A Gradual Positive Bias in Chloride Measurements with a Major Automated Chemistry Platform.

Background: Chloride is an extracellular anion commonly measured using ion-selective electrode methods standardized to traceable reference materials. This study investigates the gradual appearance of a positive bias in chloride measurements using the Siemens Healthineers Atellica CH platform that was undetectable via routine quality control or proficiency testing.

Methods: The bias was characterized using retrospective analysis of laboratory values from presumable healthy outpatients (ICD-10 code Z00.00), reference interval verifications, peer group means from the College of American Pathologists' proficiency testing C survey, and a method comparison study with an alternate platform.

Results: Retrospective analysis of patient data (2017-2024) revealed a gradual shift in median patient chloride concentrations distinct from stable trends observed for other electrolytes. In 2024, reference interval verification revealed that 64.3% of chloride results were above the reference interval, even though the verification was successful in 2021. Proficiency testing results were acceptable compared to peer method; however, revealed a mean positive bias of 3.19 mmol/L when compared to other indirect ISE methods (2019-2024). A method comparison with an alternate chemistry platform demonstrated a mean bias of +4.3 mmol/L using remnant patient samples.

Conclusions: The use of peer targets for quality control and proficiency testing concealed the detection of a gradual drift in chloride measurements in our laboratory. Analysis of aggregate patient data using ICD-10 codes represents a powerful tool for laboratories when assay performance is questioned. These findings highlight the need for vigilant monitoring of assay drift even for routine analytes to avert potential impact on patient care.