Studio promosso dal Gruppo di Studio Inter-Societario dei Marcatori Cardiaci della Società Italiana di Biochimica Clinica e Biologia Molecolare Clinica (SIBioC) e dalla Sezione Italiana della European Ligand Assay Society (ELAS)

Establishment and evaluation of cardiac troponin I reference values measured in a group of Italian healthy volunteers using high-sensitivity methods: a multi-center study. Introduction: this study compares the cardiac troponin I (cTnI) values measured with three high-sensitivity (hs) different methods in apparently healthy volunteers enrolled in a multicenter study. Methods: heparinized plasma samples were collected from 1511 volunteers in 8 Italian clinical institutions (mean age 51.5 years, SD 14.2, range 18-86, female to male ratio 0.94). All volunteers denied chronic or acute diseases and had normal values of routine laboratory tests and ECG. The reference laboratory of the study (Laboratorio Fondazione CNR Regione Toscana G. Monasterio, Pisa, Italy) assayed all plasma samples with three hs-methods: Architect hs-cTnI, Access hs-cTnI and ADVIA Centaur XPT hs-cTnI. After the exclusion of outlier values, calculation of 99th percentile (Upper Reference Limit, URL) values was performed using both robust nonparametric and bias corrected and accelerated bootstrap methods. Results: large between-method differences were found. ADVIA Centaur measured higher cTnI values (up to 2-fold) than the two other methods. cTnI values were significantly higher in men than in women, and progressively increased with age over 55 years. Moreover, 99th percentile URL values also depended on the statistical approach used for calculation (robust non-parametric versus bootstrap). All 99th percentile URL values calculated with non-parametric robust method were on average slightly lower than those suggested by manufacturers (mean difference 4.2 ng/L, standard error 1.7, p=0.0273). Conclusion: clinicians should be advised that plasma samples from the same patient should be measured for hs-cTnI in the same laboratory. Specific clinical studies are needed to establish the most appropriate statistical approach to calculate 99th percentile URL values for hs-cTnI methods.