Clinicopathological predictors for progression of chronic kidney disease in nephrosclerosis: a biopsy-based cohort study

Background Biopsy-based studies on nephrosclerosis are lacking and the clinicopathological predictors for progression of chronic kidney disease (CKD) are not well established. Methods We retrospectively assessed 401 patients with biopsy-proven nephrosclerosis in Japan. Progression of CKD was defined as new-onset end-stage renal disease, decrease of estimated glomerular filtration rate (eGFR) by ≥50% or doubling of serum creatinine, and the sub-distribution hazard ratio (SHR) with 95% confidence interval (CI) for CKD progression was determined for various clinical and histological characteristics in competing risks analysis. The incremental value of pathological information for predicting CKD progression was assessed by calculating Harrell’s C-statistics, the Akaike information criterion (AIC), net reclassification improvement and integrated discrimination improvement. Results During a median follow-up period of 5.3 years, 117 patients showed progression of CKD and 10 patients died before the defined kidney event. Multivariable sub-distribution hazards model identified serum albumin (SHR 0.48; 95% CI 0.35–0.67), hemoglobin A1c (SHR 0.71; 95% CI 0.54–0.94), eGFR (SHR 0.98; 95% CI 0.97–0.99), urinary albumin/creatinine ratio (UACR) (SHR 1.18; 95% CI 1.08–1.29), percentage of segmental/global glomerulosclerosis (%GS) (SHR 1.01; 95% CI 1.00–1.02) and interstitial fibrosis and tubular atrophy (IFTA) (SHR 1.52; 95% CI 1.20–1.92) as risk factors for CKD progression. The C-statistic of a model with only clinical variables was improved by adding %GS (0.790 versus 0.796, P Conclusions The study implies that in addition to the traditional markers of eGFR and UACR, we may explore the markers of serum albumin and hemoglobin A1c, which are widely available but not routinely measured in patients with nephrosclerosis, and the biopsy data, especially the data on the severity of interstitial damage, for the better prediction of CKD progression in patients with nephrosclerosis.