Abstract: FR-PO0259
Microvascular Health Score and Data-Driven Microvascular Phenotypes in CKD: Findings from the MAP-CKD Study
Session Information
- CKD: Omics, Systemic Stressors, and Targeted Pharmacotherapy
October 23, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2201 CKD (Non-Dialysis): Epidemiology, Risk Factors, and Prevention
Authors
- Ahmadi, Armin, University of California San Diego, La Jolla, California, United States
- Rahaman, Masfiqur, University of California San Diego, La Jolla, California, United States
- Harsh, Amol, Mohamed bin Zayed University of Artificial Intelligence, Masdar City, Abu Dhabi, United Arab Emirates
- Ghanim, Basma, University of California San Diego, La Jolla, California, United States
- Dasgupta, Subhasis, University of California San Diego, La Jolla, California, United States
- Weinreb, Robert N., University of California San Diego, La Jolla, California, United States
- Houben, Alfons Jhm, Universiteit Maastricht, Maastricht, LI, Netherlands
- Ix, Joachim H., University of California San Diego, La Jolla, California, United States
- Malhotra, Rakesh, University of California San Diego, La Jolla, California, United States
Background
Microvascular dysfunction contributors to kidney injury and progression of chronic kidney disease (CKD), yet its phenotypic diversity remains poorly defined. We developed a microvascular health score, identified phenotypes, and evaluated associations with eGFR and proteinuria.
Methods
We studied 94 adults with CKD (eGFR <90 mL/min/1.73 m2) enrolled in the MAP-CKD study. Skin capillary density and post-occlusion recruitment, reflecting microvascular structure and reserve were assessed using capillaroscopy. Heat-induced hyperemia measured by laser Doppler was used to assess microvascular perfusion. Each feature was standardized and averaged to generate a microvascular health score. Principle component analysis (PCA) weighted score was derived and unsupervised K-means clustering was used to identify microvascular phenotypes. Associations with eGFR and proteinuria were evaluated using multivariable linear regression adjusted for demographics and comorbidities.
Results
The mean (SD) age was 62 (15) years and 46% were women. Mean (SD) eGFR was 42 (23) mL/min/1.73 m2, and median (IQR) urine PCR was 0.33 mg/mg (0.12-1.60). PCA showed that capillary density, recruitment, and hyperemic perfusion contributed to the composite microvascular score. Higher microvascular health score was strongly associated with higher eGFR (adjusted β = 12; p < 0.001). The score was inversely associated with proteinuria in unadjusted analyses (β = −0.11, p < 0.01), but this was attenuated after adjustment. Cluster analysis identified four microvascular phenotypes with distinct structural and functional profiles. Severe dysfunction was more common in advanced CKD, while preserved microvascular health predominated in earlier stages, with intermediate phenotypes showing heterogeneity.
Conclusion
A microvascular health score was strongly associated with eGFR, while microvascular phenotypes revealed distinct structural-functional patterns of microvascular injury across CKD. Microvascular profiling may offer a useful framework for risk stratification and phenotypic-specific therapeutic strategies.
Table 1. Microvascular phenotypes in CKD
| Cluster | n | Capillary Density (count) | Absolute Recruitment (count) | Log PU Change (%) | eGFR (mL/min/1.73 m2) | Urine PCR (mg/mg) | Microvascular Score (Unweighted) |
| 1 – Severe microvascular dysfunction | 17 | 47 | 9 | 1.9 | 32 | 2.9 | -0.87 |
| 2 – Preserved microvascular health | 17 | 77 | 10 | 3.3 | 61 | 0.4 | 0.72 |
| 3 – Structural loss with preserved hyperemia | 35 | 47 | 8 | 3.1 | 34 | 2.2 | -0.22 |
| 4 – Structural preservation with high recruitment | 25 | 58 | 16 | 2.9 | 47 | 0.9 | 0.48 |
PU: Perfusion units
Funding
- NIDDK Support