In the Korean Type 1 Diabetes Home Care Pilot Project, patients with type 1 diabetes (T1D) receive at least two monthly remote educational consultations from a multidisciplinary team. This randomized trial evaluated the feasibility and acceptability of individualized, semi-automated coaching messages based on continuous glucose monitoring (CGM) data for glycemic and emotional management in patients with T1D as a potential alternative to current treatments. Participants who had already enrolled in the national home care program for T1D were randomized in a 1:2 ratio to the control or intervention groups. The control group continued their current treatment, while the intervention group received weekly CGM-based coaching messages for glycemic management and biweekly emotional support messages based on the Patient Health Questionnaire-9 (PHQ-9) for six weeks. Feasibility was assessed based on the difference in time in range (TIR; 70–180 mg/dL) at week 6, other CGM metrics, and patient-reported outcomes. Eighteen participants were enrolled. At week 6, the median TIR was 72.6

AIMS:Prognostic markers for microvascular complications in type 1 diabetes are needed because factors beyond hyperglycaemia contribute to this risk. The triglyceride-glucose (TyG) index is an established marker of vascular complications in type 2 diabetes; however, its clinical significance in type 1 diabetes remains unknown. We aimed to investigate the association between the TyG index and long-term renal outcomes in a nationwide cohort of adults with type 1 diabetes. MATERIALS AND METHODS:In this nationwide cohort study of 14 782 adults with type 1 diabetes from the Korean National Health Insurance Service database, we used Cox proportional hazards models to evaluate the risk of incident chronic kidney disease (CKD) and end-stage renal disease (ESRD) across quartiles of the baseline TyG index during a median follow-up of 7.04 years. RESULTS:A significant dose-response relationship was observed between the TyG index and adverse renal outcomes. After multivariable adjustment, the highest TyG quartile (Q4) had a 2.15-fold (95% confidence interval, 1.87-2.48) and a 2.90-fold (95% confidence interval, 2.25-3.75) higher risk of developing CKD and ESRD, respectively, than the lowest quartile (Q1) (p for trend <0.001 for both). These associations remained significant in a competing-risk analysis that included all-cause mortality. Additionally, higher TyG quartiles were associated with a progressively worsening distribution of 5-year CKD stages. CONCLUSIONS:The TyG index may serve as a valuable clinical indicator to identify individuals with type 1 diabetes at high risk for developing incident CKD and ESRD.

Aims/hypothesis Elevated remnant cholesterol (RC) levels have been implicated in low-grade inflammation and may contribute to depressed mood. We examined the association between RC levels and variability in the incidence of psychiatric disorders in adults with type 1 diabetes using a nationwide cohort. Methods This study enrolled adults with type 1 diabetes who participated in the Korean National Health Screening Program between 2009 and 2015, who were divided into quartiles based on baseline RC. RC variability was assessed using standard deviation (SD), coefficient of variation, variability independent of the mean, and average real variability (ARV). The primary endpoint was the incidence of composite mental disorders, including depression, bipolar disorder, persistent affective and other affective disorders, or anxiety- and stress-related disorders. Associations were assessed using Cox proportional hazard models. Results Among 10,523 participants (median follow-up of 6.2 years), the highest RC quartile had a significantly higher risk of composite mental disorders than the lowest quartile (adjusted hazard ratio [aHR] 1.43; 95% confidence interval 1.29–1.58). Corresponding aHRs were 1.45 (1.25–1.67) for depression, 1.38 (1.07–1.77) for bipolar disorder, persistent affective and other affective disorders, and 1.41 (1.25–1.58) for anxiety- and stress-related disorders. Higher RC variability, particularly SD and ARV, was independently associated with an increased risk of composite mental disorders and anxiety- and stress-related disorders. Conclusions Elevated RC levels were associated with a higher risk of incident mental disorders. RC from routine lipid profiles may identify adults with type 1 diabetes needing closer mental health assessment, pending prospective validation.

Obesity is a major risk factor for type 2 diabetes, underscoring the need for effective weight management strategies for its prevention and control. We evaluated the association between body weight time in target range (TTR) and the risk of type 2 diabetes in adults with obesity. Using the 2010–2022 Korean National Health Insurance Service database, we included 248,367 adults with obesity with no history of diabetes. Body weight TTR was defined as the proportion of time maintaining a ≥ 3

Although young-onset type 2 diabetes is more prevalent than before, evidence on the management of hypertension in this population is limited. We aimed to evaluate the association between hypertension status and the risk of cardiovascular disease. This retrospective cohort study included 173,483 patients (aged 20-39 years; median follow-up 7.1 years) with type 2 diabetes who were not on antihypertensive medication and underwent health examinations between January 2009 and December 2012. The participants were categorized according to their hypertension status as follows: normal blood pressure, elevated blood pressure, stage 1 isolated systolic hypertension (ISH), stage 1 isolated diastolic hypertension (IDH), stage 1 systolic and diastolic hypertension (SDH), stage 2 ISH, stage 2 IDH, and stage 2 SDH. Compared to those with normal blood pressure, patients with young-onset type 2 diabetes with stage 1 IDH (hazard ratio [HR] 1.14, 95% confidence interval [CI] 1.02-1.28), stage 1 SDH (HR 1.34, 95% CI 1.20-1.49), stage 2 ISH (HR 1.56, 95% CI 1.30-1.89), stage 2 IDH (HR 1.53, 95% CI 1.29-1.82), and stage 2 SDH (HR 1.94, 95% CI 1.71-2.19) showed increased risk of composite cardiovascular disease events. Among young adults with type 2 diabetes, the risks of myocardial infarction, ischemic stroke, heart failure, and CVD-specific mortality increased from stage 1 SDH and higher blood pressure categories. Our findings call attention to the early detection and strict management of hypertension in this population.

Background:Although type 2 diabetes mellitus (T2DM) and metabolic dysfunction-associated steatotic liver disease (MASLD) are associated with an increased risk of ischemic stroke, the extent to which steatotic liver disease (SLD) subtypes and advanced liver fibrosis confer additional risk in individuals with T2DM remains unclear. We aimed to investigate the association of SLD categories and/or advanced liver fibrosis with ischemic stroke risk among patients with T2DM. Methods:A total of 2,220,249 patients with T2DM were classified into five groups: no steatosis, MASLD, MASLD with other combined disease, metabolic dysfunction and alcohol-related steatotic liver disease (MetALD), and alcohol-related liver disease (ALD) with metabolic dysfunction (MD). SLD was defined using a fatty liver index (FLI) of ≥30, and advanced fibrosis was defined by a BARD score of ≥2. Results:Over a median follow-up of 11 years, 135,482 ischemic strokes (6.10%) occurred. Compared with no steatosis, adjusted hazard ratios (aHRs) for stroke were higher in MASLD (aHR, 1.10; 95% confidence interval [CI], 1.08 to 1.11), MASLD with combined disease (aHR, 1.14; 95% CI, 1.12 to 1.17), MetALD (aHR, 1.13; 95% CI, 1.11 to 1.16), and ALD with MD (aHR, 1.32; 95% CI, 1.27 to 1.37). Advanced fibrosis progressively increased stroke risk compared with non-SLD and non-fibrotic SLD. Compared with the FLI <30 group, those with FLI 30-60 and ≥60 showed increased aHRs for stroke. Very heavy drinking was associated with a further increase in risk compared with non-drinking. Conclusion:In patients with T2DM, all SLD categories and advanced liver fibrosis were associated with an increased risk of ischemic stroke. Very heavy alcohol consumption was generally associated with a higher stroke risk across FLI categories.

