Empirical estimates of Mean Aortic Pressure: Advantages, drawbacks and implications for pressure redundancy

Abstract

Mean arterial pressure (MAP) is estimated at the brachial artery level by adding a fraction of pulse pressure (form factor; =0.33) to diastolic pressure. We tested the hypothesis that a fixed form factor can also be used at the aortic root level. We recorded systolic aortic pressure (SAP) and diastolic aortic pressure (DAP), and we calculated aortic pulse pressure (PP) and the time-averaged MAP in the aorta of resting adults (n=73; age 43+/-14 years). Wave reflection was quantified using the augmentation index. The aortic form factor (range 0.35-0.53) decreased with age, MAP, PP and augmentation index (each P<0.001). The mean form factor value (0.45) gave a reasonable estimation of MAP (MAP=DAP+0.45PP; bias=0+/-2 mmHg), and the bias increased with MAP (P<0.001). An alternative formula (MAP=DAP+PP/3+5 mmHg) gave a more precise estimation (bias=0+/-1 mmHg), and the bias was not related to MAP. This latter formula was consistent with the previously reported mean pulse wave amplification of 15 mmHg, and with unchanged MAP and diastolic pressure from aorta to periphery. Multiple linear regression showed that 99% of the variability of MAP was explained by the combined influence of DAP and SAP, thus confirming major pressure redundancy. Results were obtained irrespective of whether the marked differences in heart period and extent of wave reflection between subjects were taken into account. In conclusion, the aortic form factor was strongly influenced by age, aortic pressure and wave reflection. An empirical formula (MAP=DAP+PP/3+5 mmHg) that is consistent with mechanical principles in the arterial system gave a more precise estimate of MAP in the aorta of resting humans. Only two distinct pressure-powered functions were carried out in the (SAP, DAP, MAP, PP) four-pressure set.

Full text

7Clinical Science (2002) 103, 7–13 (Printed in Great Britain)
Empirical estimates of mean aortic pressure:
advantages, drawbacks and implications
for pressure redundancy
Denis CHEMLA*, Jean-Louis HE
;
BERT*, Eduardo APTECAR†, Jean-Xavier MAZOIT*,
Karen ZAMANI*, Robert FRANK‡, Guy FONTAINE‡, Alain NITENBERG§
and Yves LECARPENTIER*
*Services de Physiologie Cardio-Respiratoire et d’Anesthe
!sie, CHU de Bice
#tre, Universite
!Paris XI, Assistance
Publique – Ho
#pitaux de Paris, UMR 7639 CNRS-Loa-Ensta-Ecole Polytechnique, 94 275 Le Kremlin-Bice
#tre, France, †Service de
Chirurgie Thoracique et Cardiovasculaire, CHU Henri Mondor, 94 010 Cre
!teil, France, ‡Service de Cardiologie, Ho
#pital Jean-
Rostand, 94 200 Ivry-sur-Seine, France, and §Service de Physiologie et d’Explorations Fonctionnelles & INSERM U251, CHU
Xavier Bichat, 75018 Paris, France
ABSTRACT
Mean arterial pressure (MAP) is estimated at the brachial artery level by adding a fraction of
pulse pressure (form factor; l0.33) to diastolic pressure. We tested the hypothesis that a fixed
form factor can also be used at the aortic root level. We recorded systolic aortic pressure (SAP)
and diastolic aortic pressure (DAP), and we calculated aortic pulse pressure (PP) and the time-
averaged MAP in the aorta of resting adults (nl73; age 43p14 years). Wave reflection was
quantified using the augmentation index. The aortic form factor (range 0.35–0.53) decreased
with age, MAP, PP and augmentation index (each P0.001). The mean form factor value (0.45)
gave a reasonable estimation of MAP (MAP lDAPj0.45PP ; bias l0p2 mmHg), and the bias
increased with MAP (P0.001). An alternative formula (MAP lDAPjPP/3j5 mmHg) gave a
more precise estimation (bias l0p1 mmHg), and the bias was not related to MAP. This latter
formula was consistent with the previously reported mean pulse wave amplification of 15 mmHg,
and with unchanged MAP and diastolic pressure from aorta to periphery. Multiple linear
regression showed that 99%of the variability of MAP was explained by the combined influence
of DAP and SAP, thus confirming major pressure redundancy. Results were obtained irrespective
of whether the marked differences in heart period and extent of wave reflection between
subjects were taken into account. In conclusion, the aortic form factor was strongly influenced
by age, aortic pressure and wave reflection. An empirical formula (MAP lDAPjPP/3
j5 mmHg) that is consistent with mechanical principles in the arterial system gave a
more precise estimate of MAP in the aorta of resting humans. Only two distinct pressure-
powered functions were carried out in the (SAP, DAP, MAP, PP) four-pressure set.
INTRODUCTION
Mean arterial pressure (MAP) is considered as the
perfusion pressure through each tissue bed, and accounts
for more than 80% of the total hydraulic load placed on
Key words: aorta, arterial pressure, pulse pressure, wave reflection.
Abbreviations : DAP, diastolic aortic pressure; MAP, mean arterial pressure ; PP, aortic pulse pressure; SAP, systolic aortic pressure.
Correspondence: Dr Denis Chemla (e-mail denis.chemla!bct.ap-hop-paris.fr).
the left ventricle [1,2]. Although true MAP is the area
under the blood pressure curve divided by cardiac cycle
length, clinical and epidemiological studies currently
approximate brachial MAP by using an empirical formula
where a fraction of pulse pressure (form factor; l0.33)
#2002 The Biochemical Society and the Medical Research Society

8 D. Chemla and others
is added to the diastolic pressure [3,4]. This empirical
formula implies that MAP is twice as sensitive to
diastolic as it is to systolic pressure (MAP l2\3
idiastolic pressurej1\3isystolic pressure). Increased
MAP is an important component of vascular overload
and cardiovascular risk, and the redundant relationship
between systolic, diastolic and mean pressure may have
implications for cardiovascular risk stratification [5–7].
The main drawbacks of this formula arise from unpre-
dictable changes in form factor related to age and
vasomotor tone [5,8,9].
According to this empirical formula, it is generally
assumed that, at a given MAP, increased pulse pressure
reflects an increased systolic pressure (and thus left
ventricular systolic load) and a decreased diastolic press-
ure, which may compromise coronary perfusion [10,11].
However, central and not brachial blood pressure reflects
the true workload placed on the left ventricle and
determines myocardial oxygen consumption and cor-
onary perfusion pressure. The MAP and diastolic arterial
pressure show little difference between central and
peripheral arteries, whereas systolic pressure increases
from aorta to periphery. This pulse wave amplification
depends on the characteristics of pressure wave trans-
mission and reflection, and is mainly influenced by
arterial compliance, body length, heart rate, MAP, age
and sex [10–14]. The form factor value, which relates to
pulse contour characteristics, is approx. 0.50 in central
arteries (sine-wave pattern), thus implying that MAP is
roughly as sensitive to diastolic aortic pressure (DAP) as
it is to systolic aortic pressure (SAP) [4]. However, the
precise relationship between the steady and pulsatile
components of blood pressure remains to be documented
at the aortic root level.
In this preliminary prospective study, we tested the
hypothesis that a single form factor value can be used to
reliably estimate MAP at the aortic root level. We also
present the advantages and drawbacks of MAP empirical
formulas, together with their pathophysiological im-
plications in terms of left ventricular load, coronary
perfusion pressure, peripheral arterial haemodynamics
and pressure redundancy.
METHODS
Patients
This prospective study included patients (nl73; 65
males) with symptoms of chest pain or other cardio-
vascular symptoms who were referred to the catheteriz-
ation laboratory for diagnostic or routine right and left
heart catheterization. Patients with end-stage heart fail-
ure, rhythm disturbances or aortic or mitral valve
insufficiency were excluded from the study. The final
diagnosis was as follows: subjects with normal cardiac
function and coronary angiograms, nl11; subjects with
miscellaneous cardiac diseases (mainly idiopathic dilated
cardiomyopathy, coronary artery disease, right ventricu-
lar disease, grafted heart and hypertrophic cardio-
myopathy), nl62. In patients receiving vasoactive
drugs, treatment was discontinued 24 h before the inves-
tigation. At the time of the study, 18 patients (25%) had
a SAP 140 mmHg at baseline. Four patients (5%)
suffered from diabetes mellitus. All patients gave in-
formed consent, and the ethical committee of our
institution approved the protocol.
Catheterization technique and recordings
Patients were studied according to our routine protocol
[15]. All patients were in the fasting state for at least 12 h
before the investigation. No premedication was admin-
istered. Lidocaine (1%) was used for local anaesthesia,
and 5000 units of heparin was administered intra-
venously. The percutaneous femoral approach was used.
The left heart pigtail catheter was an 8F single lumen
catheter with a lateral high-fidelity transducer (Cordis\
Sentron, Roden, The Netherlands) [16]. The catheter was
advanced from the femoral artery to the aortic root. After
a 5-min equilibrium period, pressure data were recorded
at base over a 15 s period. The data were computed on a
Toshiba 3200 SX with customized software (sampling
rate 500 Hz). Cardiac output was determined by the
thermodilution technique (Cardiac Output Computer
model 9520A; Edwards Laboratories).
SAP and DAP were measured automatically, and aortic
pulse pressure (PP) was calculated as PP lSAPkDAP.
Pressure wave reflection was evaluated by calculating the
pressure augmentation index [14]. In 65 out of 73 patients
(89%), a well-defined systolic pressure inflection point
(Pi) divided the aortic pressure wave form into early and
late systolic phases, and we calculated the augmentation
index as follows:
=P\PP l(SAPkPi)\PP
Subjects were divided into three groups according to the
classification proposed previously by Murgo et al. [14]:
type A (nl51), =P\PP 0.12; type B (nl11),
0=P\PP 0.12; type C (nl3), =P\PP 0. Thus,
according to this classification, 78% subjects were type
A, 17% were type B and 5% were type C. Given that
81% of patients were older than 30 years, this finding is
consistent with earlier studies [2,14]. The systolic in-
flection point could not be clearly defined in eight
subjects.
Reference MAP value
The time-averaged MAP in the aorta was taken as the
reference value. The MAP was calculated as the total area
under the pressure curve divided by the heart period.
#2002 The Biochemical Society and the Medical Research Society

9Estimation of mean aortic pressure
Estimates of MAP
MAP lDAPj(form factoriPP)
In each patient, the form factor was calculated as follows:
MAP lDAPj(form factoriPP)
Form factor l(MAPkDAP)\PP
The mean form factor value was calculated in the overall
population, so as to propose an improved, new empirical
estimation of MAP.
MAP lDAPjPP/3j5 mmHg
This formula has been proposed more recently at the
aortic root level [17].
Statistical analysis
Results are expressed as meanspS.D. Pressures and time
parameters were averaged over 10 consecutive cardiac
cycles. We tested the predictive performance of the
empirical formulae against the reference MAP value,
using single linear regression. In addition to the r#
statistics, the predictive performances of all regression
analyses were assessed by the bias, calculated as the mean
difference (pS.D.) between predicted and reference
MAP values. We tested the hypothesis that MAP was
redundant when SAP and DAP values were given. To
this end, a multiple linear regression of SAP and DAP
was applied to the whole data set. Comparisons between
groups were performed by using ANOVA. Linear
regression was studied using the least-squares method. A
Pvalue of 0.05 was considered statistically significant.
