The early decline in renal function in patients with type 1 diabetes and proteinuria predicts the risk of end-stage renal disease

Abstract

The risk of end-stage renal disease (ESRD) remains high in patients with type 1 diabetes and proteinuria; however, little is known about the rate of decline in their renal function. To help determine this, we enrolled patients with type 1 diabetes and proteinuria whose estimated glomerular filtration rate (eGFR) was normal (equal to or above 60 ml/min per 1.73 m(2)). Using a minimum of five serial measurements of serum creatinine for 161 patients, we determined individual trajectories of eGFR change and the occurrence of ESRD during 5-18 years of follow-up. The rates were linear for 110 patients, for 24 the nonlinear rate was mild enough to satisfy a linear model, and the rates were clearly nonlinear for only 27 patients. Overall, in more than one-third of patients, the eGFR decline was less than 3.5 ml/min per 1.73 m(2) per year and the lifetime risk of ESRD could be considered negligible. In the remainder of patients, eGFR declined with widely different slopes and ESRD developed within 2 to 18 years. Based on up to 5 years observation, when renal function was within the normal range, the estimates of early eGFR slope predicted the risk of ESRD during subsequent follow-up better than the baseline clinical characteristics of glycated hemoglobin, blood pressure, or the albumin to creatinine ratio. Thus, the early slope of eGFR decline in patients with type 1 diabetes and proteinuria can be used to predict the risk of ESRD.

Figure 1

Patterns of renal function decline: A, B Linear trajectories, stable renal function; C Stable renal function, clinically inconsequential non-linear trajectory; D and F Linear decliners; E Decliner with clinically inconsequential non-linear trajectory (observed time of ESRD is within 1 year from time of reaching eGFR=10 under linear model); G–I Non-linear decline; G Deceleration; H New-onset decline (acceleration); I Acceleration. Dots represent single eGFR measurements and solid lines represent linear spline function. Dotted line represents linear regression. Vertical dotted lines mark the observed time of ESRD onset.

Figure 2

Distribution of slopes of eGFR decline according to 2 ml/min intervals in 244 subjects with proteinuria (large figure). The three patients in the interval <−32 had slopes −52.5, −56.4 and −70.5 ml/min/1.73m²/year. (Inset: the distribution of slopes in the sub-set of 112 patients with linear trajectories of eGFR decline)

Figure 3

Cumulative incidence of ESRD according to various cut-points of slopes of early eGFR decline in ml/min/1.73m²/year. Cumulative incidence of ESRD is presented according to years of follow-up after the last measurement used to estimate early eGFR slope.