Emerging evidence links air pollution exposure to metabolic dysfunction; however, few studies have examined diabetes-related mortality in relation to ambient air pollutants using high-resolution exposure data at the population level. In the United States, particularly in large and geographically diverse states such as California, exposure contrasts and population heterogeneity provide an important setting to evaluate these associations.

We conducted a matched case-control analysis using California Department of Public Health (CDPH) Vital Records (2010-2021). Diabetes-related mortality events (ICD-10 E11) were identified as primary or contributory causes. Decedents (cases) were geocoded to residential addresses, and one-year rolling averages of fine particulate matter (PM_2.5) before death were assigned as individual exposures. Each death record was matched to its selected controls based on month and year of birth and race-ethnicity. Controls were identified from the same statewide CDPH mortality database and were eligible because they had not died by the corresponding case's date of death. Because the number of eligible controls varied across matched strata, controls were randomly sampled within each matched stratum to achieve an overall control-to-case ratio of approximately 2:1 for the study population. The final dataset included 60,824 diabetes-related deaths and 119,053 controls. Exposures were standardized by their interquartile range (IQR) and conditional logistic regression models estimated associations between 1 year rolling average fine particulate matter (PM_2.5) exposure and odds of diabetes-related mortality, adjusting for age, sex, race-ethnicity, marital status, and education. Nitrogen dioxide (NO₂) was included as a co-pollutant for confounding control.

PM_2.5 exposure (per 2.65 μg/m³ IQR increase) was associated with a 18% higher odds of diabetes-related mortality (OR = 1.18; 95% CI: 1.15-1.22) before traffic indicator NO₂ adjustment and showed a stronger association with 21% higher odds (OR = 1.21; 95% CI: 1.17-1.25) after NO₂ adjustment. Health economics analysis estimated that reducing PM_2.5 exposure by its IQR could avoid losses of $31.2 million per 100,000 people.

Higher ambient PM_2.5 exposure was associated with increased odds of diabetes-related mortality in California even after adjustment for NO₂ and other impact factors. These findings support the need for continued strengthening of ambient air quality regulations.

Prediabetes is an independent risk factor for diabetes complications and all-cause mortality. Non-pharmacological therapies of traditional Chinese medicine (NPTTCM) offer promising alternatives for prediabetes, yet their comparative efficacy remains unclear.

The present study aimed to evaluate the efficacies of 6 NPTTCM (acupuncture, acupoint catgut embedding (ACE), Baduanjin, electroacupuncture, massage and Taiji) in prediabetes intervention.

Eight databases were searched for randomized controlled trials (RCTs). The network meta-analyses were performed to estimate mean differences and 95% confidence intervals and surface under the cumulative ranking curve (SUCRA) was used to rank NPTTCM.

Fifty-one RCTs (4,129 participants) were included. Massage, electroacupuncture, Baduanjin and acupuncture significantly reduced fasting plasma glucose (FPG) compared with lifestyle interventions. Massage (SUCRA: 99.3%) was the most effective in FPG reduction, followed by electroacupuncture, Baduanjin, acupuncture. All therapies significantly improved 2-hour postprandial plasma glucose (2hPG), among them, massage (SUCRA: 94.1%) was the best, followed by electroacupuncture, Taiji, acupuncture, Baduanjin and ACE. Acupuncture (SUCRA: 81.8%) was the only intervention that significantly reduced HbA1c. Furthermore, electroacupuncture, massage and acupuncture significantly reduced body mass index (BMI), with electroacupuncture (SUCRA: 98.8%) the best, followed by massage and acupuncture. Electroacupuncture and Baduanjin significantly reduced total cholesterol, with electroacupuncture (SUCRA: 79.7%) the best, Baduanjin the second. ACE and Baduanjin significantly reduced triglycerides, with ACE (SUCRA: 74.9%) the best, Baduanjin the second.

NPTTCM may provide potential benefits for improving metabolic outcomes in prediabetes. Massage showed a favorable ranking for reducing FPG and 2hPG, while acupuncture showed a favorable ranking for HbA1c reduction. Electroacupuncture and ACE showed potential benefits for BMI and lipid management. However, given the limitations, these results should be interpreted with caution. Future head-to-head and rigorously designed studies are needed to provide further evidence.

https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=1129592, identifier CRD420251129592.

Contrast-induced acute kidney injury (CI-AKI) is a frequent complication of coronary angiography and percutaneous coronary intervention (PCI), particularly in patients with type 2 diabetes mellitus (T2DM). Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) may exert renoprotective effects, but evidence in the setting of contrast exposure remains limited.

To evaluate the association between GLP-1 RA therapy and CI-AKI in patients with T2DM undergoing coronary angiography or PCI for acute coronary syndromes.

This retrospective cohort study included 336 patients with T2DM who underwent coronary angiography or PCI between May 2023 and March 2025. Patients receiving stable GLP-1 RA therapy (n=149) were compared with non-users (n=187). Serum creatinine and estimated glomerular filtration rate (eGFR) were measured at baseline and 48-72 hours after the procedure. CI-AKI was defined as a serum creatinine increase of ≥0.5 mg/dL or ≥25% from baseline within 72 hours. Multivariable logistic regression was used to identify independent predictors of CI-AKI.

CI-AKI occurred less frequently among GLP-1 RA users than non-users (8.7% vs 25.7%, p<0.001). At 48-72 hours, eGFR was higher in the GLP-1 RA group (67.1 ± 12.1 vs 58.0 ± 11.8 mL/min/1.73 m², p<0.001), whereas serum creatinine did not differ significantly (1.02 ± 0.73 vs 1.17 ± 1.57 mg/dL, p=0.084). GLP-1 RA use was independently associated with lower odds of CI-AKI (OR 0.290, 95% CI 0.119-0.708; p=0.007). Increasing age (OR 1.332, 95% CI 1.214-1.462; p<0.001) and ST-segment elevation myocardial infarction (OR 5.039, 95% CI 2.368-10.722; p<0.001) were associated with higher odds, whereas higher baseline eGFR was associated with lower odds (OR 0.919, 95% CI 0.862-0.979; p=0.009).

GLP-1 RA therapy was independently associated with a lower risk of CI-AKI in patients with T2DM undergoing contrast-based coronary procedures. These findings suggest a potential renoprotective association but require confirmation in prospective multicenter studies.

To identify plasma protein biomarkers associated with incident non-fatal major adverse cardiovascular events (MACE) in diabetic kidney disease (DKD) patients.

We analyzed 317 DKD patients from the UK Biobank. Plasma proteomics and clinical data (demographics, metabolism, renal function) were integrated. In an exploratory discovery phase, three sequential Cox regression models (crude, socio-demographic-adjusted, socio-demographic-metabolic adjusted) screened non-fatal MACE-associated proteins. To prevent information leakage, the cohort was then randomly split into training (70%) and testing (30%) sets; machine-learning feature selection, hyperparameter optimization, and final model development were performed exclusively within the training set. The associated proteins were input into the four-step machine-learning pipeline (LASSO-Cox, random survival forest, Boruta, XGBoost-Cox). Predictive performance was validated using Kaplan-Meier survival analyses, longitudinal trajectory modeling, and ROC benchmarking. An interactive web application was deployed for clinical implementation.

Of 1,463 plasma proteins, 561 were associated with non-fatal MACE across Cox models, with 14 overlapping proteins. Nine core proteins (ANG, IL1R1, CXCL14, ESAM, PTGDS, HAVCR1, FGFR2, IGSF8, CCL3) were validated: ANG showed the strongest non-fatal MACE association (HR = 3.88, 95%CI 2.33-6.48, p<0.001), and all high-expression groups had elevated non-fatal MACE risk. GO/KEGG enrichment highlighted inflammatory-immune pathways like positive regulation of MAPK cascade, Cytokine-cytokine receptor interaction and PI3K-Akt signaling pathway as key mechanisms. The model integrating proteins, demographic factors, and clinical variables achieved the highest predictive performance across non-fatal MACE (AUC = 0.768), myocardial infarction (MI) (0.808), and stroke (0.816) outcomes, with superior stability in cross-validation. CoxBoost + Elastic Net framework was selected as the optimal framework via benchmarking of 101 algorithms. The model demonstrated favorable calibration in high-risk patients and yielded positive net clinical benefit across decision thresholds of 5% to 45%. The web tool (https://jiangli2941.github.io/MACE-prediction-v2/) enables input of 28 variables, outputs non-fatal MACE risk status, risk probability, and highlights abnormal indicators.

