Answers are grounded in this paper and cite their sources.
This analysis shows non-linearity in lactate responses with elevated glycolic acid levels, suggesting ethylene glycol poisoning.
Background Numerous papers have noted the difference in measurements of lactate across different platforms that can occur in ethylene glycol (EG) poisoning, due to method-dependent degree of interference of the glycolic acid (GA) metabolite of EG. This is known as the “lactate gap”. Given non-linearity of GA interference for the Roche Cobas lactate assay, we investigated whether such non-linearity might itself be used as an aid to diagnosis of EG poisoning. Methods Lactate concentration ([LAC]) measurements were made on the Roche Cobas c502 analyzer for serial dilutions (relative concentrations of 1, 0.75, 0.5, 0.25) of a pooled plasma (Li-heparin) sample spiked with 0, 5, 10, or 20 mmol/L GA, using saline as a diluent. Experiments were conducted in triplicate. Results Results of a representative experiment are shown in Figure. In this example, for the unspiked pooled plasma sample ([LAC]=4.3 mmol/L; reference interval=<2.2 mmol/L), the dilution curve was linear, as is observed for all LAC samples within the assay’s linear range (<20 mmol/L). For spiked samples, dilution curves were patently nonlinear for [GA] = 10 or 20 mmol/L. For instance, for [GA] = 10 or 20 mmol/L, Y-axis intercepts for 2-point curve segments, based on results for dilutions of 1.0 and 0.75, were significantly higher than zero (p < 0.05), whereas zero is the expected result for any segment of a dilution curve having no interference. Conclusions Non-linearity of a saline dilution series for [LAC] as shown using the Roch Cobas LAC assay is characteristic of positive assay interference by GA. In the context of investigation of possible EG poisoning, with an elevated anion gap characteristic of presence of GA, non-linearity of low dilutions (1, 0.75) in the Roche lactate assay for an elevated lactate result may specifically suggest presence of GA for GA >= 10 mmol/L.
No takes yet. Share an insight, caveat, or question.
Stickle et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: