Clin Res Cardiol (2014) 103:211–222 DOI 10.1007/s00392-013-0640-8
ORIGINAL PAPER
Association of high-sensitivity assayed troponin I with cardiovascular phenotypes in the general population: the population-based Gutenberg health study Christoph Sinning • Till Keller • Tanja Zeller • Francisco Ojeda • Michael Schlu¨ter • Renate Schnabel • Edith Lubos • Christoph Bickel • Karl J. Lackner • Patrick Diemert • Thomas Munzel • Stefan Blankenberg • Philipp S. Wild • for the Gutenberg Health Study
Received: 14 September 2013 / Accepted: 8 November 2013 / Published online: 23 November 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Background Aim of the study was to analyze the correlation of high-sensitivity assayed troponin I with cardiac and vascular structure and function in a large populationbased cohort. Methods In a sample of 4,139 subjects (2,099 men, 2,040 women, age 35–74 years) from the population-based Gutenberg Health Study, troponin I was measured with a high-sensitivity assay that had a limit of detection of 1.9 pg/mL. Results In the study cohort, 3,405 subjects had detectable troponin I concentrations [82.3 % overall, 89.9 % men (N = 1,888), 74.4 % women (N = 1,517)]. All analyses
C. Sinning and T. Keller contributed equally. S. Blankenberg and P. S. Wild were both senior authors and contributed equally.
Electronic supplementary material The online version of this article (doi:10.1007/s00392-013-0640-8) contains supplementary material, which is available to authorized users. C. Sinning T. Keller T. Zeller F. Ojeda M. Schlu¨ter R. Schnabel E. Lubos P. Diemert S. Blankenberg (&) Department of General and Interventional Cardiology, University Heart Center Hamburg, Martinistr. 52, 20246 Hamburg, Germany e-mail: [email protected] C. Bickel Department of Internal Medicine, Federal Armed Forces Central Hospital, Koblenz, Germany
were adjusted for age. The strongest correlate between detectable troponin I and measures of cardiac phenotypes was observed for left ventricular mass (p \ 0.001) and left ventricular end-diastolic diameter (p \ 0.001) for both, women and men. Left ventricular ejection fraction was inversely correlated with troponin I (p value\0.001 in men and 0.0013 in women), also measures of diastolic dysfunction as represented by Tei index and E/E0 correlated with detectable troponin I concentrations (p \ 0.001 for both gender). With respect to vascular structure and function, troponin I correlated with mean intima-media thickness of the carotid artery (p \ 0.001 in men and p = 0.013 in women) but showed only borderline correlation with measures of vascular function represented by flow-mediated dilation (p = 0.05 in women and p = 0.018 in men) and arterial stiffness. Conclusions Troponin I assessed by a high-sensitivity assay correlated with measures of left ventricular hypertrophy and systolic and diastolic function, whereas its correlation with vascular phenotypes was only of weak magnitude.
T. Munzel P. S. Wild (&) Department of Medicine II, University Medical Center, Johannes Gutenberg University Mainz, Langenbeckstr. 1, 55131 Mainz, Germany e-mail: [email protected] P. S. Wild Center for Thrombosis and Hemostasis, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany
K. J. Lackner Institute of Clinical Chemistry and Laboratory Medicine, University Medical Center Johannes Gutenberg University, Mainz, Germany
123
212
Keywords High-sensitivity assayed troponin I Cardiac structure Cardiac function Endothelial dysfunction Population study Abbreviations BMI Body mass index CAD Coronary artery disease CCA Common carotid artery CRP C-reactive protein CV Coefficient of variation eGFR Estimated glomerular filtration rate FMD Flow-mediated dilation GHS Gutenberg Health Study hs-TnI Troponin I as measured by a high-sensitivity assay hs-TnT Troponin T as measured by a high-sensitivity assay LoD Limit of detection LV Left ventricular LVEDD Left ventricular end-diastolic diameter LVEF Left ventricular ejection fraction LVM Left ventricular mass MI Myocardial infarction NT-proBNP N-terminal pro-B-type natriuretic peptide SD Standard deviation
Introduction Evolution of troponin assays has recently led to the determination of troponin via high-sensitivity assays [1, 2]. Several studies have shown in real-world settings that the application of both contemporary-sensitivity and high-sensitivity assays for troponin I and T improves the differential diagnosis of patients with suspected acute coronary syndrome [3–6]. By the use of modern, highsensitivity assays, cardiac troponin concentrations can now be reliably detected in the majority of the presumably healthy population with a substantial magnitude of variety among the various assays [1, 7, 8] [9–11]. However, the pathophysiology of those detectable very low concentrations and its clinical utility is poorly understood [3, 4, 10, 12]. In these cases, acute ischemia cannot be held responsible for troponin release in population-based cohorts and other pathophysiologic conditions must account for it [13–17]. Previous studies have already shown the distribution of hs-Troponin T in the population and association with cardiac phenotypes [13, 14]. Results from the Dallas Heart Study could show that troponin T was detectable in 25 % of the general population and
123
Clin Res Cardiol (2014) 103:211–222
additionally identified individuals with structural heart disease and increased risk of all-cause mortality [13]. Further, as older adults are at an increased risk of newonset of heart failure, hs-TnT proved to be reliable in detecting those individuals with incident heart failure and increased risk of cardiovascular death by means of baseline and serial troponin T measurements [14]. However, whereas hs-Troponin T detects only a limited percentage in the younger and healthy population, there is scarce information so far available linking the entire troponin I concentration range with intermediate phenotypes of cardiac and vascular structure and function in the general population. Therefore, a troponin I assay was applied in this study, reported to detect up to 96 % of the apparently healthy population aged 18–64 years [7]. Actual studies using hs-TnI assays in the general population of Scotland and older individuals in Sweden report that an increased troponin I concentration is related to future cardiovascular events and mortality [18, 19]. Previous results of the Framingham Heart Study could show that biomarkers reflecting cardiovascular stress are detectable in the general population and provide additional prognostic information to traditional cardiovascular risk factors [20]. For this study, an ultrasensitive assay assessing troponin I through a single-molecule counting technique was used, enabling measurement in over 80 % of the subjects [20]. In this context, a confirmatory study describing the association of troponin I measured with a high-sensitive assay and parameters of cardiac and vascular structure was not present so far. The intention of this study is to investigate these cardiovascular phenotypes in relation to hs-TnI concentration measured in 4,139 subjects of the population-based Gutenberg Health Study (GHS) [13, 14, 21].
Methods Study population Recruitment into the GHS was conducted between April 2007 and April 2012, finally including 15,010 individuals. A detailed study description has been published earlier [22, 23]. In brief, study individuals aged 35–74 years and stratified according to gender and age were selected randomly by the registration office from the city of Mainz. High-sensitivity assayed troponin I (hs-TnI) measurements were performed in 4,139 of the first 5,000 study participants. The GHS was approved by the Ethics Committee of Rhineland-Palatinate and the medical faculty of the Johannes Gutenberg-University, Mainz. Each study individual provided written informed consent before
Clin Res Cardiol (2014) 103:211–222
participating. The ethical application complied with the Declaration of Helsinki. All authors have read and approved the manuscript in its current form. Assessment of risk factors Cardiovascular risk factors and diseases Risk factors were assessed as outlined in previous publications [24]. Former history of stroke, coronary artery disease (CAD), myocardial infarction (MI), heart failure and peripheral artery disease were assessed in a standardized interview. Arterial hypertension was defined as a systolic blood pressure C140 mmHg and diastolic blood pressure C90 mmHg at rest obtained as the mean of the second and third measurement, or by taking any antihypertensive drugs within the last 2 weeks. Diabetes mellitus was defined as a fasting glucose C126 mg/dL, a spontaneous glucose concentration of C200 mg/dL, or as diagnosed by a physician. We defined dyslipidemia as a LDL/ HDL ratio of [3.5 or as diagnosed by a physician. Waist circumference C102 cm for men or C88 cm for women identified subjects with body mass index C30 and was used, therefore, for subjects classified as obese [25]. Smokers were classified into daily smokers (C1 cigarette/ day), occasional smokers (\1 cigarette/day), former smokers and non-smokers (never smoked). Any family history of MI in first-degree relatives before the age of 65 years for females and before the age of 60 years for males was defined as positive family history. For glomerular filtration rate (GFR), as the best marker for renal function in health and disease, we used the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation [26]. Since creatinine measurements according to the method described by Jaffe are not IDMS-traceable (being the ‘‘standard creatinine’’) we multiply the creatinine value by 0.95 before using the CKD-EPI equation as described by Matsushita et al. [27].
213
Myocardial function Myocardial function was assessed by LV ejection fraction [(LVEF) in %] as (end-diastolic volume-end-systolic volume)/end-diastolic volume 9 100, with volumes determined according to the Simpson method in 4- and 2-chamber views and Tei index. The Doppler-derived Tei index, also called myocardial performance index, was calculated as the sum of isovolumic contraction time and isovolumic relaxation time divided by LV ejection time. The E/E0 ratio of diastolic function was calculated by dividing the early filling velocity of transmitral Doppler (E) by the early relaxation velocity on tissue Doppler (E0 ). Vascular measures representing structure and function Intima-media thickness Intima-media thickness (IMT) of the carotid artery was assessed using an iE33 ultrasound system (Royal Philips Electronics, Amsterdam, The Netherlands) using an 11- to 3-MHz linear-array transducer. Evaluation was performed using an automatic computerized system (QLAB; Philips Medical Systems), and triggering was performed according to the Q wave on electrocardiography to enable measurement in complete relaxation of the ventricle. Mean IMT was recorded 1 cm proximal to the carotid bulb over a length of 1 cm at the far wall of both common carotid arteries (CCA). Flow-mediated dilation Flow-mediated dilation (FMD in %), defined as (mean brachial artery diameter [hyperemia]-mean brachial artery diameter [baseline])/mean brachial artery diameter [baseline] 9 100, was measured using a 5-min upper-arm occlusion, with the brachial artery diameter measured in resting condition (baseline) and after induction of local reactive hyperemia [29]. In the study cohort, 117 FMD values were missing because of measurement errors.
Cardiac phenotypes representing structure and function Infrared plethysmography Myocardial structure All subjects underwent multimodal echocardiography with an iE33 echocardiography system with an S5–1 sector array transducer (Royal Philips Electronics, Amsterdam, The Netherlands), a phased array with 80 elements and a 5- to 1-MHz operating frequency range [22]. The examinations were performed according to standard operating procedures by trained and certified medical technical assistants at a single center. Measurements were according to recommendations by the American Society of Echocardiography [28].
The digital volume pulse was continuously obtained by measuring the transmission of infrared light through the pulp of the right ring finger with a PulseTrace 2000 device (Cardinal Health/Micro Medical Limited; Rochester, UK) [29]. Laboratory analyses Routine laboratory parameters were measured by standardized methods. Troponin I was assessed using a commercially available high-sensitivity cardiac troponin assay (ARCHITECT
123
214
STAT highly sensitive Troponin I immunoassay, Abbott Diagnostics, USA, ARCHITECT i2000SR). The limit of detection (LoD) for the assay was 1.9 pg/mL (assay range 0–50,000 pg/mL). The assay has a 10 % coefficient of variation at a concentration of 5.2 pg/mL. Intra-assay and inter-assay coefficients of variation were 4.26 and 6.29 %, respectively [19]. In this context, measurable concentrations relate to concentrations above 1.9 pg/mL representing the LoD as reported by the manufacturer [19]. Further it has to be mentioned that different studies using the same assay had a different LoD ranging from 1.2 to 1.5 pg/mL resulting in a larger part of the subjects having measurable troponin I concentration [8, 18]. N-terminal pro-B-type natriuretic peptide (NT-proBNP) was measured with a commercially available assay on the ELECSYS 2010 using the Elecysys proBNP II assay (ECLIA, Roche Diagnostics, Mannheim, Germany) with an assay range of 5–35,000 pg/mL and a total imprecision of 1.5–2.5 %. Lipid values have been obtained by routine methods. All laboratory measurements were carried out blinded to the clinical course of the study subjects. Statistical analyses Analyses were stratified by gender. To take into account the stratified nature of the sample, baseline characteristics (Tables 1, 2) and analyses involving the distribution of high-sensitivity determined troponin I (Fig. 1) were weighted according to the age distribution in the study population (N = 210,867, data of the German Federal Statistical Office, Wiesbaden, 2007). In those analyses without gender stratification (categories of troponin I and values in Fig. 1) the statistics were weighted for the age and sex distribution in the study population. High-sensitivity determined troponin I was analyzed as continuous and categorical variable. To define the latter, the subjects were divided into four categories with the lowest category encompassing the individuals with troponin I concentrations below the LoD of the assay (\1.9 pg/mL 10.1 % in men and 25.6 % in women) (Fig. 1). The further concentration range was divided into thirds using tertiles (3.5, 5.2 pg/mL), the first third (and second category) was C1.9 and \3.5 pg/mL, the second third C3.5 and \5.2 pg/mL and the last category C5.2 pg/mL. A male subject with an hs-TnI value of 1,140.9 pg/mL was classified as an outlier and excluded from all analyses [30]. For the descriptive analyses of baseline cardiovascular risk factors and clinical variables, continuous variables are described by median, 25th and 75th percentile or by mean and standard deviation (SD), binary variables by percentages. Trend tests were applied via age-adjusted logistic or quantile regression, to examine the association of different
123
Clin Res Cardiol (2014) 103:211–222
phenotypes and a categorized troponin I (assigning values 1,2,3,4 to the troponin I categories). Tobit regression [31, 32], was used to take into account the fact that troponin I values below the LoD are censored (Figs. 2, 3). Regression coefficients (betas) per increase in 1 standard deviation are presented with the corresponding p values. The supplement for the information presented in Figs. 2 and 3 is expanded (Supplemental Table 1 and 2). All analyses were performed using R 2.5.2[33]. (R Development Core Team (2009). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org).
