Effect of creatinine and specific gravity normalization techniques on xenobiotic biomarkers in smokers' spot and 24-h...

Renal excretion mechanisms are xenobiotic-specific; therefore, accurate exposure assessment requires an understanding of relationships of xenobiotic biomarker concentration and excretion rate to urine flow, specific gravity and creatinine concentration. Twenty-four-hour urine collection for xenobiotic exposure assessment is considered the "gold standard" procedure. Random spot-urine collection is convenient and minimizes subject compliance concerns but requires that normalization techniques be employed to account for diuresis and diurnal variation in xenobiotic biomarker excretion. This paper examines and makes recommendations concerning normalization techniques and conditions under which spot-urine results most accurately reflect 24-h urine results. Specific gravity, creatinine, and xenobiotic biomarkers were determined in smokers' spot and 24-h urines. Normalization techniques were applied, variance-component analyses were performed to estimate variability, spot urines were pooled mathematically to simulate 24-h urines and analyses of variance were performed to evaluate spot urines' ability to reflect 24-h urine concentrations. For each xenobiotic biomarker concentration, log-linear relationships were observed with urine flow, specific gravity, and creatinine. For most xenobiotic biomarker excretion rates, log-linear relationships were observed with urine flow; creatinine, however, was unaffected by urine flow. The conventional creatinine ratio-normalization technique demonstrated greater variability (within-day, between-day and between-subject) than other normalization techniques. Comparisons of simulated 24-h urines to spot urines suggest that spot-urine collection be performed only between 2 p.m. and 2 a.m. and that the modified specific-gravity-adjusted-creatinine ratio-normalization technique and the creatinine-regression normalization technique yield the best agreement between spot- and simulated 24-h urine results.