Formation of semi-crystalline fraction, in which all diethylene glycol (DEG) is contained, during its extraction from human tissues: the probable cause of false negative results in fatal DEG poisoning cases

Abstract

Dear Editor, Diethylene glycol (DEG) can be found in commercial products such as antifreeze, brake fluid, and lubricants. In addition, DEG has been found as a contaminant of raw materials in the production of pharmaceuticals. At least ten mass DEG poisoning events have occurred over the past 70 years. The first and largest outbreak, which resulted in 105 deaths, occurred in the United States in 1937 [1]. In 1967, a mass poisoning occurred in South Africa in which 7 children died [2]. In 1992 in Argentina, 29 people died after consuming propolis syrups that contained high DEG concentrations; the drug was widely commercialized in Argentina to treat mild upper respiratory tract infections [3]. Thereafter, pediatric medicinal syrups contaminated with DEG caused the deaths of 33 children in India in 1998 [4], and 85 children in Haiti in 1995–1996 [5]. The most recent outbreak took place in Panama in 2006, in which more than 100 people died due to DEG poisoning [1]. In spite of these repeated mass DEG poisonings, only a few analytical methods for DEG analysis, by gas chromatography (GC) [3] and GC–mass spectrometry (MS) [6–8], have been reported. There is a pressing need to establish and improve the methods for analysis of DEG, especially for postmortem human samples. As a result of our extensive experience in analysis of DEG in postmortem samples, we hereby report an important characteristic of DEG that we observed during sample extraction. Common chemicals, including ethylene glycol, DEG, fatty acids, phospholipids, and cholesterols, were of the highest chromatographic purity commercially available. GC was performed on a Shimadzu GC-14 A equipped with a Shimadzu CR 4A integrator (Shimadzu, Kyoto, Japan). A J&W DB-Wax column (30 m 9 0.53 mm i.d., 1.5 lm film thickness, Agilent, Santa Clara, CA, USA) was used. The injector temperature was set at 250 C and the flameionization detector (FID) was set at 250 C. An initial oven temperature of 110 C was held for 2 min before a temperature ramp of 8 C/min was used to reach the final temperature of 210 C. The carrier gas was nitrogen (12 cm/min). For thin-layer chromatography (TLC) analysis of lipid components, an Iatroscan TLC–FID apparatus (Mitsubishi Kagaku, Iatron, Tokyo, Japan) was used . Tissue samples (whole blood, liver, and kidney) were obtained from 15 victims of massive intoxication who ingested propolis syrups contaminated with DEG. Ten millilitres of blood was obtained by puncturing the femoral vein and was analyzed after 24 h. Fifty grams each of liver and kidney was taken from the victims and stored frozen at -20 C until analysis (3 days after collection). We isolated DEG from tissues by continuous Soxhlet extraction with methanol for 12 h. The extract obtained from each sample was purified using a charcoal column and evaporated to 1 ml. Then 10 ll of the concentrated extract was used for analysis of DEG by GC–FID. When the methanol extract was stored in a refrigerator at 2–4 C for several hours, a semi-crystalline substance appeared, adhering to the glass wall of the test tube. When subjected to GC analysis, the supernatant fraction of the This article is for the special issue TIAFT2012 edited by Osamu Suzuki.