Download Blood Pressure Variability in Man: Its Relation to

Clinical Science (1980) 59,401~404s
401s
Blood pressure variability in man: its relation to high blood
pressure, age and baroreflex sensitivity
G . M A N C I A , A. F E R R A R I , L . G R E G O R I N I , G . P A R A T I , G . P O M I D O S S I ,
G . B E R T I N I E R I , G. G R A S S 1 A N D A . Z A N C H E T T I
Istituto di Ricerche Cardiovascolari and Istituto di Patologia Medica I , University of Milano, Progetto Finalizzato di
Medicina Preventiva del CNR, Italy
Summary
1. Intra-arterial blood pressure and heart rate
were recorded for 24 h in ambulant hospitalized
patients of variable age who had normal blood
pressure or essential hypertension. Mean 24 h
values, standard deviations and variation coefficient were obtained as the averages of values
separately analysed for 48 consecutive half-hour
periods.
2. In older subjects standard deviation and
variation coefficient for mean arterial pressure
were greater than in younger subjects with similar
pressure values, whereas standard deviation and
variation coefficient for heart rate were smaller.
3. In hypertensive subjects standard deviation
for mean arterial pressure was greater than in
normotensive subjects of similar ages, but this
was not the case for variation coefficient, which
was slightly smaller in the former than in the
latter group. Normotensive and hypertensive
subjects showed no difference in standard
deviation and variation coefficient for heart rate.
4. In both normotensive and hypertensive
subjects standard deviation and even more so
variation coefficient were slightly or not related
to arterial baroreflex sensitivity as measured by
various methods (phenylephrine, neck suction
etc.).
5. It is concluded that blood pressure
variability increases and heart rate variability
decreases with age, but that changes in variability
are not so obvious in hypertension. Also, differences in variability among subjects are only
marginally explained by differences in baroreflex
function.
Correspondence: Professor Giuseppe Mancia,
Istituto di Ricerche Cardiovascolari, Policlinico, Via
F. Sforza 35,20122 Milano, Italy.
Key words: age, baroreflexes, blood pressure
variability, essential hypertension, heart rate,
neck chamber, phenylephrine, trinitroglycerine.
Introduction
In the past years several studies have shown that
in man blood pressure has a marked degree of
variability which depends on the occurrence of
well-recognized behavioural conditions (sleep,
exercise, emotions etc.) (Bevan Honour & Stott,
1969; Littler, West, Honour & Sleight, 1978;
Watson, Stallard & Littler, 1979) and possibly on
other factors (Littler, Honour, Sleight & Stott,
1972; Millar-Craig, Bishop & Raftery, 1978).
This variability is an important phenomenon to
be defined both for discovering the multifold
influences normally involved in human blood
pressure regulation and for obtaining more
precise information on the existence and the
severity of a hypertensive state in a given patient
(Sokolow, Wertegard, Kain & Hinman, 1966;
Mancia & Zanchetti, 1980). We have asked
ourselves three questions concerning blood pressure variability in man: (1) is blood pressure
variability different in hypertensive subjects from
that in subjects with normal blood pressure?; (2)
does this variability change in relation to age?;
(3) are differences in variability substantially
explicable on the basis of different baroreflex
sensitivities?
Methods
We studied ambulant subjects of both sexes who
had either normal blood pressure or essential
hypertension. No subject had major diseases
(except the hypertension) and none had been
given cardiovascular drugs during the preceding
G. Mancia et ul.
402s
phenylephrine and the lengthening of the R-R
interval, as well as by the fall in systolic pressure
induced by an intravenous bolus of trinitroglycerine and the shortening of the R-R interval:
the slopes of the regressions were taken as the
measure of the baroreflex sensitivity (Pickering,
Gribbin & Sleight, 1972). In 35 subjects carotid
transmural pressure was increased and decreased
by a neck-chamber device (Ludbrook, Mancia,
Ferrari & Zanchetti, 1977) to observe the
magnitude of the resulting reflex decrease and
increase in mean arterial pressure: in this case the
baroreflex sensitivity was given by the ratio
betvieen the reflex blood pressure effect and the
magnituc‘ of the disturbance applied to the
carotid bhroreceptors, which could be evaluated
as the changes in tissue pressure outside the
carotid sinus walls (Mancia, Ludbrook, Ferrari,
Gregorini & Zanchetti, 1978). All subjects gave
free consent to each part of the study after having
had its nature and purpose explained.
3 weeks. The study was performed while the
subjects were hospitalized, in an attempt to
standardize their living and environmental
conditions.
Arterial blood pressure was measured continuously with a catheter placed in a radial artery
and an Oxford portable transducer-tape recorder
apparatus. In each subject the measurement was
started in the early evening and was terminated
24 h later. The blood pressure tracing was first
visually scrutinized and, after determination of
the good quality of the pulse wave (this was the
case in all subjects considered in this study),
analysed by a computer.
The analysis was carried out in a manner
described in a previous report (Cioffi & Di
Rienzo, 1980). Briefly, mean arterial pressure
was calculated by averaging 60 values (one every
50 ms) within consecutive periods of 3 s, systolic
pressure, diastolic pressure and heart rate being
similarly calculated as the values occurring within
these 3 s time intervals. Averages and standard
deviations were calculated for each of the 48
half-hours in which the recording period was
divided. The averages of the 48 different averages
provided mean values for the whole 24 h period,
and the averages of the 48 different standard
deviations provided the measures of the
variability occurring within half-hour time spans.
The standard deviations were also calculated as
percentages of the average 24 h values to obtain
not only the variability in absolute values but also
the variation coefficient, regardless of the baseline
differences in the various subjects.
Arterial baroreceptor reflexes were tested by
two methods. In 38 subjects, linear regressions
were calculated between the rise in systolic
pressure induced by an intravenous bolus of
Results
The results obtained in two groups of subjects
with normal and high blood pressure (their age
being similar) are shown in Table 1. It can be
seen that blood pressure variability expressed in
absolute values was greater in the hypertensive
than in the normotensive group but that this trend
was reversed when variability was expressed as
variation coefficient, which was not significantly
greater in the hypertensive than in the normotensive group. It can also be seen that these
results were not paralleled by the heart rate data.
As for blood pressure, 24 h mean heart rate
values were greater in the hypertensive than in the
normotensive subjects. Heart rate variability,
TABLE1. Variability of mean arterial pressure and heart rate in subjects of different ages and with normal or high blood
pressure
Results are expressed as means k SE. Statistical analysis was performed by the Student’s t-test. Subjects belonging to the
younger and older groups were selected among those having an age respectively equal to or less than 38 years and equal to
or greater than 48 years. Standard deviation (for half-hour periods) in absolute values; variation coefficient: standard
deviation as percentage o f 24 h mean arterial pressure or heart rate value.
Age
(years)
Arterial pressure
24 h
arterial
SD
(mrnHg)
pressure
(rnrnHg’i
Heart rate
Variation
coefficient
(9;)
24 h heart
rate
(beatdmin)
SD
(beatshin)
(9;)
~~
Normotensivegroup(n=22)
Hypertensivegroup(n=41)
P
Younger group (n = 31)
Older group (n = 26)
P
38.6 C 2.9
44.8, 1.6
<0.05
30.2 t 1 . 1
54.6 i 1.3
<0.001
88.1 C 1.7
137.953.1
<0.001
109.4 ?r 3.6
116.7 i 3.9
N.S.
6 5 f 0.3
9.2f0.4
co.001
7.6 f 0 . 3
9 . 2 i 0.4
<0.002
7.4 k 0.3
6.7C0.3
N.S.
7.1 i 0.3
8.0 i 0.3
<0.02
7 5 . 0 ?r 1.8
7 9 . 5 i 1.5
<0.01
78.0 i 1.6
76.7 i 2 . 1
N.S.
Variation
coefficient
~~~
8 . 9 ? 0.9
8.OiO.6
N.S.
9 . 8 + 0.7
6.8 i 0.4
10.002
~
~~~
11.9 k 0 . 6
10.0+_0.6
N.S.
11.7 f 0.6
8.9 + 0.6
t0.002
Human blood pressure variability
however, was not significantly different in the two
groups when expressed either in absolute values
or as variation coefficients.
Table 1 also shows the results obtained when
two groups of subjects with similar 24 h blood
pressures but with different ages were analysed.
Blood pressure variability was significantly
greater in the elderly subjects when expressed
either in absolute or in percentage values. On the
other hand, heart rate variability showed an
opposite alteration, as both its indices were
significantly less pronounced in the older than in
the younger group.
Attempts to correlate blood pressure
variability and baroreflex sensitivity gave only
partially positive results. No correlation was
found between variability indices and baroreflex
sensitivity obtained by trinitroglycerine and
positive neck pressure application. There was a
significant inverse linear relationship between
blood pressure variability expressed in absolute
value and baroreflex sensitivity obtained by
phenylephrine (r = 0-49, P < OaOl), but such
relationship was completely lost when the variation coefficient was used instead. There was no
correlation between absolute values of blood
pressure variability and the baroreflex sensitivity
obtained by the negative neck pressure application, although in this instance use of the
variation coefficient made an inverse correlation
of borderline significance to appear ( I = 0.32,
P < 0.05).
403 s
increase (this time both in absolute and percentage values) with age compared with heart rate
variability, which shows a simultaneous clear-cut
reduction. An increased blood pressure variability with age may be explained by the fact
that in older people any given change in stroke
volume is likely to induce a greater blood
pressure change owing to their less compliant
arterial walls. A reduced heart rate variability
under the same circumstances may reflect a
reduced responsiveness of the heart to various
stimuli.
The last comment refers to the inverse relationship between blood pressure variability and
arterial baroreflexes. Our data indicate that such
a relationship, whenever it can be found, is of a
weak order, and probably accounts for only a
small portion of the differences in spontaneous
pressure variability that occur in different subjects. Clearly other factors seem to be largely
involved in this phenomenon, those originating in
the brain centres and in the vascular smooth
muscle being the most likely ones.
Acknowledgment
The authors are deeply grateful to Professor A.
Pedotti and to Dr P. Cioffi and Dr M. Di Rienzo,
Centro di Bioingegneria Fondazione Don
Gnocchi, for developing and performing the
computer analysis of the data.
Discussion
Our study confirms that blood pressure
variability is greater in subjects with essential
hypertension than in normotensive subjects
(Goldberg, Raftery, Cashman & Stott, 1978) but
it also makes clear that this phenomenon is
limited to the variability expressed in absolute
values and is actually reversed when such value is
transformed into a variation coefficient (that is,
when it is expressed as percentage of the existing
blood pressure). Which of the two values better
represents the biological phenomenon of
variability is debatable. Nevertheless, it seems to
us important to underline that although absolute
blood pressure values do oscillate more in
hypertensive patients, their percentage oscillations
around the mean are in fact similar to the
percentage oscillations that can be observed in
normotensive individuals.
A second point that emerges clearly from our
study is that blood pressure variability shows an
References
BEVAN,A.T., HONOUR,A.J. & STOTT,F.H. (1969) Direct arterial
pressure recordings in unrestricted man. Clinical Science, 36,
329-344.
CIOFFI,P. & DI RIENZO,M. (1980) Monitoraggio continuo della
pressione arteriosa nelle 24 ore. Graduation Thesis, Politecnico di
Milano, Facolta di Ingegneria.
GOLDBERG,A.D., RAFTERY, E.B., CASHMAN,P.M.M. & STOTT,
F.D. (1978) Study of untreated hypertensive subjects by means of
continuous intra-arterial blood pressure recordings. British Hearr
Journal, 40,656-664.
HONOUR,A.J., SLEIGHT,P. & STOTT, F.D. (1972)
L I ~ E RW.A.,
,
Continuous recording of direct arterial pressure and electrocardiogram in unrestricted man. British Medical Journal, iii,
76-78.
WEST, M.J., HONOUR,A.J. & SLEIGHT,P. (1978)
L I ~ E RW.A.,
,
The variability of arterial pressure. American Heart Journal, 95,
180-186.
A. (1977)
LUDBROOK,
J., MANCIA,G., FERRARI,A. & ZANCHET~,
The variable-pressure neck chamber method for studying the
carotid baroreflex in ‘man. Clinical Science and Molecular
Medicine, 53, 165-171.
MANCIA, G., LUDBROOK,J., FERRARI, A., GREGORINI, L. &
ZANCHETTI,A. (1978) Baroreceptor reflex in human hypertension. Circulation Research, 43, 170-177.
MANCIA,G. & ZANCHETTI,A. (1980) Continuous arterial blood
pressure recording in human hypertension: a methodological
404s
G. Mancia et al.
approach. In: Afherosclerosis Reuiews, vol. 7, pp. 247-254. Ed.
Hegyeli, R. Raven Press, New York.
MILLAR-CRAIG,M.W., BISHOP, C.N. & RAFTERY.E.B. (1978)
Circadian variation of blood pressure. Lancer, i, 795-797.
PICKERING,
T.G., GRIBBIN,
B. & SLEIGHT,P. (1972) Comparison of
the reflex heart rate response to rising and falling arterial pressure
in man. Cardiovascular Research, 6,277-283.
SOKOLOW,
M., WERTEGARD,
D., KAIN, H.K. & HINMAN,A.T.
(1966) Relationship between level of blood pressure measured
casually and by portable recorders and severity of complications
in essential hypertension. Circulation, 34,279-298.
WATSON. R.D.S.. STALLARD,T.J. & LIITLER. W.A. (1979)
Factors determining the variability of arterial pressure in
hypertension. Clinical Science, 57 (Suppl. 5 ) , 283s-285s.