Lipids in Health and Disease最新文献

Epicardial adipose tissue, a metabolically active visceral fat depot surrounding the heart, is increasingly implicated as a key contributor to cardiovascular disease in patients with chronic kidney disease. This review explores how chronic kidney disease induced metabolic dysregulation, chronic inflammation and uremic toxin accumulation lead to epicardial adipose tissue dysfunction. Pathologically altered epicardial adipose tissue promotes atherosclerosis, myocardial fibrosis and arrhythmogenesis through paracrine signaling, oxidative stress and direct cellular interactions. Advanced imaging modalities now enable precise epicardial adipose tissue quantification, revealing its correlation with cardiovascular outcomes in chronic kidney disease. Epicardial adipose tissue is expected to become a new target for predicting and assessing cardiovascular risk, and for interventions to reduce cardiovascular risk in chronic kidney disease patients, thereby providing a new direciovascular risk, and for interventions to reduce cardiovascular risk in chronic kidney disease patients, thereby providing a new direction for future risk management and therapeutic strategies.

Background: Geriatric patients undergoing oral and maxillofacial surgery are at high risk of delayed neurocognitive recovery (dNCR), yet reliable predictive tools remain unavailable.

Methods: This prospective cohort study (July 2021-January 2025) enrolled patients aged ≥ 65 undergoing elective oral and maxillofacial surgery under general anaesthesia. Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) were assessed at baseline and postoperative 1, 3, 7, and 30 days. Serum lipidomics analysis via liquid chromatography-mass spectrometry was performed preoperatively and 24 h postoperatively. The predictive performance of lipid metabolites for dNCR was assessed using receiver operating characteristic curve analysis, with their independent association further evaluated by logistic regression.

Results: Among 160 patients, 52 patients (32.5%) developed dNCR. Preoperatively, dNCR patients exhibited significantly lower serum triglyceride (TG), particularly TG(58:7/22:5) (OR = 0.014, 95% CI 0.002 to 0.109, adjusted P < 0.001) and TG(54:2/18:1) (OR = 0.051, 95% CI 0.010 to 0.252, adjusted P = 0.002), which demonstrated strong predictive performance (AUC = 0.86, sensitivity = 0.73, specificity = 0.85). Postoperatively, reduced levels of TG(58:7/22:5) (OR = 0.067, 95% CI 0.015 to 0.309, adjusted P = 0.003) and TG(54:2/18:1) (OR = 0.034, 95% CI 0.006 to 0.176, adjusted P < 0.001) persisted in dNCR patients at 24 h, retaining predictive value for dNCR (AUC = 0.82, sensitivity = 0.75, specificity = 0.78).

Conclusions: Low serum TG(58:7/22:5) and TG(54:2/18:1) are promising biomarkers for early prediction of dNCR, supporting lipidomics-guided perioperative neurocognitive risk stratification.

Background: Acute basilar artery occlusion (BAO) represents an uncommon but highly devastating form of ischemic stroke, frequently resulting in severe disability or death, despite advances in endovascular treatment (EVT). Although insulin resistance (IR) has been connected to poorer neurological outcomes after stroke, evidence regarding the predictive usefulness of non-insulin-based IR indices in BAO patients receiving EVT is still limited.

Methods: In this study, 266 BAO patients who received EVT between 2012 and 2024 were analyzed. Three IR indices were calculated at admission: the triglyceride-glucose (TyG) index, the metabolic score for insulin resistance (METS-IR), and the ratio of triglyceride to high-density lipoprotein cholesterol (TG/HDL-C). The primary endpoint included a 90-day unfavorable outcome (modified Rankin scale score: 4-6), and the secondary endpoints included mortality and symptomatic intracranial hemorrhage (sICH). Associations were examined by using multivariable regression, and discriminative ability was quantified via receiver operating characteristic analysis.

