Printer Friendly

Short- and long-term risk stratification using a next-generation, high-sensitivity research cardiac troponin I (hs-cTnI) assay in an emergency department chest pain population.

In recent years, there has been enthusiasm for the development of more sensitive cardiac troponin (cTn) [5] assays (1-3). The reports about these efforts to date have indicated that cTn is measurable in a majority of individuals and that the 99th percentile concentration is lower than previously thought (approximately 10 ng/L), and the increased sensitivity of these new assays may allow changing patterns to be more easily recognized (4-9). The next step in the validation process is determining whether measurement with these hs-cTn assays can identify individuals at risk for an adverse outcome. In the present study, we measured hs-cTnI with a preproduction research assay in patients presenting with symptoms suggestive of acute coronary syndrome (ACS) and assessed if concentrations measured by this high-sensitivity assay were prognostic for short- and/or long-term risk of death and/or myocardial infarction (MI).

Materials and Methods

After ethics approval, in 1996, patients presenting with symptoms suggestive of ACS were enrolled in a cardiac biomarkers study (n = 448). Based on time from onset of pain, blood specimens were collected from the study population in heparin tubes and plasma was separated by centrifugation and frozen until 2003, when cTnI was measured in the specimens using the AccuTnI[TM] assay (Beckman Coulter) (10,11). Specimens with sufficient volume were thawed from storage at--70[degrees]C for a second time in 2007 and measured with a research hs-cTnI assay (Beckman Coulter) (9). For this analysis, we selected as the study cohort the 383 subjects whose earliest presentation specimens had both AccuTnI and hs-cTnI concentrations.

The hs-cTnI assay is reported to have a limit of the blank concentration of 1.03 ng/L, with a limit of detection of 2.06 ng/L, and attains 20% CV and 10% CV at concentrations of 2.95 ng/L and 8.66 ng/L, respectively (9). The hs-cTnI assay uses the same antibodies as the current AccuTnI assay, which has been demonstrated to provide consistent (stable) measurements after multiple freeze/thaws and over 10 years of storage (9, 12, 13). The increased sensitivity with this research hs-cTnI assay was obtained by increases in the incubation time and the sample volume and by changes to the microparticle capture bead (9). The hs-cTnI and AccuTnI assays were correlated (r = 0.87); however, the estimated 99th percentiles for cTn in heparin plasma appeared to be different between the assays (9.20 ng/L for hs-cTnI vs 0.04 [micro]g/L for AccuTnI) (9, 14). The preliminary estimate of the 99th percentile for the hs-cTnI assay was derived from a small group (n = 125) of younger individuals (age [less than or equal to] 55 years). Because our study cohort was older (median age 64 years; Table 1), we opted to subgroup our population into categories that divided the population nearly evenly based on AccuTnI concentrations that have been shown previously to have prognostic value (11, 13, 15). For the AccuTnI assay, we classified subjects into 3 groups: [less than or equal to] 0.01 [micro]g/L (n = 227); 0.02-0.04 [micro]g/L (n = 78); and > 0.04 [micro]g/L (n = 78). Owing to the increased analytical sensitivity of the hs-cTnI assay, we subdivided subjects into 4 groups based on hs-cTnI results: < 5.00 ng/L (n = 92); 5.00-9.99 ng/L (n = 93); 10.00-40.00 ng/L (n = 93); and > 40.00 ng/L (n = 105). To clearly distinguish which cTn concentrations were derived from the hs-cTnI assay, the reported units are in ng/L compared to the AccuTnI assay with reported units in [micro]g/L.

