Reliability of using urinary and blood trichloroacetic acid as a biomarker of exposure to chlorinated drinking water disinfection byproducts.

This study was designed to analyse the reliability of using urinary and blood trichloroacetic acid (TCAA) as a biomarker of exposure. A total of 46 healthy women consumed supplied TCAA-containing tap water for 15 days and provided urine and blood samples for TCAA measurements. The findings revealed that the reliability of measurements was excellent by using measures of TCAA ingestion, blood concentration and urinary excretion (intraclass correlation coefficients (ICC) > 0.75, p < 0.001). Volume of tap water consumption (ICC = 0.69) and creatinine-adjusted urinary concentration (ICC = 0.72) were less reliable. This indicated that the intraindividual variability was small and the interindividual reliability was high by using these measures in this cohort study. Laboratory variability did not significantly contribute to total variance (ICC > 0.95, p < 0.001). Other possible sources of variation such as bathing, showering, dishwashing and physical activities were unlikely to contribute significantly to total variance. For sampling strategies, 1-day blood sampling and 2-day urine sampling are sufficient to achieve reliability for an epidemiological study if a quasi-steady-state TCAA level in the body is reached. The results suggest that TCAA ingestion, TCAA loading in blood and urinary TCAA excretion are reliable measures for use as biomarkers in epidemiological studies.