Lipemic Interference in Basic Metabolic Panels: Increasing the Lipemia Index Threshold in Order to Decrease the Frequency of Ultracentrifugation

Abstract

Abstract Introduction Lipemia in clinical chemistry samples is a problematic form of interference. Clearing these samples for routine testing can be time consuming and increases the turn-around time for these specimens. In our laboratory, samples with a lipemia index >50 (L-index) are manually inspected and visibly lipemic specimens are cleared by ultracentrifugation. Objective The objective of this study was to determine at what L-index ultracentrifugation of lipemic BMP specimens is necessary prior to sample testing to ensure accurate results. Methods Specimens consisted of routinely ordered basic metabolic panels (BMP) that met current criteria for ultracentrifugation, which included an L-index >50 as measured on the Abbott Architect c8000 and visual lipemia. Specimens meeting these criteria were ultracentrifuged and retested. The difference of the pre-ultracentrifuged and post-ultracentrifuged result was evaluated and put into a percent to find the ‘percent difference’ and evaluated against the total allowable error (TEa) for each analyte. If the difference observed following ultracentrifugation was less than or equal to 50% of the TEa, clearance of lipemia by ultracentrifugation was considered unnecessary. Values from all BMP component tests were analyzed in order to find an L-index threshold at which samples need to be ultracentrifuged which could be applied to the entire panel. The report of lipemic indices for BMPs for the month of January 2020 were extracted from the laboratory information system to evaluate the potential impact of altering the L-index threshold for ultracentrifugation. Results Based on the acceptance criteria of ≤50% of TEa, L-index thresholds for Na, K, Cl, calcium, glucose, creatinine, CO2 and BUN were <203, <410, <287, <387, <410, <285, < 153 and <285, respectively. All the calculated differences or percent differences for each analyte did not exceed 50% of the TEa for a given analyte when the L-index was 150 or less. Adjusting the L-index to 150 and applying it to the 195 lipemic BMP samples in January 2020, would have potentially decreased the number of samples requiring ultracentrifugation to 24 lipemic BMPs (88% reduction). Conclusion These data suggest that an L-index greater than 150 can be used for all analytes within a BMP as the threshold for requiring ultracentrifugation. The BMP is one of the most frequently ordered tests in our laboratory and consistently accounts for a substantial portion of the lipemic samples that require ultracentrifugation. Increasing the L-index at which samples will be ultracentrifuged from 50 to 150 would potentially result in an 88% reduction in one month of BMP samples requiring ultracentrifugation.