Key points are not available for this paper at this time.
Abstract Background The objective of this study is to verify if the results of the Amylase and Glucose methods, measured by two different analyzers: Atellica Solution® and CI® (Siemens Healthineers), are interchangeable. Methods To carry out the intermethod comparison study of the Amylase and Glucose methods, 40 patient samples were selected for each assay, with concentrations across the entire measurement range. Statistical analysis was performed using Method Validator v.19 software, through Bland-Altman mean difference analysis and Passing-Bablok regression. Statistics are expressed with their 95% confidence intervals (CI). Results AMYLASE: After analyzing the Bland-Altman differences, a statistically significant systematic error was found, as the mean difference between the results obtained by Atellica Solution® and CI® was 28.1 with a 95% CI that does not include the null value, with higher results in CI®. Regarding the Passing Bablok analysis, systematic differences, both proportional and constant, were observed as the confidence interval of the intercept does not include the zero value, and that of the slope does not include the value 1.GLUCOSE: After Bland-Altman analysis, a systematic error was found, as the mean difference between the results obtained by Atellica Solution® and CI® was 2.1, with slightly higher results in CI®. Regarding the Passing Bablok analysis, there were no systematic differences of constant type since the confidence interval of the intercept includes the zero value, and a slight proportional systematic error based on the slope as its 95% CI does not include the value 1. Conclusions After evaluating the results, it is concluded that the Atellica Solution® and CI® analyzers are not interchangeable for the Amylase assay, so new reference values should be established. For the Glucose assay, both analyzers are interchangeable as the observed difference does not exceed our quality specification.
Garcia-Valdecasas et al. (Tue,) studied this question.