Introduction and Objective: Current medical AI benchmarks rely on single-best-answer exam accuracy (e.g. USMLE), but real-world type 2 diabetes (T2D) care involves context-dependent clinical decisions with acceptable practice variability. To discriminate real-world clinical effectiveness of AI systems, we aimed to develop a diabetologist-validated framework in T2D management. Methods: We devised a Donabedian model-based framework to assess AI clinical decision capability by evaluating clinical reasoning for patient triage/problem list, medication recommendation, treatment strategy, dose adjustment, and monitoring/education. Meta-evaluation items embedded at the end of each phase assessed the framework’s ability to discriminate the clinical effectiveness of AI systems. Reviewers rated comprehensiveness (coverage of required elements in T2D care) and clarity (unambiguous interpretation and application) on a 4-point scale, and provided free-text feedback to inform between-round revisions. 12 diabetologists completed two initial Delphi rounds; 3 senior diabetologists led the final consensus review. Results: Delphi rounds 1-2 generated 102 item-level revision comments spanning validity, clarity, coverage, feasibility, and traceability. Iterative revisions streamlined the framework from 56 to 29 evaluation items by removing redundancy and sharpening workflow-aligned criteria, while increasing content validity index from 64.4%/51.1% (comprehensiveness/clarity) in the initial round to 100%/100% in the final round. Conclusion: This diabetologist consensus-validated framework provides explicit standards to systematically assess AI-generated T2D treatment recommendations across reasoning reliability, clinical utility, and real-world feasibility. The framework demonstrates potential to serve as an evaluative benchmark for distinguishing AI systems that effectively support diabetologists' treatment decision-making. Disclosure S. Baek: None. J. Kim: None. S. Jin: None. G. Kim: None. Y. Lee: None. J. Kim: None. S. Cho: None. R. Oh: None. B. Kim: None. M. Jang: None. S. Ko: None. M. Moon: None. K. Kim: None. K. Hur: None. Funding Future Medicine 2030 Project of the Samsung Medical Center (#SMX1250111); The Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. RS-2024-00357879)

BACKGROUND:While HbA1c is the standard for monitoring long-term glycaemic control, it fails to capture glycaemic variability. We investigated the discriminatory capacity of longitudinal continuous glucose monitoring (CGM) metrics and identified CGM metric patterns associated with diabetic kidney disease (DKD) in individuals with type 1 diabetes (T1D) using machine learning (ML). METHODS:We analysed cross-sectional data from 282 T1D patients with 1-year consecutive CGM data. DKD was defined by persistent laboratory abnormalities (urine albumin-creatinine ratio ≥ 30 mg/g or estimated glomerular filtration rate < 60 mL/min/1.73 m2) confirmed by at least two measurements within the 1-year period. LightGBM, XGBoost, Random Forest, and Logistic Regression (LR) were developed. Feature importance was assessed using SHAP analysis. RESULTS:The LightGBM model achieved the highest performance (AUROC = 0.91 [95% CI, 0.88-0.93], F1 score = 0.65). All tree-based ML models outperformed the LR model. SHAP analysis identified the standard deviation (SD) of monthly time in range (TIR) and time in tight range (TITR) as the most influential features. In contrast, the CV of sensor glucose did not differ significantly between groups (p = 0.416). Even in the early DKD subgroup, the SD of monthly TIR and TITR remained significantly elevated. CONCLUSION:The SD of monthly TIR and TITR is strongly associated with DKD in T1D, whereas the CV of sensor glucose is not. ML-based integration of these longitudinal metrics offers improved discrimination of concurrent DKD status beyond conventional glycaemic markers.

