Recovery of spiked troponin I in four routine assays

INTRODUCTION This study aimed to examine the recovery of spiked human cardiac troponin I (cTnI) results measured by four routine assays, and investigate possible interference from microclots. MATERIALS AND METHODS 457 consecutive samples with cTnI concentration below limit of quantitation (12 ng/L), declared by the Vitros TnI ES assay (reference assay), were measured on Beckman Coulter Accu TnI+3, Siemens TnI-Ultra and Roche TnI STAT assays. These samples were enriched with native full-length cTnI to a concentration of 100 ng/L and retested. A post-spiking result that exceeded the critical difference at a predefined probability of 0.0005 of the target concentration (the median post-spiking result for each individual assay) was considered as outlier. To determine whether microclots were a significant cause of critically discrepant outlier results, a separate 50 samples were centrifuged twice between two post-spiking measurements using the Vitros TnI ES assay. RESULTS The median recovery of the enriched cTnI was highest with the Roche assay (271 ng/L) and lowest with the Vitros assay (29 ng/L). The Vitros assay had the highest percentage of results that exceeded the critical difference (49%), followed by the Siemens (38%), Roche (18%) and Beckman Coulter (7%) assays. None of the 50 additional samples produced a critically lower cTnI result after re-centrifugation. CONCLUSIONS Our findings underscored the variability of cTnI assays in measuring native cTnI. The lack of cTnI results that became significantly lower after re-centrifugation suggested that microclots are unlikely to be a major cause of the outlier results.