A-087 A case study on discrepancies in lactate measurement between blood gas and core laboratory instruments due to beta-hydroxybutyrate interference

Abstract Background Accurate lactate measurement is essential in medical diagnostics, especially for identifying life-threatening conditions like lactic acidosis. This condition arises when lactate builds up in the bloodstream due to hypoxia, sepsis, or impaired metabolism, leading to serious acid-base disturbances. Common methods for measuring lactate include blood gas analyzers and core laboratory instruments, which specifically measure L-lactate. However, these methods can yield differing results due to interference from other substances, variations in measurement principles, or calibration issues. We report a case involving significant discrepancies in lactate measurements in a patient with acute myeloid leukemia (AML) and diabetic ketoacidosis. Dilution experiments indicated potential interferences in instruments using the Trinder reaction, resulting in inaccurate readings. Methods Our institution uses GEM 5000 for blood gas analysis and the DXC core lab instrument for lactate measurement. The Gem 5000 measures lactate through L-lactate oxidase with electrochemical oxidation, while the DXC employs the Trinder reaction, where lactate is converted to pyruvate, producing hydrogen peroxide (H2O2) that reacts with hydrogen donor and 4-amino antipyrine (4-AAP) to form a chromophore then monitoring the change in absorbance at 560 nanometers to determine lactate concentration. We measured lactate levels in a patient with acute myeloid leukemia (AML) using both assays, which showed inconsistent results. After repeating the measurement with a new specimen, results were consistent. The patient has no history of ethylene glycol ingestion but does have type 2 diabetes mellitus. To further investigate interferences, we conducted dilution experiments and sent specimens to an outside lab with a different assay, along with spike-in experiments using a lactate calibrator with patient specimens. Results We received two specimens from the patient, both showing elevated lactate levels (∼7.5 mmol/L) on the Gem 5000, while the DXC reported much lower results (∼1.3 mmol/L). Both specimens had a high beta-hydroxybutyrate (BHOB) level of 2.90, and urinalysis indicated moderate ketones, suggesting diabetic ketoacidosis (DKA). A third specimen later tested on the DXC showed a lactate level of 5.6 mmol/L, consistent with blood gas results, but a lower BHOB level of 0.13. We performed dilution experiments (Neat=1.1 mmol/L, Dilution X2= 2.2 mmol/L, Dilution X4=3mmol/L, Dilution X8= 3.9mmol/L and Dilution X16 = 5.4 mmol/L). All three specimens were sent out for different laboratory using different lactate measurement method, which confirmed the core instrument results. A lactate calibrator gave a reading of 2.1 mmol/L, which decreased to 1.1 mmol/L when spiked with a BHOB calibrator (0.98). Spiking a patient specimen with high BHOB (2.9) and a lactate calibrator (6.5) resulted in a lactate value of 4.3 mmol/L, confirming the negative effect of high BHOB on lactate. Conclusion Dilution and spike-in experiments suggested that high beta-hydroxybutyrate (BHOB) levels interfered with the Trinder reaction, leading to falsely low lactate readings, which could explain the discrepancies in lactate measurements. Further research is needed to confirm the findings of this case report and to investigate the underlying mechanisms responsible for this interference particularly in patients with conditions like DKA.