Classifying Kidney Disease in a Vervet Model Using Spatially Encoded Contrast-Enhanced Ultrasound Perfusion...

Area of Science:

• Nephrology
• Medical Imaging
• Biomarkers

Background:

• Early-stage diabetic kidney disease (DKD) presents diagnostic challenges in type 2 diabetes patients.
• Microvascular changes in the kidney are critical indicators of early DKD progression.
• Contrast-enhanced ultrasound (CEUS) offers a non-invasive method to assess tissue microcirculation.

Purpose of the Study:

• To identify specific CEUS perfusion parameters as potential biomarkers for early DKD detection.
• To investigate the spatial patterns of kidney perfusion in different metabolic states.
• To evaluate the efficacy of CEUS in distinguishing between healthy, insulin-resistant, and diabetic states.

Main Methods:

• CEUS kidney flash-replenishment data were collected from control, insulin-resistant, and diabetic vervet monkeys (N=16).
• A mono-exponential model was employed to extract time-intensity curve parameters (blood volume, velocity, perfusion index) from 10 kidney layers.
• Linear regression models utilized perfusion parameters and blood pressure to classify the cohorts.

Main Results:

• The mono-exponential model demonstrated good performance across all cohorts (average MSEs ranging from 0.0254 to 0.0321).
• Perfusion index features, combined with blood pressure, were identified as the most effective classifiers between the study groups (p < 0.05).
• Distinct spatial perfusion patterns were observed, correlating with disease status.

Conclusions:

• CEUS has significant potential for detecting early microvascular alterations in the kidney, offering insights into DKD progression.
• The perfusion index derived from CEUS is a promising biomarker for early DKD diagnosis.
• Further research is warranted to explore the sensitivity of CEUS across various stages of DKD.