The nephropathy of non-insulin-dependent diabetes: predictors of outcome relative to diverse patterns of renal injury.

Nephropathy of non-insulin-dependent diabetes mel- litus (NIDDM) is the most common cause of end-stage renal failure (ESRF) in Western countries. This study investigates the clinical and histologic putative predictors of disease pro- gression, with the final goal to identify patients at risk who may benefit from early diagnosis and intervention. It examines by repeated measurements of BP, blood glucose, serum creat- mine, and urinary protein excretion rate 65 consecutive NIDDM patients with clinical, persistent proteinuria and biop- sy-documented typical diabetic gbomerubopathy (class I; n = 30), predominant nephroangiosclerosis (class II; n = 23), or nondiabetic type glomerulopathy (class III; n = 12), whose severity of renal tissue involvement was precisely quantified by a global histologic score. Baseline parameters and progres- sion to renal end points, i.e. , doubling of baseline serum creatinine, dialysis, or transplantation, were univariately and mubtivariately correlated by proportional hazards regression models. The median kidney survival time in the overall study population was 3.07 yr. Thirty-seven percent of patients reached an end point during a median (range) follow-up of 1.8 yr (0.4 to 5.7 yr). By univanate and multivariate analysis, kidney survival significantly correlated with baseline urinary protein excretion rate (P = 0.04 and P = 0.04, respectively) and renal tissue injury score (P = 0.0001 and P = 0.02, respectively), but not with the histologic classes. Patients with a urinary protein excretion rate �2 g/24 h, or >2 g/24 h with a histologic score 2 g/24 h and a histologic score > 13 progressed to ESRF over a median of 1 .6 yr. No differ- ences in other baseline parameters or in BP and diabetes control during follow-up accounted for these different out- comes. In NIDDM as well as in nondiabetic chronic renal disease, quantification of urinary protein excretion rate-inde- pendent of the pattern of underlying gbomerular involvement- reliably discriminates progressors from nonprogmessors and, combined with precise quantification of renal tissue injury, reliably predicts risk of ESRF. This information may be used to set guidelines for early diagnosis and appropriate intervention to reduce the number of diabetic patients who will need renal replacement therapy in years to come.