Results Distribution of high-sensitivity assayed troponin I The distribution of troponin I concentration assessed by the high-sensitivity assay in the 4,139 study individuals is displayed in Fig. 1. Detectable troponin I concentrations were found in 3,405 subjects of the study cohort (82.3 % overall, 89.9 % men (N = 1,888), 74.4 % women (N = 1,517), if weighted by age distribution according to the data of the German Federal Statistical Office, these number are 81.6 % for the overall population, 89.9 % in men and 73.1 % in women, respectively). In this study, the troponin concentrations were stratified into categories with the lowest category with troponin I concentrations below the LoD of the assay (\1.9 pg/mL, N = 523 in women and N = 211 in men). The three further categories encompassed the thirds with the first third ranging from C1.9 to \3.5 pg/mL (N = 665 in women and N = 514 in men), the second third C3.5 and\5.2 pg/mL (N = 509 in women and N = 606 in men) and the last third C5.2 pg/mL (N = 343 in women and N = 767 in men). In this cohort, troponin I concentrations increase with age to about 49 years for both genders in category 1, and 60 and 55 years for women and men, respectively, in category 4 (Table 1). Therefore, all further statistical analyses correlating troponin I with cardiac structure and function were adjusted for age. The numbers of men were higher in troponin categories 3 and 4 and women were more often present in the category below the LoD and the second category. Troponin I and classical risk factors/cardiovascular diseases In the baseline table showing the results for the ageadjusted analysis, the classical cardiovascular risk factors
Clin Res Cardiol (2014) 103:211–222
215
Table 1 Baseline characteristics according to the 4 hs-TnI categories hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
48.3 ± 10.1
50.6 ± 10.7
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
p value adjusted for age p value
Age (years) Women
Men 49.6 ± 9.2 48.2 ± 10.1 Cardiovascular risk factors and diseases
54.9 ± 10.4
59.7 ± 10.7
51 ± 10.9
55.2 ± 11.7
Hypertension (%) Women
31.4 %
38.2 %
48.8 %
58.3 %
0.020
Men
31.3 %
36.6 %
52 %
60.3 %
\0.001
Diabetes (%) Women
2.2 %
3.8 %
4.7 %
9.2 %
0.019
Men
5.8 %
5.8 %
4.3 %
12.9 %
0.056
Obesity (BMI [ 30 kg/m2) (%) Women
14.3 %
19.5 %
27 %
31.5 %
\0.001
Men
15.1 %
18.3 %
26.2 %
31 %
\0.001
Dyslipidemia (%) Women
10 %
15.6 %
26 %
40.2 %
\0.001
Men
22 %
27.6 %
40.6 %
45.4 %
\0.001
18.6 % 26.8 %
15.9 % 22.3 %
18.6 % 21.1 %
0.61 0.70
Active smoking (%) Women Men
22 % 24.3 %
Prevalent CAD (%) Women
0.9 %
1.6 %
2.1 %
3%
0.96
Men
0.9 %
3%
5.2 %
9%
\0.001
History of MI (%) Women
0.4 %
0.6 %
1.9 %
2.9 %
0.049
Men
0.8 %
1.4 %
2.4 %
7.1 %
\0.001
Family history of MI (%) Women
16.1 %
17 %
18.2 %
21.9 %
0.12
Men
16.9 %
15.9 %
17.8 %
16.5 %
0.97
Medicated heart failure (%) Women
1.1 %
0.7 %
1.3 %
2.9 %
0.91
Men
0%
0.6 %
0.6 %
1.8 %
0.024
arterial hypertension, obesity and dyslipidemia were associated with hs-TnI concentration (Table 1). For both genders, an increase in the prevalence of hypertension from troponin category 1 to category 4 was observed (men p \ 0.001 and women p = 0.02). The prevalence of diabetes was higher in subjects within the elevated troponin I categories, irrespective of gender, however, significance was only reached for women (p = 0.019) and was borderline in men (p = 0.056). The prevalence of CAD and previous MI was higher in men with higher troponin I concentrations (both p values \0.001). The results for the CV risk factors show the influence of age, BMI (both p value \0.001) and history of MI (p value in men \0.001 and 0.041 in women) (Supplemental Table 1 and 2).
Troponin I and natriuretic peptides/renal function The categorized troponin I concentration was associated with Nt-proBNP and eGFR in men (p \ 0.001 and p = 0.0081, respectively) but not in women (p = 0.12 for Nt-proBNP and p = 0.15 for eGFR) (Table 2). Also the distribution of troponin I was not associated with Nt-proBNP and eGFR in women according to the different categories, and the association between cardiovascular phenotypes and hs-TnI was more accurately assessed by censored regression analysis (Supplemental Table 1 and 2). NT-proBNP values (both genders p value of \0.001, b = 0.164 in men and b = 0.063 in women) were correlated positive with hs-TnI and eGFR calculated with the CKD-EPI equation (men p value of\0.001, b = -0.107 and women p = 0.0012, b = -0.059) was
123
216
Clin Res Cardiol (2014) 103:211–222
Table 2 Characteristics of interest according to the 4 hs-TnI categories defined hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
p value adjusted for age p value
69 (38, 118) 26 (8, 56)
80 (45, 142) 29 (10, 63)
109 (58, 235) 49 (19, 130)
0.12 \0.001
Biomarker NT-proBNP (pg/mL) Women Men
69 (39, 114) 29 (9, 52)
eGFR (mL/min for 1.73 m2) Women
93.9 ± 13.1
91.5 ± 12.5
87.8 ± 13.8
83.7 ± 14.2
0.15
Men
97 ± 10.7
97.8 ± 11.8
95.1 ± 12
90.6 ± 15.1
0.0081
Cardiac structure Left ventr. end-diastolic dia. (mm) Women
42.8 ± 3.6
43.4 ± 3.9
43.4 ± 4.1
43.4 ± 4.9
\0.001
Men
46.8 ± 4.2
46.7 ± 3.9
47.3 ± 4.1
48.2 ± 4.9
\0.001
Left ventricular mass (g) Women
120 (101, 142)
124 (106, 145)
130 (111, 154)
135 (116, 164)
Men
168 (143, 194)
167 (143, 191)
175 (153, 201)
185 (158, 221)
0.0022 \ 0.001
Cardiac function LVEF (%) 64.4 ± 5.9
64.6 ± 5.3
64.3 ± 5.4
63.8 ± 6.3
\0.001
Men 62.9 ± 5.6 Tei index
63.8 ± 5.9
63.4 ± 6.1
62.1 ± 7.4
0.13
Women
Women
0.5 (0.42, 0.58)
0.53 (0.45, 0.61)
0.54 (0.45, 0.63)
0.56 (0.47, 0.64)
0.0017
Men
0.51 (0.44, 0.59)
0.54 (0.45, 0.63)
0.54 (0.43, 0.64)
0.56 (0.47, 0.65)
0.01
E/E0 Women
6.49 (5.53, 7.72)
6.92 (5.82, 8.23)
7.25 (5.84, 9.33)
7.78 (6.41, 9.22)
0.44
Men
5.92 (5, 7.17)
6.18 (5.22, 7.53)
6.28 (5.26, 7.59)
6.9 (5.63, 8.8)
0.0082
Vascular function FMD (%) Women
11 ± 6.2
10.4 ± 5.8
9.6 ± 5.4
8.4 ± 5.7
0.16
Men
6.6 ± 4.3
6.6 ± 4.1
6.3 ± 3.8
6.2 ± 3.7
0.040
Reflection index (Post–Pre) Women
-4.5 ± 12.5
-4 ± 14.4
-4.4 ± 14.1
-3 ± 15.8
1.0
Men
-0.5 ± 9.9
-1.2 ± 12.9
-0.6 ± 12.1
-1.1 ± 10.9
0.21
0.603 ± 0.111 0.604 ± 0.116
0.633 ± 0.109 0.64 ± 0.128
0.671 ± 0.125 0.673 ± 0.136
0.0056 0.019
Vascular structure IMT (mean) (mm) Women Men
0.58 ± 0.1 0.618 ± 0.126
Lipids HDL (mg/dL) Women
65.5 ± 15.2
64.1 ± 14.9
61.3 ± 15
61.2 ± 15.5
\0.001
Men
50.7 ± 13.4
50.7 ± 12.6
48.7 ± 12.6
48.3 ± 12.2
\0.001
LDL (mg/dL) Women
122.6 ± 30.8
138.7 ± 31.5
150 ± 33.3
161.3 ± 38.9
\0.001
Men
128 ± 35.4
136.2 ± 29.9
145.7 ± 32.9
148.3 ± 38.9
\0.001
Triglycerides (mg/dL) Women
86.44 (64, 118)
93 (71, 124.26)
99.44 (77.6, 133.14)
110 (87, 143)
\0.001
Men
104.44 (69.06, 157.79)
108 (82, 149.82)
121 (87, 163.28)
125 (90, 175)
\0.001
3.2 %
2.8 %
5.2 %
Systolic dysfunction Women
123
6.1 %
0.4
Clin Res Cardiol (2014) 103:211–222
217
Table 2 continued
Men
hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
8.2 %
4.6 %
7.2 %
12.2 %
p value adjusted for age p value 0.0021
Diastolic dysfunction Women
9.4 %
17.4 %
25.4 %
35.5 %
\0.001
Men
11.7 %
11.2 %
15.4 %
24.9 %
0.034
Fig. 1 Distribution of troponin I concentration assessed by a high-sensitivity assay in 4,139 subjects (2,099 men, 2,040 women) according to four categories of troponin concentration. The data was weighted for age of the region Mainz/Mainz-Bingen. The height of the bars represents the percentage of each gender in the different troponin I categories. Concentration of highsensitivity determined troponin I is given in pg/mL. The number below the brackets represent the different hs-TnI categories with category 1 below the LoD and the remaining categories defined using tertiles of the detectable hs-TnI values with the first third being C1.9 and \3.5 pg/mL, second third C3.5 and \5.2 pg/mL and the last third C5.2 pg/mL
correlated inverse with hs-TnI. This reflects cardiac structural and functional changes mirrored by increased NT-proBNP concentration, showing myocardial stretch as encountered in systolic and diastolic heart failure. Renal function is important to decrease troponin I concentration in the circulation, therefore, with decreasing renal function, troponin I concentration increases.
categories (p \ 0.001 for LVEDD in both genders and LVM with p \ 0.001 in men and p = 0.022 in women) as given in Table 1. Figure 3 and Supplemental Table 2 show the positive correlation of left ventricular mass and left ventricular end-diastolic diameter with troponin I concentration (for both genders p \ 0.001, b for men 0.166 and 0.127; b for women 0.084 and 0.079).
Association of troponin I with cardiac structure and function
Cardiac function
Cardiac structure Both left ventricular end-diastolic diameter and left ventricular mass show a positive correlation with hs-TnI
Concentrations of hs-TnI were associated with both systolic and diastolic function of the heart. An inverse correlation was present with LVEF in the results of the censored regression (men p \ 0.001, b = -0.104 and women p = 0.0013, b = -0.049) (Fig. 3 and Supplemental
123
218
Clin Res Cardiol (2014) 103:211–222
Fig. 2 Forest plots for cardiovascular risk factors. Regression coefficients (beta) per increase in one standard deviation are depicted in the forest plots for each gender with the corresponding p value. Each regression was adjusted for age
Table 2). In this context, patients with a lower LVEF had increased hs-TnI concentrations. In a similar fashion, troponin I concentrations were positively associated with tissue Doppler E/E0 (in both genders p \ 0.001, b in men 0.099 and b 0.057 in women) as well the Tei index (for both genders p \ 0.001, b in men 0.076 and 0.054 in women) (Fig. 3 and Supplemental Table 2). Overall, categorical hs-TnI was associated with the aggregate diagnosis of diastolic dysfunction (men p = 0.038 and women p \ 0.001). Association of troponin I with vascular structure and function Vascular function Endothelial dysfunction, as a correlate of vascular function measured by FMD, was weakly associated with troponin I concentrations in women, (Fig. 3 and Supplemental Table 2, p = 0.05). A similar trend was observed in men (p = 0.018). Vascular stiffness as detected by infrared plethysmography was not associated with categorical troponin I (Table 1, men p = 0.21, women p = 1.0).
123
Vascular structure Mean IMT of the carotid artery was correlated positively with troponin I concentrations (p \ 0.001, b = 0.061 in men and p = 0.013, b = 0.047 in women) (Fig. 3 and Supplemental Table 2).
Discussion Previous studies could show that the application of highsensitivity troponin assays optimize the early diagnosis of MIs [3, 4, 21, 34–36]. The availability of such high-sensitivity troponin assays does not only improve acute cardiac care but could also enable risk stratification by troponin assessment in individuals at high risk for cardiovascular events [13, 14, 17, 18, 37]. The evaluation of a new, robust high-sensitivity troponin I assay in 4,139 individuals of the population-based GHS proves on a population basis that troponin I can be detected in over 80 % of all individuals in this large cohort representing the general population of Germany. The range of
Clin Res Cardiol (2014) 103:211–222
219
Fig. 3 Forest plots for cardiovascular phenotypes. Regression coefficients (beta) per increase in one standard deviation are depicted in the forest plots for each gender with the corresponding p value. Each regression was adjusted for age
modern, high-sensitivity assays allow measurements at the LoD of 1.9 pg/mL. However, the LoD is assay-specific and changes with a different manufacturer [38]. The reference population for assay validation studies should ideally be based on demographic characteristics that mirror the European population and it should be remembered that gender and ethnicity are not comorbidities and should be taken into account. Therefore, the selected reference population will influence the measured spectrum of troponin concentration [38]. Most of the variables presented in this study, reflecting cardiovascular phenotypes, correlated with troponin I concentration in the general population and were differently distributed for men and women and influenced by age. The results underline the cardiac-specific nature of troponin I and the correlation to cardiac structure and function resulting in a higher risk of cardiovascular events as recently reported [19]. However, that troponin I also reflects changes in the vascular structure as reflected by correlation with increasing IMT may also be shown by an increased stroke risk as suggested in population studies [19]. In the vast majority of individuals, the detected troponin I levels are far below the cut-off (i.e., 99th percentile of the general population) and are used to detect myocardial ischemia. Most likely, those detectable low troponin concentrations cannot simply be considered as ‘‘background
noise’’; it rather reflects an aggregate measure of cardiacspecific pathophysiology [39]. It is known from population-based studies that hs-TnT correlates with cardiovascular risk factors, age and impaired renal function [13, 14]. Although similar correlations are observed for the hs-TnI concentrations detected in the present population-based cohort, our results further extent current knowledge and demonstrate that low troponin I concentrations are associated with precise measures of impaired cardiac structure and function in the general population. Diastolic dysfunction and systolic dysfunction as markers of cardiac function are associated with troponin I concentrations assayed with high-sensitivity tests, mirroring abnormalities of the cardiac structure like increased end-diastolic diameter of the LV or left ventricular mass [13, 14, 40]. Important are also contributions of chronic sources of myocardial injury to troponin release, baseline concentrations of hs-TnI, below the assay range of contemporary-sensitivity assays (99th percentile below the 10 % CV), strongly associate with prevalent heart failure as reported from studies performed with hs-TnT [14]. Structural vascular correlates like IMT were also connected to increased risk load of arterial hypertension and age contributing to enlarged IMT diameter and sub-atherosclerotic changes of the vasculature [29, 41]. Chronic elevated troponin I concentration should be interpreted as result of structural heart disease, like systolic
123
220
and diastolic heart failure depicting a higher left ventricular mass, lower LVEF and increased myocardial stress, as reflected by NT-proBNP concentration [13, 42, 43]. Overall, the additional information of hs-TnI might be used to determine those subjects in the general population at risk of developing clinically relevant changes in cardiovascular structure and function, which translates into cardiovascular outcome [7, 8, 13, 37]. From a clinical perspective, the application of hs-TnI for cardiovascular risk stratification and decision making in clinical routine is not ready yet. Although improving prediction metrics, it needs to be proven that the change in medical therapies, which is based on risk reclassification, lead to improved patient care [19]. This would constitute a milestone in personalized cardiovascular medicine based on biomarker assessment. Further aspects of cardiovascular disease have already been elucidated by the use of troponin measurement as troponin can be used to evaluate heart failure patients showing correlation with systolic and diastolic function together with NT-proBNP concentration [44]. Further, measurement of troponin in elderly individuals was useful to assess incident heart failure [14]. The lack of a strong association between hs-TnI and endothelial function may be explained by the cardiacspecific origin of troponin release and by the large variation of FMD measures which is inherent in the methods [29]. There is only limited data on the association of troponin I with vascular function in the literature. A previous study with patients suffering from end-stage renal disease has shown a positive correlation of contemporary-sensitivity troponin T with vascular structure measured by IMT indicating asymptomatic atherosclerotic changes [45]. Previous studies describing associations of low levels of cardiac troponins determined by high-sensitivity assays with cardiac and vascular structure and function were mostly conducted in diseased cohorts [12–14, 21] and targeting troponin T rather than troponin I. Our results confirm the data for troponin I as demonstrated with troponin T levels regarding structural variables derived by magnetic resonance imaging [13]. The use of troponin I as a biomarker for risk prediction was earlier suggested in a study employing a biomarker score of NT-proBNP, CRP and troponin I measured with a contemporary-sensitivity assay [46]. With the data from our study that analysis can be further augmented, stressing the importance of hs-TnI concentrations between LoD and the 99th percentile to identify subjects at risk for developing cardiovascular disease. Current study results show that risk assessment of circulating biomarkers over and above the multiple pre-existing cardiovascular risk factors is of considerable interest [19, 46].
123
Clin Res Cardiol (2014) 103:211–222
Limitations Limitations of our study are (a) that individuals in the GHS represent inhabitants of a medium-sized central European city. Therefore, the present data cannot be extrapolated to other populations. Although the age range reflects a broad proportion of the population, extrapolation of the results to older or younger individuals is not possible. In addition, (b) troponin I was measured in frozen samples. (c) Echocardiography and carotid ultrasound measurements were performed by certified medical assistants working according to standard operating procedures, however, intra- and interobserver comparisons were not included into the current manuscript. (d) Detection of troponin I concentration below the 10 % CV of 5.2 ng/mL is associated with a higher imprecision. (e) Every subject was informed not to perform exercise 8 h before the study visit as this may influence biomarker concentrations but biomarker concentration could be influenced by previous exercise. (f) Repeated measurements are not present currently as the 5 year follow-up is still ongoing. Categorizing subjects from the general population to aggregate disease diagnoses like diastolic or systolic heart failure may be influenced by low numbers. Follow-up was not complete during presentation of the results of this study. An association with the primary cardiovascular endpoint can, therefore, not be reported. Conclusion In conclusion, troponin I concentration determined with a high-sensitivity assay in the general population is associated with phenotypes of cardiovascular structure, function and vascular structure. Due to its cardiac-specific origin, an association with vascular function could not be shown. In this context, circulating troponin concentration is driven by left ventricular hypertrophy and systolic and diastolic dysfunction. Our results provide pathophysiological background for the role of hs-TnI in cardiovascular risk stratification in primary prevention settings. Acknowledgments The Gutenberg Health Study (GHS) is funded through the government of Rhineland-Palatinate (‘‘Stiftung Rheinland Pfalz fu¨r Innovation,’’ contract No. AZ 961–386261/733), the research programs ‘‘Wissen schafft Zukunft’’ and ‘‘Schwerpunkt Vaskula¨re Pra¨vention’’ of the Johannes Gutenberg-University of Mainz, and its contract with Boehringer Ingelheim and Philips Medical Systems, including an unrestricted grant for the GHS. Conflict of interest Abbott Diagnostics provided test reagents for high-sensitivity troponin I measurements. Stefan Blankenberg has received honoraria from Abbott Diagnostics, Siemens, Thermo Fisher, and Roche Diagnostics and is a consultant for Thermo Fisher. All other co-authors did not report any conflict of interest.
Clin Res Cardiol (2014) 103:211–222
221
References 1. Jaffe AS (2011) The 10 commandments of troponin, with special reference to high sensitivity assays. Heart 97(11):940–946. doi:10.1136/hrt.2009.185751 2. Mueller M, Celik S, Biener M, Vafaie M, Schwoebel K, Wollert KC, Januzzi JL, Katus HA, Giannitsis E (2012) Diagnostic and prognostic performance of a novel high-sensitivity cardiac troponin T assay compared to a contemporary sensitive cardiac troponin I assay in patients with acute coronary syndrome. Clin Res Cardiol 101(10):837–845. doi:10.1007/s00392-012-0469-6 3. Keller T, Zeller T, Peetz D, Tzikas S, Roth A, Czyz E, Bickel C, Baldus S, Warnholtz A, Frohlich M, Sinning CR, Eleftheriadis MS, Wild PS, Schnabel RB, Lubos E, Jachmann N, Genth-Zotz S, Post F, Nicaud V, Tiret L, Lackner KJ, Munzel TF, Blankenberg S (2009) Sensitive troponin I assay in early diagnosis of acute myocardial infarction. N Engl J Med 361(9):868–877. doi:10.1056/NEJMoa0903515 4. Reichlin T, Hochholzer W, Bassetti S, Steuer S, Stelzig C, Hartwiger S, Biedert S, Schaub N, Buerge C, Potocki M, Noveanu M, Breidthardt T, Twerenbold R, Winkler K, Bingisser R, Mueller C (2009) Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. N Engl J Med 361(9):858–867. doi:10. 1056/NEJMoa0900428 5. Kurz K, Giannitsis E, Becker M, Hess G, Zdunek D, Katus HA (2011) Comparison of the new high sensitive cardiac troponin T with myoglobin, h-FABP and cTnT for early identification of myocardial necrosis in the acute coronary syndrome. Clin Res Cardiol 100(3):209–215. doi:10.1007/s00392-010-0230-y 6. Celik S, Giannitsis E, Wollert KC, Schwobel K, Lossnitzer D, Hilbel T, Lehrke S, Zdunek D, Hess A, Januzzi JL, Katus HA (2011) Cardiac troponin T concentrations above the 99th percentile value as measured by a new high-sensitivity assay predict long-term prognosis in patients with acute coronary syndromes undergoing routine early invasive strategy. Clin Res Cardiol 100(12):1077–1085. doi:10.1007/s00392-011-0344-x 7. Apple FS, Ler R, Murakami MM (2012) Determination of 19 cardiac troponin I and T assay 99th percentile values from a common presumably healthy population. Clin Chem 58(11):1574–1581. doi:10.1373/clinchem.2012.192716 8. Apple FS, Steffen LM, Pearce LA, Murakami MM, Luepker RV (2012) Increased cardiac troponin I as measured by a high-sensitivity assay is associated with high odds of cardiovascular death: the Minnesota Heart Survey. Clin Chem 58(5):930–935. doi:10.1373/clinchem.2011.179176 9. Giannitsis E, Kehayova T, Vafaie M, Katus HA (2011) Combined testing of high-sensitivity troponin T and copeptin on presentation at prespecified cutoffs improves rapid rule-out of non-STsegment elevation myocardial infarction. Clin Chem 57(10):1452–1455. doi:10.1373/clinchem.2010.161265 10. Keller T, Zeller T, Ojeda F, Tzikas S, Lillpopp L, Sinning C, Wild P, Genth-Zotz S, Warnholtz A, Giannitsis E, Mockel M, Bickel C, Peetz D, Lackner K, Baldus S, Munzel T, Blankenberg S (2011) Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction. JAMA 306(24):2684–2693. doi:10.1001/jama.2011.1896 11. Saunders JT, Nambi V, de Lemos JA, Chambless LE, Virani SS, Boerwinkle E, Hoogeveen RC, Liu X, Astor BC, Mosley TH, Folsom AR, Heiss G, Coresh J, Ballantyne CM (2011) Cardiac troponin T measured by a highly sensitive assay predicts coronary heart disease, heart failure, and mortality in the atherosclerosis risk in communities study. Circulation 123(13):1367–1376. doi:10.1161/CIRCULATIONAHA.110.005264 12. Giannitsis E, Becker M, Kurz K, Hess G, Zdunek D, Katus HA (2010) High-sensitivity cardiac troponin T for early prediction of
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
evolving non-ST-segment elevation myocardial infarction in patients with suspected acute coronary syndrome and negative troponin results on admission. Clin Chem 56(4):642–650. doi:10. 1373/clinchem.2009.134460 de Lemos JA, Drazner MH, Omland T, Ayers CR, Khera A, Rohatgi A, Hashim I, Berry JD, Das SR, Morrow DA, McGuire DK (2010) Association of troponin T detected with a highly sensitive assay and cardiac structure and mortality risk in the general population. JAMA 304(22):2503–2512. doi:10.1001/ jama.2010.1768 deFilippi CR, de Lemos JA, Christenson RH, Gottdiener JS, Kop WJ, Zhan M, Seliger SL (2010) Association of serial measures of cardiac troponin T using a sensitive assay with incident heart failure and cardiovascular mortality in older adults. JAMA 304(22):2494–2502. doi:10.1001/jama.2010.1708 Venge P, James S, Jansson L, Lindahl B (2009) Clinical performance of two highly sensitive cardiac troponin I assays. Clin Chem 55(1):109–116. doi:10.1373/clinchem.2008.106500 Venge P, Johnston N, Lindahl B, James S (2009) Normal plasma levels of cardiac troponin I measured by the high-sensitivity cardiac troponin I access prototype assay and the impact on the diagnosis of myocardial ischemia. J Am Coll Cardiol 54(13):1165–1172. doi:10.1016/j.jacc.2009.05.051 Frankenstein L, Remppis A, Giannitis E, Frankenstein J, Hess G, Zdunek D, Doesch A, Zugck C, Katus HA (2011) Biological variation of high sensitive Troponin T in stable heart failure patients with ischemic or dilated cardiomyopathy. Clin Res Cardiol 100(8):633–640. doi:10.1007/s00392-011-0285-4 Eggers KM, Venge P, Lindahl B, Lind L (2013) Cardiac troponin I levels measured with a high-sensitive assay increase over time and are strong predictors of mortality in an elderly population. J Am Coll Cardiol 61(18):1906–1913. doi:10.1016/j.jacc.2012. 12.048 Zeller T, Tunstall-Pedoe H, Saarela O, Ojeda F, Schnabel RB, Tuovinen T, Woodward M, Struthers A, Hughes M, Kee F, Salomaa V, Kuulasmaa K, Blankenberg S (2013) High population prevalence of cardiac troponin I measured by a high-sensitivity assay and cardiovascular risk estimation: the MORGAM Biomarker Project Scottish Cohort. Eur Heart J. doi:10.1093/ eurheartj/eht406 Wang TJ, Wollert KC, Larson MG, Coglianese E, McCabe EL, Cheng S, Ho JE, Fradley MG, Ghorbani A, Xanthakis V, Kempf T, Benjamin EJ, Levy D, Vasan RS, Januzzi JL (2012) Prognostic utility of novel biomarkers of cardiovascular stress: the Framingham Heart Study. Circulation 126(13):1596–1604. doi:10. 1161/CIRCULATIONAHA.112.129437 Eggers KM, Jaffe AS, Lind L, Venge P, Lindahl B (2009) Value of cardiac troponin I cutoff concentrations below the 99th percentile for clinical decision-making. Clin Chem 55(1):85–92. doi:10.1373/clinchem.2007.101683 Wild PS, Sinning CR, Roth A, Wilde S, Schnabel RB, Lubos E, Zeller T, Keller T, Lackner KJ, Blettner M, Vasan RS, Munzel T, Blankenberg S (2010) Distribution and categorization of left ventricular measurements in the general population: results from the population-based Gutenberg Heart Study. Circ Cardiovasc Imaging 3(5):604–613. doi:10.1161/CIRCIMAGING.109. 911933 Zeller T, Wild P, Szymczak S, Rotival M, Schillert A, Castagne R, Maouche S, Germain M, Lackner K, Rossmann H, Eleftheriadis M, Sinning CR, Schnabel RB, Lubos E, Mennerich D, Rust W, Perret C, Proust C, Nicaud V, Loscalzo J, Hubner N, Tregouet D, Munzel T, Ziegler A, Tiret L, Blankenberg S, Cambien F (2010) Genetics and beyond—the transcriptome of human monocytes and disease susceptibility. PLoS ONE 5(5):e10693. doi:10.1371/journal.pone.0010693
123
222 24. Wild PS, Zeller T, Beutel M, Blettner M, Dugi KA, Lackner KJ, Pfeiffer N, Munzel T, Blankenberg S (2012) The Gutenberg Health Study. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 55(6–7):824–829. doi:10.1007/s00103-0121502-7 25. Lean ME, Han TS, Morrison CE (1995) Waist circumference as a measure for indicating need for weight management. BMJ 311(6998):158–161 26. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150(9):604–612 150/9/604 [pii] 27. Matsushita K, Mahmoodi BK, Woodward M, Emberson JR, Jafar TH, Jee SH, Polkinghorne KR, Shankar A, Smith DH, Tonelli M, Warnock DG, Wen CP, Coresh J, Gansevoort RT, Hemmelgarn BR, Levey AS (2012) Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate. JAMA 307(18):1941–1951. doi:10. 1001/jama.2012.3954 28. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N (1986) Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 57(6):450–458 0002-9149(86)90771-X [pii] 29. Schnabel RB, Schulz A, Wild PS, Sinning CR, Wilde S, Eleftheriadis M, Herkenhoff S, Zeller T, Lubos E, Lackner KJ, Warnholtz A, Gori T, Blankenberg S, Munzel T (2011) Noninvasive vascular function measurement in the community: crosssectional relations and comparison of methods. Circ Cardiovasc Imaging 4(4):371–380. doi:10.1161/CIRCIMAGING.110.961557 30. Dixon WJ (1953) Processing data for outliers. Biometrics 9:74–89 31. Tobin J (1958) Estimation of relationships for limited dependent variables. Econometrica 26:24–36 32. Amemiya T (1984) Tobit models: a survey. J Econom 24:3–61 33. Team RDC (2011) R: a language and environment for statistical computing. R Foundation for Statistical Computing. http://www. R-project.org 34. Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA (2010) Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 56(2):254–261. doi:10.1373/ clinchem.2009.132654 35. Weber M, Bazzino O, Navarro Estrada JL, de Miguel R, Salzberg S, Fuselli JJ, Liebetrau C, Woelken M, Moellmann H, Nef H, Hamm C (2011) Improved diagnostic and prognostic performance of a new high-sensitive troponin T assay in patients with acute coronary syndrome. Am Heart J 162(1):81–88. doi:10. 1016/j.ahj.2011.04.007 36. Apple FS, Smith SW, Pearce LA, Ler R, Murakami MM (2008) Use of the Centaur TnI-Ultra assay for detection of myocardial infarction and adverse events in patients presenting with symptoms suggestive of acute coronary syndrome. Clin Chem 54(4):723–728. doi:10.1373/clinchem.2007.097162 37. Otsuka T, Kawada T, Ibuki C, Seino Y (2010) Association between high-sensitivity cardiac troponin T levels and the
123
Clin Res Cardiol (2014) 103:211–222
38.
39.
40.
41.
42.
43.
44.
45.
46.
predicted cardiovascular risk in middle-aged men without overt cardiovascular disease. Am Heart J 159(6):972–978. doi:10.1016/ j.ahj.2010.02.036 S0002-8703(10)00245-0 [pii] Korley FK, Jaffe AS (2013) Preparing the United States for highsensitivity cardiac troponin assays. J Am Coll Cardiol. doi:10. 1016/j.jacc.2012.09.069 Providencia R, Barra S, Paiva L (2013) Atrial fibrillation, elevated troponin, ischemic stroke and adverse outcomes: understanding the connection. Clin Res Cardiol 102(10):701–711. doi:10.1007/s00392-013-0591-0 Dinh W, Nickl W, Futh R, Lankisch M, Hess G, Zdunek D, Scheffold T, Barroso MC, Tiroch K, Ziegler D, Seyfarth M (2011) High sensitive troponin T and heart fatty acid binding protein: novel biomarker in heart failure with normal ejection fraction: a cross-sectional study. BMC Cardiovasc Disord 11:41. doi:10.1186/1471-2261-11-41 Flu WJ, van Kuijk JP, Hoeks SE, Kuiper R, Schouten O, Goei D, Winkel T, van Gestel YR, Verhagen HJ, Bax JJ, Poldermans D (2009) Intima media thickness of the common carotid artery in vascular surgery patients: a predictor of postoperative cardiovascular events. Am Heart J 158(2):202–208. doi:10.1016/j.ahj. 2009.05.028 Nadir MA, Rekhraj S, Wei L, Lim TK, Davidson J, MacDonald TM, Lang CC, Dow E, Struthers AD (2012) Improving the primary prevention of cardiovascular events by using biomarkers to identify individuals with silent heart disease. J Am Coll Cardiol 60(11):960–968. doi:10.1016/j.jacc.2012.04.049 Schnabel RB, Schulz A, Messow CM, Lubos E, Wild PS, Zeller T, Sinning CR, Rupprecht HJ, Bickel C, Peetz D, Cambien F, Kempf T, Wollert KC, Benjamin EJ, Lackner KJ, Munzel TF, Tiret L, Vasan RS, Blankenberg S (2010) Multiple marker approach to risk stratification in patients with stable coronary artery disease. Eur Heart J 31(24):3024–3031. doi:10.1093/ eurheartj/ehq322 Kusumoto A, Miyata M, Kubozono T, Ikeda Y, Shinsato T, Kuwahata S, Fujita S, Takasaki K, Yuasa T, Hamasaki S, Tei C (2012) Highly sensitive cardiac troponin T in heart failure: comparison with echocardiographic parameters and natriuretic peptides. J Cardiol 59(2):202–208. doi:10.1016/j.jjcc.2011.11. 012 Hojs R, Ekart R, Hojs Fabjan T, Balon BP, Gorenjak M (2005) Cardiac troponin T (cTnT) in hemodialysis patients with asymptomatic and symptomatic atherosclerosis. Arch Med Res 36(4):367–371. doi:10.1016/j.arcmed.2005.03.024 Blankenberg S, Zeller T, Saarela O, Havulinna AS, Kee F, Tunstall-Pedoe H, Kuulasmaa K, Yarnell J, Schnabel RB, Wild PS, Munzel TF, Lackner KJ, Tiret L, Evans A, Salomaa V (2010) Contribution of 30 biomarkers to 10-year cardiovascular risk estimation in 2 population cohorts: the MONICA, risk, genetics, archiving, and monograph (MORGAM) biomarker project. Circulation 121(22):2388–2397. doi:10.1161/CIRCULATIONAHA. 109.901413
ORIGINAL PAPER
Association of high-sensitivity assayed troponin I with cardiovascular phenotypes in the general population: the population-based Gutenberg health study Christoph Sinning • Till Keller • Tanja Zeller • Francisco Ojeda • Michael Schlu¨ter • Renate Schnabel • Edith Lubos • Christoph Bickel • Karl J. Lackner • Patrick Diemert • Thomas Munzel • Stefan Blankenberg • Philipp S. Wild • for the Gutenberg Health Study
Received: 14 September 2013 / Accepted: 8 November 2013 / Published online: 23 November 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Background Aim of the study was to analyze the correlation of high-sensitivity assayed troponin I with cardiac and vascular structure and function in a large populationbased cohort. Methods In a sample of 4,139 subjects (2,099 men, 2,040 women, age 35–74 years) from the population-based Gutenberg Health Study, troponin I was measured with a high-sensitivity assay that had a limit of detection of 1.9 pg/mL. Results In the study cohort, 3,405 subjects had detectable troponin I concentrations [82.3 % overall, 89.9 % men (N = 1,888), 74.4 % women (N = 1,517)]. All analyses
C. Sinning and T. Keller contributed equally. S. Blankenberg and P. S. Wild were both senior authors and contributed equally.
Electronic supplementary material The online version of this article (doi:10.1007/s00392-013-0640-8) contains supplementary material, which is available to authorized users. C. Sinning T. Keller T. Zeller F. Ojeda M. Schlu¨ter R. Schnabel E. Lubos P. Diemert S. Blankenberg (&) Department of General and Interventional Cardiology, University Heart Center Hamburg, Martinistr. 52, 20246 Hamburg, Germany e-mail: [email protected] C. Bickel Department of Internal Medicine, Federal Armed Forces Central Hospital, Koblenz, Germany
were adjusted for age. The strongest correlate between detectable troponin I and measures of cardiac phenotypes was observed for left ventricular mass (p \ 0.001) and left ventricular end-diastolic diameter (p \ 0.001) for both, women and men. Left ventricular ejection fraction was inversely correlated with troponin I (p value\0.001 in men and 0.0013 in women), also measures of diastolic dysfunction as represented by Tei index and E/E0 correlated with detectable troponin I concentrations (p \ 0.001 for both gender). With respect to vascular structure and function, troponin I correlated with mean intima-media thickness of the carotid artery (p \ 0.001 in men and p = 0.013 in women) but showed only borderline correlation with measures of vascular function represented by flow-mediated dilation (p = 0.05 in women and p = 0.018 in men) and arterial stiffness. Conclusions Troponin I assessed by a high-sensitivity assay correlated with measures of left ventricular hypertrophy and systolic and diastolic function, whereas its correlation with vascular phenotypes was only of weak magnitude.
T. Munzel P. S. Wild (&) Department of Medicine II, University Medical Center, Johannes Gutenberg University Mainz, Langenbeckstr. 1, 55131 Mainz, Germany e-mail: [email protected] P. S. Wild Center for Thrombosis and Hemostasis, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany
K. J. Lackner Institute of Clinical Chemistry and Laboratory Medicine, University Medical Center Johannes Gutenberg University, Mainz, Germany
123
212
Keywords High-sensitivity assayed troponin I Cardiac structure Cardiac function Endothelial dysfunction Population study Abbreviations BMI Body mass index CAD Coronary artery disease CCA Common carotid artery CRP C-reactive protein CV Coefficient of variation eGFR Estimated glomerular filtration rate FMD Flow-mediated dilation GHS Gutenberg Health Study hs-TnI Troponin I as measured by a high-sensitivity assay hs-TnT Troponin T as measured by a high-sensitivity assay LoD Limit of detection LV Left ventricular LVEDD Left ventricular end-diastolic diameter LVEF Left ventricular ejection fraction LVM Left ventricular mass MI Myocardial infarction NT-proBNP N-terminal pro-B-type natriuretic peptide SD Standard deviation
Introduction Evolution of troponin assays has recently led to the determination of troponin via high-sensitivity assays [1, 2]. Several studies have shown in real-world settings that the application of both contemporary-sensitivity and high-sensitivity assays for troponin I and T improves the differential diagnosis of patients with suspected acute coronary syndrome [3–6]. By the use of modern, highsensitivity assays, cardiac troponin concentrations can now be reliably detected in the majority of the presumably healthy population with a substantial magnitude of variety among the various assays [1, 7, 8] [9–11]. However, the pathophysiology of those detectable very low concentrations and its clinical utility is poorly understood [3, 4, 10, 12]. In these cases, acute ischemia cannot be held responsible for troponin release in population-based cohorts and other pathophysiologic conditions must account for it [13–17]. Previous studies have already shown the distribution of hs-Troponin T in the population and association with cardiac phenotypes [13, 14]. Results from the Dallas Heart Study could show that troponin T was detectable in 25 % of the general population and
123
Clin Res Cardiol (2014) 103:211–222
additionally identified individuals with structural heart disease and increased risk of all-cause mortality [13]. Further, as older adults are at an increased risk of newonset of heart failure, hs-TnT proved to be reliable in detecting those individuals with incident heart failure and increased risk of cardiovascular death by means of baseline and serial troponin T measurements [14]. However, whereas hs-Troponin T detects only a limited percentage in the younger and healthy population, there is scarce information so far available linking the entire troponin I concentration range with intermediate phenotypes of cardiac and vascular structure and function in the general population. Therefore, a troponin I assay was applied in this study, reported to detect up to 96 % of the apparently healthy population aged 18–64 years [7]. Actual studies using hs-TnI assays in the general population of Scotland and older individuals in Sweden report that an increased troponin I concentration is related to future cardiovascular events and mortality [18, 19]. Previous results of the Framingham Heart Study could show that biomarkers reflecting cardiovascular stress are detectable in the general population and provide additional prognostic information to traditional cardiovascular risk factors [20]. For this study, an ultrasensitive assay assessing troponin I through a single-molecule counting technique was used, enabling measurement in over 80 % of the subjects [20]. In this context, a confirmatory study describing the association of troponin I measured with a high-sensitive assay and parameters of cardiac and vascular structure was not present so far. The intention of this study is to investigate these cardiovascular phenotypes in relation to hs-TnI concentration measured in 4,139 subjects of the population-based Gutenberg Health Study (GHS) [13, 14, 21].
Methods Study population Recruitment into the GHS was conducted between April 2007 and April 2012, finally including 15,010 individuals. A detailed study description has been published earlier [22, 23]. In brief, study individuals aged 35–74 years and stratified according to gender and age were selected randomly by the registration office from the city of Mainz. High-sensitivity assayed troponin I (hs-TnI) measurements were performed in 4,139 of the first 5,000 study participants. The GHS was approved by the Ethics Committee of Rhineland-Palatinate and the medical faculty of the Johannes Gutenberg-University, Mainz. Each study individual provided written informed consent before
Clin Res Cardiol (2014) 103:211–222
participating. The ethical application complied with the Declaration of Helsinki. All authors have read and approved the manuscript in its current form. Assessment of risk factors Cardiovascular risk factors and diseases Risk factors were assessed as outlined in previous publications [24]. Former history of stroke, coronary artery disease (CAD), myocardial infarction (MI), heart failure and peripheral artery disease were assessed in a standardized interview. Arterial hypertension was defined as a systolic blood pressure C140 mmHg and diastolic blood pressure C90 mmHg at rest obtained as the mean of the second and third measurement, or by taking any antihypertensive drugs within the last 2 weeks. Diabetes mellitus was defined as a fasting glucose C126 mg/dL, a spontaneous glucose concentration of C200 mg/dL, or as diagnosed by a physician. We defined dyslipidemia as a LDL/ HDL ratio of [3.5 or as diagnosed by a physician. Waist circumference C102 cm for men or C88 cm for women identified subjects with body mass index C30 and was used, therefore, for subjects classified as obese [25]. Smokers were classified into daily smokers (C1 cigarette/ day), occasional smokers (\1 cigarette/day), former smokers and non-smokers (never smoked). Any family history of MI in first-degree relatives before the age of 65 years for females and before the age of 60 years for males was defined as positive family history. For glomerular filtration rate (GFR), as the best marker for renal function in health and disease, we used the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation [26]. Since creatinine measurements according to the method described by Jaffe are not IDMS-traceable (being the ‘‘standard creatinine’’) we multiply the creatinine value by 0.95 before using the CKD-EPI equation as described by Matsushita et al. [27].
213
Myocardial function Myocardial function was assessed by LV ejection fraction [(LVEF) in %] as (end-diastolic volume-end-systolic volume)/end-diastolic volume 9 100, with volumes determined according to the Simpson method in 4- and 2-chamber views and Tei index. The Doppler-derived Tei index, also called myocardial performance index, was calculated as the sum of isovolumic contraction time and isovolumic relaxation time divided by LV ejection time. The E/E0 ratio of diastolic function was calculated by dividing the early filling velocity of transmitral Doppler (E) by the early relaxation velocity on tissue Doppler (E0 ). Vascular measures representing structure and function Intima-media thickness Intima-media thickness (IMT) of the carotid artery was assessed using an iE33 ultrasound system (Royal Philips Electronics, Amsterdam, The Netherlands) using an 11- to 3-MHz linear-array transducer. Evaluation was performed using an automatic computerized system (QLAB; Philips Medical Systems), and triggering was performed according to the Q wave on electrocardiography to enable measurement in complete relaxation of the ventricle. Mean IMT was recorded 1 cm proximal to the carotid bulb over a length of 1 cm at the far wall of both common carotid arteries (CCA). Flow-mediated dilation Flow-mediated dilation (FMD in %), defined as (mean brachial artery diameter [hyperemia]-mean brachial artery diameter [baseline])/mean brachial artery diameter [baseline] 9 100, was measured using a 5-min upper-arm occlusion, with the brachial artery diameter measured in resting condition (baseline) and after induction of local reactive hyperemia [29]. In the study cohort, 117 FMD values were missing because of measurement errors.
Cardiac phenotypes representing structure and function Infrared plethysmography Myocardial structure All subjects underwent multimodal echocardiography with an iE33 echocardiography system with an S5–1 sector array transducer (Royal Philips Electronics, Amsterdam, The Netherlands), a phased array with 80 elements and a 5- to 1-MHz operating frequency range [22]. The examinations were performed according to standard operating procedures by trained and certified medical technical assistants at a single center. Measurements were according to recommendations by the American Society of Echocardiography [28].
The digital volume pulse was continuously obtained by measuring the transmission of infrared light through the pulp of the right ring finger with a PulseTrace 2000 device (Cardinal Health/Micro Medical Limited; Rochester, UK) [29]. Laboratory analyses Routine laboratory parameters were measured by standardized methods. Troponin I was assessed using a commercially available high-sensitivity cardiac troponin assay (ARCHITECT
123
214
STAT highly sensitive Troponin I immunoassay, Abbott Diagnostics, USA, ARCHITECT i2000SR). The limit of detection (LoD) for the assay was 1.9 pg/mL (assay range 0–50,000 pg/mL). The assay has a 10 % coefficient of variation at a concentration of 5.2 pg/mL. Intra-assay and inter-assay coefficients of variation were 4.26 and 6.29 %, respectively [19]. In this context, measurable concentrations relate to concentrations above 1.9 pg/mL representing the LoD as reported by the manufacturer [19]. Further it has to be mentioned that different studies using the same assay had a different LoD ranging from 1.2 to 1.5 pg/mL resulting in a larger part of the subjects having measurable troponin I concentration [8, 18]. N-terminal pro-B-type natriuretic peptide (NT-proBNP) was measured with a commercially available assay on the ELECSYS 2010 using the Elecysys proBNP II assay (ECLIA, Roche Diagnostics, Mannheim, Germany) with an assay range of 5–35,000 pg/mL and a total imprecision of 1.5–2.5 %. Lipid values have been obtained by routine methods. All laboratory measurements were carried out blinded to the clinical course of the study subjects. Statistical analyses Analyses were stratified by gender. To take into account the stratified nature of the sample, baseline characteristics (Tables 1, 2) and analyses involving the distribution of high-sensitivity determined troponin I (Fig. 1) were weighted according to the age distribution in the study population (N = 210,867, data of the German Federal Statistical Office, Wiesbaden, 2007). In those analyses without gender stratification (categories of troponin I and values in Fig. 1) the statistics were weighted for the age and sex distribution in the study population. High-sensitivity determined troponin I was analyzed as continuous and categorical variable. To define the latter, the subjects were divided into four categories with the lowest category encompassing the individuals with troponin I concentrations below the LoD of the assay (\1.9 pg/mL 10.1 % in men and 25.6 % in women) (Fig. 1). The further concentration range was divided into thirds using tertiles (3.5, 5.2 pg/mL), the first third (and second category) was C1.9 and \3.5 pg/mL, the second third C3.5 and \5.2 pg/mL and the last category C5.2 pg/mL. A male subject with an hs-TnI value of 1,140.9 pg/mL was classified as an outlier and excluded from all analyses [30]. For the descriptive analyses of baseline cardiovascular risk factors and clinical variables, continuous variables are described by median, 25th and 75th percentile or by mean and standard deviation (SD), binary variables by percentages. Trend tests were applied via age-adjusted logistic or quantile regression, to examine the association of different
123
Clin Res Cardiol (2014) 103:211–222
phenotypes and a categorized troponin I (assigning values 1,2,3,4 to the troponin I categories). Tobit regression [31, 32], was used to take into account the fact that troponin I values below the LoD are censored (Figs. 2, 3). Regression coefficients (betas) per increase in 1 standard deviation are presented with the corresponding p values. The supplement for the information presented in Figs. 2 and 3 is expanded (Supplemental Table 1 and 2). All analyses were performed using R 2.5.2[33]. (R Development Core Team (2009). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org).
