Use of recalibrated serum creatinine concentrations for adjustment of drug dosages: determination of values compatible with conventional dosing recommendations.

BACKGROUND: Measurement of serum creatinine concentration is central to evaluation of kidney function. Recent efforts to increase the accuracy of this evaluation have led to recalibration of laboratory instruments. Recalibrated creatinine concentrations often are lower than previously reported. OBJECTIVES: To derive a method for converting recalibrated serum creatinine (R-SCr) concentrations into values that are compatible with recommended equations for calculation of estimated creatinine clearance. METHODS: Beginning with a proprietary array of recalibrated and corresponding non-recalibrated serum creatinine (NR-SCr) numerical data provided by the instrument manufacturer, relationships were examined with exponential and linear regression analyses. The validity of derived values for NR-SCr obtained through these analyses was tested by comparison of proprietary and derived serum creatinine concentrations and calculated creatinine clearance values. RESULTS: Analyses revealed that relationships between R-SCr and NR-SCr creatinine were essentially linear. Rearranging and solving the equation for a straight line described this relationship as x = (y - b)/m, where x is the derived creatinine value, y is the R-SCr concentration, and, for our laboratory instrument, best parameters for m and b equal to 0.987 and -0.07, respectively. Use of these parameters to derive NR-SCr values was shown to significantly decrease positive bias in subsequent creatinine clearance calculations. CONCLUSIONS: As compared to R-SCr concentrations, use of derived NR-SCr values can improve the predictive performance of conventional equations used to calculate estimated creatinine clearance.