Plasma proteomics combined with machine learning identifies robust non-fatal MACE predictors in DKD.

To investigate the association between glycated hemoglobin A1c (HbA1c) and diabetic retinopathy (DR) in adults with type 2 diabetes mellitus (T2DM) at high altitude, and to explore potential effect modifiers.

This retrospective, hospital-based, cross-sectional study included 1040 adults with T2DM residing at high altitude in Qinghai Province, stratified into two altitude groups (1800-2500 m vs. ≥2500 m); no low-altitude comparison group was included. DR was diagnosed by fundus photography. Logistic regression, restricted cubic spline analyses, stratified and interaction analyses were performed.

DR prevalence was 26.6% (n = 277). After full adjustment, each 1% increase in HbA1c was associated with higher odds of prevalent DR (OR 1.19, 95% CI 1.12-1.25), with no evidence of nonlinearity. Altitude did not significantly modify the association (P for interaction = 0.213). Significant effect modification was observed for estimated glomerular filtration rate (eGFR) (P for interaction = 0.031) and a borderline interaction was observed for body mass index (BMI) (P for interaction = 0.061). The association strengthened with lower eGFR (P for trend = 0.016) and higher BMI (P for trend = 0.013), being strongest in the low eGFR plus high BMI group (OR 1.38, 95% CI 1.20-1.59).

Among high-altitude adults with T2DM, HbA1c exhibited a linear positive association with prevalent DR; lower eGFR significantly modified this association, whereas the modifying trend linked to higher BMI remains exploratory and needs additional confirmation. These findings suggest that glycemic control, weight status, and renal function may need to be considered jointly when assessing DR burden in high-altitude populations.

This study aimed to compare the strengths of association and predictive performance of systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP), pulse pressure (PP), pulse pressure index (PPI), and the SBP/DBP ratio for incident diabetes among Chinese adults with elevated BP.

This retrospective cohort study included 29,377 adults with elevated BP from a large Chinese health examination database. Cox proportional hazards models were used to estimate hazard ratios (HRs), and restricted cubic spline analyses were applied to evaluate non-linear associations.

During a median follow-up of 2.98 years, 1,550 participants (5.28%) developed diabetes. After multivariable adjustment, SBP (HR: 1.09, 95% CI 1.05-1.13), MAP (HR 1.07, 95% CI 1.02-1.13), PP (HR 1.06, 95% CI 1.03-1.10), PPI (HR 2.47, 95% CI 1.36-4.46), and the SBP/DBP ratio (HR 1.38, 95% CI 1.13-1.70) were significantly associated with incident diabetes. Restricted cubic spline analyses revealed significant non-linear relationships for SBP, DBP, MAP, and the SBP/DBP ratio. Among all BP parameters, SBP demonstrated the highest predictive performance.

Among Chinese adults with elevated BP, several BP parameters were associated with incident diabetes, and significant non-linear dose-response relationships were observed for SBP, DBP, MAP, and the SBP/DBP ratio. SBP showed the highest predictive performance, suggesting that BP parameters may provide simple supplementary information for diabetes risk stratification in this high-risk population.

The mechanisms underlying why some individuals with diabetes develop severe diabetic retinopathy (DR), whereas others remain free of retinal complications despite long-standing disease, remain incompletely understood. We aimed to identify systemic susceptibility and resilience signatures associated with severe DR using an extreme phenotype-derived machine learning framework.

An extreme phenotype cohort was established comprising 712 individuals with diabetes mellitus, including 437 patients with proliferative diabetic retinopathy (PDR; susceptible phenotype) and 275 patients with diabetes duration ≥10 years without retinopathy (resilient phenotype). Clinical and biochemical variables were integrated to develop interpretable machine learning models. The optimal model was further interpreted using SHapley Additive exPlanations (SHAP). An independent community-based diabetic cohort (n=673) was used to evaluate the distribution of susceptibility signatures in real-world populations.

LASSO regression identified 21 phenotype-associated features for model development. Among the evaluated algorithms, LightGBM demonstrated the strongest ability to discriminate susceptible and resilient phenotypes, achieving an area under the receiver operating characteristic curve (AUC) of 0.90 (95% CI, 0.84-0.95) in the internal validation cohort. SHAP analysis identified urinary albumin excretion rate (UAER), diabetes duration, serum creatinine, total protein, age, and hypertension duration as the dominant phenotype-defining features. Notably, UAER exhibited a pronounced nonlinear association with susceptibility scores, suggesting a close link between renal microvascular injury and vulnerability to severe DR. When applied to the community cohort, the susceptibility signature showed limited discrimination in the overall population (AUC = 0.54) but became progressively enriched among individuals with greater metabolic burden, reaching an AUC of 0.71 in participants with fasting blood glucose ≥9.0 mmol/L.

Using an extreme phenotype-derived machine learning framework, we identified systemic susceptibility and resilience signatures associated with severe diabetic retinopathy. Renal dysfunction, albuminuria, glycemic burden, and disease duration emerged as key phenotype-defining characteristics. These signatures became increasingly enriched in metabolically stressed individuals, supporting the concept that severe diabetic retinopathy arises through the interaction between intrinsic biological susceptibility and cumulative metabolic exposure. This framework may provide new insights into disease heterogeneity and facilitate future precision risk stratification strategies in diabetic eye disease.

Sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP-1RAs) improve glycemic control and promote weight loss, but their effects on the relative balance between skeletal muscle and fat mass remain incompletely understood. This study examined treatment-associated body composition changes in patients with type 2 diabetes mellitus (T2DM), focusing on muscle-to-fat balance.

In this multicenter retrospective cohort study, we analyzed body composition changes 12 months after initiation of SGLT2i (n = 36), GLP-1RA (n = 17), or GLP-1RA add-on to SGLT2i therapy (n = 20). Changes in appendicular skeletal muscle mass (ΔASM) and body fat mass (ΔBFM) were assessed using bioelectrical impedance analysis. The Body Recomposition Score (BRS) was defined as ΔASM - ΔBFM and used as an exploratory index of the relative balance between muscle and fat mass changes; BRS < 0 was operationally defined as indicating relatively unfavorable body recomposition. An independent external cohort of patients with T2DM (n = 148) was used for external assessment.

ΔBFM and ΔASM were positively correlated with changes in body weight (ΔBW) in all treatment groups. However, ΔASM and ΔBFM were positively correlated only in the GLP-1RA group, whereas no such correlation was observed in the SGLT2i or GLP-1RA add-on to SGLT2i groups, suggesting interindividual heterogeneity in muscle-to-fat balance changes. In the SGLT2i group, BRS was positively correlated with baseline HbA1c (r = 0.36, p = 0.029) and estimated glomerular filtration rate (eGFR; r = 0.34, p = 0.040). No baseline variables correlated significantly with BRS in the other groups. Exploratory receiver operating characteristic analyses identified Youden index-derived cut-offs for discriminating BRS < 0 in the SGLT2i group: 6.6% for baseline HbA1c and 65 mL/min/1.73 m² for eGFR. In the external cohort, BRS, calculated as the change in total skeletal muscle mass (ΔSMM) - ΔBFM at 4 weeks after SGLT2i initiation, showed significant positive correlations with baseline HbA1c and eGFR.

Lower baseline HbA1c and reduced eGFR were associated with lower BRS after SGLT2i therapy in patients with T2DM. These exploratory findings support the importance of individualized pharmacotherapy for type 2 diabetes with consideration of muscle-to-fat balance.

https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000055239, identifier UMIN000048469; https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000055242, identifier UMIN000048471.

Baduanjin and resistance training improve metabolism and function; however, it is unclear whether their combination reduces cardiometabolic risk in older adults with type 2 diabetes mellitus (T2DM). This study aimed to determine the effects of 16-week BDJ combined with resistance training on cardiometabolic risk parameters in older adults with T2DM.

An assessor-blinded, randomized controlled. A total of 123 participants were enrolled and randomly assigned to either an intervention group (IG) that received a 16-week supervised BDJ combined with resistance training exercise program (2-3 sessions/week, 60 min/session, n=60) or a control group (CG) that received traditional health education without structured exercise (n=63). Participants in the CG were instructed to maintain their usual daily activities and attend monthly health education talks. The primary outcomes were glycosylated hemoglobin (HbA1c) and fasting plasma glucose (FPG) level, body composition, and handgrip strength. The secondary endpoints were blood pressure and lipid levels. Assessments for all outcome measures were performed at baseline and at 4-month follow-up.

Statistically significant group-by-time interactions were observed for weight (B = -1.30, 95% CI [-2.17, -0.44], p = 0.003), BMI (B = -0.49, 95% CI [-0.82, -0.17], p = 0.003), waist circumference (B = -1.21, 95% CI [-2.09, -0.32], p = 0.008), Left handgrip strength (B = 2.02, 95% CI [0.98, 3.05], p < 0.001), Right handgrip strength (B = 3.75, 95% CI [2.46, 5.05], p < 0.001), FPG (B = -0.81, 95% CI [-1.43, -0.20], p = 0.010), HbA1c (B = -0.39, 95% CI [-0.64,-0.14], p = 0.003). In contrast, none of the secondary outcome measures showed statistically significant differences between the two groups at the 4-month follow-up.

This study provides evidence that a 16-week program combining BDJ with resistance training can be successfully implemented in a diabetes self-management group for older adults with T2DM. By improving glycemic control, muscle strength, and body composition, the intervention improved selected cardiometabolic risk markers in older adults with T2DM.

http://www.chictr.org.cn, identifier ChiCTR2600123608.

Gestational diabetes mellitus (GDM) is a common pregnancy complication, yet current diagnosis at 24-28 weeks may occur after metabolic changes have already begun. The atherogenic index of plasma (AIP, log₁₀[TG/HDL-C]) has shown promise for early GDM risk stratification, but it does not incorporate obesity-related parameters. We evaluated whether first-trimester AIP and its modified indices, AIP-BMI and AIP-WC, are associated with subsequent GDM and compared their predictive performance.

This secondary analysis of a prospective Korean cohort included 581 women with singleton pregnancies. AIP, AIP-BMI (AIP × BMI), and AIP-WC (AIP × waist circumference) were calculated using fasting measurements obtained at 10-14 weeks of gestation; BMI and waist circumference were algebraically reconstructed from the full-precision components of the hepatic steatosis index and fatty liver index because contemporaneous direct measurements were unavailable in the public dataset. GDM was diagnosed at 24-28 weeks using the two-step ACOG approach. Multivariable logistic regression, restricted cubic spline, sensitivity, subgroup, and receiver operating characteristic analyses were performed.

GDM occurred in 36 women (6.2%). In the fully adjusted model, AIP (per 0.1-unit increase: OR 1.66, 95% CI 1.36-2.03; Q3 vs Q1: OR 7.86, 95% CI 1.93-32.08), AIP-BMI (OR 1.20, 95% CI 1.11-1.31), and AIP-WC (OR 1.05, 95% CI 1.03-1.08) were associated with GDM (all P < 0.001). Overall associations were significant in restricted cubic spline analyses, without evidence of nonlinearity. Among individual indices, AIP-BMI had the largest apparent AUC (83.05%, sensitivity 81%, specificity 80%), followed by AIP-WC (81.32%) and AIP (78.66%). The AIP + AIP-BMI model had the largest apparent AUC among evaluated models (84.19%, sensitivity 81%, specificity 81%). Results remained broadly consistent in sensitivity analyses excluding participants with NAFLD, whereas subgroup analyses were underpowered to detect interactions.

First-trimester AIP, AIP-BMI, and AIP-WC were associated with subsequent GDM, and AIP-BMI showed higher apparent discrimination than AIP alone in this sample. These findings require confirmation in larger cohorts with directly measured anthropometry and appropriate internal and external validation.