Discrepancy between Cardiac Troponin Assays Due to Endogenous Antibodies

Abstract

Abstract Background Despite well-described analytical effects of autoantibodies against cardiac troponin (cTn) I on experimental assays, no study has systematically examined their impact on cTn assays in clinical use. We determined the effects of endogenous antibodies on 5 different cTnI assays and a cTnT assay. Methods cTn was measured by 6 methods: Siemens hs-cTnI Centaur, Siemens hs-cTnI Vista, Abbott hs-cTnI Architect, Beckman hs-cTnI Access, Beckman cTnI Access, and Roche hs-cTnT Elecsys. Measurements were repeated on 5 assays (all except Siemens hs-cTnI Vista) following immunoglobulin depletion by incubation with protein A. Low recovery of cTnI (<40%) following immunoglobulin depletion was considered positive for macro-cTnI. Protein A findings were validated by gel filtration chromatography and polyethylene glycol precipitation. Results In a sample of 223 specimens selected from a community laboratory that uses the Siemens hs-cTnI Centaur assay and from which cTn was requested, 76% of samples demonstrated increased cTnI (median, 88 ng/L; interquartile range, 62–204 ng/L). Macro-cTnI was observed in 123 (55%) of the 223 specimens. Comparisons of cTnI assays markedly improved once patients with macro-cTnI were removed. Passing-Bablok regression analysis between hs-cTnI assays demonstrated different slopes for patients with and without macro-cTnI. In patients with macro-cTnI, 89 (72%) showed no effect on the recovery of cTnT, whereas 34 (28%) had reduced recovery of cTnT. The proportion of results above the manufacturers' 99th percentile varied with the cTn assay and macro-cTnI status. Conclusion We suggest that the observed discrepancy between hs-cTnI assays may be attributed in part to the presence of macro-cTnI.