Epidemiology/Population Science

(1)

Multiple Endothelial Biomarkers and Noninvasive Vascular Function in the General Population

The Gutenberg Health Study

Renate B. Schnabel, Philipp S. Wild, Andreas Schulz, Tanja Zeller, Christoph R. Sinning, Sandra Wilde, Jan Kunde, Edith Lubos, Karl J. Lackner, Ascan Warnholtz, Tommaso Gori,

Stefan Blankenberg, Thomas Munzel, for the Gutenberg Health Study Investigators

Abstract—Vascular reactivity is reflected by blood biomarkers and noninvasive vascular function measurement. The relation of biomarkers to flow-mediated dilation and peripheral arterial tonometry in the general population is little understood. In 5000 individuals (mean age, 56 ⫾11 years; age range, 35–74 years; 49% women) of the population-based Gutenberg Health Study we simultaneously assessed 6 biomarkers of cardiovascular function (midregional proadre- nomedullin [MR-proADM], midregional pro atrial natriuretic peptide [MR-proANP], N-terminal pro B-type natriuretic peptide, copeptin, C-terminal proendothelin 1, and neopterin) in relation to flow-mediated dilation and peripheral arterial tonometry. Strongest partial correlations (adjusted for age and sex) were observed for baseline pulse amplitude with MR-proADM (r ⫽0.13) and MR-proANP (r⫽⫺0.13); hyperemic response variables showed the highest correlation for MR-proADM and peripheral arterial tonometry ratio (r ⫽⫺0.14). In multivariable linear regression models, strongest associations with baseline vascular function were observed for MR-proANP with baseline pulse amplitude ( ␤ per SD increase [99.17%], ⫺0.080 [⫺0.115 to ⫺0.044]; P⬍0.0001 after Bonferroni correction for multiple testing) and MR-proADM ( ⫺0.044 [⫺0.070 to ⫺0.017]; P⬍0.0001), as well as MR-proANP (⫺0.033 [⫺0.057 to ⫺0.009];

P ⫽0.0017) and N-terminal pro B-type natriuretic peptide (⫺0.027 [⫺0.051 to ⫺0.003]; P⫽0.015) with brachial artery diameter. For hyperemic response variables, highest associations were seen for peripheral arterial tonometry ratio with MR-proADM ( ⫺0.022 [⫺0.043 to ⫺0.004]; P⫽0.043), MR-proANP (0.016 [⫺0.0034 to 0.035]; P⫽0.18), and C-terminal proendothelin 1 ( ⫺0.025 [⫺0.043 to ⫺0.008]; P⫽0.00094]. In our large, population-based study, we identified MR-proADM and MR-proANP as circulating biomarkers of vascular function most strongly related to noninvasive measures of conduit artery and peripheral arterial performance. Whether determination of blood biomarkers helps to better understand vascular pathology and may provide prognostic information needs to be investigated in future studies. (Hypertension. 2012;60:288-295.) ● Online Data Supplement

Key Words: biomarkers 䡲 flow-mediated dilation 䡲 peripheral arterial tonometry 䡲 cross-sectional 䡲 population-based study

V ascular dysfunction has been shown to be a critical intermediate phenotype of atherosclerotic cardiovascu- lar events.

¹

Early detection of vascular impairment before clinically manifest disease has been the goal of preventive efforts.

^2,3

Methods applied for screening of vascular dysfunc- tion at the population level need to be noninvasive or minimally invasive. Most popular screening methods used recently are composed of the determination of flow-mediated dilation (FMD) of the brachial artery and peripheral arterial

tonometry (PAT).

^4–6

FMD reflects conduit artery reactivity,

⁷

and PAT represents microvascular status.

⁸

Both measures are modestly correlated with coronary endothelial function,

^9,10

and prospective data have remained controversial.

^2,3

Another approach to assess vascular status is the measure- ment of circulating biomarkers that may directly mirror vascular activation. Established and novel vasoactive pep- tides or their more stable precursors are composed of midre- gional proadrenomedullin (MR-proADM), midregional

Received January 23, 2012; first decision February 8, 2012; revision accepted May 8, 2012.

From the Department of General and Interventional Cardiology (R.B.S., C.R.S., S.W., T.Z., E.L., S.B.), University Heart Center, Hamburg-Eppendorf, Germany; Department of Medicine 2 (P.S.W., A.S., A.W., T.G., T.M.), Center of Thrombosis and Hemostasis (P.S.W.), and Institute of Clinical Chemistry and Laboratory Medicine (K.J.L.), University Medical Center Mainz, Johannes Gutenberg-University, Mainz, Germany; B · R · A · H · M · S GmbH (J.K.), Thermo Fisher Scientific, Hennigsdorf, Germany.

S.B. and T.M. contributed equally to the article.

The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.

112.191874/-/DC1.

Correspondence to Renate B. Schnabel, University Heart Center, Department of General and Interventional Cardiology, Martinistr 52, 20246 Hamburg, Germany. E-mail [email protected]

Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.112.191874 288

Downloaded from http://ahajournals.org by on April 25, 2021

(2)

proatrial natriuretic peptide (MR-proANP), N-terminal pro- B-type natriuretic peptide (Nt-proBNP), copeptin as the C-terminal (CT) part of the arginine vasopressin prohormone, CT proendothelin 1, and neopterin. Most of these blood biomarkers have been correlated with measures of vascular function in humans.

^11–15

These circulating biomarkers repre- sent diverse protective and vasoaggressive mechanisms. The natriuretic peptides MR-proANP and Nt-proBNP, as well as MR-proADM, effectively control blood pressure and plasma volume and have been shown to be indicators of cardiovas- cular risk and mortality in initially healthy individuals.

^16,17

The antidiuretic hypothalamic hormone vasopressin regulates osmotic homeostasis through water retention in the kidneys and acts directly on vascular smooth muscle cells.

¹⁸

Endo- thelin 1 is a potent paracrine vasoconstrictor produced predominantly by endothelial cells.

¹⁹

The pteridine deriv- ative neopterin is a marker of monocyte activation and mirrors elevated inflammatory states and vascular oxida- tive stress.

^20,21

Pathophysiologically, these biomarkers are strongly related to vascular homeostasis and reactivity.

The relation of the biomarker panel representing distinct vascular pathways with FMD and PAT in a large, population- based sample is less well established. The aim of our study, therefore, was to directly compare 6 circulating vasoactive peptides with arterial reactivity in a cross-sectional sample of the Gutenberg Health Study.

Methods Study Participants

We determined noninvasive measures of vascular function in the first 5000 individuals of the Gutenberg Health Study, a population-based cohort of individuals of European descent incepted in 2007 at the Department of Medicine 2, University Medical Center Mainz. The sample was composed of randomly selected individuals within a 10-year age strata from 35 to 74 years. Details on study inclusion and cardiovascular risk factor assessment have been published earlier.⁶ In brief, during a clinic visit at the Center for Cardiovascular Prevention at Johannes Gutenberg University Mainz, study participants undergo standardized computer-assisted interviews on cardiovascular risk factors, lifestyle, and socioeconomic status. Standard anthropometric data are collected. Medications are recorded by self-report and from bottles and packages the participants are asked to bring to the study center. We used the following definitions for classical cardiovascular risk factors. Smoking status was composed of nonsmokers (never smokers and former smokers) and smokers.

The diagnosis of diabetes mellitus was based on individuals report- ing a physician diagnosis of diabetes mellitus or a fasting blood glucose concentration ofⱖ126 mg/dL after at least an 8-hour fasting period or a blood glucose level ofⱖ200 mg/dL. Dyslipidemia was diagnosed from a physician diagnosis of dyslipidemia or a low- density lipoprotein/high-density lipoprotein ratio of⬎3.5. Antihy- pertensive drug treatment, a mean systolic blood pressure ofⱖ140 mm Hg, or a mean diastolic blood pressure ofⱖ90 mm Hg led to the diagnosis of hypertension. Family history of myocardial infarction was defined as myocardial infarction of male first-degree relatives until the age of 60 years or female first-degree relatives until the age of 65 years. Self-reported coronary artery disease, myocardial infarction, heart failure, or stroke indicated prevalent cardiovascular disease.

The study has been approved by the local ethics committee; all of the participants provided written, informed consent. All of the authors have read and approved the article as written.

Flow-Mediated Dilation

FMD of the brachial artery after a 5-minute upper arm occlusion was measured under standardized conditions according to guidelines.²² Diameter measurements of the brachial artery were performed on 2D high-resolution ultrasound images recorded on a Philips HD11XE CV ultrasound machine (Philips, Best, Netherlands) with an L12-5 (38 mm) linear array broadband probe. Diameters were measured offline using the commercially available Brachial Analyzer software tool, version 5.0 (Medical Imaging Applications LLC, Iowa City, IA). The means of 3 measurements at baseline and at 60 seconds after cuff release were taken for analysis. Of the 5000 consecutive individuals, 117 FMD cases were missing because of measurement errors and 119 cases were excluded from analysis because brachial diameter before or after was⬍2.4 mm or ⬎6.5 mm or FMD was

⬍0% or ⬎25%.

Peripheral Arterial Tonometry

We used the Endo-PAT2000 (Itamar Medical, Caesarea, Israel) finger tip device to acquire the pneumatic pulse amplitude of the right index finger with the left index finger serving as a control.

Results were automatically calculated using an optimized algo- rithm.⁵The first 336 consecutive data sets were missing because of technical transfer problems. A total of 189 studies were excluded because of noisy signal, inaccurate occlusion duration, or arterial pulse breakthrough during occlusion. In 96.2% we achieved high- quality data.

FMD and PAT were performed simultaneously in a single examination by technicians trained according to protocol and with continuing quality assessment. Further details and quality control data, which showed good reproducibility of the 2 methods in the Gutenberg Health Study, are available in Reference 6.

Biomarker Measurement

Fasting blood was taken before vascular function measurement.

Samples were aliquotted and stored at⫺80°C immediately after blood draw. We used routine laboratory methods for blood glucose and lipid measurements. Details of vascular biomarker determination have been reported earlier.¹⁷ In brief, we measured plasma (MR- proADM, MR-proANP, copeptin, CT proendothelin 1, and neopterin) and serum (Nt-proBNP) biomarkers using commercially available assays. Reproducibility was good, with all of the coefficients of variation (intra-assay and interassay) ⬍5% (please see Table S1, available in the online-only Data Supplement). Neopterin was missing in 1028 individuals because of limited sample volume and, thus, did not enter random forest analyses.

Statistical Methods

Data were analyzed for the total sample of 5000 individuals and stratified by sex. Available case analysis was performed. Skewed variables including all of the biomarkers were logarithmically transformed to achieve near normal distribution. Spearman rank correlation coefficients were calculated for biomarkers, vascular function parameters, and classic cardiovascular risk factors. Partial Spearman rank correlations were determined for biomarkers and measures of vascular function adjusted for age and sex. In linear regression models, biomarkers were related to vascular function parameters per 1-SD increase. Models were adjusted for age and sex, as well as for age, sex, and classic cardiovascular risk factors (body mass index, diabetes mellitus, current smoking, logarithmically transformed pulse pressure, dyslipidemia, and a positive family history of myocardial infarction). To account for renal function we performed secondary analyses adjusting for creatinine concentrations. To avoid confounding by prevalent cardiovascular disease, multivariable-adjusted models excluding individuals with manifest cardiovascular disease and heart failure were computed. Because some antihypertensive drugs and statins may affect vascular function, we also examined multivariable models adjusted for the use of

␤-blockers, angiotensin-converting enzyme inhibitors/angiotensin II type 2 receptor blockers, and dihydropyridine-type calcium channel blockers, as well as statin treatment, in secondary analyses. As a

(3)

statistical significance threshold, P⬍0.05 was used for the current analyses. To account for multiple testing we performed a Bonferroni correction for the number of tests applied in each analysis, that is, the number of biomarkers. P values provided in the Results section have been adjusted and can be interpreted at the 0.05 level.

For statistical calculations we used R software, version 2.13.0 (R Foundation for Statistical Computing, Vienna, Austria;

http://www.R-project.org).

Results

Details on the characteristics of the total study sample and stratified by sex are presented in Table 1. The mean age in the cohort was 55.5 ⫾10.9 years; half of the sample was female.

At the same mean age, cardiovascular risk factors were more

frequent in men except for a positive family history of myocardial infarction, which tended to be more common in women 18.4% versus 17.0% in men. Women showed lower baseline pulse amplitude and smaller brachial artery diame- ters compared with men, and hyperemic response seemed to be more pronounced in women. MR-proADM and CT proen- dothelin 1 concentrations were not different between sexes (median, 0.46 nmol/L and 59.0 pmol/L, respectively, for the total sample). The natriuretic peptides MR-proANP and Nt- proBNP tended to be higher in women, whereas copeptin and neopterin revealed lower plasma levels in women.

Partial Spearman rank correlations between biomarkers and vascular function measures showed moderate relations

Table 1. Characteristics of the Study Sample by Sex

Variable Total Sample (N⫽5000) Men (N⫽2540) Women (N⫽2460)

Age, y 55.5⫾10.9 56.0⫾10.9 55.0⫾11.0

Body mass index, kg/m² 26.5 (23.9/29.8) 27.2 (24.9/30.0) 25.6 (22.9/29.5)

Systolic blood pressure, mm Hg 133.7⫾17.6 136.3⫾16.3 131.0⫾18.4

Diastolic blood pressure, mm Hg 79.0⫾9.0 80.1⫾9.0 77.9⫾8.9

Heart rate, bpm 64.3⫾10.5 63.5⫾10.6 65.2⫾10.2

Pulse pressure, mm Hg 47.0 (40.0/57.0) 48.5 (42.5/57.5) 45.5 (38.0/55.5)

LDL/HDL cholesterol ratio 2.68⫾0.95 2.97⫾0.96 2.39⫾0.84

Creatinine, mg/dL 0.88 (0.79/0.98) 0.96 (0.88/1.04) 0.80 (0.74/0.87)

Diabetes mellitus, N (%) 374 (7.5%) 247 (9.7%) 127 (5.2%)

Current smoking, N (%) 959 (19.2%) 527 (20.8%) 432 (17.6%)

Hypertension, N (%) 2564 (51.3%) 1426 (56.1%) 1138 (46.3%)

Dyslipidemia, N (%) 1462 (29.3%) 923 (36.4%) 539 (21.9%)

Family history, N (%) 886 (17.7%) 433 (17.0%) 453 (18.4%)

Prevalent cardiovascular disease, N (%) 407 (8.1%) 276 (10.9%) 131 (5.3%) Vascular function measures

Baseline pulse amplitude, arbitrary units 6.06⫾0.90 6.42⫾0.74 5.67⫾0.90

PAT ratio 0.55⫾0.44 0.44⫾0.40 0.67⫾0.46

Brachial artery diameter, mm 4.32⫾0.82 4.89⫾0.59 3.74⫾0.59

FMD, % 8.08⫾4.88 6.53⫾3.73 9.69⫾5.39

Circulating biomarkers

MR-proADM, nmol/L 0.46 (0.39/0.54) 0.46 (0.39/0.54) 0.46 (0.39/0.55)

MR-proANP, pmol/L 65.9 (49.0/90.0) 61.5 (45.1/88.1) 69.7 (53.4/92.1)

Nt-proBNP, pg/mL 61.9 (28.5/124.4) 42.2 (17.9/98.0) 81.64 (45.0/146.4)

Copeptin, pmol/L 2.77 (1.79/4.43) 3.64 (2.38/5.67) 2.11 (1.47/3.16)

CT proendothelin 1, pmol/L 59.0 (50.6/68.1) 59.3 (50.8/68.3) 58.5 (50.2/67.9)

Neopterin, nmol/L 5.40 (4.70/6.40) 5.50 (4.70/6.45) 5.39 (4.70/6.33)

Medication

Antihypertensive drugs, N (%) 1559 (31.3%) 838 (33.1%) 721 (29.4%)

␤-blockers, N (%) 235 (9.3%) 206 (8.4%) 441 (8.8%)

ACE inhibitors/AT2 receptor blockers, N (%) 670 (26.5%) 486 (19.8%) 1156 (23.2%)

Calcium channel blockers, N (%) 217 (8.6%) 127 (5.2%) 344 (6.9%)

Statins, N (%) 610 (12.2%) 358 (14.1%) 252 (10.3%)

Continuous variables are described by median and 25th and 75th percentile, if they had a skewed distribution ( skewness⬎1).

Normally distributed variables are presented as mean values and SD. Categorical variables are described through relative and absolute frequencies. LDL indicates low-density lipoprotein; HDL, high-density lipoprotein; ACE, angiotensin-converting enzyme; FMD, flow-mediated dilation; PAT, peripheral arterial tonometry; MR-proADM, midregional proadrenomedullin; MR-proANP, midregional proatrial natriuretic peptide; Nt-proBNP, N-terminal pro-B-type natriuretic peptide; CT, C-terminal. Blood pressure and heart rate measurements were taken before noninvasive vascular function measurement. Prevalent cardiovascular disease contains self- reported coronary artery disease, myocardial infarction, heart failure, or stroke.

(4)

for MR-proADM with baseline pulse amplitude (r ⫽0.13) and PAT ratio (r ⫽⫺0.14), as well as for MR-proANP with baseline pulse amplitude (r ⫽⫺0.13; Figure). Highest corre- lation coefficients for FMD measures were r ⫽⫺0.074 for MR-proANP and brachial artery diameter and r ⫽⫺0.045 for MR-proADM and FMD. Overall, correlations were modest but significant at the 0.05 level. Correlation coefficients between the different blood biomarkers, classic risk factors, and noninvasive vascular function measures are provided in the online-only Data Supplement (Table S2).

In multivariable linear regression models (Table 2) with Bonferroni correction for multiple testing, greatest associa- tions with baseline vascular function were observed for MR-proANP and baseline pulse amplitude ( ␤ per SD increase [99.17% CI], ⫺0.080 [⫺0.115 to ⫺0.044]; P⬍0.0001), as well as for MR-proADM ( ⫺0.044 [⫺0.070 to ⫺0.017];

P

⬍0.0001), MR-proANP (⫺0.033 [⫺0.057 to ⫺0.009];

P

⫽0.0017), and Nt-proBNP (⫺0.027 [⫺0.051 to ⫺0.003];

P

⫽0.015) with brachial artery diameter. A switch in the sign of MR-proADM coefficients for baseline pulse amplitude and brachial artery diameter in multivariable-adjusted models was observed after adding body mass index to the model. The change in sign indicates cross-correlations and potential interactions. For hyperemic response variables, the associa- tion remained statistically significant for CT-proendothelin ( ⫺0.025 [⫺0.043 to ⫺0.008]; P⫽0.00094) with the PAT ratio. Risk factors combined with information on biomarker concentrations explained between 14% (MR-proANP and PAT ratio) and 59% (neopterin and brachial artery diameter) of the variation of the respective vascular function measures.

Biomarkers alone only accounted for 0.01% (MR-proANP and brachial artery diameter) and ⱕ10.80% (copeptin and brachial artery diameter) of the variation.

Discussion

In our contemporary population-based cohort we observed weak-to-moderate associations of circulating vascular bio-

markers with noninvasive vascular function measures. In comparative analyses, MR-proADM and the natriuretic pep- tides were the strongest correlates of both PAT and brachial artery function other than sex, body mass index, and age.

CT proendothelin 1 remained related to PAT ratio in multivariable-adjusted analyses. In general, relations were stronger with baseline vascular function parameters com- pared with hyperemic response. Overall, the amount of variability of vascular function measures explained by circu- lating biomarkers was small.

We compared all of the cardiovascular risk factors and circulating biomarkers. Sex, body mass index, and age were the strongest correlates of noninvasive vascular function.

MR-proADM and the natriuretic peptides MR-proANP and Nt-proBNP provided the most additional information and were fairly consistently related to vascular function across the spectrum of biomarkers investigated.

Vascular impairment is central to cardiovascular disease.

Noninvasive assessment of early vascular dysfunction might help to identify individuals at risk for incident cardiovascular disease and support prevention. Our data may improve the understanding of the cross-sectional relations of the measured biomarkers and vascular function. As expected, higher MR- proADM concentrations were related to worse vascular func- tion in age- and sex-adjusted models, that is, larger brachial artery diameter and baseline pulse amplitude and impaired response to hyperemia. MR-proADM reflects endothelium- dependent, as well as endothelium-independent, vasoreactiv- ity, because adrenomedullin exerts its action through NO- dependent mechanisms

^23,24

but mainly through cAMP.

²⁵

Importantly, increases in cAMP lead to substantial inhibition of vascular superoxide production, which also has been shown to stimulate endothelin 1 production in endothelial and smooth muscle cells.

^26–28

These latter mechanisms may provide insights into vascular function beyond NO response.

Because of their physiological actions, natriuretic peptides capture endothelium-dependent, NO-mediated vasoreactivity, as well as indirect vasodilative effects through activation of the cyclic guanosine monophosphate cascade.

²⁹

An associa- tion of an atrial natriuretic peptide precursor and FMD has been reported earlier.

¹²

We have examined the reliably measurable midregional fragment of the prohormone and can confirm the findings of the Framingham Heart Study and extend the knowledge toward an association with peripheral arterial function. MR-proANP was positively associated with hyperemic response. The phenomenon of higher natriuretic peptide levels with better vascular function has been de- scribed before and may be attributed to beneficial vasodila- tory properties in normal physiology and as a compensatory mechanism in disease.

^30,31

Similar to the experimental setting, atrial natriuretic pep- tide seemed to be the more potent vasodilator and more closely related to vascular reactivity in our current analyses compared with Nt-proBNP as an indicator of B-type natri- uretic peptide action.

³²

A stronger correlation with MR- proANP may indicate that BNP rather represents cardiac function, whereas MR-proANP might more closely mirror vascular reactivity.

Neopterin, nmol/L CT−pro endothelin−1, pmol/L Copeptin, pg/mL Nt−proBNP, pg/mL MR−proANP, pg/mL MR−proADM,

pmol/dL 0.090 −0.12 0.038 −0.012

<0.0001 <0.0001 0.0077 0.40

−0.095 0.036 −0.044 0.018

<0.0001 0.015 0.0021 0.21

−0.020 −0.022 −0.019 0.024

0.19 0.15 0.20 0.10

−0.00038 −0.019 −0.0061 0.019

0.98 0.21 0.67 0.19

0.024 −0.076 0.022 −0.00016

0.10 <0.0001 0.12 0.99

−0.022 −0.023 −0.015 0.034

0.18 0.17 0.35 0.039

Baseline pulse amplitude,

arbitrary units

PAT ratio, arbitrary

units

Brachial artery diameter,

mm

FMD,

%

Figure. Spearman partial correlation coefficients and P values (adjusted for age and sex) for vascular function measures and circulating biomarkers.

(5)

Table 2. Multivariable Linear Regression of Biomarker Concentrations per 1-SD Increase in Relation to Vascular Function Measures in Age- and Sex-Adjusted Models (Upper Row) and Cardiovascular Risk Factor-Adjusted Models (Lower Row) and Biomarker R²Values

Variable N ␤ 99.17% CI P Value

Baseline pulse amplitude, arbitrary units

MR-proADM, nmol/L (R²: 0.045) 4459 0.111 0.074 0.148 ⬍0.0001

4449 ⫺0.005 ⫺0.044 0.034 1.0

MR-proANP, pmol/L (R²: 0.002) 4459 ⫺0.123 ⫺0.159 ⫺0.087 ⬍0.0001

4449 ⫺0.080 ⫺0.115 ⫺0.044 ⬍0.0001

Nt-proBNP, pg/mL (R²: 0.003) 4463 ⫺0.044 ⫺0.081 ⫺0.008 0.0087

4453 ⫺0.021 ⫺0.056 0.015 0.73

Copeptin, pmol/L (R²: 0.049) 4443 0.046 0.012 0.080 0.0021

4433 0.005 ⫺0.028 0.038 1.0

CT proendothelin 1, pmol/L (R²: 0.014) 4458 0.028 ⫺0.006 0.062 0.17

4448 ⫺0.001 ⫺0.034 0.031 1.0

Neopterin, nmol/L (R²: 0.003) 3552 ⫺0.014 ⫺0.050 0.022 1.0

3546 ⫺0.021 ⫺0.056 0.014 0.68

PAT ratio

MR-proADM, nmol/L (R²: 0.033) 4460 ⫺0.066 ⫺0.086 ⫺0.047 ⬍0.0001

4450 ⫺0.022 ⫺0.043 ⫺0.0004 0.043

MR-proANP, pmol/L (R²: 0.001) 4460 0.038 0.018 0.057 ⬍0.0001

4450 0.016 ⫺0.0034 0.035 0.18

Nt-proBNP, pg/mL (R²: 0.007) 4464 0.005 ⫺0.015 0.024 1.0

4454 ⫺0.008 ⫺0.027 0.011 1.0

Copeptin, pmol/L (R²: 0.026) 4444 ⫺0.027 ⫺0.046 ⫺0.009 0.00044

4434 ⫺0.010 ⫺0.028 0.008 0.84

CT proendothelin 1, pmol/L (R²: 0.016) 4459 ⫺0.037 ⫺0.055 ⫺0.020 ⬍0.0001

4449 ⫺0.025 ⫺0.043 ⫺0.008 0.00094

3547 ⫺0.008 ⫺0.027 0.011 1.0

Brachial artery diameter, mm

MR-proADM, nmol/L (R²: 0.032) 4810 0.035 0.010 0.059 0.0015

4797 ⫺0.044 ⫺0.070 ⫺0.017 ⬍0.0001

MR-proANP, pmol/L (R²: 0.0001) 4809 ⫺0.047 ⫺0.072 ⫺0.023 ⬍0.0001

4796 ⫺0.033 ⫺0.057 ⫺0.009 0.0017

Nt-proBNP, pg/mL (R²: 0.012) 4811 ⫺0.038 ⫺0.063 ⫺0.013 0.00028

4798 ⫺0.027 ⫺0.051 ⫺0.003 0.015

Copeptin, pmol/L, (R²: 0.108) 4793 0.036 0.013 0.059 0.00023

4780 0.011 ⫺0.011 0.034 1.0

CT proendothelin 1, pmol/L (R²: 0.017) 4808 0.016 ⫺0.006 0.039 0.34

4795 ⫺0.0004 ⫺0.022 0.022 1.0

3824 ⫺0.004 ⫺0.028 0.019 1.0

FMD, %

MR-proADM, nmol/L (R²: 0.016) 4748 ⫺0.128 ⫺0.333 0.077 0.60

4736 0.123 ⫺0.100 0.347 0.88

MR-proANP, pmol/L, (R²: 0.001) 4747 0.180 ⫺0.021 0.381 0.11

4735 0.154 ⫺0.051 0.358 0.28

Nt-proBNP, pg/mL (R²: 0.001) 4750 0.163 ⫺0.039 0.366 0.20

4738 0.148 ⫺0.054 0.351 0.32

(Continued)

(6)

An interesting finding was the association of CT- proendothelin with PAT ratio as an indicator of microvascu- lar function. No correlation with baseline pulse amplitude could be shown, although a significant correlation between baseline function and PAT ratio is known.

⁶

NO belongs to a variety of physiological inhibitors of endothelin 1 synthesis.

³³

Elevated shear stress during reactive hyperemia after supra- systolic arterial occlusion is characterized by increased NO generation in healthy endothelial cells. In our sample, higher CT proendothelin 1 was related to lower hyperemic micro- vascular response and may, thus, reflect impaired capability of stimulated endothelial NO production not overt at baseline, when no relevant correlation between peripheral pulse am- plitude and the biomarker was seen. Alternatively, endothelin has been shown to be a strong stimulator of vascular superoxide production, which may reduce endothelial NO bioavailability because of its rapid bimolecular reaction with superoxide.

³⁴

Comparing circulating vascular function biomarkers, dif- ferent associations were observed with baseline and hyper- emic response variables in the 2 vascular beds. Although NO bioavailability is central to both reported noninvasive vascu- lar function measures,

^35,36

basal tone and reactivity to isch- emia and subsequent hyperemia are regulated by many other factors produced by endothelial and smooth muscle cells.

Circulating biomarkers may act through a large variety of mechanisms affecting blood pressure, vascular volume, vas- cular tone, and long-term vascular homeostasis, which is not directly correlated with the reaction to ischemia and hyper- emia. Overall, MR-proADM and the natriuretic peptides seemed to be most closely related to vascular function measures. Adrenomedullin has been described as more potent in vascular function regulation in the specific clinical setting of congestive heart failure,

³⁷

an observation that is not clearly supported by our findings on circulating intrinsic biomarkers, because differences in the strength of associations between both markers were minor. Additional investigations are nec- essary to provide more insights into differences of adre- nomedullin and atrial natriuretic peptide as regulators of vascular function.

Limitations

A limitation of our study is the single measurement of vascular function and biomarkers, which does not account for intraindividual variability and changes over time with poten- tial prognostic significance. With cross-sectional data we cannot draw final conclusions toward the clinical applicabil- ity of biomarker measurements compared with noninvasive vascular function. In addition, measurements were performed in individuals in a stable condition. Correlations between vascular function and biomarkers may profoundly differ in the setting of acute cardiovascular disease. Furthermore, our analyses were exploratory, and confirmation of our observa- tions in independent samples is necessary to rule out by- chance findings.

Among the strengths of our investigation counts the avail- ability of data on a broad age range of individuals at risk for cardiovascular disease with simultaneous determination of 2 noninvasive vascular function tests and biomarker concentra- tions. Measurement at a single time point permits a direct comparison independent of many potential confounding con- ditions. Although a range of biomarkers has been assessed in relation to FMD in the past,

¹²

we now provide high-quality blood measurements on novel and more stable biomarkers in a contemporary sample. We can, thus, extend recent obser- vations toward a different set of circulating markers with a direct pathophysiological relation to vascular tone and reac- tivity in 2 distinct noninvasive measures of vascular function.

Our study demonstrates that both blood biomarkers and noninvasive vascular function measures can be acquired under standardized conditions at a large scale. Crucial for the understanding of the value for prevention is the association with cardiovascular outcome. Results have been inconsistent for FMD.

^2,3

and large-scale data at the population level are lacking for PAT. For MR-proADM and MR-proANP,

¹⁷

initial data suggest potential relevance in population screen- ing for susceptibility for cardiovascular disease and kidney function.

³⁸

The next step will, thus, need to be the rigorous comparison of the prospective value of the vascular function markers for cardiovascular diseases in an early, potentially reversible state in the population.

Table 2. Continued

Variable N ␤ 99.17% CI P Value

Copeptin, pmol/L (R²: 0.023) 4731 ⫺0.035 ⫺0.225 0.154 1.0

4719 0.047 ⫺0.144 0.237 1.0

CT proendothelin 1, pmol/L (R²: 0.007) 4746 ⫺0.037 ⫺0.223 0.149 1.0

4734 0.016 ⫺0.171 0.202 1.0

Neopterin, nmol/L (R²: 0.002) 3782 0.072 ⫺0.130 0.274 1.0

3775 0.063 ⫺0.139 0.266 1.0

P values were Bonferroni corrected. Biomarkers were logarithmically transformed for analysis. Provided are the biomarker R² values. Risk factor adjustment was composed of age, sex, body mass index, diabetes mellitus, current smoking, logarithmically transformed pulse pressure, dyslipidemia, and a positive family history of myocardial infarction. FMD indicates flow-mediated dilation;

PAT, peripheral arterial tonometry; MR-proADM, midregional proadrenomedullin; MR-proANP, midregional proatrial natriuretic peptide;

Nt-proBNP, N-terminal pro-B-type natriuretic peptide; CT, C-terminal. Biomarker associations were comparable if stratified for sex (data not shown). Multivariable models accounting for renal function through adjustment for creatinine concentrations (Table S3) and common types of antihypertensive medication and statin therapy (Table S4) are provided in the online-only Data Supplement. The strength of association remained similar as compared with the main analyses. Regression results were similar when individuals with manifest cardiovascular disease were excluded from analyses (Table S5).

(7)

Perspectives

Among a panel of circulating biomarkers of highly vasoactive mediators, we identified the novel biomarkers MR-proADM and MR-proANP as correlates of conduit artery and micro- vascular function in the general population. Whether the information on vascular status gained from a single blood biomarker measurement has a prognostic meaning for vascu- lar function deterioration and incident cardiovascular events needs to be investigated in prospective studies.

Acknowledgments

We thank the participants and dedicated study staff of the Gutenberg Health Study for their generous contribution of time and efforts. We also give thanks to Michael Schlüter for his careful review of the article.

Sources of Funding

The Gutenberg Health Study is funded through the government of Rheinland-Pfalz (“Stiftung Rheinland Pfalz für Innovation,” contract No. AZ 961-386261/733), the research programs “Wissen Schafft Zukunft” and “Schwerpunkt Vaskuläre Prävention” of the Johannes Gutenberg-University of Mainz and its contract with Boehringer Ingelheim and PHILIPS Medical Systems, including an unrestricted grant for the Gutenberg Health Study. This work was further supported by research grants from the Brandenburg Ministry of Economics (Germany) and the European Regional Development Fund (EFRE/ERDF). The testkits for Copeptin, CT-proendothelin 1, MR-proADM, MR-proANP, and Neopterin were provided by B.R.A.H.M.S Hennigsdorf and for Nt-proBNP by Roche Diagnostics Mannheim at no cost (J.K. is an employee of BRAHMS GmbH).

R.S.B. is supported by Deutsche Forschungsgemeinschaft (German Research Foundation) Emmy Noether Program SCHN 1149/3-1.

Disclosures

None.

References

1. Verma S, Buchanan MR, Anderson TJ. Endothelial function testing as a biomarker of vascular disease. Circulation. 2003;108:2054–2059.

2. Yeboah J, Folsom AR, Burke GL, Johnson C, Polak JF, Post W, Lima JA, Crouse JR, Herrington DM. Predictive value of brachial flow-mediated dilation for incident cardiovascular events in a population-based study:

the multi-ethnic study of atherosclerosis. Circulation. 2009;120:502–509.

3. Lind L, Berglund L, Larsson A, Sundstrom J. Endothelial function in resistance and conduit arteries and 5-year risk of cardiovascular disease.

Circulation. 2011;123:1545–1551.

4. Benjamin EJ, Larson MG, Keyes MJ, Mitchell GF, Vasan RS, Keaney JF Jr, Lehman BT, Fan S, Osypiuk E, Vita JA. Clinical correlates and heritability of flow-mediated dilation in the community: the Framingham Heart Study. Circulation. 2004;109:613–619.

5. Hamburg NM, Keyes MJ, Larson MG, Vasan RS, Schnabel R, Pryde MM, Mitchell GF, Sheffy J, Vita JA, Benjamin EJ. Cross-sectional relations of digital vascular function to cardiovascular risk factors in the Framingham Heart Study. Circulation. 2008;117:2467–2474.

6. 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. Noninvasive vascular function measurement in the community: cross-sectional relations and comparison of methods.

Circ Cardiovasc Imaging. 2011;4:371–380.

7. Laurent S, Lacolley P, Brunel P, Laloux B, Pannier B, Safar M. Flow- dependent vasodilation of brachial artery in essential hypertension. Am J Physiol. 1990;258:H1004–H1011.

8. Kuvin JT, Patel AR, Sliney KA, Pandian NG, Sheffy J, Schnall RP, Karas RH, Udelson JE. Assessment of peripheral vascular endothelial function with finger arterial pulse wave amplitude. Am Heart J. 2003;146:168–174.

9. Teragawa H, Ueda K, Matsuda K, Kimura M, Higashi Y, Oshima T, Yoshizumi M, Chayama K. Relationship between endothelial function in the coronary and brachial arteries. Clin Cardiol. 2005;28:460–466.

10. Bonetti PO, Pumper GM, Higano ST, Holmes DR Jr, Kuvin JT, Lerman A. Noninvasive identification of patients with early coronary athero- sclerosis by assessment of digital reactive hyperemia. J Am Coll Cardiol.

2004;44:2137–2141.

11. Nagaya N, Satoh T, Nishikimi T, Uematsu M, Furuichi S, Sakamaki F, Oya H, Kyotani S, Nakanishi N, Goto Y, Masuda Y, Miyatake K, Kangawa K. Hemodynamic, renal, and hormonal effects of adre- nomedullin infusion in patients with congestive heart failure. Circulation.

2000;101:498–503.

12. Kathiresan S, Gona P, Larson MG, Vita JA, Mitchell GF, Tofler GH, Levy D, Newton-Cheh C, Wang TJ, Benjamin EJ, Vasan RS. Cross-sectional relations of multiple biomarkers from distinct biological pathways to brachial artery endothelial function. Circulation. 2006;113:938–945.

13. Robertson GL. The use of vasopressin assays in physiology and patho- physiology. Semin Nephrol. 1994;14:368–383.

14. Kyriakides ZS, Kremastinos DT, Bofilis E, Tousoulis D, Antoniadis A, Webb DJ. Endogenous endothelin maintains coronary artery tone by endothelin type A receptor stimulation in patients undergoing coronary arteriography. Heart. 2000;84:176–182.

15. Ray KK, Morrow DA, Sabatine MS, Shui A, Rifai N, Cannon CP, Braunwald E. Long-term prognostic value of neopterin: a novel marker of monocyte activation in patients with acute coronary syndrome. Circulation.

2007;115:3071–3078.

16. Wang TJ, Gona P, Larson MG, Tofler GH, Levy D, Newton-Cheh C, Jacques PF, Rifai N, Selhub J, Robins SJ, Benjamin EJ, D’Agostino RB, Vasan RS. Multiple biomarkers for the prediction of first major cardio- vascular events and death. N Engl J Med. 2006;355:2631–2639.

17. 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. Contribution of 30 biomarkers to 10-year cardiovascular risk estimation in 2 population cohorts: the MONICA, Risk, Genetics, Archiving, and Monograph (MORGAM) bio- marker project. Circulation. 2010;121:2388–2397.

18. Singh RG. The physiology and emerging roles of antidiuretic hormone.

Int J Clin Pract. 2002;56:777–782.

19. Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang IM, Christman BW, Weir EK, Eickelberg O, Voelkel NF, Rabinovitch M.

Cellular and molecular pathobiology of pulmonary arterial hypertension.

J Am Coll Cardiol. 2004;43:13S–24S.

20. Fuchs D, Hausen A, Reibnegger G, Werner ER, Dierich MP, Wachter H.

Neopterin as a marker for activated cell-mediated immunity: application in HIV infection. Immunol Today. 1988;9:150–155.

21. Razumovitch JA, Semenkova GN, Fuchs D, Cherenkevich SN. Influence of neopterin on the generation of reactive oxygen species in human neutrophils. FEBS Lett. 2003;549:83–86.

22. Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D, Charbonneau F, Creager MA, Deanfield J, Drexler H, Gerhard-Herman M, Herrington D, Vallance P, Vita J, Vogel R. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery:

a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol. 2002;39:257–265.

23. Nakamura M, Yoshida H, Makita S, Arakawa N, Niinuma H, Hiramori K.

Potent and long-lasting vasodilatory effects of adrenomedullin in humans.

Comparisons between normal subjects and patients with chronic heart failure. Circulation. 1997;95:1214–1221.

24. Hayakawa H, Hirata Y, Kakoki M, Suzuki Y, Nishimatsu H, Nagata D, Suzuki E, Kikuchi K, Nagano T, Kangawa K, Matsuo H, Sugimoto T, Omata M. Role of nitric oxide-cGMP pathway in adrenomedullin- induced vasodilation in the rat. Hypertension. 1999;33:689–693.

25. Ishizaka Y, Ishizaka Y, Tanaka M, Kitamura K, Kangawa K, Minamino N, Matsuo H, Eto T. Adrenomedullin stimulates cyclic AMP formation in rat vascular smooth muscle cells. Biochem Biophys Res Commun. 1994;

200:642–646.

26. Muzaffar S, Shukla N, Bond M, Sala-Newby G, Angelini GD, Newby AC, Jeremy JY. Acute inhibition of superoxide formation and Rac1 activation by nitric oxide and iloprost in human vascular smooth muscle cells in response to the thromboxane A2 analogue, U46619. Prosta- glandins Leukot Essent Fatty Acids. 2008;78:247–255.

27. Kahler J, Mendel S, Weckmuller J, Orzechowski HD, Mittmann C, Koster R, Paul M, Meinertz T, Munzel T. Oxidative stress increases synthesis of big endothelin-1 by activation of the endothelin-1 promoter.

J Mol Cell Cardiol. 2000;32:1429–1437.

28. Kahler J, Ewert A, Weckmuller J, Stobbe S, Mittmann C, Koster R, Paul M, Meinertz T, Munzel T. Oxidative stress increases endothelin-1 syn-

(8)

thesis in human coronary artery smooth muscle cells. J Cardiovasc Pharmacol. 2001;38:49–57.

29. van der ZK, Houben AJ, Kroon AA, De Mey JG, Smits PA, de Leeuw PW. Nitric oxide and potassium channels are involved in brain natriuretic peptide induced vasodilatation in man. J Hypertens. 2002;20:493–499.

30. Currie MG, Geller DM, Cole BR, Boylan JG, YuSheng W, Holmberg SW, Needleman P. Bioactive cardiac substances: potent vasorelaxant activity in mammalian atria. Science. 1983;221:71–73.

31. Vesely DL, Douglass MA, Dietz JR, Gower WR Jr, McCormick MT, Rodriguez-Paz G, Schocken DD. Three peptides from the atrial natriuretic factor prohormone amino terminus lower blood pressure and produce diuresis, natriuresis, and/or kaliuresis in humans. Circulation.

1994;90:1129–1140.

32. Nakamura M, Arakawa N, Yoshida H, Makita S, Niinuma H, Hiramori K.

Vasodilatory effects of B-type natriuretic peptide are impaired in patients with chronic heart failure. Am Heart J. 1998;135:414–420.

33. Wagner F, Buz S, Neumeyer HH, Hetzer R, Hocher B. Nitric oxide inhalation modulates endothelin-1 plasma concentration gradients fol-

lowing left ventricular assist device implantation. J Cardiovasc Pharmacol. 2004;44(suppl 1):S89–S91.

34. Pollock DM, Pollock JS. Endothelin and oxidative stress in the vascular system. Curr Vasc Pharmacol. 2005;3:365–367.

35. Nohria A, Gerhard-Herman M, Creager MA, Hurley S, Mitra D, Ganz P.

Role of nitric oxide in the regulation of digital pulse volume amplitude in humans. J Appl Physiol. 2006;101:545–548.

36. Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, Luscher TF. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation. 1995;91:

1314–1319.

37. Oya H, Nagaya N, Furuichi S, Nishikimi T, Ueno K, Nakanishi N, Yamagishi M, Kangawa K, Miyatake K. Comparison of intravenous adrenomedullin with atrial natriuretic peptide in patients with congestive heart failure. Am J Cardiol. 2000;86:94–98.

38. Maier C, Clodi M, Neuhold S, Resl M, Elhenicky M, Prager R, Moertl D, Strunk G, Luger A, Struck J, Pacher R, Hulsmann M. Endothelial markers may link kidney function to cardiovascular events in type 2 diabetes.

Diabetes Care. 2009;32:1890–1895.

Novelty and Significance

What Is New?

● We examined FMD and PAT in a contemporary population-based cohort.

● Six novel circulating biomarkers of cardiovascular function (MR-proADM, MR-proANP, N-terminal pro-B-type natriuretic peptide, copeptin, CT proendothelin 1, and neopterin) were related to vascular function.

What Is Relevant?

● The strongest clinical correlate of noninvasive vascular function other than age and sex consistently was body mass index.

● Correlations between circulating biomarkers and vascular function were weak. In multivariable-adjusted models, statistically significant associations were observed for MR-proADM and natriuretic peptides.

Summary

Among 6 cardiovascular biomarkers we identified MR-proADM and natriuretic peptides as most strongly related to noninvasive measures of arterial performance.