Adjustment of serum creatinine values should be standardized in MDRD calculations

Van Biesen W et al. (2006) The importance of standardization of creatinine in the implementation of guidelines and recommendations for CKD: implications for CKD management programmes. Nephrol Dial Transplant 21: 77–83

Guidelines suggest referral to a nephrologist if glomerular filtration rates (GFRs) fall below 60 ml/min/1.73 m². The most widely recommended tool for estimating GFRs, the abbreviated MDRD EQUATION, is based on serum creatinine values determined using the Beckman Astra 8 method. Van Biesen et al. show that measuring serum creatinine using other methods can lead to worrying variation in MDRD-based estimations of GFR, even when correction formulas are used.

Four laboratories provided a set of MDRD-based estimates of GFR and the uncorrected serum creatinine values (n = 10,108) on which the estimates were based. A further 10,617 serum creatinine values were extracted from the BIRNH DATABASE. Data from each source were derived using a different serum creatinine assay. The authors corrected these serum creatinine values to Beckman standards using four published formulas, and recalculated GFRs using the MDRD equation.

Calculated GFRs varied when different correction formulas were used. These variations persisted when patients were stratified by disease severity according to K/DOQI CRITERIA (P <0.0001). Based on their corrected recalculations, the authors determined that the number of patients who should have begun renal replacement therapy in Flanders between 2000 and 2005 far exceeded the actual figure (230–350 vs 12–14 per 100,000 residents).

It is clear that identification of patients in need of treatment, and thereby accurate prediction of future disease burden, requires a worldwide standardization of serum creatinine measurement if estimates of GFR continue to be widely relied upon.