REGISTER
Four EASD 2026 Studies Link Retinal, Renal, and Pregnancy Markers to Diabetes Risk
Four studies from the team of Yufan Wang of the First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine, presented at the 62nd European Association for the Study of Diabetes (EASD) Annual Meeting in Milan, comprise two oral and two short oral presentations on sight-threatening diabetic retinopathy and arterial stiffening, the C-reactive protein-triglyceride-glucose index (CTI) and diabetic kidney disease risk, and continuous glucose monitoring (CGM) in gestational diabetes.

Four studies from the team of Prof. Yufan Wang of the Department of Endocrinology and Metabolism at the First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine were selected for the 62nd European Association for the Study of Diabetes (EASD) Annual Meeting 2026, held Sept 28 to Oct 2, 2026, at the Allianz MiCo convention center in Milan, Italy. The team had two oral presentations and two short oral presentations addressing type 2 diabetes (T2DM) complications and gestational diabetes.

Expert Profile: Prof. Yufan Wang

- Director of the Department of Endocrinology and Metabolism, First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine; MD, chief physician, doctoral supervisor

- Research focus: endocrine disease in pregnancy and postpartum management, and regional diabetes management

- Member of the 6th Session Committee of the Endocrine and Metabolism Branch, Chinese Medical Doctor Association

- Vice-chair of the Endocrinology Branch, Shanghai Medical Association

- Recipient of the Asia Pacific CODHy Young Scientist Award; selected for the Shanghai Pujiang Talent Program and other municipal talent and clinician awards

- Principal investigator on National Natural Science Foundation projects and others; more than 70 publications as first or corresponding author, including more than 30 SCI-indexed articles in Lancet Diabetes & Endocrinology, Science Bulletin, Nature Communications, JCEM, and Thyroid

Article Key Points

- STDR linked to faster baPWV progression (beta=13.70, P=0.02) and new arterial stiffness (HR 1.28, 95% CI 1.16-1.41)

- Cohort: 7,082 T2DM patients without arterial stiffness at baseline, median follow-up 2.64 years, 1,057 incident cases

- Baseline CTI per 1-unit increase associated with incident and progressive diabetic kidney disease (RR 1.09 for both)

- CTI decline associated with lower risk (RR 0.75 incident; RR 0.74 progressive); CTI rise linked to progression (RR 1.41)

- In 897 GDM patients, neonatal hyperbilirubinemia occurred in 11.6% despite median TIR of 95%

- SD, CV, LAGE, and LI were independent risk factors for neonatal hyperbilirubinemia

- E-GDM showed greater glycemic variability (GLI, P=0.028) and higher insulin use (54.3% vs 35.3%) than L-GDM

- In L-GDM, CV was an independent risk factor for large-for-gestational-age infants

- Evidence status: conference abstracts; full results have not yet been peer-reviewed

Fig.1. Prof. Yufan Wang Presenting at the EASD 2026 Congress

Study 1: Sight-Threatening Diabetic Retinopathy and Accelerated Arterial Stiffening in T2DM (OP 16)

Background

Vascular aging determines healthy life expectancy, affecting susceptibility to cardiovascular disease and shaping the risk of functional decline across multiple organ systems. Arterial stiffness is a key phenotype of vascular aging, and brachial-ankle pulse wave velocity (baPWV) is an established noninvasive measure. Diabetic retinopathy (DR) is a window into systemic vascular injury and is associated with multiple cardiometabolic conditions. This study aimed to evaluate the association between DR severity and annual change in arterial stiffness, as well as the risk of incident arterial stiffness, in patients with T2DM.

Methods

This prospective study screened 23,525 patients with T2DM from two national metabolic management centers (MMCs) between 2017 and 2025. A total of 7,082 patients with regular follow-up and no arterial stiffness at baseline were enrolled. Patients were stratified by DR severity as non-DR, mild non-proliferative DR (NPDR), moderate NPDR, and sight-threatening DR (STDR). Arterial stiffness was defined as baPWV of at least 1,774 cm/s, corresponding to the upper quartile in this study population. Cox proportional hazards models assessed the relationship between DR severity and risk of arterial stiffness.

