Verificación de la intercambiabilidad de resultados entre equipos duplicados
- Enrique Prada de Medio
- Raquel Blázquez
- Carmen Perich
- Gabriela Gutiérrez Bassini
- Daniel Pineda Tenor
- Ana Isabel Álvarez Ríos
- Ángel Salas García
- José María Jou Turallas
- María Ángeles Cuadrado Cenzual
- Jorge Morancho Zaragoza
- Carmen Ricós Aguilá
- María José Cobo del Hoyo
- Rafael Calafell Clar
- Francisco Ramón Bauzá
ISSN: 1888-4008
Year of publication: 2014
Volume: 7
Issue: 1
Pages: 17-24
Type: Article
More publications in: Revista del laboratorio clínico
Abstract
A fairly common situation in clinical laboratories is the existence of different analyzers used interchangeably for determining the same tests. It must be ensured that the level of agreement between the results obtained is independent of the systems or analytical platforms used, and that the measurements are comparable. Otherwise, we can create clinician misunderstanding and a possible misinterpretation of the results. Safety of the patient may be at risk. For this reason, this workgroup established two different situations: implementation of new equipment and review of previously verified equipment. In the first case, an analysis of the differences should be performed using the biological variability (VB) as a means to establish the maximum difference allowed. When VB data are not available, the statistical criterion should used for this calculation. A second possibility would be the use of linear regression for the study of comparability of analyzers. The use of the Clinical and Laboratory Standards Institute Protocol (CLSI), which allows up to 10 analyzers to be verified with a maximum of 2 analytical series, is proposed in the case of previously checked equipment. If the results provided by the different analyzers do not exceed the maximum deviation established, they would be considered as a single virtual equipment, otherwise, the results are not interchangeable, and the laboratory report should indicate clearly the analyzer used in each patient, as well as the appropriate reference range.