Morphometry Predicts Early GFR Change in Primary Proteinuric Glomerulopathies: A Longitudinal Cohort Study Using...

Area of Science:

• Nephrology
• Medical Imaging
• Biostatistics

Background:

• Kidney disease progression models often lack structural data.
• Previous use of structural variables was typically semi-quantitative.

Purpose of the Study:

• To assess the predictive utility of quantitative structural parameters from kidney biopsies.
• To incorporate these variables into models predicting estimated glomerular filtration rate (eGFR) decline.

Main Methods:

• Quantitative structural parameters were measured from digital kidney biopsy images.
• Longitudinal statistical models predicted eGFR change over 55 months.
• Models included structural variables and confounding demographic/clinical factors.

Main Results:

• Four models incorporating quantitative structural variables achieved a cross-validated R2 of approximately 62%.
• Structural variables demonstrated strong associations with eGFR changes.
• These structural predictors generally outperformed confounding variables, except for baseline eGFR.

Conclusions:

• Quantitative assessment of renal biopsies shows significant predictive power for kidney function loss.
• These findings support the potential role of quantitative biopsy analysis in risk stratification.
• This approach may enhance future clinical studies and practice for kidney disease management.