Download Abnormal arterial flow pattern in untreated

Clinical Science ( 1990) 78,43 1-435
43 1
Abnormal arterial flow pattern in untreated essential
hypertension: possible link with the development of
atherosclerosis
CHRISTOPHER J. H. JONES'**,DONALD R. J. SINGER3, NICHOLAS V. WATKINS',
GRAHAM A. MAcGREGOR~AND COLIN G. CARO'
'Physiological Flow Studies Unit, Imperial College, 'Brompton Hospital, and "Blood
School, London
Pressure Unit, St George's Hospital Medical
(Received 29 August 1989/3 January 1990; accepted 16 January 1990)
SUMMARY
1. We compared the velocity waveforms in the superficial femoral artery measured by multichannel Doppler
ultrasound in 45 subjects: 21 patients with untreated
essential hypertension and 24 normal subjects of similar
age and sex.
2. The pattern of arterial flow was abnormal in hypertensive patients, with the acceleration time, the duration
of reverse flow and the time to maximum flow reversal
being abbreviated. The internal arterial diameter, calculated from the velocity profile, was reduced despite raised
pressure, suggesting altered arterial wall mechanics in
essential hypertension.
3. These abnormalities will influence the wall shear
stress, a major determinant of arterial function. The
abnormal arterial wall mechanics and abnormal blood
flow pattern may contribute to the increased risk of
arterial disease in patients with untreated hypertension.
Key words: atherosclerosis, blood flow, Doppler ultrasound, femoral artery, hypertension, shear stress.
INTRODUCTION
The pattern of blood flow near the wall in large arteries
has been shown to influence many aspects of the biology
of the wall [l, 21. It may, in addition, influence the
development of atherosclerosis: the disease exhibits a
predilection for regions where the time-averaged wall
shear stress is expected to be low [3]. Essential hypertension is a major risk factor for the development of
atherosclerosis [4], although the mechanisms by which the
Correspondence: Dr C. J. H. Jones, Physiological Flow
Studies Unit, Imperial College of Science, Technology and
Medicine, Prince Consort Road, London SW7 2AZ.
risk is increased are unknown. Since the pattern of blood
flow may be altered by pharmacological interventions
which modify the mechanical properties of the arterial
wall [5], and arterial wall mechanics are abnormal in
hypertension [6],we compared the arterial flow pattern in
a group of patients with untreated essential hypertension
with that in a group of normotensive subjects. T h l
measurements were made non-invasively with multichannel Doppler ultrasound equipment [7]. A preliminary
account of the findings has been presented to the Physiological Society [8].
METHODS
We studied 21 patients (13 males, eight females; age
53 k 12 years, mean f SD) with established, uncomplicated
essential hypertension diagnosed using a semi-automated
ultrasound sphygmomanometer (Arteriosonde) with
attached recorder [9]. The patients had all been off antihypertensive treatment for the previous 3 weeks. We also
studied 24 normal subjects (16 males, eight females, age
48 zk 13 years), who had a normal resting blood pressure
and no past history of hypertension, and most of whom
were unfamiliar with the investigation. No account was
taken of female menstrual status in either group. All
measurements were made in the late morning in a room
with the ambient temperature controlled at 21 f 1"C. The
heart rate and blood pressure were measured noninvasively (Dynamap) immediately after the blood
velocity measurements were made, after the subjects had
rested supine for a period of 30 min. The measurement
site was the superficial femoral artery 10 cm below the
bifurcation of the common femoral artery. The
measurements were made with a 4.8 MHz range-gated
Doppler vessel-imaging system (Picker Internatio-naiLtd,
London). Velocity profiles across the diameter Of the
vessel were obtained by measurement of the instan-
'
C. J. H. Jones et al.
432
100
-1
- 60
Fig. 1. A representative velocity waveform of a normal subject obtained by plotting the spatially
averaged blood velocity, calculated from the velocity profile, against time. The acceleration time,
from the onset of flow to time of the peak forward flow velocity, is shown and the mean (time and
spatially averaged) velocity is represented by the horizontal broken line.
taneous blood velocity in adjacent sample volumes of
0.64 mm length at 25 ms intervals. T h e ensembleaveraged, spatially averaged blood velocity for 20 consecutive, ectopic free beats was plotted against time to
give the velocity waveform (Fig. 1). Internal arterial
diameter was defined as the width of the velocity profile at
peak forward velocity, obtained by the application of a
cubic spline algorithm. The volume flow rate was calculated. The pulsatility index (defined as the excursion
between maximum and minimum velocities during the
cardiac cycle divided by the spatially and temporally
averaged blood velocity) was also calculated.
Velocity profile data and the values of the spatially and
time averaged velocities, volume flow rate, arterial
diameter and pulsatility index were recorded on to floppy
disc using a BBC microcomputer. The velocity waveform
was calculated from the spatially averaged velocities
using a cubic spline interpolation and the following
measurements were taken: the maximum forward and
reverse velocities, the durations of initial forward flow
and reverse flow, and the acceleration time, defined as the
time from the onset of forward flow to peak forward flow,
and the time to the maximum reverse velocity.
The results were expressed as means k SD. The means
of the results from the hypertensive patients and the
normal subjects were compared using an unpaired t-test.
Correlation between variables measured in each group
was tested by calculation of Pearson’s product moment
correlation coefficient.
Table 1. Blood pressure, heart rate and arterial diameter
in hypertensive patients and normal subjects
Results are means fSD.Abbreviation: BP, blood pressure.
Systolic BP (mmHg)
Diastolic BP (mmHg)
Heart rate (beatslmin)
Arterial diameter (mm)
Hypertensive
patients
subjects
(n=21)
(n=24)
+
155 14
98 f 10
74 f I3
5.1 f 0.9
Pvalue
Normal
117 rt 10
73+9
<0.001
<0.001
<0.01
< 0.05
63f8
5.7
+ 0.9
RESULTS
Blood pressure, heart rate and arterial diameter (Table 1)
Blood pressure in the hypertensive group was
1 5 5 + 1 4 / 9 8 f 1 0 mmHg and in the normal group
117 10/73 f 9 mmHg ( P < 0.001). Heart rate was higher
in the hypertensive group than in the normal group
( 7 4 + 13 versus 6 3 + 8 beats/min; P<O.Ol). Arterial
diameter was lower in the hypertensive group (5.1 & 0.9
mm) compared with the normal subjects (5.7 0.9 mm)
( P < 0.05).
+
+
Arterial blood velocity waveform (Table 2)
Mean blood velocity, averaged both temporally and
spatially, was similar in the two groups (hypertensive
Arterial flow in hypertension
433
Table 2. Indices of arterial flow in hypertensive patients and normal subjects
Results are means f SD.Abbreviation: NS, not significant.
Hypertensive
subjects ( n= 2 1)
Mean velocity (cm/s)
Volume flow rate (ml/min)
Peak forward velocity (cm/s)
Peak reverse velocity (cm/s)
Acceleration time (ms)
Forward flow duration (ms)
Reverse flow duration (ms)
Time to peak reverse velocity (ms)
Pulsatility index
patients 7.0 If:3.2 cm/s, normal subjects 6.1 f3.3 cm/s;
(not significant).Although arterial diameter was lower in
the hypertensive patients, the volume flow rates were
similar in the two groups (hypertensive patients 92 k 5 1
ml/min, normal subjects 91 k 51 ml/min (not significant).
Peak forward and reverse velocities were similar in the
two groups, as was the pulsatility index.
Differences were observed, however, in the time course
of change of arterial velocity. Acceleration time was
reduced in patients with hypertension compared with
normal subjects (100 f25 versus 125 f25 ms; P< 0.01).
The duration of initial forward flow was normal in the
hypertensive patients, but the duration of reverse flow
was reduced (hypertensive patients 195 f40, normal
subjects 2 2 0 f 3 5 ms; P<0.05). The time interval
between the onset of flow and the peak reverse velocity
was also reduced in the hypertensive group ( 3 4 0 f 4 5
versus 370 k 30 ms; P< 0.01).
Relationships between blood flow pattern and blood
pressure
In the hypertensive group, but not in the normal
subjects, diastolic blood pressure correlated negatively
with mean blood velocity and reverse flow duration (both
P<O.Ol).There was no correlation between systolic or
diastolic blood pressure and acceleration time or the
duration of forward flow in either group.
There was no significant correlation between any of the
blood velocity measurements obtained and age, sex or
heart rate.
DISCUSSION
The present study has demonstrated significant abnormalities in the pattern of blood flow in the femoral artery
in patients with essential hypertension. The aceleration
time, the time to peak reverse flow velocity and the
duration of reverse flow are abbreviated. In addition,
internal artqial diameter is diminished. An abnormal
arterial flow pattern will be associated with abnormal
shear forces at the arterial wall and the changes may
thus contribute to the increased risk of arterial disease in
hypertension.
The blood velocity measurements were made in a rela-
7.0 f 3.2
92-151
. 48f16
-14f7
100 25
265 k 50
195f40
340 f 45
9.5 f 2.3
*
Normal
subjects (n= 24)
P value
6.1 f 3.3
91 f 5 1
44f15
-12f6
125f25
280 f 35
220 f 35
370 f 30
10.6 f 4.5
NS
NS
NS
NS
<0.01
NS
< 0.05
<0.01
NS
tively straight section of the femoral artery at some
distance from any major branches. The velocity vectors
would therefore be mainly parallel to the axis of the vessel
so that secondary motions would not give rise to appreciable errors in determining the spatially averaged velocity.
However, high-pass filtering, used to suppress the large
Doppler signals generated by movement of the strongly
reflective vessel walls, would suppress signals from the
low velocity flow near the wall, in practice flow with
velocity less than 5 cm/s. Thus there may have been
errors in determining the location of the wall and in the
calculation of the spatially averaged velocity during
periods in the cardiac cycle when blood velocities were
low. For these reasons, we, like other investigators in this
field, have made no attempt to calculate the velocity
gradient at the wall, the wall shear rate. These errrors
should not affect significantly the calculation of internal
arterial diameter which is determined at the time of peak
forward velocity when the low velocity regions are
expected to be small.
This study demonstrated differences between the timecourse of change of blood velocity during the cardiac
cycle in the patients with hypertension and the normal
subjects despite remarkable similarity of the spatially
averaged velocity and velocity excursion in the two
groups. Many factors contribute to the time-course of
change of arterial blood velocity, including the pattern of
left ventricular ejection and the characteristics of pulse
wave propagation and reflection. A higher resting heart
rate, an unexplained finding in our hypertensive group,
would not be expected to influence the velocity or
acceleration of upstream blood flow [lo, 111, and is
probably not a primary determinant of the propagation of
the peripheral pulses. Due to the use of electrocardiogram-triggered ensemble-averaging, the measurements
themselves are not heart rate dependent. The wave speed
is generally a function of pressure and is known to be
elevated in patients with essential hypertension [ 121.
Increased wave speed or an altered relationship between
wave speed and pressure may cause steepening of the
pressure pulse, and may in turn have caused the rather
'spiky' velocity waveforms that we observed in the
patients with hypertension. We observed a similar waveform in normal subjects after cigarette smoking, which
also caused an increase in arterial blood pressure and
434
C. J. H..Jones et al.
pulse wave velocity [13]. In the present study we
observed a negative correlation between diastolic blood
pressure and mean blood velocity in hypertensive
patients. O n e mechanism which may in part account for
this is increased downstream impedance caused by
structural and functional changes in resistance arteries
[ 141. Furthermore, despite the elevated blood pressure,
superficial femoral arterial diameter was decreased in the
hypertensive patients. T h er e have been previous
measurements of arterial diameter in hypertension made
with ultrasound. Significant increases are reported in
aortic [15] and brachial artery [16, 171 diameter, but
common carotid diameter is consistently found t o be
unchanged [16, 181. Th es e findings indicate that there is
regional variation in the structural response of the arterial
wall in hypertension [17]. As far as we know, no multichannel Doppler measurements of femoral artery
diameter have been reported previously. We believe that
our finding is consistent with wall thickening by medial
hypertrophy, a previously proposed mechanism of autoregulation of wall stress in hypertension [6].
