Noninvasive Determination of Age-Related Changes in the Human Arterial Pulse

1. 1652
Noninvasive Determination of Age-Related
Changes in the Human Arterial Pulse
R. Kelly, MB, FRACP, C. Hayward, MB, BSc,
A. Avolio, PhD, and M. O'Rourke, MD, FACC
Arterial pressure waves were recorded noninvasively from the carotid, radial, femoral, or all
three of these arteries of 1,005 normal subjects, aged 2-91 years, using a new transcutaneous
tonometer containing a high fidelity Millar micromanometer. Waves were ensemble-averaged
into age-decade groups. Characteristic changes were noted with increasing age. In all sites,
pulse amplitude increased with advancing age (carotid, 91.3%; radial 67.5%; femoral, 50.1%
from first to eighth decade), diastolic decay steepened, and diastolic waves became less
prominent. In the carotid pulse, there was, in youth, a second peak on the downstroke of the
waves in late systole. After the third decade, this second peak rose with age to merge with and
dominate the initial rise. In the radial pulse, a late systolic wave was also apparent, but this
occurred later; with age, this second peak rose but not above the initial rise in early systole, even
at the eighth decade. In the femoral artery, there was a single systolic wave at all ages. Aging
changes in the arterial pulse are explicable on the basis of both an increase in arterial stiffness
with increased pulse-wave velocity and progressively earlier wave reflection. These two factors
may be separated and effects of the latter measured from pressure wave-contour analysis using
an "augmentation index," determined by a computer algorithm developed from invasive
pressure and flow data. Changes in peak pressure in the central (carotid) artery show increasing
cardiac afterload with increasing age in a normal population; this can account for the cardiac
hypertrophy that occurs with advancing age (even as other organs atrophy) and the predispo-
sition to cardiac failure in the elderly. Identification of mechanisms responsible offers a new
approach to reduction of left ventricular afterload. (Circulation 1989;80:1652-1659)
A lthough the arterial pulse is the most funda- manometer. Despite classic texts on the pressure
mental of physical signs and has been used pulse contour by Mackenzie3 and by Wiggers,4
by clinicians for hundreds of years, it was written in the early twentieth century, the sphyg-
not until the last century that the contour of the momanometer was readily embraced as being more
pressure pulse was first recorded noninvasively in ,"scientific" because it was able to quantify blood
humans. The sphygmograph introduced by Marey pressure in terms of systolic and diastolic numbers.
in 18601 was sufficiently simple to be applied in This ascendancy of sphygmomanometric values was
clinical practice; this was done by Mahomed,2 who, also due to an absence of theory to describe or
utilizing the instrument, first described the clinical interpret the pulse contour, an inability to use pulse
entity of essential hypertension, and by Mackenzie,3 recordings usefully in clinical practice, and to prob-
who described changes in both pulse contour and lems with artifacts inherent in the available mechan-
rhythm observed in his general practice. This early ical recording systems. Thus, the use of the sphyg-
work on pulse-contour recordings was soon eclipsed mograph in describing pressure-pulse waves
by the introduction of the now ubiquitous sphygmo- declined, even as Einthoven's electrocardiograph
was being quickly accepted for its description of
From St. Vincent's Hospital (M.F.O'R.), Concord Hospital waves of electric activity.
(C.H.), and University of New South Wales (A.A., M.F.O'R.), With subsequent clinical use of the sphygmoma-
Sydney, Australia.
R.K. was a Scholar of the National Health and Medical nometer, brachial arterial pressure has been
Research Council of Australia and is now Overseas Research described in terms of the two extremes between
Fellow with the National Heart Foundation of Australia at The which it fluctuates, the systolic and diastolic pres-
Johns Hopkins University, Baltimore, Maryland. sures, and aging changes characterized as an increase
Address for correspondence: Michael F. O'Rourke, Medical
Professorial Unit, St. Vincent's Hospital, Victoria Street, Dar- in systolic pressure with little change in diastolic
linghurst, Sydney 2010, Australia. pressure within the brachial artery.5 Furthermore, it
Received October 27, 1988; revision accepted July 20, 1989. has been assumed that such change is the same in

2. Kelly et al Age-Related Pulse Changes 1653
APPLANATION TONOMETRY torted signal. Recordings are taken only when a
reproducible signal is obtained with high amplitude
excursion. Large amounts of overlying tissue make
optimal waveforms harder to obtain; the technique
is best applicable to superficial, large arteries such
as the radial, carotid, and femoral. The sensor is a
stiff ceramic beam on which are mounted piezo-
resistive elements forming two arms of a Wheat-
stone bridge. The remaining two arms are housed in
the connector. This is attached to a standard Millar
preamplifier box that amplifies the induced voltage
change by a factor of 100. This amplified, electri-
FIGURE 1. Diagram of applanation tonometry process.