Results Distribution of high-sensitivity assayed troponin I The distribution of troponin I concentration assessed by the high-sensitivity assay in the 4,139 study individuals is displayed in Fig. 1. Detectable troponin I concentrations were found in 3,405 subjects of the study cohort (82.3 % overall, 89.9 % men (N = 1,888), 74.4 % women (N = 1,517), if weighted by age distribution according to the data of the German Federal Statistical Office, these number are 81.6 % for the overall population, 89.9 % in men and 73.1 % in women, respectively). In this study, the troponin concentrations were stratified into categories with the lowest category with troponin I concentrations below the LoD of the assay (\1.9 pg/mL, N = 523 in women and N = 211 in men). The three further categories encompassed the thirds with the first third ranging from C1.9 to \3.5 pg/mL (N = 665 in women and N = 514 in men), the second third C3.5 and\5.2 pg/mL (N = 509 in women and N = 606 in men) and the last third C5.2 pg/mL (N = 343 in women and N = 767 in men). In this cohort, troponin I concentrations increase with age to about 49 years for both genders in category 1, and 60 and 55 years for women and men, respectively, in category 4 (Table 1). Therefore, all further statistical analyses correlating troponin I with cardiac structure and function were adjusted for age. The numbers of men were higher in troponin categories 3 and 4 and women were more often present in the category below the LoD and the second category. Troponin I and classical risk factors/cardiovascular diseases In the baseline table showing the results for the ageadjusted analysis, the classical cardiovascular risk factors
Clin Res Cardiol (2014) 103:211–222
215
Table 1 Baseline characteristics according to the 4 hs-TnI categories hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
48.3 ± 10.1
50.6 ± 10.7
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
p value adjusted for age p value
Age (years) Women
Men 49.6 ± 9.2 48.2 ± 10.1 Cardiovascular risk factors and diseases
54.9 ± 10.4
59.7 ± 10.7
51 ± 10.9
55.2 ± 11.7
Hypertension (%) Women
31.4 %
38.2 %
48.8 %
58.3 %
0.020
Men
31.3 %
36.6 %
52 %
60.3 %
\0.001
Diabetes (%) Women
2.2 %
3.8 %
4.7 %
9.2 %
0.019
Men
5.8 %
5.8 %
4.3 %
12.9 %
0.056
Obesity (BMI [ 30 kg/m2) (%) Women
14.3 %
19.5 %
27 %
31.5 %
\0.001
Men
15.1 %
18.3 %
26.2 %
31 %
\0.001
Dyslipidemia (%) Women
10 %
15.6 %
26 %
40.2 %
\0.001
Men
22 %
27.6 %
40.6 %
45.4 %
\0.001
18.6 % 26.8 %
15.9 % 22.3 %
18.6 % 21.1 %
0.61 0.70
Active smoking (%) Women Men
22 % 24.3 %
Prevalent CAD (%) Women
0.9 %
1.6 %
2.1 %
3%
0.96
Men
0.9 %
3%
5.2 %
9%
\0.001
History of MI (%) Women
0.4 %
0.6 %
1.9 %
2.9 %
0.049
Men
0.8 %
1.4 %
2.4 %
7.1 %
\0.001
Family history of MI (%) Women
16.1 %
17 %
18.2 %
21.9 %
0.12
Men
16.9 %
15.9 %
17.8 %
16.5 %
0.97
Medicated heart failure (%) Women
1.1 %
0.7 %
1.3 %
2.9 %
0.91
Men
0%
0.6 %
0.6 %
1.8 %
0.024
arterial hypertension, obesity and dyslipidemia were associated with hs-TnI concentration (Table 1). For both genders, an increase in the prevalence of hypertension from troponin category 1 to category 4 was observed (men p \ 0.001 and women p = 0.02). The prevalence of diabetes was higher in subjects within the elevated troponin I categories, irrespective of gender, however, significance was only reached for women (p = 0.019) and was borderline in men (p = 0.056). The prevalence of CAD and previous MI was higher in men with higher troponin I concentrations (both p values \0.001). The results for the CV risk factors show the influence of age, BMI (both p value \0.001) and history of MI (p value in men \0.001 and 0.041 in women) (Supplemental Table 1 and 2).
Troponin I and natriuretic peptides/renal function The categorized troponin I concentration was associated with Nt-proBNP and eGFR in men (p \ 0.001 and p = 0.0081, respectively) but not in women (p = 0.12 for Nt-proBNP and p = 0.15 for eGFR) (Table 2). Also the distribution of troponin I was not associated with Nt-proBNP and eGFR in women according to the different categories, and the association between cardiovascular phenotypes and hs-TnI was more accurately assessed by censored regression analysis (Supplemental Table 1 and 2). NT-proBNP values (both genders p value of \0.001, b = 0.164 in men and b = 0.063 in women) were correlated positive with hs-TnI and eGFR calculated with the CKD-EPI equation (men p value of\0.001, b = -0.107 and women p = 0.0012, b = -0.059) was
123
216
Clin Res Cardiol (2014) 103:211–222
Table 2 Characteristics of interest according to the 4 hs-TnI categories defined hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
p value adjusted for age p value
69 (38, 118) 26 (8, 56)
80 (45, 142) 29 (10, 63)
109 (58, 235) 49 (19, 130)
0.12 \0.001
Biomarker NT-proBNP (pg/mL) Women Men
69 (39, 114) 29 (9, 52)
eGFR (mL/min for 1.73 m2) Women
93.9 ± 13.1
91.5 ± 12.5
87.8 ± 13.8
83.7 ± 14.2
0.15
Men
97 ± 10.7
97.8 ± 11.8
95.1 ± 12
90.6 ± 15.1
0.0081
Cardiac structure Left ventr. end-diastolic dia. (mm) Women
42.8 ± 3.6
43.4 ± 3.9
43.4 ± 4.1
43.4 ± 4.9
\0.001
Men
46.8 ± 4.2
46.7 ± 3.9
47.3 ± 4.1
48.2 ± 4.9
\0.001
Left ventricular mass (g) Women
120 (101, 142)
124 (106, 145)
130 (111, 154)
135 (116, 164)
Men
168 (143, 194)
167 (143, 191)
175 (153, 201)
185 (158, 221)
0.0022 \ 0.001
Cardiac function LVEF (%) 64.4 ± 5.9
64.6 ± 5.3
64.3 ± 5.4
63.8 ± 6.3
\0.001
Men 62.9 ± 5.6 Tei index
63.8 ± 5.9
63.4 ± 6.1
62.1 ± 7.4
0.13
Women
Women
0.5 (0.42, 0.58)
0.53 (0.45, 0.61)
0.54 (0.45, 0.63)
0.56 (0.47, 0.64)
0.0017
Men
0.51 (0.44, 0.59)
0.54 (0.45, 0.63)
0.54 (0.43, 0.64)
0.56 (0.47, 0.65)
0.01
E/E0 Women
6.49 (5.53, 7.72)
6.92 (5.82, 8.23)
7.25 (5.84, 9.33)
7.78 (6.41, 9.22)
0.44
Men
5.92 (5, 7.17)
6.18 (5.22, 7.53)
6.28 (5.26, 7.59)
6.9 (5.63, 8.8)
0.0082
Vascular function FMD (%) Women
11 ± 6.2
10.4 ± 5.8
9.6 ± 5.4
8.4 ± 5.7
0.16
Men
6.6 ± 4.3
6.6 ± 4.1
6.3 ± 3.8
6.2 ± 3.7
0.040
Reflection index (Post–Pre) Women
-4.5 ± 12.5
-4 ± 14.4
-4.4 ± 14.1
-3 ± 15.8
1.0
Men
-0.5 ± 9.9
-1.2 ± 12.9
-0.6 ± 12.1
-1.1 ± 10.9
0.21
0.603 ± 0.111 0.604 ± 0.116
0.633 ± 0.109 0.64 ± 0.128
0.671 ± 0.125 0.673 ± 0.136
0.0056 0.019
Vascular structure IMT (mean) (mm) Women Men
0.58 ± 0.1 0.618 ± 0.126
Lipids HDL (mg/dL) Women
65.5 ± 15.2
64.1 ± 14.9
61.3 ± 15
61.2 ± 15.5
\0.001
Men
50.7 ± 13.4
50.7 ± 12.6
48.7 ± 12.6
48.3 ± 12.2
\0.001
LDL (mg/dL) Women
122.6 ± 30.8
138.7 ± 31.5
150 ± 33.3
161.3 ± 38.9
\0.001
Men
128 ± 35.4
136.2 ± 29.9
145.7 ± 32.9
148.3 ± 38.9
\0.001
Triglycerides (mg/dL) Women
86.44 (64, 118)
93 (71, 124.26)
99.44 (77.6, 133.14)
110 (87, 143)
\0.001
Men
104.44 (69.06, 157.79)
108 (82, 149.82)
121 (87, 163.28)
125 (90, 175)
\0.001
3.2 %
2.8 %
5.2 %
Systolic dysfunction Women
123
6.1 %
0.4
Clin Res Cardiol (2014) 103:211–222
217
Table 2 continued
Men
hs-TnI \ 1.9 pg/mL (women/men N = 523/211)
1.9 pg/mL B hsTnI \ 3.5 pg/mL (women/ men N = 665/514)
3.5 pg/mL B hsTnI \ 5.2 pg/mL (women/ men N = 509/606)
hs-TnI [ 5.2 pg/mL (women/men N = 343/767)
8.2 %
4.6 %
7.2 %
12.2 %
p value adjusted for age p value 0.0021
Diastolic dysfunction Women
9.4 %
17.4 %
25.4 %
35.5 %
\0.001
Men
11.7 %
11.2 %
15.4 %
24.9 %
0.034
Fig. 1 Distribution of troponin I concentration assessed by a high-sensitivity assay in 4,139 subjects (2,099 men, 2,040 women) according to four categories of troponin concentration. The data was weighted for age of the region Mainz/Mainz-Bingen. The height of the bars represents the percentage of each gender in the different troponin I categories. Concentration of highsensitivity determined troponin I is given in pg/mL. The number below the brackets represent the different hs-TnI categories with category 1 below the LoD and the remaining categories defined using tertiles of the detectable hs-TnI values with the first third being C1.9 and \3.5 pg/mL, second third C3.5 and \5.2 pg/mL and the last third C5.2 pg/mL
correlated inverse with hs-TnI. This reflects cardiac structural and functional changes mirrored by increased NT-proBNP concentration, showing myocardial stretch as encountered in systolic and diastolic heart failure. Renal function is important to decrease troponin I concentration in the circulation, therefore, with decreasing renal function, troponin I concentration increases.
categories (p \ 0.001 for LVEDD in both genders and LVM with p \ 0.001 in men and p = 0.022 in women) as given in Table 1. Figure 3 and Supplemental Table 2 show the positive correlation of left ventricular mass and left ventricular end-diastolic diameter with troponin I concentration (for both genders p \ 0.001, b for men 0.166 and 0.127; b for women 0.084 and 0.079).
