We developed a mathematical model describing the interaction between the he
art and the arterial system. The model was constructed and tested on basis
of invasive hemodynamic data in six sheep. Data from a first group of three
animals (49 cardiac cycles) were used to assess a template time-varying el
astance curve for the left ventricle, while the baseline steady-state data
of a second group of three animals were used to assess reference cardiac an
d arterial parameters in sheep. The model is fully characterized by nine pa
rameters, which were converted into 6 dimensionless numbers using the Bucki
ngham II theorem. The model was then used to generate LV pressure and volum
e and aortic pressure and how for 86 conditions obtained by varying paramet
ers 50 to 200% of their reference value. Systolic (SBP) and diastolic (DBP)
blood pressure and stroke volume (SV) were determined from these model-gen
erated curves and multiple linear regression analysis yielded the following
expressions: SEP = P-isovolumic[0.638 - 0.0773 E-max C+ 0.0507 RC/T] (r(2)
=0.89); DBP=P-isovolumic [0.438 -0.0712 EmaxC+ -0.0655RC/T] (r(2)=0.88) and
SV=LVEDV [1.165-1.040 LVEDV/(LVEDV -V-d) + 0.125 E-max C-0.0777RC/T] (r(2)
= 0.93) with P-isovolumic = E-max(LVEDV - V-d), E-max and V-d being the sl
ope and intercept of the end-systolic pressure-volume relation, R and C the
total peripheral resistance and compliance, LVEDV the left ventricular end
-diastolic volume, and T the cardiac cycle length. These expressions were v
alidated using data from the second group of three animals obtained during
vena cava occlusion at baseline and during administration of dobutamine (61
cycles). The correlation between measured and predicted values was 0.98, 0
.97 and 0.92 for SEP, DBP and SV, respectively. Compared to the measured va
lues, SEP and DBP were, on average, underestimated by 5 and 6 mmHg, respect
ively, and SV overestimated by 1.4 ml. We conclude that the derived express
ions for blood pressure and stroke volume remain valid in the intact sheep
for various hemodynamic conditions, and, taking into account their dimensio
nless form, may hold in other species and in humans. (C) 2000 Elsevier Scie
nce Ltd. All rights reserved.