The major toxic alcohols that pose a risk to human health and for which measurement of levels is requested from the clinical laboratory include ethanol, ethylene glycol, isopropanol, and methanol. Most clinical laboratories are equipped to accurately quantitate ethanol levels, mainly by an enzymatic assay on an automated chemistry analyzer platform. However, the measurement of the other 3 alcohols is more challenging.1,2 Some relevant laboratory tests that assist in the differential diagnosis are listed in Table 1.
An increase in the osmolal gap is evident in the presence of all of the alcohols. In addition, methanol, ethanol, and ethylene glycol also result in an increase in metabolic acidosis with an increased anion gap, separating them from isopropanol. A relevant clue to isopropanol ingestion is positivity for acetone in the urine or blood without attendant hyperglycemia. In addition to an increase in the osmolal gap and anion gap acidosis, ethylene glycol ingestion can result in a positive urine analysis for the presence of calcium oxalate crystals (monohydrate and dihydrate) because it is converted to oxalate, which can chelate calcium and result in hypocalcemia. However, it needs to be emphasized that that the deposition of calcium oxalate can occur in many tissues, including the kidney, and can result in acute renal failure, pulmonary dysfunction, myocardial dysfunction, and impaired neurologic function. Furthermore, the documentation of calcium oxalate crystals is hampered the fact that it is only detected in around 50% of patients.
During metabolism, ethylene glycol breaks down to glycolic acid and then to oxalic acid by the action of alcohol dehydrogenase. Oxalate binds calcium, leading to calcium oxalate crystals, which are deposited in many tissues but are most readily detected in the urine. The glycolic acids contribute to central nervous system manifestations, mortality, and a significant metabolic acidosis (anion gap acidosis).3 These metabolites also contribute to significant renal tubular necrosis. Thus, while the tests provide valuable clues to clinicians for differential diagnosis, measurement of the relevant alcohol is the most definite test to rule in or exclude a certain alcohol poisoning or ingestion.
The preferred assays are summarized in Table 1. Thus, for methanol and isopropanol, gas chromatography (GC) is the confirmatory assay. For ethylene glycol, the preferred method is GC–flame ionization detection (FID). However, these instruments and the needed expertise are not readily available in most clinical laboratories.
In this issue of the Journal, Jeunke et al4 present the validation of a modified kinetic time and data analysis technique in a rapid enzymatic assay that can be adopted on open chemistry analyzer platforms for the rapid measurement of ethylene glycol levels. The Catachem assay is based on a bacterial enzyme, glycerol dehydrogenase.5–7 This enzyme oxidizes ethylene glycol in the presence of NAD (nicotinamide adenine dinucleotide), generating NADH (the reduced form of NAD), resulting in an increase in absorbance at 340 nm, which is detected spectrophotometrically by automated chemistry analyzers. In the original method proposed by Catachem (Bridgeport, CT), the difference between the absorbance readings at 2 time points was used to determine the ethylene glycol concentration.
In this study, the modification by Juenke et al4 was to determine the slope of the line by measuring absorbance differences at several points, starting at a later time point than the original 2-point design. For samples containing ethylene glycol, the difference between the 2 determinations is minimal; however, for samples containing compounds that mimic ethylene glycol (eg, propylene glycol), the difference between the 2 methods is substantial because the slope of the line flattens after an initial increase in absorbance. In this way, Juenke et al4 were able to demonstrate that the modified kinetic parameters applied to the Catachem reagent system accurately distinguished between propylene glycol, 3-butanediol, and ethanol from the true measurement of ethylene glycol. The modified assay, in addition to being precise (intra-assay coefficient of variation, <3.0%; interassay coefficient of variation, <8.0% ) and linear (up to 300 mg/dL), was also able to eliminate false-positives from a variety of interfering substances, including combinations of ethylene glycol, glycerol, propylene glycol, 2,3-butanediol, formic acid, n-propanol, isopropanol, acetone, methanol, ethanol, glycolic acid, polyethylene glycol, oxalic acid, glyoxal solution, glyoxylic acid, 1,2-butanediol, 1,4-butanediol, 1,3-propanediol, 1-butanol, 1,3-butanediol, DOT 3 brake fluid, and 1-octanol. These were generally flagged with a rate error. Also, an important observation is that there was no interference with fomepizole, commonly used for treatment of ethylene glycol poisoning.8 Juenke et al4 also performed patient sample comparison and showed excellent correlation with the “gold standard” method, GC-FID. Furthermore, there seemed to be no interference with glycerol under the defined assay conditions.
The advantage of this automated assay over the GC method is a substantial decrease in labor costs and a significant reduction in turnaround times by approximately 10 hours. Results can be made available in 30 minutes instead of 3 hours. Thus, this assay is clearly advantageous to an emergency department physician dealing with this serious disorder. Future studies will need to confirm the performance and usefulness of this automated testing in other clinical laboratories. Also, assays for glycolic acid need to be more readily available because levels correlate with symptomatology and mortality. Testing for glycolic acid remains a clear advantage of GC, which can readily assay ethylene glycol and its toxic metabolite.
The advent of this modified enzymatic assay goes a long way in enhancing the role of the laboratory in the management of the life-threatening condition, ethylene glycol poisoning. The assay needs to be embraced by laboratorians to confirm its validity in identifying this serious form of intoxication, which constituted 2% of poisonings reported in the United States in 1999.9