We obtained health outcomes in the study cohort via linkage to the Registered Persons Data Base for mortality outcomes and the Canadian Institute for Health Information Discharge Abstract Database for Ontario hospital discharges associated with MI, which has been shown to be accurate and is consistent in our previous analyses (11, 16). Based on the earliest subsequent readmission for MI and/or date of death, we created indicators to reflect whether an event (readmission or death) occurred within 30 days, 6 months, 1, 2, 5, and 10 years postpresentation. If a patient died without previous MI, follow-up was censored at the date of death. The outcomes were captured as events postpresentation (i.e., either during index hospitalization or afterward). We assessed the time to an adverse event by Kaplan-Meier survival curves with differences between groups determined by the log rank test. We used the Cox proportional hazard model to compare time to an event while adjusting for age and sex. Hazard ratios (HRs) were generated for each cTn group relative to the lowest concentration group ([less than or equal to] 0.01 [micro]g/L for AccuTnI and < 5.00 ng/L for hs-cTnI) and were derived by partial likelihood estimation. Significance of the association was based on the Wald [chi square] statistic. Between-group comparisons of central tendency (means, medians) were based on ANOVA and the Kruskal-Wallis test. We used the Pearson [chi square] test statistic to compare proportions. ROC curve analyses were performed for end point death/MI at 30 days and 1 and 10 years, with logistic regression modeling used to find the optimal concentration cutoffs. Briefly, we derived the optimal cutoff concentrations for cTnI and hs-cTnI by first building logistic models with the cTnI or hs-cTnI variable as the covariate, for each value observed to obtain a c-statistic, with the maximum c-statistic from all models used to identify the optimal cutoff. As the distributions of hs-cTnI and the AccuTnI concentrations were heavily skewed, we used log transformation first, then performed univariate analysis in the logistic regression model for log(hs-cTnI) or log(AccuTnI) using death or MI as the outcome. For the final models, we calculated c-statistics with 95% CIs for each model and compared c-statistics among the different models using a nonparametric approach.

All statistical analyses were performed using SAS and Graphpad prism software. We considered P values < 0.05 statistically significant.

Results

The median time from onset of pain to the first specimen was 3 h (interquartile range 2-6) (Table 1). Kaplan-Meier survival curves for death alone, MI alone, and the combined end point death/MI indicated differences in event-free survival between the 4 hs-cTnI groups up to 10 years after presentation (Fig. 1). Cox proportional hazards analyses indicated that those participants with cTn > 40.00 ng/L or 0.04 [micro]g/L were at higher risk for the combined end point death/MI at 30 days compared with the referent groups (P < 0.01; Table 2). Analysis for either MI or death alone indicated that only those subjects with cTn >40.00 ng/L or 0.04 [micro]g/L were at higher risk for MI within 30 days (HR 13.2, 95% CI 1.73-99.9, P = 0.01 for hs-cTnI; HR 26.1, 95% CI 5.91-115, P < 0.01 for AccuTnI) but not death (P > 0.10). At 6 months and 1 year, in addition to the > 40.00 ng/L groups, only the 10.00-40.00 ng/L group was at higher risk for the combined end point (death/MI) at these time points. Further analysis at 1 and 2 years indicated that participants with hs-cTnI concentrations in the 10.00-40.00 ng/L group were at higher risk for death alone (HR 3.64, 95% CI 1.04-12.7, P = 0.04 at 1 year; HR 4.06, 95% CI 1.54-10.7, P < 0.01 at 2 years); however, participants in the group with AccuTnI concentrations of 0.02-0.04 [micro]g/L were not found to be at increased risk(HR 1.56,95% CI 0.69-3.54, P = 0.29 at 1 year; HR 1.64, 95% CI 0.86-3.11, P = 0.13 at 2 years). Using the combined end point (death/MI) at 2, 5, and 10 years, participants presenting with AccuTnI [greater than or equal to] 0.02 [micro]g/L or hs-cTnI [greater than or equal to] 10.00 ng/L were at higher risk compared with the referent groups (P < 0.05; Table 2). ROC curve analysis for death/MI at 30 days had an area under the curve of 0.74 (95% CI 0.66-0.82) for the hs-cTnI assay compared with 0.81 (95% CI 0.73-0.90) for the AccuTnI assay (P = 0.05). The derived optimal cutoff concentrations of hs-cTnI and AccuTnI at 30 days were 12.68 ng/L (sensitivity 0.83; specificity 0.57) and 0.03 [micro]g/L (sensitivity 0.80; specificity 0.77), respectively (Fig. 2). At 1 and 10 years, using the same approach, the optimal cutoff hs-cTnI concentrations were 12.68 ng/L and 9.30 ng/L, respectively, and for the AccuTnI these concentrations were 0.02 [micro]g/L and 0.01 [micro]g/L.

[FIGURE 1 OMITTED]

Discussion

These data add to our understanding of novel high-sensitivity troponin assays. The increased analytical sensitivity of the hs-cTnI assay allowed us to identify that individuals who presented with hs-cTnI concentrations [greater than or equal to] 10.00 ng/L were at higher risk of death/MI within 1 year after their presentation. For short-term risk (i.e., within 30 days), only those individuals with increased cTn (> 40 ng/L or > 0.04 [micro]g/L) were at higher risk compared to those with low cTnI concentrations. This inconsistency in the relationship of concentrations of cTn to short- and long-term risks could reflect the fact that a relative paucity of short-term events may have led to an underpowered analysis; however, despite this inconsistency, the concentration cutoff at the 99th percentile derived for the AccuTnI assay (i.e., 0.04 [micro]g/L) seems suitable for short-term risk stratification. This finding is supported by logistic regression modeling for AccuTnI, where the optimal cutoff was 0.03 [micro]g/L for death/MI within 30 days; for the hs-cTnI assay, however, this concentration cutoff was 12.68 ng/L. A larger study is necessary to confirm the optimal cutoff for short-term risk stratification using this hs-cTnI assay. Beyond 30 days, the hs-cTnI assay provided additional information regarding risk in the early time frame (i.e., [less than or equal to] 1 year), whereas detectable concentrations at presentation by the AccuTnI assay (e.g., 0.02- 0.04 [micro]g/L) were not able to provide such discriminatory utility in this early time frame. Consistent with other findings (13, 17), these detectable concentrations with the AccuTnI assay were prognostic at 2, 5, and 10 years. As noted by Morrow and Antman (18), the increased sensitivity of these newer assays requires that we refine what we consider "normal." Alternatively, one could start using risk prediction in place of cut values (19).

The data we present are preliminary, and additional studies will be required to confirm these findings and assess the relationship between these lower concentrations and available therapies. Specific limitations include our inability to separate noncardiovascular death from cardiovascular death in our population. Moreover, limited sample volumes in the stored samples precluded our ability to retest those samples with divergent cTn concentrations (i.e., specimens with AccuTnI concentrations [less than or equal to] 0.01 [micro]g/L that had corresponding hs-cTnI concentrations >40.00 ng/L); such divergence in results between the assays might have negatively affected the specificity of the research hs-cTnI assay. Also, the fact that only the admission samples were used in this analysis is another limitation, because serial changes as well as peak concentrations maybe more clinically relevant for an ACS population (20). Additional work using this research hs-cTnI assay in a larger ACS population with multiple specimens to assess both change and peak concentrations over the first 24 h after pain onset would provide a richer dataset to truly assess both the diagnostic and prognostic power of this hs-cTnI assay. Despite these limitations, the findings of our current study extend our understanding and offer promise that novel high-sensitivity cardiac troponin assays will improve our ability to triage and treat patients with possible cardiac injury.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures of Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: A.R. MacRae, Beckman Coulter; A.S. Jaffe, Siemens, Beckman Coulter, Ortho Diagnostics, Singulex, Nanosphere, Inverness, Critical Diagnostics, GlaxoSmithKline, and Novartis.

Stock Ownership: None declared.

Honoraria: P.A. Kavsak, Beckman Coulter.

Research Funding: Canadian Institutes of Health Research. P.A. Kavsak, Beckman Coulter. A.S. Jaffe, Siemens and Beckman Coulter. Reagents were provided as an unrestricted grant by Beckman Coulter.

