Preliminary evaluation of a quantitative ethylene glycol test in dogs and cats

Abstract

Ethylene glycol (EG) toxicity is commonly encountered in dogs and cats. The purpose of the current study was to determine if the Catachem test kit (Catachem Inc., Oxford, Connecticut) could precisely and accurately detect the presence of EG added to serum and plasma from 6 dogs and 4 cats. Serum and plasma samples were spiked at various concentrations of EG (0, 20, 60, and 100 mg/dl) and analyzed using the Catachem kit. Twenty randomly selected samples were also submitted for gas chromatography–mass spectroscopy (GC-MS) analysis of EG concentration, which was considered the gold standard. Inter- and intra-assay coefficients of variation (CVs) were calculated. Bland–Altman analysis was performed to compare the Catachem results to the GC-MS analyses. Analysis of serum samples showed a bias of 8.48 mg/dl (95% limits of agreement: 17.8 to −0.9 mg/dl) while spiked plasma samples had a bias of 7.32 mg/dl (18.1 to −3.5 mg/dl). Intra-assay CV was 0.7%. Interassay CV ranged from 1.2% to 2.0%. For all samples, the Catachem kit read higher than GC-MS values and slightly overestimated in vitro concentrations. The Catachem test kit is an accurate quantitative test for EG in dogs and cats that may aid in timely recognition of EG exposure. Because of the positive bias in all samples, some pets may receive treatment unnecessarily. However, animals with blood EG concentrations at or above the published lethal serum or plasma concentration will be readily identified so that treatment may be initiated.

Keywords

Cats Catachem dogs ethylene glycol quantitative toxicity

Introduction

Ingestion of ethylene glycol (EG) is a frequently encountered toxicity in dogs and cats.² Ethylene glycol is most commonly found in automotive antifreeze and is also used as a precursor to industrial polymers. It is a colorless, odorless, and sweet-tasting solution.¹¹ After ingestion, the solution is rapidly absorbed and reaches peak serum concentrations within 4 hr and has a half-life of 3 hr in human beings.¹¹ In dogs, the peak serum concentration is 2 hr and the half-life is 3.4 hr.⁹ The published minimum lethal dose in cats is 1.4 ml/kg and in dogs is 4.4–6.6 ml/kg of body weight⁴; death is usually a result of acute renal failure. Mortality after ingestion of EG is 96–100% in cats and 59–70% in dogs.^2,16

Clinical signs of EG toxicosis occur in 3 stages. In the first stage (30 min–12 hr postingestion), central nervous system signs predominate, and animals may display neurologic depression, ataxia, seizures, coma, or death.⁸ There are reports that cats can become unresponsive in as little as 9 hr after ingestion.⁵ These signs are thought to be a result of the aldehyde metabolites, hyperosmolarity, and metabolic acidosis, and resemble those of alcohol ingestion.⁸ The second stage occurs from 12 to 24 hr following ingestion. There is resolution of the initial neurologic signs with development of cardiopulmonary signs such as tachypnea or tachycardia. Common necropsy findings include pulmonary edema, congestion and hyperemia.⁸ The third and final stage of EG intoxication occurs 24–72 hr following ingestion of a lethal dose and is characterized by oliguric renal failure and associated clinical signs (anorexia, vomiting, and other signs of uremia),⁸ which may occur within 12 hr in some cats. The majority of the clinical signs in stages 2 and 3 are due to the metabolites of EG, which are largely dependent on the presence of available alcohol dehydrogenase (ADH) and the subsequent metabolism of EG to glycoaldehyde.⁷ Competitive inhibition of ADH can be used as a treatment for EG ingestion if it is administered early in the course of the disease.⁷

Four-methylpyrazole (4-MP) may be administered as a treatment for EG toxicity by acting as an inhibitor of ADH, and can successfully alleviate EG toxicity if administered early in the course of intoxication.¹⁸ Ethanol, an alternative treatment for EG ingestion, competes with EG for the binding site on ADH. Due to its higher affinity for ADH, ethanol competes with EG and allows excretion of unchanged EG.³ Treatment and hospitalization, especially if azotemia and oligoanuria has developed, may be costly and ultimately unsuccessful. Due to the mortality associated with EG intoxication, rapid and accurate diagnostic tests are essential for early diagnosis so that therapy may be instituted before life-threatening damage occurs.

To date, there is only 1 semiquantitative point-of-care EG test available to the veterinary market.^a The test requires 20 µl of plasma and is reported to detect levels of EG as low as 20 mg/dl (the feline lethal serum concentration). Unfortunately, at the time of publication, there is no published data to support this claim, and a previous study suggests that the accuracy of this test is poor.¹ Another point-of-care qualitative test^b is advertised to accurately detect EG levels above 50 mg/dl (the canine lethal serum concentration), which is above the lethal serum concentration in cats.⁸ Gas chromatography–mass spectroscopy (GC-MS) is considered to be the gold standard to detect serum or plasma levels of EG in human and veterinary patients,¹³ but this test is not a cage-side test, which may result in delays in diagnosis and initiation of therapy. A urine fluorescence test for sodium fluorescein, a common additive to some commercially available antifreeze solutions, may aid in the rapid detection of an antifreeze exposure, but this compound may be undetectable greater than 4–6 hr after ingestion and may cause false-negative results. Also, administration of other drugs such as benzodiazepines may cause false-positive results due to the florescence of their urine metabolites.¹⁹ Due to lack of available point-of-care testing, the decision to treat may be entirely based on clinical history of ingestion of EG or clinicopathologic data suggestive of intoxication (increased anion gap or hyperosmolarity).⁷

The Catachem EG reagent^c was developed in response to the perceived need for a rapid quantitative test to replace the more cumbersome GC-MS methodologies. The test was designed to measure serum or plasma EG levels using existing clinical chemistry analyzers.^d The test utilizes a bacterial-sourced glycerol dehydrogenase enzyme to catalyze an oxidation-reduction reaction of EG in the presence of nicotinamide adenine dinucleotide (NAD⁺). NADH produced by this reaction is then detected at a 340 nm wavelength where the absorption signal is proportional to the concentration of EG in the sample.¹⁷ To eliminate propylene glycol and glycerol interference, the test is programmed into the analyzer in such a way that measurement readings for EG do not begin until after any endogenous glycerol or propylene glycol has been consumed by glycerol dehydrogenase.¹² The chemistry is then run as a two-point kinetic test with the absorbance difference between the first and second read points being directly proportional to the level of EG in the sample.¹²

The purpose of the current study was to determine if the Catachem EG kit^c would quantitatively determine the amount of EG present in canine and feline serum and plasma with in vitro addition of EG. It was hypothesized that the Catachem EG kit would precisely and accurately detect the presence of EG at various concentrations associated with toxicity in companion animals.

Materials and methods

Analyzer and assay

A commercial assay^c for EG was installed on a clinical chemistry analyzer^d in accordance with manufacturer’s specifications. Intra-assay imprecision (within-assay) was estimated by determining EG concentration in pooled canine serum samples obtained as convenience samples from 5 healthy dogs (on the basis of complete blood cell counts and serum biochemistry screening) that had serum submitted to the clinical laboratory of the University of Georgia (Athens, Georgia) prior to anesthesia for elective orthopedic procedures. Linearity of the assay was determined by evaluating the pooled serum samples to which serial dilutions of EG were added, for final analytical concentrations of 250, 167, 125, 100, and 83 mg/dl.

Precision

Intra-assay (within-assay) precision was evaluated using 20 replicate measurements of EG concentration in each of 3 pooled serum samples containing 155 mg/dl of EG, within the same run. The EG solution (155 mg/dl) was provided by the manufacturer. Each assay was run in triplicate at the University of Georgia clinical pathology laboratory. The coefficient of variation (CV; [standard deviation/mean] × 100) for each sample was calculated.

Interassay (between-run) precision was calculated by determining EG concentrations in the pooled canine serum samples. The concentration of the EG in the tested solutions were 56 mg/dl, 155 mg/dl, and 248 mg/dl and were provided by the manufacturer. These samples were run twice daily on 14 consecutive days; all analyses were performed in triplicate. The CV for these assays was calculated as described above.

Analytical sensitivity

All procedures involving collection of blood were approved by the University of Georgia Animal Care and Use Committee. Six random-source adult mixed-breed dogs (estimated ages 4–8 years old) and 4 random-source adult Domestic Shorthair cats (estimated ages 2–5 years old) were used for donation of serum and plasma for study. All animals were healthy adults of both genders (canine: 4 spayed females, 2 neutered males; feline: 4 neutered males) that were not receiving any medications at the time of sampling. The animals were determined to be healthy on the basis of physical examination, complete blood cell count,^e and serum chemistry analysis.^d Blood was collected by jugular venipuncture^f from dogs without sedation, and from cats sedated with intramuscular ketamine^g (2 mg/kg) and dexmedetomidine^h (1µg/kg). Following blood collection, dexmedetomidine was antagonized using atipamezoleⁱ (5 µg/kg, intramuscularly). A total of 16–20 ml of blood was collected from each animal by jugular venipuncture.^f Eight to ten milliliters were anticoagulated using lithium heparin,^j and the remaining 8–10 ml were added to a collection tube with no additive.^j After a room temperature (22–24°C) rest of approximately 30 min, the samples were centrifuged at 1,500 × g for 10 min, and the plasma and sera were decanted and stored in 1-ml aliquots at −80°C until analysis.

Serum and plasma samples were subsequently thawed at room temperature, and EG^k was added to achieve calculated serum and plasma EG concentrations of 0, 20, 60, and 100 mg/dl. To avoid errors associated with measurement of small quantities of EG, the stock solution (99.7% EG) was diluted with 0.9% saline^l to working concentrations of 2% and 5% EG, immediately prior to addition to samples. Using these working concentrations, no more than 20 µl and no less than 10 µl of EG solution was added per milliliter of serum or plasma. After addition of EG, the samples were mixed using a vortex mixer for 5 sec.

Samples were batch analyzed in duplicate. A single, blinded investigator performed the analyses. Twenty samples were randomly selected^m and submitted for GC-MS evaluation of EGⁿ concentration to verify accuracy of the Catachem kit.

Interferences

Additional samples of canine serum that were not used in the in vitro studies described above were pooled, and these samples were spiked with a drug containing propylene glycol to determine if this substance interferes with the results of the EG assay. To achieve this, diazepam^o was added to serum to achieve concentrations of 2.5, 5, and 25 µg/ml of diazepam. The concentrations were based on multiples of the maximum concentration (Cmax) of diazepam achieved after a clinically relevant anticonvulsant dose (0.5 mg/kg) administered to dogs by intravenous injection.¹⁴ The samples were batch analyzed in duplicate by a single, blinded investigator.

Statistical analysis

All statistical analyses were performed with statistical software programs.^p,q Linearity and intra- and interassay precision data were evaluated using CVs, as described above. Ethylene glycol concentrations reported by GC-MS analysis were compared to those obtained from the Catachem kit using Bland–Altman and simple linear regression analysis. A Pearson correlation coefficient was also calculated to determine overall concordance between the Catachem method and GC-MS, the gold standard method.

Results

The CV for the intra-assay precision was 0.76%. The CV for the interassay precision was 2.1% for the low concentration, 1.7% for the medium concentration, and 1.2% for the high concentration (Table 1). All individual samples had EG detected, even the samples with no EG added. The highest EG concentration reported in samples without EG was 2.3 mg/dl in 1 cat plasma sample. The CV for the individual sample concentrations are listed in Table 2 for canine samples, and Table 3 for feline samples.

Table 1.

Interassay variables for low, medium, and high levels of ethylene glycol performed with the Catachem ethylene glycol test kit.*

Concentration	Low	Medium	High
Mean	62.4	156.1	226.8
Standard deviation	1.3	2.6	2.8
Coefficient of variation (%)	2.1	1.7	1.2
Replicates	20	20	20

Catachem Inc., Oxford, Connecticut.

Table 2.

Ethylene glycol values from plasma and serum samples in dogs as reported by the Catachem ethylene glycol test kit.*

	Serum concentration (mg/dl)				Plasma concentration (mg/dl)
	0	20	60	100	0	20	60	100
Mean (mg/dl)	0.3	28.1	80.0	121.5	0.6	25.5	73.9	119.0
Range	0.2–0.5	26.2–30.3	73.0–95.1	118.6–128.6	0.2–0.9	23.2–27.7	72.7–75.9	118.0–122.5
Standard deviation	0.1	1.3	7.7	3.7	0.2	1.3	0.9	1.7
Coefficient of variation (%)	33.3	4.6	9.6	3.0	33.3	5.1	1.2	1.4
N	6	6	6	6	6	6	6	6

Catachem Inc., Oxford, Connecticut. Samples were spiked with 0, 20, 60, and 100 mg/dl of ethylene glycol.

Table 3.

