Page:NIOSH Manual of Analytical Methods - 8322.pdf/4

 TRICHLOROACETIC ACID IN URINE: METHOD 8322, Issue 1, dated 17 April 2015 - Page 4 of 5

NOTE:	If the creatinine value is available, the concentration may be reported as µg/g creatinine if desired. EVALUATION OF METHOD: This method was evaluated over a range of 0.9-30 µg/mL. This range covers 3x, 10x, 30x, and 100x of the estimated LOQ. Six replicates were prepared and analyzed at each concentration level. The average recoveries for each of the concentration levels were 94.8% (3 x LOQ), 102.3% (10 x LOQ), 110.4% (30 x LOQ), and 97.1% (100 x LOQ). Recoveries were determined by comparison against spiked and derivatized liquid standards (standards prepared in deionized water). The upper concentration range was extended to 100 µg/mL during testing by an independent laboratory. Five samples were analyzed at this concentration and the average recovery was 105.0%. Overall accuracy was calculated to be 7.5%; bias was 0.0113, and overall precision was 0.0366 [6]. The limit of detection (LOD) and LOQ were determined by analyzing a series of derivatized spiked standards, with the data fitted to a quadratic curve, then estimated according to the Burkart method [17]. A long-term storage stability study was carried out at the 10x, 30x, and 100x LOQ levels. Urine samples were spiked with trichloroacetate and stored at -17 ºC for 7, 14, 21, 30, and 46 days and then analyzed. Recoveries at 30 and 46 days were all greater than 90% [6]. During the testing performed by the independent laboratory (“User Check”), broad, interfering carryover peaks from the urine matrix were noticed. The lab found that raising the final temperature of the GC program to 240 ºC (instead of 180 ºC) and adding a longer hold time 10 min (instead of 7 min) reduced the carryover problem, allowing more precise and accurate measurement of the peak of interest. This adjusted GC program would now be: 80 ºC for 0.5 min, heat to 240 ºC at 20 ºC/min, and hold for 10 min. Either set of conditions may be used. NOTE:	While the overall accuracy and precision for the User-Check samples were within acceptable limits [6], there were spurious results in 10% of the samples (2 out of 20.) No reason is known for these outliers, nor were the samples able to be re-injected, re-extracted, or re-analyzed. To improve user confidence in the results obtained by this method, it is suggested to randomly run duplicate analyses of 10-20% of the samples and to randomly re-inject 10-20% of the samples. If the method is used in an on-going manner and no problems or spurious results are noted, this recommendation could be lowered or eliminated. REFERENCES: [1]	 ACGIH [2014]. TLVs and BEIs based on the documentation of the Threshold Limit Values for chemical substances and physical agents and Biological Exposure Indices. Cincinnati, Ohio: American Conference of Governmental Industrial Hygienists. www.acgih.org (website accessed May 2014) [2]	 Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area [2014]. List of MAK and BAT values 2014; Maximum concentrations and biological tolerance values at the workplace. DFG: Deutsche Forschungsgemeinschaft, http://onlinelibrary.wiley.com/ book/10.1002/9783527682027 (website accessed January 2015) [3]	 SUVA (Swiss National Accident Insurance Fund) [2013]. Protection de la santé au poste de travail. Valeurs limites d’exposition aux postes de travail. 3rd Ed. http://www.suva.ch/english/startseite-ensuva/praevention-en-suva/publications-en-suva.htm (website accessed May 2014) [4]	 Institut für Arbeitsschutz der Deutschen Gesetzlichen Unfallversicherung (IFA) http://www.dguv. de/ifa/Gefahrstoffdatenbanken/GESTIS-Internationale-Grenzwerte-für-chemische-Substanzenlimit-values-for-chemical-agents/index-2.jsp (website accessed January 2015) [5]	 CRC Handbook of Chemistry and Physics [2013]. 94th Edition, CRC Press, Taylor and Francis Group, Boca Raton, FL.

NIOSH Manual of Analytical Methods (NMAM), Fifth Edition