Biological variation estimates for spot urine analytes and analyte/creatinine ratios in 33 healthy subjects

Urine samples are frequently used in the clinical practice. In our study, we aimed to calculate the biological variations (BV) of analytes and analyte/creatinine ratios measured in spot urine.

Second-morning spot urine samples were collected from 33 (16 female, 17 male) healthy volunteers once weekly for 10 weeks and analyzed in the Roche Cobas 6,000 instrument. Statistical analyzes were performed using BioVar, an online BV calculation software. The data were evaluated in terms of normality, outliers, steady state, homogeneity of the data, and BV values were obtained by analysis of variance (ANOVA). A strict protocol was established for within-subject (CV_I) and between-subject (CV_G) estimates for both genders.

There was a significant difference between female/male CV_I estimates of all analytes except potassium, calcium and magnesium. No difference was found in CV_G estimates. When the analytes that had a significant difference in CV_I estimates in spot urine analytes were compared to creatinine, it was observed that the significant difference between the genders disappeared. There was no significant difference between female/male CV_I and CV_G estimates in all spot urine analyte/creatinine ratios.

Since the CV_I estimates of analyte/creatinine ratios are lower, it would be more reasonable to use them in result reporting. Reference ranges should be used with caution, since II values of almost all parameters are between 0.6 and 1.4. The CV_I detection power of our study is 1, which is the highest value.