| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Hennepin County Medical Center, Minneapolis, MN
aAddress correspondence to this author at: Hennepin County Medical Center, Clinical Laboratories P4, 701 Park Ave., Minneapolis, MN 55415. Fax 612-904-4229; e-mail apple004{at}umn.edu.
To the Editor:
Clinical and laboratory medicine associations have endorsed the implementation of the 99th percentile reference value for cardiac troponin I (cTnI) and T (cTnT) for the detection of myocardial injury (1)(2). In the presence of ischemia, cTn above the 99th percentile has become the cornerstone for the definition of acute myocardial infarction (MI) in non–ST-elevation MI (1)(2), leading manufacturers of in vitro diagnostic tests to improve low-end analytic sensitivity and precision. We determined the 99th percentile values for 3 2nd-generation cTnI assays in serum and/or plasma.
After receiving appropriate institutional review board approval, we obtained frozen (–20 °C up to 30 days) serum samples from 2992 apparently healthy volunteers. Because this group of samples was collected for another study, data on donor age and sex were unavailable, a study limitation. We also obtained 2000 plasma (heparin) samples from a separate group of healthy individuals, 75% male, age 18–66 years. These fresh specimens were refrigerated for 24–48 h before analysis. cTnI concentrations were measured by use of the following US Food and Drug Administration–cleared assays: the Ortho-Clinical Diagnostics (OCD) Vitros Troponin I-ES assay on the Vitros ECi /ECiQ System, the Abbott Diagnostics Architect STAT Troponin-I on the i2000SR System, and the Beckman Coulter Access Accu TnI on the Access system. Results were reported to 3 decimals for the Abbott and OCD assays and to 2 decimals for the Beckman assay. All specimens were analyzed on the 3 different assays on the same day for serum or plasma. According to package inserts, limits of detection were 
0.010 µg/L for the Abbott, 0.01 µg/L for the Beckman, and 0.012 µg/L for the OCD assay. As determined in the current study, 99th percentiles for plasma (heparin) were 0.012 µg/L (n = 224) for Abbott, 0.04 µg/L (n = 254) for Beckman, and 0.031 µg/L (n = 2000) for OCD. We determined 99th percentile reference cutoffs by nonparametric statistics following Clinical and Laboratory Standards Institute guidelines C28-A2. Our laboratory has documented <10% cTnI variability within specimens for cTnI in serum and plasma frozen at –20 °C for 30 days.
Fig. 1
shows the histogram distributions for serum cTnI by OCD (top), Beckman (middle), and Abbott (bottom) assays. The 99th percentiles were 0.034 µg/L, 0.04 µg/L, and 0.025 µg/L, respectively. Maximum concentrations were OCD 0.108 µg/L, Beckman 0.31 µg/L, and Abbott 1.124 µg/L (confirmed by repeat analysis; assay results for the same specimen were 0.01 µg/L for Beckman and 0.003 µg/L for OCD). The skewed histograms demonstrated that 88% (n = 2637) of the OCD, 98% (n = 2930) of the Abbott, and 33% (n = 979) of the Beckman assay results were below the limit of detection. Each assay also showed different numbers of samples (a) between the limit of detection and 99th percentile limits [Abbott n = 33 (1%), OCD n = 319 (11%), Beckman n = 1994 (67%)] and (b) greater than the 99th percentile [Abbott n = 29 (1%), Beckman n = 19 (1%), OCD n = 36 (1%)]. Only 3 samples had cTnI concentrations greater than the 99th percentile according to all 3 assays. Similar observations were found for plasma, with a 99th percentile of 0.031 µg/L for OCD and 0.04 µg/L for Beckman (histograms not shown because of nonsignificant differences with serum). The Abbott assay was not analyzed for plasma, because of lack of funding and inadequate volumes. There were no significant cTnI differences between samples from donors of different sexes or across donor age by decade, with age limited to 66 years.
|
Our findings represent the largest database for 99th percentile values determined on 2nd generation cTnI assays for serum and plasma. The nonsignificant differences between cutoffs for serum and plasma and between assays are encouraging regarding the ease of interpretation by clinicians and laboratories. Also encouraging were improvements in the analytical sensitivity of these newer-generation assays. Interestingly, only 3 samples were identified as above the 99th percentile by all 3 assays. The mechanisms responsible for this lack of agreement are not known, but antibody differences and the cTnI epitopes recognized on the different circulating cTnI forms are the most likely causes (4). The discrepancies between assays in regard to individual samples identified as above the 99th percentile raise questions as to whether healthy individuals can be better characterized for use in defining reference limits. A prototype cTnI assay has shown sensitivity to 0.001 µg/L with gaussian-distributed results (3). As assays become more sensitive, increasing numbers of MIs will be detected, but an increased prevalence of myocardial injury not related to ischemic pathologies will also be seen (5). The latter may complicate schemes for selecting healthy individuals when defining reference limits. The 99th percentile reference cutoffs derived here in 2 large apparently healthy populations for 3 2nd-generation cTnI assays should prove useful in clinical practice.
Acknowledgments
Grant/funding support: We thank Sheryl Sullivan, Jody Parsells, Michele Steinmann, and Theresa Tubbs (all from Ortho-Clinical Diagnostics) for their assistance and for the partial financial support provided by OCD for this study.
Financial disclosures: F.S.A. has both consulted for and received research grant support from Abbott, Beckman, and Ortho-Clinical Diagnostics.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  H. D White Evolution of the definition of myocardial infarction: what are the implications of a new universal definition? Heart, June 1, 2008; 94(6): 679 - 684. [Full Text] [PDF]
|
|
|
|
 |
|
 J. R Tate, W. Ferguson, R. Bais, K. Kostner, T. Marwick, and A. Carter The determination of the 99th centile level for troponin assays in an Australian reference population Ann Clin Biochem, May 1, 2008; 45(3): 275 - 288. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