Expert Testimony: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.

Acknowledgments: Special thanks to the Clinical Research and Clinical Trials Laboratory, Hamilton, for performing the laboratory measurements.

References

(1.) Thygesen K, Alpert JS, White HD. Joint ESC/AACF/ AHA/WHF Task Force for the Redefinition of Myocardial Infarction. Universal definition of myocardial infarction. J Am Coll Cardiol 2007;50:217395.

(2.) Jaffe AS. Chasing troponin: how low can you go if you can see the rise? J Am Coll Cardiol 2006; 48:1763-4.

(3.) Wu AH, Jaffe AS. The clinical need for high sensitivity cardiac troponin assays for acute coronary syndromes and the role for serial testing. Am Heart J 2008;155:208-14.

(4.) Wu AH, Fukushima N, Puskas R, Todd J, Goix P. Development and preliminary clinical validation of a high sensitivity assay for cardiac troponin using a capillary flow (single molecule) fluorescence detector. Clin Chem 2006;52:2157-9.

(5.) Wu AH, Smith A, Schultz K, Lu A, Todd J, Wians F. Short- and long-term biological variation for cardiac troponin I using a high sensitivity assay: implications for clinical practice. Clin Chem 2009; 55:52-8.

(6.) Latini R, Masson S, Anand IS, Missov E, Carlson M, Vago T, et al. Prognostic value of very low plasma concentrations of troponin T in patients with stable chronic heart failure. Circulation 2007;116:1242-9.

(7.) Kurz K, Giannitsis E, Zehelein J, Katus HA. Highly sensitive cardiac troponin T values remain constant after brief exercise- or pharmacologic-induced reversible myocardial ischemia. Clin Chem 2008;54:1234-8.

(8.) Mingels A, Jacobs L, Michielsen E, Swaanenburg J, Wodzig W, Dieijen-Visser M. Reference population and marathon runner sera assessed by highly sensitive cardiac troponin T and commercial cardiac troponin T and I assays. Clin Chem 2009;55:101-8.

(9.) Kavsak PA, MacRae AR, Yerna MJ, Jaffe AS. Analytical and clinical utility of a next generation, highly sensitive cardiac troponin I assay for early detection of myocardial injury. Clin Chem 2009; 55:573-7.

(10.) Kavsak PA, MacRae AR, Lustig V, Bhargava R, Vandersluis R, Palomaki GE, et al. The impact of the ESC/ACC redefinition of myocardial infarction and new sensitive troponin assays on the frequency of acute myocardial infarction. Am Heart J 2006;152:118-25.

(11.) Kavsak PA, Newman AM, Lustig V, MacRae AR, Palomaki GE, Ko DT, et al. Long-term health outcomes associated with detectable troponin I concentrations. Clin Chem 2007;53:220-7.

(12.) Venge P, Lindahl B, Wallentin L. New generation cardiac troponin I assay for the access immunoassay system. Clin Chem 2001;47:959-61.

(13.) Zethelius B, Johnston N, Venge P. Troponin I as a predictor of coronary heart disease and mortality in 70-year-old men: a community-based cohort study. Circulation 2006;113:1071-8.

(14.) Apple FS, Murakami MM. Serum and plasma cardiac troponin I 99th percentile reference values for 3 2nd-generation assays. Clin Chem 2007; 53:1558-60.

(15.) Eggers KM, Laqerqvist B, Venge P, Wallentin L, Lindahl B. Persistent cardiac troponin I elevation in stabilized patients after an episode of acute coronary syndrome predicts long-term mortality. Circulation 2007;116:1907-14.

(16.) Austin PC, Daly PA, Tu JV. A multicentre study of the coding accuracy of hospital discharge administrative data for patients admitted to cardiac care units in Ontario. Am Heart J 2002;144: 290-6.

