Key words
High-sensitivity cardiac troponin I - type 2 diabetes mellitus - intima-media thickness,
cardiovascular disease
Introduction
Type 2 diabetes mellitus (T2DM) promotes the development of atherosclerosis thereby
increasing the risk for cardiovascular disease (CVD), the leading cause of morbidity
and mortality in T2DM [1]
[2]. Compared with persons without diabetes, patients with T2DM have a twofold higher
risk developing CVD and diabetes-related macrovascular complications. This atherogenic
risk is known to be markedly larger in men compared to women [3]
[4]. Nowadays, the combination of clinical markers with biomarkers allows the identification
of individuals at increased CVD. This can enable appropriate treatment tailored to
patients with an already existing macrovascular disease, such as coronary heart disease.
Cardiac-specific troponins (cTn) are powerful biomarkers to detect subclinical myocardial
damage and revolutionized treatment in patients with acute coronary syndrome (ACS)
[5]
[6]
[7]. Subclinical myocardial damage leads to release of cardiac troponin I (cTnI) and
T (cTnT) that are detectable from the bloodstream. The development and introduction
of high-sensitivity (hs) assays for the determination of cTn not only fundamentally
improved the diagnostic performance and clinical utility, but also allowed the quantification
of very low cTn concentrations seen in asymptomatic individuals and the detection
of subclinical myocardial damage in individuals without overt acute cardiac disease
[8]
[9]. The term high-sensitivity is defined as the ability to measure troponin concentrations
in>50% of a healthy reference population with a coefficient of variation<10% at the
99th percentile [10]
[11]
[12]. Besides the tremendous role in the management of patients with ACS, recent studies
demonstrated prognostic implications of hs-cTn determination in individuals without
acute cardiac morbidities [13]
[14]
[15]
[16]: Hs-cTn is recognized as predictor of all-cause mortality and cardiovascular death
in the general population. Currently, there are three hs-cTn assays available. Roche
hs-cTnT assay (Roche Diagnostics, Mannheim, Germany) and Abbott hs-cTnI assay (Abbott,
Chicago, United States) have been used for several years. Numerous studies focused
on diagnostic performance and clinical utility of hs-cTn determination in patients
with ACS and in the general population [13]
[14]
[15]
[16]
[17]
[18]. Recently, a novel hs-cTnI assay by Siemens (Siemens Healthineers, Eschborn, Germany),
following referred to as ADVIA Centaur hs-cTnI, has become available. In a study by
Boeddinghaus and colleagues, this novel hs-cTnI assay showed high diagnostic accuracy
and clinical performance in diagnosing ACS and a comparable performance to the aforementioned
hs-cTn assays [8].
Patients with diabetes mellitus show elevated plasma cTn concentrations and various
diabetes associated factors contributing to the increased CVD risk including hyperglycemia,
dyslipidemia and alterations in the hemostasis system [19]
[20]
[21]
[22]. A useful tool to assess early forms of atherosclerosis in clinical practice is
the determination of carotid intima-media thickness (cIMT) by high-resolution ultrasound.
cIMT increases with age and is known to be larger in patients with T2DM compared to
healthy controls [23]
[24]. While cIMT is a risk marker for future CVD events in the general population [24], cIMT progression was not associated with future cardiovascular risk in patients
with diabetes [25].
We now addressed the contribution of hs-cTnI, measured by the novel ADVIA Centaur
hs-cTnI assay, to cardiovascular risk which was assessed by cIMT in T2DM. We furthermore
compared this to classical clinical CVD risk factors.
Methods
Cohort
234 patients with T2DM were included in the analysis. 100 of the patients were women
(43%) and 134 men (58%) with a median age of 65 years (57–71) and the median duration
of diabetes mellitus was 10 years (6–17). Anthropometric data, clinical characteristics
and laboratory parameters are shown in [Table 1]. The study was conducted at the Department of Internal Medicine of the University
Hospital Tübingen. Data were obtained from inpatients taking part at a diabetes mellitus
education program according to recent guidelines [26]. Patients with type 1 diabetes mellitus and those who had a record of diabetes secondary
to other causes (e. g. genetic defects of beta cell function, monogenic forms of diabetes,
diabetes mellitus due to other pancreatic diseases) were excluded from the study.
A detailed history including demographics and time since diagnosis of T2DM was taken
from all subjects, followed by routine laboratory testing and physical examination
including anthropometric data acquisition such as body mass index. In addition, patients
were asked for history or presence of macrovascular disease (coronary artery disease,
peripheral vascular disease, cerebrovascular disease). Blood pressure was measured
according to Riva-Rocci by auscultation after resting for 10 min in sitting position.
The study was conducted in accordance with the ethical standards of the Declaration
of Helsinki from 1964 and its later amendments. Written informed consent was obtained
from all study participants. The local ethics committee approved the protocol.
Table 1 Patients’ clinical and laboratory characteristics (n=234).
|
Variable
|
Total cohort (n=234)
|
Men (n=134)
|
Women (n=100)
|
p-value men vs. women
|
|
Age, years
|
65 (57–71)
|
65 (57–71)
|
64 (56–71)
|
0.9566
|
|
Body-Mass-Index, kg/m2
|
32.8 (28.4–39.0)
|
33.3 (29.0–38.8)
|
31.4 (27.0–39.3)
|
0.1070
|
|
Time since diabetes diagnosis, years
|
10 (6–17)
|
11 (6–16)
|
9 (5–18)
|
0.7619
|
|
Systolic blood pressure, mmHg
|
130 (122–140)
|
130 (120–140)
|
130 (125–140)
|
0.1546
|
|
Diastolic blood pressure, mmHg
|
80 (75–80)
|
80 (75–80)
|
80 (75–82)
|
0.3778
|
|
Glucose (fasting), mmol/l
|
8.6 (7.2–10.6)
|
8.6 (7.2–11.0)
|
8.4 (6.9–10.0)
|
0.1626
|
|
Glycated haemoglobin (HbA1c), %
|
7.6 (6.9–9.1)
|
7.7 (6.9–9.2)
|
7.6 (6.8–8.9)
|
0.3062
|
|
Glycated haemoglobin (HbA1c), mmol/mol
|
59.6 (51.9–76.0)
|
60.7 (51.9–76.0)
|
59.6 (50.8–73.8)
|
|
|
Creatinine, mg/dl
|
0.8 (0.7–1.0)
|
0.9 (0.8–1.2)
|
0.7 (0.6–0.9)
|
<0.0001
|
|
eGFR (MDRD), ml/min/1,73 m²
|
84 (61–100)
|
84 (60–99)
|
83 (62–102)
|
0.8781
|
|
Triglycerides, mmol/l
|
1.9 (1.4–2.4)
|
1.9 (1.3–2.2)
|
1.9 (1.4–2.7)
|
0.5163
|
|
Total cholesterol, mmol/l
|
4.6 (4.0–5.3)
|
4.5 (3.8–5.2)
|
4.8 (4.1–5.5)
|
0.0048
|
|
HDL-Cholesterol, mmol/l
|
1.1 (0.9–1.3)
|
1.1 (0.9–1.3)
|
1.2 (1.0–1.4)
|
0.0029
|
|
LDL-Cholesterol, mmol/l
|
3.0 (2.4–3.9)
|
3.0 (2.3–3.8)
|
3.0 (2.6–4.1)
|
0.0523
|
|
Apolipoprotein AI, mg/dl
|
149±22
|
144±20
|
154±22
|
0.0002
|
|
Apolipoprotein B, mg/dl
|
95 (80–112)
|
93 (77–110)
|
99 (82–118)
|
0.0379
|
|
Apolipoprotein B/Apolipoprotein AI
|
0.6 (0.5–0.8)
|
0.6 (0.5–0.8)
|
0.6 (0.5–0.8)
|
0.9065
|
|
C-reactive Protein, mg/dl
|
0.33 (0.09–0.67)
|
0.23 (0.07–0.62)
|
0.45 (0.16–0.91)
|
0.0030
|
|
Macroangiopathy#, % (n)
|
31 (74)
|
41 (56)
|
18 (18)
|
<0.0001
|
|
Smoking, % (n)
|
15 (35)
|
16 (21)
|
14 (14)
|
0.7405
|
|
Carotid intima-media thickness, mm
|
0.75 (0.70–0.95)
|
0.80 (0.70–0.95)
|
0.73 (0.65–0.85)
|
0.0021
|
|
High-sensitivity cardiac troponin I*, ng/l
|
4.0 (2.0–10.0)
|
5.0 (3.0–12.0)
|
3.0 (2.0–6.8)
|
0.0005
|
#Macroangiopathy: evidence for coronary artery syndrome, cerebrovascular disease or
peripheral arterial disease. *measurable in 226 patients. eGFR, estimated glomerular
filtration rate; HDL, high density lipoprotein; LDL, low density lipoprotein.
Laboratory assessment and clinical parameters
Determination of hs-cTnI was performed on an ADVIA Centaur XPT Immunoassay System
(Siemens Healthineers, Eschborn, Germany) using re-thawed lithium-heparinized plasma
samples. The limit of detection is indicated according to the manufacturer as 2.21 ng/l,
the limit of blank as 0.90 ng/l and the 99th percentile concentrations are reported as 57 and 37 ng/l for men and women, respectively.
All other clinical chemistry analyses (i. e. creatinine [enzymatic method], glucose
[hexokinase method], total cholesterol, high-density lipoprotein-cholesterol [HDL-cholesterol],
low-density lipoprotein-cholesterol [LDL-cholesterol], triglycerides and C-reactive
protein) were measured on an ADVIA Chemistry XPT analyzer following manufacturer’s
instructions, except for apolipoprotein AI and apolipoprotein B, which were measured
on a BN ProSpec System (Siemens Healthineers). All laboratory parameters were measured
after an overnight fasting period. Glycated hemoglobin (HbA1c) was measured using
EDTA whole blood on a Tosoh G8 HPLC Analyzer (Tosoh Bioscience, Griesheim, Germany).
Estimated glomerular filtration rate (eGFR) was calculated using a modified MDRD (Modification
of Diet in Renal Disease)-formula based on plasma creatinine concentration, gender
and age [27].
Measurement of carotid intima-media thickness
The cIMT of the common carotid artery was measured using high-resolution ultrasound
with a 10–13 MHz-linear array transducer (AU5 Harmonic, Esaote Biomedica, Munich,
Germany). Ultrasound of both left and right common carotid artery was performed according
to the European Mannheim carotid intima-media thickness consensus [28]. In detail, B-mode cIMT was assessed on the far wall of the artery approximately
10 mm proximal to the carotid bulb as described before [29].
Statistical analysis
Continuous data were presented as mean±standard deviation for normally distributed
variables and as median (lower quartile - upper quartile) for non-normally distributed
variables. Normal distribution was tested for all continuous variables using the Shapiro-Wilk-W
test. Non-normal distributed data were transformed with a logarithmic function prior
to statistical analyses. The non-parametric rank sum test was used to compare two
groups. The association of hs-cTnI results with clinical and laboratory variables
were analysed using simple and multiple linear regression analysis. A p-value<0.05
was considered to indicate significant difference. For the selection of variables
in the multiple regression analysis, stepwise linear regression (forward selection)
was performed. Statistical analyses were conducted with JMP 14 (SAS Institute, Cary,
United States). [Figure 1] was created using GraphPad Software (version 8; GraphPad Software Inc., San Diego,
United States).
Fig. 1 Association of high-sensitivity cardiac troponin I (hs-cTnI) with carotid intima
media thickness in type 2 diabetes. Shown are results of linear regression and correlation
analyses separated by gender. A vertical line indicates the gender-specific 99th percentile cut-off concentration for hs-cTnI.
Results
High-sensitivity cardiac troponin I in patients with type 2 diabetes mellitus
Plasma hs-cTnI concentrations were measurable in 226 patients (97%) with a median
hs-cTnI concentration of 4.0 ng/l (2.0–10.0) as presented in [Table 1]. Among all T2DM patients, 9 individuals (4%) had hs-cTnI concentrations above the
99th gender-specific percentile. 217 individuals (93%) had concentrations below these
thresholds and 72 individuals (31%) showed concentrations below the limit of detection
(2.21 ng/ml) of which 8 individuals (3%) had concentrations below the limit of blank
(0.90 ng/l). Median concentration of hs-cTnI was higher in men compared to women (5.0
vs. 3.0 ng/l; p=0.0005). Total cholesterol (4.5 vs. 4.8 mmol/l; p=0.0048) and HDL-cholesterol
(1.1 vs. 1.2 mmol/l; p=0.0025) concentrations were lower in men compared to women.
In addition, mean concentration of apolipoprotein AI (154 vs. 144 mg/dl; p=0.0002),
median concentrations of apolipoprotein B (99 vs. 93 mg/dl; p=0.0379) and C-reactive
protein (0.45 vs. 0.23 mg/dl; p=0.0030) were higher in women than in men. Although
median creatinine concentration (0.9 vs. 0.7 mg/dl; p<0.0001) was significantly higher
in men compared to women, the calculated eGFR (MDRD) did not differ between genders
(p=0.8781).
137 patients (59%) used lipid-lowering drugs. Metformin was taken by 156 patients
(67%) and 137 patients (59%) were on insulin therapy. 35 patients (15%) reported to
be smokers and 75 patients (35%) had a history of smoking. Median cIMT was 0.75 mm,
ranging from 0.70–0.95 mm with higher cIMT in men compared to women (0.80 vs. 0.73 mm;
p=0.0021). Macrovascular disease was present in 56 of the men (41%) and in 18 of the
women (18%) with T2DM, with a significant difference comparing the presence of macrovascular
disease by gender (p<0.0001).
Associations between hs-cTnI concentrations and markers of cardiovascular risk and
vascular damage
For the understanding of the association of hs-cTnI concentrations with cIMT and traditional
cardiovascular risk factors, we performed simple linear regression analyses ([Table 2]). In the entire cohort, hs-cTnI was significantly associated with age (effect size
ßst=0.25; p=0.0001), gender (ßst=−0.24; p=0.0003), macrovascular disease (ßst=−0.23; p=0.0005), eGFR (ßst=−0.29; p<0.0001) and cIMT (ßst=0.27; p<0.0001). As cIMT interacted with gender on hs-cTnI (p=0.0256), subsequent
analyses were performed separately for both genders. In men, significant associations
of hs-cTnI with age (ßst=0.29; p=0.0009), HDL-cholesterol (ßst=0.18; p=0.0452), eGFR (ßst=−0.39; p<0.0001) and cIMT (ßst=0.34; p<0.0001) were detected in univariate analysis. In women, only age was significantly
associated with hs-cTnI (ßst=0.24; p=0.0178). CIMT and eGFR showed no significant association with hs-cTnI in
women ([Table 2] and [Figure 1]). To identify independent predictors of hs-cTnI multiple linear regression analyses
were performed. Therefore, stepwise regression analysis was utilized to identify variables
for the multivariable regression model ([Table 3]). In the final analyses only gender (ßst=−0.22; p=0.0007), eGFR (ßst=−0.20; p=0.0028) and C-reactive protein (ßst=0.14; p=0.0234) were significantly associated with hs-cTnI in the entire cohort.
Analysis separated by gender revealed eGFR (ßst=−0.24; p=0.0126) and cIMT (ßst=0.24; p=0.0081) to be independently associated with hs-cTnI in men ([Table 3]). Adjusting for the same covariates in women, age (ßst=0.24; p=0.0280) remained as a single factor significantly associated with hs-cTnI.
Finally, the influence of traditional cardiovascular risk factors, such as body mass
index, systolic blood pressure, LDL-cholesterol and smoking on the final model was
tested. CIMT (ßst=0.25; p=0.0067) and eGFR (ßst=−0.23; p=0.0169) still remained significantly associated with hs-cTnI in men and,
again, only the association between hs-cTnI and age in women (ßst=0.27; p=0.0234).
Table 2 Predictors of high-sensitivity cardiac troponin I in patients with type 2 diabetes.
Shown are effect sizes (ßst) and p-values of univariate relationships between high-sensitivity cardiac troponin
I and anthropometric or laboratory parameters in the entire cohort and separated by
gender.
|
Variable
|
High-sensitivity cardiac troponin I
|
|
Total cohort (n=226)
|
Men (n=130)
|
Women (n=96)
|
|
Gender (female=0, male=1)
|
−0.24
p=0.0003
|
-
|
-
|
|
Age, years
|
0.25
p=0.0001
|
0.29
p=0.0009
|
0.24
p=0.0178
|
|
Body-Mass-Index, kg/m2
|
−0.01
p=0.8925
|
−0.12
p=0.1831
|
0.09
p=0.4050
|
|
Time since diabetes diagnosis, years
|
0.06
p=0.3872
|
0.05
p=0.5381
|
0.07
p=0.4736
|
|
Systolic blood pressure, mmHg
|
0.03
p=0.6275
|
0.02
p=0.7816
|
0.10
p=0.3372
|
|
Diastolic blood pressure, mmHg
|
−0.07
p=0.3192
|
−0.07
p=0.4054
|
−0.01
p=0.9066
|
|
Glucose (fasting), mmol/l
|
−0.05
p=0.4953
|
−0.05
p=0.5508
|
−0.08
p=0.4375
|
|
Glycated haemoglobin, %
|
−0.09
p=0.1980
|
−0.09
p=0.2991
|
−0.12
p=0.2399
|
|
eGFR (MDRD), ml/min/1,73 m²
|
−0.29
p<0.0001
|
−0.39
p<0.0001
|
−0.18
p=0.0735
|
|
Triglycerides, mmol/l
|
0.01
p=0.9988
|
−0.02
p=0.8433
|
0.04
p=0.6832
|
|
Total cholesterol, mmol/l
|
−0.03
p=0.6199
|
0.02
p=0.8621
|
−0.01
p=0.8952
|
|
HDL-Cholesterol, mmol/l
|
0.03
p=0.6137
|
0.18
p=0.0452
|
−0.03
p=0.7792
|
|
LDL-Cholesterol, mmol/l
|
−0.02
p=0.7448
|
0.02
p=0.8377
|
−0.02
p=0.8672
|
|
Macrovascular disease (yes=1)
|
−0.23
p=0.0005
|
−0.24
p=0.0050
|
−0.07
p=0.4993
|
|
Smoking (yes=1)
|
0.09
p=0.6156
|
−0.07
p=0.6959
|
0.17
p=0.7896
|
|
C-reactive Protein, mg/dl
|
0.09
p=0.1561
|
0.17
p=0.0531
|
0.08
p=0.4329
|
|
Carotid intima-media thickness, mm
|
0.27
p<0.0001
|
0.34
p<0.0001
|
0.05
p=0.6142
|
Table 3 Multiple linear regression analysis of associations with high-sensitivity cardiac
troponin I in patients with type 2 diabetes. Shown are effect sizes (ßSt) and p-values of multivariable regression analyses with all indicated variables.
|
Variable
|
High-sensitivity cardiac troponin I
|
|
Total cohort (n=226)
|
Men (n=130)
|
Women (n=96)
|
|
Gender (female=0, male=1)
|
−0.22
0.0007
|
-
|
-
|
|
Age, years
|
0.11
0.1064
|
0.04
0.7094
|
0.24
0.0280
|
|
Glycated haemoglobin, mmol/mol
|
−0.12
0.0619
|
−0.10
0.2384
|
−0.18
0.0927
|
|
eGFR, ml/min/1.73 m²
|
−0.20
0.0028
|
−0.24
0.0126
|
−0.14
0.1764
|
|
Macrovascular disease (yes=1)
|
−0.09
0.1957
|
−0.07
0.4142
|
−0.06
0.5490
|
|
C-reactive Protein
|
0.14
0.0234
|
0.12
0.1317
|
0.17
0.1176
|
|
Carotid intima-media thickness, mm
|
0.12
0.0837
|
0.24
0.0081
|
−0.09
0.3850
|
Discussion
We here investigated links between high-sensitivity cardiac troponin I (hs-cTnI) and
cardiovascular risk factors and early signs of arteriosclerotic disease in patients
with type 2 diabetes. Our results revealed important gender-specific differences:
Men showed higher cIMT compared to women as shown before [29]
[30]. In men, cIMT was significantly associated with elevated hs-cTnI concentrations,
a marker of subclinical myocardial damage, independently of traditional CVD risk factors,
such as age, body mass index, smoking and LDL-cholesterol. The association between
subclinical myocardial damage and cIMT was independent of the presence of macrovascular
disease. Interestingly, women did not show the association between cIMT and hs-cTnI,
whereby cIMT gender differences, to the detriment of men, has shown early before [31]. Age was the single independently associated risk factor of hs-cTnI in women with
T2DM. This is in line with several other studies reporting age as the main risk factor
of cardiovascular death in women [32]
[33]
[34]. A previous study investigated the association of Abbott hs-cTnI and an additional
marker of vascular damage by using brachial-arterial pulse wave velocity [35]. A significant association between hs-cTnI and brachial-arterial pulse wave velocity
(ba-PWV), a marker of arterial stiffness, was demonstrated, but not with cIMT. Unfortunately,
the study did not report gender-specific associations. Nevertheless, their results
are in line with our findings in the entire cohort.
T2DM itself is associated with an increased cIMT and gender differences of cIMT are
also well documented [23]
[36]
[37]
[38]. Consequently, gender-specific risk stratification in T2DM should be performed,
which may have important implications in prognosis and potentially the treatment of
these patients.
Further interesting findings were observed in our study: the majority of T2DM patients
in our study showed measurable hs-cTnI concentrations below the gender-specific 99th percentile. Hs-cTn assays are frequently used today and provide results with strongly
increased diagnostic accuracy even at very low troponin concentrations within the
reference interval [8]. Several other studies consistently showed that elevated troponin plasma concentrations
are present in patients with T2DM patients compared to non-diabetics either using
older troponin assays or hs-cTn assays [39]
[40]. It was shown that even slightly increased hs-cTn concentrations, still within the
reference interval, are associated with subclinical myocardial damage and are therefore
predictive for future cardiovascular events in the general population and in patients
attending an emergency room without obvious cardiac diseases [13]
[14]
[16]. This link emphasizes the need to include hs-cTn assays in the risk stratification
of patients with T2DM to identify individuals at increased risk for cardiovascular
events.
We found increased hs-cTnI concentrations in men compared to women as it was previously
demonstrated with the introduction of the first hs-cTn assays gender-specific differences
of hs-cTn concentrations [41]
[42]. In a recent study, Boeddinghaus and colleagues confirmed gender-specific differences
determined by the same hs-cTnI assay we used in our study in patients presenting with
suspicion of acute coronary syndrome to the emergency unit [8]. We observed nearly 2-fold higher hs-cTnI in men as it was consistently reported
by the aforementioned studies. It is now widely accepted that troponin concentrations,
in line with other biomarkers, are influenced by gender and that gender-specific 99th percentile must be taken into account [5]
[11].
Renal function, determined by eGFR (MDRD), was found to be another parameter, which
is associated with hs-cTnI in our study. And again, this association was only true
in men. Similar findings were reported in previous publications showing significant
association between renal function and hs-cTn concentrations in patients admitting
the emergency room [43]
[44]. Studies using hs-cTnT and hs-cTnI assays in parallel showed stronger correlation
of renal function to hs-cTnT compared to hs-cTnI [43]
[44]. Unfortunately, we did not measure hs-cTnT in our study to focus on this specific
issue. In our study, the majority of T2DM patients had an eGFR within the reference
interval. It is therefore not possible to make detailed statements to the association
of hs-cTnI and chronic kidney disease, especially, since we do not have an additional
marker for early nephropathy such as albumin in urine measurement. However, there
is evidence that hs-cTnI is not increased in patients with dialysis-dependent renal
failure. In a recent study it was demonstrated that hs-cTnI concentrations were below
the 99th percentile in the majority of these patients [45].
Strength of our study is the observation in a real-world clinical setting in a well
characterized cohort of patients with T2DM. We assessed subclinical myocardial damage
using a novel hs-cTnI assay and investigated the association to cIMT, an established
marker of atherosclerosis. Limitations of the study include the cross-sectional design
and the lack of detailed data about cardiac parameters (i. e. cardiac function). Regarding
cIMT as an early marker of subclinical atherosclerosis, a point of limitation is that
we used this tool in a cohort including patients with known macrovascular disease.
Nevertheless, we are looking at cIMT as a continuous variable and an increment of
cIMT is associated with an elevated CVD risk [24]. We did not examine more distal segments of the carotid artery (i. e. the internal
carotid artery) or plaque burden. Latter is known to be an important marker of advanced
atherosclerosis [46]. Regarding laboratory parameters, the MDRD formula for estimating GFR used in this
study is not accurate for values within the normal range and results should therefore
be carefully interpreted. Finally, results of the study need to be validated in larger
cohorts, ideally combined with follow-up measurements to investigate the prognostic
utility of the association between hs-cTnI and cIMT.
In conclusion, the present study in a real-world clinical setting demonstrated that
the majority of patients with T2DM showed measurable hs-cTnI concentrations using
the novel ADIVA centaur hs-cTnI assay. CIMT was significantly associated with hs-cTnI
in men and is therefore suitable as predictor of subclinical myocardial damage in
men, but not in women.
