ASN's Mission

ASN leads the fight to prevent, treat, and cure kidney diseases throughout the world by educating health professionals and scientists, advancing research and innovation, communicating new knowledge, and advocating for the highest quality care for patients.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005


The Latest on Twitter

Kidney Week

Abstract: PO0912

Urine Biomarkers of Tubulointerstitial Injury in Diabetic Kidney Disease

Session Information

Category: Diabetic Kidney Disease

  • 601 Diabetic Kidney Disease: Basic


  • Zhang, Xiaolan, The Ohio State University, Columbus, Ohio, United States
  • Zhang, Li, The Ohio State University, Columbus, Ohio, United States
  • Cassol, Clarissa Araujo, The Ohio State University, Columbus, Ohio, United States
  • Rovin, Brad H., The Ohio State University, Columbus, Ohio, United States

Group or Team Name

  • The Ohio State Univeristy

Diabetic kidney disease (DKD) is a serious complication of diabetes. Chronic damage to the tubulointerstitium often predicts loss of kidney function, but kidney biopsies are seldom obtained during the routine clinical care of patients with presumed DKD. This study was done to identify non-invasive urine biomarkers that reflect kidney histology in DKD.


Urine and serum were collected from 34 diabetic patients at the time of kidney biopsy, and from 30 healthy volunteers who served as controls. Biopsies were read by a nephropathologist. The severity of interstitial fibrosis/tubular atrophy (IFTA) was reported semi-quantitatively using Tervaert’s Score. Epidermal growth factor (EGF), NGAL, and complement component C5a were measured in urine and serum by ELISA. Analyte levels were compared between patients and controls and correlated to histologic lesions using ANOVA, the Wilcoxon test, linear regression, and logistic model fitting.


Compared to controls, urine (u)C5a and uNGAL levels were about 359 and 20-fold higher, and uEGF levels were 4-fold lower in the DKD cohort (all p<0.0001). Serum and urine C5a and EGF were not correlated, but there was a modest correlation between serum and urine NGAL (r2 0.41, p 0.0024). uEGF was negatively correlated with IFTA (r2 0.41, p < 0.0001) while uC5a, uNGAL, and serum (s)NGAL were positively correlated (r2 0.21, p=0.009; r2 0.17, p=0.02; r2 0.46, p=0.0007). Receiver operating characteristic (ROC) curves were modeled to determine the best combination of biomarkers to distinguish between mild, moderate, and severe IFTA in DKD. The model with a combination of uEGF+sNGAL yielded an area under the ROC curve (AUC) of 1.00 to differentiate mild from moderate IFTA, an AUC of 0.86 to differentiate mild from severe IFTA. The model with uEGF+uNGAL+uC5a gave an AUC of 0.94 to differentiate moderate from severe IFTA. These composite biomarkers outperformed serum creatinine (AUC 0.9 mild-moderate; AUC 0.82 mild-severe; AUC 0.74 moderate-severe) and proteinuria (AUC 0.76 mild-moderate; AUC 0.67 mild-severe; AUC 0.60 moderate-severe).


Urine EGF, C5a and serum and urine NGAL reflect chronic tubulointerstitial damage in patients with DKD. These biomarkers may be useful for monitoring the effectiveness of treatment to slow progression of DKD.