The shear stress at the arterial wall, and therefore the
pattern of arterial flow, exerts an important influence on
arterial physiology. Recent studies have suggested that the
endothelium may act as a transducer for both pressure
and flow, sensing both the stretch and the shear
stress at the wall. Alterations in the pattern of shear stress
are known t o influence numerous properties of the endothelial cells including their shape and orientation, stress
fibre content, membrane potential, turnover rate, solute
uptake and metabolism and endothelium-derived relaxing
factor release [21-231. Many of these properties a re
affected particularly by the rate of change of shear stress.
T h e observed differences in the time course of blood flow
between the normal subjects and the patients with hypertension are also likely t o have been associated with alteration of mixing and the residence time of blood flowing
near the vessel wall [24].
Local haemodynamic factors appear t o play a n important role in the development of atherosclerosis. T h e
disease exhibits a predilection for outer walls at sites of
branching and the inner wall at sites of curvature. The
mechanisms which account for this distribution remain
unclear, but the sites at which the lesions occur at
branches and curves coincide with sites where the wall
shear is on average low and undergoes large changes of
direction during the cardiac cycle [19,20].
In summary, the pattern of blood flow is found t o be
abnormal in the femoral artery of patients with untreated
essential hypertension. Acceleration time, the time to
peak reverse flow and the duration of reverse flow are all
lower than in normal subjects and femoral artery diameter
is diminished. These differences, although small, would
still result in changes in the shear stress at the wall and in
blood-vessel wall interactions in the arteries, including the
regions of complex geometry where atherosclerotic
lesions occur. The present study raises the possibility that
changes in the pattern of arterial flow in patients with
essential hypertension may thus contribute t o the atherosclerotic process.
ACKNOWLEDGMENTS
We thank Nirmala Markandu, SRN, for her help throughout the study. We gratefully acknowledge the support of
the following: the M.R.C., the National Heart Research
Fund, the National Kidney Research Fund, Bayer U.K.,
Schwarz Pharma and A B Hassle. C.J.H.J. and D.R.J.S.
a re supported by the British Heart Foundation.
REFERENCES
1. Lansman, J.B. Going with the flow. Nature (London) 1988;
331,481-2.
2. Caro, C.G. & Lever, M.J. Factors influencing arterial wall
transport. Biorheology 1984; 21, 197-205.
3. Svindland, A.D. & Walloe, L. Localisation of early atherosclerotic lesions in carotid and coronary bifurcations in
humans - a bifurcation of the high shear stress hypothesis.
In: Schettler, G., Nerem, R.M., Schmid-Schonbeim, H.,
Mod, H. & Diehm, C., eds. Fluid dynamics as a localising
factor for atherosclerosis. Berlin: Springer-Verlag, 1983;
212.
4. The Pooling Project Research Group. Relationship of blood
pressure, serum cholesterol, smoking habit, relative weight
and ECG abnormalities to incidence of major coronary
events. Final reDort of the Pooling Proiect. J. Chronic Dis.
1978; 31,201-jO6.
5. Mahler. F.. Brunner. H.H.. Bollineer. A.. Castv. M. &
Anliker, M: Changes ’in phasic femo;al’artery flow ‘induced
by various stimuli: a study with percutaneous pulsed
Doppler ultrasound. Cardiovasc. Res. 1977; 11,454-60.
6. Folkow, B. Structure and function of the arteries in hypertension. Am. Heart J. 1987; 114,938-48.
7. Fish, P.J. Multichannel direction-resolving Doppler
angiography. In: Proceedings of the Second European
Congress on Ultrasonics in Medicine. Excerpta Medica:
Amsterdam, Oxford, 1975: 153-9.
8. Caro, C.G., Jones, C.J.H., MacGregor, G.A., Singer, D.R.J.
& Watkins, N. Arterial blood velocity waveforms in
essential hypertension in man. J. Physiol. (London) 1989;
413-2151?
9. George, C.F., Lewis, P.J. & Petrie, A. Clinical experience
with the use of the ultrasound sphygmomanometer. Br.
Heart J. 1975; 37,804-7.
10. Wallmeyer, K., Wann, S.L., Sagar, K.B., Kalbfleisch, J. &
Kopfenstein, H.S. The influence of preload and heart rate
on Doppler echocardiographic indexes of left ventricular
performance: comparison with invasive indexes in an
experimental preparation. Circulation 1986; 74, 181-6.
11. Harrison, M.R., Smith, M.D., Nissen, S.E., Grayburn, P.A. &
DeMaria, A.N. Use of Doppler echocardiography to
evaluate cardiac drugs: effects of propranolol and verapamil
on aortic blood flow velocity and acceleration. J. Am. Coll.
Cardiol. 1988; 11,1002-9.
12. Safar, M.E., Bouthier, J.A., Levenson, J.A. & Simon. A.
Peripheral large arteries and the response to antihypertensive treatment. Hypertension 1983; 5 (Suppl. III), III63-8.
13. Caro, C.G., Lever, M.J., Parker, K.H. & Fish, P.J. Effect of
cigarette smoking on the pattern of arterial blood flow:
possible insight into the mechanisms underlying the
development of atherosclerosis. Lancet 1987; ii, 11-3.
14. Aalkjaer, C., Heagerty, A.M., Petersen, K.K., Swales, J.D. &
Mulvaney, M.J. Evidence for increased medial thickness,
increased neuronal amine uptake and depressed excitation-contraction coupling in isolated resistance vessels
from essential hypertensives. Circ. Res. 1987; 61, 181-6.
15. Hugue, Ch.J., Safar, M.E., Aliefierakis, Ch., Asmar, R.G. &
London, G.M. The ratio between ankle and brachial systolic
pressure in patients with sustained uncomplicated essential
hypertension. Clin. Sci. 1988; 74, 179-82.
Arterial flow in hypertension
16. Laurent, S., Lacolley, P., London, G. & Safar, M. Hemodynamics of the carotid artery after vasodilation in essential
hypertension. Hypertension 1988; 11, 134-40.
17. Safar, M.E., Peronneau, P.A., Levenson, J.A., TotoMoukouo, J.A. & Simon, A.Ch. Pulsed Doppler: diameter,
blood flow velocity and volumic flow of the brachial artery
in sustained essential hypertension. Circulation 1981; 63,
393-400.
18. van Merode, T., Hick, P.J.J., Hoeks, A.P.G., Rahn, K.H. &
Reneman, R.S. Vessel-wall properties of the carotid artery
in normotensive and borderline hypertensive young male
volunteers. J. Hypertens. 1987; 5 (Suppl. 5 ) ,S471-3.
19. Friedman, M.H. & Deters, O.J. Correlation among shear
rate measures in vascular flows. J. Biomechan. Eng. 1987;
109,254.
20. Ku, D.N., Giddens, D.P., Zarins, C.K. & Glagov, S. Pulsatile
’
21.
22.
23.
24.
435
flow and atherosclerosis in the human carotid bifurcation.
Positive correlation between plaque location and low and
oscillating shear stress. Arteriosclerosis 1985; 5,293-302.
Dewey, C.F. Effects of fluid flow on living vascular cells. J.
Biomechan. Eng. 1984; 106,31-5.
Yoshida, Y., Caro, C.G., Glagov, S. & Nerem, R.M., eds.
The role of blood flow in atherogenesis. Tokyo: SpringerVerlag, 1988: 153-62.
Griffith. T.M., Lewis, M.J.. Newbv, A.C. & Henderson. A.H.
Endothelium-derived relaxing factor. J. Coll. Cardiol. 1988;
12,797-806.
Jorgensen, L., Packham, M.A., Rowsell, H.C. & Mustard,
J.F. Deposition of formed elements of blood on the intima
and signs of intimal injury in the aorta of rabbit, pig and
man. Lab. Invest. 1972; 27,341-50.