Ethylene glycol values from plasma and serum samples in cats as reported by the Catachem ethylene glycol test kit.*

	Serum concentration (mg/dl)				Plasma concentration (mg/dl)
	0	20	60	100	0	20	60	100
Mean (mg/dl)	1.5	26.1	72.4	117.2	1.4	26.8	75.4	116.1
Range	1.0–1.9	24.9–27.7	64.6–76.5	113.2–122.5	0.6–2.4	26.3–28.0	73.2–78.9	110.5–120.9
Standard deviation	0.4	0.9	4.7	3.6	0.7	0.6	2.0	3.7
Coefficient of variation (%)	26.7	3.4	6.5	3.1	50.0	2.2	2.7	3.2
Replicates	4	4	4	4	4	4	4	4

Catachem Inc., Oxford, Connecticut. Samples were spiked with 0, 20, 60, and 100 mg/dl of ethylene glycol.

Bland–Altman analysis of the concentration data obtained from GC-MS compared to all of the values generated by the Catachem kit showed a bias of 7.83 mg/dl (95% confidence interval [CI]: –2.1 to 17.8 mg/dl), with the Catachem kit reading higher than the GC-MS values at all points. Similar bias and limits of agreement were obtained when the serum samples and the plasma samples were evaluated separately (Table 4). Plasma and serum samples from 10 dogs and 8 cats were analyzed to determine the agreement between results obtained from the Catachem EG reagent and the gold standard method, GC-MS. Differences between both methods ranged from 1.8 mg/dl to 19.7 mg/dl, with an average difference of 7.83 mg/dl. The Bland–Altman plot showed agreement between the reference method and the Catachem reagent with the majority of EG concentrations (Fig. 1). Based on the plot, only 1 of the differences in measured concentrations were outside of the 95% acceptability limits. Overall concordance between the methods determined by Pearson correlation test was 0.9939.

Table 4.

Comparison of the Catachem ethylene glycol test kit* to gas chromatography–mass spectroscopy samples of plasma and serum.

	Serum samples	Plasma samples	All samples
Bias (mg/dl)	7.32	8.48	7.83
Standard deviation	5.5	4.8	5.1
95% limits of agreement	−3.50 to 18.14	−0.89 to 17.84	−2.14 to 17.81

Catachem Inc., Oxford, Connecticut.

Figure 1.

Bland–Altman difference plot of the data comparing ethylene glycol concentrations (mg/dl) measured with the Catachem ethylene glycol reagent and by the gold standard, gas chromatography–mass spectroscopy (GC-MS), in 18 paired samples. The solid line indicates the bias (defined as the mean difference between the methods) of the results obtained by the 2 methods; the dashed lines represent the 95% confidence intervals.

Linear regression analysis revealed a slope of 0.917 (95% CI: 0.863–0.971; P < 0.0001) and a y-intercept of −1.946 (95% CI: –6.177 to 2.286; P = 0.344; Fig. 2). A slope value significantly close to 1 implies that the new method is unbiased and that proportional systematic error is absent. The 95% CI for the y-intercept contains 0; therefore indicating that deviation from the expected value of zero (constant systematic error) is not significant¹⁰ (http://www.westgard.com/lesson44.htm#3). Samples that were spiked with a propylene glycol additive did have positive values for EG; however, the highest reported value of EG was 10.0 mg/dl in the most concentrated sample (Table 5).

Figure 2.

Linear regression plot of the data comparing ethylene glycol concentrations (mg/dl) measured with the Catachem ethylene glycol reagent and by the gold standard, gas chromatography–mass spectroscopy (GC-MS), in 18 samples. The regression line and its equation are shown.

Table 5.

Results of the Catachem ethylene glycol (EG) test* with samples spiked with diazepam, a propylene glycol additive.

	Concentration of diazepam (mcg/ml)
	2.5	5.0	25
Mean EG concentration test result (mg/dl)	1.4	2.8	10.0

Catachem Inc., Oxford, Connecticut.

Discussion

The Catachem EG test kit was both precise and accurate in detecting EG in serum and plasma samples. The positive bias when compared to GC-MS is not necessarily a disadvantage. While this may lead to treatment of EG toxicity in some animals that have not ingested a toxic dose, it will ensure that all animals with actual EG ingestion are detected and treated appropriately.

Compared to other commercially available EG test kits,¹ the Catachem kit has been shown to be more accurate (bias 7.83 of all samples) and precise (CV 0.76%). A previous study¹ evaluated a commercial EG test and found the sensitivity for identification of low concentration samples (20 mg/dl of EG) to be between 56% and 67%. This may result in a failure to identify up to 40% of animals who test positive for EG.¹ A 2008 study¹³ tested a point-of-care qualitative EG test kit,^r and compared results to GC-MS. This study reported that the sensitivity and specificity were 100% and 88%, respectively. The lowest concentration of EG that tested positive was 27 mg/dl.¹³ The 2008 study suggested that the EGT kit^r may be useful as a point-of-care test for EG in human beings.

