Measurement of pulse wave velocity using a beat-sampling technique.
Explore the source record for details and available documents.
Blood Flow Velocity↗
Biomedical subjects
N Westerhof
Publications and source records attributed to N Westerhof.
Linear and nonlinear one-dimensional models of pulse wave transmission at high Womersley numbers.
The accuracy of nonlinear and linear one-dimensional models in describing pulse wave propagation in a uniform cylindrical viscoelastic tube, with Womersley's parameter alpha equal to 7.6 at 1 Hz, was evaluated. To this end calculations of wave propagation using these models were compared with the experimentally determined propagation of the pressure wave in the tube. The experimentally generated pressure pulse had an amplitude of 9.0 kPa and caused a relative radius change of about 17%. The static pressure vs cross-sectional area relation was found to be nonlinear for these pressure changes. Maximum fluid velocity was about 2.9 ms-1, while the phase velocity was about 5.4 ms-1. The radius change and the ratio of fluid and phase velocities violated the linear model assumptions. The nonlinear model with viscous fluid friction modelled on the basis of Poiseuille's law and treating the tube wall as purely elastic, underestimated the damping of the pulse wave and predicted the formation of shock waves, which were not found experimentally. In the linear models, the viscous friction of the blood was modelled according to either Poiseuille's law or Womersley's theory and the tube wall was treated as either linearly elastic or linearly viscoelastic. A description of the viscous friction of the blood based on Poiseuille's law underestimated damping. Disregarding the viscoelasticity of the tube wall resulted in an underestimation of both phase velocity and damping. In spite of the nonlinearity of the system, the linear viscoelastic Womersley model described the pulse wave propagation satisfactorily.
Animals↗
Varying elastance concept may explain coronary systolic flow impediment.
We measured phasic arterial coronary inflow in the blood-perfused isolated cat heart (n = 5) with a balloon in the left ventricle under well-defined conditions, i.e., constant perfusion pressure, constant vasomotor tone (maximal vasodilation), and heart rate. The normalized amplitude (A) between systolic flow (Fs) and diastolic flow (Fd) [A = (Fd - Fs)/Fd] was related to systolic left ventricular pressure (Ps, range 1.6-17 kPa, 1 kPa = 7.5 mmHg) for different isovolumic beats obtained by changes in balloon volume and for low load isobarically ejecting beats (pressure 0.2 kPa). The data were fitted to A = a + bPs with a = 0.70 +/- 0.15 (SD) and b = 0.005 +/- 0.005 kPa-1. This relation indicates a very weak effect of left ventricular systolic pressure on normalized flow amplitude. Thus the hypothesis that left ventricular pressure is the sole determinant impeding coronary flow could not be confirmed. However, our data could be explained on basis of the time-varying elastance concept (H. Suga, K. Sagawa, and A. A. Shoukas. Circ. Res. 32: 314-322, 1973). The intravascular and luminal (cavity) compartments both are assumed to be subject to a time-varying elastance. The time-varying luminal elastance is similar for isovolumic and isobaric beats. We assume that the elastance of the vascular compartment also behaves the same for these beats, and therefore coronary flow is affected similarly.(ABSTRACT TRUNCATED AT 250 WORDS)
Animals↗
Contractility is the main determinant of coronary systolic flow impediment.
We measured the relation between coronary flow amplitude (delta F = Fd-Fs; where d is diastolic and s is systolic) and developed left ventricular pressure (delta PLV = Ps-Pd) at a constant perfusion pressure of 75 mmHg (10 kPa) in the maximally vasodilated blood-perfused isolated cat heart for different steady-state levels of contractility (protocol A) and during transients in contractility (protocol B). Contractility was defined as the slope of the end-systolic pressure-volume relation (Emax). From protocol A it appeared that the coronary flow amplitude was only weakly related to left ventricular pressure at each steady-state level of contractility studied. However, the coronary flow amplitude was strongly related to the different levels of contractility. In protocol B, contractility was changed over a wide range of values (0-100%) but developed pressure and contractility changed simultaneously. Using multiple linear regression analysis, we found that contractility has approximately 10 times (range: 2.8-57.3) stronger effect than left ventricular pressure on coronary flow amplitude (n = 10 experiments). These data and our earlier observations suggest that it is the difference in stiffness of cardiac muscle between systole and diastole that determines coronary flow amplitude.
Animals↗
Does the endothelium play a role in flow-dependent constriction? A study in the isolated rabbit femoral artery.
We studied the role of the endothelium in diameter changes as a function of flow of the isolated femoral artery of the rabbit (n = 15) perfused and superfused with a physiological salt solution (37 degrees C). In 10 vessels, diameters were studied before and after exposure to gossypol, an agent that impairs the endothelial function pharmacologically. In 5 of these 10 vessels we added albumin (1.5%) to the perfusion solution. The mean external diameter (+/- SEM) after equilibration for 60 min at a transmural pressure of 50 cm H2O (n = 10) was: 1,426 +/- 34 microns. Vessels were then constricted with norepinephrine (1.0-1.5 microM in the superfusion solution) to 70% of the resting diameter, acetylcholine was used to check endothelial function. All vessels constricted as flow was increased (p less than 0.001), irrespective of the impairment of the endothelial function by gossypol or the presence of albumin. It is therefore unlikely that the flow-induced constriction results from a 'wash away' effect of endothelium-derived relaxing factor (EDRF). To test whether EDRF could still play a role after gossypol, we used hemoglobin (n = 5) to bind EDRF. Flow-dependent constriction was still observed, although the mean diameter was decreased. We conclude that flow-dependent constriction is either mediated via the endothelial cells, but not via EDRF, or that the endothelial cells are not involved.
Acetylcholine↗
Left ventricular time varying elastance behavior does not reflect a basic property of cardiac muscle.
Explore the source record for details and available documents.
Animals↗
Feline left ventricular oxygen consumption is not affected by volume expansion, ejection or redevelopment of pressure during relaxation.
We studied the dependency of myocardial oxygen consumption on the mechanical events during left ventricular relaxation in isolated supported cat hearts. The volume of the left ventricle was controlled by means of a balloon connected to a membrane pump. Oxygen consumption (MVO2 in cm3.min-1.100 g-1) for three protocols (PROT) performed at peak isovolumic pressure, was studied: (1) rapid ejection to zero pressure, (2) partial rapid ejection followed by redevelopment of pressure, (3) volume expansion during relaxation, and compared with oxygen consumption of isovolumic (ISOV) beats. We found (mean +/- SD): (table; see text) In the protocols 1 and 3 the differences were not significant (paired Student's t-test, p greater than 0.05). In protocols 1 and 2 left ventricular volume was decreased up to 2.15 cm3 (i.e. stroke volume, SV) during the pressure release. We studied the specific effect of ejection (i.e., wall muscle shortening) in a fourth protocol in which the ventricle ejected up to 2.7 cm3 under nearly zero pressure load (isobaric contraction). There was a small amount of oxygen consumption associable with this unloaded ejection i.e. MVO2 = 3.38 (+/- 0.47) + 0.30 (+/- 0.16) SV, but it was too small to compensate for a decrease in MVO2 expected from the pressure release according to the tension time index. These findings suggest that oxygen consumption does not depend on the mechanical events during ventricular relaxation.
Animals↗
Influence of geometric taper on the derivation of the true propagation coefficient using a three point method.
We studied the effect of geometric taper on the derivation of the true propagation coefficient from three pressures determined 10 cm apart ('three-point method'). For this purpose the true propagation coefficients of a uniform latex tube (length 50 cm, outer diameter 12.73 mm, Womersley phase velocity 6.23-6.42 ms-1 (1-10 Hz), Womersley damping coefficient 0.05-0.14 m-1 (1-10 Hz) and of a tapered latex tube (length 50 cm, outer diameter varying from 15.88 to 9.45 mm, in the middle section with same properties as the uniform tube) were determined. The real part of the true propagation coefficient (the damping coefficient) was compared with apparent damping, and with the damping coefficient calculated using Womersley's theory. The imaginary part of the true propagation coefficient (the phase coefficient) was expressed in terms of phase velocity. True phase velocity was compared with measurements of apparent phase velocity, foot-to-foot velocity, and calculations of phase velocity parameters Womersley's theory and the Moens-Korteweg equation. The results show that in the uniform tube the three-point propagation coefficient is in agreement with all other estimates. Taper causes an error in the three-point propagation coefficient. At some frequencies the damping is reversed to amplification (values up to -2 m-1) and the phase velocity may be both overestimated or underestimated (up to 50%). The overestimation of true damping as reported in the literature cannot be explained from vessel taper.
Arteries↗
Matching between feline left ventricle and arterial load: optimal external power or efficiency.
We tested the hypothesis that the feline left ventricle normally works at optimal external power as opposed to optimal efficiency by (re)analyzing data from five isolated, blood-perfused cat hearts and 39 open-thorax cats. In the isolated hearts, we measured pump function, external steady power, myocardial oxygen consumption, and efficiency. Optimal external power and optimal efficiency were found at different left ventricular outputs (6.94 +/- 0.33 and 8.35 +/- 0.37 ml/s, respectively; P less than 0.001). In the in situ cat hearts the working point was found at an output of 4.72 +/- 0.32 ml/s, whereas optimal external power was found at 4.84 +/- 0.26 ml/s. These values were not significantly different. Assuming that the point of optimal efficiency was located at the same fraction of the maximal unloaded left ventricular output (Fmax) as in the isolated hearts, i.e., 0.7, we found the point of optimal efficiency for the in situ heart at a flow of 5.83 +/- 0.32 ml/s, which was significantly different (P less than 0.001) from the flow in the working point. Our data therefore indicate that the left ventricle in the open-thorax cat is matched to the arterial load such that its external power output rather than efficiency is optimized.
Algorithms↗
Pump perfusion abolishes autoregulation possibly via prostaglandin release.
The influence of pump perfusion on autoregulation was studied in the hindleg of the halothane- and chloralose-anesthetized cat. Flow was measured with an electromagnetic flow probe in a tube between aorta and the vascularly isolated, denervated leg and varied with a calibrated occluder. Perfusion pressure was measured via a T-piece distal to the occluder. The steady-state pressure-flow relations could be fit with a sigmoidal curve. The mean closed-loop gain (0 less than Gc less than 1) for autoregulation in six cats was 0.46 +/- 0.11 (SD). When in these cats a roller pump was used, an almost linear pressure-flow relation was found (Gc = 0.01 +/- 0.09), while the resistance at control flow was decreased by 15 +/- 4%. Administration of indomethacin (5 mg/kg iv), a cyclooxygenase inhibitor, partly restored autoregulation during pumping (Gc = 0.34 +/- 0.09) and slowly increased the resistance to above its original value (20 +/- 13%). In six other cats, pump perfusion had no influence on autoregulation when started after indomethacin administration but resistance increased. This increase could not be prevented with ketanserin, a specific serotonin 2 receptor blocker. We conclude that pump perfusion abolishes autoregulation and decreases resistance via a process that involves prostaglandins. Blockade of the prostaglandin synthesis unmasks a slow vasoconstrictor influence in the bed.
Animals↗
Changes in coronary pressure-flow relation after transition from blood to Tyrode perfusion.
In six isolated, diastolic-arrested, maximally vasodilated cat hearts, we studied changes in coronary pressure-flow relations (zero-flow pressure intercept, resistance) during the first 25 min, after a change of perfusate, from blood to Tyrode. The apparent intercept (zero-flow) pressure changed from 2.0 +/- 0.94 (+/- SD) kPa during blood perfusion to 2.5 +/- 0.55, 2.6 +/- 0.68, 2.5 +/- 0.94, and 2.7 +/- 1.34 kPa during Tyrode perfusion for 2:15, 5:30, 10:30, and 25:00 min, respectively. Intercept pressures during Tyrode perfusion were significantly different from the intercept pressure during blood perfusion, except for the one measured after 25 min of Tyrode perfusion (P less than 0.05). Resistance (defined as the ratio of perfusion pressure and flow at 10 kPa perfusion pressure) steadily rose to approximately 170% of the value during blood perfusion. The observation that the apparent intercept pressure is maintained, when a particle-free (Newtonian) isotonic perfusate is used, may indicate that this intercept is not a result of blood rheology alone. The increase rather than decrease in resistance suggests an effect of edema, which increases interstitial volume at the expense of intravascular volume.
Animals↗
Aortic input impedance during Mueller maneuver: an evaluation of "effective length".
Aortic input impedance was calculated in seven subjects in the control state (normal reflection) and during the Mueller maneuver (increased reflection) to evaluate "effective arterial length" under altered physiological conditions. Regional foot-to-foot pulse wave velocities and pressure waveforms along the aorta were used to define an "apparent anatomic length" or distance to a dominant discrete site of reflection "seen" by the ejecting ventricle. Time of wave travel was taken to be one-half the interval from the foot of the incident wave to the midsystolic inflection point. Knowing the time of travel from the returning reflection and velocity, distances calculated to the "apparent anatomic length" were 35 +/- 2 and 34 +/- 2 during control and Mueller maneuver, respectively (P = NS). The frequency of the first minimum of the modulus (fmin) and the first zero crossing of the phase angle (f phi) were determined from the input impedance spectra. During baseline conditions, fmin (3.9 +/- 0.2 Hz) approximately equaled f phi (4.2 +/- 0.2 Hz), and the resulting "effective lengths" calculated using the quarter-wavelength formula were similar to the apparent anatomic length. These data suggested that the aortic region incorporating the renal arterial branches as a site of discrete reflection and that terminal load was not significantly frequency dependent. During Mueller maneuver, however, f min (3.3 +/- 0.2 Hz) and f phi (5.1 +/- 0.2 Hz) were significantly discordant, the terminal load became strongly frequency dependent, and effective length calculated from f min was dissimilar (P less than 0.05) from the unchanged apparent anatomic length.(ABSTRACT TRUNCATED AT 250 WORDS)
Adult↗
Ventricular/vascular coupling and regional arterial dynamics in the chronically hypertensive baboon: correlation with cardiovascular structural adaptation.
Ventricular/vascular coupling dynamics and regional hemodynamics of five hypertensive baboons with concentric left ventricular (LV) hypertrophy (mean arterial pressure +/- SD, 148 +/- 16 mm Hg; LV mass/body weight ratio 3.42 +/- 0.8) were compared with five normotensive controls (mean arterial pressure 89 +/- 3 mm Hg; LV mass/body wt ratio 2.73 +/- 0.5) at different mean arterial pressures. Ventricular/vascular dynamics were assessed by aortic input impedance, pulsatile/total power ratio, effective arterial elastance and compliance from a three-element Windkessel "lumped" model of the circulation. Regional arterial dynamics were assessed by pulse-wave velocities and local reflection coefficients. Systemic arterial compliance was similarly decreased with elevated pressure in both groups but was significantly more reduced for the hypertensive group compared with control animals at control (0.49 +/- 0.16 vs. 0.96 +/- 0.09 ml/mm Hg; p less than 0.05) and acutely lowered arterial pressure (0.62 +/- 0.26 vs. 1.41 +/- 0.24 ml/mm Hg, respectively). Changes in compliance were paralleled by differences in effective arterial elastance derived from cineventriculographic pressure-volume ratios. Regional foot-foot and apparent phase pulse-wave velocities were significantly increased for distal aortic segments of the hypertensive animals during elevated pressures compared with controls (cff, 17.5 +/- 7.5 vs. 8.7 +/- 3.0 m/sec; p less than 0.05). Histology of the aorta revealed significant increases in collagen content (microgram/mg dry wt) from proximal to distal aortic segments (27 +/- 2 vs. 38 +/- 6; p less than 0.005) in hypertensive animals but not in controls (27 +/- 2 vs. 32 +/- 6; NS). With pharmacological normalization of systemic arterial pressures, hypertensive baboons developed aortic wave speeds similar to controls but manifested significantly reduced compliance compared with controls. In contrast, with acute elevations of pressure, systemic arterial aortic compliances were similar for both groups, but distal pulse-wave velocities were significantly increased for hypertensive animals compared with controls. We conclude that measures of ventricular/vascular coupling and arterial dynamics are determined by both the level of arterial pressure and the physical characteristics of the cardiovascular system in chronic systemic hypertension and pressure overload ventricular hypertrophy.
Adaptation, Physiological↗
Mechanical determinants of myocardial energy turnover.
Energy turnover of the left ventricle does not differ in isovolumic contractions and contractions where pressure is released from peak to zero. This experimental result corresponds to predictions from a time varying elastance model of the mechanical and energetic properties of the left ventricle. To assess the validity of this model for cardiac muscle in general, experiments were designed to investigate whether mechanical and energetic behaviour of isolated cardiac muscle preparations could also be predicted from the time varying elastance model. The results obtained so far indicate, however, that not all experimental results can be accommodated by the model. This suggests that the value of the model may be limited.
Animals↗
Short-term regulation of arterial pressure and the calculation of open-loop gain in the intact anesthetized dog.
Open-loop gain of the short-term systemic pressure regulation was determined under closed-loop conditions in the closed chest anesthetized dog (n = 5). For this purpose, cardiac output and mean systemic pressure were varied by ventricular pacing after the production of complete heart block. From the pressure-flow data resistance gain (the ratio of peripheral resistance change to pressure change in the steady state) was obtained by means of a simple model. The value of this gain was automatically estimated by fitting the pressure-flow relation described by the model to the experimental data. The model allows the pressure-flow relation to be straight or curved with or without a zero-flow pressure intercept. The best fit was obtained when the pressure-flow curve was convex to the pressure axis and had no intercept. When the model was linearized about the control values of pressure and flow (operating point), open-loop gain could be calculated from resistance gain. Its averaged value in the control condition, 1.63 +/- 0.45, is in agreement with values found by other investigators in open-loop conditions. During vasoconstriction open-loop gain, at the (new) operating point, increased to 2.51 +/- 0.51; during vasodilation it decreased to 1.17 +/- 0.27. Open-loop gain about an operating point thus can be determined in the intact animal from measurements of mean pressure and mean flow in the steady state.
Animals↗
Dynamic model of the short-term regulation of arterial pressure in the cat.
Explore the source record for details and available documents.
Animals↗
Dynamics of the short-term regulation of arterial pressure: frequency dependence and role of arterial compliance.
Explore the source record for details and available documents.
Animals↗
Total systemic arterial compliance and aortic characteristic impedance in the dog as a function of pressure: a model based study.
Total arterial compliance and aortic characteristic impedance as a function of pressure in the anesthetized closed chest dog (n = 5) were studied. The three-element windkessel (consisting of a peripheral resistance, a total systemic arterial compliance, and an aortic characteristic impedance) was assumed as an arterial model. Aortic pressure was varied by pacing the heart at different rates after the production of atrioventricular block and by administration of Angiotensin and Hydralazine. Model parameters were estimated by two different methods. The first was based on a computerized optimization procedure using all the information contained in the aortic pressure and flow waveforms. The second method used the diastolic decay of aortic pressure to compute total arterial compliance and used the arterial input impedance spectrum to compute aortic characteristic impedance. Total arterial compliance and aortic characteristic impedance changed with pressure. The parameter optimization procedure yielded values of total arterial compliance ranging from 0.20 to 1.4 ml/mmHg and values of aortic characteristic impedance ranging from 0.05 to 0.42 mmHg sec/ml. Values of parameters estimated on the basis of the impedance spectrum and diastolic pressure decay were similar. Compliance values as a function of mean aortic pressure could be fitted with a bell-shaped curve similar to that found from in vitro studies of aortic segments. Characteristic impedance values as a function of mean aortic pressure could be fitted with a parabolic function the minimum of which was found in the range of control to high pressures (90-160 mmHg).
Animals↗