Association of troponin I with cardiac structure and function
Cardiac function
Cardiac structure Both left ventricular end-diastolic diameter and left ventricular mass show a positive correlation with hs-TnI
Concentrations of hs-TnI were associated with both systolic and diastolic function of the heart. An inverse correlation was present with LVEF in the results of the censored regression (men p \ 0.001, b = -0.104 and women p = 0.0013, b = -0.049) (Fig. 3 and Supplemental
123
218
Clin Res Cardiol (2014) 103:211–222
Fig. 2 Forest plots for cardiovascular risk factors. Regression coefficients (beta) per increase in one standard deviation are depicted in the forest plots for each gender with the corresponding p value. Each regression was adjusted for age
Table 2). In this context, patients with a lower LVEF had increased hs-TnI concentrations. In a similar fashion, troponin I concentrations were positively associated with tissue Doppler E/E0 (in both genders p \ 0.001, b in men 0.099 and b 0.057 in women) as well the Tei index (for both genders p \ 0.001, b in men 0.076 and 0.054 in women) (Fig. 3 and Supplemental Table 2). Overall, categorical hs-TnI was associated with the aggregate diagnosis of diastolic dysfunction (men p = 0.038 and women p \ 0.001). Association of troponin I with vascular structure and function Vascular function Endothelial dysfunction, as a correlate of vascular function measured by FMD, was weakly associated with troponin I concentrations in women, (Fig. 3 and Supplemental Table 2, p = 0.05). A similar trend was observed in men (p = 0.018). Vascular stiffness as detected by infrared plethysmography was not associated with categorical troponin I (Table 1, men p = 0.21, women p = 1.0).
123
Vascular structure Mean IMT of the carotid artery was correlated positively with troponin I concentrations (p \ 0.001, b = 0.061 in men and p = 0.013, b = 0.047 in women) (Fig. 3 and Supplemental Table 2).
Discussion Previous studies could show that the application of highsensitivity troponin assays optimize the early diagnosis of MIs [3, 4, 21, 34–36]. The availability of such high-sensitivity troponin assays does not only improve acute cardiac care but could also enable risk stratification by troponin assessment in individuals at high risk for cardiovascular events [13, 14, 17, 18, 37]. The evaluation of a new, robust high-sensitivity troponin I assay in 4,139 individuals of the population-based GHS proves on a population basis that troponin I can be detected in over 80 % of all individuals in this large cohort representing the general population of Germany. The range of
Clin Res Cardiol (2014) 103:211–222
219
Fig. 3 Forest plots for cardiovascular phenotypes. Regression coefficients (beta) per increase in one standard deviation are depicted in the forest plots for each gender with the corresponding p value. Each regression was adjusted for age
modern, high-sensitivity assays allow measurements at the LoD of 1.9 pg/mL. However, the LoD is assay-specific and changes with a different manufacturer [38]. The reference population for assay validation studies should ideally be based on demographic characteristics that mirror the European population and it should be remembered that gender and ethnicity are not comorbidities and should be taken into account. Therefore, the selected reference population will influence the measured spectrum of troponin concentration [38]. Most of the variables presented in this study, reflecting cardiovascular phenotypes, correlated with troponin I concentration in the general population and were differently distributed for men and women and influenced by age. The results underline the cardiac-specific nature of troponin I and the correlation to cardiac structure and function resulting in a higher risk of cardiovascular events as recently reported [19]. However, that troponin I also reflects changes in the vascular structure as reflected by correlation with increasing IMT may also be shown by an increased stroke risk as suggested in population studies [19]. In the vast majority of individuals, the detected troponin I levels are far below the cut-off (i.e., 99th percentile of the general population) and are used to detect myocardial ischemia. Most likely, those detectable low troponin concentrations cannot simply be considered as ‘‘background
noise’’; it rather reflects an aggregate measure of cardiacspecific pathophysiology [39]. It is known from population-based studies that hs-TnT correlates with cardiovascular risk factors, age and impaired renal function [13, 14]. Although similar correlations are observed for the hs-TnI concentrations detected in the present population-based cohort, our results further extent current knowledge and demonstrate that low troponin I concentrations are associated with precise measures of impaired cardiac structure and function in the general population. Diastolic dysfunction and systolic dysfunction as markers of cardiac function are associated with troponin I concentrations assayed with high-sensitivity tests, mirroring abnormalities of the cardiac structure like increased end-diastolic diameter of the LV or left ventricular mass [13, 14, 40]. Important are also contributions of chronic sources of myocardial injury to troponin release, baseline concentrations of hs-TnI, below the assay range of contemporary-sensitivity assays (99th percentile below the 10 % CV), strongly associate with prevalent heart failure as reported from studies performed with hs-TnT [14]. Structural vascular correlates like IMT were also connected to increased risk load of arterial hypertension and age contributing to enlarged IMT diameter and sub-atherosclerotic changes of the vasculature [29, 41]. Chronic elevated troponin I concentration should be interpreted as result of structural heart disease, like systolic
123
220
and diastolic heart failure depicting a higher left ventricular mass, lower LVEF and increased myocardial stress, as reflected by NT-proBNP concentration [13, 42, 43]. Overall, the additional information of hs-TnI might be used to determine those subjects in the general population at risk of developing clinically relevant changes in cardiovascular structure and function, which translates into cardiovascular outcome [7, 8, 13, 37]. From a clinical perspective, the application of hs-TnI for cardiovascular risk stratification and decision making in clinical routine is not ready yet. Although improving prediction metrics, it needs to be proven that the change in medical therapies, which is based on risk reclassification, lead to improved patient care [19]. This would constitute a milestone in personalized cardiovascular medicine based on biomarker assessment. Further aspects of cardiovascular disease have already been elucidated by the use of troponin measurement as troponin can be used to evaluate heart failure patients showing correlation with systolic and diastolic function together with NT-proBNP concentration [44]. Further, measurement of troponin in elderly individuals was useful to assess incident heart failure [14]. The lack of a strong association between hs-TnI and endothelial function may be explained by the cardiacspecific origin of troponin release and by the large variation of FMD measures which is inherent in the methods [29]. There is only limited data on the association of troponin I with vascular function in the literature. A previous study with patients suffering from end-stage renal disease has shown a positive correlation of contemporary-sensitivity troponin T with vascular structure measured by IMT indicating asymptomatic atherosclerotic changes [45]. Previous studies describing associations of low levels of cardiac troponins determined by high-sensitivity assays with cardiac and vascular structure and function were mostly conducted in diseased cohorts [12–14, 21] and targeting troponin T rather than troponin I. Our results confirm the data for troponin I as demonstrated with troponin T levels regarding structural variables derived by magnetic resonance imaging [13]. The use of troponin I as a biomarker for risk prediction was earlier suggested in a study employing a biomarker score of NT-proBNP, CRP and troponin I measured with a contemporary-sensitivity assay [46]. With the data from our study that analysis can be further augmented, stressing the importance of hs-TnI concentrations between LoD and the 99th percentile to identify subjects at risk for developing cardiovascular disease. Current study results show that risk assessment of circulating biomarkers over and above the multiple pre-existing cardiovascular risk factors is of considerable interest [19, 46].
123
Clin Res Cardiol (2014) 103:211–222
Limitations Limitations of our study are (a) that individuals in the GHS represent inhabitants of a medium-sized central European city. Therefore, the present data cannot be extrapolated to other populations. Although the age range reflects a broad proportion of the population, extrapolation of the results to older or younger individuals is not possible. In addition, (b) troponin I was measured in frozen samples. (c) Echocardiography and carotid ultrasound measurements were performed by certified medical assistants working according to standard operating procedures, however, intra- and interobserver comparisons were not included into the current manuscript. (d) Detection of troponin I concentration below the 10 % CV of 5.2 ng/mL is associated with a higher imprecision. (e) Every subject was informed not to perform exercise 8 h before the study visit as this may influence biomarker concentrations but biomarker concentration could be influenced by previous exercise. (f) Repeated measurements are not present currently as the 5 year follow-up is still ongoing. Categorizing subjects from the general population to aggregate disease diagnoses like diastolic or systolic heart failure may be influenced by low numbers. Follow-up was not complete during presentation of the results of this study. An association with the primary cardiovascular endpoint can, therefore, not be reported. Conclusion In conclusion, troponin I concentration determined with a high-sensitivity assay in the general population is associated with phenotypes of cardiovascular structure, function and vascular structure. Due to its cardiac-specific origin, an association with vascular function could not be shown. In this context, circulating troponin concentration is driven by left ventricular hypertrophy and systolic and diastolic dysfunction. Our results provide pathophysiological background for the role of hs-TnI in cardiovascular risk stratification in primary prevention settings. Acknowledgments The Gutenberg Health Study (GHS) is funded through the government of Rhineland-Palatinate (‘‘Stiftung Rheinland Pfalz fu¨r Innovation,’’ contract No. AZ 961–386261/733), the research programs ‘‘Wissen schafft Zukunft’’ and ‘‘Schwerpunkt Vaskula¨re Pra¨vention’’ of the Johannes Gutenberg-University of Mainz, and its contract with Boehringer Ingelheim and Philips Medical Systems, including an unrestricted grant for the GHS. Conflict of interest Abbott Diagnostics provided test reagents for high-sensitivity troponin I measurements. Stefan Blankenberg has received honoraria from Abbott Diagnostics, Siemens, Thermo Fisher, and Roche Diagnostics and is a consultant for Thermo Fisher. All other co-authors did not report any conflict of interest.
