Diabetes research and clinical practice最新文献

Aim: To compare real-world glycaemic and clinical outcomes in adults with Type 1 Diabetes (T1DM) using Automated Insulin Delivery (AID) vs. those using manual insulin delivery.

Methods: Demographic and diabetes-related glycaemic and clinical data were prospectively collected via a survey from consecutive participants with T1DM attending TIDM clinics in Australia during 2024-25.

Results: Of 406 participants surveyed (233 females [57.4%], age 45.6 ± 16.5 years). AID was used by 141 participants (34.8%), with 50.2% of non-users expressing interest in AID use. AID use vs. non-use was associated with lower HbA1c (7.2 ± 1.0% [63 ± 19 mmol/mol] vs 7.9 ± 1.6% [63 ± 18 mmol/mol], p < 0.001), Glucose Management Indicator (GMI) (7.2 ± 0.8% [55 ± 8 mmol/mol vs 8.0 ± 1.4% [63 ± 15 mmol/mol], p < 0.001), and higher Time In Range (TIR) (69.21 ± 14.79% vs 50.53 ± 21.8%, p < 0.001), with fewer severe hypoglycaemia episodes (n = 3 [2.1%] vs n = 31 [11.7%], p < 0.001). These associations were observed irrespective of Socio-Economic Indexes for Areas (SEIFA) group.

Conclusion: AID use was associated with better glycaemic and clinical outcomes irrespective of socio-economic status. AID use tended to be more prevalent among the socio-economically advantaged. We strongly advocate for equitable AID access based on clinical need rather than financial means.

Background: Existing methods for estimating GFR in people with diabetes have shown inaccuracies when compared to mGFR measurements. We developed and validated an artificial neural network - RenoTrue to improve estimating GFR in people with diabetes.

Methods: 5,619 individuals from five international cohorts with type 1 and type 2 diabetes was split into training (70%), validation (10%) and test (20%) datasets. RenoTrue was developed to estimate GFR using age, sex, and serum creatinine. The performance was evaluated in the test dataset by estimating agreement, bias (mean difference), and accuracy (p30), and compared to CKD-EPI estimates through a multi-level mixed effect regression model.

Findings: Median mGFR was 75 ml/ min per 1.73 m² [IQR: 49, 100] and median age was 59 years [IQR: 38, 69]. RenoTrue demonstrated high agreement (ICC: 0.87 (95% CI: 0.78, 0.93)), low bias (-0.57 (95% CI: -1.59, 0.46) ml/min per 1.73 m²) and p30 of 81% (95% CI: 79%, 83%) compared to mGFR measurements. The 2009 CKD-EPI equation had an ICC of 0.86 (95% CI: 0.77, 0.92), bias of 4.17 (95% CI: 3.14, 5.20) ml/min per 1.73 m² and p30 of 74% (95% CI: 72%, 77%).

Conclusion: For people with diabetes, RenoTrue demonstrated better performance compared to the 2009 CKD-EPI equation in terms of estimating GFR across the full range of GFR.

Aims: Metformin is a cost-effective alternative to insulin for gestational diabetes mellitus (GDM), yet concerns regarding potential teratogenicity persist. This study aimed to evaluate the association between prenatal metformin exposure and multisystem congenital malformations (CMs), and to explore biologically relevant pathways.

Methods: This study first conducted a meta-analysis of RCTs and cohort studies assessing maternal metformin use and fetal CMs. Subsequently, drug-target Mendelian randomization (DTMR) examined genetically proxied associations between metformin pharmacodynamic targets (eQTLs) and 64 CMs (FinnGen), reflecting lifelong target perturbation, adjusting for maternal confounders and validating with placental eQTL data.

Results: Meta-analysis showed a protective effect of metformin versus insulin on overall CMs (RR = 0.83, 95% CI 0.71-0.99). In DTMR, following Bonferroni correction and covariate adjustment, seven of 92 target genes (e.g., NDUFS5, NDUFA2) showed significant associations, primarily exhibiting protective effects against circulatory and musculoskeletal system. Validation in placental eQTLs corroborated the direction of effects for 17 gene-outcome pairs, reinforcing the robustness of key safety signals.

Conclusion: By integrating clinical and genetic evidence, this study is consistent with the overall clinical safety of metformin use during pregnancy with respect to congenital malformations and provides hypothesis-generating insights into relevant biological pathways.

Objective: Whether respiratory syncytial virus (RSV) infection contributes to the development of type 2 diabetes (T2DM) or serves as an early warning indicator of T2DM risk remains unclear.

Methods: The study utilized TriNetX US Collaborative Database between January 1, 2022, and March 31, 2024. Patients with a history of T2DM diagnosis, antidiabetic medication use, or HbA1c ≥ 6.5 were excluded. RSV infection was designated as the index event with propensity score matching. The risk of T2DM was assessed using Cox proportional hazards regression models. Sensitivity analyses were conducted for two periods: 2010-2015 and 2016-2019, and across different databases.

Results: A total of 3,052,016 patients, including 15,205 with RSV (mean age, 51.1 years; 59.9% female) and 3,036,811 without RSV (mean age, 46.9 years; 53.4% female). T2DM incidence was 5.69% in the RSV group vs 2.48% in controls, HR 2.684 [95% CI: 2.378-3.030], E-value 4.81. Risk was significantly increased from infection to 3 and 6 months (HR 3 months: 2.697 [95% CI: 2.332-3.119]; HR 6 months: 2.271 [95% CI: 1.911-2.699]). All sensitivity analyses consistently showed a positive trend.

Conclusions: Our findings suggest an association between RSV infection and subsequent T2DM. Prospective studies and mechanistic investigations are warranted to validate these observations and elucidate the underlying pathways.

Aim: To longitudinally investigate early indicators of cardiovascular disease (CVD) in youths with Type 1 Diabetes Mellitus (T1DM) and their associations with novel metrics derived from continuous glucose monitoring (CGM) systems.

Patients and methods: Eighty-seven patients (mean age: 10.72 ± 3.35 years), completed 3 visits at six-to-twelve-month intervals. In each visit, Pulse Wave Velocity (PWV) and Augmentation Index @75 (AIx@75) were quantified using a validated, non-invasive method, while glycemic parameters such as HbA1c, time in range (TIR), time above range (TAR) and time below range (TBR) were assessed in the trimester prior to the assessment. Patients were categorized according to TIR into TIR improvers: patients with constantly TIR ≥ 70% or constantly TIR ≥ 60% and improved by +≥10% from baseline versus TIR non-improvers and additionally according to HbA1c into HbA1c improvers: patients with constantly HbA1c ≤ 7% or constantly HbA1c ≤ 8% and improved by -≥0.8% from baseline versus HbA1c non-improvers.

Results: TIR improvers showed significant improvement in Δ PWV Z-score according to age and in Δ Systolic Blood Pressure index. No significant difference was observed between HbA1c improvers and non-improvers.

Conclusion: New glycemic metrics seem to serve as more sensitive and early predictors of CVD in young patients with T1DM. Further studies are needed to replicate and confirm these preliminary results.

Aims: Early identification of individuals at risk for type 2 diabetes (T2D) is essential for prevention. We evaluated a novel model-derived disposition index without insulin (mDI-woI), which requires only glucose values from a three time-point oral glucose tolerance test (OGTT: 0, 60, 120 min).

Methods: Among 5,742 healthy Koreans (age 51.2 ± 8.6 years, BMI 24.5 ± 3.1 kg/m²) followed biennially for up to 14 years with repeated OGTTs, we compared baseline mDI-woI with current diabetes biomarkers and the oral disposition index (oDI) using AUC-ROC analyses.

Results: mDI-woI and mean OGTT glucose (mean G) showed the strongest prediction for incident T2D (AUC = 0.79 each), outperforming fasting plasma glucose (0.67), 1 h-PG (0.77), 2 h-PG (0.72), HbA1c (0.71), and oDI (0.68; all P < 0.001). In individuals who progressed to T2D, baseline mDI-woI, mean G, and 1-PG exceeded their thresholds while fasting and 2 h glucose were still below prediabetes cutoffs, indicating earlier risk detection. Moreover, the novel marker mDI-woI is the earliest one, 4 years earlier than mean G and 4.5 years earlier than 1 h-PG, the next two earliest.

Conclusions: Using only three glucose measurements without measuring insulin, mDI-woI provides a simple, sensitive, and clinically practical early marker that outperforms current diabetes criteria for predicting T2D progression, with strong potential for large-scale studies.

Aims: The C-reactive protein-triglyceride-glucose index (CTI) is a novel biomarker of insulin resistance and inflammation. This study aimed to investigate the association between baseline CTI and all-cause and cardiovascular (CVD) mortality in patients with diabetic foot (DF), providing a scientific basis for clarifying this relationship in clinical practice.

Methods: A total of 827 inpatients meeting inclusion criteria were enrolled from a tertiary hospital DF department in Tianjin (2020-2021), with median follow-up of 47 months. Outcomes included all-cause and CVD mortality. Multivariable Cox regression, restricted cubic splines (RCS), Kaplan-Meier analysis, and ROC curves were used to evaluate associations and predictive performance.

Results: During follow-up, 325 all-cause deaths occurred, including 158 CVD deaths. Fully adjusted models showed that high CTI was associated with 57% higher all-cause mortality (HR = 1.57, 95% CI:1.24-2.00) and 50% higher CVD mortality (HR = 1.50, 95% CI:1.06-2.11). RCS indicated nonlinear associations (P < 0.05), and ROC supported CTI's predictive accuracy for mortality.

Conclusions: A higher CTI level was independently associated with an increased risk of all-cause and cardiovascular mortality in patients with diabetic foot. The CTI may serve as a promising and easily accessible risk stratification tool, though its clinical utility requires validation in prospective studies.