Results: In 266 patients, 149 (56.0%) demonstrated unfavorable outcomes, and 77 (28.9%) died within 90 days. Higher TyG values showed a robust association with unfavorable outcomes, mortality, and sICH; moreover, it provided the strongest predictive ability for functional outcomes (area under the curve comparisons, P < 0.05). METS-IR was associated with mortality and demonstrated a predictive ability comparable to that of TyG but weaker for functional outcome. TG/HDL-C was not significantly related to any outcome.

Conclusions: Among BAO patients who underwent EVT, the TyG index could strongly predict unfavorable functional recovery and mortality, whereas the METS-IR was mainly associated with mortality. These findings indicate that TyG may help in refining risk stratification after EVT, whereas METS-IR appears to be less informative.

Background: High-altitude hypoxia is closely linked to dysregulated lipid metabolism, particularly elevated triglyceride (TG) levels, which increase cardiovascular and metabolic risks. This study proposes an interpretable deep learning model to predict TG levels in high-altitude migrants based on clinically accessible indicators.

Methods: Data were collected from low-altitude residents (n = 96) and high-altitude migrants (n = 388). An Uncertainty-driven Gated Feature Selection Network (UGFS-Net) was developed for TG prediction, incorporating an uncertainty-driven sample re-weighting and hard example mining strategy. The model was trained via modern optimization techniques and stratified partitioning by the TG distribution. Performance was evaluated using accuracy and calibration metrics, and interpretability was assessed via SHapley Additive exPlanations (SHAP). Five benchmark machine learning models with PCA or LASSO dimensionality reduction were used for comparison.

Results: The UGFS-Net demonstrated a notable performance gain through uncertainty estimation, yielding an increase in R² from 0.7294 to 0.8776 for TG levels prediction under high-altitude. Predicted uncertainty showed significant correlations with errors and effectively distinguished low- from high-reliability samples, with strong calibration (bin-wise r = 0.9164). SHAP analysis highlighted that lipid metabolism, glucose metabolism, and the erythrocyte system collectively form a network that drives TG alterations in high-altitude environments and the Pearson correlation coefficient between gated the attention weights and SHAP importance scores was 0.9093. UGFS-Net consistently outperformed conventional machine learning models.

Conclusions: This study developed UGFS-Net, an interpretable deep learning model that accurately predicts triglyceride levels in high-altitude migrants (R² = 0.8776) and provides well-calibrated uncertainty estimates, with identified key biomarkers offering clinical insights.

Introduction: Noncommunicable diseases, primarily cardiovascular diseases (CVDs), are an emerging cause of morbidity in sub-Saharan Africa. Dyslipidaemia, a key modifiable CVD risk factor, is increasing but remains underdiagnosed and undertreated, particularly among people living with HIV (PLHIV), where HIV-related chronic inflammation, antiretroviral therapy (ART), and sociodemographic factors may contribute. This study assessed the prevalence, correlates of dyslipidaemia, and estimated 10-year CVD risk among PLHIV and people without HIV (PWoH) adults in Rwanda.

Methods: We analysed baseline data from 1,546 adults (1,234 PLHIV and 312 PWoH) enrolled in the NCOHIRWA Cohort from 12 Rwandan health facilities. Data were collected via standardised World Health Organisation (STEP) questionnaires. Dyslipidaemia was defined according to the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) criteria, and 10-year CVD risk was calculated via the Framingham risk score. For group comparisons, chi-square tests were used. Multivariate logistic regression identified independent predictors of dyslipidaemia, with adjusted odds ratios and 95% confidence intervals reported.

Results: Overall, 979 participants (61.9%) had dyslipidaemia, with comparable prevalence rates among PLHIV (744/1234; 60.29%) and PWoH participants (205/312; 65.06%). The PWoH participants had higher LDL-C levels (59 (18.9%) vs. 120 (9.7%), p < 0.0001). Independent predictors of dyslipidaemia included female sex, older age, obesity, and widowhood. PLHIV had lower odds of having elevated total cholesterol (aOR = 0.68, 0.49-0.95). Based on the estimated 10-year CVD risk, 3.95% of the participants had high and 21.47% had very high 10-year CVD, respectively, which was concentrated among older, widowed women with low education levels.