cally calibrated signal was then recorded on mag-
Flattening of curved pressure containing structure allows netic tape. The accuracy of the probe has been
accurate registration of transmitted pressure, because previously validated in animals and in human sub-
stresses inherent in wall of curved surface are balanced jects with indwelling radial artery lines.9 Compari-
when it is flattened. sons were made between directly and indirectly
recorded radial artery pressure waves in 62 humans,
between directly and indirectly recorded femoral
all arteries. The effects of advancing age on the pressure waves in three dogs, and between directly
arterial pulse contour, however, have not been recorded ascending aortic pressure waves and indi-
precisely defined. rectly recorded carotid pressure waves in 17 humans.
The aim of this study was to document age-related Comparison both in the time domain and by spec-
changes in the arterial pulse contour, both in the tral analysis show excellent correspondence with
central and peripheral arteries, using an instrument theoretic predictions. The tonometer provides a
that would noninvasively register pressure wave high fidelity recording of arterial pressure wave
contour without artifact and with such accuracy that contour under a wide variety of clinical conditions
characteristics of the contour could be clearly and and pulse pressures. The stringent requirements
quantitatively defined, much as we can now clearly delineated by more recent theoretical studies8 apply
define features of electrocardiographic waves. Such to internally calibrated systems attempting to obtain
precisely documented changes in the normal pulse accurate noninvasive measurements of absolute sys-
with age will allow valid determination of the effects tolic and diastolic pressure levels. When, however,
of disease states on the pressure wave contour. applanation tonometry is used to record pressure
wave contour to supplement sphygmomanometric
Methods readings, these exacting conditions are not neces-
The Instrument sary. Accurate recordings of wave contour can
To document aging changes in pulse contour in indeed be achieved under standard clinical condi-
large numbers of the general population, an accu- tions.
rate noninvasive method of measuring pressure- The tonometer voltage signal registers a pressure
pulse waves was needed. The instrument devised is wave with harmonic content that does not signifi-
a pencil-shaped probe held on the skin over the cantly differ from that of an intra-arterially recorded
maximal arterial pulsation either by hand or by a wave.9 The output of the Millar preamplifier box
micromanipulator. The probe incorporates a Millar electrically calibrates the signal in mm Hg (1 mV/
micromanometer in its tip and has the same high mm Hg) in the same way as the conventional Millar
fidelity response as conventional Millar catheters.6 unit. The use of the tonometer on an exposed vessel
Use of the probes is based on the principle of indeed records a waveform of amplitude virtually
applanation tonometry, as is used in ocular tonom- identical to that recorded intra-arterially.9 Percuta-
etry for registration of intraocular pressure. Essen- neous use of the probe records morphological fea-
tially, if one can flatten or applanate (Figure 1) the tures of the wave that are accurately reproduced
curved surface of a pressure-containing structure, over a wide range of pulse pressures9 although
then the circumferential stresses in the wall of the absolute pulse pressure recorded might be less
structure are balanced and the pressure registered reliable. Accordingly, the output of the instrument
by the sensor is the true intra-arterial pressure.7'8 is reported in this manuscript with an arbitrary
Like echocardiography and other noninvasive diag- amplitude scale of millivolt (mV) units.
nostic techniques, applanation tonometry requires
some training but readily can be mastered. Having Registration and Analysis
located the point of maximal arterial pulsation, the Waveforms were recorded on an FM TEAC
probe is placed over the vessel and pressed down magnetic tape recorder, digitized by a 12-bit analog
on the artery against underlying bone. If no flat- to digital converter with an acquisition rate of 1 kHz
tening is achieved, no consistent signal can be and entered into storage on disk by an IBM-AT
registered. Excessive flattening produces a dis- computer. For each subject studied, six to eight

3. 1654 Circulation Vol 80, No 6, December 1989
A1 4.
