二甲基甲醯胺（職業性）暴露生物偵測分析方法之探討及驗證

Method Development and Validation for Biological Monitoring of Occupational Exposure to Dimethylformamide

二甲基甲醯胺（dimethylformamide，DMF）是常用的有機溶劑，它能迅速穿透皮膚與黏膜而進入人體，其毒性標的器官為肝臟及睪丸，美國政府工業衛生師協會（American Conference of Governmental Industrial Hygienists, ACGIH）及德國國家科學技術委員（Deutche Forschungsgemeinschaft, DFG）對暴露於DMF之生物偵測皆建議以氣相層析（GC/FID）分析尿中單甲基甲醯胺（NMF）之量作為指標，目前ACGIH對其暴露參考值正準備由40mg/g肌酸酐降為20mg/g；德國的參考值則是15mg/L尿。人及大鼠暴露於DMF後，尿中排出的主要代謝物為DMF-OH，此代謝物不是很穩定，目前無市售之標準品，出現在GC圖譜的是其加熱後之裂解物NMF，因此DMF之生物偵測分析方法是先以兩份冷甲醇（含內標）來處理尿檢體，經離心後進行GC分析，以NMF製備檢量線。本研究使用單甲基乙醯胺（NMA）作為內標，檢量線範圍為1~9μg/mL，全部的異日變異係數（CV, n=4）都≦10％，全部的相對誤差都≦±10％，回收率以斜率比計算為100.3±10.2％。儀器偵測極限為0.056ng in 2μL（0.028μg/mL），三種濃度接近參考值之品管樣品之製備，是以未經暴露之人尿稀釋經DMF暴露之鼠尿而得。在-30°C的儲存穩定度是以品管樣品來探討，發現穩定度至少達38天，用此方法分析所收取的39個下班前工人之檢體，只發現其中一人含有微量的NMF，含量為0.289μg/mL （低於目前的檢量線），約相當於0.120μg/g肌酸酐。本分析方法經不同實驗室不同分析員使用不同GC儀器及不同線性範圍驗證，發現再現性佳，相對誤差均小於±10％，但因氣相層析儀的感度較差而無法有效地分析≦4μg/mL 之NMF。

Dimethylformamide (DMF) is an important industrial organic solvent. It is readily absorbed through mucosa and skin. DMF is hepatotoxic and may induce tumors in testis. To monitor workers' exposure, both American Conference of Governmental Industrial Hygienists (ACGIH) and Deutche Forschungsgemeinschaft (DFG) recommend a gas chromatography /flame ionization detector (GC/FID) analysis of urinary monomethylformamide (NMF). The recommended reference value is 15mg/L urine by DFG and in the process of decreasing from 40 to 20mg/g creatinine by ACGIH. NMF is the thermal decomposition product of DMF-OH, the major urinary metabolite, which is unstable and currently not available commercially. Our monitoring method is different from the others in two aspects; (1) a brief centrifugation to remove particulate after the treatment with cold methanol and (2) analysis with an internal standard, monomethylacetamide. The method established had (1) a linear range of 1~9μg NMF/mL, (2) all interday CVs≦10%, (10 sets, with and without media, n=4) (3) all relative errors≦10%, (4) a recovery of, as slope ratio, 100.3 ±10.2% (n=4), and (5) a detection limit of 0.028μg/mL. Three series of QC samples, with concentrations close to the reference vaiues, were prepared by diluting the DMF-exposed rat urines with unexposed human urine. Sample stability was found to have 38 days at -30°C. An OP was written and used to evaluate 39 end of shift samples. Four of them showed a small signal at the retention time of NMF. Standard addition method was used to confirm its identity. Only one of the 4 actually really was NMF. Its concentration was estimated, by extrapolation, to be 0.29 μg/mL. The method was quasi-verified, without simultaneously analyzing a group of identical samples. We've found that (1) the method worked satisfactorily by various hands, (2) all interday CVs and all relative errors were≦10%, and (3) some GC/FID can not effectively analyze samples of urinary NMF of ≦4μg/mL.