A protocol for testing the stability of biochemical analytes. Technical document

• Published in: Clinical chemistry and laboratory medicine Vol. 57; no. 12; p. 1829
• Main Authors: Gómez-Rioja, Rubén, Segovia Amaro, Marta, Diaz-Garzón, Jorge, Bauçà, Josep Miquel, Martínez Espartosa, Débora, Fernández-Calle, Pilar
• Format: Journal Article
• Language: English
• Published: Germany 26-11-2019
• Subjects: Blood Specimen Collection - methods; Humans; Laboratories; Specimen Handling - methods; Specimen Handling - standards; Time Factors; Urine Specimen Collection - methods; Urine Specimen Collection - standards; preanalitycal phase; sample stability; guideline
• ISSN: 1437-4331
• DOI: 10.1515/cclm-2019-0586

Stability of a measurand in a specimen is a function of the property variation over time in specific storage conditions, which can be expressed as a stability equation, and is usually simplified to stability limits (SLs). Stability studies show differences or even inconsistent results due to the lack of standardized experimental designs and heterogeneity of the chosen specifications. Although guidelines for the validation of sample collection tubes have been published recently, the measurand stability evaluation is not addressed. This document provides an easy guideline for the development of a stability test protocol based on a two-step process. A preliminary test is proposed to evaluate the stability under laboratory habitual conditions. The loss of stability is assessed by comparing measurement values of two samples obtained from the same patient and analyzed at different time points. One of them is analyzed under optimal conditions (basal sample). The other is stored under specific stability conditions for a time set by the laboratory (test sample). Differences are expressed using percentage deviation (PD%) to facilitate comparison with specifications. When the preliminary test demonstrates instability, a comprehensive test is proposed in order to define the stability equation and to specify SLs. Several samples are collected from a set of patients. The basal sample is analyzed under optimal conditions, whereas analysis of test samples is delayed at time intervals. For each patient PD% is calculated as the difference between measurements for every test sample and its basal one and represented in a coordinate graph versus time.