In the current study, all of the samples for EG were positive, even the unadulterated samples. However, the mean concentration that was reported for all canine and feline samples without added EG was 0.45 mg/dl and 1.45 mg/dl, respectively, both much lower than the reported toxic plasma or serum concentrations for dogs and cats.⁸ While this does lower the specificity of the Catachem test, it should not affect the clinical course of treatment for any animal. If the serum or plasma concentration is less than the reported lethal serum or plasma concentration (20 mg/dl in cats and 50 mg/dl in dogs), consideration may be given to treat the animal supportively and not administer ethanol or 4-MP. However, due to the fact that EG levels will decrease dramatically after it has been metabolized, treatment may by warranted based on the clinical signs seen in the particular patient. Given that this test relies of the absorption of light following a chemical reaction, it is possible that other substances in the plasma may cause interference. Additional testing in animals with elevations in serum bilirubin or other analytes is indicated before this test may be recommended for use in those patients. Because clinicopathologic signs of renal insufficiency (e.g., elevations in serum creatinine) do not occur until days following intoxication, this test should be accurate for identifying animals in the critical early stages of disease without interference from other substances in the blood.

Because seizures are a common presenting clinical sign of EG intoxication, animals presenting to the emergency room may receive therapy with intravenous diazepam prior to additional diagnostics and diagnosis of EG ingestion. Diazepam is solubilized in propylene glycol, which may theoretically interact with the methodology of the Catachem test. For this reason, plasma samples were also spiked with diazepam in a concentration (based on diazepam dose and distribution) that a patient might receive clinically, as well as concentrations that were 100 times the reported Cmax (in mg of diazepam).¹⁴ For these samples, the Catachem analyzer did report a positive EG value. However, a dose that is 100 times higher than the recommended dose for a seizuring dog or cat¹⁵ did not result in a report of EG levels that would be considered toxic in dogs or cats. Propylene glycol is also present in other commonly used pharmaceuticals, such as phenobarbital, lorazepam, etomidate, sorbitol, digoxin, and trimethoprim–sulfamethoxazole,⁶ and the possibility for a small degree of inaccuracy should be taken into account when using these products. Based on the findings in the current study, the propylene glycol that is used as a carrier for diazepam would not interfere with interpretation of the EG test kit.

The Catachem EG test kit is designed as a module on the Hitachi-P chemistry analyzer,^d which is a large, specialized chemistry analyzer, and may only be present at universities, reference laboratories, or large specialty referral practices. This application was reasonably easy to install with the manufacturer’s specifications. At the start of the present study, a qualitative point-of care EG test kit from Catachem using similar methodology was in the process of being released (David Templeton, Catachem Inc., personal communication, 2012).

One limitation to the accurate use of the Catachem EG test kit is that the results are based on a circulating level of EG. Due to rapid metabolism of the parent compound, it is possible that in vivo, the EG would be broken down to its metabolites and the quantitative value would read lower than concentrations associated with toxicity. In this case, it is indicated to use clinical signs such as patient history, physical examination, and the presence of oliguria along with laboratory changes such as calcium oxalate crystals on urinalysis, azotemia, a high anion gap metabolic acidosis, or an increased serum osmotic concentration to support a diagnosis of EG intoxication. As toxicity progresses, it may be difficult to detect EG in the blood after toxic metabolites have been produced. However, because 20% of ingested EG is excreted in the urine,⁸ with peak urine concentration in dogs occurring within 6 hr after ingestion, analysis of urine may be another option to diagnose EG exposure, although this was not studied in the current report.^7,11

Diagnosis of EG in the clinical setting must be made based on clinical signs as well as history and clinicopathologic findings. The Catachem EG test kit may help to confirm clinical suspicions and may aid in quantifying the amount of EG ingested by the animal. While it may occasionally lead to unnecessary treatment, all animals that ingest a toxic dose of EG should be identified and able to receive treatment based on the Catachem test results. Therefore, this test may be a valuable screening test to determine if a patient has ingested EG.

Footnotes

a.

Kacey ethylene glycol test, Kacey Inc., Asheville, NC.

b.

EGT test kit, Allelic Biosystems, Kearneysville, WV.

c.

Catachem Inc., Oxford, CT.

d.

Hitachi P-modular 800, Roche Diagnostics, Indianapolis, IN.

e.

Advia, Bayer Healthcare, Shawnee Mission, MO.

f.

