ROC AMI cutpoint
99% upper reference limit (ESC/ACC) 97.5% upper reference limit
0.01 Assay detection limit
0.00 —I Protein-free sample
Fig. 7. Comparison ofcutoffconcentrations for cardiac troponin according to various criteria. (Adapted from ref. 27.)
the upper reference limit and the ROC-determined AMI cutoff concentration. Because cardiac patients with troponin concentration in this region were also shown to have a high incidence of poor outcomes, ajoint committee ofthe ESC and ACC recommended lowering the cutoffs for cardiac markers to the upper 99% of healthy individuals (Fig. 7). Peak B in Fig. 2 shows that a patient with unstable angina may have a troponin concentration that is below the AMI decision limit but exceeds the upper reference limit of a healthy population. Use of this lower cutoff concentration posed a problem because most assays do not have the sensitivity to consistently detect troponin in the blood ofhealthy individuals and, therefore, the 99% cutoff could not be computed with any acceptable degree of analytic precision. Previously, the National Academy of Clinical Biochemistry had recommended that the assay imprecision be <10% at the cutoff concentration (2). A subcommittee ofthe ESC/ACC suggested that the AMI cutoff should be the marker concentration that first produces a 10% imprecision (coefficient ofvariation [CV]) (9) until new assays for troponin are developed with the necessary sensitivity. Figure 8 illustrates data for several commercial troponin I assays and the value and the concentration at the 10% CV level (28). Figure 9 compares the 99th percentile (obtained from the manufacturer's
Fig. 8. Imprecision vs analyte concentration profiles for cTnI. A: AIA21; B, C: Access and Access 2; D: Stratus CS; E: Centaur; F: Immulite; G: Dimension; H: ACS:180; I: Immuno 1; J: ECi; K: Liaison; L: Opus Plus; M: Vidas; N: AxSym; O: Alpha Dx. (Reproduced from ref. 28, with permission from the American Association for Clinical Chemistry.)
Was this article helpful?