No increase in the incidence of cardiac troponin I concentration above the 99th percentile by Siemens Centaur high-sensitivity compared to the contemporary assay.

• Siemens Centaur hs-cTnI assay provided more values above LOQ than the contemporary assay. • Using the hs-cTnI assay did not increase the incidence of elevated troponin above URLs. • There is proportional bias between the hs-cTnI and the contemporary assays. • Analytical comparisons between the two assays were consistent in two hospitals. This study aimed to compare the incidence of cardiac troponin I (cTnI) concentrations above the limit of quantification (LOQ) and the sex-specific 99th percentile upper reference limits (URLs) between the Ultra contemporary cTnI assay and the high-sensitivity (hs-cTnI) assay on Siemens Centaur in patients evaluated in the emergency department (ED) and inpatient at a U.S. urban academic hospital. A retrospective study was performed in an unselected patient cohort who presented to the hospital with symptoms suggestive of myocardial injury. All clinically ordered samples for cTnI assay (n = 1,056, LOQ 0.03 µg/L, URL 0.04 µg/L) were simultaneously tested on the hs-cTnI assay (LOQ 2.5 ng/L; URL 58 ng/L and 39 ng/L for male and female, respectively). The incidence of elevated cTnI above the 99th percentile URL in males measured by the hs-cTnI assay was significantly lower compared to the cTnI assay (31.4% vs. 38.7%, p = 0.016), whereas there was no difference in females (27.4% vs. 30.2%, p = 0.35) in all the patient samples. In ED patient samples (n = 718), the incidence of elevated cTnI above the sex-specific 99th percentile URL was not significantly different between the hs-cTnI and contemporary cTnI assays in either sex (male: hs-cTnI 16.6% vs. cTnI 21.5%, p = 0.13; female: hs-cTnI 19.6% vs. cTnI 21.1%, p = 0.66). The agreement between the two assays was 93.5% (kappa = 0.798). Results were confirmed in an independent patient cohort measured by the same instruments at another hospital. Our study suggests that implementation of the hs-cTnI assay would not lead to an increase in the proportion of elevated cTnI above the 99th percentile in the emergency department and other inpatient units. [ABSTRACT FROM AUTHOR]