Clin Physiol Funct Imaging (2008) 28, pp8695

doi: 10.1111/j.1475-097X.2007.00774.x

Pulse wave analysis on ngertip arterial pressure: effects

of age, gender and stressors on reected waves and their
relation to brachial and femoral artery blood ow
Catherine Ahlund1, Knut Pettersson1 and Lars Lind2
1

AstraZeneca R&D Molndal, Sahlgrenska University Hospital, Goteborg, and 2Department of Medical Sciences, University Hospital, Uppsala, Sweden

Summary
Correspondence
Catherine Ahlund, Department of Medical Sciences,
AstraZeneca, SE-431 83 Molndal, Sweden
E-mail: catherine.ahlund@astrazeneca.com

Accepted for publication

Received 26 April 2007;
accepted 4 October 2007

Key words
exercise; mental stress test; non-invasive blood
pressure; pressure pulse reflection; pulse contour
analysis

Objectives: Analysis of the contour of the arterial pressure pulse (pulse wave
analysis; PWA) adds information about arterial stiffness etc., beyond that
obtained from absolute pressures. Peripheral pulses normally show an anterograde systolic peak and two reflected peaks: one in systole and one in diastole.
The amplitudes and timings of these were estimated from finger pressure
recordings in three study groups. We studied the usefulness of continuous digital
pressures for PWA.
Methods: First, PWA from intra-arterial (brachial) and non-invasive finger pressure
recordings was compared. Secondly, stress-induced (mental arithmetics and cold
pressor test) changes in pressure pulse reflection were compared with blood flow
changes in brachial and femoral arteries (ultrasound). Thirdly, the influence of
age and gender on digital pulse pressures was investigated at rest and during
exercise.
Results and conclusion: Pulse wave analysis results from brachial and digital pressures
correlated strongly. Stress induced changes in systolic reflection were associated with
changes in brachial artery flow patterns, whereas diastolic reflection was associated
with femoral artery flow changes. At rest, age increased systolic reflection without
affecting diastolic reflection. Exercise increased systolic reflection and reduced
diastolic reflection more in older subjects (>40 years) than in younger (<40 years).
In conclusion, PWA from continuous, digital pressure recordings is a convenient
technique to study the arterial function at rest and during exposure to stressors in
broad populations. The two reflected waves are differently regulated, which may
indicate different anatomical origin.

Introduction
The arterial pulse represents the summation of left ventricular
outflow and reflected pressure waves from the periphery. The
waveform varies throughout the arterial tree, and especially the
aortic systolic pressure may differ markedly from peripheral
arterial systolic pressure (Pauca et al., 1992). Such differences are
important as cardiac workload is determined by aortic rather
than peripheral blood pressure, and thus cannot directly be
estimated from peripheral pressure measurements. However,
aortic pressure waveforms can be generated from peripheral
waveforms such as the radial artery waveforms obtained by
applanation tonometry by the application of a general transfer
function, and cardiac workload can be estimated by analysis of
pressure waveforms (Karamanoglu & Feneley, 1996). Increased
86

systolic reflection because of reduced aortic compliance is an

important risk factor for premature cardiovascular (CV) disease
(Benetos et al., 2002; Weber et al., 2004).
Pulse wave analysis (PWA) of peripheral arterial pulses is
mainly used to estimate systolic augmentation and aortic systolic
(SBP) and diastolic blood pressure (DBP) (ORourke et al.,
2001), but it can also be used to analyse changes in the
peripheral circulation (Duprez et al., 2001; Ahlund et al., 2003).
The shape of peripheral pressure waveforms, recorded
non-invasively, differs markedly between individuals (Fig. 1).
In principal, the peripheral reflections are summed up into one
continuous wave, divided into a systolic and a diastolic portion
by the dicrotic notch; the systolic and diastolic portion can be
distinguished, each having a separate peak. Because of the
timing of these two reflected peaks, and the knowledge that the

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

Pulse wave analysis of ngertip blood pressure, C. Ahlund et al. 87

Figure 1 Different shapes of peripheral

pressure waveforms, recorded non-invasively
with the Portapres device. The analysis (PWA)
made to describe the pulse wave times and
amplitudes used in this study are listed. For
each cycle, the base line was defined as a
straight line between the end-diastolic
pressures preceding the systolic phases of two
subsequent heart cycles. Time indices are the
times from onset of the systolic wave to the
peak of the systolic, the reflected systolic and
the reflected diastolic waves respectively. AIx,
ba; RIx, da; SRt, a ) 0; At, b ) 0; Rt, d ) 0.

0 = Onset of systole
a = Peak systole
b = Reflected wave in systole
d = Reflected wave in diastole

aortic pressure pulse wave velocity usually is in the range of

710 m s)1, it can be assumed that the detectable systolic
reflected wave (At) originates from the middle part of the aorta,
whereas the first diastolic pressure peak is reflected from a more
distal part of the body. In young healthy volunteers, we have
used digital arterial pulse pressure waveforms recorded noninvasively to analyse the effects of stressful interventions often
encountered in daily life (Ahlund et al., 2003). In terms of
timing and amplitude, the reflected systolic and diastolic
pressure peaks did not always change in the same direction
during the stressful interventions, and we therefore concluded
that PWA of the digital arterial pressure is a simple method to
study vascular reactivity non-invasively, with a potential to
discriminate between different reflection sites. A major advantage with digital pressure recordings is that it can easily be used
during exercise, which is more difficult by using applanation
tonometry.
Arterial flow also changes markedly during a cardiac cycle,
and the shape of the flow wave depends on the site of
measurement (Nichols & ORourke, 1998). The relationship
between flow and pressure in the arterial system is complex, but
reflection in the vascular system can be seen both in pressure
and flow curves. Thus, if the diastolic reflection seen in the
digital pressure pulse originates from the lower extremities, this
should also be detectable as changes in the shape of the flow
wave in the extremities. Similarly, changes in the systolic
reflection should also affect both flow and pressure pulses.
Indeed, flow patterns vary markedly during the cardiac cycle
between measuring sites, and brief episodes of blood flow
reversal are often seen (Nichols & ORourke, 1998).

d
AIx = Relative amplitude of systolic reflected wave
RIx = Relative amplitude of the first diastolic reflected wave
SRt = Time to systolic peak
At = Time to systolic reflected wave
Rt = Time to the first diastolic reflected wave

The aim of this study was to further investigate the usefulness

of PWA of finger blood pressure recordings as a useful tool to
study peripheral circulation. It has been shown earlier that there
is a quantitative relationship between systolic and diastolic
pressures measured in the finger and brachial artery (Bos et al.,
1996; Voogel & van Montfrans, 1997; Eckert & Horstkotte,
2002), but a relationship between the reflected waves in the
finger and the brachial artery pulse needs to be further
established. The first aim of this study was therefore to
investigate if there is a correlation between timing and
amplitude of the reflected waves measured invasively in the
brachial artery and measured non-invasively in the finger.
The second aim was to provide evidence for our assumption
on the regional origin of the systolic and diastolic reflected
waves (Rt). For this, we reperformed challenges with mental
and painful stress in healthy volunteers, but this time we also
measured blood flow in the brachial and femoral arteries, based
on the hypothesis that the changes in the two reflected pressure
waves identified at the peripheral pulse recordings during the
stressful interventions are paralleled by changes in blood flow.
Using Doppler recordings, it is possible to quantify changes
both in the amplitude of blood flow and in the timing of the
different flow peaks, and it is also possible to relate these
changes to those seen during peripheral pulse pressure wave
recordings. We have earlier used PWA from digital pressure
measurements only in a young healthy population. Age and
gender are well known to affect the arterial pressure waveform,
and the third aim of the study was therefore to explore how age
and gender affect the digital pressure waveform during rest and
exercise.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

88 Pulse wave analysis of ngertip blood pressure, C. Ahlund et al.

Methods and materials

Study protocol 3

This study was performed in three separate series of experiments. The local Ethics Committee approved all the three
protocols. All subjects included gave written informed consent.

The aim was to investigate differences of pulse wave characteristics during rest (supine), and bicycle exercise due to age and
gender. Sixteen females and 17 males were included in this
protocol. They were free from cardiovascular and other serious
illnesses and free of regular medication. The subjects were
divided into groups with respect to age, younger (<40 years)
and older (>40 years) and gender. All examinations took place
in the morning after an overnight fast.
After positioning of ECG electrodes and the digital blood
pressure measurement equipment, the subject rested in supine
position for 30 min. During the last 5 min of rest, heart rate and
blood pressure were continuously recorded. These registrations
were followed by bicycle exercise. The subjects were instructed
to pedal with a rate of 60 turns per minute and the workload
was adjusted until the subjects reached a heart rate of 100 beats
per minutes, and digital pressure was recorded for 5 min after
steady state heart rate was achieved.

Study protocol 1
To investigate if PWA finger tip pressure measurements of
reflected waves produce results similar to those from invasively
measured blood pressure, pulse waves were simultaneously
recorded with a Portapres (FMS, Finapres Medical Systems BV,
Amsterdam, The Netherlands) device in a finger, and invasively
through an indwelling arterial catheter in the brachial artery in
the contra-lateral arm. Both pressure measurements were
sampled with the same frequency and analysed in the Pharmlab
software described below. This procedure was performed in 11
young healthy subjects (seven males and four females), with a
mean age of 24 years. All examinations took place in the
morning after an overnight fast.
After positioning the ECG electrodes, the digital blood
pressure measurement equipment and the arterial catheter, the
subjects rested in the supine position for 30 min. During the last
5 min of rest, heart rate and blood pressure waveforms were
continuously recorded. Coefficients of variability of the PWA
estimates were performed on eight subjects (mean age
22 years). Measurements were performed on resting subjects
at four occasions separated by at least 2 days.

Methods
Invasive pressure recordings
The intra-arterial blood pressure recordings were obtained by a
catheter (20G 110 45 mm) connected to a pressure tube
(150 mm) and a transducer (DTX Plus Transducer DT-XX,
Franklin Lakes, NJ, USA). The signals were registered by GE
Marquette Eagle4000 (Stockholm, Sweden).

Study protocol 2
The aim of this study was to investigate if changes in the two
reflected pressure waves during stressful interventions were
correlated to changes in blood flow. This study was performed
in 12 male subjects. All examinations took place in the morning
after an overnight fast.
After positioning the ECG electrodes and the digital blood
pressure measurement equipment, the subjects rested in the
supine position for 30 min. During the last 5 min of rest,
heart rate and the digital blood pressure were continuously
recorded and a Doppler examination was performed (see
below). After that baseline examination, the subjects were
subjected to two different stressors, 5 min of mental stress
(mental arithmetics, MAT) and 5 min of painful stress (the
cold pressor test, COP) during which heart rate and blood
pressure were continuously recorded and Doppler examinations were reperformed. Between these interventions, there
was at least 15 min of rest to allow haemodynamic variables
to return to baseline.
During MAT stress, the subjects were instructed to subtract 13
or 17 from a three-digit number as quickly and accurate as
possible. During this test, the subjects were intentionally
frustrated by being corrected frequently and by playing loud
unpleasant music. During the COP test, one foot was placed in
ice-cold water for 5 min.

Digital pressure recordings

The digital pressure waveform measurement was performed by
a Portapres device (Finapres Medical Systems BV) as previously
described (Ahlund et al., 2003). The experiments were performed at room temperature (2123C). To avoid vasoconstriction, the measuring hand was warmed by a heat pad.
Data collection and analysis
Data sampling and analyses from both pressure signals and ECG
were performed similarly. The pressure waveform obtained
with both the Eagle4000 and the Portapres was sampled with a
frequency of 200 Hz and stored digitally for later analyses in the
custom-made program package Pharmlab 3.0 (Axenborg J,
Hirsch I, AstraZeneca R&D Molndal, Sweden). The reported
heart rate and blood pressure levels and the characteristics of the
pulse waveform are the mean of the recordings of the 5-min
long interventions or the last 5 min of the resting period. Heart
rate was defined by the frequency of the systolic peaks. Systolic
and diastolic blood pressure, time intervals and amplitudes of
the two reflected waves were derived from analyses of the first
and second derivate of the pulse pressure waveform. The
definitions of the variables measured to analyse the pressure
waveform are given in Fig. 1. The amplitudes of the two major

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

Pulse wave analysis of ngertip blood pressure, C. Ahlund et al. 89

details). For the femoral artery measurements, the primary

outcome variable was the FVI phase II to FVI phase I ratio. Also,
the time difference between these two blood flow peaks was
recorded (femoral phases III). Phase II was followed by another
anterograde wave.

reflected waves were always analysed in relation with the first

systolic peak (denoted as a in Fig. 1) (McNeill et al., 2000;
Millasseau et al., 2000; Weinberg et al., 2001).
Blood ow measurements
The brachial and femoral artery blood flow was assessed by
external B-mode ultrasound imaging and Doppler 23 cm
above the elbow and at 23 cm distal of the inguinal ligament,
respectively, equipped with a 75-MHz linear transducer
(Acuson, Mountain View, CA, USA). From Doppler recordings,
the mean flow velocity integral (FVI) was calculated from three
to four cardiac cycles. A 60 angle correction was applied for
blood flow measurements. The shape of the flow waves is
shown in Fig. 2, where the evaluation of the effects of
provocations is also indicated. Measurements of reflected blood
flow were always related to the primary systolic blood flow, like
the pressure pulse wave estimates (Fig. 1).
For brachial recordings, FVI for the ejected stroke volume was
denoted as phase I, the first reversed blood flow wave, possibly
originating from the forearm, was denoted phase II and FVI for
the second anterograde blood flow wave was defined as phase III
(Fig. 2). For the brachial artery measurements, the primary
outcome variables were the FVI phase III to FVI phase I ratio.
Also the time difference between these two blood flow peaks
were recorded (brachial phases IIII), in analogy with the time
differences used for the pulse pressure wave. A second
retrograde phase followed by a third anterograde phase was
also seen in diastole.
For femoral recordings, FVI for the ejected stroke volume was
denoted phase I, the first reversed blood flow, possibly
originating from the leg, was denoted phase II (see Fig. 2 for

Statistics
In study protocol 1 and 2, differences between means were
calculated by paired t-test, and relationships between variables
were evaluated by linear regression analysis. In protocol 3,
effects of age, gender and exercise were analysed by two-way
ANOVA. P<005 was regarded as significant.

Results
Study protocol 1
Results obtained are shown in Table 1. Digital SBP and DBP
were lower than those obtained invasively. Systolic reflection
index (AIx; P<00001) and diastolic reflection index (RIx; P =
00009) were also lower when measured in the finger. Systolic
rise time (SRt, P<00001) and time to the Rt (P = 0001)
were both shorter when measured in the finger than in the
brachial artery. The time from the onset of systole to the At
(P = 0520 ns) did not differ between the two methods.
Importantly, all pulse wave characteristics were significantly
correlated when the intra-arterial and digital registrations were
compared (Table 1). When PWA was analysed at four
different occasions to investigate the reproducibility of the
method the CV for AIx was 16%, RIx 10%, At 6% and 4%
for Rt.

Brachial artery

Figure 2 Doppler recordings of blood flow in

the brachial and femoral artery. On the brachial
flow, three peaks in the anterograde direction
(I, III and V) and two in the retrograde
direction (II and IV) are easily identified. The
major peak I is due to the cardiac emptying,
followed by what is probably a reflected wave
from the lower arm and hand (II). A second
anterograde wave (III) is associated to systolic
augmentation AIx. Peaks IV and V have no
significant correlations to any other measure
obtained in PWA. In the femoral artery, the
systolic flow surge (peak I) is followed by a
flow reversal (peak II), and then by a third peak
in the anterograde direction. The magnitude of
peak II I is associated with the reflectance
index RIx.

Femoral artery

III

II

IV

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

III

II

90 Pulse wave analysis of ngertip blood pressure, C. Ahlund et al.

Table 1 Intra-brachial and non-invasive digital pressures.

Mean
Invasive
HR (bpm)
SBP (mmHg)
DBP (mmHg)
AIx
RIx
SRt (ms)
At (ms)
Rt (ms)

50
114
64
063
056
134
274
487

(8)
(11)
(6)
(008)
(007)
(12)
(16)
(22)

Non-invasive
49
107
56
049
049
121
271
478

(7)
(10)
(8)
(010)
(007)
(10)
(20)
(20)

P-value

012
0015
00032
<00001
00009
<00001
051
00011

098
074
057
070
075
084
075
094

HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure and pulse wave characteristics (see Fig. 1 for definitions) obtained
through invasive measurements in the brachial artery and non-invasive
measurements in the finger.
Mean and SD (in brackets) are shown. Statistical analysis showed significant differences in all measures except HR and At. However, there
was a strong correlation between all pressure contour derived measures.
Abbreviations as in Fig. 1.

Study protocol 2
Table 2 shows that SBP and DBP were increased by the stressful
interventions, with the most pronounced increase during COP.
Also HR differed significantly between the studied periods, with
the most pronounced increase seen during MAT. Both AIx and
RIx were significantly changed by the interventions. AIx was
mainly increased during COP, but also increased slightly during
MAT. RIx was reduced during MAT only. Also At and Rt were
significantly changed by the interventions. At was reduced more
during MAT than COP, Rt was reduced during MAT only.
There were positive associations between changes in flow and
pressure waves obtained during MAT and COP. Brachial artery
FV1 phase III to phase I ratio were significantly increased by
both interventions, the increase was larger for COP than MAT

Baseline
SBP (mmHg)
DBP (mmHg)
HR(bpm)
AIx
RIx
At (ms)
Rt (ms)
Brachial phase III I ratio
Brachial phase III I (ms)
Femoral phase II I ratio
Femoral Phase II I (ms)

120
60
64
044
042
246
422
018
218
041
197

Mental stress

(13)
(12)
(7)
(011)
(005)
(16)
(15)
(015)
(61)
(012)
(15)

135
74
85
047
036
223
399
027
190
027
170

(15)
(11)
(14)
(010)
(007)
(21)
(19)
(017)
(61)
(012)
(19)

(as the changes in AIx). Also the time between the phase I and
phase III FV1 peaks changed significantly during the interventions. Similar reductions in this time index were seen during
both MAT and COP.
Femoral artery FV1 phase II to phase I ratio was significantly
changed by the interventions. This ratio was mainly reduced
during MAT. Also the time between the phase I and phase II FV1
peaks changed significantly during the interventions, with the
most pronounced reduction seen during MAT. This pattern was
the same as that for RIx and Rt (Table 2). Thus, changes in
brachial artery flow in phase III were associated to changes in
systolic reflection, and the magnitude of flow reversal in the
femoral arteries was associated to diastolic reflection.
Study protocol 3
Resting conditions
Baseline characteristics of the subjects are shown in Table 3. The
males had higher blood pressure than females, but no significant
effect on blood pressure by age was found. Heart rate did not
differ between the groups.
Results from PWA at rest are shown in Table 3 and Figs 3 and
4. There was a significant positive correlation between AIx and
age (P<00001), while there was no significant effect of age on
the RIx. SRt increased significantly with age (P = 00003), while
on the contrary the time to the At decreased (P = 0002). There
were no significant interactions between age and gender
regarding these indices of PWA. Time from onset of systole to
the Rt, did not vary significantly with age. Females tended to
have higher AIx than males, although the difference was not
significant (P = 011). Also At tended to be shorter in male
subjects (P = 007), but Rt was not affected by gender.
During exercise, both diastolic (P = 00004) and systolic
blood pressure (P = 00003) were significantly higher in males
(Table 3). SBP was significantly higher in the older groups

Cold pressor
148
84
68
064
041
241
413
031
188
035
184

(15)
(11)
(13)
(015)
(007)
(22)
(19)
(012)
(59)
(012)
(21)

ANOVA

Table 2 Arterial pressures and flows at rest and

during stress.

00001
00001
00001
00001
0003
00001
00001
0002
003
00006
00002

PWA from digital pressure and brachial and femoral artery blood flow pulses during rest, mental
arithmetics and cold pressor tests. Mean values and SD (in brackets) are shown. All variables were
significantly affected by the stressors. Significant correlations were obtained between changes in
systolic reflection (AIx and At) and changes in brachial artery phase IIII measures, and between
changes in diastolic reflection (RIx and Rt) and changes in femoral artery phase II-I measures.
Abbreviations as in Figs 1 and 2.
 2007 The Authors
Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

Pulse wave analysis of ngertip blood pressure, C. Ahlund et al. 91

Table 3 PWA of finger pressures from older and younger males and females.
Female, younger
Age (years)
Height (cm)
Weight (kg)
HR (bpm) rest
SBP (mmHg) rest
DBP (mmHg) rest
SBP (mmHg) exercise
DBP (mmHg) exercise
SRt (ms) rest
SRt (ms) exercise

24
170
65
52
102
53
151
66
116
87

(3)
(7)
(10)
(7)
(8)
(5)
(18)
(10)
(8)
(7)

Female, older
53
164
60
54
97
52
174
70
138
95

Male, younger

(8)
(5)
(8)
(4)
(8)
(6)
(14)
(4)
(14)
(4)

25
182
72
53
111
60
180
79
123
89

Male, older

(4)
(6)
(10)
(7)
(14)
(6)
(20)
(11)
(13)
(8)

54
185
83
55
116
63
199
83
136
95

(8)
(8)
(13)
(9)
(17)
(8)
(17)
(11)
(12)
(8)

Age

Gender

Interaction

06859
04846
04408
09058
06092
00071
02489
00003
00054

<00001
00009
06421
00024
00002
00003
00004
05709
05689

01000
00414
08726
02902
03916
06821
09990
02831
06131

Blood pressure, SRt and descriptive charachteristics of the younger and older subjects studied at rest and during exercise. Mean values and SD (in
brackets) are shown. Statistical analysis showed that blood pressure was higher in males than in females both at rest and during exercise. Systolic blood
pressure during exercise was higher in the older groups. Systolic rise time was longer in the older subjects.

Reflectance index

Augmentation index
Rest

Exercise

Rest

Exercise

09
08
07

06
05
04

03
02

compared with the young. SRt was longer (P = 0005) in the

older groups, but was not affected by gender. Both AIx and RIx
fell markedly in both sexes during exercise. AIx was significantly
higher (P = 00001) and the Rt RIx lower (P = <00001) in the
older groups than in the younger groups. Also the time indices
were significantly shorter during exercise than in rest in both
sexes (P<00001). At was significantly shorter in the older
group compared with the young group (P = 00011), while Rt
was longer in the older than in the younger groups. Age did not
affect the length of the time indices during exercise.
Age and gender did not affect significantly the exercise
induced changes in AIx and At. However, exercise induced
changes in RIx was affected both by age (P = 002) and gender
(P = 001). Rt responses were not significantly affected by age

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

01
Female younger

Figure 3 Changes in augmentation and

reflectance indices at rest and during exercise
with age. Where indicated, significant differences between younger and older individuals
were found. In addition exercise induced
significant reductions in both AIx and RIx in all
groups. AIx was significantly increased by age
both at rest and during exercise, while RIx was
similar in all groups at rest but significantly
lower in older subjects during exercise.
*P<005.

or gender, although there was a trend towards a smaller

reduction in Rt changes in the older subjects.

Discussion
In this study, we provide further evidence for PWA from
continuous digital blood pressure measurements as a simple and
non-invasive technique to investigate arterial properties. First,
there was a direct and quantitative relationship between results
from PWA performed on recordings from invasive brachial
artery recordings and finger pressure measurements. Secondly,
support for different origins for the major reflected waves was
obtained by linking brachial and femoral artery blood flow
patterns to results from PWA. Thirdly, by extending previous

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

92 Pulse wave analysis of ngertip blood pressure, C. Ahlund et al.

Rt

At
Rest

Rest

Exercise

Exercise

500
*

Time (ms)

400
*

300

200

investigations in young subjects to a larger population including

different age groups of both genders, we conclude that the
technique is feasible to use in large populations and will provide
robust results.
Blood pressure falls continuously from the larger conduit
arteries to the smaller peripheral arteries, and systolic and
diastolic pressures measured in the finger are reported to be
515 mmHg lower than pressures simultaneously measured in
the brachial artery (Parati et al., 1989; Bos et al., 1992;
Karamanoglu & Feneley, 1997; Voogel & van Montfrans,
1997; Eckert & Horstkotte, 2002; Guelen et al., 2003)
(Table 1). Brachial artery pressures can be estimated from
digital pressures by the application of transfer function (Bos
et al., 1996; Guelen et al., 2003). Also, induced changes in
non-invasive digital pressures reflects changes in aortic and
brachial pressures (Parati et al., 1989; Eckert & Horstkotte,
2002). However, PWA goes beyond the determination of
systolic and diastolic pressure only, e.g. by analysis of the
timing and amplitude of the reflected wave(s) in relation with
the systolic wave (Fig. 1). Table 1 shows that although there
are differences in waveforms between the invasive and
non-invasive recordings in absolute values, there are indeed
significant correlations between the amplitudes and timings of
the systolic and Rt, and that AIx and RIx estimated from finger
pressure measurements are valid substitutes for invasively
obtained values. This is not surprising, as it is previously
shown that, similarly to intrabrachial artery pressure, finger tip
pressure measurements can be used to generate ascending aortic
pressures by the generation of a specific transfer function, and
e.g. aortic input impedance can thus be estimated from digital
pressures (Karamanoglu, 1997).
Potential reflecting sites in the arterial system are branching
points, areas of alterations in arterial distensibility and high-

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

Female younger

Female older
Male older

Male younger

Female younger

100

Figure 4 Changes in time intervals at rest and

during exercise with age. Where indicated,
significant differences between younger and
older individuals were found. In addition
exercise induced significant reductions in both
At and Rt in all groups. At was significantly
reduced by age both at rest and during exercise,
while Rt was similar in all groups at rest but
significantly longer in older subjects during
exercise. *P<005.

resistance arterioles (Nichols & ORourke, 1998). It is shown

that the distance from the site where pressure is measured to the
reflection site (Lp) can be estimated roughly from the equation
Lp = co*D(t)2, where co is pulse wave velocity, and D(t) is the
time from systolic upstroke to the upstroke of the reflected wave
(Murgo et al., 1980; Latham et al., 1985). Based on known pulse
wave velocity in young healthy individuals and the time analysis
of the digital pressure wave we suggested that the systolic and
diastolic reflected waves originated from the middle part of the
aorta and in the legs, respectively (Ahlund et al., 2003), this in
line with earlier findings of significant reflection sites at the
level of the renal arteries (Latham et al., 1985) and at the level
of the iliac bifurcation or its branches (Alexander, 1953;
Remington, 1963). Alterations in wave reflection should also
result in alterations in the arterial flow wave. The relation
between flow and pressure waves is complicated and require
estimates of phase and modulus to be analysed (Nichols &
ORourke, 1998). This could not be performed in the present
experiments, but we were able to find positive associations
between stress-provoked changes in flow reflection in the
femoral arteries and RIx, as well as between brachial artery flow
and AIx. The results are summarized in Table 2.
The MAT stress test is a widely used, well-characterized stress
test mimicking the fight or flight-reaction. During this test
blood pressure is raised mainly because of an increase in heart
rate and cardiac output. A vasodilatation in the limb skeletal
muscles is seen together with vasoconstriction in the renal and
intestinal vascular beds (Freyschuss et al., 1988; Folkow, 1997).
If we postulate that the At originates from the middle part of the
aorta and the diastolic reflection mainly originates from the
lower limbs, the well-known divergent effect on vascular
resistance in these vascular beds during MAT could affect AIx
and RIx differently. Indeed, while there was no major change in

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

Pulse wave analysis of ngertip blood pressure, C. Ahlund et al. 93

AIx during MAT, there was a marked reduction in RIx.

Furthermore, there was also a marked reduction in femoral
artery blood flow reversal, this finding supports the concept
that the diastolic reflected wave seen on the fingertip blood
pressure contour originates from reflection in the lower limbs.
Chowienczyk et al. (1999) recorded the blood flow pulse in the
finger using photoplethysmography, and they also concluded
that diastolic blood flow in the finger partly reflected circulation
in the lower body including limbs. However, we were unable to
find any clear association between changes in any brachial
artery blood flow waves and changes in the magnitude of flow
reversal in the femoral artery, in line with conclusions from
Karamanoglou et al. (Karamanoglu et al., 1995).
Cold pressor test is a widely used, highly standardized stress
test to examine the effects of painful stress. COP increases blood
pressure mainly through a generalized increase in vascular
resistance, more pronounced in the renal and splanchnic
circulation than in the limbs (Hines & Brown, 1936; Zbrozyna
& Westwood, 1990; Chaudhuri et al., 1991; Boddi et al., 1996;
Fennessy et al., 2003). The major catecholamine released during
this painful stress is noradrenaline, which previously has been
shown to increase AIx, as well as to reduce At (Wilkinson et al.,
2001; Millasseau et al., 2003). COP increased AIx in parallel
with an increase in brachial artery phase III flow (Fig. 2). No
major alterations in RIx or in the magnitude of the reversed
blood flow in the femoral artery were seen, being consistent
with the fact that only minor alterations in vascular resistance in
the lower limbs are seen during COP. The minor effects of COP
on RIx again support the view of the limbs as the reflection site
for the diastolic wave.
In these experiments, we could not use exercise, as we were
unable to record blood flow during bicycle exercise. However,
in a couple of subjects we were able to measure femoral artery
blood flow together with finger pressure while shifting from
supine to standing position, this was associated with major
reductions in both RIx and femoral phase II I ratio, supporting
the conclusion above (not shown). Changes in AIx were
unrelated to changes in the femoral blood flow pattern.
Altogether, the data presented suggest that analysis of digital
blood pressure profiles is a simple, non-invasive way to
investigate properties in both a central aortic and a more
peripheral arterial site. If RIx correlates to, or even is determined
by, resistance vessels in the limbs, or to other arteries of larger
calibre remains to be investigated. As peripheral pressure
recordings can also be used to estimate aortic input impedance,
etc. if a transfer function that allows generation of ascending
aorta pressure profiles is generated (Karamanoglu, 1997),
digital pressure recordings can be used for a wide variety of
investigations of the CV system.
We have in an earlier study investigated how some different
stressors affected the pulse wave contour in a young, healthy
population (Ahlund et al., 2003). In this study, we divided the
subjects into four equally sized groups, younger and older
adults, females and males. When divided in these coarse groups,
systolic and diastolic blood pressure was slightly higher in male

than in female subjects, but in this healthy population, blood

pressure was not markedly different between the age groups.
Both the timing and amplitude of the At were, however,
significantly affected by age. The At arrived earlier and with
higher amplitude in the older group. Systolic reflection has been
investigated earlier with applanation tonometry and with direct
intravascular measurements with similar results (Nichols &
ORourke, 1998; ORourke et al., 2001). Timing and amplitude
of the diastolic reflected wave was unaffected by age, and there
was only minor gender differences. This clearly shows that the
systolic and diastolic reflected waves are not regulated by the
same mechanisms, indicating that age dependent changes occurs
to a lesser extent in the distal vasculature than in central parts of
the arterial tree. Several (but not all) investigators has shown
that aortic systolic augmentation is higher in women than in
men (Hayward & Kelly, 1997; Gatzka et al., 2001; Hope et al.,
2002; Kohara et al., 2005) and that there is a strong age
dependence on AIx (Hayward & Kelly, 1997; Kohara et al.,
2005). In this study, we noted a similar strong age-dependency,
while the difference between women and men did not reach
statistical significance.
Finally, it was investigated how a powerful change in central
and peripheral haemodynamics affect the arterial pulse contour.
Exercise induces vasodilatation in the working skeletal muscle,
but vasoconstriction in the renal and splanchnic arteries
(Middlekauff et al., 1997; Momen et al., 2003) together with a
profound sympathetic activation with tachycardia (Middlekauff
et al., 1997; Momen et al., 2003). We therefore expected major
changes in the pulse wave contour during exercise compared
with resting. Both the SBP and DBP was increased during
exercise and the reflected waves arrived earlier and had reduced
amplitudes compared with rest in all groups investigated. A
reduced systolic augmentation during exercise could thus at
least partly compensate for the increased cardiac workload at the
higher blood pressure level. However, the differences caused by
age and gender principally followed the same pattern as during
rest (Figs 3 and 4). The one difference found was that RIx was
slightly lower in older than in younger subjects during exercise,
this was not found at rest and may imply a higher degree of
limb vasodilatation in older than in younger subjects. This
finding can simply be a consequence of a proportionally larger
increase in pulse pressure in the older subjects by exercise.
However, in spite of a reduction in At with age, Rt was slightly
increased. If this was caused by a change in the reflection site
toward a more peripheral location, this would explain why Rt
was increased, and may also explain the effect on RIx. As
continuous finger pressure recordings were easily obtained
during exercise, the techniques used here would be very
interesting to apply during exercise challenges of patients with
cardiovascular disorders, such as angina pectoris or peripheral
artery disease.
In conclusion, we have with this study provided further
support for contour analysis of continuous, digital blood
pressure measurement as a practical tool to investigate arterial
function in different anatomical locations, and that the method

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

94 Pulse wave analysis of ngertip blood pressure, C. Ahlund et al.

can be used to study vascular reactions during stress provocations of the circulation. Furthermore, age and gender related
differences in systolic and diastolic reflection at rest are largely
maintained also during exercise, showing that the technique
used is well suited for studies of the arterial system in broad
populations.

References
Ahlund C, Pettersson K, Lind L. Influence of different types of stressors
on the waveform of the peripheral arterial pulse in humans. Blood Press
(2003); 12: 291297.
Alexander RS. The genesis of the aortic standing wave. Circulation (1953);
1: 145151.
Benetos A, Adamopoulos C, Bureau JM, Temmar M, Labat C, Bean K,
Thomas F, Pannier B, Asmar R, Zureik M, Safar M, Guize L. Determinants of accelerated progression of arterial stiffness in normotensive subjects and in treated hypertensive subjects over a 6-year period.
Circulation (2002); 105: 12021207.
Boddi M, Sacchi S, Lammel RM, Mohseni R, Serneri GG. Age-related and
vasomotor stimuli-induced changes in renal vascular resistance detected
by Doppler ultrasound. Am J Hypertens (1996); 9: 461466.
Bos WJ, Imholz BP, van Goudoever J, Wesseling KH, van Montfrans GA.
The reliability of noninvasive continuous finger blood pressure
measurement in patients with both hypertension and vascular disease.
Am J Hypertens (1992); 5: 529535.
Bos WJ, van Goudoever J, van Montfrans GA, van den Meiracker AH,
Wesseling KH. Reconstruction of brachial artery pressure from noninvasive finger pressure measurements. Circulation (1996); 94: 1870
1875.
Chaudhuri KR, Thomaides T, Hernandez P, Alam M, Mathias CJ. Noninvasive quantification of superior mesenteric artery blood flow
during sympathoneural activation in normal subjects. Clin Auton Res
(1991); 1: 3742.
Chowienczyk PJ, Kelly RP, MacCallum H, Millasseau SC, Andersson TL,
Gosling RG, Ritter JM, Anggard EE. Photoplethysmographic assessment of pulse wave reflection: blunted response to endotheliumdependent beta2-adrenergic vasodilation in type II diabetes mellitus.
J Am Coll Cardiol (1999); 34: 20072014.
Duprez DA, De Buyzere MM, De Bruyne L, Clement DL, Cohn JN. Small
and large artery elasticity indices in peripheral arterial occlusive disease (PAOD). Vasc Med (2001); 6: 211214.
Eckert S, Horstkotte D. Comparison of Portapres non-invasive blood
pressure measurement in the finger with intra-aortic pressure measurement during incremental bicycle exercise. Blood Press Monit (2002);
7: 179183.
Fennessy F, Casey RG, Bouchier-Hayes D. Peripheral and central
arterial haemodynamic interactions are early abnormalities in
young male cigarette smokers. Eur J Vasc Endovasc Surg (2003); 25:
152158.
Folkow B. Physiological aspects of the defence and defeat reactions.
Acta Physiol Scand Suppl (1997); 640: 3437.
Freyschuss U, Hjemdahl P, Juhlin-Dannfelt A, Linde B. Cardiovascular
and sympathoadrenal responses to mental stress: influence of betablockade. Am J Physiol (1988); 255 (6 part 2): H1443H1451.
Gatzka CD, Kingwell BA, Cameron JD, Berry KL, Liang YL, Dewar EM,
Reid CM, Jennings GL, Dart AM. Gender differences in the timing of
arterial wave reflection beyond differences in body height. J Hypertens
(2001); 19: 21972203.
Guelen I, Westerhof BE, Van Der Sar GL, Van Montfrans GA, Kiemeneij
F, Wesseling KH, Bos WJ. Finometer, finger pressure measurements

with the possibility to reconstruct brachial pressure. Blood Press Monit

(2003); 8: 2730.
Hayward CS, Kelly RP. Gender-related differences in the central arterial
pressure waveform. J Am Coll Cardiol (1997); 30: 18631871.
Hines E, Brown G. The cold pressor test for measuring the reactibility of
pressure: data concerning 571 normal and hypertensive subjects. Am
Heart J (1936); 11: 19.
Hope SA, Tay DB, Meredith IT, Cameron JD. Comparison of generalized and gender-specific transfer functions for the derivation of
aortic waveforms. Am J Physiol Heart Circ Physiol (2002); 283:
H1150H1156.
Karamanoglu M. A system for analysis of arterial blood pressure waveforms in humans. Comput Biomed Res (1997); 30: 244255.
Karamanoglu M, Feneley MP. Derivation of the ascending aortic-carotid
pressure transfer function with an arterial model. Am J Physiol (1996);
271 (6 part 2): H2399H2404.
Karamanoglu M, Feneley MP. On-line synthesis of the human ascending
aortic pressure pulse from the finger pulse. Hypertension (1997); 30:
14161424.
Karamanoglu M, Gallagher DE, Avolio AP, ORourke MF. Pressure wave
propagation in a multibranched model of the human upper limb. Am J
Physiol (1995); 269 (4 part 2): H1363H1369.
Kohara K, Tabara Y, Oshiumi A, Miyawaki Y, Kobayashi T, Miki T.
Radial augmentation index: a useful and easily obtainable parameter
for vascular aging. Am J Hypertens (2005); 18 (1 part 2): 11S14S.
Latham RD, Westerhof N, Sipkema P, Rubal BJ, Reuderink P, Murgo JP.
Regional wave travel and reflections along the human aorta: a study
with six simultaneous micromanometric pressures. Circulation (1985);
72: 12571269.
McNeill KL, Fontana L, Russell-Jones DL, Rajman I, Ritter JM, Chowienczyk PJ. Inhibitory effects of low-density lipoproteins from men
with type II diabetes on endothelium-dependent relaxation. J Am Coll
Cardiol (2000); 35: 16221627.
Middlekauff HR, Nitzsche EU, Nguyen AH, Hoh CK, Gibbs GG. Modulation of renal cortical blood flow during static exercise in humans.
Circ Res (1997); 80: 6268.
Millasseau SC, Guigui FG, Kelly RP, Prasad K, Cockcroft JR, Ritter JM,
Chowienczyk PJ. Noninvasive assessment of the digital volume pulse.
Comparison with the peripheral pressure pulse. Hypertension (2000);
36: 952956.
Millasseau SC, Patel SJ, Redwood SR, Ritter JM, Chowienczyk PJ. Pressure
wave reflection assessed from the peripheral pulse: is a transfer
function necessary? Hypertension (2003); 41: 10161020.
Momen A, Leuenberger UA, Ray CA, Cha S, Handly B, Sinoway LI. Renal
vascular responses to static handgrip: role of muscle mechanoreflex.
Am J Physiol Heart Circ Physiol (2003); 285: H1247H1253.
Murgo JP, Westerhof N, Giolma JP, Altobelli SA. Aortic input impedance
in normal man: relationship to pressure wave forms. Circulation
(1980); 62: 105116.
Nichols WW, ORourke ME. McDonalds Blood Flow in Arteries. Theoretical, Experimental and Clinical Principles (1998). Arnold, London.
ORourke MF, Pauca A, Jiang XJ. Pulse wave analysis. Br J Clin Pharmacol
(2001); 51: 507522.
Parati G, Casadei R, Groppelli A, Di Rienzo M, Mancia G. Comparison of
finger and intra-arterial blood pressure monitoring at rest and during
laboratory testing. Hypertension (1989); 13 (6 part 1): 647655.
Pauca AL, Wallenhaupt SL, Kon ND, Tucker WY. Does radial artery
pressure accurately reflect aortic pressure? Chest (1992); 102: 1193
1198.
Remington JW. The Physiology of the Aorta and Major Arteries. American Physiological Society Handbook of Physiology, Section 2, Circulation (1963). American
Physiology Society, Washington, D.C., 2: 799838.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695

Pulse wave analysis of ngertip blood pressure, C. Ahlund et al. 95

Voogel AJ, van Montfrans GA. Reproducibility of twenty-four-hour

finger arterial blood pressure, variability and systemic hemodynamics.
J Hypertens (1997); 15 (12 part 2): 17611765.
Weber T, Auer J, ORourke MF, Kvas E, Lassnig E, Berent R, Eber B.
Arterial stiffness, wave reflections, and the risk of coronary artery
disease. Circulation (2004); 109: 184189.
Weinberg PD, Habens F, Kengatharan M, Barnes SE, Matz J, Anggard
EE, Carrier MJ. Characteristics of the pulse waveform during altered
nitric oxide synthesis in the rabbit. Br J Pharmacol (2001); 133:
361370.

Wilkinson IB, MacCallum H, Hupperetz PC, van Thoor CJ, Cockcroft JR,
Webb DJ. Changes in the derived central pressure waveform and pulse
pressure in response to angiotensin II and noradrenaline in man.
J Physiol (2001); 530 (part 3): 541550.
Zbrozyna A, Westwood D. Habitation and recovery of vascular
responses in calf and forearm and of the level of pain sensation
during the cold pressor test in man. Eur J Appl Physiol Occup Physiol
(1990); 61: 106111.

 2007 The Authors

Journal compilation  2007 Blackwell Publishing Ltd Clinical Physiology and Functional Imaging 28, 2, 8695