(17.) Eggers KM, Jaffe AS, Lind L, Venge P, Lindahl B. Value of cardiac troponin I cutoff concentrations below the 99th percentile for clinical decision making. Clin Chem 2009;55:85-92.

(18.) Morrow DA, Antman EM. Evaluation of high-sensitivity assays for cardiac troponin. Clin Chem 2009;55:5-8.

(19.) Vickers AJ, Lilja H. Cutpoints in clinical chemistry: time for fundamental reassessment. Clin Chem 2009;55:15-7.

(20.) Apple FS, Pearce LA, Smith SW, Kaczmarek JM, Murakami MM. Role of monitoring changes in sensitive cardiac troponin I assay results for early diagnosis of myocardial infarction and prediction of risk of adverse events. Clin Chem 2009;55: 930-7.

Peter A. Kavsak, [1] * Xuesong Wang, [2] Dennis T. Ko, [2] Andrew R. MacRae, [3] and Allan S. Jaffe [4]

[1] Department of Pathology and Molecular Medicine, McMaster University, Hamilton, ON, Canada; [2] Institute for Clinical Evaluative Sciences, University of Toronto, ON, Canada; [3] Department of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, MB, Canada; and [4] Cardiovascular Division and Division of Laboratory Medicine, Mayo Clinic, Rochester, MN.

[5] Nonstandard abbreviations: cTn, cardiac troponin; ACS, acute coronary syndrome; MI, myocardial infarction; HR, hazard ratio.

* Address correspondence to this author at: Hammilton Regional Laboratory Medicine Program, Henderson General Hospital (Core Lab Section), 711 Concession St., Hamilton, ON, Canada. Fax 905-575-2581; e-mail kavsakp@mcmaster.ca.

Received March 12, 2009; accepted July 15, 2009.

Previously published online at DOI: 10.1373/clinchem.2009.127241
Table 1. Study cohort characteristics based on hs-cTnl concentration
groupings. (a)
 hs-cTnI, ng/L

Variable <5.00 5.00-9.99

n 92 93
Demographics
Age at presentation
 Mean (SD) 60.3 (14.0) 62.6(14.7)
 Median (IQR)b 60 (49-71) 63 (52-74)
Sex, n (%)
 Female 40 (43.5) 40 (43.0)
 Male 52 (56.5) 53 (57.0)
Previous MI, n (%) 20(21.7) 22 (23.7)
1996 MI diagnosis, n (%) <6 <6
Median cTn concentration
 hs-cTnI, ng/L (IQR) 3.95 (3.11-4.44) 6.82 (5.59-8.02)
 AccuTnI, [micro]g/L
 (IQR) 0.00(0.00-0.01) 0.01 (0.00-0.02)
Median time from onset 4 (2-5) 3 (2-9)
 until measurement, h
 (IQR)
Outcome, n (%)
MI
 Within 30 days <6 <6
 Within 6 months <6 (1.1) <6 (3.2)
 Within 1 year <6 <6
 Within 2 years <6 7 (7.5)
 Within 5 years 10(10.9) 12 (12.9)
 Within 10 years 16(17.4) 18(19.4)
Death
 Within 30 days <6 <6
 Within 6 months <6 7 (7.5)
 Within 1 year <6 8 (8.6)
 Within 2 years <6 11 (11.8)
 Within 5 years 17(18.5) 26 (28.0)
 Within 10 years 26 (28.3) 38 (40.9)
MI/death
 Within 30 days <6 <6
 Within 6 months <6 8 (8.6)
 Within 1 year 6 (6.5) 10(10.8)
 Within 2 years 8 (8.7) 15(16.1)
 Within 5 years 24 (26.1) 31 (33.3)
 Within 10 years 37 (40.2) 44 (47.3)

