DISSERTATIONS | AAPO VEIJALAINEN | CLINICAL CORRELATES OF ARTERIAL STIFFNESS, TONE, AND... | No 381
uef.fi
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND Dissertations in Health Sciences
ISBN 978-952-61-2297-7 ISSN 1798-5706
Dissertations in Health Sciences
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND
AAPO VEIJALAINEN
CLINICAL CORRELATES OF ARTERIAL STIFFNESS, TONE, AND REACTIVITY IN CHILDREN
This dissertation focused on the reproducibility and associations of cardiovascular risk factors with arterial stiffness, tone, and reactivity. Study population
was 6-8 year old children from PANIC- study.
Arterial stiffness showed good reproducibility.
Clustering of cardiometabolic risk factors, low amount of physical activity, poor cardiorespiratory fitness, and adiposity were
associated with high arterial stiffness and partly with poor reactivity. Results support
need of early interventions.
AAPO VEIJALAINEN
Clinical correlates of arterial stiffness, tone,
and reactivity in children
AAPO VEIJALAINEN
Clinical correlates of arterial stiffness, tone, and reactivity in children
To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in Canthia CA100, Kuopio, on Friday, November 25th 2016,
at 12 noon
Publications of the University of Eastern Finland Dissertations in Health Sciences
Number 381
Physiology/ Institute of Biomedicine
Faculty of Health Sciences, University of Eastern Finland Kuopio
2016
Grano Oy Jyväskylä, 2016
Series Editors:
Professor Tomi Laitinen, M.D., Ph.D.
Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences
Professor Hannele Turunen, Ph.D.
Department of Nursing Science Faculty of Health Sciences
Professor Kai Kaarniranta, M.D., Ph.D.
Institute of Clinical Medicine, Ophthalmology Faculty of Health Sciences
Associate Professor (Tenure Track) Tarja Malm, Ph.D.
A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences
Lecturer Veli-Pekka Ranta, Ph.D. (pharmacy) School of Pharmacy
Faculty of Health Sciences
Distributor:
University of Eastern Finland Kuopio Campus Library
P.O.Box 1627 FI-70211 Kuopio, Finland http://www.uef.fi/kirjasto
ISBN (print): 978-952-61-2297-7 ISBN (pdf): 978-952-61-2298-4
ISSN (print): 1798-5706 ISSN (pdf): 1798-5714
ISSN-L: 1798-5706
III
Author’s address: Institute of Biomedicine, physiology/School of Medicine University of Eastern Finland
KUOPIO FINLAND
Supervisors: Professor Timo Lakka, M.D., Ph.D.
Institute of Biomedicine, physiology/School of Medicine University of Eastern Finland
KUOPIO FINLAND
Professor Tomi Laitinen, M.D., Ph.D.
Institute of Clinical Medicine/Clinical Physiology and Nuclear Medicine University of Eastern Finland
Kuopio University Hospital KUOPIO
FINLAND
Technical supervisor:
Tuomo Tompuri, M.D., M.Sc. (Health Sciences)
Institute of Biomedicine, physiology/School of Medicine University of Eastern Finland
KUOPIO FINLAND
Reviewers: Professor Päivi Korhonen, M.D., Ph.D.
Faculty of Medicine/General practice University of Turku
TURKU FINLAND
Adjunct professor Kari Kalliokoski, Ph.D., M.Sc. (Sport and Health Sciences) Faculty of Medicine/Turku PET Centre
University of Turku TURKU
FINLAND
Opponent: Professor Urho Kujala, M.D., Ph.D.
Department of Health Sciences University of Jyväskylä JYVÄSKYLÄ
FINLAND
IV
V
Veijalainen, Aapo
Clinical correlates of arterial stiffness, tone, and reactiv ity in children University of Eastern Finland, Faculty of Health Sciences
Publications of the University of Eastern Finland. Dissertations in Health Sciences 381. Year. 2016 p. 93.
ISBN (print): 978-952-61-2297-7 ISBN (pdf): 978-952-61-2298-4 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
ABSTRACT:
Cardiovascular diseases are a major clinical, public health, and economical problem. The roots of cardiovascular diseases are in childhood, when a healthy lifestyle is recommended for the prevention of these diseases. Arterial stiffness and impaired arterial reactivity can be considered early risk markers of cardiovascular diseases.
The purpose of this thesis was to study the reproducibility and clinical correlates of arterial stiffness, arterial tone, and arterial reactivity assessed by digital pulse contour analysis (PCA) and finger skin temperature (FST) before and after a maximal cycle ergometer exercise test.
Arterial stiffness was assessed using the stiffness index (SI) before exercise test, arterial tone using the reflection index (RI) and FST before exercise test, and arterial reactivity using the relative change in RI (RI∆%) and FST (FST∆%) during exercise.
In total 230 children 6-8 years of age who participated in the baseline examinations of the Physical Activity and Nutrition in Children (PANIC) study underwent PCA and comprised the subjects of this thesis. The reproducibility of SI, RI, and FST after 5-14 days were studied in 33-36 children. The associations of the metabolic syndrome score (MetS score) and its components with SI, RI, FST, RI∆%, and FST∆% were investigated in 173 children. The independent and combined associations of cardiorespiratory fitness (CRF, assessed using maximal workload during exercise test divided by lean body mass), physical activity (PA, assessed using a questionnaire), and body fat percentage (BF%, assessed using dual-energy X-ray absorptiometry) with SI, RI, and RI∆% were examined in 160 children.
The reproducibility of SI before the exercise test (coefficient of variation [CV] 6.3%, intraclass correlation [ICC] 0.438) and SI after the exercise test (CV 4.8%, ICC 0.548) was good.
RI after exercise test had a higher ICC (0.689) but also a higher CV (28.6%) than RI before exercise test (ICC 0.416, CV 18.7%). The reproducibility of FST after exercise test (CV 5.7%, ICC 0.509) was better than that of FST before exercise test (CV 10.0%, ICC 0.503). RI decreased 49.3% (p<0.001) and FSTincreased 17.8% (p<0.001) in response to the exercise test, whereas SI did not change significantly. MetS score (Partial correlation coefficient r=0.26, p=0.001), fasting insulin (r=0.24, p=0.002), fasting triglycerides (r=0.20, p=0.009), systolic blood pressure (r=0.24, p=0.002), and diastolic blood pressure (r=0.19, p=0.013) correlated positively with SI.
Poorer CRF (standardized regression coefficient β=-0.297, P<0.001), lower unstructured PA (β=-0.162, P=0.042), and higher BF% (β=0.176, P=0.044) were related to higher SI. However, only CRF remained a statistically significant correlate of SI when all these variables were entered simultaneously in the linear regression model. Poorer CRF was also related to a lower arterial reactivity (smaller RI∆%, β=0.316, P<0.001).
The reproducibility of SI is good, and cardiometabolic risk factors are associated with SI among children. SI could thus be used as a risk marker of cardiovascular diseases already in childhood. Long-term intervention studies are needed to investigate whether SI could be decreased by lifestyle modification among children.
National Library of Medicine Classification: WK 820, QT 256, QT 250, W G 120, WG 141, WG 550, WS 290 Medical Subject Headings: Cardiovascular Diseases; Vascular Stiffness; Exercise; Physical Fitness; Adipose Tissue; Risk Factors; Metabolic Syndrome X; Child; Finland; Skin Temperature; Arteries/physiology; Elasticity
VI
VII
Veijalainen, Aapo
Clinical correlates of arterial stiffness, tone, and reactivity in children University of Eastern Finland, Faculty of Health Sciences
Publications of the University of Eastern Finland. Dissertations in Health Sciences 381. 2016. 93 s.
ISBN (print): 978-952-61-2297-7 ISBN (pdf): 978-952-61-2298-4 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
TIIVISTELMÄ:
Sydän- ja verisuonisairaudet ovat merkittävä kliininen, kansanterveydellinen ja -
taloudellinen ongelma jotka alkavat kehittyä jo lapsuusiässä, jolloin myös ennaltaehkäisy tulisi aloittaa terveellisten elämäntapojen keinoin. Valtimoiden jäykkyyttä ja
reaktiivisuuden häiriötä pidetään näiden sairauksien riskitekijöinä.
Tavoitteena on tutkia valtimoiden jäykkyyden, supistustilan sekä reaktiivisuuden mittausten toistettavuutta sekä kliinisiä yhteyksiä käyttämällä sormen pulssiaaltoanalyysin (PCA) sekä lämpötilan (FST) mittauksia ennen ja jälkeen rasitustestin. Valtimojäykkyyttä mitattiin SI-indeksillä, supistumistilaa RI-indeksillä sekä FST:llä ennen rasitustestiä sekä reaktiivisuutta RI:n ja FST:n suhteellisella muutoksella rasituskokeessa (RI∆% ja FST∆%).
Lasten liikunta ja ravitsemus- tutkimuksen alkumittauksiin osallistuneilta 6-8 vuotiailta lapsilta 230:ltä mitattiin PCA. SI:n, RI:n ja FST:n toistettavuutta tutkittiin 33-36:lla lapsella 5- 14 päivää alkumittausten jälkeen. Metabolisen oireyhtymän summamuuttujan (MetS score) sekä osatekijöiden yhteyksiä SI:n, RI:n, FST:n, RI∆%:n ja FST∆%:n kanssa tutkittiin 173:lla lapsella. Kardiorespiratorisen kunnon (CRF, mitaten maksimaalisen suorituskyvyn suhdetta rasvattomaan massaan), fyysisen aktiivisuuden (PA, mitaten kyselylomakkeella) ja rasvaprosentin (BF%, mitaten kaksois-röntgen-absorptiometri-laitteella) itsenäisiä ja yhdysvaikutuksia SI:n, RI:n ja RI∆%:n kanssa tutkittiin 160 lapsella.
SI:n toistettavuus oli hyvä sekä ennen rasituskoetta (variaatiokerroin, CV 6.3%, toistettavuuskerroin, ICC 0.438) että sen jälkeen (CV 4.8%, ICC 0.548). RI:llä oli parempi ICC (0.689) mutta suurempi CV (28.6%) rasituskokeen jälkeen kuin ennen (ICC 0.416, CV%
18.7%). FST:n toistettavuus oli parempi jälkeen (CV% 5.7, ICC 0.509) kuin ennen rasituskoetta (CV 10.0%, ICC 0.503). RI laski 49.3% rasituskokeessa (p<0.001) ja FST nousi 17.8% (p<0.001).
Suurempi MetS score (osittaiskorrelaatiokerroin r=0.26, p=0.001), paastoinsuliini (r=0.24, p=0.002), paastotriglyseridi (r=0.20, p=0.009), systolinen verenpaine (r=0.24, p=0.002) ja diastolinen verenpaine (r=0.19, p=0.013) olivat yhteydessä suurempaan valtimojäykkyyteen.
Huonompi CRF (Stand. regressiokerroin β=-0.297, P<0.001), vähäisempi omatoiminen PA (β=-0.162, P=0.042), ja suurempi BF% (β=0.176, P=0.044) olivat yhteydessä suurempaan SI:hin.
Vain CRF:n vaikutus oli itsenäinen. Huonompi CRF oli yhteydessä vähäisempään valtimoiden reaktiivisuuteen (pienempi RI∆%, β=0.316, P<0.001).
Valtimojäykkyyden mittaus SI:n avulla on hyvin toistettava ja valtimoiden jäykkyys on yhteydessä haitallisiin kardiometabolisiin riskitekijöihin jo lapsilla. Pitkäaikaiset elämäntapainterventiotutkimukset toisivat lisätietoa jäykistymisen ehkäisystä.
Luokitus: WK 820, QT 256, QT 250, WG 120, WG 141, WG 550, WS 290
Yleinen Suomalainen asiasanasto: sydän- ja verisuonitaudit; metabolinen oireyhtymä; suorituskyky;
rasituskokeet; riskitekijät; fyysinen aktiivisuus; rasvaprosentti; lapset; Suomi
VIII
IX
Acknowledgements
This study was carried out at the Department of Biomedicine, University of Eastern Finland in 2009–2016 as a part of the Physical Activity and Nutrition in Children (PANIC) Study. I am grateful to all individuals who have contributed to this study. I wish to express my deepest gratitude especially to:
my prinicipal supervisor, Professor Timo Lakka, for his comprehensive guidance in writing manuscripts and sharing his expertise in physiology and epidemiology.
my second supervisor, Professor Tomi Laitinen, for challenging me all along the way to learn more about physiology and for his support.
my third supervisor, Tuomo Tompuri (specialist in Clinical Physiology and Nuclear Medicine), who introduced me to the PANIC study and in “the world of science” and supported me throughout this study.
pre-examiners, Professor Päivi Korhonen and adjunct professor Kari Kalliokoski, for their constructive comments and advice.
PANIC group for their always helpful attitude and all co-authors, Virpi Lindi, David Laaksonen, Eero Haapala, Niina Lintu, Jarmo Jääskeläinen, Juuso Väistö, Anna Viitasalo and Hanna-Maaria Lakka, for commenting and improving this study.
Finally, loving thanks to my parents and my lovely wife Reetta for support and encouragement. Special thanks also to my boys Nooa and Eeli for bringing joy in my days.
This study was financially supported by Saara Kuusisto and Salme Pennanen Foundation, Aarne and Aili Turunen Foundation, The Finnish Medical Foundation and University of Eastern Finland (PANIC Study).
Naantali, September 2016
Aapo Veijalainen
X
XI
List of the original publications
This dissertation is based on the following original publications:
I Veijalainen A, Tompuri T, Lakka H-M, Laitinen T and Lakka T A. Reproducibility of pulse contour analysis in children before and after maximal exercise stress test:
the Physical Activity and Nutrition in Children (PANIC) study. Clin Physiol Funct Imaging. 2011 31: 132-8.
II Veijalainen A, Tompuri T, Laitinen T, Lintu N, Viitasalo A, Laaksonen D E, Jääskeläinen J and Lakka T A. Metabolic risk factors are associated with stiffness index, reflection index and finger skin temperature in children--Physical Activity and Nutrition in Children (PANIC) study. Circ J. 2013 77: 1281-8.
III Veijalainen A, Tompuri T, Haapala E A, Viitasalo A, Lintu N, Väistö J, Laitinen T, Lindi V and Lakka T A. Associations of cardiorespiratory fitness, physical
activity, and adiposity with arterial stiffness in children. Scand J Med Sci Sports.
2015 26: 943-950.
The publications were adapted with the permission of the copyright owners.
XII
XIII
1 INTRODUCTION ... 1
2 REVIEW OF THE LITERATURE ... 3
2.1 BASICS OF ARTERIAL STRUCTURE AND FUNCTION ... 3
2.2. PATHOPHYSIOLOGY OF ARTERIOSCLEROSIS AND ATHEROSLECROSIS ... 4
2.2.1 Arterial stiffness... 5
2.2.2 Endothelial dysfunction ... 6
2.3 NON-INVASIVE ASSESSMENT OF ARTERIAL STIFFNESS... 6
2.3.1 Pulse wave velocity ... 7
2.3.2 Digital pulse contour analysis... 7
2.3.3 Systolic pulse contour analysis ... 8
2.3.4 Diastolic pulse contour analysis ... 9
2.3.5 Aortic distensibility ... 9
2.3.6 Carotid artery distensibility ... 9
2.3.7 Summary and reproducibility of different non-invasive methods assessing arterial stiffness ... 9
2.4 NON-INVASIVE ASSESSMENTS OF ENDOTHELIAL FUNCTION... 10
2.4.1 Brachial artery flow-mediated dilation ... 10
2.4.2 Pulse contour analysis... 10
2.4.3 Reactive hyperemia peripheral artery tonometry ... 10
2.4.4 Digital thermal monitoring ... 11
2.4.5. Summary and reproducibility of different non-invasive methods assessing endothelial function ... 11
2.5 NONINVASIVE ASSESSMENT OF ARTERY WALL THICKNESS... 15
2.6 RISK FACTORS FOR ARTERIAL STIFFNESS AND ENDOTHELIAL DYSFUNCTION IN CHILDHOOD ... 15
2.6.1 Traditional cardiovascular risk factors ... 15
2.6.1.1 Overweight and obesity ... 15
2.6.1.2 Hypercholesterolemia and dyslipidemia... 16
2.6.1.3 Elevated blood pressure ... 16
2.6.1.4 Gender ... 17
2.6.2 Clustering of cardiometabolic risk factors... 19
2.6.3 Life-style related risk factors ... 20
2.6.3.1 Physical activity, sedentary behaviour, and physical fitness... 20
2.6.3.2 Diet ... 21
2.6.3.3 Smoking... 22
2.6.4 Genetic factors ... 23
2.6.5 Fetal and early childhood development ... 24
3 AIMS OF THE STUDY... 29
XIV
4 METHODS ... 31
4.1. STUDY DESIGN AND POPULATION ... 31
4.1.1 Physical Activity and Nutrition in Children Study ... 31
4.1.2 Study I... 31
4.1.3 Study II ... 31
4.1.4 Study III ... 32
4.2. ASSESSMENTS ... 32
4.2.1 Assessment of stiffness index, reflection index, and finger skin temperature... 32
4.2.2 Assessment of physical activity and sedentary behaviour... 33
4.2.3 Assessment of cardiorespiratory fitness ... 33
4.2.4 Assessment of body composition ... 33
4.2.5 Assessment of clustering of cardiometabolic risk factors and metabolic syndrome ... 34
4.2.6 Assessments of confounding factors ... 35
4.2.7 Statistical methods ... 35
5 RESULTS ... 39
5.1 REPRODUCIBILITY OF STIFFNESS INDEX, REFLECTION INDEX, AND FINGER SKIN TEMPERATURE BEFORE AND AFTER EXERCISE TEST ... 39
5.1.1 Basic characteristics and exercise stress test variables of subjects... 39
5.1.2 SI, RI, and FST before and after the exercise test ... 40
5.1.3 Reproducibility of SI, RI, and FST before and after the exercise test ... 41
5.2 ASSOCIATIONS OF CARDIOMETABOLIC RISK FACTORS WITH ARTERIAL STIFFNESS AND REACTIVITY ... 43
5.2.1 Characteristics of children ... 43
5.2.2 SI, RI, and FST before and after the exercise test ... 44
5.2.3 Associations of arterial stiffness, arterial tone, and arterial reactivity with cardiometabolic risk factors... 44
5.3 ASSOCIATIONS OF CARDIORESPIRATORY FITNESS, PHYSICAL ACTIVITY, AND BODY FAT PERCENTAGE WITH ARTERIAL STIFFNESS AND ARTERIAL REACTIVITY ... 47
5.3.1 Characteristics of children ... 47
5.3.2 Determinants of stiffness index ... 47
5.3.3 Determinants of reflection index ... 47
5.3.4 Determinants of percentage change in reflection index during exercise test ... 48
5.3.5 Interactions between cardiorespiratory fitness, unstructured physical activity, adiposity, stiffness index, and change in reflection index during exercise test ... 49
6 DISCUSSION ... 51
6.1 METHODOLOGICAL STRENGTHS AND LIMITATIONS ... 51
6.1.1 Study population and study design ... 51
6.1.2 Assessment of arterial stiffness, arterial tone, and arterial reactivity ... 51
6.1.3 Assessment of cardiometabolic risk factors ... 52
XV
6.1.4 Assessment of cardiorespiratory fitness ... 52
6.1.5 Assessment of physical activity ... 53
6.2 MAIN FINDINGS ... 53
6.2.1 Reproducibility of assessments of arterial stiffness and reactivity before and after the exercise test (Study I and II) ... 53
6.2.2 Association of arterial stiffness and arterial reactivity with cardiometabolic risk factors (Study II)... 55
6.2.3. Association of arterial stiffness and arterial reactivity with cardiorespiratory fitness, physical activity, sedentary behaviour, and body fat percentage (Study III) ... 56
7 CONCLUSIONS ... 61
7.1 CLINICAL IMPLICATIONS ... 61
7.2 FUTURE PERSPECTIVES... 61
REFERENCES... 63
XVI
Abbreviations
Aix augmentation index BF% body fat percentage BP blood pressure BMI body mass index IMT intima-media thickness CRF cardiorespiratory fitness CV% coefficient of variation DBP diastolic blood pressure FMD flow-mediated dilation FST finger skin temperature FST∆ absolute change in FST during the exercise test FST∆% percentage change in FST during the exercise test HDL high-density lipoprotein hs-CRP high-sensitivy C-reactive protein
ICC intraclass correlation coefficient IDF International Diabetes Federation
LDL low-density lipoprotein LM lean body mass
MetS metabolic syndrome NCEP National Cholesterol Education Program NO nitric oxide
PA physical activity
PANIC Physical Activity and Nutrition in Children
PCA pulse contour analysis PWV pulse wave velocity
RH-PAT reactive hyperemia peripheral artery tonometry
RI reflection index
RI∆ absolute change in RI during the exercise test
RI∆% percentage change in RI during the exercise test
SBP systolic blood pressure SD standard deviation SI stiffness index WC waist circumference
1 Introduction
Cardiovascular diseases are the leading cause of death globally, responsible for 17.3 million deaths per year, 10-18% of loss of disability-adjusted life years and costs of almost one trillion per year (1,2). In the European Union cardiovascular diseases cause 1.5 million deaths per year and cost 169 billion euros annually (3).
Arterial stiffness (4–6) and endothelial dysfunction (7) have been recognized as pathophysiological mechanisms for cardiovascular diseases and have been associated with cardiovascular risk factors, such as physical inactivity, obesity, insulin resistance, dyslipidemia, and metabolic syndrome already in childhood (8–12). Arterial stiffness and endothelial dysfunction can be assessed non-invasively and are thus considered suitable measures for cardiovascular risk in children (13). There are many different methods for the assessment of arterial stiffness and endothelial dysfunction (14,15). One of these methods is the digital pulse contour analysis (PCA), where operator-independent measurement can be done easily from the fingertip, without special training, and with a cheap device (16).
The development of cardiovascular diseases begins in childhood, and so should also the prevention of these diseases by adapting healthy lifestyle (13). The American Heart Association stated over a decade ago that the prevention of cardiovascular diseases should begin already in childhood (17). This guideline gives detailed health promotion goals and recommendations for physical activity, diet, and smoking. The definition of childhood ideal cardiovascular health has also been proposed (18). Although health information is given to parents in child health clinics, and to children and adolescents in school health care and health education and physical education, biology, and home economics lessons in Finland (19), adherence to the guidelines for the prevention of cardiovascular diseases is poor (20).
Early measurable changes in arterial structure and function may identify children at increased risk for future cardiovascular disease (13). Objective measures of increased cardiovascular risk may increase motivation to improve lifestyle and benefit intervention.
However, the measurement of arterial stiffness or endothelial function are not in used in clinical practice. Guidelines by the European Society of Hypertension from 2007 recommended measuring arterial stiffness to predict risk for subclinical organ damage in adults with hypertension (21). A recent statement from the American Heart Association also stated that “it is reasonable to measure arterial stiffness to provide incremental information beyond standard cardiovascular risk factors in the prediction of future cardiovascular events” (22). The statement also highlighted the need for normative and longitudinal data and the validation of methods for the assessment of arterial stiffness in children (22).
Further studies on the noninvasive assessment and health implications of arterial stiffness and endothelial function in children are needed. This thesis aims to provide a thorough literature review on the pathogenesis and assessment of arterial stiffness and endothelial dysfunction and risk factors for increased arterial stiffness and endothelia l dysfunction in children. The thesis also aimed to provide novel information on the
2
reproducibility of measures of arterial stiffness, tone, and reactivity obtained from the PCA and finger skin temperature (FST) measurements and the associations of physical activity, sedentary behaviour, cardiorespiratory fitness, body fat content, and clustering of cardiometabolic risk factors with arterial stiffness, tone, and reactivity in 6-8-year-old Finnish children.
3
2 Review of the literature
2.1 BASICS OF ARTERIAL STRUCTURE AND FUNCTION
Artery walls consist of three different layers: the innermost is called the tunica intima, the outermost is the tunica adventitia and between the intima and adventitia is the tunica media. Based on their different structures and functions, arteries can be divided into elastic and muscular arteries (23).
The arterial tree in the systemic circulation begins from the left ventricle of the heart as the aorta, the largest elastic artery, where the media is formed by alternating elastic laminae and circularly arranged smooth muscle cells (23,24). In the thoracic aorta, up to 40% of the artery wall consist of these elastic fibers (25).
When traveling from the aorta to the more distal arteries, the elasticity of the arteries decreases as the proportion of the elastin fibers to the stiffer collagen fibers decreases (25,26).
The proportion of smooth muscle cells also increases in the more distal arteries, thus called muscular arteries (23). The muscular arteries regulate blood flow and blood pressure (BP) by increasing and decreasing their diameter to fulfil the needs of organs and tissues (27,28).
Peripheral arteries in organs and tissues, called as arterioles, have a diameter of less than 0.15 millimeters, lack elastic laminae and consist only of a thin layer of smooth muscle cells.
Vessels distally from the arterioles are called capillaries which consist only of tunica intima (23).
The tunica intima consists of a single layer of squamous endothelial cells which has many critical functions in maintaining normal hemodynamics (23). Endothelial cells are responsible for modulating the tone of smooth muscle and maintaining a non-adhesive luminal surface. They also have antithrombotic, anticoagulant, and fibrinolytic properties (29).
When the left ventricle of the heart pumps blood to the aorta in systole, the increased intraluminal pressure distends the elastic wall of the aorta. Only up to 40% of stroke volume during systole is forwarded directly to the peripheral arteries, whereas the rest of stroke volume distends the elastic artery walls. This distension is released in diastole thus buffering the pressure generated by the left ventricle during systole and the loss of pressure during diastole (30,31). These changes in the aortic wall during cardiac cycle transform the intermittent blood flow to a more steady blood flow in more distal elastic arteries (32).
The pressure pulse generated in systole travels forward as a wave and is influenced naturally by the surroundings and distance of its journey. Because arteries become stiffer along the way in the arterial tree, the strength of the pressure pulse becomes dampened and the pulse wave changes in shape. When the propagating wave encounters any change in impedance, such as in branch points or points where lumen diameter or arterial material type changes, part of it reflects backwards. This physiological phenomenon is ideal to boost diastolic blood flow in the aorta and coronary arteries (33). The measured morphology of
4
the pressure wave is a summation of the wave going forward and the wave reflected (27,30,34).
2.2. PATHOPHYSIOLOGY OF ARTERIOSCLEROSIS AND ATHEROSLECROSIS
Arterial aging could be described as a physiological process because throughout species, including the human, remodeling of arteries as a function of age is seen (35). Arteries become stiffer by thickening of intima and media layers and also more dilated both in luminal and overall, thus hindering the optimal coupling between the heart and the arteries (35–37). The arteries begin to lose their distensibility already after birth and this seems to be related to changes in vessel wall (22,38,39).
The remodeling of the elastin and collagen ratio due to aging, where stiffer collagen fibers gradually become more responsible for bearing mechanical load than elastin, is explained by age-associated fractures in elastin caused by repeated cyclic stress (40,41). There are also differencies in the chemical profiles of aging arteries (41). Overactivation of different matrix metalloproteases decreases the elastin to collagen ratio (42). Accumulation of advanced glycation end products (43) and calcium (44) also stiffens the arteries. The term age- associated arterial secretory phenotype presents the complex secretory phenotype where overexpression of angiotensin II signaling cascade plays a major role in the cascade that increases arterial aging (35). The endothelial cells initiate the remodeling of arteries by observing blood flow and pressure inside the artery lumen and converts this message further to the smooth muscle cells and fibroblast cells (45).
Sympathetic nervous system (46) and endothelium-derived nitric oxide (NO) are also involved in arterial stiffening (47), as they control the vascular tone via smooth muscle cells and increased tone increases arterial stiffness (48).
In general, these aforementioned changes involve adventitia and media and are characterized by diffuse breakdown of elastin, increased amount of collagen and matrix deposition, and vascular smooth muscle cell hypertrophy. These changes cause arterial stiffening and can be called as arteriosclerosis (34,40). The pathogenesis of arterioscler osis shares many features with the pathogenesis of atherosclerosis that is characterized by activation of inflammatory cells and formation of oxidized radicals (42,49). However, it is important to distinguish arteriosclerosis from atherosclerosis. Atherosclerosis is a focal and occlusive disease starting in the intima and causing ischaemic consequences with plaques clinically seen in coronary, carotid, renal, and lower limb arteries. Arteriosclerosis in turn is a diffuse disease involving adventitia and media in the whole vasculature, is dilatory in nature, increases left ventricular workload, and hinders coupling between the heart and the arteries (34,50).
Atherosclerosis, although it is more a stenotic disease in the lumen of the artery, also begins in artery wall (50–55). The initial step is dysfunctional endothelium, causing it to pathologically overexpress many substances that enhance monocyte intake. Once monocytes have been taken into the artery wall, they differentiate into macrophages. Also because of the injury of endothelium, it leaks low-density lipoprotein (LDL) particles into
5
the intimal layer of artery wall, where they become oxidized which makes them highly atherogenic by attracting more monocytes and impairing endothelium-depen dent relaxation. Oxidized LDL then interacts with macrophages to make foam cells. The accumulation of foam cells produces macroscopically observable fatty streaks, which can be considered, perhaps because of being visible, the first step of atherosclerosis. These cholesterol-filled cells releases part of their cholesterol back to vasculature in high-density lipoprotein (HDL) -particles. However, the accumulation of foam cells and secreted cytokines cause the smooth muscle from the medial layer of arterial wall cells to migrate into the intimal layer of artery wall where they proliferate and secrete extracellular matrix proteins, thus forming the next stage of the disease, the fibrous plaques. This formation involves plenty of cellular and humoral responses of mahcrophages and T cells mimicking chronic inflammation. The formed plaque continues growing, thus narrowing the vessel lumen. Inside in this growing plaque, the apoptosis of the smooth muscle cell and macrophages starts to occur. This weakens the plaque, which may become unstable prone to rupture of the plaque, causing potentially lethal thrombosis (50).
The progression of atherosclerotic disease from fatty streaks to plaque formation is slow and begins in childhood (17,56). Fatty streaks could be seen even in infants and fibrous plaque with increasing prevalence after the age of 15 and with an increasing severity with an increase in cardiovascular risk factors (57). Increasing in arterial stiffness in childhood with an increase in cardiovascular risk factors is also seen, but the association between childhood arterial stiffness and adulthood consequences is not yet fully determined (22).
2.2.1 Arterial stiffness
Healthy arteries are vital for the optimal utilization of the pressure generated by heart. As discussed earlier, a significant amount of this pressure is spent on dilating the aorta. The amount of pressure needed to dilate the aorta (the efficiency of heart-vascular coupling) is dependent on arterial wall viscoelastic properties, and this work should naturally be minimized to avoid the overwork of heart. This viscoelastic property is called the distensibility of the artery, however more often considering about its reciprocal, the stiffness of the artery. Arterial stiffness is expressed as the relationship of needed pressure to change in volume (30,58):
𝐴𝑟𝑡𝑒𝑟𝑖𝑎𝑙 𝑠𝑡𝑖𝑓𝑓𝑛𝑒𝑠𝑠 =𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒
𝑉𝑜𝑙𝑢𝑚𝑒 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒 (Formula 1)
When considering arterial stiffness, one must take into account the whole tube-like arterial system and the pulse wave phenomenon (instead of only considering the pump effect of heart and cushioning effect of the aorta): when the pressure generated by heart travels as a wave forward, part of this pressure comes back as an echoed wave (31,33,59). In stiffer arteries, these reflected waves come earlier and thus effecting also upstream in the vascular tree (33).
As the arteries become stiffer, the reduction in the elasticity of the aorta increases early systolic blood pressure (SBP) and earlier wave reflection increases late SBP (instead of diastolic boost) (60). This increases the SBP and pulse pressure, which has many adverse
6
effects on the heart. The increase in afterload results in growing demand in pump function, and the lack of diastolic enhancement results in limited coronary perfusion. This can cause or aggravates congestive heart failure, subendocardial ischemia, myocardial hypertrophy, the impairment of diastolic myocardial function, and left ventricular ejection (61–66).
Furthermore, the stiffening of the arteries accelerates because of damage due higher SBP and pulse pressure (59). Arterial stiffening is also associated independently with coronary heart disease and stroke in the healthy population and all-cause and cardiovascular mortality in hypertensive patients (67,68).
The stiffening of arteries due to age is not uniform in the arterial tree, but is more evident in central arteries than peripheral arteries (69). After 60 years of age central artery stiffness starts to exceed peripheral artery stiffness, again hindering the normal physiology and allowing harmful (direct) pressure waves to be higher in the periphery, where it can cause damage to end organs such as the kidney and brain (69,70). Arterial stiffness has been associated with peripheral arterial disease (71), stroke (67), and advanced kidney disease (72). This non-uniform distribution of stiffening could be due to a different amount of stretch-caused stress in central versus peripheral arteries (36).
2.2.2 Endothelial dysfunction
The key element in endothelial dysfunction is the impairment of endothelium-depen dent vasodilation, which occurs due to reduced bioavailability of vasodilators, most importantly NO, and increased production of endothelium-derived vasoconstrictor factors (7,73). NO also contributes to platelet adhesion and aggregation, thrombogenicity, and cell proliferation linking it to the early stage of atherosclerosis (74–78). Excessive production of reactive oxygens species, which are by-products of aerobic cell metabolism, could reduce NO bioavailability and is considered a major mechanism behind endothelial dysfunction (79,80).
Impaired endothelium function may play a key role in atherosclerosis, as it has been shown that preserving normal endothelium function was protective against developing an increase in carotid intima-media thickness (cIMT, a measure of structural atherosclerosis) (81,82) due to the exposure of cardiovascular risk factors (83). It also plays an important role in arterial stiffening (47,84).
Endothelial dysfunction can also be considered a systemic disease (85). However, the endothelium is not uniform through the vasculature, but instead differs in function and structure to serve its needs in different vascular beds (86,87).
2.3 NON-INVASIVE ASSESSMENT OF ARTERIAL STIFFNESS
The pulse wave, manifested as artery wall movement, is a complex physiologica l phenomenon observed in the blood circulation, as discussed earlier. As the pulse wave travels through the arteries, changes in pressure (pressure pulse), blood flow velocity (flow pulse), and transverse profile (volume pulse) can be observed and arterial stiffness measured (88) (Figure 1). The pulse wave varies in different parts of circulation (89), and it is physiologically dependent on the age and BP (90).
7
2.3.1 Pulse wave velocity
The pulse wave velocity (PWV) between the carotid and femoral artery is described as the
´gold standard´ measurement of arterial stiffness (14). It is typically measured with a pressure transducer or applanation tonometry, but also ultrasound and impedance are used (91,92). The Moens-Korteweg equation relates the velocity of pulse wave travel in a vessel to the distensibility of that vessel (93):
𝑐₀ = √2 𝑅𝜌𝐸ℎ (Formula 2)
where c₀ = wave speed, E = Young’s modulus in circumferential direction, h = wall thickness, R = artery radius at the end of diastole, and 𝜌 = density of fluid.
With PWV between carotid and femoral artery, we can only measure large artery segments, with no insight into the status of smaller blood vessels. Examining the smaller arteries has been proposed to allow us to identify diseases earlier (94) because their behaviour has an influence on larger arteries (95).
Figure 1. Simplified illustration of movement of pulse wave and its different and measurable effect to artery.
2.3.2 Digital pulse contour analysis
The central pressure waveform can be estimated from the radial pressure pulse (96,97). The contour of the radial pressure pulse is similar to the contour of the reconstructed digital volume pulse (98). The digital volume pulse provides estimate for arterial stiffness as discussed below and can be easily measured by means of photoplethysmography, which measures the transmission of infra-red light through the finger pulp (16).
The contour of the digital volume pulse is formed as a result of a complex interaction between the function of the left ventricle and systemic circulation (89) and consists of the
8
systolic component (first peak, a) and the diastolic component (second peak, b) or the point of inflection (Figure 6, page 41). The first peak is due to a systolic pressure wave starting from the left ventricle and ending to the finger. The second peak is caused mainly by reflected pressure waves thought to originate from the lower body: waves travel from the aorta to the small arteries in the lower body, where they reflect back via the aorta to the fingers. Peak-to-peak time is the time between the first and the second peak and is related to the time where the pressure waves travel from the subclavian artery to the aortic bifurcation and correlated well with measures done with ultrasound (r=0.75, p<0.0001) (99).
When dividing this estimate for time which is consumed in the movement of central pulse wave from one place to another with the subject height, an estimate for central pulse wave velocity is achieved. This estimate is called stiffness index (SI) and it is calculated by dividing the height of the subject (in meters) by peak-to-peak time (in seconds) and is influenced predominantly by the stiffness of the large elastic arteries. The correlation between SI and PWV between carotid and femoral artery is strong (r=0.65, p<0.0001), they are both independently influenced by similar factors (age and BP), and behave in a similar manner to vasoactive drugs (with no great influence) (90). Sensitivity (87%) and specificity (87%) to detect older subject with angiographically confirmed coronary heart disease (compared with younger without coronary heart disease as identified with no history of coronary heart disease or risk factors for cardiovascular disease) is reported to be good (100).
As it is an assessment done in the periphery, it could not provide measures identical to central PWV, and could thus be confounded by e.g. characteristics of ventricular ejection and the sites of wave reflections (90). Another study also found a moderate correlation between SI and PWV (r=0.55, p<0.0001) but in Bland-Altman plotting the consistency with PWV was worse (101). In a recent study, where SI was measured in a model of an artery network, SI was affected by changes in all conduit arteries, whereas central PWV was affected only by central stiffness (102).
The strong association of SI and the augmentation index (AIx) (r=0.80, p<0.001) and moderate association of SI and the small artery elastic index (r=-0.65 p<0,001) has also be reported (103). Small artery elastic index (and AIx) are suggested to be dependent upon small artery compliance, implying that SI is also dependent on that. This is supported by the fact that wave reflection does not appear in central but in distal arteries (99,103). It may therefore identify diseases earlier than assessment of PWV (94).
2.3.3 Systolic pulse contour analysis
The systolic part of the radial artery pressure pulse wave is estimated with applanation tonometry and calibrated with brachial BP. It estimates the central (aortic) pressure pulse wave with a special, mathematically generated general transfer function. Central arterial stiffness is then estimated from this wave by calculating the difference between the second (reflected wave) and the first (direct) systolic peaks and expressed as a percentage of central pulse pressure and called as AIx (104,105). It correlates weakly with PWV (r=0.29, p<0.005) (106), and changes differently than PWV after vasoconstrictive stimulation (107,108), according to age (109), and according to the effect of heart rate and BP, suggesting different pathophysiological mechanisms behind these measurements, as it is affected of wave reflection (106,110).
9
2.3.4 Diastolic pulse contour analysis
The diastolic part of radial artery pressure pulse wave, as estimated with semiautomatic tonometry calibrated by oscillatory BP, is used to assess large and small arterial stiffness by means of a special algorithm and modified Windkessel model (111). It provides indices called small artery elastic index and large artery elastic index. This technique does not consider the reflection of waves and is oversimplified with an assumption that all segments of small and large arteries are pressurized simultaneously (112).
2.3.5 Aortic distensibility
Cardiac magnetic resonance imaging provides a novel method to assess arterial stiffness. A magnetic resonance imaging device is used to obtain images from the ascending aorta, the proximal ascending aorta (at the site of the crossing of the pulmonary artery), and the distal descending aorta (level of 10cm below the diaphragm). Arterial distensibility is then calculated by dividing the difference of aortic area at the end of the systole and diastole by product of aortic area at the end of diastole and brachial pulse pressure (113). It is based on the Bramwell and Hill equation whis has been derived from the Moens-Korteweg equation (Formula 2) (114):
𝑃𝑊𝑉 = 1
√𝑏𝑙𝑜𝑜𝑑 𝑑𝑒𝑛𝑠𝑖𝑡𝑦 ×𝑑𝑖𝑠𝑡𝑒𝑛𝑠𝑖𝑏𝑖𝑙𝑖𝑡𝑦 (Formula 3)
PWV correlates strongly with values provided by cardiac magnetic resonance imaging technique in different regions of the arterial tree (correlation coefficient ranged 0.57-0.72 (p<0.01) (113).
2.3.6 Carotid artery distensibility
Local arterial stiffness can also be measured in the common carotid artery by ultrasound.
Analogically to the distensibility of the aorta, the change in carotid artery diameter during the heart cycle is measured with ultrasound and related to pulse pressure achieved with brachial BP measurements (92). It can be converted to carotid stiffness with the Moens- Korteweg equation. Carotid artery stiffness correlates weakly with aortic PWV (r=0.34 (p<0.001) (115), and the correlation weakens when the number of cardiovascular risk factors increased (92,115).
2.3.7 Summary and reproducibility of different non-invasive methods assessing arterial stiffness
The PWV between carotid and femoral artery is the “gold standard” for measuring arterial stiffness. Assessment of PWV in clinical settings has been encouraged (116), more recently by American Heart Association (22), as supported also by a recent meta-analysis of 17 635 subjects, where it remained a predictor of coronary heart disease, stroke, and cardiovascular events after adjusting for conventional risk factors (117). Measurements such as SI and AIx are a reasonable surrogate for PWV because of their ease of use but they might reflect different pathophysiological states. On the other hand, this could also be their advantage.
Measures of diastolic pulse contour analysis have lesser agreement with PWV (100). Cardiac magnetic resonance imaging technique is still a novel method and is expensive and often
10
unavailable. Measurements of arterial stiffness can also be done locally from a carotid artery by measuring its distensibility or using a novel approach of measuring longitudinal motion of the common carotid artery by ultrasound (118). These measurements are technically challenging and carotid and aortic stiffness cannot, however, be considered identical as aorta seems to be more vulnerable for stiffening due hypertension and type 2 diabetes than carotid artery (14,115,119).
Reproducibility of assessments of arterial stiffness is presented in Table 1 and seems to be in order of cardiac magnetic resonance imaging technique, PWV, SI/AIx, small artery elastic index, large artery elastic index, and carotid artery distensibility. The cardiac magnetic resonance imaging technique showing the best reproducibility with a coefficient of variation 1-2%. However, reproducibility studies are quite scarce, and methods assessing of reproducibility differed, making comparison harder. Also results differed within age groups and study population again hindering comparison.
2.4 NON-INVASIVE ASSESSMENTS OF ENDOTHELIAL FUNCTION
2.4.1 Brachial artery flow-mediated dilation
Endothelium function was first assessed non-invasively with a high-resolution ultrasound by Celermajer et al. in 1992 (78). It is based on the fact that increasing flow in artery causes the dilation of the vessel via endothelium-released NO (120–122), and that dilation is diminished, or even reversed, in endothelial dysfunction (123–125). This increase in flow is induced by reactive hyperemia after releasing a pneumatic cuff. The reported result is the proportion of observed dilation with baseline measurements. It is often compared with the vasodilation caused by an endothelium-independent vasodilator such as glyceryl trinitrate.
It is measured often in brachial artery and is thus called brachial artery flow-mediat ed dilation (FMD).
2.4.2 Pulse contour analysis
Endothelial function can also be assessed by means of digital PCA with the use of the reflection index (RI). RI is calculated as the percentage of the height of the reflected wave of the height of the direct wave (Figure 6, page 41). The common way is to study the effect of vasoactive drug (inhaled salbutamol) on RI, which has been reported showing blunted endothelium-dependent vasodilation in those with endothelial dysfunction (16,126).
Change in RI due to vasoactive drug has shown moderate correlation with FMD (r=0.44, p<0.01) and 88% sensitivity together with 79% specificity to detect endothelial dysfunction (126).
However, local vasodilator showed no effect to RI assessed from periphery. RI is also independent of arterial stiffness (99).
2.4.3 Reactive hyperemia peripheral artery tonometry
Reactive hyperemia peripheral artery tonometry (RH-PAT) is analogical to the measurement of FMD. It measures the amplitude of a pulse wave, achieved with a plethysmograph placed around finger, before and after cuff-induced reactive hyperemia (127). It correlated moderately with FMD (r=0.55, p<0.0001) (127).
11
2.4.4 Digital thermal monitoring
Finger skin temperature (FST) can be assessed with thermal probes and a lower increase in FST after cuff-induced reactive hyperemia was seen in persons with coronary heart disease than in those without coronary heart disease. This has been suggested to represent endothelial dysfunction (128). Changes in FST correlated with FMD (correlation coefficient 0.45, p<0.01) (129).
2.4.5. Summary and reproducibility of different non-invasive methods assessing endothelial function
FMD has robust a foothold in the field of measuring endothelial function, but it requires a trained physician and an accurate (and expensive) ultrasound device (130). Alternative methods such as digital PCA and RH-PAT are therefore tempting because they are operator- independent. Measurement of RH-PAT has shown to be more reproducible than measurements with digital PCA (Table 2). Novel methods such as the measurement of FST after occlusion should be used with certain caveats as it is confounded by body thermo- balance (131). However the measurements of these microcirculatory reactions by RH-PAT and FST could provide information on pathological states different from measurements in a conduit artery, such the brachial artery (132). For example, microvascular dysfunction may better reflect metabolic abnormalities (such metabolic syndrome) (133) and brachial FMD coronary heart disease (134). Thus, the overall “winner” cannot be concluded (15,135).
12
Table 1. Reproducibilityof measurementsof arterial stiffness.
StudyCohortAge (mean (SD or range) nDevice and MethodIntra-class correlationcoefficiency (95%CI) Coefficient of variation (%) Correlationcoefficient / Other
Pulse Wave Velocity between carotid and femoral artery(136) Pop75.7 (4.6)79VP-1000 Plus (app tonometry)0.70 (0.59-0.81)(137) Pop44.4 (19.9)60Complior (mechanotransducer)0.94-0.983.6-4.8%
(138) Pop(male) 30.9 (9.0)8Sphygmocor (app tonometry+ECG) Intra-/interstudy 4.5-6.9%
(91) Het21-8210Ultrasound+ECGIntra-/Interobserver 0.97-0.98 (0.89-0.99)
(139) Het 51 (20) 68Pulsepen (tonometry+ECG) Intra-/Interobserver 7.2-7.94%(101) Het 49.3(19.6)50Complior (mechanotransducer)12.40%(101) Het 49.3(19.6)50Pulsepen12.30%(90) Healthy(male) 22-518Sphygmocor 8.80%
(140) Healthy 11.4 (8.4-14.8) 79PulseTrace PWV (ultrasound+ECG) 7.40%
Stiffness Index(90) Healthy(male) 22-518Pulse Trace PCA29.60%
(101) Het 49.3 (19.6)50Pulse Trace PCA214.50%
(140) Healthy 11.4 (8.4-14.8) 79Pulse Trace PCA24.00%(Table 1continues)