Results

Over a median follow-up of 2.64 years, 1,057 patients developed incident arterial stiffness. The annual change in baPWV increased progressively with DR severity. Median annual absolute changes in baPWV were 23.39, 25.27, 30.96, and 66.11 cm/s in the non-DR, mild NPDR, moderate NPDR, and STDR groups, with corresponding median annual relative changes of 1.68%, 1.94%, 1.98%, and 4.66%. Compared with patients without DR, STDR was significantly associated with larger annual absolute change (beta=13.70, P=0.02) and annual relative change (beta=0.88, P=0.02) in baPWV. STDR was also independently associated with higher risk of incident arterial stiffness (HR 1.28, 95% CI 1.16-1.41, P<0.0001). Subgroup analysis showed the association between STDR and arterial stiffness risk was consistent across traditional risk factor subgroups.

Conclusion

STDR was associated with accelerated baPWV progression and higher risk of incident arterial stiffness. Routine retinal imaging may therefore be one approach to assessing systemic vascular aging.

Study 2: CTI and Risk of Kidney Disease in T2DM (OP 32)

Background

The C-reactive protein-triglyceride-glucose index (CTI) is an emerging marker reflecting inflammation and insulin resistance, and has been closely linked to T2DM, chronic kidney disease, cardiovascular disease, and all-cause mortality. However, most previous studies excluded patients with diabetes. This study examined the association between CTI and incident and progressive diabetic kidney disease in patients with T2DM.

Methods

This longitudinal study enrolled 7,076 patients with T2DM, divided into an incident diabetic kidney disease cohort (4,773 patients without diabetic kidney disease at baseline) and a disease progression cohort (2,303 patients with diabetic kidney disease at baseline). Incident diabetic kidney disease was defined as estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m2 or urine albumin-to-creatinine ratio (UACR) of at least 30 mg/g developing during follow-up. Progression was defined as eGFR decline of at least 30% from baseline, or proteinuria progression with absolute values exceeding twice baseline. Modified Poisson regression assessed the association between baseline CTI and longitudinal CTI change with kidney disease outcomes.

Results

Over a median follow-up of 1.88 years, 1,124 patients developed incident diabetic kidney disease and 641 experienced progression. After multivariable adjustment, each 1-unit increase in baseline CTI was associated with higher risk of incident diabetic kidney disease (RR 1.09, 95% CI 1.03-1.15) and of disease progression (RR 1.09, 95% CI 1.02-1.16). Restricted cubic spline analysis indicated a positive dose-response relationship for both outcomes. Longitudinal CTI change was significantly associated with outcomes. Compared with the stable CTI group, the declining CTI group had lower risk of incident diabetic kidney disease (RR 0.75, 95% CI 0.62-0.91) and of progression (RR 0.74, 95% CI 0.58-0.94), while the rising CTI group had higher risk of progression (RR 1.41, 95% CI 1.02-1.94).

Conclusion

Higher baseline CTI was associated with higher risk of both incident and progressive diabetic kidney disease in patients with T2DM. Longitudinal CTI change provided additional prognostic information. Monitoring baseline CTI and its dynamic change may help refine risk stratification for diabetic kidney disease.

Study 3: Glycemic Variability and Neonatal Hyperbilirubinemia in Well-Controlled GDM (SO 094)

Background

Glycemic elevation in gestational diabetes mellitus (GDM) is closely associated with neonatal hyperbilirubinemia, but the relationship between glycemic variability and hyperbilirubinemia is less well defined. This study used CGM-derived glycemic variability metrics to examine the relationships between glucose levels, variability, and neonatal hyperbilirubinemia.

Methods

This retrospective study enrolled 897 patients with singleton GDM pregnancies seen at the First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine between November 2017 and April 2025. All patients wore CGM for at least 72 hours. General information and glucose and lipid metabolic indices were collected, and CGM metrics were calculated, including standard deviation (SD), coefficient of variation (CV), largest amplitude of glycemic excursion (LAGE), mean of daily differences (MODD), lability index (LI), and time in range (TIR). Patients were divided by whether their infants developed hyperbilirubinemia requiring phototherapy. Differences between groups were compared, and multivariable logistic regression assessed the association between variability metrics (LI, CV, and LAGE) and hyperbilirubinemia. ROC curve analysis evaluated predictive performance, with the Youden index determining optimal cut-offs.

