Abstract
INTRODUCTION: The intraindividual variability in urinary creatinine excretion is notoriously large. The aims of this study were to investigate the variability of duplicate consecutive 24-hour urinary creatinine excretions in patients and to develop a model for the detection and correction of discrepant creatinine excretions. MATERIALS AND METHODS: A group of 270 patients (82 men and 188 women) were included in the study. We collected the following data: urinary 24-hour volumes (volumetric/gravimetric) and urinary creatinine concentrations (Jaffé/enzymatic) on both collection days. We performed specific calculations to detect discrepant creatinine excretions. RESULTS: In 60 patients (22%) discrepant collections were found. Among the remaining 78%, 22% of the patients collected very accurately (almost identical urinary creatinine excretions). In this subgroup the volume ratios and the creatinine concentration ratios behave inversely as in a dilution curve. A theoretical model and six collection scenarios were developed to detect, interpret and correct discrepant collections. Practical examples are given to illustrate the use of the model in successful correction of creatinine and other analytes for under- or overcollection. CONCLUSIONS: We conclude that missed or overcollected urine volumes are the largest source of variation in creatinine excretion. Discrepancies in consecutive duplicate 24-hour creatinine excretions can be detected and corrected with specific calculations by means of the presented model. The effectiveness of these corrections is demonstrated with examples from daily practice. These calculations can be easily automated.