Conclusion: Dyslipidaemia and elevated CVD risk are highly prevalent among the study participants, with disparities based on HIV status, sex, and social vulnerability. Routine lipid screening and integrated HIV-CVD care, particularly for high-risk subgroups such as older women, are essential for reducing the long-term CVD burden.

Background: Acute ischemic stroke (AIS) remains a major determinant for both mortality and enduring functional impairment. High density lipoprotein cholesterol (HDL) and Uric acid (UA) have opposing effects on oxidative stress and vascular protection. The UA/HDL ratio (UHR) may reflect systemic oxidative-lipid imbalance. This study examined whether admission UHR is linked to unfavorable prognosis among individuals with AIS.

Methods: Between October 2022 and September 2024, 822 consecutive AIS patients admitted to Bethune Hospital of Shanxi were retrospectively analyzed. Inclusion required first-ever AIS confirmed by imaging findings within 72 h. The principal endpoint was the functional status assessed at 90 days. Baseline demographics, clinical features, and laboratory values were collected. To account for confounding factors, the association of UHR with outcomes was analyzed through restricted cubic spline (RCS) models, multivariable logistic regression and subgroup analyses.

Results: Among 822 patients, 262 (31.9%) had unfavorable outcomes. These patients exhibited higher UA, lower HDL, and elevated UHR. Multivariable analysis confirmed that UHR was independently associated with unfavorable outcomes (adjusted OR: 1.0021; P = 0.0013). Membership in the top UHR quartile conferred an approximately twofold higher risk compared with the bottom quartile (adjusted OR = 1.97). RCS models demonstrated a linear positive association, with consistent findings across subgroups defined by stroke severity, sex, smoking, alcohol use, hypertension, and diabetes.

Conclusions: Admission UHR is independently linked to unfavorable prognosis among individuals with AIS and holds promise as a simple, economical tool for early risk stratification and personalized management. Its use could support timely interventions and personalized management, contributing to improved patient outcomes and efficient allocation of healthcare resources. Prospective research is needed to validate its prognostic value and guide targeted interventions.

Background: Brown adipose tissue (BAT) regulates metabolic homeostasis, yet its role in tumor presence remains undefined. This study investigates the associations among BAT, body mass index (BMI), and tumor presence in a large Chinese cohort.

Methods: We retrospectively analyzed 1,736 adults who underwent ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography (¹⁸F-FDG PET/CT) at Shanghai East Hospital (2016-2024). Patients were grouped by BAT status with propensity score matching and sex stratification. BAT was identified by FDG uptake at typical depots with CT density criteria. Statistical analyses included chi-square tests, regression models, and mediation analysis (structural equation modeling with bootstrapping) to assess the role of BMI in the BAT-tumor presence association.

Results: The presence of BAT was associated with younger age, female sex, and lower BMI. In multivariable logistic regression, age (p < 0.001) and BMI (p = 0.014) were independently associated with tumor presence. After propensity score matching (PSM), sex-stratified mediation analysis showed that in females, BAT was indirectly associated with a lower likelihood of tumor presence (Natural Indirect Effect OR = 0.88, 95% CI 0.72-0.98), whereas no significant mediation effect was observed in males. Among cancer patients, individuals with BAT had a significantly lower BMI compared to individuals without BAT (overall: 21.08 vs. 23.34; females: 21.33 vs. 22.97; males: 19.65 vs. 24.68).

Conclusions: Our findings indicate that BAT and tumor presence are statistically associated in a sex-specific manner, with BMI potentially playing a mediating role in females. The consistently lower BMI among BAT(+) cancer patients further indicates a plausible link between BAT and body weight regulation. These observations warrant further investigation to elucidate the underlying mechanisms.