Evaluating agreement between separate capillary sampling sites and venous concentrations of β-hydroxybutyrate measured by a point-of-care device and liquid chromatography tandem mass spectrometry.

Abstract

β-hydroxybutyrate (BHB) is recommended as a measure of ketosis and is often assessed by capillary samples on point-of-care (POC) meters. However, liquid chromatography tandem mass spectrometry (LC-MS/MS) is considered the gold standard for assessing venous samples. The POC device KetoSure^TM is recommended only for capillary samples from finger pricks. So far, KetoSure^TM has not been compared to LC-MS/MS just as it has not been evaluated if the sampling site influences the BHB concentration. Blood samples were collected from 16 healthy, fasting individuals before and after ingestion of ketone monoester. BHB concentrations were measured in capillary samples from the earlobe and fingertip, and in venous blood using KetoSure^TM. Venous plasma samples were collected for BHB quantification using LC-MS/MS. No sign of significant difference between values of BHB measured from the two capillary sampling sites were found. Interestingly, significantly higher values of BHB were measured in capillary samples compared to venous samples reflecting a systematic proportional relationship. No systematic difference was observed in the measured BHB concentrations when comparing KetoSure^TM and LC-MS/MS results: However, a significant mean bias of 32% reflects a skewness at very low BHB concentrations. In conclusion, capillary BHB concentration did not exhibit variation between the earlobe and fingertip. Conversely, a significant bias was observed between venous and capillary blood and between the POC and LC-MS/MS methods. It is recommended that caution be exercised if individual monitoring of BHB changes encompasses both capillary and venous sampling.