Histologic predictors of renal outcome in diabetic nephropathy: Beyond renal pathology society classification

Abstract

The prognostic utility of histologic features in patients with diabetic nephropathy (DN) classified according to the Renal Pathology Society (RPS) classification is controversial. Therefore, we aimed to evaluate the relationship between histologic changes and renal outcome in DN patients.We examined the renal outcome at November 30, 2017 of 74 adult patients (median age of 54.6 years, 69% male, 81% diabetes mellitus (DM) type 2, estimated GFR (eGFR) 29.6 mL/min) with biopsy proven DN between 2010 and 2015. The primary endpoint was renal replacement therapy (RRT) initiation.Half of the patients progressed to end stage renal disease (ESRD) during follow-up; they had lower eGFR, increased proteinuria, hematuria and serum cholesterol. Regarding the pathologic features, they were more frequently in class III and IV, had higher interstitial fibrosis and tubular atrophy score (IFTA), increased interstitial inflammation, more frequent arteriolar hyalinosis and higher glomerular basement membrane (GBM) thickness. The mean kidney survival time was 2.7 (95%CI 2.1, 3.3) years. In univariate time-dependent analyses, higher RPS DN class, increased IFTA, the presence of arteriolar hyalinosis and arteriosclerosis were associated with RRT initiation.In the fully adjusted model, the clinical characteristics associated with poor renal survival were longer duration of DM, lower eGFR, increased proteinuria and higher hematuria and the only pathologic lesions to remain significant were the GBM thickness and the IFTA.In conclusion, in this European cohort, the severity of glomerular lesions evaluated with the RPS DN classification had limited utility in predicting RRT initiation. However, IFTA and GBM thickness were significantly associated with renal survival.

Figure 1

Kaplan–Meier curves of renal survival rate in patients with diabetic nephropathy. (A) Renal survival rates in glomerular pathologic classes. (B) Renal survival rates in IFTA. (C) Renal survival rates in interstitial inflammation. (D) Renal survival rates in extracapillary hypercellularity. (E) Renal survival rates in arteriolar hyalinosis. (F) Renal survival rates in arteriosclerosis.

Figure 2

Multivariate Cox proportional hazards models for ESRD of 59 diabetic nephropathy patients with RPS DN and IFTA introduced as categorical variables: glomerular model (left) and fully adjusted model (right).