J Appl Physiol 130: 149–159, 2021.

First published October 29, 2020; doi:10.1152/japplphysiol.00502.2020

RESEARCH ARTICLE

Physiology of Thermal Therapy

Distinct contributions of skin and core temperatures to flow-mediated dilation

of the brachial artery following passive heating
Geoff B. Coombs,1 Joshua C. Tremblay,1 Daria A. Shkredova,1,2 Jay M. J. R Carr,1 Denis J. Wakeham,3
Alexander Patrician,1 and Philip N. Ainslie1
1
Centre for Heart, Lung and Vascular Health, School of Health and Exercise Sciences, University of British Columbia
Okanagan, Kelowna, British Columbia, Canada; 2Department of Physiology, Radboud University Medical Centre, Nijmegen,
The Netherlands; and 3Cardiff School of Sport and Health Sciences, Cardiff Metropolitan University, Cardiff, United Kingdom

Abstract
We measured acute vascular responses to heat stress to examine the hypothesis that macrovascular endothelial-dependent
dilation is improved in a shear-dependent manner, which is further modified by skin temperature. Twelve healthy males per-
formed whole body heating ( þ 1.3 C esophageal temperature), bilateral forearm heating (38 C skin temperature), and a
time-matched (60 min) control condition on separate days in a counterbalanced order. Bilateral assessments of blood flow
and brachial artery flow-mediated dilation (FMD) were performed before and 10 min after each condition with duplex
Doppler ultrasound. To isolate the influence of shear stress, a pneumatic cuff was inflated (90 mmHg) around the right
forearm during each condition to attenuate heat-induced rises in blood flow and shear stress. After forearm heating, FMD
increased [cuffed: 4.7 (2.9)% to 6.8 (1.5)% and noncuffed: 5.1 (2.8)% to 6.4 (2.6)%] in both arms (time P < 0.01). Whole body
heating also increased FMD in the noncuffed arm from 3.6 (2.2)% to 9.2 (3.2)% and in the cuffed arm from to 5.6 (3.0)% to
8.6 (4.9)% (time P < 0.01). After the time control, FMD decreased [cuffed: 6.3 (2.4)% to 4.7 (2.2)% and noncuffed: 6.1 (3.0)%
to 4.5 (2.6)%] in both arms (time P = 0.03). Multiple linear regression (adjusted R2 = 0.421 P = 0.003) revealed that changes
in esophageal temperature, skin temperatures, and heart rate explained the majority of the variance in this model (34%,
31%, and 21%, respectively). Our findings indicate that, in addition to shear stress, skin and core temperatures are likely im-
portant contributors to passive heating-induced vascular adaptations.
NEW & NOTEWORTHY The primary determinant of vascular adaptations to lifestyle interventions, such as exercise and heat
therapy, is repeated elevations in vascular shear stress. Whether skin or core temperatures also modulate the vascular adapta-
tion to acute heat exposure is unknown, likely due to difficulty in dissociating the thermal and hemodynamic responses to heat.
We found that skin and core temperatures modify the acute vascular responses to passive heating irrespective of the magnitude
of increase in shear stress.

endothelial function; heat stress; reactive hyperemia; ultrasound; vascular function

INTRODUCTION The precise mechanisms mediating improved cardiovas-

Recent reports from large-scale, prospective population
studies suggest that regular exposure to passive heat stress via
various methods (e.g., sauna, hot tubs) is associated with lower
risk of cardiovascular diseases (CVDs) and related mortality
(1, 2). Several studies have investigated potential physiological
mechanisms underlying the reduction in CVD risk with
repeated heat exposures (3–6). For example, Brunt et al. (3)
demonstrated in a sham-controlled study that blood pressure
and arterial stiffness were reduced and endothelial-dependent
dilation (flow-mediated dilation, FMD) was improved following
8 wk of repeated whole body hot water immersion via hot tub.
cular function are unclear in part due to the difficulty of iso-
lating the effects of temperature from the associated
hemodynamic responses. Acutely, regional limb heating
improves endothelial function of the brachial artery (7, 8)
and whole body heating additionally reduces blood pressure
and aortic stiffness (6, 9). However, lower limb/body heating
is often accompanied by increased core temperature (10–12)
confounding interpretation of underlying mechanisms. The
importance of shear stress as a hemodynamic stimuli has
been highlighted (13) and it is widely reported as the main
stimulus for heat-induced vascular adaptation (7, 11, 12).
Some studies have used pneumatic cuff inflation around one

Correspondence: G. B. Coombs (geoff.coombs91@gmail.com).

Submitted 16 June 2020 / Revised 23 September 2020 / Accepted 21 October 2020

http://www.jap.org 8750-7587/21 Copyright © 2021 the American Physiological Society 149

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES
limb to reduce regional flow with the aim of attenuating the
increase in shear rate compared to the contralateral limb.
For example, using this experimental paradigm, Tinken
et al. (8) observed greater brachial artery FMD in the non-
cuffed arm immediately following acute forearm heating
thereby demonstrating a role for shear rate in vascular adap-
tations to passive heating. This finding was further supported
by subsequent studies with chronic repeated increases in
local shear rate (14) and moderate ( þ 0.6 C) increases in core
temperature (15); however, the FMD response of the brachial
artery to acute and large elevations in core temperature
remains unclear. Although both shear rate and skin tempera-
ture are involved in adaptation to 8 wk of local forearm heat-
ing in the cutaneous microvasculature (16), the interaction of
local shear rate, skin temperature, and core temperature on
brachial artery endothelium-dependent dilation requires fur-
ther investigation. Indeed, rises in skin temperature increase
local shear stress, but increased core temperature imposes a
greater systemic cardiovascular stress similar to mild exer-
cise, such that a larger cumulative increase in shear stress
occurs.
Therefore, the primary aim of this study was to determine
the thermal and hemodynamic factors that contribute to
vascular responses to acute heat stress. To delineate these
interactions, using a within-subject design, we bilaterally
assessed vascular function in the brachial artery before and
after: 1) bilateral local forearm heating (to increase shear rate
and skin temperature); 2) whole body heating (to increase
shear rate, as well as skin and core temperatures); and 3)
thermoneutral time control. During each condition, forearm
blood flow (and therefore shear rate) was attenuated to one
limb via forearm cuff inflation. It was hypothesized that
FMD is greater following passive heating in a shear-depend-
ent manner, and that the increases in FMD are greater dur-
ing whole body heating compared to forearm heating, and in
the noncuffed arm compared to the cuffed arm.
Furthermore, we hypothesized that FMD is greater following
forearm heating compared to whole body heating in the
cuffed arm due to higher skin temperature. As previously
established, we also anticipated that FMD is reduced follow-
ing the time control condition where greater retrograde and
low mean shear stress will be a result of immobilization (17)
and cuff inflation (18).
METHODS
Participants
Based on recently published data (7) reporting improve-
ments in FMD after forearm heating [5.8% standard devia-
tion (SD) = 2.2 vs. 8.4% SD = 3.6], an a priori sample size
estimate (a = 0.05, b = 0.80) using G*Power version 3.1
(University of Dusseldorf, Dusseldorf, Germany) required a
minimum of 11 participants for this study (19). We therefore
recruited 12 white male volunteers to account for potential
loss of data inherent with vascular ultrasound. All partici-
pants [age: 25 (SD = 5) yr; body mass index (BMI): 23.1 (2.6)
kg/m2; peak oxygen consumption (V_ o2peak): 41 (8) mg/kg/
min] completed the study and provided written, informed
consent before any experimental measures. Approval of the
experimental protocol was obtained from the Clinical
150 J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org
Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
Research and Ethics Board at the University of British
Columbia (H17-01817) and all procedures conformed to the
Declaration of Helsinki, except for registration in a database.
None of the participants reported a history of cardiovascular,
respiratory, metabolic, or neurological disease and were not
taking any medication for such conditions.
Measurements
Before baseline measurements, participants were instru-
mented with general purpose probes (RET-1, Physitemp
Instruments, Clifton, NJ) inserted 40 cm into the esopha-
gus for the measurement of esophageal temperature (Tes).
During the insertion of the esophageal probe, 500 mL of
water was consumed for participant comfort and to ensure
euhydration before baseline measures. Skin temperature
(Tsk) probes (MLT422/A, ADInstruments, Colorado Springs,
CO) were placed on the ventral surface of each forearm and
secured with medical tape. Heart rate (HR) was monitored
via lead II electrocardiogram and peripheral blood pressure
was measured via automated brachial artery auscultation
(BP5100, Omron Healthcare Canada, Burlington, ON,
Canada). A segmental cuff (SC5, Hokanson, Bellevue, WA)
was placed immediately distal to the elbow and antecubi-
tal IV catheter on the right arm to attenuate increases in
blood flow to that arm. Skin and core temperature probes
were interfaced with a data acquisition system (PowerLab
16/35, ADI) via thermistor and T-type pods, respectively
(ML309 and ML312, ADI), whereas the ECG signal was
connected via a Dual Bio Amp (FE232, ADI).
All vascular sonography was performed using high resolu-
tion duplex ultrasound (Terson uSmart 3300/T3200, Teratech,
Burlington, MA) interfaced with a 4– 15-MHz linear array trans-
ducer (15L4 Smart Mark, Teratech). An angle of insonation of
60 was maintained throughout all scans. The same experi-
enced sonographer performed all scans on the same arm
during each trial for each participant. Endothelial-de-
pendent dilation was assessed via FMD according to inter-
national guidelines (20, 21), which consisted of a 1-min
baseline recording of brachial artery diameter and velocity
followed by a forearm occlusion period of 5 min using
rapid cuff inflation 220 mmHg. During occlusion, record-
ing was paused but imaging of the vessel was continued to
ensure consistent location and angle of the image. Before
the end of the occlusion period, recording was resumed
and the cuff was rapidly deflated to induce reactive hyper-
emia while the recording was continued for 3 min. At each
time point of the protocol, recordings of brachial artery di-
ameter and velocity were performed for 1 min or 20 car-
diac cycles. Recordings were made using screen capture
software (Camtasia Studio, TechSmith, Okemos, MI) and
saved for offline analysis.
Experimental Protocol
The participants visited the laboratory for one preliminary
visit where V_ o2peak was determined while upright cycling.
The protocol consisted of a 2-min warm up at 50 W with sub-
sequent increases of 20 W every minute at a rate of 0.33 W/s
until volitional exhaustion, inability to maintain >50 rpm,
or a plateau in oxygen consumption determined via expired
gases (Vmax Encore Metabolic Cart, Carefusion, San Diego,
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES
CA). The study consisted of three separate visits to the labo-
ratory for experimental sessions. All participants arrived
between 07:00 AM and 09:00 AM (except one at 12:00 PM)
after 6 h of fasting, having avoided alcohol and strenuous
exercise for 24 h and caffeine for 12 h. The start time of the
protocol was consistent within each participant for all visits,
with at least 48 h between sessions. After instrumentation,
the participants rested supine for 20 min before baseline
measures were performed, which included three blood pres-
sure recordings on the left arm (i.e., noncuffed) followed by
simultaneous assessment of FMD on both arms. The partici-
pant was then raised into a semirecumbent position on the
bed and rested quietly for 5 min before three more blood
pressure recordings and a 1-min brachial artery ultrasound
recording. These measures were repeated halfway and at the
end of the protocol. Immediately before commencing the ex-
perimental intervention, the cuff around the right forearm
was inflated and maintained at 90 mmHg throughout to
attenuate the increase in blood flow during heating (14–16)
and to induce disturbed flow patterns during the time-con-
trol session (18, 22).
On the whole body heating day, the participant donned a
water-perfused suit (Med-End, Ottawa, ON, Canada) cover-
ing the entire skin surface area except for the head, feet, and
arms below the shoulder (see Fig. 1). A Coghlan’s emergency
blanket was placed on top to prevent any further heat losses.
The suit was circulated with water maintained at 49 C using
two heaters (A2.2-120V-US Sous Vide, Anova, San Francisco,
CA) and a magnetic drive pump (WMD-20RLT-492GPH,
Iwaki America, Holliston, MA) with the aim of increasing
esophageal temperature by 1.5 C. Forearm heating was per-
formed with the same method while only covering the fore-
arms (including the occlusion cuff) with the aim of rapidly
increasing forearm Tsk to 38 C. The whole body and fore-
arm heating sessions were completed in a counterbalanced
order to match the duration of whole body heating
(60 min) as closely as possible. The time control sessions
were completed last to match the durations of the heating
sessions. All trials were time-matched to the first visit for
each participant to ensure there was no intraindividual vari-
ation in protocol duration. The forearm heating and time
control sessions were performed wearing regular clothing
(i.e., pants and t-shirt). Core temperatures were not meas-
ured during the time control because it is not expected that
core temperature will change in a thermoneutral environ-
ment (3) and to maintain participant comfort. At the end of
the intervention, the participant was returned to the supine
position for 10 min before the final FMD measurement.
Data Analysis
All temperature data and heart rate were sampled at
400 Hz from PowerLab into LabChart Pro software (v 7,
ADI); data are reported as 5-min averages. At each time
point, blood pressure measurements were averaged and
mean arterial pressure (MAP) was calculated as the sum of
two thirds of diastolic pressure and one third of systolic
pressure (Eq. 1).
MAP ¼ 1=3  SBP þ 2=3  DBP ðmmHgÞ
The saved ultrasound recordings were analyzed using
semiautomated edge detection software (FMD/BloodFlow,
J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org
Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
v 5.1, Reed C, Perth, WA, Australia) (23). Regions of interest
were placed around the highest quality portion of the
B-mode longitudinal image of the artery and the Doppler
tracing of blood velocity. This software tracks the vessel
walls and peak envelope velocity within their respective
regions at 30 Hz. The product of calculated cross-sectional
area (CSA; Eq. 2) and mean blood velocity (MBV; Eq. 3) were
used to determine brachial blood flow (Eq. 4).
CSA ¼ p ð0:5  diameterÞ2 ðcm2 Þ ð2Þ
MBV ¼ ðpeak envelope velocity=2Þ ðcm=sÞ  60 ðs=minÞ
ð3Þ
Flow ¼ CSA  MBV ðmL=minÞ ð4Þ
Diameter postcuff occlusion was measured automatically
using a 3-s moving window-smoothed average where the
maximum median value was determined as the peak diame-
ter and FMD was then calculated as the absolute and relative
difference between peak and baseline diameters (Eq. 5).
Baseline following WBH was considered as the last 30 s of
cuff occlusion considering the influence of the large shear
stimulus on preocclusion diameter. Postcuff occlusion diam-
eter and velocity were interpolated from 30 Hz into 3-s bins
where peak reactive hyperemia was determined as an index
of forearm microvascular function (24).
FMD ¼ ðpeak  baseline diameterÞ=baseline  100 ð%Þ
ð5Þ
Shear rates (SR) were estimated with Eq. 6 using ante-
grade, retrograde, and mean blood velocities. The oscillatory
shear index (OSI) was calculated per Eq. 7.
SR ¼ 4  MBV=artery diameter ðs1 Þ ð6Þ
OSI ¼ jretrograde SRj=ðjantegrade SRj þ jretrograde SRjÞðauÞ
ð7Þ
Statistical Analyses
All data are presented as means (SD). Data were analyzed
using two-factor linear mixed models with a compound sym-
metry covariance structure where time (pre/mid/post inter-
vention) and arm (cuffed/noncuffed limb) were repeated
variables. Central hemodynamics and Tes were compared
between trials, whereas the forearm Tsk and vascular
responses were compared within trials only. A Bonferroni
correction was used for multiple comparisons between main
effects. When significance (a = 0.05) was observed, simple
main effects were determined with post hoc testing using a
paired sample t test with a Holm–Bonferroni correction. The
model for FMD was run with logged changes in diameter as
the dependent variable and logged baseline diameter and
SRAUC as covariates. Corrected group means and SD for
FMD were back calculated from the estimated means (EM)
and standard errors of the model using the formula: (eEM  1)
 100 (25, 26). The delta values for FMD between conditions
were compared with a one-way ANOVA. Before correlational
ð1Þ
analyses, the data were screened for normality by calculating
skewness and kurtosis, where values within ±2.00 were con-
sidered acceptable. Repeated-measures correlations were
151
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES

Figure 1. Experimental setup demonstrating the location of ultrasound measurements of the brachial artery and the skin surface areas covered by the
water-perfused suit in black. Created with BioRender.com.

performed between selected variables (Tes, forearm Tsk, HR,
antegrade SR, retrograde SR, and OSI) and the change in
uncorrected FMD using the rmcorr package for R (27) to
determine the influence of intraindividual changes of the in-
dependent variables on FMD. A standard multiple linear
regression analysis with the aforementioned independent
variables was then performed on the change in uncorrected
FMD values during each heating intervention. The abso-
lute value of each standardized b coefficient from the
model was used to calculate the relative contribution of
each independent variable as the quotient of b for a given
variable and the sum of b coefficients from all variables
[e.g., ba/(ba þ bb þ bc)  100] (28). To avoid unaccept-
able multicollinearity, only independent variables with
tolerance >0.1 were accepted. Our Durbin–Watson value
of 1.96 indicates that there is no autocorrelation between
observations. All statistical analyses were performed in
either SPSS v 24 (IBM, Armonk, NY) or R, and Figs. 2–4
were generated with GraphPad version 6.0 (Prism, La
Jolla, CA).
RESULTS
Central Hemodynamics and Thermometry
Heart rate, MAP, and Tes are presented in Table 1. There
was a time by arm interaction for heart rate (P < 0.01), with
greater increases during whole body heating compared to
forearm heating and time control. There were no differences
in MAP between conditions or before and after heating
(interaction P = 0.83). There was also a time by arm interac-
tion for Tes (P < 0.01), which increased more during whole
body heating [D1.3 (0.25) C] compared to forearm heating
[D0.11 (0.19) C]. Skin temperatures are presented in Fig. 2

Figure 2. Forearm skin temperatures during each intervention in the cuffed and noncuffed arms. *P < 0.05 vs. contralateral arm.

152 J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES

Figure 3. Antegrade (top) and retrograde (bottom) shear rates during each intervention in the cuffed and noncuffed arms. *P < 0.05 vs. contralateral
arm.

During forearm heating, there was a main effect of time for
forearm Tsk (P < 0.01) which increased by 8 C on both
arms with no time by arm interaction (P = 0.42). During
whole body heating, there was a time by arm interaction (P =
0.01) where forearm Tsk increased by 3 C in the noncuffed
arm compared to 0.8 C in the cuffed arm. During the time
control, there was a time by arm interaction (P = 0.02) where
forearm Tsk decreased by 1.5 C in the cuffed arm but not in
the noncuffed arm.
Brachial Artery Hemodynamics
During whole body heating, blood flow in the brachial ar-
tery increased more (interaction P < 0.01) in the noncuffed
arm [24 (13) to 240 (114) mL/min] compared to the cuffed
arm [23 (10) to 115 (75) mL/min]. During forearm heating,
brachial blood flow also increased more (interaction P <
0.01) in the noncuffed arm [31 (28) to 142 (77) mL/min] com-
pared to the cuffed arm [30 (19) to 59 (37) mL/min]. Brachial

Figure 4. Flow-mediated dilation (FMD) before and after each intervention in the cuffed and noncuffed arms. Top displays actual values (individual val-
ues in symbols and group means in gray bars), and bottom displays the group means corrected for baseline diameter and the shear stimulus (i.e.,
SRAUC). *P < 0.05 vs. preintervention. dP < 0.05 vs. cuffed arm. SRAUC, shear rate area under the curve.

J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org 153

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES

Table 1. Core temperature and cardiovascular variables

Whole Body Heating Forearm Heating Time Control ANOVA
Pre Mid End Pre Mid End Pre Mid End P value

Tes, C 36.99 (0.28) 37.70 (0.29)* 38.28 (0.50)* 36.98 (0.31) 36.99 (0.28) 37.10 (0.27) - - - <0.01
HR, beats/min 55 (8) 84 (17)*† 95 (16)*† 57 (8) 62 (9) 64 (11)† 55 (7) 58 (9) 54 (8) <0.01
MAP, mmHg 84 (7) 80 (9) 83 (7) 84 (4) 84 (6) 84 (6) 84 (6) 83 (6) 86 (7) 0.83
HR, heart rate; MAP, mean arterial pressure; Tes, esophageal temperature. Data are means ± SD. Tes was not measured during time con-
trol for participant comfort. *P < 0.05 vs. forearm heating. †P < 0.05 vs. time control.

blood flow decreased more (interaction P = 0.03) during the
time control protocol in the cuffed arm [32 (20) to 7 (12) mL/
min] compared to the noncuffed arm [42 (31) to 36 (24) mL/
min].
Antegrade and retrograde shear rate responses are pre-
sented in Fig. 3 During forearm heating, there was a time by
arm interaction for antegrade SR (P < 0.01) where the cuffed
arm increased by 57% and the noncuffed arm increased by
240%. There was a time by arm interaction for retrograde SR
(P < 0.01) where the cuffed arm increased by 217% and the
noncuffed arm decreased by 99%. There were also time by
arm interactions during whole body heating (both P < 0.01)
for both antegrade and retrograde SR. In the cuffed arm,
antegrade SR increased by 400% and retrograde SR
increased by 130%. In the noncuffed arm, antegrade SR
increased by 641% and retrograde SR decreased by 466%.
During the time control, antegrade SR did not change (P =
0.52), whereas there was a time by arm interaction for retro-
grade SR (P < 0.01). Retrograde SR increased in the cuffed
arm by 488% and by 33% in the noncuffed arm.
Flow-Mediated Dilation
Brachial artery characteristics before and after each inter-
vention are presented in Table 2 Individual values as well as
allometrically scaled and shear-corrected mean FMD values
are presented in Fig. 4 There was a main effect of time where
FMD increased by 1.5–2 percentage points in both arms after
forearm heating (P < 0.01). There was a main effect of time
(P < 0.01) where whole body heating increased FMD by 5.5
percentage points in the noncuffed arm and 3 percentage
points in the cuffed arm. There was a main effect of time
where FMD decreased in both arms after the time control
(P = 0.03) by 1.5 percentage points. In the noncuffed arm,
the change in FMD was greater with whole body heating
compared to forearm heating (P < 0.01) and time control
(P < 0.01); however, forearm heating tended to induce a
greater change in FMD compared to time control (P = 0.06).
In the cuffed arm, the change in FMD with whole body heat-
ing was greater than time control (P < 0.01) but not that
induced by forearm heating (P = 1.0); however, forearm heat-
ing induced a greater change in FMD compared to time con-
trol (P = 0.04).
Reactive Hyperemia
Postocclusion peak reactive hyperemia (RH) values are
presented in Table 2 There was a main effect of time (P <
0.01) where peak RH increased by 33% and 23% in both the
cuffed and noncuffed arms, respectively, after forearm heat-
ing. There was also a main effect of time (P < 0.01) where
whole body heating increased peak RH by 37% and 53% in
both the cuffed and noncuffed arms, respectively. During
time control, there was a main effect of arm only (P < 0.01)
where peak RH was 23% higher in the noncuffed arm com-
pared to the cuffed arm.
Correlations of Selected Variables and Regression
Analyses
The changes in FMD pooled from all three trials demon-
strated significant positive correlations with changes in Tes,
Tsk, HR, and antegrade SR, whereas there was a significant

Table 2. Brachial artery baseline characteristics and endothelium-dependent dilation before and after each
intervention
Whole Body Heating Forearm Heating Time Control
Variable Arm Pre Post Pre Post Pre Post
Baseline diameter, mm C 3.97 (0.62) 4.28 (0.69)* 4.10 (0.68) 4.18 (0.70) 3.98 (0.55) 3.90 (0.59)
NC 3.93 (0.42) 4.12 (0.38)* 3.96 (0.32) 3.98 (0.38) 3.90 (0.46) 3.93 (0.56)
FMD, mm C 0.21 (0.10) 0.35 (0.17)* 0.18 (0.09) 0.28 (0.05)* 0.25 (0.10) 0.18 (0.08)
NC 0.14 (0.09) 0.37 (0.12)* 0.20 (0.11) 0.25 (0.10) 0.23 (0.11) 0.17 (0.09)
FMD, % C 5.6 (3.0) 8.6 (4.9) 4.7 (2.9) 6.8 (1.5) 6.3 (2.4) 4.7 (2.2)
NC 3.6 (2.2) 9.2 (3.2) 5.1 (2.8) 6.4 (2.6) 6.1 (3.0) 4.5 (2.6)
Corrected FMD, % C 5.9 (3.5) 8.9 (3.1)* 3.8 (0.7) 6.9 (1.9)* 6.4 (2.3) 4.7 (2.7)*
NC 4.1 (3.1) 8.7 (3.5)* 4.6 (2.4) 6.5 (2.4) * 5.9 (2.3) 4.3 (2.7)*
SRAUC C 15,855 (10,256) 32,797 (16,922)* 20,031 (9,988) 33,316 (22,285)* 21,768 (9750) 14,931 (8,112)*
NC 18,568 (5,564) 47,399 (17,470)*† 23,644 (8,267)† 37,493 (12,348)* 27,099 (9,215) 28,311 (8,960)†
Peak RH, mL/min C 247 (84) 338 (130)* 248 (68) 330 (139)* 266 (91)† 245 (77)†
NC 258 (91) 395 (128)* 285 (84) 353 (86)* 321 (94) 308 (137)
Data are means ± SD. Post hoc Holm-Bonferroni corrections: *P < 0.05 vs. pre-intervention, †P < 0.05 vs. contralateral arm. C, cuffed;
NC, noncuffed; FMD, flow-mediated dilation corrected for baseline diameter and shear stimulus; RH, reactive hyperemia; SRAUC, shear
rate area under the curve.

154 J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES

Figure 5. Repeated-measures correlations between selected variables on the x-axis and the change in flow-mediated dilation (FMD) following each
intervention. The colored circles and lines represent individual participants and their correlation, respectively, and the dashed black line is the group av-
erage regression. Ante SR, antegrade shear rate; HR, heart rate; OSI, oscillatory shear index; retro SR, retrograde shear rate; Tes, esophageal tempera-
ture, Tsk, forearm skin temperature.

negative correlation with OSI and no correlation with retro-
grade SR (Fig. 5). The multiple regression model (adjusted
R2 = 0.421, P < 0.001) shows that the relative contributions to
the explained variance were 33.6% for Tes, 31.1% for forearm
Tsk, 21.1% for HR, 12.2% for antegrade SR, 1.4% for retrograde
SR, and 0.70% for OSI (Table 3).
DISCUSSION
The purpose of this study was to determine the thermal and
hemodynamic factors that contribute to vascular responses to
acute heat stress. The main findings were that: 1) FMD
improved following whole body passive heating in the non-
cuffed arm from 3.6 (2.2)% to 9.2 (3.2)% and in the cuffed arm
from to 5.6 (3.0)% to 8.6 (4.9)%, irrespective of lower shear rates
in the cuffed versus noncuffed arms; 2) FMD improved in the
cuffed arm from 4.7 (2.9)% to 6.8 (1.5)% and in the noncuffed
arm from 5.1 (2.8)% to 6.4 (2.6)% following forearm heating-
induced rises in skin temperature, irrespective of lower shear
rates in the cuffed versus noncuffed arms; and 3) FMD
decreased following a time-matched control condition where
mean shear rates were reduced. Overall, in a multiple linear
regression model with independent variables of both thermal
and hemodynamic nature (Table 3), Tes, forearm Tsk, and HR
were the largest predictors of the change in FMD during 60 min
of whole body or local heating and a time-control period.
Our findings of increased FMD with forearm heating are
consistent with existing literature in young adults (7, 8);
however, the improved FMD with whole body heating is a
novel observation. The difference between the current study
and others likely relates to differences in measured limbs (e.
g., arm vs. leg) (11, 30) and recovery time (e.g., immediate in
the current study vs. 30–60 min in others) (9, 11, 30, 31).
Previous studies have also demonstrated that increased
shear stress is important for vascular adaptation with acute
(8) and chronic forearm heating (14) as well as repeated and

J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org 155

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES

Table 3. Multiple regression model for changes in FMD across whole-body heating, forearm heating, and time
control
Unstandardized b SE P value Tolerance Explained Variance, %
DTes,  C 2.178 1.518 0.175 0.194 33.6
DTsk,  C 0.318 0.129 0.018 0.723 31.1
DHR, beats/min 0.044 0.055 0.427 0.170 20.9
DAnte shear rate, s1 0.002 0.003 0.483 0.388 12.2
DRetro shear rate, s1 0.001 0.017 0.938 0.264 1.40
DOSI, au 0.148 4.600 0.974 0.360 0.70
Adjusted r2 0.421 - - - -
ante, antegrade; FMD, flow-mediated dilation; HR, heart rate; OSI, oscillatory shear index; retro, retrograde; r2, partial contribution to
total variance; SE, standard error of the slope coefficient; Tes, esophageal temperature; Tsk, skin temperature; b, unstandardized regres-
sion coefficients. Tolerance values >0.1 indicate acceptable collinearity.

episodic increases in core temperature (15). The obligatory
role of shear stress was demonstrated using similar experimental
designs to the current study where increases in shear rates
were attenuated in one arm via cuff inflation, and improve-
ments in FMD were only observed in the noncuffed arm. For
instance, in the cutaneous circulation, both increases in skin
blood flow (noncuffed arm only) and skin temperature con-
tribute to vascular adaptation (16). However, the impact of
changes in skin temperature on conduit arteries has not been
previously investigated. Moreover, improved FMD in older
but not young adults following local heating of the legs occurs
alongside increased core temperature (11); therefore, direct
comparison of increases in skin versus core temperature is
necessary in an attempt to delineate these mechanisms. Our
current data reveal that attenuation of shear rates did not
modify FMD for a given heating stimulus. For example, the
increase in FMD following whole body heating was not differ-
ent whether shear rates were permitted to fully increase or
not. The fact that—in the noncuffed arm—the increase in
FMD was 2.5-fold greater following whole body heating com-
pared to forearm heating (P < 0.01) further indicates an inde-
pendent influence of core temperature on FMD (Fig. 4).
The similar FMD responses to matched increases in
forearm Tsk but differing shear rates during forearm heat-
ing, suggest an independent influence of Tsk on FMD.
However, Tes explained the largest portion of variance of
the multiple regression model (34%), which is consistent
with greater changes in FMD after whole body heating.
Increased bioavailability of nitric oxide (NO) due to high
temperature (32, 33) also appears important for both
macro and microvascular function (Table 2). The view that
greater NO bioavailability drives increases in FMD is sup-
ported by in vivo cutaneous microvascular and in vitro
studies. For example, inhibition of endothelial nitric oxide
synthase (eNOS) with N-x-nitro-L-arginine (L-NNA) abol-
ished improvements in cutaneous microvascular dilator
function following 8 wk of whole body heating (34) as well
as angiogenesis of endothelial cells cultured with serum
taken from the same study participants (4). In that study,
the increased endothelial tubule formation (i.e., angiogen-
esis) was related to greater abundance of eNOS protein (4).
This finding suggests that circulating factors are integral
to NO-mediated improvements in vascular function.
Conversely, FMD was attenuated with increased retro-
grade shear rates or decreased mean shear rates in the
time control condition of the current study (Fig. 4). These
results support a role for shear stress in determining the
hemodynamic milieu and predominant cell phenotype
(22, 35–37).
Considering that the increase in forearm Tsk in the non-
cuffed arm was about half of that during forearm heating,
the larger improvement in FMD is also likely due to the
660% increase in antegrade SR in whole body heating com-
pared to the 250% increase during forearm heating (Fig. 3).
However, the influence of SR on FMD is limited by design
in the multiple regression model due to the attenuated SR
in the cuffed arm. In fact, shear rates were not a major con-
tributor to the explained variance of the model (12%), but
this should be interpreted cautiously considering the pur-
pose of this statistical model was to quantify the contribu-
tions to FMD from factors other than shear stress. In the
absence of appreciable changes in stroke volume during
passive heat stress (38), changes in cardiac output are
mediated by HR, and the change in HR explained 21% of
the model (Table 3). Moreover, the change in HR was
related to the change in FMD (r = 0.58, Fig. 5C). As a result,
whole body heating appears to be a more potent stimulus
for eliciting increased systemic shear stress [and poten-
tially circulating factors related to temperature (29)], due
to the greater increases in core temperature and HR (i.e.,
cardiac output). Indeed, whole body heating increased HR
by 40 beats/min in comparison to an increase of 7 beats/
min during forearm heating (Table 1). Because shear stress
increases with cardiac output, the obligatory rise in shear
for vascular adaptation—which has been demonstrated
many times (14, 15, 39–41)—should not be discounted or
diminished in this study.
Brunt et al. (29) showed that both temperature per se and
circulating factors reduced basal superoxide production,
whereas only temperature upregulated heat shock protein
70 in cultured human endothelial cells. The effects of heat
shock proteins on macrovascular function are unclear, but
heat shock protein 90, at least, is involved in the activation
of eNOS (42, 43). Consistent with the findings of Brunt et al.
(29), the current data support a role for the direct effect of
temperature in vascular responses to heating. Indeed, Tes
explained 34% of the variance of the model (Table 3) and
was related to the change in FMD (r = 0.59, Fig. 5F). It has
also been demonstrated that reactive oxygen species (ROS)
impair FMD and prevention of ROS (via vitamin C) can
reverse endothelial dysfunction in aging (44). There is evi-
dence to suggest that heat shock protein 70 (which is

156 J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org

Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES
upregulated by heat) may be protective against inflamma-
tory and ROS responses to in vitro stress (e.g., hypoxia-reoxy-
genation) (29), and that heat therapy could therefore be
protective against ischemia-reperfusion in vivo (31). However,
though the role of heat shock proteins requires further
research, it has been speculated that upregulation of heat
shock proteins in patient groups via heat therapy could be
beneficial for insulin resistance and related vascular dysfunc-
tion (45, 46).
It has been acknowledged that a better understanding of
the factors that cause improvements in vascular function is
required in order to prescribe optimal temperatures, dura-
tions, and modes of heating (47, 48). Indeed, passive heating
has been evaluated in individuals with CVD [e.g., heart fail-
ure (49, 50), peripheral artery disease (51)], and those at risk
of CVD [e.g., aging (11), coronary risk factors (52), obesity (5),
and polycystic ovary syndrome (53)]; however, there is little
consistency between studies regarding passive heating pro-
tocols. Traditionally, heat acclimation protocols aim to
increase core temperature above 38.5 C for 60 min (54, 55)
and this approach has been used in short [e.g., 7 days (56)]
and longer term [e.g., 8 wk (3)] studies to induce both ther-
moregulatory and vascular adaptations, respectively.
However, such stringent protocols may be difficult to imple-
ment depending on the mode of heating, patient capacity/
tolerance, or motivation. Therefore, it is important to deter-
mine the primary contributors to vascular adaptation to
determine how best to modify acclimation parameters (tem-
perature, mode, and time) for various groups. We have dem-
onstrated herein that protocols increasing skin temperature
and core temperature—and not only shear—are likely to be
beneficial. The implications of these observations indicate
that interventions increasing skin temperature rapidly (e.g.,
sauna) but with low tolerance times may also be of benefit to
vascular function compared to the traditional prolonged
core temperature >38.5 C. However, evaluation of the time
required to elicit responses are scarce. For example, as little
as 10–30 min of forearm heating has been reported to
increase FMD in young healthy participants (7, 8), whereas
10–20 min of sauna bathing did not change FMD in older
healthy people (9). Conversely, 45–60 min of lower leg heat-
ing improved femoral artery FMD in older but not young
adults (11) and may be of benefit in individuals with limited
exercise capacity (e.g., individuals with spinal cord injury)
(10). Clearly, the need for additional evidence on the dura-
tion and intensity of heating is important, but it is likely that
even small doses of heat repeated over time are of benefit for
the reduction of CVD risk (1, 49).
Experimental Considerations
The inclusion of a time control condition in our study is an
important strength. Not only do our data support previous
reports of a detrimental impact of retrograde shear rates on
endothelial function (18, 22), it also demonstrates that the
improvements in FMD during heating are not simply due to
time or other factors during our experimental protocol.
Although FMD was measured supine before and after heating,
hemodynamic responses were performed in a semirecumbent
position at baseline and during heating to account for the
effects of supine posture. Another important consideration in
J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org
Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
our study is that blood viscosity is a key determinant of shear
stress at the arterial wall and this could be influenced by
losses in plasma volume due to sweating. For instance, viscos-
ity-related changes in FMD have been observed following re-
moval of 1 L of blood volume (57), but the 1%–2% loss of
plasma volume during passive heat stress via sweating (56) is
unremarkable in comparison. Moreover, the effects of hemo-
concentration related to sweating are offset by hyperthermia
resulting in no change of blood viscosity (58).
Limitations
The shear dependency of blood viscosity is well-established
(59), but it was recently demonstrated that blood exhibits
shear-thinning properties where viscosity decreases at higher
shear rates (60). Given that shear rates vary widely in heating
protocols, future studies should aim to match the shear rates
experienced in the study conditions to those used for viscosity
measures using a range of shear rates and an exponential
decay model. The current results are only relevant to the
brachial artery and the forearm, thus differences could exist
between limbs and these data should be replicated in the leg
(e.g., superficial femoral, popliteal arteries). Although the use
of a water-perfused suit is not a common mode of heating in
everyday life, the passive heating of skin temperature resem-
bles that of natural heating modes (e.g., sauna, hot baths) but
the medium of heat transfer does not directly mimic air or
water. Lastly, but certainly not least, we cannot extrapolate
our findings to female participants or other racial groups.
Previous reports have suggested that sex differences exist in
endothelial function following interventions that manipulate
shear stress (61, 62), so it is entirely possible that women could
respond differently to our current protocols. It is not entirely
clear whether the hemodynamic changes to heating would be
the same between sexes or across the menstrual cycle when
women demonstrate phase-dependent fluctuations in core
temperature (63), but these questions should be addressed in
the future. Few studies have addressed racial differences in
vascular function but evidence indicates that cutaneous vaso-
dilation in response to local heating (64) and vascular con-
ductance to forearm exercise (65) are lower in black compared
to white men. Further studies are required to replicate our
findings in other ethnic and racial groups.
Conclusion
This study demonstrates that increases in forearm skin
and core temperatures are important determinants of the
vascular responses of the brachial artery to passive heat
stress. In addition to heat-induced increases in shear stress,
limb heating likely increases local factors such as nitric oxide
to improve flow-mediated dilation. Future studies should
address different combinations of skin/core temperatures
and shear stress, with a range of acute and chronic dura-
tions, on vascular outcomes. Future focus should now be
directed particularly to populations at risk of CVD and with
limited exercise capacity.
ACKNOWLEDGMENTS
The authors thank the volunteers for time and Connor Howe
for help with data collection.
157
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES
GRANTS
This project was supported by a Natural Sciences and
Engineering Research Council of Canada (NSERC) Discovery
Grant and a Canada Research Chair to P. N. Ainsle. G. Coombs
was supported by NSERC and Killam doctoral scholarships.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by
the authors.
AUTHOR CONTRIBUTIONS
G.B.C. and P.N.A. conceived and designed research; G.B.C.,
J.C.T., D.A.S., J.M.J.R.C., D.J.W., and A.P. performed experiments;
G.B.C. analyzed data; G.B.C., J.C.T., D.A.S., J.M.J.R.C., D.J.W., A.P.,
and P.N.A. interpreted results of experiments; G.B.C. and A.P. pre-
pared figures; G.B.C. drafted manuscript; G.B.C., J.C.T., J.M.J.R.C.,
D.J.W., A.P., and P.N.A. edited and revised manuscript; G.B.C.,
J.C.T., D.A.S., J.M.J.R.C., D.J.W., A.P., and P.N.A. approved
final version of manuscript.
REFERENCES
1. Laukkanen T, Khan H, Zaccardi F, Laukkanen JA. Association
between sauna bathing and fatal cardiovascular and all-cause mor-
tality events. JAMA Intern Med 175: 542–548, 2015. doi:10.1001/
jamainternmed.2014.8187.
2. Ukai T, Iso H, Yamagishi K, Saito I, Kokubo Y, Yatsuya H, Muraki I,
Eshak ES, Sawada N, Tsugane S. Habitual tub bathing and risks of
incident coronary heart disease and stroke. Heart 106: 732–737,
2020. doi:10.1136/heartjnl-2019-315752.
3. Brunt VE, Howard MJ, Francisco MA, Ely BR, Minson CT. Passive
heat therapy improves endothelial function, arterial stiffness and
blood pressure in sedentary humans. J Physiol 594: 5329–5342,
2016. doi:10.1113/JP272453.
4. Brunt VE, Weidenfeld-Needham KM, Comrada LN, Francisco MA,
Eymann TM, Minson CT. Serum from young, sedentary adults who
underwent passive heat therapy improves endothelial cell angiogene-
sis via improved nitric oxide bioavailability. Temperature (Austin) 6:
169–178, 2019. doi:10.1080/23328940.2019.1614851.
5. Hoekstra SP, Bishop NC, Faulkner SH, Bailey SJ, Leicht CA. Acute
and chronic effects of hot water immersion on inflammation and me-
tabolism in sedentary, overweight adults. J Appl Physiol (1985) 125:
2008–2018, 2018. doi:10.1152/japplphysiol.00407.2018.
6. Laukkanen T, Kunutsor SK, Zaccardi F, Lee E, Willeit P, Khan H,
Laukkanen JA. Acute effects of sauna bathing on cardiovascular
function. J Hum Hypertens 32: 129–138, 2018. doi:10.1038/s41371-
017-0008-z.
7. Cheng JL, Au JS, MacDonald MJ. Peripheral artery endothelial func-
tion responses to altered shear stress patterns in humans. Exp
Physiol 104: 1126–1135, 2019. doi:10.1113/EP087597.
8. Tinken TM, Thijssen DH, Hopkins N, Black MA, Dawson EA, Minson
CT, Newcomer SC, Laughlin MH, Cable NT, Green DJ. Impact of
shear rate modulation on vascular function in humans. Hypertension
54: 278–285, 2009. doi:10.1161/HYPERTENSIONAHA.109.134361.
9. Gravel H, Coombs GB, Behzadi P, Marcoux-Cle ment V, Barry H,
Juneau M, Nigam A, Gagnon D. Acute effect of Finnish sauna bath-
ing on brachial artery flow-mediated dilation and reactive hyperemia
in healthy middle-aged and older adults. Physiol Rep 7: e14166,
2019. doi:10.14814/phy2.14166.
10. Coombs GB, Barak OF, Phillips AA, Mijacika T, Sarafis ZK, Lee AH,
Squair JW, Bammert TD, DeSouza NM, Gagnon D, Krassioukov
AV, Dujic Z, DeSouza CA, Ainslie PN. Acute heat stress reduces
biomarkers of endothelial activation but not macro- or microvascular
dysfunction in cervical spinal cord injury. Am J Physiol Heart Circ
Physiol 316: H722–H733, 2019. doi:10.1152/ajpheart.00693.2018.
11. Romero SA, Gagnon D, Adams AN, Cramer MN, Kouda K, Crandall
CG. Acute limb heating improves macro- and microvascular dilator
158 J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org
Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
function in the leg of aged humans. Am J Physiol Heart Circ Physiol
312: H89–H97, 2017. doi:10.1152/ajpheart.00519.2016.
12. Thomas KN, van Rij AM, Lucas SJ, Cotter JD. Lower-limb hot-water
immersion acutely induces beneficial hemodynamic and cardiovascular
responses in peripheral arterial disease and healthy, elderly controls.
Am J Physiol Regul Integr Comp Physiol 312: R281–R291, 2017.
doi:10.1152/ajpregu.00404.2016.
13. Green DJ, Hopman MTE, Padilla J, Laughlin MH, Thijssen DH.
Vascular adaptation to exercise in humans: role of hemodynamic stimuli.
Physiol Rev 97: 495–528, 2017. doi:10.1152/physrev.00014.2016.
14. Naylor LH, Carter H, FitzSimons MG, Cable NT, Thijssen DH,
Green DJ. Repeated increases in blood flow, independent of exer-
cise, enhance conduit artery vasodilator function in humans. Am J
Physiol Heart Circ Physiol 300: H664–H669, 2011. doi:10.1152/
ajpheart.00985.2010.
15. Carter HH, Spence AL, Atkinson CL, Pugh CJ, Naylor LH, Green
DJ. Repeated core temperature elevation induces conduit artery ad-
aptation in humans. Eur J Appl Physiol 114: 859–865, 2014.
doi:10.1007/s00421-013-2817-2.
16. Carter HH, Spence AL, Atkinson CL, Pugh CJ, Cable NT, Thijssen
DH, Naylor LH, Green DJ. Distinct effects of blood flow and temper-
ature on cutaneous microvascular adaptation. Med Sci Sports Exerc
46: 2113–2121, 2014. doi:10.1249/MSS.0000000000000349.
17. Restaino RM, Walsh LK, Morishima T, Vranish JR, Martinez-Lemus
LA, Fadel PJ, Padilla J. Endothelial dysfunction following prolonged
sitting is mediated by a reduction in shear stress. Am J Physiol Heart
Circ Physiol 310: H648–H653, 2016. doi:10.1152/ajpheart.00943.2015.
18. Thijssen DH, Dawson EA, Tinken TM, Cable NT, Green DJ. Retrograde
flow and shear rate acutely impair endothelial function in humans.
Hypertension 53: 986–992, 2009. doi:10.1161/HYPERTENSIONAHA.
109.131508.
19. Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: a flexible sta-
tistical power analysis program for the social, behavioral, and bio-
medical sciences. Behav Res Methods 39: 175–191, 2007. doi:10.
3758/BF03193146.
20. Thijssen DH, Black MA, Pyke KE, Padilla J, Atkinson G, Harris RA,
Parker B, Widlansky ME, Tschakovsky ME Green DJ. Assessment
of flow-mediated dilation in humans: a methodological and physio-
logical guideline. Am J Physiol Heart Circ Physiol 300: H2–H12,
2011. doi:10.1152/ajpheart.00471.2010.
21. Thijssen DH, Bruno RM, van Mil AC, Holder SM, Faita F, Greyling
A, Zock PL, Taddei S, Deanfield JE, Luscher T, Green DJ, Ghiadoni
L. Expert consensus and evidence-based recommendations for the
assessment of flow-mediated dilation in humans. Eur Heart J 40:
2534–2547, 2019. doi:10.1093/eurheartj/ehz350.
22. Jenkins NT, Padilla J, Boyle LJ, Credeur DP, Laughlin MH, Fadel
PJ. Disturbed blood flow acutely induces activation and apoptosis of
the human vascular endothelium. Hypertension 61: 615–621, 2013.
doi:10.1161/HYPERTENSIONAHA.111.00561.
23. Woodman RJ, Playford DA, Watts GF, Cheetham C, Reed C, Taylor
RR, Puddey IB, Beilin LJ, Burke V, Mori TA, Green D. Improved
analysis of brachial artery ultrasound using a novel edge-detection
software system. J Appl Physiol (1985) 91: 929–937, 2001.
doi:10.1152/jappl.2001.91.2.929.
24. Rosenberry R, Nelson MD. Reactive hyperemia: a review of methods,
mechanisms, and considerations. Am J Physiol Regul Integr Comp
Physiol 318: R605–R618, 2020. doi:10.1152/ajpregu.00339.2019.
25. Atkinson G, Batterham AM. The percentage flow-mediated dilation
index: a large-sample investigation of its appropriateness, potential
for bias and causal nexus in vascular medicine. Vasc Med 18: 354–
365, 2013. doi:10.1177/1358863X13508446.
26. Shenouda N, Gillen JB, Gibala MJ, MacDonald MJ. Changes in brach-
ial artery endothelial function and resting diameter with moderate-inten-
sity continuous but not sprint interval training in sedentary men. J Appl
Physiol (1985) 123: 773–780, 2017. doi:10.1152/japplphysiol.00058.2017.
27. Bakdash JZ, Marusich LR. Repeated measures correlation. Front
Psychol 8:456, 2017. doi:10.3389/fpsyg.2017.00456.
28. Vargas NT, Chapman CL, Johnson BD, Gathercole R, Schlader ZJ.
Skin wettedness is an important contributor to thermal behavior dur-
ing exercise and recovery. Am J Physiol Regul Integr Comp Physiol
315: R925–R933, 2018. doi:10.1152/ajpregu.00178.2018.
29. Brunt VE, Wiedenfeld -Needham K, Comrada LN, Minson CT.
Passive heat therapy protects against endothelial cell hypoxia-
 DETERMINANTS OF HEAT-INDUCED VASCULAR RESPONSES
reoxygenation via effects of elevations in temperature and circulat-
ing factors. J Physiol 596: 4831–4845, 2018. doi:10.1113/JP276559.
30. Thomas KN, van Rij AM, Lucas SJ, Gray AR, Cotter JD. Substantive
hemodynamic and thermal strain upon completing lower-limb hot-
water immersion; comparisons with treadmill running. Temperature
(Austin) 3: 286–297, 2016. doi:10.1080/23328940.2016.1156215.
31. Brunt VE, Jeckell AT, Ely BR, Howard MJ, Thijssen DH, Minson CT.
Acute hot water immersion is protective against impaired vascular
function following forearm ischemia-reperfusion in young healthy
humans. Am J Physiol Regul Integr Comp Physiol 311: R1060–R1067,
2016. doi:10.1152/ajpregu.00301.2016.
32. Kellogg DL Jr, Liu Y, Kosiba IF, O’Donnell D. Role of nitric oxide in
the vascular effects of local warming of the skin in humans. J Appl
Physiol (1985) 86: 1185–1190, 1999. doi:10.1152/jappl.1999.86.4.1185.
33. Minson CT, Berry LT, Joyner MJ. Nitric oxide and neurally mediated
regulation of skin blood flow during local heating. J Appl Physiol
(1985) 91: 1619–1626, 2001. doi:10.1152/jappl.2001.91.4.1619.
34. Brunt VE, Eymann TM, Francisco MA, Howard MJ, Minson CT.
Passive heat therapy improves cutaneous microvascular function in
sedentary humans via improved nitric oxide-dependent dilation. J
Appl Physiol (1985) 121: 716–723, 2016. doi:10.1152/japplphysiol.
00424.2016.
35. Chistiakov DA, Orekhov AN, Bobryshev YV. Effects of shear stress
on endothelial cells: go with the flow. Acta Physiol (Oxf) 219: 382–
408, 2017. doi:10.1111/apha.12725.
36. Laughlin MH, Newcomer SC, Bender SB. Importance of hemody-
namic forces as signals for exercise-induced changes in endothelial
cell phenotype. J Appl Physiol (1985) 104: 588–600, 2008. doi:10.
1152/japplphysiol.01096.2007.
37. Tremblay JC, Grewal AS, Pyke KE. Examining the acute effects of
retrograde versus low mean shear rate on flow-mediated dilation. J
Appl Physiol (1985) 126: 1335–1342, 2019. doi:10.1152/japplphysiol.
01065.2018.
38. Crandall CG, Wilson TE. Human cardiovascular responses to passive
heat stress. Compr Physiol 5: 17–43, 2015. doi:10.1002/cphy.c140015.
39. Green D, Cheetham C, Mavaddat L, Watts K, Best M, Taylor R,
O’Driscoll G. Effect of lower limb exercise on forearm vascular func-
tion: contribution of nitric oxide. Am J Physiol Heart Circ Physiol 283:
H899–H907, 2002. doi:10.1152/ajpheart.00049.2002.
40. Green DJ, Carter HH, Fitzsimons MG, Cable NT, Thijssen DH,
Naylor LH. Obligatory role of hyperaemia and shear stress in micro-
vascular adaptation to repeated heating in humans. J Physiol 588:
1571–1577, 2010. doi:10.1113/jphysiol.2010.186965.
41. Green DJ, Maiorana A, O'Driscoll G, Taylor R. Effect of exercise
training on endothelium-derived nitric oxide function in humans. J
Physiol 561: 1–25, 2004. doi:10.1113/jphysiol.2004.5611, 10.1113/
jphysiol.2004.068197. doi:10.1113/jphysiol.2004.068197.
42. Fujii N, Zhang SY, McNeely BD, Nishiyasu T, Kenny GP. Heat shock
protein 90 contributes to cutaneous vasodilation through activating
nitric oxide synthase in young male adults exercising in the heat. J
Appl Physiol (1985) 123: 844–850, 2017. doi:10.1152/japplphysiol.
00446.2017.
43. Shastry S, Joyner MJ. Geldanamycin attenuates NO-mediated dila-
tion in human skin. Am J Physiol Heart Circ Physiol 282: H232–
H236, 2002. doi:10.1152/ajpheart.2002.282.1.H232.
44. Wray DW, Nishiyama SK, Harris RA, Zhao J, McDaniel J,
Fjeldstad AS, Witman MA, Ives SJ, Barrett-O’Keefe Z,
Richardson RS. Acute reversal of endothelial dysfunction in
the elderly after antioxidant consumption. Hypertension 59:
818–824, 2012. doi:10.1161/HYPERTENSIONAHA.111.189456.
45. McCarty MF, Barroso-Aranda J, Contreras F. Regular thermal ther-
apy may promote insulin sensitivity while boosting expression of en-
dothelial nitric oxide synthase – effects comparable to those of
exercise training. Med Hypotheses 73: 103–105, 2009. doi:10.1016/j.
mehy.2008.12.020.
46. Molina MN, Ferder L, Manucha W. Emerging role of nitric oxide and
heat shock proteins in insulin resistance. Curr Hypertens Rep 18:1,
2016. doi:10.1007/s11906-015-0615-4.
47. Cheng JL, MacDonald MJ. Effect of heat stress on vascular out-
comes in humans. J Appl Physiol 126: 771–781, 2019. doi:10.1152/
japplphysiol.00682.2018.
48. Ely BR, Clayton ZS, McCurdy CE, Pfeiffer J, Minson CT. Meta-inflam-
mation and cardiometabolic disease in obesity: Can heat therapy
J Appl Physiol  doi:10.1152/japplphysiol.00502.2020  www.jap.org
Downloaded from journals.physiology.org/journal/jappl (185.016.025.051) on October 31, 2023.
help? Temperature (Austin) 5: 9–21, 2018. doi:10.1080/23328
940.2017.1384089.
49. Kihara T, Biro S, Imamura M, Yoshifuku S, Takasaki K, Ikeda Y,
Otuji Y, Minagoe S, Toyama Y, Tei C. Repeated sauna treatment
improves vascular endothelial and cardiac function in patients with
chronic heart failure. J Am Coll Cardiol 39: 754–759, 2002.
doi:10.1016/S0735-1097(01)01824-1.
50. Sobajima M, Nozawa T, Fukui Y, Ihori H, Ohori T, Fujii N, Inoue H.
Waon therapy improves quality of life as well as cardiac function and
exercise capacity in patients with chronic heart failure. Int Heart J
56: 203–208, 2015. doi:10.1536/ihj.14-266.
51. Akerman AP, Thomas KN, van Rij AM, Body ED, Alfadhel M, Cotter
JD. Heat therapy vs. supervised exercise therapy for peripheral arterial
disease: a 12-wk randomized, controlled trial. Am J Physiol Heart Circ
Physiol 316: H1495–H1506, 2019. doi:10.1152/ajpheart.00151.2019.
52. Imamura M, Biro S, Kihara T, Yoshifuku S, Takasaki K, Otsuji Y,
Minagoe S, Toyama Y, Tei C. Repeated thermal therapy improves
impaired vascular endothelial function in patients with coronary risk
factors. J Am Coll Cardiol 38: 1083–1088, 2001. doi:10.1016/S0735-
1097(01)01467-X.
53. Ely BR, Francisco MA, Halliwill JR, Bryan SD, Comrada LN, Larson
EA, Brunt VE, Minson CT. Heat therapy reduces sympathetic activ-
ity and improves cardiovascular risk profile in obese women who
are obese with polycystic ovary syndrome. Am J Physiol Regul
Integr Comp Physiol 317: R630–R640, 2019. doi:10.1152/ajpregu.
00078.2019.
54. Fox RH, Goldsmith R, Kidd DJ, Lewis HE. Acclimatization to heat in
man by controlled elevation of body temperature. J Physiol 166:
530–547, 1963. doi:10.1113/jphysiol.1963.sp007121.
55. Taylor NAS. Human heat adaptation. Compr Physiol 4: 325–365,
2014. doi:10.1002/cphy.c130022.
56. Barry H, Chaseling GK, Moreault S, Sauvageau C, Behzadi P,
Gravel H, Ravanelli N, Gagnon D. Improved neural control of body
temperature following heat acclimation in humans. J Physiol 598:
1223–1234, 2020. doi:10.1113/JP279266.
57. Tremblay JC, Hoiland RL, Howe CA, Coombs GB, Vizcardo-
Galindo GA, Figueroa-Mujíca RJ, Bermudez D, Gibbons TD, Stacey
BS, Bailey DM, Tymko MM, MacLeod DB, Gasho C, Villafuerte FC,
Pyke KE, Ainslie PN. Global REACH 2018: high blood viscosity and
hemoglobin concentration contribute to reduced flow-mediated dila-
tion in high-altitude excessive erythrocytosis. Hypertension 73:
1327–1335, 2019. doi:10.1161/HYPERTENSIONAHA.119.12780.
58. Buono MJ, Krippes T, Kolkhorst FW, Williams AT, Cabrales P.
Increases in core temperature counterbalance effects of haemocon-
centration on blood viscosity during prolonged exercise in the heat.
Exp Physiol 101: 332–342, 2016. doi:10.1113/EP085504.
59. Dormandy JA. Measurement of whole-blood viscosity. In: Clinical
Aspects of Blood Viscosity and Cell Deformability, edited by Lowe
GD, Barbenel JC, Forbes CD. London, UK: Springer, 1981, p. 67–78.
60. Leo JA, Simmonds MJ, Sabapathy S. Shear-thinning behaviour of
blood in response to active hyperaemia: implications for the assess-
ment of arterial shear stress-mediated dilation. Exp Physiol 105:
244–257, 2020. doi:10.1113/EP088226.
61. Tremblay JC, Stimpson TV, Pyke KE. Evidence of sex differences in
the acute impact of oscillatory shear stress on endothelial function. J
Appl Physiol (1985) 126: 314–321, 2019. doi:10.1152/japplphysiol.00729.
2018.
62. Vranish JR, Young BE, Kaur J, Patik JC, Padilla J, Fadel PJ.
Influence of sex on microvascular and macrovascular responses to
prolonged sitting. Am J Physiol Heart Circ Physiol 312: H800–H805,
2017. doi:10.1152/ajpheart.00823.2016.
63. Charkoudian N, Stachenfeld NS. Reproductive hormone influences
on thermoregulation in women. Compr Physiol 4, 793–804, 2014.
doi:10.1002/cphy.c130029.
64. Patik JC, Curtis BM, Nasirian A, Vranish JR, Fadel PJ, Brothers
RM. Sex differences in the mechanisms mediating blunted cutane-
ous microvascular function in young black men and women. Am J
Physiol Heart Circ Physiol 315: H1063–H1071, 2018. doi:10.1152/
ajpheart.00142.2018.
65. Barbosa TC, Kaur J, Stephens BY, Akins JD, Keller DM, Brothers
RM, Fadel PJ. Attenuated forearm vascular conductance responses
to rhythmic handgrip in young African-American compared with
Caucasian-American men. Am J Physiol Heart Circ Physiol 315:
H1316–H1321, 2018. doi:10.1152/ajpheart.00387.2018.
159