PP
m:mHg]
10
AUGMENTATION INDEX = APIPP
10cm/sec]
-_ j
FIGURE 2. Graphic representation of augmentation index
defined as ratio of AP/PP%. AP, pressure from shoulder
to late peak; PP, pulse pressure.
consecutively recorded pulses were averaged, their
alignment being triggered at the occurrence of the
maximum rise time determined by the smoothed
first differential of the signal (dP/dt). The averaged
pulse was then taken as the representative pulse for
that individual. An ensemble average of similar
representative waves from 30-70 other subjects in FIGURE 3. Graphic plotting of representative ascending
the same age decade was obtained to form a repre- aortic pressure (solid line above) and flow (dashed line
sentative wave and 2 SDs for that particular age above). Below is shown fourth derivative of pressure
decade. Thus, a series of decade waves was estab- (solid line). First zero crossing offourth derivative corre-
lished, each being the average of approximately 50 sponds to beginning of pressure wave upstroke (AA).
representative waveforms from individuals of simi- Second zero crossing in same direction (from above to
lar age. Subjects with heart rates greater than 100 below) corresponds to beginning of second wave, which
beats/min were excluded to avoid excessive trunca- is close to peak of flow (BB).
tion of the age-decade wave (to that of the subject
with the shortest heart period) during the averaging
process. ascending aortic pressure wave, the time deriva-
Carotid waveforms were further analyzed to mea- tives of pressure, and the peak of simultaneously
sure the "shoulder" and "peak" of the waves. recorded flow. The simultaneously recorded pres-
Such analysis of wave contour has been done sure and flow-velocity data were digitized at 0.01-
previously by Murgo et al0 and Takazawa"l for second intervals and were plotted with the first four
invasively recorded ascending aortic waves and, derivatives of pressure. When plotted against pres-
also, by Fujii et a112 for noninvasively recorded sure derivatives, the timing of the shoulder was
waves. We defined the augmentation index for each reliably indicated by a local minimum in the first
wave as the ratio of height of the peak above the derivative that was in the range from 0 to 50 msec of
shoulder of the wave to the pulse pressure (Figure the peak of flow (the mean estimate from interpola-
2). This can be done from visual inspection as has tion between sample points determined from 13
been used previously.10-12 Such identification, how- patients) being 27 msec once correction had been
ever, can be highly subjective because the shoulder made for the frequency response of the flowmeter
is often not a clearly defined point but sometimes a (10 msec). To simplify the algorithm for detection of
less well defined plateau region on the systolic this point, higher-order derivatives were used to
upstroke. We, therefore, chose to identify the shoul- identify the zero-crossing point equivalent to the
der automatically from the time derivative of the local minimum of the first derivative. The first zero
pressure wave, as herein described. crossing of the fourth derivative (in a direction from
above to below the zero line) corresponded to the
Measurement and Justification of beginning of the pressure wave upstroke (line AA,
Augmentation Index Figure 3). The second zero crossing in the same
From invasive ascending aortic pressure and flow direction corresponded to the shoulder at the begin-
velocity data previously reported,'3 we have noted ning of the second wave, which constituted the late
there exists a correspondence between the peak of systolic peak (line BB, Figure 3). A good correla-
flow and the shoulder of pressure. The shoulder of tion was found between the time to the second zero
the pressure wave is defined as the first concavity crossing of the fourth derivative (x) and the timing
on the upstroke of the wave and separates the initial of the peak of flow (y) in the patients studied
pressure rise from the late systolic peak that occurs (y=0.91+1.31x; R=0.75).
in middle aged and older subjects.10 A computer Such a relation between peak flow and the shoul-
algorithm was written to identify this shoulder in a der of the pressure wave is to be expected if the
more objective, automated way than visual inspec- shoulder of the wave indicates the pressure rise
tion by using time derivatives of the pressure wave. resulting from peak flow input into the vasculature
Reanalysis of the invasive data was performed to before the effects of wave reflection.10"13,'4 Hence,
ascertain the relation between the shoulder of the we decided to use the algorithm as described to

4. Kelly et al Age-Related Pulse Changes 1655
RADIAL PULSE CONTOUR CAROTID PULSE CONTOUR
I 10 mV UNITS' DECADE I10 mv UNfTS
DECADE
2
2
3
3
4 4.
5 5
6
6
7
7
8
a
0 1 50 300 450 600 750
0 1 50 300 450 600 7S0
TIME(msec) TIME(msec)
FIGURE 4. Contours showing averaged radial waves FIGURE 5. Contours showing averaged carotid waves
from 420 subjects. Radial waveforms are displayed above from 407 subjects. Carotid pulse contour from first dec-
each other from infants in the first decade (1), through ade (1) to eighth decade (8), each pulse being average of
children, to adults up to the eighth decade (8). Each pulse 40-70 individual pulses. *Amplitude is expressed in mV
is average of 40-70 individual pulses. *Amplitude is units.
expressed in uncalibrated voltage (m V) units although
the Millar preamplifier box electrically calibrates the
output signal in mm Hg. Results
The radial pulse contour recorded in 420 subjects
detect the contribution of wave reflection to chang- showed characteristic changes with increasing age
ing carotid pulse contour. The method used is auto- (Figure 4). The radial pulse contour in children
matic and excludes subjective bias. It identifies a shows multiple prominent fluctuations. With advanc-
point on the pressure wave very close to peak flow in ing age, these become less distinct and the systolic
the ascending aorta, and is justifiable on the theoret- peaks progressively broader, although the maxi-
ical basis of wave travel and reflection in the aorta. mum still usually occurs in early systole. In con-
trast, the carotid wave recorded in 407 subjects
Subjects shows not so much a broadening but the emergence
Volunteers studied were from the out-of-hospital of the late systolic peak, which determines the
community-based population considered to repre- systolic pressure level (Figure 5). With increasing
sent the normal population. Volunteers were Cau- aging, the principal change in the carotid pulse is a
casians from both urban and rural areas who were progressive rise in the second systolic peak that,
2-91 years old. They were screened on the basis of after the third decade, merges with and dominates
cardiovascular history and examination; those with the initial rise. The femoral waveforms (Figure 6)
valvular heart disease or chronically treated cardio- also show a progressive rise in the systolic wave
vascular disease were excluded. Subjects found to and loss of any diastolic wave.
have mild hypertension (diastolic blood pressure The amplitude (pulse pressure) of the carotid
>95 and <105 mm Hg), 4.8% of the sample, were wave increases by 91.3% from the first to eighth
included in analysis as part of the spectrum in the decade compared with a 67.5% increase in the
general population. Radial pressure waveforms were radial pulse and a 50.1% increase in the femoral
obtained from 420 subjects (207 men and 213 wom- pulse. This greater increase is due to change in two
en), of whom 38 were smokers; carotid waveforms parts of the carotid wave. After the third decade of
were from a further 407 subjects (181 men and 226
women), of whom 82 were smokers; and femoral life, a late systolic peak becomes dominant, which
waveforms were from 178 subjects of similar age adds to the initial pressure rise and so defines the
(110 men and 68 women), of whom five were shoulder on the upstroke of the wave. The total 91%
smokers. Of all the subjects studied, 28 were on increase in carotid pulse pressure from the first to
monotherapy for hypertension at the time of study. the eighth decade is due to an increase in both the
rise to this initial shoulder (by an average 20 mV
Statistical Analysis units or 53% of the increase) and the height of the
Statistical analysis was performed using a two-tail late peak above the shoulder (averaging 18 mV
unpaired Student's t test with a significant difference units) (Table 1). This change in carotid pulse con-
being at the level ofp equaling less than 0.05. tour was quantified by the augmentation index

5. 1656 Circulation Vol 80, No 6, December 1989
FEMORAL PULSE CONTOUR TABLE 2. Ascending Aortic Blood Flow Velocity Profile
DECADE Age Age
<30 yr >60 yr p
Average peak flow
velocity (cm/sec) 94+10 67+13 0.01
2
Average acceleration
(cm/sec2) 921±+ 113 650+280 0.05
3
Mean ejection time
4 (msec) 290+30 293+38 NS
Values are mean+one SD.
5
6
although rate of flow acceleration and peak flow
velocity are reported to decline with increasing
age.16 We sought to verify that our study population
8 concurred with these results by studying a sample
of subjects using Doppler flow techniques. From
0 1 50 300 450 600 750
the original cohort, a group of 12 older (>60 years)
TIME(msec) and 13 younger (<30 years) subjects were studied.
FIGURE 6. Contours showing averaged waves from 178 Parameters compared were peak flow velocity (cm/
subjects. Femoral pulse contour from first decade (1) to sec), acceleration of systolic upstroke, and ejection
eighth decade (8). *Amplitude is expressed in mV units. time. This group spanned a wider range of age
(19-80 years) than the data previously reported by
Nichols et al.14 Ultrasonography was done with a
measured by the algorithm developed using inva- Toshiba Sonolayer SSH 65 A phased array ultra-
sive aortic pressure and flow data. sonograph and a 2.5-MHz pulsed-wave Doppler
When this algorithm was applied to the carotid probe. Recordings were made with subjects in a
wave age-decade data, it was found that although supine, rested position. Recordings of ascending
the timing of the shoulder was within 14 msec for all aortic flow were taken from the suprasternal notch
eight decades (range, 102-116 msec, after the onset (where sampling was done near the inner curvature
of systole), the height of the shoulder and the late at the aortic root), the right parasternal, and the
peak both showed a substantial increase with age. apical positions. Signals analyzed were those that
(Table 1). The height of the shoulder increased from gave the best delineated velocity profile at any of
39 to 59 mV units above diastolic pressure from the the three positions.
first to the eighth decade, whereas the late peak As has been found by other workers, ejection
increased from less than 1 to 19 mV units above the time remained constant with increasing age, although
shoulder. The corresponding augmentation index peak flow velocity and acceleration did decline with
increased from 1.6% to 24.1%. increasing age (Table 2). As will be discussed, the
Changes in pressure-pulse contour are due to changes in pressure wave contour with increasing
changes in the systemic vasculature itself or to age are explicable mainly on the basis of changes in
differing flow input into the vasculature from the the systemic vasculature with increasing age because
heart. Previous studies have shown that at rest there were few changes in flow input into the
there is little change in stroke volume or ejection system as detected by ascending Doppler flow mea-
fraction with increasing age.15 Aortic flow-velocity surements (Figure 7).
profile does not alter in contour with increasing age,
Discussion
TABLE 1. Carotid Waveform Analysis These characteristic age-decade pulse waveforms
Height of represent the largest sample of noninvasive record-
Height of peak Augmen- ings of the pulse yet made in human subjects. The
Time to shoulder above tation accuracy of the recordings is dependent on both the
Age shoulder above foot shoulder index principle of applanation tonometry, which is based
(yr) (msec) (mV units) (mV units) (%) on well established theory,8 and also the use of high
1-10 116 39.4 0.6 1.6 fidelity Millar micromanometers incorporated into
11-20 116 50.1 1.6 3.0 the tip of the probes. The technique has been shown
21-30 110 51.3 2.7 4.9 to be accurate in registering intra-arterial pressure
31-40 110 46.5 3.7 7.4 waves as compared in the time domain and by
41-50 102 42.6 8.4 16.5 spectral analysis.9 Previous noninvasive pulse
51-60 102 43.0 11.1 20.5 recordings have used wave tracings obtained by
61-70 102 47.1 14.9 24.1 displacement of a volume of air or fluid contained in
71+ 106 59.0
a capsule fixed over the arterial pulsation.'7,18 Such
18.7 24.1
volume displacement techniques are limited by

6. Kelly et al Age-Related Pulse Changes 1657
cess used to obtain age-decade waveforms resulted
in obvious smoothing of high frequency compo-
100 nents of the wave such as the incisura. Never-
theless, there is still a clear change in pressure wave
contour with increasing age. There is a progressive
U
rise in a late systolic fluctuation in both so that it
-
becomes the late peak of pressure in the carotid
c artery after the third decade, and that becomes
equal to the initial pressure peak in the radial pulse
at the eighth decade. The result is a greater increase
in carotid pulse pressure with increasing age than in
either of the peripheral pulses because, in the
carotid, the late peak is added to the initial pressure
rise. The rise in this late systolic fluctuation in both
the radial and carotid pulses is explicably due to
faster return of wave reflection from the lower
body. Such altered timing of wave reflection has
40 mVT 80 mV been discussed at length elsewhere and is largely
wL UNITS1 UNITS' responsible for the suboptimal coupling of the vas-
a. ' .
culature to left ventricular function that is known to
occur with increasing age.14
As aging occurs, there are dramatic changes in
the vasculature that increase cardiac afterload.14,19
YOUNG OLD Both nonpulsatile (peripheral resistance) and pulsa-
FIGURE 7. Upper: Ascending aortic blood flow velocity tile (characteristic impedance and wave reflections)
profile recorded by Doppler ultrasonography in 19- components of arterial load increase with increasing
year-old male (young) and 72-year-old male (old). Lower: age. The age-related increase in left ventricular
Carotid pressure pulse contour recorded by arterial tonom- pressure load is primarily due to increased aortic
etry in same individuals. With increasing age, there is stiffness (characteristic impedance) that increases
marked increase in pulse pressure, mostly due to late aortic and left ventricular pressure at its early
systolic peak, with concomitant lesser reduction in peak systolic peak or shoulder and secondarily due to
flow velocity. *Amplitude is expressed in mV units. augmented early wave reflection that raises pres-
sure from the shoulder to the late systolic peak.10"4
movement artifact, by errors due to the large size of The results in Table 1 show these two effects on the
the capsule relative to the size of the artery, by carotid pressure pulse as they increase from the first
inability to separate carotid and venous pulsations, compared with the eighth decade. In the intervening
by need for a significant "hold-down" force for years, however, the increase in pulse pressure with
optimal signal registration, and by a high damping increasing age is primarily due to an increase in
coefficient causing attenuation of high frequency wave reflection (height of the late systolic peak).
characteristics. Nevertheless, these techniques have Although the 20-mV-units increase in shoulder height
given valuable information and extended the earlier across the full age spectrum is consistent with the
work of Marey,' Mahomed,2 and Mackenzie.3 In known increase in aortic stiffness of large arteries
contrast to such capsules that rely on magnification with increasing age, its constancy in middle age
of vessel wall displacement, tonometry is not depen- may be explained by the concomitant decrease in
dent on volume changes. It requires minimal hold- peak velocity that occurs with increasing age.'6
down force and is much less subject to artifact Although, in middle years the progressive increase
related to placement over the pulse because the in arterial stiffness is balanced by a gradual decrease
micromanometer in the tip of the probe is very in peak flow from the heart, with the passage of
small (0.5 x 1.0 mm, o.d.) compared to the diameter time, the degenerative vascular changes are greater
of the artery. Tonometry, thus, registers the arterial and, by the seventh and eighth decades, there is an
pressure pulse much as the clinician feels it by increased characteristic impedance and consequent
palpation. increased height of the shoulder above diastolic
The changes in the femoral pulse with age are pressure. This study shows that, with increasing
characterized by a progressive increase in the size age, increased aortic stiffness and increased early
of the systolic peak with a resultant increase in wave reflection contribute approximately equally to
pulse pressure, whereas diastolic fluctuations dimin- ventricular load as assessed by the central pressure
ish. There are no secondary systolic fluctuations in pulse (20 versus 18 mV units).
the femoral pulse. In contrast, the carotid and radial The increase in the late systolic pressure peak
waves show an increase in pulse amplitude with with increasing age is substantial and represents an
age, as well as important changes in the contour of increase of 25% of pulse pressure in the carotid
secondary systolic fluctuations. The averaging pro- artery between the ages of 30 and 60 years (Figure

7. 1658 Circulation Vol 80, No 6, December 1989
AUGMENTATION INDEX - CAROTID ARTERY in part, reversible by the use of nitroglycerin, which
reduces the reflected wave component.13,23-24
These characteristic age-related changes in the
60 pulse waveform have established a profile of normal-
aging changes that will allow the effects of disease
-
40 - states to be more precisely defined. The increase in
systolic and pulse pressure, particularly in the cen-
z
20 - tral carotid pulse, have important implications for
W
0
cc
the aging population. Increase in peak systolic
0. pressure and increased pulsation around the mean
pressure level result in increased cyclic stresses on
the structure of arterial walls and account for the
propensity of large arteries to rupture or develop
5 15 25 35 45 55 65 75 +
atheromatous occlusive disease with increasing
AGE
age.25 The findings in this study concur with previ-
FIGURE 8. Graph of augmentation index for carotid
ous epidemiologic data, showing the importance of
artery as function of age. systolic pressure as a risk factor for cardiovascular
morbidity and mortality with advancing age.26'27 In
addition, these findings provide important answers
8). This may be compared with previous data of to the question posed by the most recent of these
Murgo et al10 and Takazawa"l for invasive data from studies,28 regarding the determinants of systolic
the ascending aorta. These two studies show a 40% pressure levels in both central and peripheral
increase in pulse pressure between the ages of 30 arteries.
and 60 years (Figure 9). Comparison of Figures 8 Furthermore, the fact that the radial pulse shows
and 9 shows that although the carotid pressure different but complementary changes to those in the
pulse tends to underestimate the magnitude of age- carotid pulse has important implications for diag-
related changes, the carotid pulse detected by appla- nostic approaches and treatment. It is clear that the
nation tonometry is a good noninvasive guide to increase in radial (and femoral) systolic pressures
events in the central aorta. underestimate the age-related rise in central systolic
The relevance of these pressure changes to left pressure because it does not include the character-
ventricular afterload is apparent when one consid- istic rise in the late systolic peak seen in the central
ers known changes in left ventricular structure and arteries of mature adults. Hence, in subjects aged
function with increasing age. The increase in the 40-70 years, measurements of sphygmomanometric
late systolic peak in the carotid pulse occurs from pressure might substantially underestimate the effect
the third decade onward, which is the same time at of vasoactive drugs on ventricular afterload because
which the age-related increase in left ventricular a reduction in the late carotid pressure peak might
mass commences.20,21 This increase in mass with
increasing age is accompanied by changes in left occur with little or no change in peripheral systolic
ventricular function.22 Furthermore, recent studies pressure. This is because, in the peripheral pulse,
have shown that the increased ventricular load and the reflected wave is on the downstroke of systole
suboptimal ventricular-vascular coupling repre- rather than at its peak and, although reduced by
sented by these pressure-pulse changes are, at least vasoactive drugs, it has little effect on peak systolic
pressure. On the other hand, further investigation
is warranted to see if a predictive index of central
AUGMENTATION INDEX - ASCENDING AORTA systolic pressure changes might be obtained from
80
analysis of the late systolic fluctuation in the
60 peripheral (radial) pressure pulse waveform.
The determination of age-related changes in the
J 40 r arterial pulse by high fidelity applanation tonome-
try, thus, provides important supplementary infor-
20
z
W mation to that obtained by use of the sphygmoma-
cc
W' 0-
nometer. The use of this modern-day sphygmograph
enhances new investigations of the ill effects of
-20v TWZAWA aging and disease states on cardiovascular function
and extends the earlier observations of Mahomed,2
-40* Mackenzie,3 and Freis et al.17
0 10 20 30 40 50 60 70 80
AGE Acknowledgments
FIGURE 9. Scatterplot ofaugmentation indexfor ascend- We acknowledge J. Ganis, BSc, and J. Daley,
ing aorta as function of age; from data published by RN, for assistance in data collection and prepara-
Murgo et a110 (n) and Takazawa" (A). tion of this manuscript.

8. Kelly et al Age-Related Pulse Changes 1659
The arterial tonometer was developed with the 16. Gardin JM, Burn CS, Childs WJ, Henry WL: Evaluation of
assistance of Mr. Huntley Millar, Houston, and Dr. blood flow velocity in the ascending aorta and main pulmo-
nary artery of normal subjects by Doppler echocardiogra-
D. Winter, South-West Research Institute, San phy. Am Heart J 1984;107:310-319
Antonio, Texas. 17. Freis ED, Heath WC, Luchsinger PC, Snell RE: Change in
the carotid pulse which occur with age and hypertension.
References Am Heart J 1966;71:757-765
1. Marey EJ: Recherches sur le Pouls au Moyen d'un Nouvel 18. Craige E, Smith D: Heart sounds: Phonocardiographs;
Apareil Enregistreur-le Sphygmographe. Paris, E. Thunot carotid, apex and jugular venous pulse tracing; and systolic
et Cie, 1860 time intervals, in Braunwald E (ed): A Textbook of Cardio-
2. Mahomed F: The aetiology of Bright's disease and the vascular Medicine. Philadelphia, Saunders, 1988, pp 41-64
prealbuminuric stage. Med Chir Trans 1874;57:197-228 19. Yin FCP: Aging and vascular impedance, in Yin FCP (ed):
3. Mackenzie J: The Study of the Pulse. Edinburgh, McMillan, Ventricular/Vascular Coupling. New York, Springer-
1902, pp 8-28, 67-82 Verlag, 1988, pp 115-137
4. Wiggers CJ: The Pressure Pulses in the Cardiovascular 20. Linzbach AJ, Akuomoa-Boateng E: Alternsveranderungen
System. London, Longmans, Green and Co, 1928, pp 65-90 des menschlichen Herzens. 1. Das Herzgewicht in Alter.