 hs-cTnI, ng/L

Variable 10.00-40.00 >40.00

n 93 105
Demographics
Age at presentation
 Mean (SD) 65.7 (13.4) 62.6 (13.7)
 Median (IQR) (b) 68 (56-76) 65 (51-74)
Sex, n (%)
 Female 32 (34.4) 41 (39.0)
 Male 61 (65.6) 64 (61.0)
Previous MI, n (%) 24 (25.8) 35 (33.3)
1996 MI diagnosis, n (%) 15(16.1) 38 (36.2)
Median cTn concentration
 hs-cTnI, ng/L (IQR) 17.74 (12.80-25.89) 112.9 (69.77-270.2)
 AccuTnI, [micro]g/L
 (IQR) 0.02 (0.01-0.04) 0.08(0.01-0.32)
Median time from onset 3 (2-5) 4(2-12)
 until measurement, h
 (IQR)
Outcome, n (%)
MI
 Within 30 days <6 15(14.3)
 Within 6 months 7 (7.5) 20 (19.0)
 Within 1 year 10(10.8) 21 (20.0)
 Within 2 years 15(16.1) 23(21.9)
 Within 5 years 18(19.4) 28 (26.7)
 Within 10 years 20(21.5) 34 (32.4)
Death
 Within 30 days <6 <6
 Within 6 months 12 (12.9) 8 (7.6)
 Within 1 year 15(16.1) 12 (11.4)
 Within 2 years 24 (25.8) 17(16.2)
 Within 5 years 36 (38.7) 26 (24.8)
 Within 10 years 52 (55.9) 48 (45.7)
MI/death
 Within 30 days 9 (9.7) 17(16.2)
 Within 6 months 18(19.4) 25 (23.8)
 Within 1 year 24 (25.8) 29 (27.6)
 Within 2 years 35 (37.6) 35 (33.3)
 Within 5 years 46 (49.5) 45 (42.9)
 Within 10 years 60 (64.5) 65 (61.9)

Variable All P value

n 383
Demographics
Age at presentation
 Mean (SD) 62.8 (14.0) 0.070
 Median (IQR) (b) 64 (51-74) 0.064
Sex, n (%)
 Female 153 (39.9) 0.559
 Male 230 (60.1)
Previous MI, n (%) 101 (26.4) 0.261
1996 MI diagnosis, n (%) 57 (14.9) <0.001
Median cTn concentration
 hs-cTnI, ng/L (IQR) 10.67 (5.18-50.73) <0.001
 AccuTnI, [micro]g/L
 (IQR) 0.01 (0.00-0.03) <0.001
Median time from onset 3 (2-6) 0.069
 until measurement, h
 (IQR)
Outcome, n (%)
MI
 Within 30 days 22 (5.7) <0.001
 Within 6 months 31 (8.1) <0.001
 Within 1 year 30 (10.2) <0.001
 Within 2 years 49 (12.8) <0.001
 Within 5 years 68 (17.8) 0.016
 Within 10 years 88 (23.0) 0.053
Death
 Within 30 days 8 +(2.1) 0.367
 Within 6 months 30 (7.8) 0.112
 Within 1 year 38 (9.9) 0.029
 Within 2 years 57 (14.9) 0.001
 Within 5 years 105 (27.4) 0.018
 Within 10 years 164 (42.8) 0.002
MI/death
 Within 30 days 30 (7.8) <0.001
 Within 6 months 55 (14.4) <0.001
 Within 1 year 69 (18.0) <0.001
 Within 2 years 93 (24.3) <0.001
 Within 5 years 146 (38.1) 0.006
 Within 10 years 206 (53.8) 0.001

(a) Privacy constraints prohibit the display of cells from groups of
<6 individual patients.

(b) IQR, interquartile range.

Table 2. Cox proportional hazard model of time to death/MI using
presentation specimen (after adjusting for age and sex).

