Abstract
Measurement of creatinine clearance (CCr) is widely used for the evaluation of kidney function. However, CCr measured with diuresis for 24 hours (24hrCCr) is time-consuming. While there have been reports on measurement of CCr with samples of urine collected in shorter intervals, there has been few reports on an one-hour urine sample for CCr measurement (1hrCCr). We compared 24hrCCr, as the standard measure, with 1hrCCr. Forty-two patients had 56 complete sets of 24hrCCr and 1hrCCr values. Agreement between methods was assessed using Spearman's correlation coefficients and Bland-Altman plots. Multivariate analysis was performed by multiple linear regression to identify independent etiologic factors associated with the percentage change between 24hrCCr and 1hrCCr. There was a significant correlation between 24hrCCr and 1hrCCr (r=0.72, P<0.0001) . Bias was 14.2 ml/min, and 95% limits of agreement were -59.8 to 88.2 ml/min. After multivariate analysis, the factors significantly associated with percentage change between 24hrCCr and 1hrCCr was urine creatinine concentration in the one-hour urine collection (P=0.015) , change in urine volume before the one-hour urine collection (P<0.0001) and change in estimated glomerular filtration rate (eGFR) (P=0.0002) . 1hrCCr is not an adequate substitute for 24hrCCr in an ICU setting. The discrepancy between 24hrCCr and 1hrCCr may be caused by measurement error and dynamic changes in kidney function.
