OBJECTIVES: Serum creatinine (Scr) is not a reliable marker of renal function in critically ill patients because of an enhancement of protein catabolism, which makes it difficult to adjust the dosage of renally eliminated drugs such as antibiotics. This study aimed to investigate whether serum cystatin C (Scys-C) could be used as a reliable marker of renal function. METHODS: We investigated whether Scys-C was a reliable marker of renal function in 56 critically ill patients. Subsequently, the usefulness of Scys-C to determine the initial loading and the maintenance dose of vancomycin was examined in 18 patients. Crea- tinine clearance (Ccr) was assessed from Scr and creatinine in urine collected over 24 h (24-h Ccr). KEY FINDINGS: There was a good correlation between 24-h Ccr and 1/Scys-C (r(2) = 0.616), whereas less marked correlation was observed between 24-h Ccr and 1/Scr (r(2) = 0.221). On the other hand, vancomycin concentration was predicted from population pharmacokinetic parameters based on a two-compartment linear model. There were significant correlations between real trough concentrations of vancomycin and the values predicted from Scys-C using various equations (r(2) = 0.416-0.488), while less pronounced relationships were observed between real concentrations and the values predicted from Scr (r(2) = 0.134-0.187). CONCLUSIONS: These findings suggest that Scys-C is a reliable marker reflecting renal function in critically ill patients and is applicable to determine the initial loading dose as well as the maintenance dose of vancomycin.
OBJECTIVES: Serum creatinine (Scr) is not a reliable marker of renal function in critically illpatients because of an enhancement of protein catabolism, which makes it difficult to adjust the dosage of renally eliminated drugs such as antibiotics. This study aimed to investigate whether serum cystatin C (Scys-C) could be used as a reliable marker of renal function. METHODS: We investigated whether Scys-C was a reliable marker of renal function in 56 critically illpatients. Subsequently, the usefulness of Scys-C to determine the initial loading and the maintenance dose of vancomycin was examined in 18 patients. Crea- tinine clearance (Ccr) was assessed from Scr and creatinine in urine collected over 24 h (24-h Ccr). KEY FINDINGS: There was a good correlation between 24-h Ccr and 1/Scys-C (r(2) = 0.616), whereas less marked correlation was observed between 24-h Ccr and 1/Scr (r(2) = 0.221). On the other hand, vancomycin concentration was predicted from population pharmacokinetic parameters based on a two-compartment linear model. There were significant correlations between real trough concentrations of vancomycin and the values predicted from Scys-C using various equations (r(2) = 0.416-0.488), while less pronounced relationships were observed between real concentrations and the values predicted from Scr (r(2) = 0.134-0.187). CONCLUSIONS: These findings suggest that Scys-C is a reliable marker reflecting renal function in critically illpatients and is applicable to determine the initial loading dose as well as the maintenance dose of vancomycin.