 AccuTnI HR relative to
Time since concentration, AccuTnI [less than 95% CI
presentation [micro]g/L or equal to] 0.01

 30 days >0.04 14.68 4.96-43.48
 0.02-0.04 3.37 0.89-12.75

 6 months >0.04 7.78 4.03-15.04
 0.02-0.04 2.03 0.88-4.68

 1 year >0.04 5.42 3.09-9.53
 0.02-0.04 1.91 0.97-3.77

 2 years >0.04 4.37 2.69-7.10
 0.02-0.04 2.09 1.22-3.60

 5 years >0.04 2.73 1.85-4.03
 0.02-0.04 1.78 1.18-2.67

 10 years >0.04 2.34 1.68-3.26
 0.02-0.04 1.70 1.21-2.39

 hs-cTnI HR relative
Time since P value concentration, to hs-cTnI
presentation ng/L < 5.00

 30 days <0.01 >40.00 7.20
 0.07 10.00-40.00 3.60
 5.00-9.99 0.91
 6 months <0.01 >40.00 5.82
 0.10 10.00-40.00 3.77
 5.00-9.99 1.83
 1 year <0.01 >40.00 4.58
 0.06 10.00-40.00 3.38
 5.00-9.99 1.50
 2 years <0.01 >40.00 4.32
 0.01 10.00-40.00 4.01
 5.00-9.99 1.70
 5 years <0.01 >40.00 1.94
 0.01 10.00-40.00 1.89
 5.00-9.99 1.19
 10 years <0.01 >40.00 1.85
 <0.01 10.00-40.00 1.66
 5.00-9.99 1.15

Time since 95% CI P value
presentation

 30 days 1.66-31.21 0.01
 0.76-16.94 0.11
 0.13-6.50 0.93
 6 months 2.02-16.75 <0.01
 1.26-11.27 0.02
 0.55-6.08 0.33
 1 year 1.90-11.04 <0.01
 1.37-8.33 0.01
 0.55-4.14 0.43
 2 years 2.00-9.32 <0.01
 1.85-8.70 <0.01
 0.72-4.02 0.23
 5 years 1.18-3.18 0.01
 1.15-3.11 0.01
 0.70-2.03 0.53
 10 years 1.23-2.77 <0.01
 1.10-2.52 0.02
 0.74-1.79 0.53

Fig. 2. ROC curve analysis for death/MI at 30 days,
1 year, and 10 years for hs-cTnl (A) and AccuTnl (B).
AUC, area under the curve.

A
 Outcome Lower 95% Upper 95%
 (death/MI) AUC CI of AUC CI of AUC

death/MI within 30 days within 30 days 0.74 0.66 0.82
death/MI within 1 days within 1 days 0.68 0.61 0.74
death/MI within 10 days within 10 days 0.61 0.55 0.67

 Optimal
 hs-cTnl (ng/L) Sensitivity Specificity

death/MI within 30 days 12.68 0.83 0.57
death/MI within 1 days 12.68 0.72 0.60
death/MI within 10 days 9.30 0.65 0.57

B
 Outcome Lower 95%
 (death/MI) AUC CI of AUC

death/MI within 30 days within 30 days 0.81 0.73
death/MI within 1 days within 1 days 0.74 0.68
death/MI within 10 days within 10 days 0.68 0.62

 Upper 95% Optimal AccuTnl
 CI of AUC ([micro]g/L)

death/MI within 30 days 0.90 0.03
death/MI within 1 days 0.81 0.02
death/MI within 10 days 0.73 0.01

 Sensitivity Specificity

death/MI within 30 days 0.83 0.57
death/MI within 1 days 0.72 0.66
death/MI within 10 days 0.65 0.57
COPYRIGHT 2009 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Proteomics and Protein Markers
Author:Kavsak, Peter A.; Wang, Xuesong; Ko, Dennis T.; MacRae, Andrew R.; Jaffe, Allan S.
Publication:Clinical Chemistry
Date:Oct 1, 2009
Words:4386
Previous Article:Development of an immunoassay for the quantification of soluble LR11, a circulating marker of atherosclerosis.
Next Article:Design and optimization of reverse-transcription quantitative PCR experiments.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters