The reliability of a single protocol to determine endothelial, microvascular and
autonomic function in adolescents
Bert Bond PhD, Craig A. Williams PhD and Alan R. Barker PhD
Children’s Health and Exercise Research Centre, Sport and Health Sciences, College of
Life and Environmental Sciences, University of Exeter, Exeter, EX1 2LU.
Corresponding author:
Dr Alan R. Barker
Children’s Health and Exercise Research Centre
Sport and Health Sciences
College of Life and Environmental Sciences
University of Exeter
St Luke's Campus
Exeter
EX1 2LU
Tel: 44 (0)1392 722766
Email: [email protected]
Short Title: Reliability of novel CVD risk factors in adolescents
Word count: 3704
1
Background: Impairments in macrovascular, microvascular and autonomic function are
present in asymptomatic youths with clustered cardiovascular disease risk factors. This
study determines the within- and between-day reliability of a single protocol to noninvasively assess these outcomes in adolescents. Methods: Forty 12-15 year old
adolescents (20 boys) visited the laboratory in a fasted state on two occasions,
approximately one week apart. One hour after a standardised cereal breakfast,
macrovascular function was determined via flow mediated dilation (FMD). Heart rate
variability (root mean square of successive R-R intervals; RMSSD) was determined
from the ECG-gated ultrasound images acquired during the FMD protocol prior to cuff
occlusion. Microvascular function was simultaneously quantified as the peak (PRH) and
total (TRH) hyperaemic response to occlusion in the cutaneous circulation of the
forearm via laser Doppler imaging. To address within-day reliability, a subset of twenty
adolescents (10 boys) repeated these measures 90 minutes afterwards on one occasion.
Results: The within- and between-day typical error expressed as a coefficient of
variation of these outcomes is as follows: ratio-scaled FMD, 5.1 and 10.6%;
allometrically-scaled FMD, 4.4 and 9.4%; PRH, 11 and 13.3%; TRH, 29.9 and 23.1%;
and RMSSD, 17.6 and 17.6%. The within- and between-day test-retest correlation
coefficients for these outcomes were all significant (r>0.54 for all). Conclusion:
Macrovascular, microvascular and autonomic function can be simultaneously and noninvasively determined in adolescents using a single protocol with an appropriate degree
of reproducibility. Determining these outcomes may provide greater understanding of
the progression of cardiovascular disease and aid early intervention.
Key Words: Flow mediated dilation (FMD), reactive hyperaemia, heart rate variability,
novel CVD risk factors, young people.
2
INTRODUCTION
Whilst overt cardiovascular disease (CVD) is not apparent until later life, the underlying
atherosclerotic process originates in childhood (Stary 1989). Accordingly, the ability to
non-invasively determine the earliest manifestations of this disease in paediatric groups
is likely important for the primary prevention of CVD. An impairment in endothelial
function is a sentinel event in the sequelae of atherosclerosis and may be a prerequisite
for structural changes to the vessel wall (Juonala et al. 2004). Endothelial function can
be determined via flow mediated dilation (FMD), which uses high resolution ultrasound
imaging to quantify brachial artery vasodilation following 5 minutes of forearm cuff
occlusion. In a seminal study, Celermajer et al. (1992) demonstrated that FMD is lower
in asymptomatic children and adolescents with CVD risk factors. Consequently, FMD
has since been adopted in several acute (Bond et al. 2015b; Mills et al. 2013) and
chronic (Bond et al. 2015a; Hopkins et al. 2012) exercise intervention studies in this age
group, which assess FMD both within and between days. However, to our knowledge
no data are available regarding the within- and between- day reliability of the FMD
technique in a paediatric cohort when FMD is performed by the same practitioner and
according to current methodological guidelines (Harris et al. 2010; Thijssen et al. 2011).
A key methodological consideration when determining FMD is to account for pulsatile
changes in arterial lumen width by synchronising the acquisition of ultrasound images
at end diastole using an electrocardiogram (ECG) (Harris et al. 2010). Given that recent
guidelines recommend that baseline arterial diameter be determined for at least 1 minute
prior to cuff occlusion (Thijssen et al. 2011), and that the time-domain analysis of heart
rate variability (HRV) can be accurately determined from just 1 minute of R-R intervals
(Flatt and Esco 2015), it is possible to also assess autonomic function using the ECG-
3
gated images acquired during the FMD protocol (Bond et al. 2015a). Low HRV is
associated with clustered CVD risk factors in adolescents (Farah et al. 2014), and
changes in autonomic function may account for some of the reduction in CVD risk with
exercise (Bond et al. 2015a; Joyner and Green 2009). Therefore there is merit in
identifying the reproducibility of this outcome when determined simultaneously with
FMD.
The hyperaemic response of the cutaneous circulation following cuff occlusion during
the FMD protocol also provides insight into microvascular function. Such an
assessment is important as the earliest changes in endothelial function due to the
metabolic syndrome may manifest in the capillary and arteriole beds rather than conduit
arteries (Pinkney et al. 1997). Furthermore, a lack of association has been demonstrated
between FMD and post-ischaemic microvascular reactivity in adults (Shamim-Uzzaman
et al. 2002) and adolescents (Bond et al. 2015b). Thus, simultaneous assessment of
cutaneous post-occlusive reactive hyperaemia during the FMD protocol may offer
further insight into vascular health, although the within- and between- day reliability of
this approach needs to be established.
Typically, assessments of macrovascular, microvascular and autonomic function are
performed in isolation. The purpose of this study was to identify the within- and
between-day reproducibility of a single protocol (in line with contemporary guidelines)
to simultaneously and non-invasively elucidate these outcomes in adolescents.
METHODS
Participants
4
Forty 12-15 year old adolescents (20 boys) volunteered to take part in this study.
Written participant assent and parental consent were obtained before participation in the
project, which was approved by the institutional ethics committee. Exclusion criteria
included the use of any medication or substance known to influence vascular function.
Participants were familiarised to all measures on a separate visit prior to the
commencement of the study. During this visit, maximal oxygen uptake (𝑉̇ O2 max) was
determined using a combined ramp and supramaximal exercise protocol (Barker et al.
2011). Age- and sex- appropriate 𝑉̇ O2 max (Adegboye et al. 2011) and body mass index
(BMI) (Cole et al. 2000) cut points were used to define low fitness and
overweight/obesity respectively. Pubertal status was determined by a self-assessment of
secondary sexual characteristics using adapted drawings of the five Tanner stages
(Tanner 1962) of pubic hair development.
Study protocol
Participants completed two experimental conditions, separated by approximately one
week. With parental supervision, participants were asked to replicate their evening meal
prior to each laboratory visit. Participants also completed a food diary during the 48
hour period immediately preceding each visit, which were subsequently assessed for
total energy and macronutrient intake (CompEat Pro, Nutrition Systems, UK).
Participants were instructed to avoid strenuous exercise and wear a tri-axial
accelerometer on their wrist (GENEActiv, Activinsights Ltd, Cambridge, UK) during
the 48 hour prior to each visit. Time spent performing moderate to vigorous activity was
determined using established cut points for paediatric groups (Phillips et al. 2013).
5
Following a ~ 12 h overnight fast, participants were transported to the laboratory at
08:00 and then consumed 30 g of commercially available Corn Flakes with 130 mL of
skimmed milk. The macronutrient contribution of this breakfast is unlikely to have
influenced endothelial function (Vogel et al. 1997). At 08:45, participants rested in a
darkened, temperature-controlled (24°C) room for 15 min before the simultaneous
assessment of endothelial (FMD), microvascular and autonomic (HRV) function
(methods described below). A subgroup of 20 adolescents (10 boys) repeated these
measures 90 minutes after the first assessment to assess within-day variability of these
outcomes. In this group, the location of the probe during the first scan was marked on
the arm in order to minimise error.
Assessment of FMD
High-resolution Doppler and B-mode images of the brachial artery were simultaneously
assessed (Sequoia 512, Acuson, Siemens Corp, Aspen, USA) with a 13 MHz linear
array transducer in duplex mode, in accordance with recent guidelines (Thijssen et al.
2011) and our earlier work (Bond et al. 2015a; Bond et al. 2015b). Following a 10
minute acclimatisation period to the temperature controlled room (24°C) in the supine
position, baseline arterial diameter was measured for 1.5 minutes. Endotheliumdependent vasodilation was calculated as the percentage increase in arterial diameter
after a 5 minute ischaemic stimulus induced by rapid forearm pneumatic cuff inflation
(Hokanson, Bellevue, USA) to 220 mmHg. Baseline and post occlusion brachial artery
diameter was assessed during end diastole using validated ECG-gating software
(Medical Imaging Applications LLC, Coralvile USA) (Mancini et al. 2002; Thijssen et
al. 2011). All analyses were performed by the primary investigator who was blinded to
6
the measurement point. The area under the curve for estimated shear rate was calculated
from the point of cuff deflation until the time of peak dilation (SRAUC) (Thijssen et al.
2011).
Assessment of microvascular function
During the FMD protocol, microvascular function was simultaneously assessed using a
laser Doppler perfusion imager (Periscan PIM II, Perimed, Järfälla, Sweden) positioned
20 cm away from a reproducible point on the distal third of the forearm. High resolution
data were collected at 4.33 Hz, and then interpolated to 1 s averages before being
smoothed using a 5 s moving average. Peak reactive hyperaemia (PRH) was defined as
the highest point after occlusion (Cracowski et al. 2006). The time taken to achieve
PRH (PRHt) was also determined as an indicator of vascular resistance (Wahlberg et al.
1994). Total reactive hyperaemia (TRH) was calculated by determining the area under
the post-occlusive reactive hyperaemic curve minus the baseline (pre-occlusion) blood
flow (expressed as a percentage of PRH), multiplied by the time taken for reactive
hyperaemia to return to baseline (Wong et al. 2003). When calculated in this manner,
TRH is known to be nitric oxide independent (Wong et al. 2003), and accounts for
differences in baseline skin perfusion.
Assessment of HRV
HRV was determined using the time intervals between each ECG-gated image of the
brachial artery (i.e. the R-R interval) during the 1.5 minutes prior to cuff occlusion.
These data were screened for ectopic beats, and artefacts were removed and replaced by
7
the mean of the adjacent beats (Task force of the European Society of Cardiology
1996). The root mean square of the squared differences between adjacent normal R-R
intervals (RMSSD) was calculated using the Kubios HRV software (Biosignal Analysis
and Medical Imaging Group, Joensuu, Finland), which has recently been used to
establish reference values in adolescent boys (Farah et al. 2014).
Statistical Analyses
In accordance with recent statistical guidelines (Atkinson et al. 2013), the primary
outcome for FMD was the difference between log-transformed peak and baseline
arterial diameter, adjusted allometrically for baseline diameter. However, the widelyused ratio-scaled FMD statistic has also been included for reference. Similarly, both the
allometric and ratio-scaled FMD statistic were normalised for SRAUC, even though they
were not related to SRAUC (r<0.23, P>0.54 for all measures), as this outcome is
frequently reported in the paediatric literature. The lack of relationship between SRAUC
and FMD is consistent with other paediatric data (Hopkins et al. 2015).
Parameters of macro- and micro- vascular function were initially analysed using a
mixed model ANOVA with visit (between-day) or assessment (within-day) and sex
(male, female) as the main effects. The ANOVA model did not reveal a significant main
effect or interaction effect for these variables. Data were subsequently pooled for sex,
and the mean within- and between-day differences in outcomes of interest were
analysed using paired samples t tests with significance set at P<0.05. The
reproducibility of these outcomes was explored using the typical error, the typical error
8
expressed as a coefficient of variation (CV) and Pearson’s correlation coefficient (r)
(Hopkins 2000).
RESULTS
Baseline participant characteristics for the within-day (n=20) and between-day (n=40)
groups are presented in Table 1.
Within-day reliability
The maturation status for boys and girls included in the within-day reliability analysis
was as follows; Tanner stage 2, n=1 and n=0; stage 3, n=3 and n=0; stage 4, n=5 and
n=7; stage 5, n=1 and n=3. Macro- and microvascular data from 3 participants were
excluded from analysis due to movement of the arm/probe following cuff deflation. The
within-day reproducibility of parameters of macrovascular, microvascular and
autonomic function are presented in Table 2. No significant mean differences were
apparent between assessments 1 and 2 for any outcome. Significant correlations were
apparent between assessments 1 and 2 for all outcomes (0.54<r<0.97) apart from PRHt
(P=0.81).
Between-day reliability
The maturation status for boys and girls was as follows; Tanner stage 2, n=2 and n=0;
stage 3, n=6 and n=0; stage 4, n=10 and n=14; stage 5, n=2 and n=6. No significant
differences in energy intake, individual macronutrient contributions or time spent
9
performing moderate to vigorous physical activity were apparent during the 48 hours
preceding each laboratory visit (Table 3).
Suitable macro- and micro-vascular data were obtained from 35 participants for both
assessments. No significant differences were apparent between visits for any outcome
measure (Table 4). Significant correlations were observed between visits for all
outcomes (0.64<r<0.94).
DISCUSSION
Initial guidelines for the ratio-scaled FMD technique suggest that a CV of 20-30% is an
acceptable level of reproducibility for the ratio-scaled FMD outcome (Corretti et al.
2002). Since this early publication, the development of software to continuously track
the intima-lumen interface has improved the reproducibility of this technique almost
fourfold (Woodman et al. 2001), and has been incorporated into the most recent
methodological guidelines (Harris et al. 2010; Thijssen et al. 2011). Currently, no
paediatric data are available regarding the within- and between-day reliability of the
ratio-scaled FMD statistic when determined in this manner. However, the between-day
variability of this outcome in the present study is consistent with a CV of 10.9%
(Donald et al. 2010) and Pearson correlation coefficient of 0.88 (Leeson et al. 1997)
reported in paediatric investigations where FMD was not performed in accordance with
the current methodological guidelines (Harris et al. 2010; Thijssen et al. 2011). Our data
also compare favourably with the most conservative within-day (CV ~ 5 – 10%)
(Donald et al. 2008; Ghiadoni et al. 2012; Meirelles et al. 2007; Onkelinx et al. 2012)
10
and between-day (CV ~ 7 – 16%) (Charakida et al. 2013; Donald et al. 2008; Ghiadoni
et al. 2012; Hashimoto et al. 1995; Meirelles et al. 2007; Onkelinx et al. 2012) values
reported in adult studies, and it is likely that the good reproducibility we have
demonstrated here is related to our minimising of known sources of error. These include
1) controlling for preceding diet (Vogel et al. 1997), physical activity (Bond et al.
2015a; Bond et al. 2015b) and use of supplements (Harris et al. 2009); 2) standardising
the time of assessment (ter Avest et al. 2005), adopting a high (13 MHz) frequency
probe (Herrington et al. 2001), utilising continuous Duplex mode imaging postocclusion (Donald et al. 2008), and analysing ECG-gated images (Corretti et al. 2002)
using validated edge-detection software (Woodman et al. 2001); and 3) removing intersonographer variation and “reader” error (Charakida et al. 2013; Donald et al. 2010).
In accordance with recent criticisms of the ratio-scaled FMD statistic (Atkinson et al.
2013), we also present allometrically scaled FMD in order to partition out the
confounding effect of vessel calibre. To our knowledge, this is the first time the withinand between-day reproducibility of allometrically-scaled FMD has been presented, and
our findings demonstrate that the reliability of this outcome is comparable, if not
marginally better, than the ratio-scaled FMD statistic. In contrast, the SRAUC
demonstrated poorer reproducibility, but is consistent with similar measures of
hyperaemic blood flow reported elsewhere (Charakida et al. 2013; Ghiadoni et al.
2012). Considering that both our group and others (Hopkins et al. 2015) have failed to
observe a meaningful relationship between SRAUC and FMD in a paediatric population,
and that normalising the FMD statistic for SRAUC ultimately results in poorer withinand between-day variation, there does not appear to be a strong rationale for expressing
FMD in this manner.
11
It is noteworthy that one ultrasound scan was rejected from within-day and between-day
analysis in 3 and 5 participants respectively, due to excessive movement by the
participant or poor image quality. This rejection rate of ~ 10% in the present
investigation is in line with the expected level of data loss when performing FMD with
adults (Charakida et al. 2013) and children (Leeson et al. 1997) and should be
considered by researchers when contemplating sample sizes.
A key feature of this investigation is the simultaneous assessment of microvascular
function, which provides further insight regarding vascular health as FMD shows little
association with skin reactive hyperaemia (Shamim-Uzzaman et al. 2002). Currently,
the mechanisms underlying the cutaneous post-ischaemic response are not well defined,
however Wong et al. (2003) demonstrated that the area under the total reactive
hyperaemic curve versus time following a 5 minute ischaemic stimulus is nitric-oxide
independent. These authors reported a between-day coefficient of variation for this
outcome of 25%, which is consistent with our TRH data. Elsewhere, the within- and
between-day variation in PRH and PRHt have been reported to be < 10% (Yvonne-Tee
et al. 2005), which we were unable to replicate. The variability in these measures are
mostly related to measurement site rather than between-day variability (Kubli et al.
2000), therefore this disparity is probably related to our use of a laser positioned 20 cm
away from the forearm, rather than attaching a laser-emitting probe to the skin surface.
However, our data show that the simultaneous acquisition of cutaneous PRH and TRH
during the FMD protocol is achievable.
To our knowledge, no other group have presented HRV data alongside the FMD
outcome, despite appropriately determining brachial artery diameter from ECG-gated
ultrasound images. We find this surprising as HRV provides prognostic information
12
regarding cardiovascular event risk beyond “traditional” CVD risk factors (Tsuji et al.
1996), which is consistent with the rationale for performing FMD (Green et al. 2011).
The within- and between-day reproducibility of the RMSSD outcome presented here is
in line with the work of others (Salo et al. 1999; Sandercock et al. 2005), therefore we
encourage researchers to use the ECG-triggered ultrasound data to determine HRV.
Additionally, it may be prudent to measure baseline artery diameter for longer than the
1.5 minute period utilised in the present investigation in order to include both time and
frequency domain analyses and thus provide a more comprehensive analysis of
sympathovagal balance.
The strengths of this investigation include controlling for physical activity and diet
during the 48 hours preceding each trial, the inclusion of allometrically-scaled FMD,
the adoption of the most recent FMD guidelines (Thijssen et al. 2011). We were unable
to control for potential confounding influence of the menstrual cycle on FMD
(Hashimoto et al. 1995), although this is not a consistent finding regarding HRV and
microvascular function. Furthermore, any influence of the menstrual cycle would
logically add error to our between-day measures. Secondly, we cannot comment on the
reliability of frequency-domain HRV analyses where 5 minutes of data capture is
recommended (Task force of the European Society of Cardiology 1996).
We have previously demonstrated that favourable changes in FMD and HRV following
exercise training may be independent of modifications to traditional CVD risk factors in
adolescents (Bond et al. 2015a). Considering that impairments in macrovascular
(Celermajer et al. 1992), microvascular (Khan et al. 2003) and autonomic (Farah et al.
2014) function are already apparent in children and adolescents with CVD risk factors,
13
there is a clear rationale for future studies to identify how lifestyle interventions can
favourably modify these outcomes.
CONCLUSION
The present investigation sought to address the within- and between-day reliability of a
single protocol to simultaneously and non-invasively assess endothelial, microvascular
and autonomic function. Our data indicate suitable within- and between-day reliability
of these measures. Given that the methodological considerations for the FMD technique
are commensurate with the assessment of post-occlusive reactive hyperaemia in the
cutaneous circulation (Cracowski et al. 2006; Thijssen et al. 2011) and HRV
assessment, and that these outcomes may account for some of the “risk factor gap”
between physical activity and CVD risk (Bond et al. 2015a; Joyner and Green 2009),
we hope that other researchers performing FMD will adopt similar protocol designs in
order to maximise their insight regarding how exercise modulates CVD risk.
Acknowledgements: The research team would like to thank Ms Lucy Corless, Ms
Yasmin Pratt and Ms Siobhan Hind for help with data collection, and the pupils who
volunteered from Blundell’s School and Exmouth Community College. This study was
supported by the Physiological Society.
Conflicts of interest: The authors have no conflicts of interest.
14
REFERENCES
Adegboye AR et al. (2011) Recommended aerobic fitness level for metabolic health in
children and adolescents: a study of diagnostic accuracy Br J Sports Med 45:722-728
Atkinson G, Batterham AM, Thijssen DH, Green DJ (2013) A new approach to improve
the specificity of flow-mediated dilation for indicating endothelial function in
cardiovascular research J Hypertens 31:287-291
Barker AR, Williams CA, Jones AM, Armstrong N (2011) Establishing maximal
oxygen uptake in young people during a ramp cycle test to exhaustion Br J Sports Med
45:498-503
Bond B et al. (2015a) Two weeks of high-intensity interval training improves novel but
not traditional cardiovascular disease risk factors in adolescents Am J Physiol Heart
Circ Physiol 309:H1039-1047
Bond B, Hind S, Williams CA, Barker AR (2015b) The Acute Effect of Exercise
Intensity on Vascular Function in Adolescents Med Sci Sports Exerc 47:2628-2635
Celermajer DS et al. (1992) Non-invasive detection of endothelial dysfunction in
children and adults at risk of atherosclerosis Lancet 340:1111-1115
Charakida M et al. (2013) Variability and reproducibility of flow-mediated dilatation in
a multicentre clinical trial European heart journal 34:3501-3507
Cole TJ, Bellizzi MC, Flegal KM, Dietz WH (2000) Establishing a standard definition
for child overweight and obesity worldwide: international survey BMJ 320:1240-1243
Corretti MC et al. (2002) Guidelines for the ultrasound assessment of endothelialdependent flow-mediated vasodilation of the brachial artery: a report of the
International Brachial Artery Reactivity Task Force J Am Coll Cardiol 39:257-265
Cracowski JL, Minson CT, Salvat-Melis M, Halliwill JR (2006) Methodological issues
in the assessment of skin microvascular endothelial function in humans Trends in
pharmacological sciences 27:503-508
Donald AE et al. (2010) Determinants of vascular phenotype in a large childhood
population: the Avon Longitudinal Study of Parents and Children (ALSPAC) European
heart journal 31:1502-1510
Donald AE et al. (2008) Methodological approaches to optimize reproducibility and
power in clinical studies of flow-mediated dilation J Am Coll Cardiol 51:1959-1964
Farah BQ, Barros MV, Balagopal B, Ritti-Dias RM (2014) Heart rate variability and
cardiovascular risk factors in adolescent boys J Pediatr 165:945-950
15
Flatt AA, Esco MR (2015) Heart rate variability stabilization in athletes: towards more
convenient data acquisition Clinical physiology and functional imaging
Ghiadoni L et al. (2012) Assessment of flow-mediated dilation reproducibility: a
nationwide multicenter study J Hypertens 30:1399-1405
Green DJ, Jones H, Thijssen D, Cable NT, Atkinson G (2011) Flow-mediated dilation
and cardiovascular event prediction: does nitric oxide matter? Hypertension 57:363-369
Harris RA, Nishiyama SK, Wray DW, Richardson RS (2010) Ultrasound assessment of
flow-mediated dilation Hypertension 55:1075-1085
Harris RA, Nishiyama SK, Wray DW, Tedjasaputra V, Bailey DM, Richardson RS
(2009) The effect of oral antioxidants on brachial artery flow-mediated dilation
following 5 and 10 min of ischemia Eur J Appl Physiol 107:445-453
Hashimoto M et al. (1995) Modulation of endothelium-dependent flow-mediated
dilatation of the brachial artery by sex and menstrual cycle Circulation 92:3431-3435
Herrington DM et al. (2001) Brachial flow-mediated vasodilator responses in
population-based research: methods, reproducibility and effects of age, gender and
baseline diameter J Cardiovasc Risk 8:319-328
Hopkins ND et al. (2015) Age and sex relationship with flow-mediated dilation in
healthy children and adolescents J Appl Physiol (1985) 119:926-933
Hopkins ND, Stratton G, Cable NT, Tinken TM, Graves LE, Green DJ (2012) Impact of
exercise training on endothelial function and body composition in young people: a study
of mono- and di-zygotic twins Eur J Appl Physiol 112:421-427
Hopkins WG (2000) Measures of reliability in sports medicine and science Sports Med
30:1-15
Joyner MJ, Green DJ (2009) Exercise protects the cardiovascular system: effects
beyond traditional risk factors J Physiol 587:5551-5558
Juonala M, Viikari JS, Laitinen T, Marniemi J, Helenius H, Ronnemaa T, Raitakari OT
(2004) Interrelations between brachial endothelial function and carotid intima-media
thickness in young adults: the cardiovascular risk in young Finns study Circulation
110:2918-2923
Khan F, Green FC, Forsyth JS, Greene SA, Morris AD, Belch JJ (2003) Impaired
microvascular function in normal children: effects of adiposity and poor glucose
handling J Physiol 551:705-711
Kubli S, Waeber B, Dalle-Ave A, Feihl F (2000) Reproducibility of laser Doppler
imaging of skin blood flow as a tool to assess endothelial function J Cardiovasc
Pharmacol 36:640-648
16
Leeson CP, Whincup PH, Cook DG, Donald AE, Papacosta O, Lucas A, Deanfield JE
(1997) Flow-mediated dilation in 9- to 11-year-old children: the influence of
intrauterine and childhood factors Circulation 96:2233-2238
Mancini GB, Yeoh E, Abbott D, Chan S (2002) Validation of an automated method for
assessing brachial artery endothelial dysfunction Can J Cardiol 18:259-262
Meirelles CM, Leite SP, Montenegro CAB, Gomes PSC (2007) Reliability of brachial
artery flow-mediated dilatation measurement using ultrasound Arquivos Braileiros de
Cardiologia 89:176-183
Mills A et al. (2013) The effect of exergaming on vascular function in children J Pediatr
163:806-810
Onkelinx S, Cornelissen V, Goetschalckx K, Thomaes T, Verhamme P, Vanhees L
(2012) Reproducibility of different methods to measure the endothelial function Vasc
Med 17:79-84
Phillips LR, Parfitt G, Rowlands AV (2013) Calibration of the GENEA accelerometer
for assessment of physical activity intensity in children Journal of science and medicine
in sport / Sports Medicine Australia 16:124-128
Pinkney JH, Stehouwer CD, Coppack SW, Yudkin JS (1997) Endothelial dysfunction:
cause of the insulin resistance syndrome Diabetes 46 Suppl 2:S9-13
Salo TM, Voipio-Pulkki LM, Jalonen JO, Helenius H, Viikari JS, Kantola I (1999)
Reproducibility of abnormal heart rate variability indices: the case of hypertensive sleep
apnoea syndrome Clin Physiol 19:258-268
Sandercock GR, Bromley PD, Brodie DA (2005) The reliability of short-term
measurements of heart rate variability International journal of cardiology 103:238-247
Shamim-Uzzaman QA et al. (2002) Altered cutaneous microvascular responses to
reactive hyperaemia in coronary artery disease: a comparative study with conduit vessel
responses Clin Sci (Lond) 103:267-273
Stary HC (1989) Evolution and progression of atherosclerotic lesions in coronary
arteries of children and young adults Arteriosclerosis 9:I19-32
Tanner JM (1962) Growth at Adolescence. 2nd edn. Blackwell Scientific Publications,
Oxford
Task force of the European Society of Cardiology (1996) Heart rate variability.
Standards of measurement, physiological interpretation, and clinical use. European heart
journal 17:354-381
17
ter Avest E, Holewijn S, Stalenhoef AF, de Graaf J (2005) Variation in non-invasive
measurements of vascular function in healthy volunteers during daytime Clin Sci
(Lond) 108:425-431
Thijssen DH et al. (2011) Assessment of flow-mediated dilation in humans: a
methodological and physiological guideline Am J Physiol Heart Circ Physiol 300:H212
Tsuji H, Larson MG, Venditti FJ, Jr., Manders ES, Evans JC, Feldman CL, Levy D
(1996) Impact of reduced heart rate variability on risk for cardiac events. The
Framingham Heart Study Circulation 94:2850-2855
Vogel RA, Corretti MC, Plotnick GD (1997) Effect of a single high-fat meal on
endothelial function in healthy subjects Am J Cardiol 79:350-354
Wahlberg E, Line PD, Olofsson P, Swedenborg J (1994) Correlation between peripheral
vascular resistance and time to peak flow during reactive hyperaemia European journal
of vascular surgery 8:320-325
Wong BJ, Wilkins BW, Holowatz LA, Minson CT (2003) Nitric oxide synthase
inhibition does not alter the reactive hyperemic response in the cutaneous circulation J
Appl Physiol 95:504-510
Woodman RJ et al. (2001) Improved analysis of brachial artery ultrasound using a novel
edge-detection software system J Appl Physiol (1985) 91:929-937
Yvonne-Tee GB, Rasool AH, Halim AS, Rahman AR (2005) Reproducibility of
different laser Doppler fluximetry parameters of postocclusive reactive hyperemia in
human forearm skin J Pharmacol Toxicol Methods 52:286-292
18
Table 1: Participant characteristics
Age (y)
Body mass (kg)
Stature (m)
BMI (kg∙m2)
Overweight (n)
𝑉̇ O2 max (mL∙min-1∙kg-1)
Low fit (n)
Within-day (n = 20)
Between-day (n = 40)
14.4 ± 0.6
57.8 ± 10.9
1.65 ± 0.09
21.1 ± 3.2
3
41.6 ± 6.7
14.3 ± 0.6
58.0 ± 11.2
1.66 ± 0.09
21.0 ± 3.0
6
41.3 ± 6.7
7
15
BMI = body mass index; 𝑉̇ O2 max = maximal oxygen uptake. Data presented as mean ±
SD.
19
Table 2: Within-day reproducibility of novel CVD outcomes
Assessment 1
Mean ± SD
Assessment 2
Mean ± SD
Change in
mean
P value
Typical
error
Typical error
as CV (%)
r
3.12 ± 0.33
3.40 ± 0.34
8.8 ± 0.8
7.6 ± 1.1
871 ± 266
0.011 ± 0.003
0.009 ± 0.003
3.11 ± 0.32
3.39 ± 0.33
8.9 ± 0.7
7.6 ± 1.1
825 ± 221
0.012 ± 0.003
0.010 ± 0.003
-0.01
-0.01
0.1
0.1
-46
0.001
0.001
0.61
0.75
0.49
0.46
0.21
0.26
0.27
0.06
0.07
0.4
0.4
96
0.001
0.001
1.8
2.1
5.1
4.4
12.0
12.0
12.3
0.97
0.96
0.74
0.90
0.86
0.83
0.89
PRH (AU)
PRHt (s)
TRH (AU)
2.05 ± 0.34
30 ± 16
219 ± 72
1.97 ± 0.28
23 ± 11
211 ± 69
-0.08
-7
-7
0.29
0.14
0.61
0.22
13
41
11.0
79.7
29.9
0.54
0.09
0.65
RMSSD (ms)
79.9 ± 27.3
82.8 ± 23.9
2.9
0.45
11.0
17.6
0.82
Baseline diameter (mm)
Peak diameter (mm)
Ratio-scaled FMD (%)
Allometric FMD (%)
SRAUC
Ratio-scaled FMD/SRAUC
Allometric FMD/SRAUC
CV, coefficient of variation; FMD, flow mediated dilation; SRAUC, area under the post-occlusive curve until peak dilation for shear rate; PRH,
peak reactive hyperaemia; PRHt, time taken to achieve peak reactive hyperaemia; TRH, total reactive hyperaemia; RMSSD, root mean square of
successive R-R intervals. All correlation coefficients are statistically significant, apart from PRHt.
20
Table 3: Accelerometer and food diary data during the 48 hours preceding each laboratory visit
Moderate-vigorous activity (min day-1)
Total energy intake (kcal day-1)
Energy from carbohydrates (%)
Energy from fat (%)
Energy from protein (%)
Visit 1
Visit 2
P value
95% CI
36 ± 14
1903 ± 366
47 ± 5
37 ± 5
16 ± 4
31 ± 13
1919 ± 358
47 ± 5
37 ± 5
16 ± 3
0.14
0.94
0.42
0.58
0.77
-2 to 13
-130 to 140
-3 to 1
-2 to 3
-1 to 2
95% CI = 95% confidence intervals for the true difference
Data have been pooled as ANOVA analysis revealed no main effect for sex
21
Table 4: Between-day reproducibility of novel CVD outcomes
P value
Typical
error
Typical
error as
CV (%)
r
0.02
0.02
0.1
<0.1
25
0.001
0.001
0.35
0.34
0.84
0.91
0.53
0.36
0.39
0.09
0.10
1.0
0.9
162
0.003
0.003
2.9
2.7
10.6
9.4
29.6
25.7
25.9
0.94
0.94
0.78
0.78
0.70
0.72
0.73
1.84 ± 0.42
23 ± 14
208 ± 62
0.01
-1
10
0.80
0.56
0.50
0.22
9
32
13.3
68.2
23.1
0.73
0.64
0.76
70.4 ± 23.9
-1.7
0.48
10.7
17.6
0.80
Visit 1
Mean ± SD
Visit 2
Mean ± SD
Change in
mean
3.16 ± 0.33
3.43 ± 0.35
8.6 ± 1.3
7.3 ± 1.1
691 ± 310
0.014 ± 0.006
0.012 ± 0.005
3.18 ± 0.37
3.45 ± 0.39
8.6 ± 2.1
7.3 ± 1.9
716 ± 270
0.013 ± 0.006
0.011 ± 0.005
1.83 ± 0.41
24 ± 15
197 ± 69
72.1 ± 24.1
Flow mediated dilation
Baseline diameter (mm)
Peak diameter (mm)
Ratio-scaled FMD (%)
Allometric FMD (%)
SRAUC
Ratio-scaled FMD/SRAUC
Allometric FMD/SRAUC
Laser Doppler perfusion imaging
PRH (AU)
PRHt (s)
TRH (AU)
Heart rate variability
RMSSD (ms)
CV, coefficient of variation; FMD, flow mediated dilation; SRAUC, area under the post-occlusive curve until peak dilation for shear rate; PRH,
peak reactive hyperaemia; PRHt, time taken to achieve peak reactive hyperaemia; RMSSD, root mean square of successive R-R intervals. All
correlation coefficients are statistically significant.
22