was selected for exploratory interpretation using SHapley Additive exPlanations (SHAP).

LASSO regression identified five predictors: hypertension, white blood cell count, lymphocyte percentage, basophil percentage, and monocyte count. Several models showed comparable discrimination in the held-out internal test set. The neural network model achieved an AUC of 0.883 in the training set and 0.852 in the held-out internal test set, with 95% confidence intervals of 0.864-0.902 and 0.819-0.885, respectively. Based on the ROC curve analysis, the neural network model showed relatively favorable discrimination and was therefore selected as the final representative model for exploratory SHAP-based interpretation. SHAP analysis indicated that lymphocyte percentage, hypertension, monocyte count, white blood cell count, and basophil percentage were the major contributors to the predictions of the final model, with lymphocyte percentage showing the highest mean absolute SHAP value.

This study developed and internally evaluated an explainable machine-learning model for identifying existing TBAD in a single-center CTA-confirmed retrospective case-control cohort. Given the lack of external validation, this study should be regarded as exploratory. External validation in clinically relevant acute symptomatic populations is required before clinical implementation.

Cerebral ischemia-reperfusion injury (CIRI) represents a critical pathological cascade that paradoxically exacerbates neurological damage following revascularization therapy for acute ischemic stroke (AIS). The pathogenesis of CIRI is intricately linked to dysregulated neuroinflammation, with microglia-the resident innate immune cells of the central nervous system-serving as central orchestrators of this response. Emerging evidence indicates that microglia undergo profound metabolic reprogramming encompassing glucose metabolism, the tricarboxylic acid (TCA) cycle, fatty acid metabolism, and NAD⁺ homeostasis, which fundamentally dictates their functional polarization and consequent neuroinflammatory outcomes. Rather than existing as discrete pro-inflammatory versus reparative phenotypes (classically referred to as M1/M2), microglia exhibit a continuum of activation states with distinct metabolic signatures that evolve dynamically across spatiotemporal dimensions following CIRI. Here, we systematically synthesize current knowledge on the core molecular mechanisms underlying microglial metabolic reprogramming, including the ACOD1/itaconate pathway, the glycolysis-OxPhos balance, and NAMPT-mediated NAD⁺ homeostasis. We critically examine the intricate crosstalk between these metabolic pathways and neuroinflammatory signaling cascades, revealing how metabolic checkpoints serve as integrative nodes that decode microenvironmental cues into functional outputs. Building on this mechanistic foundation, we evaluate emerging intervention strategies targeting metabolic reprogramming, stratified by intervention modality and translational readiness, with emphasis on agents in active clinical development. Finally, we identify prevailing challenges-including spatiotemporal heterogeneity, cell-specific targeting requirements, and clinical translation barriers-and outline future directions integrating single-cell omics, systems biology approaches, and advanced delivery systems. This comprehensive analysis aims to provide a refined conceptual framework and highlight promising therapeutic avenues for mitigating CIRI through strategic modulation of microglial immunometabolism.

The calculation of the left atrioventricular coupling index (LACI) by the ratio between left atrial (LA) and left ventricular (LV) end-diastolic volumes reflects the close interaction between LA and LV.

The study aimed to assess LACI using echocardiography before and after percutaneous coronary intervention (PCI) compared with LV ejection fraction (LV-EF) in acute coronary syndrome (ACS).

This retrospective cohort analysis included 193 ACS patients (65.1 ± 12.5 years, 71.6% males), who underwent PCI. Both LACI and LV-EF were calculated before and after PCI.

At presentation, 156 (80.8%) patients had LV-EF reduction; mild in 61 (39.1%), moderate in 77 (49.3%) and severe in 18 (11.5%). Using LACI < 25% as the cutoff value for normal, 40 (20.7%) patients had impaired LACI. Out of the 153 patients with normal LACI, 127 (83%) had reduced LV-EF. Patients were classified into three groups: I: 51 (26.4%) with STEMI, II: 82 (42.5%) with NSTEMI and III: 60 (31.1%) with unstable angina. Baseline LV-EF and LACI did not show significant differences between the three groups. Over a 12.8 ± 8.2 month follow-up post PCI, LV-EF improved in 68 (43.6%) out of 156 patients becoming normal in 28 (17.9%), mildly reduced in 65 (41.7%), moderately reduced in 55 (35.2%) and severely reduced in 8 (5.1%). Post PCI, LACI normalized in 19 (47.5%) out of 40 patients with impaired baseline LACI.

