Simultaneous Determination of Ethylene Glycol and Glycolic Acid in Serum by Gas Chromatography-Mass Spectrometry

Abstract

We describe a gas chromatographic-mass spectrometric (GC-MS) procedure for the simultaneous determination of ethylene glycol (EG) and its major toxic metabolite, glycolic acid (GA), in serum. In this method, serum (50 microL) is treated with 150 microL of glacial acetic acid/acetonitrile (1:10, v/v; contains internal standard, 1,3-propanediol, 15 mg/dL) to precipitate protein. After centrifugation, 10 microL of supernate is treated with 500 microL of 2,2-dimethoxypropane/dimethylformamide (80:20, v/v) to convert water to methanol, and the volume is then reduced to < 100 microL of dimethylformamide (but not to dryness). After formation of tertbutyldimethylsilyl derivatives, analysis is performed by capillary column GC-MS in selected ion mode. The method gives a linear response to 1000 mg/L each EG and GA (16.1 mmol/L and 13.2 mmol/L, respectively) and has a lower limit of detection and a lower limit of quantitation of 10 mg/L each EG and GA (0.16 mmol/L and 0.13 mmol/L, respectively). Total assay imprecision is CV < or = 6.4% (200 and 800 mg/L EG and GA [3.2 and 12.9 mmol/L EG; 2.6 and 10.5 mmol/L GA, respectively]). Absolute recovery from human serum was 91.1% for EG and 77.6% for GA. The procedure is free from any known interference. A complete analysis set (three calibrators, patient serum neat, patient serum diluted 1:5 (v/v), and two controls) may be completed in about 2 h. A preliminary result, based on a single calibrator and patient serum diluted 1:5 (v/v), is complete in about 1 h. The method has been used to aid the diagnosis and management in 34 cases of EG intoxication. Selected cases are briefly reviewed.

Ethylene glycol: an estimate of tolerable levels of exposure based on a review of animal and human data

“…Porter et al 1999 in the estimate of the minimum lethal dose in humans at about 1600 mg/kg body weight. Tolerable levels of ingested EG are derived from animal data.…”

“…However, interferences with the enzymatic assay are to be expected from glycerol or propylene glycol (used as solvents for parenteral drugs) or b-hydroxybutyrate (present in serum from patients with diabetic ketoacidosis; Allen and Hamlin Porter et al 1999 1993; Blandford and Desjardins 1994;Malandain and Cano 1995;Nilsson and Jones 1992). Determination of GA by HPLC-MC (Hewlett et al 1983;Petrarulo et al 1991) involves complex and rather laborious methodology.…”

“…Neither of these compounds is naturally occurring or present in medications. The same laboratory recently reported an adaptation of this procedure using the more specific mass spectral detection instead of flame ionization detection (Porter et al 1999). In this procedure EG and GA are specifically analyzed as the tbutyldimethylsilyl derivatives.…”

“…Patients reported to have presented with high levels of EG (up to 55.3 mM) and GA below detection limit exemplify this statement. These patients had ethanol present in serum at the time of admission, explaining both the absence of GA and of significant symptoms (Porter et al 1999). On the other hand, if therapeutic strategy is based on the EG concentration, appropriate measures (ethanol or fomepizole) are generally recommended at plasma EG levels higher than 3.2 mM (>20 mg/dl; Porter et al 1999).…”

“…These patients had ethanol present in serum at the time of admission, explaining both the absence of GA and of significant symptoms (Porter et al 1999). On the other hand, if therapeutic strategy is based on the EG concentration, appropriate measures (ethanol or fomepizole) are generally recommended at plasma EG levels higher than 3.2 mM (>20 mg/dl; Porter et al 1999). This value is considered the threshold of toxicity for systemic exposure, including renal failure.…”