1 23

European Journal of AppliedPhysiology ISSN 1439-6319Volume 112Number 10 Eur J Appl Physiol (2012) 112:3619-3627DOI 10.1007/s00421-012-2349-1

G tolerance vis-à-vis pressure-distensionand pressure-flow relationships of legarteries

Ola Eiken, Igor Mekjavic, PatrikSundblad & Roger Kölegård

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag. This e-offprint is for personal use only

and shall not be self-archived in electronic

repositories. If you wish to self-archive your

work, please use the accepted author’s

version for posting to your own website or

your institution’s repository. You may further

deposit the accepted author’s version on a

funder’s repository at a funder’s request,

provided it is not made publicly available until

12 months after publication.

ORIGINAL ARTICLE

G tolerance vis-a-vis pressure-distension and pressure-flowrelationships of leg arteries

Ola Eiken • Igor Mekjavic • Patrik Sundblad •

Roger Kolegard

Received: 16 September 2011 / Accepted: 31 January 2012 / Published online: 16 February 2012

� Springer-Verlag 2012

Abstract During increased gravitoinertial (G) load in the

head-to-foot direction, pressures in dependent vascular

beds are commonly raised to levels capable of distending

precapillary vessels, which, in turn, may reduce arterial

pressure, and hence compromise the capacity to withstand

G load (G tolerance). We hypothesized that distensibility in

precapillary leg vessels would be lower in a group of

subjects possessing high G tolerance (H; n = 7; relaxed G

tolerance = 6.6 ± 0.8 G) than in a group with low G tol-

erance (L; n = 8; G tolerance = 3.9 ± 0.3 G). The groups

were matched with regard to gender, age, weight, height,

and resting arterial pressure. Arterial pressure-distension

and pressure-flow experiments were performed with the

subject supine in a pressure chamber with a lower leg

protruding to the outside. Increased intravascular pressure

in the blood vessels of the outside leg was accomplished

by stepwise increasing chamber pressure to 240 mmHg.

Diameter and flow in the posterior tibial artery were

measured by ultrasonographic/Doppler techniques. Pres-

sure-induced increments in arterial diameter and flow were

more pronounced (p \ 0.03) in the L (14.1 ± 4.2% and

32 ± 21 ml/min respectively) than in the H (1.7 ± 5.0%

and 1.6 ± 25 ml/min) group, and the pressure thresholds at

which these increments commenced were lower (by 52 and

48 mmHg, respectively) in the L than in the H group

(p \ 0.04). Negative correlations were observed between

G tolerance and the increments in diameter and flow

(p \ 0.02). Thus, the wall stiffness of precapillary leg

vessels is greater in individuals with high relaxed G

tolerance; whether a causal relationship exists remains to

be established.

Keywords Acceleration � Arterial stiffness �Distensibility � G-level tolerance � Precapillary vessels �Total peripheral resistance

Introduction

Pilots flying high-performance aircraft are exposed to high-

sustained acceleration with the gravitoinertial vector

directed head-to-foot (?Gz). An individuals’ capacity to

withstand increased ?Gz load is termed G-level tolerance

(in the following, G tolerance refers to G-level tolerance

unless otherwise stated) and is governed by his/her arterial

pressure response; the predominant challenge for the cir-

culatory system is to maintain adequate perfusion of the

brain in the face of the G-induced exaggerated pressure

drop along the arteries from the heart to the head (Wood

and Lambert 1952; Wood 1987; Burton and Whinnery

2008). G tolerance varies considerably in relaxed individ-

uals, especially when the G load is increased at a slow rate

(Burton and Whinnery 2008). Experience from our own

laboratory agrees well with that of others (for reviews see

Burton and Whinnery 2008) that relaxed G tolerance

amongst individuals of the same gender, and who are also

fairly well matched with regards to age and body mass,

may vary from 2.5 to 8.5 G. The mechanisms underlying

Communicated by Dag Linnarsson.

O. Eiken (&) � P. Sundblad � R. Kolegard

Department of Environmental Physiology, School of Technology

and Health, Royal Institute of Technology, Berzelius v 13,

Solna, 171 65 Stockholm, Sweden

e-mail: ola.eiken@sth.kth.se

I. Mekjavic

Department of Automation Biocybernetics and Robotics,

Jozef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia

123

Eur J Appl Physiol (2012) 112:3619–3627

DOI 10.1007/s00421-012-2349-1

Author's personal copy

the large interindividual variation in G tolerance are only

partly understood. Factors known to affect G tolerance

include resting arterial pressure and the vertical distance

between the heart and head (Klein et al. 1969). It is com-

monly assumed that also arterial baroreflex sensitivity

determines relaxed G tolerance even though evidence

to support this notion is indirect and inconclusive (cf

Newman et al. 1998; Convertino 1998).

We reasoned that the distensibility of precapillary blood

vessels in the lower body might influence relaxed G toler-

ance. Marked elevations of intravascular pressures distend

not only veins, but also peripheral arteries and presumably

arterioles (Eiken and Kolegard 2001). In the sitting position,

high ?Gz loads substantially increase pressures in depen-

dent vessels, since large hydrostatic pressure components

add to local intravascular pressures. Judging from previous

experiments, the wall stiffness of precapillary vessels in the

legs may not be sufficient to withstand such G-induced

increments in local intravascular pressure. Thus, once

intravascular pressure exceeds 250–270 mmHg in the lower

leg, local arteries distend and local arterial flows increase

promptly; the latter finding has been attributed to arteriolar

pressure distension (Eiken and Kolegard 2001, 2004; Eiken

et al. 2008). Pressure distension of precapillary vessels in

the lower body may result in a substantial drop in total

peripheral blood-flow resistance, which, in turn, may

compromise arterial pressure regulation and hence limit G

tolerance.

Accordingly, the purpose of the present study was to

compare the distensibility of precapillary leg vessels of

individuals with high versus low relaxed G-tolerance. We

hypothesized that precapillary wall stiffness would be

higher in individuals with high G tolerance than in those

with low G tolerance.

Methods

Subjects

The protocol and experimental procedures of the study

were approved by the Regional Human Ethics Committee

in Stockholm. After giving their informed consent, 15

healthy males took part in the study. They were recruited

amongst individuals who had previously participated in

G-tolerance experiments. Subjects were assigned to two

groups, high G tolerance (H; n = 7) and low G tolerance

(L; n = 8) individuals. The inclusion criteria were a

relaxed gradual onset-rate (see below) G tolerance of

C5.5 G for the H-group and of B4.2 G for the L-group,

these thresholds corresponding to the 80th and 20th per-

centile, respectively, of G-tolerance values obtained in our

laboratory.

Study protocol

Each subject reported to the laboratory on two different

days. On the first day, the subject underwent criterion tests

followed by G-tolerance tests in the human-use centrifuge.

On the second day, the pressure-distension and pressure–

flow relationships of an artery in his lower leg were

determined using a pressure-chamber model.

Criterion tests

Body height and mass were measured using standard

techniques. Resting systolic and diastolic arterial pressures

(SAP, DAP) were measured with the subject in a supine

position using a standard sphygmomanometric (Riva

Rocci) technique, with a 13 cm wide pressure cuff placed

around the right upper arm and with the heart as the

pressure reference level.

G-tolerance tests

G-level tolerance was determined employing gradual onset-

rate (GOR) G-time profiles. The G load was increased by

0.1 G/s, starting from ?1.4 Gz and terminating at the G

load resulting in impaired vision (see below). G tolerance

was defined as the average of three such GOR trials with an

inter-trial variation of B0.25 G, consecutive tests being

interspersed by about 5 min. If the EMG recordings (see

below) revealed that the subject did not sit relaxed during

a G-tolerance test, the test was terminated and restarted

after 5 min.

Centrifuge

The experiments were conducted in a 7.25 m radius human-

use centrifuge (ASEA, Sweden) at the Royal Institute of

Technology in Stockholm. During the trials the subject sat

in the centrifuge gondola in a seat with the back angle

reclining 28� from the vertical. During the tests, he wore a

short-sleeved T-shirt and long cotton trousers. Air temper-

ature in the gondola was maintained within the range

23–25�C. The centrifuge was controlled by an ‘‘open-loop’’

system employing a pre-set computer-derived G-time pro-

file. The subject, who was monitored via closed circuit

television, was not provided with any visual G-time feed-

back signal. Gz was measured by means of an analogue

accelerometer mounted in front of the subject at a vertical

level approximately corresponding to that of his heart.

Central and peripheral vision

A light bar comprising a central red light and two green

lights positioned at a 60� angle in relation to the subjects’

3620 Eur J Appl Physiol (2012) 112:3619–3627

123

Author's personal copy

eyes was used for assessment of impairments in central and

peripheral vision. Central and peripheral vision was rated

using a four-point scale (clear, dim, gray, light loss; Buick

et al. 1995). The subject was instructed to terminate the G

exposure by releasing a handgrip switch once both or one

of the peripheral lights were no longer visible (light loss)

and/or once the central light was gray/white.

Heart rate (HR) and mean arterial pressure (MAP)

Heart rate (HR) was derived from electrocardiographic

recordings using a cardiograph (Datex-Engstrom, Instru-

mentation Corp, Helsinki, Finland), electrodes being

positioned in a precordial 5-lead arrangement. mean arte-

rial pressure (MAP) was measured using a volume-clamp

technique (Portapres, TNO, Amsterdam, The Netherlands)

with the pressure cuff placed around the middle-phalanx of

the third or fourth finger of the right hand and the reference

pressure transducer taped to the skin of the temple, at the

level of the eyes. The right arm was supported by an

armrest adjusted so that the distal portions of the fingers

were at a vertical level corresponding to the jugulum sterni.

Prior to each experiment, arterial pressure values obtained

with the volume-clamp technique were compared, and

found to agree well, with those obtained from the right

brachial artery using a standard Riva Rocci method.

Electromyographic (EMG) activity was monitored

continuously in the right vastus lateralis and rectus abdo-

minis muscles using a Bagnoli-4 EMG system in combi-

nation with EMG-works computer software (DelSys Inc.

Boston, MA). The EMG signals were detected by bipolar

surface electrodes positioned over the mid-portion of the

muscles. Maximal voluntary isometric contractions (MVC;

attempted knee extension at 0� knee angle and hip flexion

at 90� hip angle) were performed prior to each experiment,

and the EMG recordings obtained during these maneuvers

were used as reference values for the EMG values obtained

during the experiments. To ensure that the subject kept

his glottis open, and hence refrained from performing

straining maneurers, he was instructed to sit as relaxed as

possible and to breathe quietly (to continuously chat with

the experimenters) during the G exposures. If the root-

mean-square EMG activity increased by [15% of MVC

from baseline either in the vastus lateralis or rectus abdo-

minis muscles, the experiment was disregarded and the

G-time profile was repeated in 5 min. EMG values were

used solely as data inclusion/exclusion criteria and were

not further analyzed.

All signals were transmitted via slip rings and continu-

ously monitored and/or recorded on magnetic tape (TEAC

DAT recorder RD-135T, TEAC Corp, Japan), video tape

(AG 7350, Panasonic, Japan) or on a computer (Optiplex

Gx1, Dell, USA).

Vascular pressure-distension relationships

Methods and instrumentation

To increase the transmural pressures in the blood vessels of

a subject’s lower leg, a method described in more detail

previously (Eiken and Kolegard 2001, 2004) was used,

Briefly, the experiments were carried out with the subject

positioned supine in a pressure chamber with the lower

portion of a leg (here termed the ‘‘test leg’’) extended

through a port in the chamber door. The test leg was sup-

ported at the level of the heart by means of a stand, and was

hermetically sealed to the door hole slightly proximally of

the knee, by use of a short self-sealing rubber sleeve.

A special harness was used to stabilize the trunk and to

prevent involuntary muscle activity and movements in the

leg as the pressure in the chamber was elevated. Ambient

temperature was maintained at 25�C (range: 23–27�C) at

the site of the test leg. As chamber pressure is raised,

pressure increases in all tissues enclosed in the chamber

and the pressure is also transmitted virtually without dis-

tortion to the blood vessels of the unexposed test leg, so

that transmural pressures in all vessels (arteries, capillaries

and veins) of the test leg are elevated by the same mag-

nitude as the applied chamber pressure (cf. Green et al.

2007; Kolegard 2010).

The diameter of the tibial posterior artery was measured

5–10 cm proximal to the medial malleolus using ultraso-

nography. Measurements were conducted in B-mode image

during end-diastole (determined from the ECG), as wall-to-

wall distance in the sagittal section, using a 6.0–11.0 MHz

linear array transducer (Aspen, Acuson, Mountain View,

CA, USA). Volume flow was estimated in the tibial artery

by simultaneous measurements of vessel diameter and

mean flow-velocity, employing an ultrasound/Doppler

technique (6.0–11.0 MHz transducer; Aspen, Acuson).

Assuming that the artery had a circular cross-section, flow

was subsequently calculated by multiplying vessel cross-

sectional area by the time integral of the mean flow-

velocity.

Heart rate (HR) and arterial pressures (SAP, DAP) were

measured using ECG and volume-clamp techniques,

respectively, as described above. The level of the heart was

used as the point of reference for the arterial pressure

measurements. Arterial distending pressure (DP) was cal-

culated, and arteriolar DP was approximated, by adding

chamber pressure to DAP (cf. Eiken and Kolegard 2001,

2004; Eiken et al. 2008). Since blood flow in the legs is

mainly controlled by local vascular resistance, it can be

argued that flow should be treated as a function of arteriolar

rather than arterial DP. However, since the average arte-

riolar DP is not readily determined, flow patterns in the

tibial artery were compared by treating them as functions

Eur J Appl Physiol (2012) 112:3619–3627 3621

123

Author's personal copy

of peak arteriolar DP (i.e. DP at the upstream end of the

arterioles, corresponding to arterial DP).

Each subject rated his perceived pain using a ratio scale

(Borg 1982), in which pain could be rated from 0 (no pain)

to 10 (very, very strong pain, almost intolerable).

Experimental protocol

Each experiment started with a 10-min baseline period with

normal atmospheric pressure in the chamber. Thereafter

chamber pressure was increased every 2.5 min in steps of

60 mmHg up to 180 mmHg above atmospheric, and there-

after in steps of 30 mmHg up to a maximum of 240 mmHg

above atmospheric. Chamber pressure was then rapidly

released and a 2.5-min recovery period at atmospheric

pressure ensued. Recordings of HR, SAP, DAP, tibial artery

diameter and flow as well as rating of perceived pain were

obtained during the last min at each pressure level.

Analyses

To determine whether differences in pressure-distension

relationships were due to changes in threshold or gain of the

distension, linear functions were fitted to the initial four and

final three data points, respectively, of every DP-D diameter

and DP-flow curve. The DP corresponding to the crossing

point of the two regression lines was defined as the threshold

DP and the slope of the regression line of the last three points

was defined as the gain of the pressure distension.

The statistical significance of intergroup differences was

tested using analysis of variance or a two-sample t test for

all variables except for perceived pain, which was evalu-

ated by a Mann–Whitney signed-rank test (Statistica Stat-

soft, Tulsa OK, USA). p \ 0.05 were regarded statistically

significant.

Results

Criterion variables

Subjects’ physical characteristics and resting arterial pres-

sures are given in Table 1. As evident from the table, the

groups were well matched with regard to age, mass, body

height and resting arterial pressures.

G-tolerance trials

Relaxed G tolerance was substantially higher (p \ 0.01) in

the H group (6.6 ± 0.8 G = mean ± SD) than in the L

group (3.9 ± 0.3 G). Baseline values for HR and MAP,

obtained with the subject sitting in the centrifuge gondola

prior to the G-tolerance trials, were similar in the two

groups (Fig 1). During the trials, HR increased with

increasing G load in a similar fashion in the two groups,

whereas the MAP response was considerably stronger

(p \ 0.01) in the H group than in the L group with higher

values in the H group at any given G load (Fig 1).

Pressure-distension relationships

Arterial diameters

The baseline lumen diameter of the tibial artery, determined

at normal atmospheric chamber pressure prior to the pres-

sure provocation, was similar in the H group (2.4 ±

0.3 mm) and L group (2.2 ± 0.3 mm). In both groups,

lumen diameter remained unaltered at slight and moderate

elevations of DP, but increased (p \ 0.001) promptly at the

highest levels of DP (Fig. 2). Distension at the highest

DP that could be attained by all subjects in both groups

(about 290 mmHg) was more pronounced (p \ 0.001) in

the L-group (14.1 ± 4.2%) than in the H group (1.7 ± 5.0%)

(Fig. 2). A negative correlation was observed between G

tolerance and the relative increase in diameter at the highest

attainable DP (R = 0.79; p = 0.001).

Linear regression of the three final data points of the

DP-D diameter relationship, obtained during each pressure-

distension experiment, revealed that the inter-group differ-

ence in arterial distensibility was due both to a difference

in gain (slope) and threshold (intercept) of the pressure-

distension relationship, with lower gain and higher pressure

threshold in the H than in the L group. (Table 2).

Arterial flow

In both groups, tibial artery flow remained unaltered at

slight and moderate elevations of DP, whereas at the

highest levels of DP a prompt increase (p \ 0.001) in flow

occurred (Fig. 2). At the highest DP that could be attained

by all subjects in both groups, the increase in flow from the

Table 1 Subjects’ physical characteristics and resting systolic and

diastolic arterial pressures (SAP, DAP) in the high G-tolerance

(H) and low G-tolerance (L) groups

Variable Subject group

H Difference L

Age (year) 27 ± 4 n.s. 27 ± 3

Height (cm) 179 ± 10 n.s. 183 ± 5

Mass (kg) 84 ± 14 n.s. 75 ± 12

SAP (mmHg) 126 ± 9 n.s. 119 ± 9

DAP (mmHg) 74 ± 11 n.s. 69 ± 7

Values are mean ± SD

n = 7 in the H group and n = 8 in the L group

3622 Eur J Appl Physiol (2012) 112:3619–3627

123

Author's personal copy

baseline value was more pronounced (p = 0.026) in the L

group (32 ± 21 ml/min) than in the H group (2 ± 25 ml/

min) (Fig. 2). A negative correlation was observed between

G tolerance and the increase in flow at the highest attain-

able DP (r = 0.59; p = 0.019).

Linear regression of the three final data points of the DP-

flow relationship, obtained during each pressure-distension

experiment, revealed that the inter-group difference in flow

response was due mainly to higher pressure-flow threshold

in the H than in the L group; there was a tendency, albeit

not statistically significant, also for reduced gain of the

pressure-flow relationship in the H compared to in the L

group (Table 2).

Heart rate and arterial pressure responses

In both groups, HR remained relatively stable during the

course of the pressure provocation. In the H group, it was

74 ± 22 beats/min under baseline conditions (atmospheric

chamber pressure) and 80 ± 19 beats/min at the highest

chamber pressure that could be attained by both groups; in

the L group, it was 68 ± 10 and 77 ± 6 beats/min at the

corresponding chamber pressures.

During the course of the pressure provocation (atmo-

spheric to highest attainable chamber pressure) SAP/DAP

increased (p = 0.012) from 119 ± 9/69 ± 7 to 132 ± 12/

77 ± 8 mmHg in the L group and from 126 ± 9/74 ± 11

to 140 ± 11/86 ± 8 mmHg in the H group. There were no

intergroup differences as regards HR, SAP or DAP

responses during the pressure provocations.

Perceived pain

Exposure to high intravascular pressures induced pain

in both groups. Pain increased with increasing chamber

pressure (p \ 0.001) and was stronger in the L-group than

Fig. 1 Mean arterial pressure (MAP) at eye level and heart rate (HR)

as functions of headward acceleration (increased G load in the head-

to-foot direction) in the high and low G-tolerance groups. Values are

means (SD)

Fig. 2 Changes in tibial artery diameter (upper panel) and flow

(lower panel) as functions of distending pressure (see also ‘‘Methods’’

section) in the high and low G-tolerance groups. Values are means

(SD)

Eur J Appl Physiol (2012) 112:3619–3627 3623

123

Author's personal copy

in the H-group at any given level of markedly increased

intravascular pressure; at the highest chamber pressure

attained by all subjects in both groups (195 mmHg) pain

was rated 7 (3–9) [=median (range)] in the L group and 4

(1–5) in the H-group (p = 0.028). At that chamber pressure

the arterial distension was 6.9 ± 5.7% in the L group and

1.1 ± 3.7% in the H group.

Discussion

The present results demonstrated that pressure-induced

distension and flow in leg arteries were less pronounced in

individuals possessing high-relaxed G tolerance than in

those with low G tolerance. Arterial DP was calculated by

adding applied chamber pressure to DAP and changes in

flow were also evaluated as functions of arterial DP.

Because, in both groups, the pressure provocation induced

only minute increments in MAP, with no difference

between groups, the observed pressure-induced increments,

and inter-group differences, in tibial artery flow are pre-

dominantly attributable to regional changes/differences in

blood-flow resistance and hence presumably in arteriolar

distensibility. Thus, when applying the present procedure

to increase local intravascular pressures, a considerable

share of the pressure-induced increase in arterial flow

appears to be accommodated by the capillary beds since the

flow increase invariably occurs in concert with increased

rate of tissue oedema formation, presumably resulting from

increased capillary filtration pressure (Eiken and Kolegard

2001, 2004, 2011). The alternative explanation to a sudden

flow increase, namely that it would solely be attributable to

shunting of blood via arteriovenous anastomoses would

probably not result in augmented tissue oedema formation.

Present results do not allow us to draw any firm

conclusions as regards mechanisms underlying the inter-

group differences in arterial and arteriolar wall stiffness,

but different possibilities should be considered. Differ-

ences in vascular pressure resistance may be caused by:

(1) differences in smooth muscle tone, due to differences in

myogenic responsiveness or in local or systemic release

of vasoconstrictive substances, (2) differences in passive

elastic recoil resulting from differences in the structure of

the vessel wall, or (3) differences in the counter pressure

exerted by tissues adjacent to the vessel. Pressure in tissues

surrounding the vessels undoubtedly may affect vessel

distensibility, and since we did not measure it we cannot

exclude that pressure surrounding the tibial artery to some

extent affected the present results. However, there are

several reasons to doubt that it had any major impact.

Firstly, the two groups were well matched as regards

physical characteristics, and care was taken to ensure

that each subject refrained from performing leg-muscle

contractions during the pressure-distension measurements.

At the site of examination, the posterior tibial artery is

surrounded by loose connective tissue. Notwithstanding,

it is enclosed in a septal compartment and hence the sur-

rounding tissue pressure may exceed that of the ambient

atmosphere. In a lower leg septal compartment, the inter-

individual variation in tissue pressure in healthy resting

horizontal subjects is only a few mmHg (Ludbrook 1966;

Lundin and Styf 1998); once intravascular pressure is

increased, compartmental tissue pressure may increase due

to transmission of pressure from engorged veins, and

eventually due to oedema formation. The pressure eleva-

tion in tissues surrounding the vessels following an acute

intravascular pressure increase of about 100 mmHg may

vary slightly between individuals from about 7 to

14 mmHg (Ludbrook 1966; Lundin and Styf 1998).

Although it cannot be excluded that the interindividual

variation in compartmental pressure is somewhat greater at

higher intravascular pressures, it appears improbable that it

would be sufficient to explain the 50 mmHg discrepancies

between the present H and L groups as regards pressure

thresholds for diameter and flow increments (Table 2); for

leg veins, both the magnitude and interindividual variation

of the pressure distension response are more pronounced in

the pressure range 0–100 mmHg than in the range

100–200 mmHg (cf. Kolegard 2010). Lastly, it appears

highly unlikely that the present prompt pressure-induced

increase in tibial artery diameter was determined by the

surrounding pressure because such arterial diameter

increments (1) are never accompanied by prompt incre-

ments in venous diameters (Eiken and Kolegard 2001,

2004), and (2) would indicate a sudden and prompt drop in

tissue pressure, which could only be explained by a septal

rupture; that the arterial and venous pressure–distension

relationships are highly repeatable during consecutive

pressure exposures (cf. Kolegard 2010) contradicts the

notion of pressure-induced septal damage.

Instead the shapes of the present pressure-diameter and

pressure-flow curves are typical for arteries with preserved

Table 2 Gain and threshold for the distending pressure (DP)-Ddiameter and DP-flow relationships in the posterior tibial artery in the

high (H) and low (L) G-tolerance groups

H group Difference L group

Diameter

Gain (mm/mmHg 9 103) 2.7 ± 3.6 p = 0.01 8.1 ± 3.2

Threshold (mmHg) 323 ± 17.5 p = 0.004 271 ± 36

Flow

Gain (ml/min/mmHg) 0.4 ± 0.4 p = 0.08 0.8 ± 0.3

Threshold (mmHg) 294 ± 19 p = 0.002 246 ± 26.7

Values are mean ± SD

n = 7 in the H group and n = 8 in the L group

3624 Eur J Appl Physiol (2012) 112:3619–3627

123

Author's personal copy

myogenic tone (Folkow et al. 1970a, b). It is possible that

differences in vascular myogenic tone contributed to the

inter-group differences in arterial/arteriolar pressure resis-

tance. Changes in vascular myogenic responsiveness,

observed in rats have, in part, been attributed to changed

expression of endothelial and neuronal nitric oxide (NO)

and of inducible NO synthase (Jasperse et al. 1999; Vaziri

et al. 2000). It has been demonstrated that local NO-med-

iated dilatation of precapillary vessels is reduced in

hypertensive patients (Muiesan et al. 2009).

It is also possible that the inter-group differences in

arterial/arteriolar wall stiffness reflected differences in the

elastic recoil properties of the vessel wall resulting from

differences in wall thickness (Aalkjaer et al. 1987), wall

thickness-to-lumen ratio (Folkow et al. 1958; Folkow 1990)

and/or in the content and arrangement of collagen and

elastin (Intengan and Schiffrin 2000). Notably, the DP

threshold for increasing flow was substantially lower in the

L than in the H group, whereas the gain (slope) of the

DP-flow response only tended to be marginally higher in

the L than in the H group. Media hypertrophy in the arte-

rioles is expected to reduce the gain of the DP-flow curve,

whereas increased myogenic reactivity is likely to pre-

dominantly increase the pressure threshold at which flow

commences to increase (cf Folkow and Sivertsson 1968;

Folkow 1990; Eiken and Kolegard 2011). Even though

these findings support the notion that the inter-group dif-

ference in precapillary distensibility was predominantly

attributable to differences in local myogenic activity, it

should be emphasized that further investigations are needed

to establish to what degree the diminished arterial/arteriolar

distension in individuals possessing high G tolerance

reflects smooth muscle hypertrophy and increased myo-

genic activity, respectively.

Regardless of what mechanisms govern arterial/arterio-

lar distensibility, it appears that the magnitude of this

modality is not permanent. Instead, the wall stiffness of

arteries and precapillary resistance vessels adapts to meet

the long-term demands imposed by the hydrostatic pressure

acting locally on the vessel walls. Thus, removal of gravity-

dependent intravascular pressure components, by exposure

to prolonged horizontal bedrest, increases distensibility

of leg arteries/arterioles (Eiken et al. 2008). Inversely,

repeated exposures to moderately increased intravascular

pressures render peripheral arteries/arterioles more pressure

resistant (Eiken and Kolegard 1999, 2011). It cannot be

excluded that similar pressure habituation effects on arter-

ies/arterioles of the lower body may contribute to the

increase in G tolerance in response to repeated exposures to

high G loads (i.e. G training) (for review see Burton and

Smith 1996).

The finding that the distensibility of leg arteries and

arterioles was greater in the L group than in the H group,

and that negative correlations were found between G tol-

erance and both arterial and arteriolar distensibility raises

the question of whether a causal relationship exists

between relaxed G tolerance and the stiffness of precapil-

lary vessels of the lower body. As mentioned previously,

an individuals’ G-level tolerance is predominantly gov-

erned by his/her arterial pressure response. In the seated

position, increased ?Gz load creates exaggerated hydro-

static pressure gradients in blood vessels oriented longitu-

dinally in the body. To maintain adequate cerebral

perfusion, and hence consciousness, at high G loads, arte-

rial pressure at heart level must increase to the extent that it

overcomes the hydrostatic pressure drop in the arteries

from the heart to the head (Wood and Lambert 1952; Wood

1987; Eiken et al. 2007). Heart level arterial pressure, in

turn, is the product of cardiac output and total peripheral

blood-flow resistance. Assuming that the distension of

arteries/arterioles observed in our experiments reflected a

general response of dependent precapillary vessels, then it

is likely that during the G exposures, pressure distension in

dependent vessels diminished total peripheral resistance

and hence compromised the control of heart-level arterial

pressure. In the present study, initial signs of arterial/arte-

riolar distension were observed at DPs ranging from 250 to

300 mmHg. In the vasculature of the lower body, pressure

increases in direct proportion to the increased ?Gz load. It

can be estimated that, with the subject positioned as in the

present centrifuge experiments - which corresponds to the

position of a pilot flying the JAS 39 Gripen fighter air-

craft—a critical DP of 250–300 mmHg is attained in the

arteries/proximal arterioles of the legs at G loads ranging

from about 3–4 G in the ankles to 5–6 G in the thighs. This

is in the approximate range of the present G-tolerance

values. Thus, it seems likely that the observed difference

in arterial/arteriolar wall stiffness between the H- and

L-groups contributed to the intergroup difference in G

tolerance.

This does not exclude contribution by other mecha-

nisms, such as arterial baroreflex sensitivity, commonly

assumed to affect G tolerance (Newman et al. 1998;

Convertino 1998), to the inter-group difference in G tol-

erance. Thus, high arterial baroreflex sensitivity may reflect

high responsiveness in the effector organs, including high

myogenic responsiveness in arterioles. The observation

that, during the G-tolerance trials, the MAP response was

considerably stronger in the H group, even at slight and

moderate elevations of the G load, is compatible with the

notion of increased myogenic responsiveness in arterioles

of G tolerant individuals. Other factors that have been

shown to affect relaxed G-level tolerance in previous

studies include baseline arterial pressure and distance

between the heart and brain (Klein et al. 1969). Our find-

ings that both resting SAP/DAP and body height were

Eur J Appl Physiol (2012) 112:3619–3627 3625

123

Author's personal copy

similar in the H and L groups suggest that neither baseline

arterial pressure nor the heart-to-brain distance, constituted

distinguishing features between the two groups.

Present results showed that markedly increased pressure

in the hyperbaric chamber induced pain in the test leg,

which, at any given pressure, tended to be greater in the L

group than in the H group. Such pressure-induced pain

is most likely of vascular origin and akin to the pain

commonly experienced in the arms by pilots flying high-

performance fighter aircraft (Eiken and Kolegard 2001).

Thus, the magnified hydrostatic pressure gradients that

act along the vessels during exposure to high-sustained

acceleration may cause severe pain locally in the arms

(Green 1997; Watkins et al. 1998); in contrast to the leg

vessels, the arm vessels are typically not supported by the

external counter pressure of the anti-G suit during expo-

sures to high sustained G forces. Under certain circum-

stances, G-induced pain may develop also in the lower

legs/feet, for example in pilots/centrifuge subjects who are

not wearing anti-G suits or who are exposed to substantial

variations in G-suit pressure during the G-time profiles

(Paul 1996). There is ample evidence to support the notion

that G-induced arm pain is due mainly to local overdis-

tension of veins, presumably via activation of perivascular

mechanoreceptors that respond to stretch and to direct

mechanical stimulation (for review see Kolegard 2010). If,

or to what extent, pressure-induced distension of arteries

and arterioles contributes to the development of such pain

is less clear. There is some evidence to suggest that in

certain conditions distension of arteries may also induce

pain (Malliani and Lombardi 1982; Wooley et al. 1998).

Thus, our finding that the pressure-induced pain tended to

be attenuated in the H group compared to in the L group

may either suggest that distensibility of peripheral veins

differed between the groups in a similar manner as did

precapillary distensibility, or that the degree of pressure-

induced pain in fact reflected the degree of arterial/arte-

riolar distension.

The present pressure-distension and pressure-flow

examinations were limited to those of the posterior tibial

artery and the vascular circuits supplied by this artery. To

obtain a more comprehensive view of the relationship

between relaxed G tolerance and vascular distensibility,

future studies should include other, and larger, vascular

beds in dependent parts of the body. Another delimitation

of the present study was that only males were investigated.

In conclusion, present results showed that the in vivo

wall stiffness of precapillary leg vessels is higher in indi-

viduals with high-relaxed G tolerance. Whether a causal

relationship exists between wall stiffness in dependent

precapillary vessels and G tolerance remains to be estab-

lished, as do the mechanisms governing interindividual

differences in precapillary wall stiffness.

Acknowledgments This study was supported by grants from the

Swedish Armed Forces and the Gosta Fraenekel Foundation.

Conflict of interest The authors declare that they have no conflict

of interest.

References

Aalkjaer C, Heagerty AM, Petersen KK, Swales JD, Mulvany MJ

(1987) Evidence for increased media thickness, increased

neuronal amine uptake, and depressed exitation-contraction

coupling in isolated resistance vessels from essential hyperten-

sives. Circ Res 61:181–186

Borg GA (1982) Psychophysical bases of perceived exertion. Med Sci

Sports Exer 14:377–381

Buick F, Wood EH, Pecaric M, Maloan J (1995) Methods for

measuring physiological responses and protection in man

exposed to high ?Gz. In: AGARD-LS-202, Current Concepts

on G-protection research and development, North Atlantic

Treaty Organization, pp 8-1 to 8-15

Burton RR, Smith AH (1996) Adaptation to acceleration environ-

ments. In: Fregly and Blatteis (eds) Handbook of physiology

environmental physiology. sect. 4, vol II. Oxford University

Press, New York, pp 943–970

Burton RR, Whinnery JE (2008) Biodynamics: sustained acceleration.

In: DeHart RL (ed) Fundamentals of aerospace medicine, 3rd

edn. Williams & Wilkins, Baltimore, pp 201–260

Convertino VA (1998) High sustained ?Gz acceleration: physiolog-

ical adaptation to high-G tolerance. J Gravit Physiol 5(1):51–54

Eiken O, Kolegard R (1999) Pain in the arms induced by markedly

increased intravascular pressure decreases after repeated exposures

to moderately increased pressures. J Gravit Physiol 6(1):35–36

Eiken O, Kolegard R (2001) Relationship between arm pain and

distension of arteries and veins caused by elevation of transmural

pressure in local vascular segments. Aviat Space Environ Med

72:427–431

Eiken O, Kolegard R (2004) Comparison of vascular distensibility in

the upper and lower extremity. Acta Physiol Scand 181:281–287

Eiken O, Kolegard R (2011) Repeated exposures to moderately

increased intravascular pressure increases stiffness in human

arteries and arterioles. J Hypertens 29:1963–1971

Eiken O, Kolegard R, Bergsten E, Gronkvist M (2007) G protection:

interaction of straining maneuvers and positive pressure breath-

ing. Aviat Space Environ Med 78:392–398

Eiken O, Kolegard R, Mekjavic IB (2008) Pressure-distension

relationship in arteries and arterioles in response to 5 wk of

horizontal bedrest. Am J Physiol Heart Circ Physiol 295:H1296–

H1302

Folkow B (1990) ‘‘Structural factor’’ in primary and secondary

hypertension. Hypertension 16:89–101

Folkow B, Sivertsson R (1968) Adaptive changes in ‘‘reactivity’’ and

wall/lumen ratio in cat blood vessels exposed to prolonged

transmural pressure difference. Life Sciences 7:1283–1289

Folkow B, Grimby G, Thulesius O (1958) Adaptive structural

changes of the vascular walls in hypertension and their relation

to the control of the peripheral resistance. Acta Physiol Scand

44:255–272

Folkow B, Hallback M, Lundgren Y, Weiss L (1970a) Structurally

based increase of flow resistance in spontaneously hypertensive

rats. Acta Physiol Scand 79:373–378

Folkow B, Hallback M, Lundgren Y, Weiss L (1970b) Background of

increased flow resistance and vascular reactivity in spontane-

ously hypertensive rats. Acta Physiol Scand 80:93–106

3626 Eur J Appl Physiol (2012) 112:3619–3627

123

Author's personal copy

Green ND (1997) Arm arterial occlusion cuffs as a means of

alleviating high ?Gz-associated arm pain. Aviat Space Environ

Med 68:715–721

Green ND, Brown MD, Coote JH (2007) Pain and changes in

peripheral resistance at high vascular transmural pressure in the

human forearm. Eur J Appl Physiol 100:627–635

Intengan HD, Schiffrin EL (2000) Structure and mechanical proper-

ties of resistance arteries in hypertension: role of adhesion

molecules and extracellular matrix determinants. Hypertension

36:312–318

Jasperse JL, Woodman CR, Price EM, Hasser EM, Laughlin MH

(1999) Hindlimb unweighting decreases ecNOS gene expression

and endothelium-dependent dilation in rat soleus feed arteries.

J Appl Physiol 87:1476–1482

Klein KE, Bruner H, Jovy D, Vogt L, Wegmann HM (1969) Influence

of stature and physical fitness on tilt-table and acceleration

tolerance. Aerospace Med 40:293–297

Kolegard R (2010) Distensibility in arteries, arterioles and veins in

humans: adaptation to intermittent or prolonged change in

regional intravascular pressure. Dissertation, Royal Institute of

Technology. ISBN: 978-91-7415-788-9

Ludbrook J (1966) The musculovenous pumps of the human lower

limb. Am Heart J 71:635–641

Lundin O, Styf JR (1998) Intramuscular pressure in the leg and thigh

related to tensile strap force during knee brace wear: an

experimental study in man. Am J Sports Med 26:567–570

Malliani A, Lombardi F (1982) Consideration of the fundamental

mechanisms eliciting cardiac pain. Am Heart J 103:575–578

Muiesan ML, Salvetti M, Paini A, Agabiti-Rosei E (2009) Prognostic

significance of flow-mediated dilatation of the brachial artery in

hypertensive patients; possible role of central blood pressure.

J Hypertension 27:903–904

Newman DG, White SW, Callister R (1998) Evidence of baroreflex

adaptation to repetitive ?Gz in fighter pilots. Aviat Space

Environ Med 69:446–451

Paul M (1996) Extended-coverage-bladder G-suits can provide

improved G-tolerance and high Gz foot pain. Aviat Space

Environ Med 67:253–255

Vaziri ND, Ding Y, Sangha DS, Purdy RE (2000) Upregulation of

NOS by simulated microgravity, potential cause of orthostatic

intolerance. J Appl Physiol 89:338–344

Watkins SM, Welch L, Whitey P, Forster E (1998) The design of arm

pressure covers to alleviate pain in high G maneuvers. Aviat

Space Environ Med 69:461–467

Wood EH (1987) Development of Anti-G suits and their limitations.

Aviat Space Environ Med 58:699–706

Wood EH, Lambert EH (1952) Some factors which influence the

protection afforded by pneumatic anti-G suits. J Aviat Med

23:218–228

Wooley CF, Sparks EH, Boudoulas H (1998) Aortic pain. Prog

Cardiovasc Dis 40:563–589

Eur J Appl Physiol (2012) 112:3619–3627 3627

123

Author's personal copy