Assessing LACI in ACS provided a better understanding of LA and LV dynamics and its reduction following PCI indicates a better prognosis.

Hypertensive disorders of pregnancy (HDP) are a leading cause of maternal and perinatal morbidity worldwide, and their pathogenesis involves complex interactions among vascular dysfunction, placental ischemia and immune dysregulation. The gut microbiota has been highlighted as a key upstream modulator, with microbial metabolites serving as key functional mediators rather than microbial composition alone. The present review focused on the roles of short‑chain fatty acids (SCFAs) and tryptophan metabolites in HDP pathogenesis. SCFAs and tryptophan‑derived metabolites modulate vascular function, immune tolerance and placental development via G protein‑coupled receptors, histone deacetylase inhibition and aryl hydrocarbon receptor (AhR) signaling. Their crosstalk underscores integrated regulatory networks at the maternal‑fetal interface. For example, butyrate promotes regulatory T cell differentiation via histone deacetylase inhibition, while indole‑3‑lactic acid activates the AhR; these pathways may synergistically enhance immune tolerance, yet competition for AhR binding between different tryptophan metabolites could produce antagonistic effects. Despite key progress, notable challenges remain regarding causal inference, methodological standardization and translational barriers. The present review aimed to synthesize current mechanistic insights and evaluate the clinical translation potential of targeting microbial metabolites for HDP prevention and management.

To explore the value of handheld simplified venous excess ultrasound (VExUS) grades in bedside venous congestion assessment in general intensive care unit (ICU) patients and to examine its associations with disease severity scores and fluid overload indicators.

This study was a secondary analysis based on data from a single-center prospective observational ICU cohort database. The data were derived from the existing ICU cohort database of the ICU of China-Japan Friendship Hospital. The original study included critically ill patients admitted to the ICU of China-Japan Friendship Hospital from June 14 to August 21 in 2024, all of whom underwent handheld ultrasound and conventional ultrasound VExUS-related examinations. Demographic data, severity of illness scores, laboratory indicators, fluid balance, and supportive treatment information were also collected simultaneously. The main ultrasound variables included the handheld simplified VExUS grading and its key component parameters (inferior vena cava short-to-long axis ratio, hepatic vein S/D ratio, and portal vein pulsatility fraction). Descriptive analysis was used to evaluate the overall distribution of handheld simplified VExUS indicators. Spearman rank correlation analysis was employed to evaluate the relationship between the handheld simplified VExUS grading and its key component parameters with the Acute Physiology And Chronic Health Evaluation II (APACHE II) score, Sequential Organ Failure Assessment (SOFA) score, N-terminal pro-brain natriuretic peptide (NT-proBNP), and fluid balance within 24 hours after ICU admission. The differences in clinical indicators among patients with different handheld simplified VExUS grading were compared, and the primary analysis of continuous ultrasound parameters employed the post-hoc mean values of independent measurements taken by two operators, and sensitivity analysis was conducted separately using the original measurements of the two operators. Multiple comparison correction for the primary correlation analysis was performed using the Benjamini-Hochberg false discovery rate (FDR).

The original study included a total of 80 critically ill patients. After excluding 2 minor patients from this study, a total of 78 adult patients were finally included in the overall analysis. The overall disease severity of the 78 adult ICU patients was moderate, with an APACHE II score of 20.0 (15.3, 25.0) and a SOFA score of 7.0 (4.0, 11.0). The handheld simplified VExUS grading was mainly at grades 0 and 1, with 48 cases (62%) at grade 0, 26 cases (33%) at grade 1, and 4 cases (5%) at grade 2. The average values of the inferior vena cava short-to-long axis ratio, hepatic vein S/D ratio, and portal vein pulsatility fraction were 0.763 (0.668, 0.809), 1.358 (1.252, 1.571), and 0.243 (0.180, 0.303), respectively. In the primary correlation analyses, the mean inferior vena cava short-to-long axis ratio was positively correlated with APACHE II score and SOFA score (r values were 0.302 and 0.252, with unadjusted P values of 0.007 and 0.026, respectively), while portal vein pulsatility fraction was positively correlated with 24-hour fluid balance (r=0.290, unadjusted P value was 0.010). However, no significant correlation was observed between the handheld simplified VExUS grading and either the severity of illness scores or the volume overload indices. After Benjamini-Hochberg FDR correction, the correlation between the mean inferior vena cava short-to-long axis ratio and both the APACHE II score and the SOFA score did not reach statistical significance (FDR-corrected P values were 0.080 and 0.139, respectively). Similarly, the correlation between the mean portal vein pulsatility fraction and 24-hour fluid balance also failed to maintain statistical significance (FDR-corrected P value was 0.080). After stratification by the simplified handheld VExUS grading, there were no statistically significant differences in APACHE II score, SOFA score, NT-proBNP level, and 24-hour fluid balance among patients with different grades (all P>0.05). A sensitivity analysis was conducted using the original measurements from two operators on the core pairs with original correlation signals in the main analysis, and the results were largely consistent with the main analysis. However, the previously observed associations did not remain statistically significant after Benjamini-Hochberg FDR correction.

In general ICU patients, the handheld simplified VExUS grade showed limited associations with disease severity scores and volume overload indicators. Some key component parameters showed exploratory signals related to disease severity or fluid balance in the original correlation analysis, but did not maintain statistical significance after FDR correction.

To investigate the relationship between serum retinol-binding protein 4 (RBP4) and cystatin C (CysC) levels and cardiac function as well as ventricular remodeling in patients with myocardial infarction.

A case control study was conducted. A total of 185 patients with first diagnosed acute myocardial infarction (AMI) who were treated at Harrison International Peace Hospital between December 2022 and January 2025 were enrolled as the research subjects, and 150 healthy subjects who underwent physical examination at the same hospital during the same period were enrolled as the controls. Fasting peripheral venous blood samples were collected on the second day after admission for AMI patients and on the morning of physical examination for healthy controls. Serum levels of RBP4 and CysC were measured by an automatic biochemical analyzer. Cardiac function indexes [left ventricular ejection fraction (LVEF), left ventricular end-diastolic diameter (LVEDD)] and ventricular remodeling indexes [left ventricular end-systolic volume (LVESV), left ventricular end-systolic posterior wall thickness (PWS), left ventricular remodeling index (LVRI), left ventricular mass (LVM)] were detected by color Doppler ultrasound. Pearson correlation analysis was used to analyze the correlations of serum RBP4 and CysC levels with cardiac impairment and ventricular remodeling in AMI patients. Receiver operator characteristic curve (ROC curve) was plotted to evaluate the predictive value of serum RBP4, CysC alone and their combination for cardiac impairment and ventricular remodeling in AMI patients.

No significant differences in gender and age were observed between the two groups (both P>0.05). Serum levels of RBP4 and CysC, as well as LVEDD, LVESV, PWS and LVM were significantly higher in the AMI group than those in the healthy control group [RBP4 (mg/L): 55.37±8.31 vs. 43.12±6.47, CysC (mg/L): 1.58±0.49 vs. 0.72±0.11, LVEDD (mm): 48.02±7.17 vs. 37.19±5.58, LVESV (mL): 67.42±10.11 vs. 48.61±7.29, PWS (mm): 13.52±1.03 vs. 9.36±1.10, LVM (g): 244.96±36.74 vs. 167.35±25.10, all P<0.05]. In contrast, LVEF and LVRI were significantly lower in the AMI group than those in the healthy control group [LVEF: 0.421±0.063 vs. 0.654±0.098, LVRI (g/mL): 0.85±0.13 vs. 1.24±0.19, both P<0.05]. Among 185 AMI patients, 64 were classified as Killip class I, 58 as class II, 44 as class III, and 19 as class IV. With the increase of Killip classification, serum RBP4 and CysC levels, LVEDD, LVESV, PWS and LVM were gradually elevated, while LVEF and LVRI were gradually decreased (all P<0.05). Pearson correlation analysis showed that serum RBP4 and CysC were negatively correlated with LVEF (r values were -0.453 and -0.424, respectively) and LVRI (r values were -0.479 and -0.443, respectively, all P<0.05), and positively correlated with LVEDD (r values werre 0.202 and 0.297, respectively), LVESV (r values were 0.248 and 0.267, respectively), PWS (r values were 0.285 and 0.272, respectively) and LVM (r values were 0.291 and 0.228, respectively, all P<0.05). ROC curve analysis indicated that serum RBP4 and CysC alone had moderate predictive value for cardiac impairment and ventricular remodeling in AMI patients, and the combined detection achieved higher predictive efficacy. The area under the ROC curve (AUC) with 95% confidence interval (95%CI) was 0.924 (0.863-0.984) and 0.913 (0.849-0.976), with sensitivity of 89.91% and 90.14%, and specificity of 88.92% and 88.68%, respectively.

Serum RBP4 and CysC levels are significantly elevated in AMI patients and are closely related to the severity of cardiac impairment (Killip classification, LVEF, LVEDD) and ventricular remodeling (LVESV, PWS, LVM, LVRI). Combined detection of serum RBP4 and CysC shows high predictive value for cardiac impairment and ventricular remodeling in AMI patients.

Isolated distal deep vein thrombosis (IDDVT) is thrombosis that occurs in the lower-limb venous system, limited to the calf veins, with no extension into the popliteal veins. Proximal propagation or proximal DVT (PDVT) has well-defined treatment protocols in which anticoagulation shows clear benefit. IDDVT, however, does not always require treatment because the risk-benefit relationship of anticoagulation is not well established. The divergence in IDDVT treatment has led to two ultrasound approaches: whole-leg and proximal- only ultrasound, with repeat examination when indicated.

To investigate the prevalence of IDDVT using the current ultrasound approach at Universitas Academic Hospital, in order to advise on possible adaptation and tailored approaches in distinct subpopulations.

A retrospective file review of ultrasound investigations performed from 1 January 2015 to 31 December 2020 at the vascular suite in Universitas Academic Hospital, Bloemfontein, South Africa. Demographic data, HIV status and the anatomical site and anatomical extent of the thrombosis were collected.

In total, 2 208 files were reviewed, which included 1 203 (54.5%) cases negative for DVT, with an alternative diagnosis possible in 113 (9.4%) of the negative cases. Of the 1 005 (45.5%) positive cases, 983 had adequate data for analysis, with 845 (86.0%) PDVT and 138 (14.0%) IDDVT cases. A female predominance was observed, with a male-to-female ratio of 1:2.02. Black African and mixed-race patients had significantly less IDDVT than white patients. Nearly two-thirds (66.0%) of the patients were HIV positive. However, HIV as a risk factor for PDVT did not meet statistical significance (p=0.053).

The low prevalence of IDDVT in the study cohort suggests that a proximal-only ultrasound approach can be adopted. Additionally, black African and mixed-race ethnicity, as well as male sex, were associated with a higher incidence of PDVT. Further prospective evaluation of the propagation rate of IDDVTs might advocate for tailored ultrasound approaches in different demographic groups.

As human longevity continues to increase globally, the prevalence of age-related deterioration and associated chronic lung diseases has risen in parallel. Unlike many organs, the lung is continuously exposed to environmental insults, leading to cumulative cellular damage, impaired epithelial progenitor activity, and chronic inflammation, which contribute to diseases such as pulmonary fibrosis and chronic obstructive pulmonary disease. Under physiological conditions, lung homeostasis is maintained by epithelial progenitor cells, particularly alveolar type II cells. However, this regenerative capacity is often disrupted in chronic lung diseases, resulting in impaired repair and fibrotic remodeling. Fibroblast growth factors (FGFs) and their receptors (FGFRs) play critical roles in epithelial regeneration, cell survival, and tissue repair. Despite their therapeutic potential, clinical translation remains limited by poor protein stability, rapid proteolytic degradation, low delivery efficiency across biological barriers, and insufficient cell-type specificity due to broad FGFR expression. Recently, several delivery strategies have been evaluated in preclinical lung disease models, including protein engineering, nanoparticle and liposome-based platforms, protein transduction domain- and cell-penetrating peptide-based intracellular delivery, and mesenchymal stem cell-based approaches. Therefore, this review summarizes the roles of FGF signaling in lung regeneration and age-associated diseases, and discusses emerging delivery strategies aimed at bridging the gap between preclinical efficacy and clinical application.

To evaluate the efficacy of establishing a nurse-coordinated and guideline-driven care by advanced stroke nurses for acute ischemic stroke (AIS).

Specialised nursing care is usually not integrated enough to provide high quality of interdisciplinary care for patients with AIS.

A prospective study with retrospective controls.

A total of 231 patients with AIS were prospectively enrolled to receive the nurse-coordinated and guideline-driven care for intravenous thrombolysis (prospective cohort). A retrospective cohort using propensity-score matching at a 1:1 ratio retrieved reviews of 231 cases. The primary endpoint was early major improvement in the National Institute of Health Stroke Scale (NIHSS) scores at 24 h after thrombolysis. Secondary outcome measures included treatment time of each link at different stages, functional independence, EuroQoL 5-dimension three-level scale (EQ-5D-3L) scores and adverse events.

The rate of early major improvement in NIHSS score was 57.19% and 42.0% in the prospective and retrospective cohorts with a rate ratio of 1.361 (95% CI: 1.127-1.643) indicating superiority (p = 0.002). A higher rate of functional independence was observed in the prospective cohort as compared to controls (77.9% vs. 68.8%, p = 0.035). The mean time from admission was 17.11 ± 9.64 min for imaging, 21.75 ± 8.42 min for femoral artery puncture and 51.43 ± 9.64 min for thrombolysis in the prospective cohort, which were shorter than those in controls (all p < 0.001). EQ-5D-3L scores were better in prospective cases than in controls within 90 days after thrombolysis (all p < 0.05). A lower incidence of symptomatic intracranial haemorrhage was observed in the prospective cohort, while no significant difference in mortality up to 90 days was found between the two cohorts.

Although patients and the public were not directly involved in the design or conduct of this study, the research findings provide critical evidence that nursing-led, guideline-driven coordination directly improves the efficacy and safety in the management of patients with AIS by ensuring timely reaction, consistent application of evidence-based protocols and enhanced multidisciplinary teamwork.

Chinese Clinical Trial Registry: ChiCTR2200056818.

Precision medicine offers the opportunity to improve the benefit-risk profile of new therapies by prospectively identifying patients most likely to respond or least likely to experience harm; however, its systematic integration into drug development remains inconsistent outside oncology. Key barriers include timely generation of robust predictive biomarker hypotheses (i.e., hypotheses regarding treatment-by-biomarker interactions), appropriate validation strategies, and coordinated development of companion diagnostics within increasingly complex FDA, European Medicines Agency, and In Vitro Diagnostic Regulation regulatory frameworks. This review presents a pragmatic operational framework to guide the incorporation of precision medicine into drug development across therapeutic areas. We outline structured approaches for early biomarker hypothesis generation and prephase II evidence development; define five clinical development scenarios based on the strength of the biomarker signal-front-loaded enrichment, precision-medicine-enabled phase II, adaptive phase II, adaptive phase III, and back-loaded confirmatory enrichment; and summarize regulatory and lifecycle considerations for companion diagnostics. Examples from oncology and emerging applications in cardiovascular and metabolic diseases illustrate evolving regulatory expectations and common industry challenges. We also discuss opportunities to extend precision medicine beyond predictive biomarkers to diagnostic, prognostic, and safety biomarkers, as well as digitally derived signatures. Treating precision medicine as a core component of drug development-rather than an optional enhancement-can improve clinical trial efficiency, support commercial viability, and ensure that patients derive meaningful clinical benefit. Early, structured biomarker planning; integrated clinical-diagnostic strategies; and iterative collaboration among sponsors, regulators, HTA bodies, payers, and patients are critical for translating biomarker insights into de-risked pivotal trials and aligned regulatory and market-access decisions.