Results

Of 897 patients, 104 (11.6%) had infants who developed hyperbilirubinemia, and the median TIR was 95%. In patients with TIR of at least 95%, the incidence of neonatal hyperbilirubinemia was similar to that in patients with TIR below 95% (11.0% vs 12.2%, P=0.632). Within the TIR of at least 95% group, those whose infants developed hyperbilirubinemia had significantly greater glycemic variability than those without adverse outcomes (SD: P=0.004; CV: P=0.026; LAGE: P=0.002; MODD: P=0.044; LI: P<0.001). SD (OR 3.733, 95% CI 1.276-10.921), CV (OR 1.097, 95% CI 1.030-1.170), LAGE (OR 1.442, 95% CI 1.169-1.779), and LI (OR 1.147, 95% CI 1.037-1.269) were independent risk factors for neonatal hyperbilirubinemia. Subgroup analysis showed no significant interaction of CV or LAGE with age, prepregnancy BMI, or insulin use during pregnancy. Optimal cut-offs were 0.60 mmol/L for SD, 11.71% for CV, 4.29 mmol/L for LAGE, and 1.69 (mmol/L)2/min for LI; exceeding these values was associated with significantly higher risk.

Conclusion

Neonatal hyperbilirubinemia remained common even among pregnant women with well-controlled GDM. Reducing glycemic variability may help lower the incidence of neonatal hyperbilirubinemia.

Study 4: CGM Metrics and Pregnancy Outcomes in E-GDM Compared With L-GDM (SO 094)

Background

Early-onset GDM (E-GDM) is considered a higher-risk subtype, but whether its glycemic characteristics and associated factors differ from late-onset GDM (L-GDM) remains unclear. This study used CGM to compare the relationships between key CGM metrics and maternal and neonatal outcomes across GDM subtypes.

Methods

This prospective cohort study enrolled 672 patients with singleton GDM pregnancies at the First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine between November 2017 and April 2025. All wore CGM for at least 72 hours. General information and glucose and lipid metabolic indices were collected, and CV, mean amplitude of glycemic excursions (MAGE), glucose lability index (GLI), and M value were calculated. Patients were classified into E-GDM and L-GDM groups according to whether the diagnosis of GDM occurred before 20 weeks of gestation. Differences in baseline clinical characteristics, metabolic indices, and CGM metrics were compared. Multivariable logistic regression analyzed the association between CGM metrics and adverse pregnancy outcomes.

Results

Of 672 patients, 92 were classified as E-GDM. Compared with L-GDM, E-GDM patients had higher prepregnancy BMI and HbA1c levels (both P<0.05) and a significantly higher insulin treatment rate (54.3% vs 35.3%, P<0.001). On CGM, glycemic variability was greater in the E-GDM group than in the L-GDM group (GLI: P=0.028). Multivariable logistic regression further showed that CGM metrics, particularly variability-related parameters, were independent risk factors for multiple adverse outcomes, and that these associations differed between groups. In the E-GDM group, MAGE was positively associated with the need for insulin therapy (OR 2.494, 95% CI 1.190-5.229), and the M value was significantly associated with risk of neonatal intensive care unit admission (OR 1.452, 95% CI 1.015-2.077). In the L-GDM group, CV was an independent risk factor for large-for-gestational-age infants (OR 1.033, 95% CI 1.001-1.066).

Conclusion

E-GDM and L-GDM patients have different CGM glycemic profiles. CGM metrics may help identify adverse pregnancy outcome risk across different GDM types and add supporting evidence for stratified GDM management.

Expert Profile: Prof. Liping Gu

- Deputy chief physician, Department of Endocrinology and Metabolism, First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine; MD

- Research focus: early screening and comprehensive management of diabetes complications

- Committee member and secretary of the Basic Science Group, and member of the Primary Care Endocrinology and Metabolism Group, Chinese Medical Association Endocrinology Branch

- Member of the Type 2 Diabetes Group, Endocrine and Metabolism Branch, Chinese Medical Doctor Association

- Visiting scholar at the National University of Singapore; responsible for a national standardized metabolic disease management center (MMC) in Shanghai; more than 20 SCI-indexed publications as first or corresponding author

Yanan Hou

- Resident physician, Department of Endocrinology and Metabolism, First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine

Jichen Hu

- Master’s student, Department of Endocrinology and Metabolism, First Affiliated Hospital of Shanghai Jiao Tong University School of Medicine

Closing Note

Taken together, these four EASD 2026 presentations, spanning retina, kidney, pregnancy, and the newborn, are grounded in clinical practice in China and address early prediction of T2DM complications and the clinical application of CGM in GDM. As the studies progress, they may provide more precise tools for early detection, prevention, and individualized management of diabetes and its complications.

0
0
Loading