Multimodal Exploratory Analysis of Persistent Diabetic Macular Edema: Integrating Aqueous Humour Metabolomics, Clinical and OCT Features
Authors' conclusion
PDME is characterized by elevated BMI, serum triglyceride, low SRF incidence, but a higher rate of DRIL and vitreoretinal interface abnormality. Purine metabolism and DNA repair pathways were associated with treatment response and correlated with OCT structural changes.
Study context
- Population
- DME patients treated with monthly anti-VEGF injections over 3 months146 eyes
- Participants
- 103 patients
- Comparison
- Persistent diabetic macular edema (pDME) vs non-pDME after monthly anti-VEGF injections over 3 months
- Outcome
- Clinical, OCT and aqueous humour metabolomic features of pDME vs non-pDME
Methodological detailsexploratory, untargeted metabolomics, multimodal
- Exploratory
- Untargeted metabolomics
- Multimodal
Main finding
As reported in the abstract
- Differentially expressed metabolites
- 405
- Purine metabolism pathway enrichment
- p = 0.001, Q = 0.024
- DNA repair pathway enrichment
- p < 0.001, Q = 0.013
- LOOCV AUC for pDME: metabolomic markers (purine score) vs clinical vs OCT features
- 0.963 vs. 0.716 and 0.505
Observational data: associations, not effects.
Further findings
Associations of OCT features with metabolomic scores
| Estimate | ||
|---|---|---|
| Vitreoretinal interface abnormality | OR (purine metabolism) 5.820 (95% CI 1.136–29.811) | |
| Vitreoretinal interface abnormality | OR (DNA repair score) 3.790 (95% CI 1.096–13.109) | |
| DRIL | OR (DNA repair score) 0.214 (95% CI 0.046–0.990) | |
| 0.050.21520Ratio (log scale) |
| Outcome | pDME | non-pDME | P value |
|---|---|---|---|
| BMI | 25.20 | 23.27 | P = 0.004 |
| Triglyceride | 1.80 | 1.50 | P = 0.003 |
- OCT features
- P = 0.026
- OCT features
- P = 0.009
- OCT features
- P = 0.019
Original publication
Multimodal Exploratory Analysis of Persistent Diabetic Macular Edema: Integrating Aqueous Humour Metabolomics, Clinical and OCT Features. Diabetes Obes Metab. Published 20 September 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →