Serum cardiovascular risk biomarkers in pre-pubertal obese children

Article

Reference

Serum cardiovascular risk biomarkers in pre-pubertal obese children

MAGGIO, Albane Bertha, et al.

Abstract

Childhood obesity is associated with premature cardiovascular complications. However, little is known about the effect of a family-based behavioural intervention on the relationship between arterial function, blood pressure and biomarkers in pre-pubertal children with obesity.

MAGGIO, Albane Bertha, et al . Serum cardiovascular risk biomarkers in pre-pubertal obese children. European Journal of Clinical Investigation , 2018, vol. 48, no. 9, p. e12995

PMID : 29992540

DOI : 10.1111/eci.12995

Available at:

http://archive-ouverte.unige.ch/unige:128682

Disclaimer: layout of this document may differ from the published version.

1 / 1

O R I G I N A L A R T I C L E

Serum cardiovascular risk biomarkers in pre ‐ pubertal obese children

Albane B. R. Maggio

| Nathalie J. Farpour-Lambert

| Yacine Aggoun

| Katia Galan

| Fabrizio Montecucco

| François Mach

| Maurice Beghetti

1Santé et Mouvement Consultation, Service of Paediatric Specialties, Department of Child and Adolescent, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

2Obesity Prevention and Care Program“Contrepoids”, Service of Therapeutic Education for Chronic Diseases, Department of Community Medicine, Primary Care and Emergency, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

3Paediatric Sports Medicine Consultation, Service of General Paediatrics, Department of Child and Adolescent, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

4Paediatric Cardiology Unit, Service of Paediatric Specialties, Department of Child and Adolescent, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

5Cardiology Division, Foundation for Medical Research, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

6Cardiology Service, Department of Internal Medicine, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

7First Clinic of Internal Medicine, Department of Internal Medicine, University of Genoa, Genoa, Italy

8Ospedale Policlinico San Martino, Genoa, Italy

9Centre of Excellence for Biomedical Research (CEBR), University of Genoa, Genoa, Italy

Correspondence: Albane B. R. Maggio, Santé et Mouvement Consultation, Service of Paediatric Specialties, Department of Child and Adolescent, University Hospitals of Geneva, 6, rue Willy-Donze, CH-1211 Geneva 14, Switzerland (albane.maggio@hcuge.ch).

Funding information

Fondation Gustave et Simone Prévot;

Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Grant/Award Number:

#310030-118245, #32003B-134963/1, 3200B0-120437; Fondation Ernst et Lucie Schmidheiny

Abstract

Background: Childhood obesity is associated with premature cardiovascular complications. However, little is known about the effect of a family‐based beha- vioural intervention on the relationship between arterial function, blood pressure and biomarkers in pre‐pubertal children with obesity.

Design: This was a single centre randomized controlled trial (RCT) including 74 children randomized to a 6‐month behavioural intervention to treat obesity. In 48 children (13 controls and 35 interventions), we assessed: serum level of cytokine (CCL2), adiponectin, and neutrophil product (MMP‐8), as well as carotid intima‐ media thickness, flow‐mediated dilation (FMD), nitroglycerin‐mediated dilation;

arterial stiffness (incremental elastic modulus, Einc), pulse wave velocity (PWV), resting and 24‐hour blood pressure (BP).

Results: At baseline, resting systolic BP was positively associated with MMP‐8 levels which was significantly higher in children with hypertension (P= 0.033).

Biochemical markers were not related to endothelial function at baseline, but they globally increased after 6 months in the intervention group. The significant increase of CCL2 levels in the intervention group was associated with a decrease in diastolic BP. Furthermore, adiponectin change was positively related to a change in FMD and negatively to change in Einc and PWV.

Conclusions: The usefulness of serum biomarkers for the detection of cardiovas- cular diseases is not well established in children. In our population, MMP‐8

Received: 23 April 2018

|

|

Eur J Clin Invest.2018;48:e12995.

https://doi.org/10.1111/eci.12995

wileyonlinelibrary.com/journal/eci © 2018 Stichting European Society for Clinical Investigation Journal Foundation

| 1 of 10

concentration was higher in hypertensive children. Furthermore, behavioural inter- ventions resulted in a paradoxical increase in some biomarkers in children, with potentially beneficial effects detected with CCL2 changes. Caution should be taken when using nonspecific serum biomarkers for the clinical monitoring of children with obesity.

K E Y W O R D S

biomarker, blood pressure, children, endothelial function, obesity, randomized controlled trial

1 | INTRODUCTION

Cardiovascular diseases are major contributors to the global burden of noncommunicable diseases in the world. Atherosclerosis is a complex multifactorial dis- ease known to start early in life.¹ Childhood obesity is recognized as a risk factor that has been associated with premature cardiovascular complications such as endothelial dysfunction and systemic hypertension.^2-4 One of the major pathophysiological mechanisms of atherosclerosis is the recruitment of circulating leuco- cytes onto the vessel wall and their subsequent migra- tion into the sub‐endothelial space, with activation of inflammation biomarkers.⁵ During inflammation or other damage to the vascular wall, neutrophils recruitment is triggered by chemokines (ie, CXCL8), and mediated by interaction with endothelial adhesion molecules.^5,6 Other noninflammatory mediated mediators, like adipokines, exert an effect on endothelial cells and therefore are interesting in the physiopathology of car- diovascular diseases.²

More recently, new markers of neutrophil activation have been studied as potential biomarkers of cardiovascu- lar risk and atherogenesis. Among them, myeloperoxidase (MPO),⁷ matrix metalloproteinase (MMP‐8; MMP‐9)^8,9 and human neutrophil elastase (HNE)¹⁰ are the most promising to detect early vascular changes, as they are directly secreted by activated neutrophils. In addition, MMP‐9 has been described as a possible systemic marker of atherosclerotic plaque vulnerability.¹¹ However, those markers have been poorly studied in children with obesity, especially during a behavioural intervention.

The first aim of this study was to measure these new biochemical markers of neutrophil activation, as well as adipocytokines in pre‐pubertal children with obesity, and to assess their relationship with arterial function and systemic blood pressure. Secondly, we aimed to determine the effect of a 6‐month family‐based beha- vioural intervention, including supervised physical activity, on these biomarkers in pre‐pubertal children with obesity.

2 | MATERIALS AND METHODS 2.1 | Study design and subjects

This study (n = 48 subjects) was nested in a randomized control trial (RCT), which determined the effectiveness of two 6‐month interventions combining supervised exercise training and a family‐based behavioural intervention (healthy diet, physical activity and well‐being) in individual or in group setting, on body mass index, body fatness, car- dio‐metabolic and psychological co‐morbidities in pre‐pub- ertal children with obesity. The full RCT included 74 pre‐ pubertal children with obesity aged 7.5‐12 years and their mother. Families were randomly distributed in one of the following groups: (a) control (standard care, n = 22); or (b) intervention (n = 52). Sealed opaque envelopes contain- ing 2/3 of intervention and 1/3 of control were used. Fami- lies could then choose between an exercise training program twice a week with a behavioural intervention either in an individual setting (7 sessions, one paediatrician and one dietician) or in a group setting (14 sessions, multi- disciplinary team).

Subjects were recruited at the Pediatric Obesity Clinic of the University Hospital of Geneva. Obesity was defined as a body mass index (BMI) above the 97th percentile by age and gender.¹²

Subjects were excluded from the study if they had any of the following criteria: (a) a pubertal Tanner stage≥2;

(b) a history of familial dyslipidemia or hypertension; (c) a history of recent illness, bone fracture or viral syndrome within 2 weeks of the blood draw; (d) taken any medica- tions or hormones which might influence cardiovascular function, body composition, lipid or glucose metabolism in the previous 6 months; (e) a genetic disorder or a chronic disease.

The Mother and Child Ethics Committee of the Univer- sity Hospitals of Geneva approved this study and a written informed consent was obtained from both parent and child.

For this nested study, we analysed only children with available blood marker data at baseline and at 6 months (48 children: controls, n = 12; group therapy, n = 18; indi- vidual therapy, n = 18).

2.2 | Measures

All subjects underwent identical testing protocols at the Pediatric Research Platform. The protocol was repeated at 6 months. The health professionals of the Pediatric Research Platform and at the Pediatric Cardiology Unit were blind to group allocation, whereas subjects and inter- vention delivery staff could not be blinded.

2.3 | Anthropometrics and body composition

Standing height was measured to the nearest 0.1 cm on a wall‐mounted stadiometer. Weight in kilograms, with light clothes on but shoes off, was measured on scales (Seca).

Body mass index (BMI) was calculated as weight/height² (kg/m²) and z‐scores were derived using the World Health Organization references.¹³Fat mass (FM, kg), fat‐free mass (FFM, kg), percentage of total body fat (%) and percentage of abdominal fat (%) were measured using dual‐energy X‐ ray absorptiometry (DXA – GE Lunar Prodigy^™, Lunar Corp., Madison, WI, USA).

2.4 | Biochemical markers

Serum level of matrix metalloprotease (MMP)‐8, chemo- kine ligand 2 (CCL2) and interleukin 8 (CXCL8) were measured by colorimetric enzyme‐linked immunosorbent assay (ELISA) (RayBiotech, Inc., Norcross, GA, USA).

Serum Human Neutrophil Elastase (HNE) levels were mea- sured by ELISA (IBL International, Hamburg, Germany).

Serum myeloperoxidase (MPO) levels were measured by ELISA (Immunology Consultants Laboratory, Inc, Portland, OR, USA). Serum adiponectin and MMP‐9 were measured by ELISA (R&D Systems, Minneapolis, MN, USA). All ELISA assays were performed following the manufacturer's instructions. The limits of detection was 8.23 pg/mL for MMP‐8, 2.0 pg/mL for CCL2, 0.8 pg/mL for CXCL8, 0.16 ng/mL for HNE, 1.875 ng/mL for MPO, 0.312 ng/mL for MMP‐9 and 3.9 mg/mL for adiponectin. Mean intra‐ and inter‐assay coefficients of variation (CV) were below 15% for all markers.

2.5 | Systemic blood pressure

In all subjects, the left brachial resting blood pressure (BP) was assessed by oscillometric method (Dinamap, GE Healthcare, Chicago, IL, USA) after 10 minutes of recum- bence. In addition, 24‐hour ambulatory BP monitoring (ABPM) was performed at baseline using an automatic mon- itor with a standing position captor (Dyasis Integra II, Physi- cor S.A., France) validated by the British Hypertension Society. The BP z‐scores were calculated according to the LMS method and HTN was defined as az‐score above 2.¹⁴

2.6 | Arterial function

Noninvasive measurements of arterial function were per- formed with a real time B‐mode ultrasound imager (Vingmed CFM800C system Ltd, Norway) using a 10 MHz linear high‐resolution vascular probe as previ- ously described.³ Intima‐media thickness (IMT) values were categorized according to their percentile, either above or under the 97th percentile for age.¹⁵ The pulse wave of the radial artery was assessed to estimate nonin- vasively central aortic pressure. This parameter allowed to determine mechanical indices and arterial stiffness using the incremental elastic modulus (Einc). We also assessed the arterial pressure wave at the carotid and femoral arteries to calculate the pulse wave velocity (PWV) according to a standardized method.¹⁶ The PWV and Einc were further categorized according to their per- centile, either above or under both the 25th and 75th percentiles for age.^15,17 Higher Einc and PWV indicated stiffer arteries. Noninvasive assessment of endothelium‐ dependent dilation (flow‐mediated dilation ‐ FMD) and endothelium‐independent dilatation (dilatation response to 300μg sublingual nitroglycerin, NTG) of the right bra- chial artery (RBA) was performed by the same echo- graphic vascular linear probe. After baseline measure, the dilation of the RBA in response to increased flow or NTG was measured and FMD or NTGMD was calcu- lated as absolute and percentage maximum increase in vessel size from baseline.

2.7 | Statistical analysis

Statistical analyses were performed using SPSS software 22.0 (Chicago, IL, USA). For an anticipated effect size of 0.1 for BMI z‐score (SD 0.1), a sample size of 16 sub- jects in each group was required to detect a statistically significant difference at P< 0.05 with a statistical power of 80% (β = 0.80). Due to missing biomarker data, the number of subjects per groups was not sufficient to allow adequate power when assessing intervention changes or comparing groups. Statistical power of intervention changes was 20%‐70% for biomarkers, 5%‐64% for blood pressure and 6%‐39% for endothelial function variables.

Power to compare intervention (individual and group together) and control groups was better although still moderate: 45%‐70% for endothelial function and 60%‐70%

for biochemical markers. Despite this limitation, we chose to perform the analyses knowing that our principal aims were to assess the correlations between biochemical mark- ers and other variables.

Data were screened initially for normalcy, using skew- ness and kurtosis tests. Data are presented as mean and standard deviation (SD) or interquartile range for

MAGGIOET AL.

|

nonnormaly distributed variables (designed with an ^a in Table 1). Since the majority of biomarkers were not nor- mally distributed, we used Spearman's correlations (two‐ tailed) to describe relationships among variables, Mann‐ Whitney test to compare groups and Wilcoxon Signed Ranks test to compare variables before and after the inter- vention. For other normally distributed variables, we used Student'st test, pairedt test and chi‐square when appropri- ate. Diagnostic efficiency receiver operating characteristic (ROC) curve was performed with calculation of the area under the curve (AUC) and the determination of a cut‐off point for MMP‐8 concentration between hypertensive and nonhypertensive children. Differences were considered sig- nificant ifP< 0.05.

3 | ^RESULTS

3.1 | Evaluation at baseline 3.1.1 | Patients ’ characteristics

The characteristics of the 48 children selected for this study are presented in Table 1.

3.1.2 | Relationship between biochemical markers and blood pressure

Resting systolic and diastolic BP z‐scores were positively related to MMP‐8 (SBP: r= 0.491, P< 0.001; DBP:

r= 0.352, P= 0.019), with significantly higher MMP‐8 concentration in children with 24‐hour systolic HTN (50.2 ± 42.6 vs 19.6 ± 14.9 ng/mL,P= 0.033; Figure 1A).

The ROC curve (Figure 1B) showed an AUC of 0.778 (95% CI: 0.582‐0.974; P= 0.034) with a cut‐off point estimated at 18.3 ng/mL. With this cut‐off, the specificity for detecting systolic HTN in children with MMP‐8 con- centration higher than 18.3 ng/mL was 60% (18/30), the sensitivity 83% (5/6), the positive predictive value 95%

(18/19) and the negative predictive value 29% (5/17). Fur- thermore, CXCL8 was also positively correlated with SBP and DBP z‐scores (SBP: r= 0.309, P= 0.041; DBP:

r= 0.329, P= 0.029) but did not differ between hyper- tensive and nonhypertensive children. There was no corre- lation between any of the markers and 24 hours‐BP measures.

3.1.3 | Relationship between biochemical markers and endothelial function

We found no association between marker levels and endothelial function. Furthermore, there was no difference in marker concentrations among children at the different percentiles of Einc, PWV or IMT.

3.2 | Evaluation of changes after intervention and correlation with change in adiposity

We investigated the change in biomarker concentra- tions, blood pressure measurement and endothelial

T A B L E 1 Physical characteristics, biochemical markers, blood pressure and endothelial function

n = 48

Age (years) 9.7 ± 1.0

Gender female n, (%) 20 (42)

Intervention group n, (%) 35 (73)

Anthropometrics

Height (cm) 140.8 ± 6.8

Weight (kg) 50.4 ± 9.0

BMI (kg/m²) 25.3 ± 3.0

BMIz‐score 2.9 ± 0.8

Body composition

Percentage of total body fat (%) 43.3 ± 4.7 Percentage of abdominal fat (%) 51.4 ± 5.5

Total body fat mass (kg) 21.5 ± 5.6

Fat‐free mass (kg) 27.5 ± 3.7

Chemokines and adipokines

CCL2 (pg/mL) 185.2 ± 64.0

CXCL8 (pg/mL) 25.5 ± 33.5^a

Adiponectin (microg/mL) 4.8 ± 2.5^a

Neutrophil products

MMP‐8 (ng/mL) 18.1 ± 12.7^a

MMP‐9 (ng/mL) 430.8 ± 228.4^a

MPO (ng/mL) 225.5 ± 200.2^a

HNE (ng/mL) 571.4 ± 570.9^a

Blood pressure

Resting SBP (mm Hg) 110.8 ± 9.2

Resting DBP (mm Hg) 69.4 ± 9.0

Resting SBPz‐score 0.74 ± 0.82

Resting DBPz‐score 0.73 ± 0.78

24‐h SBP (mm Hg) 114.7 ± 10.3

24‐h DBP (mm Hg) 69.6 ± 7.7

24‐h SBPz‐score 0.75 ± 1.25

24‐h DBPz‐score 0.60 ± 1.40

Systolic HTN (%) 12.5

Diastolic HTN (%) 14.6

Systolic and diastolic HTN (%) 6.3

Endothelial function

IMT (mm) 0.53 ± 0.06

(Continues)

function between baseline and 6 months. As the magnitude of change between individual and group therapies was similar (P> 0.05 for all), we analysed all subjects (n = 35) together in the intervention group.

The changes in the control and intervention groups, as well as the effects of the intervention, are presented in Table 2. A greater BMI z‐score decrease was observed in the intervention group compared to con- trols.

3.2.1 | Biochemical markers

Globally, chemokines and neutrophil products increased in the intervention group and decreased in the control group, but the difference was significant only for MMP‐ 8, MPO and CXCL8 levels. MPO, MMP‐9, CCL2 and adiponectin changes were significantly negatively related to change in percentage of abdominal fat (P> 0.05 for all).

3.2.2 | Blood pressure and endothelial function

There was no change in blood pressure measures or vascu- lar function after the intervention. However, in the control group, the significant increase in percentage of abdominal fat was related to a decrease in FMD (r=−0.738, P= 0.037). There was no relationship between change in adiposity and change in blood pressure.

3.2.3 | Correlation between changes in biomarkers and blood pressure

As only eight children performed the 24‐hour ambulatory BP monitoring at the two time points, we did not analyse the effect of the intervention on those data. However, at rest, a decrease in resting DBP z‐scores was associated

with an increase in CCL2 concentrations (r= −0.326, P= 0.035). This association was found in the intervention group only (r=−0.370, P= 0.041).

3.2.4 | Correlation between changes in biomarkers and endothelial function

The increase in CCL‐2 concentration was associated with an increase in NTGMD (r= 0.522, P= 0.005) while the decrease in adiponectin level was associated with a decrease in FMD (r= 0.525, P= 0.005, Figure 2A) as well as an increase in both Einc (r=−0.418, P= 0.030, Figure 2B) and PWV (r=−0.494, P= 0.009, Figure 2C).

4 | DISCUSSION

Our study showed that in pre‐pubertal children with obe- sity, endothelial function parameters were not related to inflammatory biomarkers. However, two of these biomark- ers (MMP‐8 and CXCL8) correlated with blood pressure at rest, MMP‐8 being significantly higher in children with systolic hypertension. A behavioural intervention, including nutrition education and regular physical activity, led not only to an improvement of the BMIz‐score but also to an unexpected increase in the levels of these markers. Their increase was shown to be related to a decrease in the per- centage of abdominal fat and to an improvement of dias- tolic blood pressure and endothelial function (NTGMD) for CCL2. Furthermore, changes in adiponectin, a molecule with anti‐atherosclerotic properties, were linked to change in endothelial function in children.

Obesity is recognized as a risk factor for atherosclero- sis through dyslipidemia, hypertension and chronic inflammation. This inflammation leads to the secretion of many biochemical markers known to augment the risk for plaque fissuring and adverse cardiovascular outcomes in adults.7,10,11,18-21 In children, CXCL8, MPO, MMP‐8 and MMP‐9 levels have been shown to be higher in chil- dren with obesity compared to children with normal weight,^22-26 MMP‐9 being higher in hypertensive than in nonhypertensive children with obesity ^23,27 and MPO cor- relating with cardiovascular risk factors.^22,26 In our study, although there was no control group of children with normal weight to compare biomarkers levels, their rela- tionship with cardiovascular health indicators, such as blood pressure and vascular function, was examined.

Resting blood pressure was found to be normal in 85%

of children at baseline. However, systolic and diastolic blood pressure z‐scores were positively related to MMP‐8 and CXCL8 levels. To our knowledge, contrarily to MMP‐9, correlations between these biomarkers and blood pressure have never been described in children. In adults, T A B L E 1 (Continued)

n = 48

Einc (mm Hg/10²) 8.7 ± 6.4^a

PWV (m/s) 4.4 ± 0.9^a

FMD (%) 4.3 ± 4.1^a

NTGMD (%) 23.3 ± 8.3

BMI, body mass index; CCL2, Chemokine ligand 2; CXCL8, interleukin 8;

DBP, diastolic blood pressure; Einc, elastic incremental modulus; FMD, flow‐ mediated dilation; HNE, Human Neutrophil Elastase; IMT, intima‐media thick- ness; MMP‐8 and 9, matrix metalloprotease; MPO, myeloperoxidase; NTGMD, nitroglycerin‐mediated dilation; PWV, aortic pulse wave velocity; SBP, systolic blood pressure.

Results are expressed as mean and SDS or median and interquartile range for non‐normally distributed variables (^a).

MAGGIOET AL.

|

MMP‐8 has been recognized as a modulator of angioten- sin I cleavage with lower blood pressure in MMP‐8 knockout mice.²⁸ In the present paper, MMP‐8 was sig- nificantly higher in children with systolic hypertension assessed through 24‐hour BP measurement. Based on the difference in MMP‐8 levels between children with and without HTN, we proposed a cut‐off value with good sensitivity and average specificity. The very high positive predictive value we found (95%) indicates that a value above 18.3 ng/mL suggests HTN, which can be helpful

when a 24‐hour BP recording cannot be performed. How- ever, this cut‐off is valid in pre‐pubertal children with obesity only.

Furthermore, we explored endothelial function through the evaluation of dependent (FMD) and independent (NTGMD) dilatation of the carotid artery, which reflects the early stage of alteration in vasomotor activity as well as arterial stiffness (Einc and PWV) and IMT. Despite the lack of control group at baseline, it is reasonable to assume, given our findings in a previous publication on a F I G U R E 1 A, MMP‐8 concentrations’ comparison between hypertensive and nonhypertensive children. Figure refers to Box and Whisker plots (median value, 25%

and 75% percentiles). Dotted line: cut‐off value at 18.3 ng/mL. B, Receiver operating characteristic curve and area under the curve (AUC) for MMP‐8 concentrations’ comparison between hypertensive and nonhypertensive children. AUC = 0.778;

95% CI: 0.582‐0.974;P= 0.034

similar sample of prepubertal children with obesity,³ that subjects in the present study had impaired endothelial func- tion. Despite this, we did not find any relationship between measured markers and vascular function, or any difference in markers concentration through vascular measures per- centiles. This may suggest that those markers are not appropriate to detect early vascular changes or that their variability is too high to be able to detect such subtle changes. One pediatric study performed in older subjects with obesity showed that MMP‐9 levels and IMT of the carotid arteries were higher compared to a control group and that they were correlated to each other.²⁵ To our knowledge, no other study has investigated the relationship between biomarkers and vascular function, especially in pre‐pubertal children with obesity.

In addition, the effect of behavioural interventions on these inflammatory biochemical markers has never been studied in children. In our sample, despite an improvement

of BMI z‐score, children in the intervention group showed unexpected proportional increase in inflammatory markers (CCL2, CXCL8) and proteases (MPO, MMP‐8) when com- pared with the control group. Indeed, almost all the mark- ers (except one: CCL‐2) did decrease in the control group, with all of them increasing in the intervention group. These different trends highlight a controversial effect of the inter- vention on those markers. We might speculate on different reasons why these markers decreased only in the control group. First of all, the control group was treated with a

“standard treatment care” that could have induced some modifications and partial improvement on the levels of inflammatory biomarkers. These data are rather observa- tional since, differently from adult with obesity, these mod- ifications have not been clearly associated with clinical outcomes in children. On the other hand, the observed increase in the intervention group could be explained by a paradoxical effect of lifestyle change on inflammation. One T A B L E 2 Effect of the intervention on changes in biochemical markers, blood pressure and endothelial function

Variables

Mean changes

Controls Intervention Intervention

effectP‐value

(n = 13) (n = 35)

BMIz‐score −0.07 (0.15) −0.14 (0.25)** 0.554

% abdominal fat 1.4 (2.2)* −0.6 (4.2) 0.066

Chemokines and adipokines

CCL2 42.4 (190) 75.0 (198)* 0.668

CXCL8 −27.8 (41.1) 38.6 (283.5)* 0.006^¥¥

Adiponectin −0.9 (1.4) −0.3 (1.7) 0.117

Neutrophil products

MMP‐8 −16.4 (32.1) 2.3 (16.8) 0.022^¥¥

MMP‐9 −93.3 (408)* 4.6 (286) 0.737

MPO −125.8 (167) 25.2 (184) 0.012^¥

HNE −223.0 (707.3) 33.7 (1143.5) 0.093

Blood pressure (n = 11) (n = 31)

Rest. SBPz‐score −0.14 (1.0) −0.02 (0.9) 0.844

Rest.DBPz‐score −0.12 (0.6) −0.2 (1.0) 0.714

Endothelial function (n = 9) (n = 18)

IMT −0.04 (0.06) 0.004 (0.05) 0.076

Einc 2.1 (3.5) 2.4 (8.5) 0.820

PWV −0.9 (2.8) −0.07 (0.5) 0.731

FMD −2.2 (4.5) 0.4 (3.4) 0.145

NTGMD −2.3 (13.0) −2.7 (9.4) 1.0

Mean change is defined as the variable change from baseline to 6 mo in each groups and is indicated as mean and SD.

CCL2, chemokine ligand 2; CXCL8, interleukin 8; DBP, diastolic blood pressure; Einc, elastic incremental modulus; FMD, flow‐mediated dilation; HNE, Neutrophil Elastase; IMT, intima‐media thickness; MMP‐8 and 9, matrix metalloprotease; MPO, myeloperoxidase; NTGMD, nitroglycerin‐mediated dilation; PWV, aortic pulse wave velocity; SBP, systolic blood pressure.

*P< 0.05; *^*P< 0.01 for intra‐group change.

¥P< 0.05;^¥¥P< 0.01;^¥¥¥P< 0.001 for inter‐groups changes.

MAGGIOET AL.

|

component of the intervention was the introduction of regu- lar physical activity twice a week for a total amount of 3 hours per week. This factor, even though not alone, may have had an effect on those markers. Indeed, a study per- formed in adults with artheriopathy showed that a short‐ term exercise (treadmill test) induced an activation of tissue metalloproteinases and an increased expression of adhesion molecules and inflammatory substances.²⁹ These results may suggest that regular physical activity during a short period of time could have negative effects on cardio‐meta- bolic parameters. However, the clinical implication of this increase is difficult to ascertain, especially in pre‐pubertal children without known artheriopathy. To our knowledge, no studies have investigated the effect of a long‐term per- iod of physical activity in human. However, in rats, a 2‐ week program of physical activity, which may be represen- tative of a long‐term period for human, induced some dam- age to skeletal muscles and connective tissue, which lead to activation of tissue metalloproteinases.³⁰ In our study, we looked at the clinical impact of the intervention, which included healthy lifestyle changes as well as physical activ- ity, on cardiovascular health indicators such as blood pres- sure or vascular function, and their relationship with biomarkers changes. We could demonstrate that increases in CCL2 levels were related to an improvement in diastolic blood pressure and smooth muscle cell function (increase in NTGMD) in the intervention group. Therefore, these markers may have a long‐term beneficial effect, and cau- tion is required in the interpretation of this increase, espe- cially in the light of the complex and unclear roles of these markers and their regulators.³¹ The direct effect of physical activity could not be evaluated in our study first because we did not monitor the total amount of physical activity performed by children and also because of other potential metabolic changes induced by the various lifestyle changes that occurred during the intervention. Indeed it is known that there is a multitude of hormonal and biochemical fac- tors that are implicated in vascular control, inflammation and weight regulation, which may have been influenced by lifestyle changes.^32,33

The behavioural intervention did not have a significant effect on the endothelial function, probably due to the small number of subjects. However, a decrease in adipo- nectin level was related to a worsening of the endothelial function (decreased FMD), as well as an increase in Einc and PWV. Adiponectin is an adipocyte‐derived protein known to suppress the attachment of monocytes to endothelial cells and lipid accumulation in macrophage and is consequently considered as an anti‐atherogenic protein.³⁴ Thus, our findings seem to confirm its protective role on vascular walls.^35,36 It is worth noting that one study per- formed in children showed no relationship between adipo- nectin and FMD,³⁷ however their adiponectin level may F I G U R E 2 Relationship between adiponectin concentration and

vascular function. A, Flow‐mediated dilation. N = 27;r= 0.525;

P= 0.005. B, Elastic incremental modulus. N = 27;r=−0.418;

P= 0.030. C, Pulse wave velocity. N = 27;r=−0.494;P= 0.009

have been too high to reveal an association with impaired endothelial function. The latter study also corroborates our finding that adiponectin is not yet related to IMT in young children, contrarily to results found in adolescent stud- ies.^38,39 Furthermore, in our control group, the percentage of abdominal fat mass increased after 6 months with a deleterious effect on vascular function (decreased FMD).

This study has some limitations. First, the sample size was probably too low to observe significant changes between the two groups and the impact of the intervention.

Second, only few children consented to do the 24 hours‐ ambulatory BP test at the end of the trial.

In conclusion, our results suggested that among the new biomarkers, only MMP‐8 may be used to detect high blood pressure in pre‐pubertal children with obesity. Fur- thermore, a “short‐term” exercise training program, com- bined with a family‐based behavioural intervention, may increase low‐grade inflammation with the secretion of neu- trophil products, with an unexpected beneficial impact on blood pressure and endothelial function, especially for CCL2. Furthermore, the beneficial effect of lifestyle inter- ventions on adiposity may prevent the decrease in adipo- nectin concentration and consequently preserve arterial wall properties. Numerous adult studies have demonstrated the beneficial long‐term effect of regular physical activity on vascular health. Further pediatric studies are now needed to investigate the effect of interventions of different types, durations, frequencies and intensities on the secretion of vascular biomarkers and changes of vascular function.

A C K N O W L E D G E M E N T S

We thank the subjects for volunteering for the study, Xavier Martin, Sophie Bucher Della Torre, Lydia Lanza Von Haller, Maude Bessat, Jérôme Martin and the staff of the Pediatric Research Platform for their assistance.

C O N F L I C T O F I N T E R E S T

This study was supported financially by the “Schmidheiny Foundation”, the Swiss National Foundation (no 3200B0‐ 120437, Clinical trials NCT01506245), and was entitled:

“Exercise Training and Family‐based Behavioural Treat- ment in Pre‐pubertal Obese Children and their Mother: A randomized Controlled Trial”, the Swiss National Science Foundation (grant #310030‐118245 to Dr. F. Mach and grant #32003B‐134963/1 to Dr. F. Montecucco) and by the Foundation“Gustave and Simone Prévot”to Dr. F. Monte- cucco. The funders were not involved in the study design, data collection, analysis, interpretation, or in the manuscript preparation and decision to publish. The authors have no conflicts of interest to declare.

O R C I D

Albane B. R. Maggio http://orcid.org/0000-0002-9110- 4467

Fabrizio Montecucco http://orcid.org/0000-0003-0823- 8729

R E F E R E N C E S

1. Berenson GS, Srinivasan SR, Bao W, Newman WP 3rd, Tracy RE, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study.N Engl J Med. 1998;338:1650‐1656.

2. Sowers JR. Obesity as a cardiovascular risk factor. Am J Med.

2003;115(suppl 8A):37S‐41S.

3. Aggoun Y, Farpour-Lambert NJ, Marchand LM, Golay E, Mag- gio AB, Beghetti M. Impaired endothelial and smooth muscle functions and arterial stiffness appear before puberty in obese children and are associated with elevated ambulatory blood pres- sure.Eur Heart J. 2008;29(6):792‐799.

4. Maggio AB, Aggoun Y, Marchand LM, et al. Associations among obesity, blood pressure, and left ventricular mass.J Pedi- atr. 2008;152:489‐493.

5. Price DT, Loscalzo J. Cellular adhesion molecules and atherogen- esis.Am J Med. 1999;107:85‐97.

6. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis.

Circulation. 2002;105:1135‐1143.

7. Tang WH, Wu Y, Nicholls SJ, Hazen SL. Plasma myeloperoxi- dase predicts incident cardiovascular risks in stable patients undergoing medical management for coronary artery disease.Clin Chem. 2011;57:33‐39.

8. Mallat Z. Matrix metalloproteinase‐8 and the regulation of blood pressure, vascular inflammation, and atherosclerotic lesion growth.Circ Res. 2009;105:827‐829.

9. Welsh P, Whincup PH, Papacosta O, et al. Serum matrix metallo- proteinase‐9 and coronary heart disease: a prospective study in middle‐aged men.QJM. 2008;101:785‐791.

10. Pradhan-Palikhe P, Vikatmaa P, Lajunen T, et al. Elevated MMP‐8 and decreased myeloperoxidase concentrations associate signifi- cantly with the risk for peripheral atherosclerosis disease and abdominal aortic aneurysm.Scand J Immunol. 2010;72:150‐157.

11. Zeng B, Prasan A, Fung KC, et al. Elevated circulating levels of matrix metalloproteinase‐9 and ‐2 in patients with symptomatic coronary artery disease.Intern Med J. 2005;35:331‐335.

12. Rolland-Cachera MF, Cole TJ, Sempe M, Tichet J, Rossignol C, Charraud A. Body Mass Index variations: centiles from birth to 87 years.Eur J Clin Nutr. 1991;45:13‐21.

13. WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Length/Height‐for‐Age, Weight‐for‐Age, Weight‐for‐Length, Weight‐for‐Height and Body Mass Index‐for‐ Age: Methods and development. Geneva: World Health Organiza- tion; 2006. 312 pp.

14. Wuhl E, Witte K, Soergel M, Mehls O, Schaefer F. Distribution of 24‐h ambulatory blood pressure in children: normalized reference val- ues and role of body dimensions.J Hypertens. 2002;20:1995‐2007.

15. Jourdan C, Wuhl E, Litwin M, et al. Normative values for intima‐media thickness and distensibility of large arteries in healthy adolescents.J Hypertens. 2005;23:1707‐1715.

MAGGIOET AL.

|

16. Weber T, Ammer M, Rammer M, et al. Noninvasive determina- tion of carotid‐femoral pulse wave velocity depends critically on assessment of travel distance: a comparison with invasive mea- surement.J Hypertens. 2009;27:1624‐1630.

17. Reusz GS, Cseprekal O, Temmar M, et al. Reference values of pulse wave velocity in healthy children and teenagers.Hyperten- sion. 2010;56:217‐224.

18. Meuwese MC, Stroes ES, Hazen SL, et al. Serum myeloperoxi- dase levels are associated with the future risk of coronary artery disease in apparently healthy individuals: the EPIC‐Norfolk Prospective Population Study.J Am Coll Cardiol. 2007;50:159‐ 165.

19. Wong ND, Gransar H, Narula J, et al. Myeloperoxidase, subclini- cal atherosclerosis, and cardiovascular disease events.JACC Car- diovasc Imaging. 2009;2:1093‐1099.

20. Tuomainen AM, Nyyssonen K, Laukkanen JA, et al. Serum matrix metalloproteinase‐8 concentrations are associated with car- diovascular outcome in men. Arterioscler Thromb Vasc Biol.

2007;27:2722‐2728.

21. Kato R, Momiyama Y, Ohmori R, Taniguchi H, Nakamura H, Ohsuzu F. Plasma matrix metalloproteinase‐8 concentrations are associated with the presence and severity of coronary artery dis- ease.Circ J. 2005;69:1035‐1040.

22. Olza J, Aguilera CM, Gil-Campos M, et al. Myeloperoxidase is an early biomarker of inflammation and cardiovascular risk in prepubertal obese children.Diabetes Care. 2012;35:2373‐2376.

23. Glowinska-Olszewska B, Urban M. Elevated matrix metallopro- teinase 9 and tissue inhibitor of metalloproteinase 1 in obese chil- dren and adolescents.Metabolism. 2007;56:799‐805.

24. Belo VA, Souza-Costa DC, Lana CM, et al. Assessment of matrix metalloproteinase (MMP)‐2, MMP‐8, MMP‐9, and their inhibitors, the tissue inhibitors of metalloproteinase (TIMP)‐1 and TIMP‐2 in obese children and adolescents. Clin Biochem.

2009;42:984‐990.

25. Andrade C, Bosco A, Sandrim V, Silva F. MMP‐9 levels and IMT of carotid arteries are elevated in obese children and adolescents compared to non‐obese.Arq Bras Cardiol. 2017;108:198‐203.

26. Correia-Costa L, Sousa T, Morato M, et al. Association of myeloperoxidase levels with cardiometabolic factors and renal function in prepubertal children.Eur J Clin Invest. 2016;46:50‐59.

27. Martinez-Aguayo A, Campino C, Baudrand R, et al. Cortisol/cor- tisone ratio and matrix metalloproteinase‐9 activity are associated with pediatric primary hypertension.J Hypertens. 2016;34:1808‐ 1814.

28. Laxton RC, Hu Y, Duchene J, et al. A role of matrix metallopro- teinase‐8 in atherosclerosis.Circ Res. 2009;105:921‐929.

29. Signorelli SS, Mazzarino MC, Di Pino L, et al. High circulating levels of cytokines (IL‐6 and TNFalpha), adhesion molecules (VCAM‐1 and ICAM‐1) and selectins in patients with peripheral arterial disease at rest and after a treadmill test. Vasc Med.

2003;8:15‐19.

30. Carmeli E, Moas M, Lennon S, Powers SK. High intensity exer- cise increases expression of matrix metalloproteinases in fast skeletal muscle fibres.Exp Physiol. 2005;90:613‐619.

31. Newby AC. Dual role of matrix metalloproteinases (matrixins) in intimal thickening and atherosclerotic plaque rupture. Physiol Rev. 2005;85:1‐31.

32. Poulain-Godefroy O, Lecoeur C, Pattou F, Fruhbeck G, Froguel P. Inflammation is associated with a decrease of lipogenic factors in omental fat in women.Am J Physiol Regul Integr Comp Phys- iol. 2008;295:R1‐R7.

33. Sarvottam K, Yadav RK. Obesity‐related inflammation & cardio- vascular disease: efficacy of a yoga‐based lifestyle intervention.

Indian J Med Res. 2014;139:822‐834.

34. Ouchi N, Kihara S, Arita Y, et al. Adipocyte‐derived plasma pro- tein, adiponectin, suppresses lipid accumulation and class A scav- enger receptor expression in human monocyte‐derived macrophages.Circulation. 2001;103:1057‐1063.

35. Torigoe M, Matsui H, Ogawa Y, et al. Impact of the high‐mole- cular‐weight form of adiponectin on endothelial function in healthy young men.Clin Endocrinol (Oxf). 2007;67:276‐281.

36. Tan KC, Xu A, Chow WS, et al. Hypoadiponectinemia is associ- ated with impaired endothelium‐dependent vasodilation. J Clin Endocrinol Metab. 2004;89:765‐769.

37. Arnaiz P, Acevedo M, Barja S, et al. Adiponectin levels, car- diometabolic risk factors and markers of subclinical atherosclero- sis in children.Int J Cardiol. 2010;138:138‐144.

38. Pilz S, Horejsi R, Moller R, et al. Early atherosclerosis in obese juveniles is associated with low serum levels of adiponectin. J Clin Endocrinol Metab. 2005;90:4792‐4796.

39. Beauloye V, Zech F, Tran HT, Clapuyt P, Maes M, Brichard SM. Determinants of early atherosclerosis in obese children and adolescents.J Clin Endocrinol Metab. 2007;92:3025‐3032.

How to cite this article:Maggio ABR, Farpour- Lambert NJ, Aggoun Y, et al. Serum cardiovascular risk biomarkers in pre‐pubertal obese children. Eur J Clin Invest. 2018;48:e12995.https://doi.org/10.1111/

eci.12995