5. Roberts J (ed): Blood Pressure Levels of Persons 6-74 Klin Wochenschr 1973;51:156-163
Years. United States, 1971-1974 21. Gerstenblith G, Fredericksen J, Yin FCP, Fortuin NJ,
6. Murgo JP, Giolma JP, Altobelli SA: Signal acquisition and Lakatta EG, Weisfeldt ML: Echocardiographic assessment
processing for human haemodynamic research. Proc IEEE of a normal adult aging population. Circulation 1977;
1977;65:696-702 56:273-278
7. Mackay RS, Marg E, Oechsli R: Automatic tonometer with 22. Nichols WW, O'Rourke MF, Avolio AP, Yaginuma T,
exact theory: Various biological applications. Science 1960; Murgo JP, Pepine CJ, Conti CR: Age-related changes in left
131: 1688-1689 ventricular/arterial coupling, in Yin FCP (ed): Ventricular!
8. Drzewiecki GM, Melbin J, Noordergraaf A: Deformational Vascular Coupling. New York, Springer-Verlag, 1988, pp
forces in arterial tonometry. IEEE Front Eng Comput Health 79-114
Care 1984;28:642-645 23. Latson TW, Hunter WC, Katoh N, Sagawa K: Effect of
9. Kelly R, Hayward C, Ganis J, Daley J, Avolio A, O'Rourke nitroglycerin on aortic impedance, diameter and pulse wave
M: Non-invasive registration of the arterial pressure pulse velocity. Circ Res 1988;62:884-890
waveform using high-fidelity applanation tonometry. J Vasc 24. Fitchett D, Simkus G, Beaudiy J, Marpole D: Reflected
pressure waves in the ascending aorta: Effect of glyceryl
Med Biol 1989;3:142-149
10. Murgo JP, Westerhof N, Giolma JP, Altobelli SA: Aortic trinitrate. Cardiovasc Res 1988;22:494-500
input impedance in normal man: Relationship to pressure 25. O'Rourke MF: Hypertension, in O'Rourke MF (ed): Arterial
waveforms. Circulation 1980;62:105-116 Function in Health and Disease. Edinburgh, Churchill Liv-
11. Takazawa K: A clinical study of the second component of ingstone, 1982, pp 185-195, 210-224
left ventricular systolic pressure. J Tokyo Med Coll 1987; 26. Kannel WB, Castelli WP, McNamara PM, McKee PA,
45:256-270 Feinlab M: Role of blood pressure in the development of
12. Fujii M, Yaginuma T, Takazawa K, Noda T, Komatsu H, congestive heart failure. N Engl J Med 1972;287:781-787
Katsui T, Watabiki H, Kawada Y, Hosoda S: Non-invasive 27. Kannel WB, Wolf PA, McGee DL, Dawber TR, McNamara
detection for reflection wave in the arterial system, in P, Castelli WP: Systolic blood pressure, arterial rigidity and
Automedica vol. 9. New York, Gordon & Breach, 1987, p 49 risk of stroke. The Framingham Study. JAMA 1981;
13. Yaginuma T, Avolio AP, O'Rourke MF, Nichols WW, 245:1225-1229
Morgan JP, Roy P, Baron D, Branson J, Feneley M: Effects 28. Rutan GH, Kuller LH, M Neaton JD, Wentworth DN,
of glyceryl trinitrate on peripheral arteries alters left ventric- McDonald RH, McFate-Smith W: Mortality associated with
ular hydraulic load in man. Cardiovasc Res 1986;20:153-160 diastolic hypertension and isolated systolic hypertension
among men screened for Multiple Risk Factor Intervention
14. Nichols W, O'Rourke M, Avolio A, Yaginuma T, Murgo J, Trial. Circulation 1988;77:504-514
Pepine CJ, Conti CR: Effects of age on ventricular-vascular
coupling. Am J Cardiol 1985;55:1179-1184
15. Weisfeldt M, Lakatta EG, Gerstenblith G: Aging and cardiac
disease, in Braunwald E (ed): A Textbook of Cardiovascular KEY WORDS * applanation tonometry * arteriosclerosis .
Medicine. Philadelphia, Saunders, 1988, pp 1650-1662 wave reflection