Evaluation of proteinuria and GFR to diagnose and classify kidney disease: systematic review and proof of concept

Abstract

Chronic kidney disease is often not associated with significant symptoms or abnormalities in common laboratory test results. Diagnosis is supposedly facilitated by calculating the glomerular filtration rate (GFR) from serum creatinine. A reference range GFR, however, does not exclude renal disease, because renal disease causes the subsequent decrease of renal function. Thorough analysis of proteinuria, however, requires a profound knowledge of the renal handling of the different marker proteins of glomerular and tubular origin. This paper summarizes the scientific basis, explains the diagnostic rationale and proves the concept by analyzing 5669 samples, where GFR and proteinuria work-up were available. 63% (1446 of 2287) of the samples with a GFR above 60 showed either glomerular (37.8%, n=865) or tubular proteinuria (25.4%, n= 581). The quantity of proteinuria increased severely with decreasing kidney function. The rate of glomerular proteinuria remained nearly constant in the different GFR groups, while primarily tubular proteinuria increased from 23% to 63%. A proteinuria pattern indicating a good response to therapy was frequently combined with a high GFR (selective glomerular proteinuria/ incomplete tubular proteinuria), while the severe forms of unselective or complete tubular proteinuria associated with a severe GFR decrease. Regression analysis showed a better inverse correlation of GFR with tubular (r=-0.643) than glomerular markers (r=-0.360; combined r=-0.646). We believe that this complex interrelated laboratory information must be delivered most effectively, i.e. with the use of a knowledge based system in combination with improved, visual oriented laboratory output.