Analysis of type 2 diabetes heterogeneity with a tree-like representation: insights from the prospective German Diabetes Study and the LURIC cohort
Authors' conclusion
These data define the pathophysiological underpinnings of the tree structure, which has the potential to stratify diabetes-related complications on the basis of routinely available variables and thereby expand the toolbox of precision diabetes diagnosis.
Study context
- Population
- Individuals aged 18–69 years with recent-onset T2D from the German Diabetes Study (GDS), followed up for up to 16 years; plus individuals with T2D undergoing invasive coronary diagnostics (LURIC cohort)GDS
- Participants
- 927 individuals
- Outcome
- Stratification of key pathophysiological components and diabetes-related complications across the tree
- Follow-up
- Up to 16 years
Methodological detailstree-like representation by reversed graph-embedded dimensionality reduction, validation in LURIC cohort
- Tree-like representation by reversed graph-embedded dimensionality reduction
- Validation in LURIC cohort
Main finding
As reported in the abstract
Gradients across the tree (dimensions 1 and 2)
- Clamp-measured insulin sensitivity gradient
- both dimensions: p<0.0001
- Insulin secretion gradient
- pdim1<0.0001, pdim2=0.00097
- Lowest insulin sensitivity region: highest hepatic lipid content
- n=205, pdim1<0.0001, pdim2=0.037
- Lowest insulin sensitivity region: highest IL-6
- n=348, pdim1<0.0001, pdim2=0.013
- Lowest insulin sensitivity region: highest IL-18
- n=350, pdim1<0.0001, pdim2=0.38
- Lowest insulin sensitivity region: elevated cardiovascular risk
- nevents=143, pdim1=0.14, pdim2<0.00081
- Lowest insulin secretion branch: more prone to require insulin therapy
- nevents=85, pdim1=0.032, pdim2=0.12
- Lowest insulin secretion branch: highest risk of diabetic sensorimotor polyneuropathy
- nevents=184, pdim1=0.012, pdim2=0.044
- Lowest insulin secretion branch: highest risk of cardiac autonomic neuropathy
- nevents=118, pdim1=0.0094, pdim2=0.06
Observational data: associations, not effects.
Further findings
LURIC cohort: all-cause mortality and HF
- All-cause mortality highest in tree branch showing insulin resistance
- nevents=488, pdim1=0.12, pdim2=0.0032
- Significant gradients differentiated HF with preserved vs reduced ejection fraction
Original publication
Analysis of type 2 diabetes heterogeneity with a tree-like representation: insights from the prospective German Diabetes Study and the LURIC cohort. Lancet Diabetes Endocrinol. Published 21 December 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →