BackgroundDiabetic nephropathy (DN) is a leading cause of end stage renal disease and renal replacement therapy worldwide. A pathophysiological hallmark of DN is glomerular basal membrane (GBM) thickening, whereas this feature is absent in minimal change disease. Theoretically, a thicker GBM will impede the diffusion of middle-molecules such as cystatin C, potentially leading to a lower estimated GFR (eGFR) from cystatin C compared to that of creatinine. Here we test the hypothesis that thickening of the glomerular filter leads to an increased diffusion length, and lower clearance, of cystatin C.MethodsTwenty-nine patients with a kidney biopsy diagnosis of either DN (n=17) or minimal change disease (MCD) (n=12) were retrospectively included in the study. GBM thickness was measured at 20 separate locations in the biopsy specimen and plasma levels of cystatin C and creatinine were retrieved from health records. A modified two-pore model was used to simulate the effects of a thicker GBM on glomerular water and solute transport.ResultsThe mean age of the patients was 52 years, and 38% were women. The mean eGFRcystatin C/eGFRcreatinine-ratio was 74% in DKD compared to 98% in MCD (P < 0.001). Average GBM thickness was strongly inversely correlated to the eGFRcystatin C/eGFRcreatinine-ratio (Pearson’s r=-0.61, P < 0.01). Two-pore modeling predicted a eGFRcystatin C/eGFRcreatinine-ratio of 78% in DN.ConclusionsWe provide clinical and theoretical evidence suggesting that thickening of the glomerular filter, increasing the diffusion length of cystatin C, lowers the eGFRcystatin C/eGFRcreatinine-ratio in DN.Significance statementIncreased thickness of the glomerular basement membrane (GBM) is an early structural abnormality in diabetes, and has been identified as an independent risk factor for progression to diabetic nephropathy (DN). Increased GBM thickness will increase the diffusion length of mid-sized molecules like cystatin C across the renal filter, potentially leading to increased plasma concentrations and lower estimated GFR from cystatin C. The authors show theoretically that estimated GFR from cystatin C (eGFRcystatin C) is lower than that of creatinine (eGFRcreatinine) in DN to a degree directly depending on GBM thickness. In line with theory, they provide clinical data showing a lower eGFRcystatin C in diabetic patients with the eGFRcystatin C/eGFRcreatinine-ratio being inversely correlated to GBM thickness assessed from electron micrographs.