Development and validation of equations estimating glomerular filtration rates in Chinese patients with type 2 diabetes.

Abstract

Existing equations to calculate the estimated glomerular filtration rate (eGFR) were derived from nondiabetic Caucasian patients with chronic kidney disease. Here, we developed formulas to more accurately predict the eGFR in Chinese patients with type 2 diabetes and validated their performance in 202 type 2 diabetic and 46 nondiabetic individuals. Within the diabetic cohort, 135 were randomly assigned to a training group, whereas the remaining 67 diabetic and all of the nondiabetic patients were assigned to a validation group. Reference GFR was measured by (51)Cr-EDTA plasma clearance. The new eGFR-estimating formulas, derived using a stepwise regression model, were compared with existing prediction equations in the validation group. The formulas are: 313 x (Age)-0.494 (years) x [SCr]-1.059 (mg/dl) x [Alb]+0.485 (g/dl) for men, and 783 x (Age)-0.489 (years) x [SCr]-0.877 (mg/dl) x [SUN]-0.150 (mg/dl) for women. Compared with existing equations, the new formulas were more accurate and precise in calculating eGFR in diabetic patients, but, similar to other equations, were less accurate in the nondiabetic cohort. Our newly developed equations are simple to use and can be applied in routine clinical practice to calculate eGFR in Chinese patients with type 2 diabetes.