19 g 3/4′′ Transiva winged infusion set, Abbott Laboratories, N. Chicago, IL.

g.

Pfizer Animal Health, New York, NY.

h.

Pfizer Inc., New York, NY.

i.

Pfizer Inc., New York, NY.

j.

BD Vacutainer, Franklin Lakes, NJ.

k.

Sigma-Aldrich, St. Louis, MO.

l.

Hospira Inc., N. Chicago, IL.

m.

, accessed 1/20/2012.

n.

Diagnostic Center for Population and Animal Health, Michigan State University, Lansing, MI.

o.

Diazepam (5 mg/ml), Abbott Laboratories, N. Chicago, IL.

p.

Prism 5.04, GraphPad Software, La Jolla, CA.

q.

EP-evaluator, Data Innovations, South Burlington, VT.

r.

PRN Pharmacal Inc., Pensacola, FL.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Partial funding for this study was provided by Catachem Inc., Oxford, CT.

References

Acierno

Serra

Johnson

Mitchell

: 2008, Preliminary validation of a point-of-care ethylene glycol test for cats. J Vet Emerg Crit Care (San Antonio)18:477–479.

Barton

Oehme

: 1981, The incidence and characteristics of animal poisonings seen at Kansas State University from 1975 to 1980. Vet Hum Toxicol 23:101–102.

Connally

Hamar

Thrall

: 2000, Inhibition of canine and feline alcohol dehydrogenase activity by fomepizole. Am J Vet Res 61:450–455.

Connally

Thrall

Hamar

: 2010, Safety and efficacy of high-dose fomepizole compared with ethanol as therapy for ethylene glycol intoxication in cats. J Vet Emerg Crit Care (San Antonio) 20:191–206.

Dial

Thrall

Hamar

: 1994, Comparison of ethanol and 4-methylpyrazole as treatments for ethylene glycol intoxication in cats. Am J Vet Res 55:1771–1782. Erratum in Am J Vet Res 1995 56:825.

Gatselis

Liamis

Makaritsis

Dalekos

: 2012, Metabolic acidosis during treatment of mushroom poisoning: a diagnostic pitfall. Intern Med 51:1077–1080.

Gaynor

Dhupa

: 1999, Acute ethylene glycol intoxication. Parts I and II. Compend Contin Educ Vet Pract 21:1014–1023 and 21:1124–1133.

Grauer

Thrall

: 1982, Ethylene glycol (antifreeze) poisoning in the dog and cat. J Am Anim Hosp Assoc 18:492–496.

Hewlett

Jacobsen

Collins

McMartin

: 1989, Ethylene glycol and glycolate kinetics in rats and dogs. Vet Hum Toxicol 31:116–120.

10.

Hopkins

: 2004, Bias in Bland-Altman but not regression validity analyses. Sportscience 8:42–46.

11.

Jammalamadaka

Raissi

: 2010, Ethylene glycol, methanol and isopropyl alcohol intoxication. Am J Med Sci 339:276–281.

12.

Juenke

Hardy

McMillin

Horowitz

: 2011, Rapid and specific quantification of ethylene glycol levels: adaptation of a commercial enzymatic assay to automated chemistry analyzers. Am J Clin Pathol 136:318–324.

13.

Long

Nelson

Hoffman

: 2008, A rapid qualitative test for suspected ethylene glycol poisoning. Acad Emerg Med 15:688–690.

14.

Platt

Randell

Scott

.: 2000, Comparison of plasma benzodiazepine concentrations following intranasal and intravenous administration of diazepam to dogs. Am J Vet Res 61:651–654.

15.

Plumb

: 2011, Plumb’s veterinary drug handbook, 7th ed., pp. 303–308. PharmaVet, Stockholm, WI.

16.

Rowland

: 1987, Incidence of ethylene glycol intoxication in dogs and cats seen at Colorado State University Veterinary Teaching Hospital. Vet Hum Toxicol 29:41–44.

17.

Standefer

Blackwell

: 1991, Enzymatic method for measuring ethylene glycol with a centrifugal analyzer. Clin Chem 37:1734–1736.

18.

Tart

Powell

: 2011, 4-Methylpyrazole as a treatment in naturally occurring ethylene glycol intoxication in cats. J Vet Emerg Crit Care (San Antonio) 21:268–272.

19.

Winter

Ellis

Snodgrass

: 1990, Urine fluorescence using a Wood’s lamp to detect the antifreeze additive sodium fluorescein: a qualitative adjunctive test in suspected ethylene glycol ingestions. Ann Emerg Med 19:663–667.