False Increases of Troponin I Attributable to Incomplete Separation of Serum

In the process of establishing a troponin I assay in our laboratory, we found limited data on reference intervals for plasma. Because our laboratory relies heavily on plasma rather than serum to expedite analysis (1), we undertook a parallel study of serum and plasma troponin I concentrations with the Abbott AxSYMTM assay. To our surprise, we found several discordant results (Table 1). Reanalysis of the samples reproduced the discrepant results. However, when one of the serum samples was recentrifuged, reanalysis produced results essentially the same as those obtained from the matching plasma sample. After the other discordant serum samples were recentrifuged and reanalyzed, all serum and plasma results were similar. Additional samples, purposely allowed to contain tiny fibrinstrands, also produced increased troponin I results. Troponin I results on serum samples following initial centrifugation and after recentrifugation compared with matched plasma samples. NA, plasma sample not available. Troponin I results on serum samples following initial centrifugation and after recentrifugation compared with matched plasma samples. NA, plasma sample not available. We believe that the antibody either binds nonspecifically to fibrin or that the indicator enzyme is physically trapped by fibrin in the separation matrix. Because incomplete separation of serum can leave fibrin in the sample, falsely increased troponin I concentrations can result. We concluded that the use of plasma rather than serum essentially eliminates this problem and also allows the samples to be placed on the AxSYM faster, expediting turnaround time by at least 10–15 min. However, if serum is used, the samples must be completely clotted and centrifuged sufficiently to ensure complete separation of the serum from the clot. In our laboratory, we have found that centrifugation at 2000g for 10 min is required for the preparation of fibrin-free serum. Brad Combs Glenda Abbott Abbott Laboratories, 200 Abbott Park Rd., Abbott Park, IL 60064 Address correspondence to this author at: Abbott Laboratories, 200 Abbott Park Rd., Building AP31, Department 9JW, Abbott Park, IL 60064. This letter highlights the important issue of sample preparation and that poor sample integrity can adversely affect assay methodologies, regardless of the highly specific nature of the reagents. For assays using serum, it is important to ensure that complete clot formation has taken place before centrifugation. It should also be noted that some specimens, especially those from patients receiving anticoagulant or thrombolytic therapy, may exhibit increased clotting time. If specimens are centrifuged before complete clot formation, the presence of fibrin may alter results.