Impact and frequency of IV fluid contamination on basic metabolic panel results using quality metrics

Objectives Clinical laboratories invest substantial time and resources to mitigate measurement error but potential errors during the preanalytical phase of testing are not subjected to the same level of scrutiny. Herein, we assess the proportions of intravenous (IV) fluid contamination sufficient to exceed common performance metrics and compare it to contaminated results flagged by current protocols. Methods Basic metabolic panels performed between 01/2017 and 07/2022 were extracted from the laboratory information system (n=928,742). Contamination was simulated for common IV fluid types. The thresholds at which contaminated results exceeded total allowable error (TEa), reference change values (RCV), or changed normality/critical flags were calculated. The mixture ratio of IV fluid contamination detected by technologists during routine analysis was estimated. Results The TEa and RCV was exceeded at a mixture ratio ≤0.10 for chloride, glucose, calcium, and potassium for both normal saline (NS) and 5 % dextrose in water (D5W). At a simulated mixture ratio of 0.10, 51.39 % of calcium and 21.17 % of potassium results would be expected to be incorrectly reported with an abnormal/critical flag with NS contamination and 99.74 % of sodium and 100 % of glucose results to be incorrectly flagged with D5W. Retrospective results flagged as contaminated revealed a median mixture ratio of 0.18 and 0.24 for D5 and non-D5 fluids. Conclusions At a mixture ratio of at least 0.10, IV fluid contamination causes relevant error between patients’ true concentrations and those reported. However, current procedures cannot reliably detect 10 % contamination.