Colchicine, a well-known anti-inflammatory drug, has emerged as a therapeutic option in various inflammatory diseases. However, its role in chronic kidney disease (CKD) remains unclear. This study aims to investigate the reno-protective effects of colchicine in an experimental CKD model and its potential role in modulating the NOD-like receptor, pyrin domain containing protein 3 (NLRP3) inflammasome pathway. A CKD animal model was established in C57BL/6 mice by feeding a 0.2

Objective: Discordance between the glucose management indicator (GMI) and hemoglobin A1c (HbA1c) is frequently observed in diabetes, yet its physiological basis remains unclear. This study investigated how specific glucose excursion patterns captured by continuous glucose monitoring (CGM) contribute to this discordance in individuals with type 1 diabetes. Research Design and Methods: Ninety-day CGM traces from 611 adults with type 1 diabetes were paired with HbA1c results obtained within ±15 days. Glucose excursions were quantified using the Glucose Rate Increase Detector (GRID) algorithm with varied peak-glucose and time-to-peak thresholds. Discordance was defined using GMI/HbA1c and updated GMI (uGMI)/HbA1c ratios, and associations with GRID-derived excursion metrics were evaluated alongside conventional CGM-derived variability metrics. Results: Excursions with peak glucose ≥250 mg/dL and time-to-peak ≥90 min were significantly associated with higher uGMI/HbA1c ratios after adjustment for age, sex, estimated glomerular filtration rate, and HbA1c group, with consistent findings across CGM devices (Sensor type 1: β = 0.174, 95% CI 0.147–0.201; Sensor type 2: β = 0.102, 95% CI 0.068–0.136; both P < 0.001) and alternative GMI formulations. In restricted cubic spline analyses, adjustment for GRID-derived excursion metrics differentially reshaped the associations of HbA1c, GMI, and uGMI with albuminuria and elevated triglyceride-glucose (TyG) index in an outcome- and context-dependent manner, preferentially enhancing the informativeness of GMI and uGMI—but not HbA1c. Conclusions: Frequent high and prolonged glucose excursions were consistently associated with GMI–HbA1c discordance across devices, HbA1c strata, and analytic conditions. GRID-derived excursion metrics modify the relationship between GMI/uGMI and glycemia-associated risk.

Photoplethysmography (PPG)-based glucose estimation has relied on time-domain or basic spectral features, achieving 13–18% mean absolute relative difference (MARD), which remains insufficient for reliable screening use. Using smartphone-based remote photoplethysmography (rPPG) from 100 subjects (603 recordings, subject-independent validation), we derived time–frequency spectrograms and identified glucose-discriminant features in low-frequency bands (0.5–1.5 Hz) via cluster-based permutation testing. These features showed a non-monotonic, biphasic relationship with glycaemic status – power decreased in prediabetic recordings and increased in diabetic recordings relative to normoglycaemic levels – explaining why linear correlation captures only modest association (r=0.19), and motivating non-linear modelling. A compact convolutional neural network trained on these spectral signatures achieved 8.40% MARD overall (92.7% Clarke Zone A), with strongest performance in the normoglycaemic and prediabetic ranges (RMSE 9.94, 8.02 mg/dL) and weaker, exploratory performance in the underrepresented diabetic range (7.5% of recordings). These findings indicate that biphasic spectral signatures, captured via non-linear integration, enable glucose estimation beyond linear or time-domain approaches.

Introduction and Objective: Type 2 diabetes (T2D) care involves multifactorial clinical decisions that integrate comorbidities, weight, safety, glycemia, adherence, and cost. Guideline updates and expanding trial evidence increase burden and variation in treatment decisions. We developed a diabetologist workflow-aligned agentic AI providing evidence-cited recommendations. Methods: The system generates an evidence-cited report aligned with diabetologist workflow, detailing clinical reasoning for patient triage/problem list, medication recommendation, treatment strategy, dose adjustment, and monitoring/education. Synthetic T2D cases were developed and validated for clinical relevance by 3 senior diabetologists. 12 diabetologists performed double-blind qualitative assessment on AI recommendations for synthetic cases (N=48) using a Delphi-verified 29-item assessment framework. Results: On a 5-point Likert scale, the proportion of ratings ≥4 were: patient triage/problem list 96.1%, medication recommendation 90.9%, treatment strategy 85.4%, dose adjustment 89.2%, monitoring/education 87.8%, reasoning reliability 100%, clinical utility 97.0%, real-world feasibility 90.9%. Conclusion: The agentic AI produced evidence-cited recommendations with high clinician acceptability across core T2D decision components. The system may help standardize treatment decisions in real-world clinical practice. Disclosure S. Baek: None. J. Kim: None. S. Jin: None. G. Kim: None. Y. Lee: None. J. Kim: None. S. Cho: None. R. Oh: None. B. Kim: None. M. Jang: None. S. Ko: None. M. Moon: None. K. Kim: None. K. Hur: None. Funding Future Medicine 2030 Project of the Samsung Medical Center (#SMX1250111); The Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. RS-2024-00357879).

AIMS:This study aimed to evaluate the association of stigma related to type 1 diabetes with CGM-derived data and psychological outcomes in adults with type 1 diabetes. METHODS:In this cross-sectional study, 104 adults with type 1 diabetes undergoing continuous glucose monitoring (CGM) completed the Type 1 Diabetes Stigma Assessment Scale (DSAS-1), Patient Health Questionnaire-9 (PHQ-9), generalized anxiety disorder-7 (GAD-7), and Diabetes Distress Scale (DDS). Thirty-day standard CGM data with ≥70% sensor wear time of CGM was analysed. Linear regression was used to evaluate the potential relationship between the DSAS-1 scores and CGM-derived hypoglycaemia metrics. RESULTS:Higher DSAS-1 total score was independently associated with increased time below range <3.0 mmol/L (adjusted β = 0.011 per point; 95% CI: 0.0018,0.0202; p = 0.019) but not with <3.9 mmol/L. Elevated stigma associated with anxiety (adjusted OR, 1.086; 95% CI:, 1.035,1.152; p = 0.002) with no significant link to depression. Item-level analyses identified DSAS-1 items related to differential treatment (items 15 and 19) and blame/judgement (items 11, 14, and 17) as being significantly associated with clinically significant hypoglycaemia. Associations were consistent across subgroups, especially among participants with a longer diabetes duration and a higher coefficient of variation of CGM glucose levels, calculated as glucose standard deviation divided by mean glucose and expressed as a percentage. CONCLUSIONS:In adults with type 1 diabetes using CGM, perceived stigma was significantly correlated with more time spent in hypoglycaemia and greater anxiety. Further studies are needed to identify causal relationships between stigma and clinically significant hypoglycaemia in people with type 1 diabetes.

Increased intestinal permeability can occur in patients with diabetes mellitus. Previous studies demonstrated a correlation between impaired intestinal barrier function, elevated blood glucose levels, and diminished protective capacity of intestinal epithelial cells. However, few studies have explored gut-barrier disruption using three-dimensional (3D) in vitro models. In this study, we developed and optimized a 3D intestinal organoid model that mimics diabetic conditions by exposing the organoids to high glucose (HG) and palmitic acid (PA) levels. Human intestinal organoids derived from samples of both healthy individuals and patients with diabetes mellitus were analyzed. We evaluated the transcript levels of tight junction proteins and inflammation-related genes in ex vivo mouse intestinal organoids cultured under HG and PA conditions for 48 h. Human intestinal organoids from patients with diabetes mellitus exhibited reduced expression of genes associated with intestinal function and barrier integrity compared with those from healthy individuals. In mouse intestinal organoids, PA treatment induced cytotoxicity and significantly reduced the expression of intestinal stem cells and tight junction proteins, including zonula occludens-1 and occludin, compared with the control and HG-treated groups. Furthermore, treatment with HG and PA resulted in increased levels of inflammatory factors compared with those in the control group. Our in vitro model using 3D intestinal organoids can be used to investigate the impact of diabetic conditions and provide insights into gut barrier disruption.