Clin Res Cardiol (2014) 103:211–222
221
References 1. Jaffe AS (2011) The 10 commandments of troponin, with special reference to high sensitivity assays. Heart 97(11):940–946. doi:10.1136/hrt.2009.185751 2. Mueller M, Celik S, Biener M, Vafaie M, Schwoebel K, Wollert KC, Januzzi JL, Katus HA, Giannitsis E (2012) Diagnostic and prognostic performance of a novel high-sensitivity cardiac troponin T assay compared to a contemporary sensitive cardiac troponin I assay in patients with acute coronary syndrome. Clin Res Cardiol 101(10):837–845. doi:10.1007/s00392-012-0469-6 3. Keller T, Zeller T, Peetz D, Tzikas S, Roth A, Czyz E, Bickel C, Baldus S, Warnholtz A, Frohlich M, Sinning CR, Eleftheriadis MS, Wild PS, Schnabel RB, Lubos E, Jachmann N, Genth-Zotz S, Post F, Nicaud V, Tiret L, Lackner KJ, Munzel TF, Blankenberg S (2009) Sensitive troponin I assay in early diagnosis of acute myocardial infarction. N Engl J Med 361(9):868–877. doi:10.1056/NEJMoa0903515 4. Reichlin T, Hochholzer W, Bassetti S, Steuer S, Stelzig C, Hartwiger S, Biedert S, Schaub N, Buerge C, Potocki M, Noveanu M, Breidthardt T, Twerenbold R, Winkler K, Bingisser R, Mueller C (2009) Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. N Engl J Med 361(9):858–867. doi:10. 1056/NEJMoa0900428 5. Kurz K, Giannitsis E, Becker M, Hess G, Zdunek D, Katus HA (2011) Comparison of the new high sensitive cardiac troponin T with myoglobin, h-FABP and cTnT for early identification of myocardial necrosis in the acute coronary syndrome. Clin Res Cardiol 100(3):209–215. doi:10.1007/s00392-010-0230-y 6. Celik S, Giannitsis E, Wollert KC, Schwobel K, Lossnitzer D, Hilbel T, Lehrke S, Zdunek D, Hess A, Januzzi JL, Katus HA (2011) Cardiac troponin T concentrations above the 99th percentile value as measured by a new high-sensitivity assay predict long-term prognosis in patients with acute coronary syndromes undergoing routine early invasive strategy. Clin Res Cardiol 100(12):1077–1085. doi:10.1007/s00392-011-0344-x 7. Apple FS, Ler R, Murakami MM (2012) Determination of 19 cardiac troponin I and T assay 99th percentile values from a common presumably healthy population. Clin Chem 58(11):1574–1581. doi:10.1373/clinchem.2012.192716 8. Apple FS, Steffen LM, Pearce LA, Murakami MM, Luepker RV (2012) Increased cardiac troponin I as measured by a high-sensitivity assay is associated with high odds of cardiovascular death: the Minnesota Heart Survey. Clin Chem 58(5):930–935. doi:10.1373/clinchem.2011.179176 9. Giannitsis E, Kehayova T, Vafaie M, Katus HA (2011) Combined testing of high-sensitivity troponin T and copeptin on presentation at prespecified cutoffs improves rapid rule-out of non-STsegment elevation myocardial infarction. Clin Chem 57(10):1452–1455. doi:10.1373/clinchem.2010.161265 10. Keller T, Zeller T, Ojeda F, Tzikas S, Lillpopp L, Sinning C, Wild P, Genth-Zotz S, Warnholtz A, Giannitsis E, Mockel M, Bickel C, Peetz D, Lackner K, Baldus S, Munzel T, Blankenberg S (2011) Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction. JAMA 306(24):2684–2693. doi:10.1001/jama.2011.1896 11. Saunders JT, Nambi V, de Lemos JA, Chambless LE, Virani SS, Boerwinkle E, Hoogeveen RC, Liu X, Astor BC, Mosley TH, Folsom AR, Heiss G, Coresh J, Ballantyne CM (2011) Cardiac troponin T measured by a highly sensitive assay predicts coronary heart disease, heart failure, and mortality in the atherosclerosis risk in communities study. Circulation 123(13):1367–1376. doi:10.1161/CIRCULATIONAHA.110.005264 12. Giannitsis E, Becker M, Kurz K, Hess G, Zdunek D, Katus HA (2010) High-sensitivity cardiac troponin T for early prediction of
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
evolving non-ST-segment elevation myocardial infarction in patients with suspected acute coronary syndrome and negative troponin results on admission. Clin Chem 56(4):642–650. doi:10. 1373/clinchem.2009.134460 de Lemos JA, Drazner MH, Omland T, Ayers CR, Khera A, Rohatgi A, Hashim I, Berry JD, Das SR, Morrow DA, McGuire DK (2010) Association of troponin T detected with a highly sensitive assay and cardiac structure and mortality risk in the general population. JAMA 304(22):2503–2512. doi:10.1001/ jama.2010.1768 deFilippi CR, de Lemos JA, Christenson RH, Gottdiener JS, Kop WJ, Zhan M, Seliger SL (2010) Association of serial measures of cardiac troponin T using a sensitive assay with incident heart failure and cardiovascular mortality in older adults. JAMA 304(22):2494–2502. doi:10.1001/jama.2010.1708 Venge P, James S, Jansson L, Lindahl B (2009) Clinical performance of two highly sensitive cardiac troponin I assays. Clin Chem 55(1):109–116. doi:10.1373/clinchem.2008.106500 Venge P, Johnston N, Lindahl B, James S (2009) Normal plasma levels of cardiac troponin I measured by the high-sensitivity cardiac troponin I access prototype assay and the impact on the diagnosis of myocardial ischemia. J Am Coll Cardiol 54(13):1165–1172. doi:10.1016/j.jacc.2009.05.051 Frankenstein L, Remppis A, Giannitis E, Frankenstein J, Hess G, Zdunek D, Doesch A, Zugck C, Katus HA (2011) Biological variation of high sensitive Troponin T in stable heart failure patients with ischemic or dilated cardiomyopathy. Clin Res Cardiol 100(8):633–640. doi:10.1007/s00392-011-0285-4 Eggers KM, Venge P, Lindahl B, Lind L (2013) Cardiac troponin I levels measured with a high-sensitive assay increase over time and are strong predictors of mortality in an elderly population. J Am Coll Cardiol 61(18):1906–1913. doi:10.1016/j.jacc.2012. 12.048 Zeller T, Tunstall-Pedoe H, Saarela O, Ojeda F, Schnabel RB, Tuovinen T, Woodward M, Struthers A, Hughes M, Kee F, Salomaa V, Kuulasmaa K, Blankenberg S (2013) High population prevalence of cardiac troponin I measured by a high-sensitivity assay and cardiovascular risk estimation: the MORGAM Biomarker Project Scottish Cohort. Eur Heart J. doi:10.1093/ eurheartj/eht406 Wang TJ, Wollert KC, Larson MG, Coglianese E, McCabe EL, Cheng S, Ho JE, Fradley MG, Ghorbani A, Xanthakis V, Kempf T, Benjamin EJ, Levy D, Vasan RS, Januzzi JL (2012) Prognostic utility of novel biomarkers of cardiovascular stress: the Framingham Heart Study. Circulation 126(13):1596–1604. doi:10. 1161/CIRCULATIONAHA.112.129437 Eggers KM, Jaffe AS, Lind L, Venge P, Lindahl B (2009) Value of cardiac troponin I cutoff concentrations below the 99th percentile for clinical decision-making. Clin Chem 55(1):85–92. doi:10.1373/clinchem.2007.101683 Wild PS, Sinning CR, Roth A, Wilde S, Schnabel RB, Lubos E, Zeller T, Keller T, Lackner KJ, Blettner M, Vasan RS, Munzel T, Blankenberg S (2010) Distribution and categorization of left ventricular measurements in the general population: results from the population-based Gutenberg Heart Study. Circ Cardiovasc Imaging 3(5):604–613. doi:10.1161/CIRCIMAGING.109. 911933 Zeller T, Wild P, Szymczak S, Rotival M, Schillert A, Castagne R, Maouche S, Germain M, Lackner K, Rossmann H, Eleftheriadis M, Sinning CR, Schnabel RB, Lubos E, Mennerich D, Rust W, Perret C, Proust C, Nicaud V, Loscalzo J, Hubner N, Tregouet D, Munzel T, Ziegler A, Tiret L, Blankenberg S, Cambien F (2010) Genetics and beyond—the transcriptome of human monocytes and disease susceptibility. PLoS ONE 5(5):e10693. doi:10.1371/journal.pone.0010693
123
222 24. Wild PS, Zeller T, Beutel M, Blettner M, Dugi KA, Lackner KJ, Pfeiffer N, Munzel T, Blankenberg S (2012) The Gutenberg Health Study. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 55(6–7):824–829. doi:10.1007/s00103-0121502-7 25. Lean ME, Han TS, Morrison CE (1995) Waist circumference as a measure for indicating need for weight management. BMJ 311(6998):158–161 26. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J (2009) A new equation to estimate glomerular filtration rate. Ann Intern Med 150(9):604–612 150/9/604 [pii] 27. Matsushita K, Mahmoodi BK, Woodward M, Emberson JR, Jafar TH, Jee SH, Polkinghorne KR, Shankar A, Smith DH, Tonelli M, Warnock DG, Wen CP, Coresh J, Gansevoort RT, Hemmelgarn BR, Levey AS (2012) Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate. JAMA 307(18):1941–1951. doi:10. 1001/jama.2012.3954 28. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N (1986) Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 57(6):450–458 0002-9149(86)90771-X [pii] 29. Schnabel RB, Schulz A, Wild PS, Sinning CR, Wilde S, Eleftheriadis M, Herkenhoff S, Zeller T, Lubos E, Lackner KJ, Warnholtz A, Gori T, Blankenberg S, Munzel T (2011) Noninvasive vascular function measurement in the community: crosssectional relations and comparison of methods. Circ Cardiovasc Imaging 4(4):371–380. doi:10.1161/CIRCIMAGING.110.961557 30. Dixon WJ (1953) Processing data for outliers. Biometrics 9:74–89 31. Tobin J (1958) Estimation of relationships for limited dependent variables. Econometrica 26:24–36 32. Amemiya T (1984) Tobit models: a survey. J Econom 24:3–61 33. Team RDC (2011) R: a language and environment for statistical computing. R Foundation for Statistical Computing. http://www. R-project.org 34. Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA (2010) Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 56(2):254–261. doi:10.1373/ clinchem.2009.132654 35. Weber M, Bazzino O, Navarro Estrada JL, de Miguel R, Salzberg S, Fuselli JJ, Liebetrau C, Woelken M, Moellmann H, Nef H, Hamm C (2011) Improved diagnostic and prognostic performance of a new high-sensitive troponin T assay in patients with acute coronary syndrome. Am Heart J 162(1):81–88. doi:10. 1016/j.ahj.2011.04.007 36. Apple FS, Smith SW, Pearce LA, Ler R, Murakami MM (2008) Use of the Centaur TnI-Ultra assay for detection of myocardial infarction and adverse events in patients presenting with symptoms suggestive of acute coronary syndrome. Clin Chem 54(4):723–728. doi:10.1373/clinchem.2007.097162 37. Otsuka T, Kawada T, Ibuki C, Seino Y (2010) Association between high-sensitivity cardiac troponin T levels and the
123
Clin Res Cardiol (2014) 103:211–222
38.
39.
40.
41.
42.
43.
44.
45.
46.
predicted cardiovascular risk in middle-aged men without overt cardiovascular disease. Am Heart J 159(6):972–978. doi:10.1016/ j.ahj.2010.02.036 S0002-8703(10)00245-0 [pii] Korley FK, Jaffe AS (2013) Preparing the United States for highsensitivity cardiac troponin assays. J Am Coll Cardiol. doi:10. 1016/j.jacc.2012.09.069 Providencia R, Barra S, Paiva L (2013) Atrial fibrillation, elevated troponin, ischemic stroke and adverse outcomes: understanding the connection. Clin Res Cardiol 102(10):701–711. doi:10.1007/s00392-013-0591-0 Dinh W, Nickl W, Futh R, Lankisch M, Hess G, Zdunek D, Scheffold T, Barroso MC, Tiroch K, Ziegler D, Seyfarth M (2011) High sensitive troponin T and heart fatty acid binding protein: novel biomarker in heart failure with normal ejection fraction: a cross-sectional study. BMC Cardiovasc Disord 11:41. doi:10.1186/1471-2261-11-41 Flu WJ, van Kuijk JP, Hoeks SE, Kuiper R, Schouten O, Goei D, Winkel T, van Gestel YR, Verhagen HJ, Bax JJ, Poldermans D (2009) Intima media thickness of the common carotid artery in vascular surgery patients: a predictor of postoperative cardiovascular events. Am Heart J 158(2):202–208. doi:10.1016/j.ahj. 2009.05.028 Nadir MA, Rekhraj S, Wei L, Lim TK, Davidson J, MacDonald TM, Lang CC, Dow E, Struthers AD (2012) Improving the primary prevention of cardiovascular events by using biomarkers to identify individuals with silent heart disease. J Am Coll Cardiol 60(11):960–968. doi:10.1016/j.jacc.2012.04.049 Schnabel RB, Schulz A, Messow CM, Lubos E, Wild PS, Zeller T, Sinning CR, Rupprecht HJ, Bickel C, Peetz D, Cambien F, Kempf T, Wollert KC, Benjamin EJ, Lackner KJ, Munzel TF, Tiret L, Vasan RS, Blankenberg S (2010) Multiple marker approach to risk stratification in patients with stable coronary artery disease. Eur Heart J 31(24):3024–3031. doi:10.1093/ eurheartj/ehq322 Kusumoto A, Miyata M, Kubozono T, Ikeda Y, Shinsato T, Kuwahata S, Fujita S, Takasaki K, Yuasa T, Hamasaki S, Tei C (2012) Highly sensitive cardiac troponin T in heart failure: comparison with echocardiographic parameters and natriuretic peptides. J Cardiol 59(2):202–208. doi:10.1016/j.jjcc.2011.11. 012 Hojs R, Ekart R, Hojs Fabjan T, Balon BP, Gorenjak M (2005) Cardiac troponin T (cTnT) in hemodialysis patients with asymptomatic and symptomatic atherosclerosis. Arch Med Res 36(4):367–371. doi:10.1016/j.arcmed.2005.03.024 Blankenberg S, Zeller T, Saarela O, Havulinna AS, Kee F, Tunstall-Pedoe H, Kuulasmaa K, Yarnell J, Schnabel RB, Wild PS, Munzel TF, Lackner KJ, Tiret L, Evans A, Salomaa V (2010) Contribution of 30 biomarkers to 10-year cardiovascular risk estimation in 2 population cohorts: the MONICA, risk, genetics, archiving, and monograph (MORGAM) biomarker project. Circulation 121(22):2388–2397. doi:10.1161/CIRCULATIONAHA. 109.901413