RESULTS
The characteristics of the study population are given in
Table 1. Significant linear relationships were found
between the four aortic pressures under study (SAP,
DAP, PP, MAP), and the correlation coefficients are
Table 1 Characteristics of the study population (nl73)
Parameter MeanpS.D. Range
Age (years) 43p14 19–77
Height (cm) 171p7 146–192
Body surface area (m2) 1.84p0.21 1.30–2.47
SAP (mmHg) 127p23 83–181
DAP (mmHg) 78p12 50–109
MAP (mmHg) 100p15 66–134
PP (mmHg) 49p15 27–89
Heart period (ms) 838p179 506–1267
Stroke index (ml/m2)39p14 15–87
Form factor 0.45p0.04 0.35–0.53
Pressure augmentation index (
n
l65) 0.28p0.16 0–0.53
Table 2 Correlation matrix (nl73)
All relationships are
P
0.0001 except *
P
l0.0003.
r
SAP DAP MAP PP
SAP 0.794 0.937 0.883
DAP 0.945 0.415*
MAP 0.668
summarized in Table 2. The augmentation index was
positively related to age (rl0.52), MAP (rl0.56) and
PP (rl0.72) (each P0.001) and was negatively related
to height (rlk0.39, P0.01), similar to previous
findings [8,18,19]. Age and PP were also positively related
(rl0.39, P0.01), but there was no relationship
between age and MAP (rl0.18).
MAP lDAPj(form factoriPP)
On average, the fraction of PP that must be added to
DAP so as to obtain the time-averaged MAP (i.e. the
form factor) was 0.45. The form factor (range 0.35–0.53)
decreased linearly with age (rlk0.69), MAP (rl
k0.41), PP (rlk0.73) and =P\PP (rlk0.83) (each
P0.001) (Figure 1). The form factor increased with
height (rl0.37, P0.01), and was unrelated to body
weight, surface area, heart rate and cardiac index.
The form factor was higher in patients with SAP 
140 mmHg (0.46p0.04) than in the remaining subjects
(0.42p0.04; P0.001). The form factor was similar in
controls (0.46p0.03) and in patients with miscellaneous
cardiac diseases (0.45p0.04; Pl0.28). By applying this
single form factor value of 0.45 to the overall population,
we obtain:
MAP lDAPj0.45PP
This formula gave a precise estimation of MAP
(100p16 mmHg ; mean bias l0p2 mmHg). The bias
ranged from k3toj8 mmHg and increased with MAP
(rl0.40, P0.001) (Figure 2, upper panel), age (rl
0.67, P0.001) and =P\PP (rl0.76, P0.001). The
bias was not influenced by heart rate (rlk0.24).
MAP lDAPjPP/3j5 mmHg
This formula gave a precise estimation of MAP
(100p15 mmHg ; bias l0p1 mmHg). The bias ranged
from k3toj3 mmHg. The bias was not influenced by
MAP (rlk0.19; Figure 2, lower panel) or heart rate
(rlk0.14). The bias increased with age (rl0.61,
P0.001) and =P\PP (rl0.30, P0.01). This formula
can be rewritten as follows:
MAP l0.67DAPj0.33SAPj5 mmHg
Pressure redundancy
Multiple linear regression indicated that 99% of the
variability of MAP was explained by the combined
#2002 The Biochemical Society and the Medical Research Society

10 D. Chemla and others
Figure 1 Relationships between form factor and age (nl73), PP (nl73), pressure augmentation index (nl65) and
height (nl73)
All relationships are
P
0.001, except for that with height (
P
0.01).
influence of DAP and SAP (multiple r#l0.9914). The
equation was as follows:
MAP l0.70DAPj0.32SAPj4 mmHg
This equation is very close to the formula given above
and gave a precise estimation of MAP (100p15 mmHg;
bias l0p1 mmHg). The bias ranged from k3to
j3 mmHg and was not influenced by MAP (rlk0.13,
Pl0.27). Assuming that the pressure bias is small
enough to be negligible, this implies that two subjects
with the same DAP and SAP had the same MAP,
irrespective of their pressure waveform and heart rate
(Figure 3). The addition of age, =P\PP or both to
the multiple regression increased the multiple r#value
slightly (0.9949, 0.9928 and 0.9950 respectively). The
SAP value was predicted accurately when MAP and DAP
values were given [bias l0.7p4.2 mmHg (i.e. 0p3%);
range k9 to 10 mmHg].
DISCUSSION
The fraction of PP that must be added to DAP so as to
obtain the time-averaged MAP was 0.45 at the aortic root
level. Although this gave a reasonably good estimation of
MAP, the form factor was influenced by age and aortic
pressure, and a better estimate was obtained by using an
alternative formula (MAP lDAPjPP\3j5 mmHg).
The unusual accuracy of this empirical formula suggested
that two pressures were enough to decribe the (SAP,
DAP, PP, MAP) four-pressure set. Pressure redundancy
was confirmed by multiple linear regression, showing
that 99% of the variability of MAP was explained by the
combined influence of DAP and SAP.
In the peripheral arteries of adults, MAP is estimated
by adding 0.33i(pulse pressure) to diastolic pressure [3,
4]. The form factor may depend on arterial pressure, age
and arterial location. In newborns, it has been recom-
mended that form factors of 0.40 and 0.50 are used for
the tibial artery and the radial artery respectively [20]. In
adults, the assumed weakness of the empirical estimate
of MAP is illustrated throughout aging, where the form
factor of 0.33 becomes closer to 0.50. This could be
responsible for the levelling off of estimated MAP after
age 50–60 years, thus also explaining why MAP is no
longer a surrogate measurement of vascular resistance [5].
Validation of the form factor value is hampered by the
small calibre of the brachial artery and by the need for
miniaturized high-fidelity pressure recording systems.
Overall, these limitations could explain the difficulty in
#2002 The Biochemical Society and the Medical Research Society

11Estimation of mean aortic pressure
Figure 2 Empirical estimates of MAP
Upper panel: correlation between the true time-averaged MAP and the pressure
bias obtained using a single form factor of 0.45 [bias l(DAPj0.45PP)kMAP]
(
n
l73). Mean bias and 95% confidence interval are indicated. The bias
increased with MAP. Lower panel: correlation between the true time-averaged MAP
and the pressure bias obtained using the alternative empirical formula [bias l
(DAPj0.33PPj5 mmHg)kMAP] (
n
l73). Mean bias and 95% confidence
interval are indicated. The bias was not influenced by MAP.
precisely documenting the role of MAP in cardiovascular
risk. Such a role is, however, intuitive, as MAP accounts
for more than 80% of the total hydraulic load put on the
left ventricle and is considered as the perfusion pressure
through each tissue bed [1,2].
Central aortic pressure reflects the workload put on
the left ventricle, and determines myocardial oxygen
consumption, coronary perfusion pressure and the pre-
vailing pressure for aortic baroreflexes. Furthermore,
arterial pressure at any point of the vascular tree depends
on both central aortic pressure and the characteristics of
pressure wave transmission and reflection. Both the large
aorta diameter and the possibility of using high-fidelity
pressure catheters make it possible to reliably compare
the true MAP and empirical estimates.
Figure 3 Pressure redundancy
A typical example is presented, where two subjects with the same DAP (67 mmHg)
and SAP (105 mmHg) have the same time-averaged MAP (85 mmHg), despite
marked differences in pressure waveform, dicrotic notch pressure and cycle length.
Subject F1, beat no. 1 (dotted line) and subject F20, beat no. 6 (solid line) are
presented. In the overall population, MAP could be estimated from DAP and SAP
only, assuming that the pressure bias (0p1 mmHg) was small enough to be
negligible.
Aortic form factor
At the aortic root level, the form factor value we report
here (0.45) lies between the classical empirical value (0.50)
[4] and experimental values obtained from carotid pulse
tracings (0.43) [3] and ascending aorta recordings (0.41)
[21]. Meaney et al. [21] recently documented the aortic
form factor in cardiac patients (mainly with coronary
artery diseases; 67% male). These authors used fluid-
filled catheters and electronic damping of the signal for
MAP calculation. When compared with their study [21],
our slightly higher form factor value could be explained
by differences in the catheters (we used high-fidelity
pressure catheters), our method for calculating the ref-
erence MAP (time-averaging) and the characteristics of
the study population. We suggest that aortic MAP could
be estimated using the following formula:
MAP lDAPj0.45PP
Although reasonably accurate (bias l0p2 mmHg), this
formula has several drawbacks. First, there is no theo-
retical\physiological background upon which to predict
such a formula. Secondly, the formula gives the false
impression that the form factor value is constant. Con-
versely, we observed that MAP was as sensitive to DAP
as it was to SAP in young patients, and in patients with a
low PP value and a small extent of wave reflection (form
factor approx. 0.50). The relative contribution of DAP
increased in older patients, and in those with a high PP
value and enhanced wave reflections, in whom MAP
tended to be twice as sensitive to DAP as it was to SAP
(form factor approx. 0.33). The pathophysiological im-
plications may be important. The left ventricular systolic
wall stress and coronary perfusion pressure depend
#2002 The Biochemical Society and the Medical Research Society

12 D. Chemla and others
strongly on SAP and DAP respectively [1,2,22]. Our
results may reflect the coupling between arterial load and
the myocardial oxygen supply\demand ratio. For a given
MAP, the form factor decreased from "0.50 to "0.33 in
cases where PP was increased, and this may reflect
beneficial protection of the coronary circulation, with
preserved (instead of decreased) DAP values when
pulsatile stress is increased.
MAP lDAPjPP/3j5 mmHg
This formula has three advantages. First, the accuracy of
this formula was excellent (bias l0p1 mmHg), and this
confirmed previous results [17]. Secondly, there is a
strong theoretical\physiological background upon which
to predict such a formula. This formula is consistent with
mechanical principles in the arterial system, if one
assumes a form factor of 0.33 in peripheral arteries.
Indeed, MAP and diastolic arterial pressure show little
difference between central and peripheral arteries, while
systolic pressure increases from aorta to periphery
[10–14]. Two studies have shown that radial artery pulse
pressure is 15 mmHg higher than aortic PP in the overall
population (mean value) [12,13]. The j5 mmHg extra
factor is in fact foreseeable, as it encompassed one-third
of the pulse wave amplification:
MAP lDAPj(peripheral PP\3)
lDAPj(aortic PPj15)\3
lDAPj(aortic PP\3)j5 mmHg
Thirdly, this formula was very similar to the equation
obtained by multiple linear regression (see the Results
section), and thus can be viewed as a simplified formali-
zation of pressure redundancy, as discussed below.
Implications
The novelty of our result is not that aortic MAP could be
approximated by using DAP and SAP only, but rather
that an empirical formula gave an unusually accurate
estimation of MAP. Pressure redundancy was confirmed
by multiple linear regression, showing that 99% of the
variability of MAP was explained by the combined
influence of SAP and DAP. For a given DAP, MAP
contains virtually no additional information independent
of SAP, assuming that the pressure bias (0p1 mmHg) is
small enough to be negligible. Results were obtained
irrespective of whether the marked differences in wave
reflection and heart period were taken into account.
Because all clinical and epidemiological studies use a
fixed form factor of 0.33 to estimate MAP at rest, pressure
redundancy is commonly admitted (although not yet
demonstrated) for peripheral arteries, and it is likely that
this has clinical implications in terms of risk stratification
[5–7]. Pressure redundancy may appear as an unifying
concept [23,24] for central and peripheral haemo-
dynamics, and may relate to the general physiology of the
human circulation at rest. One explanation could be that
redundancy reflects the overlap of the main haemo-
dynamic variables regulating SAP, DAP, MAP and PP
[22,25]. Efforts must be directed towards the con-
firmation of pressure redundancy at the peripheral
arterial level. If confirmed, this could indicate that two
pressures [i.e. either (SAP, DAP) or (PP, MAP)] are
enough to characterize the four-pressure set.
The practical implications of our results must be
discussed. Our study improved estimation of central
MAP, and thus may help carotid artery pressure cali-
bration using applanation tonometry. Further studies are
needed to confirm this point. Finally, in a population
similar to ours, we suggest that central SAP could be
reasonably estimated from peripheral MAP and diastolic
pressure values (mean bias l0p3 %) if one assumes
unchanged MAP and diastolic pressure from aorta
to periphery. However, further studies are needed to
confirm this point, given that peripheral arterial pressure
was not measured in our study.
Strengths and limitations of the study
To the best of our knowledge, this is the first study to
document form factor values using high-fidelity pressure
catheters. For an invasive study, the number of subjects
was large (nl73) and likely to be sufficient to justify the
conclusions we have drawn from the data. The results
pertain strictly to the population under study. The design
of our study (i.e. prospective and invasive) explains why
we have included control subjects and patients with
various forms of cardiac diseases, ranging in age from 19
to 77 years. Our results are thus strengthened by the fact
that data were obtained from a heterogeneous population
and over a wide range of cardiac function (stroke index
ranging from 15 to 87 ml\m#), heart period (from 506 to
1267 ms) and extent of wave reflection (pressure aug-
mentation index ranging from 0 to 0.53). Since only 5%
of patients were Murgo’s type C [14], our results need to
be confirmed in patients with small or diffuse reflections.
Results were also obtained over a wide range of SAP
(from 83 to 181 mmHg), MAP (from 66 to 134 mmHg)
and PP (from 27 to 89 mmHg) values. Although 25 % of
patients had a SAP 140 mmHg at baseline, further
studies are needed to confirm our results in patients with
hypertension. For ethical reasons, we did not test the
acute effects of vasodilators, and the lowest DAP was
50 mmHg. Thus further studies are needed to document
the form factor value in patients with very low pressures
(e.g. patients in intensive care units). Finally, our results
apply strictly to aortic pressure at rest. They do not apply
to other arterial sites (e.g. brachial artery), nor to dynamic
conditions.
#2002 The Biochemical Society and the Medical Research Society

13Estimation of mean aortic pressure
Conclusions
In conclusion, in the aortic root of resting humans, the
form factor was strongly influenced by age, aortic
pressure and wave reflection. The high accuracy of the
empirical formula (MAP lDAPjPP\3j5 mmHg) was
consistent with the previously reported mean pulse wave
amplification of 15 mmHg, and with unchanged MAP
and diastolic pressure from aorta to periphery. Aortic
pressure redundancy was demonstrated by multiple
linear regression, showing that 99% of the variability of
MAP was explained by the combined influence of DAP
and SAP in resting humans. The (SAP, DAP) pressure set
and the (MAP, PP) pressure set were redundant, and only
two pressure-powered functions were carried out in the
(SAP, DAP, MAP, PP) four-pressure set.
ACKNOWLEDGMENTS
We thank Dr Karsten Plamann and Sheila Carrodus for
helpful discussion. We also thank the nurses of Kremlin-
Bice
#tre and Bichat Hospital.
REFERENCES
1 Milnor, W. R. (1982) Hemodynamics, William and
Wilkins, Baltimore
2 Nichols, W. and O’Rourke, M. (1998) McDonald’s Blood
Flow in Arteries. Theoretical, Experimental and Clinical
Principles, 4th edn, Edward Arnold, London
3 Wezler, K. and Bo
$ger, A. (1939) Die Dynamik des
arteriellen systems. In Ergebnisse der Physiologie
(Asher, L. and Sapiro, K., eds), pp. 292–606, Verlag
von JF Bergmann, Mu
$nchen
4 Gauer, O. H. (1960) Kreislauf des Blutes. In Lehrbuch
der Physiologie des Menschen (Landois, L. and
Rosemann, H.-U., eds), pp. 95–135, Von Urban and
Schwarzenberg, Munchen and Berlin
5 Franklin, S. S. and Weber, M. A. (1994) Measuring the
hypertensive cardiovascular risk: the vascular overload
concept. Am. Heart J. 128, 793–803
6 Sesso, H. D., Stampfer, M. J., Rosner, B. et al. (2000)
Systolic and diastolic blood pressure, pulse pressure, and
mean arterial pressure as predictors of cardiovascular
disease risk in men. Hypertension 36, 801–807
7 Verdecchia, P., Schillaci, G., Reboldi, G., Franklin, S. S.
and Porcellati, C. (2001) Different prognostic impact of
24-hour mean blood pressure and pulse pressure on
stroke and coronary artery disease in essential
hypertension. Circulation 103, 2579–2584
Received 15 October 2001/13 February 2002; accepted 25 March 2002
8 Kelly, R. P., Hayward, C. S., Avolio, A. P. and
O’Rourke, M. F. (1986) Non-invasive determination of
age-related changes in the human arterial pulse.
Circulation 80, 1652–1659
9 Franklin, S. S., Gustin, W., Wong, N. D., Larson, M. G.,
Weber, M. A., Kannel, W. B. and Levy, D. (1997)
Hemodynamic patterns of age-related changes in blood
pressure. The Framingham heart study. Circulation 96,
308–315
10 Smulyan, H. and Safar, M. E. (1997) Systolic blood
pressure revisited. J. Am. Coll. Cardiol. 29, 407–413
11 O’Rourke, M. and Frohlich, E. D. (1999) Pulse pressure.
Is this a clinically useful risk factor? Hypertension 34,
372–374
12 Simkus, G. J. and Fitchett, D. H. (1990) Radial artery
pressure measurements may be a poor guide to the
beneficial effects of nitroprusside on left ventricular
systolic pressure in congestive heart failure. Am. J.
Cardiol. 66, 323–326
13 Pauca, A. L., Wallenhaupt, S. L., Kon, N. D. and Tucker,
W. Y. (1992) Does radial artery pressure accurately reflect
aortic pressure? Chest 102, 1193–1198
14 Murgo, J. P., Westerhof, N., Giolma, J. P. and Altobelli,
S. A. (1980) Aortic input impedance in normal man:
relationships to pressure waveform. Circulation 65,
105–115
15 Chemla, D., He
!bert, J.-L., Coirault, C. et al. (1998) Total
arterial compliance estimated by the stroke volume-to-
aortic pulse pressure ratio in humans. Am. J. Physiol. 271,
H1287–H1295
16 Aubert, A. E., Vrolix, M., de Geest, H. and van de Herf,
F. (1995) In vivo comparison between two tip pressure
transducer systems. Int. J. Clin. Monit. Comput. 12,
77–83
17 Chemla, D., He
!bert, J.-L., Zamani, K., Coirault, C. and
Lecarpentier, Y. (1999) A new formula for estimating
mean aortic pressure. Lancet 353, 1069–1070
18 Saba, P. S., Roman, M. J., Pini, R., Spitzer, M., Ganau, A.
and Devereux, R. (1993) Relation of arterial pressure
waveform to left ventricular and carotid anatomy in
normotensive subjects. J. Am. Coll. Cardiol. 22,
1873–1880
19 Smulyan, H., Marchais, S. J., Pannier, B., Guerin, A. P.,
Safar, M. E. and London, G. M. (1998) Influence of body
height on pulsatile arterial hemodynamic data. J. Am.
Coll. Cardiol. 31, 1103–1109
20 Gevers, M., Hack, M. W. M., van Genderingen, H. R. and
Westerhof, N. (1995) Calculated mean arterial pressure in
the posterior tibial and radial artery pressure wave in
newborn infants. Basic Res. Cardiol. 90, 247–251
21 Meaney, E., Alva, F., Meaney, A., Alva, J. and Webel, R.
(2000) Formula and nomogram for the
sphygmomanometer calculation of the mean arterial
pressure. Heart 84,64
22 Berne, R. M. and Levy, M. N. (1998) The arterial system.
In Physiology (Berne, R. M. and Levy, M. N., eds),
pp. 415–428, Mosby Inc., St Louis
23 Pine, M. and Pine, J. (1995) Standardization of terms and
analystic methods for performance evaluation: achievable
goal or impossible dream? Manag. Care Q. 3, 7–14
24 Soldati, T., Geissler, H. and Schwartz, E. C. (1999) How
many is enough? Cell. Biochem. Biophys. 30, 389–411
25 Sunagawa, K., Maughan, W. L., Burkhoff, D. and Sagawa,
K. (1983) Left ventricular interaction with arterial load
studied in isolated canine ventricle. Am. J. Physiol. 245,
H773–H780
#2002 The Biochemical Society and the Medical Research Society