Identification and Stratification of Diabetic Kidney Disease Using Serum Cystatin C and Serum Creatinine Based Estimating Equations in Type 2 Diabetes: A Comparative Analysis.

BACKGROUND: Cystatin C is a low molecular weight protein produced by all nucleated human cells, with a stable production rate. Its levels are not influenced by inflammation, infections, hepatic or renal diseases, or by dietary or constitutional factors. We compared serum cystatin C and serum creatinine based predicting equations to estimate glomerular filteration rate (GFR) in type 2 diabetes mellitus, using the staging of chronic kidney disease (CKD) defined by the National Kidney Foundation. We also explored the relationship of urine albumin, GFR, serum creatinine and cystatin C concentrations.

METHODS: A cross-sectional study was performed at a tertiary care hospital in New Delhi. Consecutive patients with type 2 diabetes mellitus above the age of 35 years were enrolled. Fasting and 2-hour-postprandial blood glucose, fasting lipid profile, lipoprotein(a), haemoglobin, microalbuminuria, glycated haemoglobin (HbA1c), liver and renal function tests were assessed. Serum levels of Cystatin C were measured using immune-turbidometric method (Dade Behring analyzer BN2). Estimated GFR (eGFR) was calculated using Cockcroft-Gault formula, Modification of Diet in Renal Disease (MDRD) and Chronic Kidney Disease-Epidemiology (CKD-EPI) Cys C formula. The three sets of eGFR were compared using repeated measure ANOVA. Linear regression analysis was performed to find the factor that affects the albumin excretion rate (AER) and e-GFR levels using all three equations.

RESULTS: We assessed 172 patients with type 2 diabetes mellitus. Mean age of the patients was 61.4 ± 9.6 years with mean duration of diabetes of 11.40 ± 7.5 years. Approximately 70% of patients had hypertension. A family history of diabetes was present in 53.4% of subjects and a history of CAD in first degree relatives in 20.9%. The prevalence of coronary artery disease was 17.4%. Albumin excretion correlated with e-GFR estimated using each of the three equations. The best correlation was seen with the CKD-EPI equation derived e-GFR. The CKD-EPI equation also identified the maximum number of patients in the normo-albuminuria group as having CKD. Albuminuria correlated with blood urea levels (p = 0.014) and serum cystatin C levels (p < 0.005).

CONCLUSIONS: The new cystatin C based Chronic Kidney Disease Epidemiology equation identifies more patients in early CKD and also patients with normo-albuminuric CKD compared to the creatinine based Cockcroft-Gault equation or the Modification of Diet in Renal Disease formulae.