Objective – To evaluate the sensitivity and specificity of a newly available, semi-quantitative, cage-side test for the detection of ethylene glycol (EG) toxicosis in cats.
Design – Prospective, laboratory study.
Setting – University teaching hospital.
Animals – This study utilized samples from 57 cats, whose blood had been anticoagulated with EDTA and submitted to the hospital's laboratory for a complete blood count. Samples were centrifuged, and the plasma separated, aliquoted, and immediately frozen at −30 °C.
Interventions – Samples were randomly divided into 2 primary groups (Group 1: no EG added, Group 2: EG added). Twenty microliters of plasma from each of the Group 1 samples was applied directly to the test strip. Plasma samples from Group 2 had EG added at different concentrations to achieve approximate final concentrations of 20, 60, or 80 mg/dL. These samples were then applied to the test strip.
Measurements – Two readers who were blinded to the sample preparation procedure and isolated from each other were asked to categorically interpret the colorimetric reaction on the randomly presented test strips.
Main Results – The agreement of the 2 reviewers at the 3 different levels of EG concentrations (20, 60, 80 mg/dL) were 0.7, 0.7, and 0.5, respectively. Thus, the readers demonstrated substantial agreement while reading the 2 lower concentrations, while at 80 mg/dL the level of agreement was moderate. Overall, the sensitivity of the assay increased as the concentration of EG increased (reviewer 1: 67%, 67%, 86%; reviewer 2: 56%, 89%, 100%), while the specificity of the assay decreased with increasing concentrations of EG (reviewer 1: 77%, 45%, 50%; reviewer 2: 77%, 53%, 25%).
Conclusions: Because of the likelihood for false negatives and false positives, results from this test must be viewed in light of clinical data and should not be relied upon as a lone diagnostic test.