Systemic vascular resistance incompletely describes left ventricular afterload because of the phasic nature of arterial pressure and blood flow. Aortic input impedance is an experimental description of left ventricular afterload that incorporates the frequency- dependent characteristics and viscoelastic properties of the arterial system. The effects of propofol on aortic input impedance were examined using three variables derived from the three-element Windkessel model: characteristic aortic impedance, total arterial compliance, and total arterial resistance.
Eight dogs were chronically instrumented for measurement of aortic pressure, left ventricular pressure, +dP/dt, subendocardial segment length, and aortic blood flow. Systemic hemodynamics and aortic blood pressure and flow waveforms were recorded in the conscious state and after a bolus of 5 mg x kg(-1) propofol and infusion for 15 min at 25, 50 and 100 mg x kg(-1) x h(-1). Aortic input impedance spectra were generated using power spectral analysis of aortic pressure and flow waveforms corrected for the phase responses of the pressure and flow transducers. Characteristic aortic impedance, total arterial resistance, and total arterial compliance were calculated from the aortic input impedance spectrum and the aortic pressure waveform. Parameters describing the net site and magnitude or arterial wave reflection were determined from aortic impedance.
Propofol decreased total arterial resistance (3.05 +/- 0.20 during control to 2.29 +/- 0.18 dynes x s x cm(-5) x 10(3) at the high dose) and increased total arterial compliance (0.53 +/- 0.04 during control to 1.15 +/- 0.17 ml x mmHg(-1) at the high dose) in a dose- related manner. Propofol increased characteristic aortic impedance (1.49 +/- 0.15 during control to 2.20 +/- 0.20 dynes x s x cm(-5) x 10(2) at the high dose). The net site and the magnitude of arterial wave reflection were unchanged by the propofol.
In chronically instrumented dogs, propofol decreased total arterial resistance, a property of arteriolar resistance vessels, consistent with the known actions of this drug on systemic vascular resistance. Propofol also increased total arterial compliance and characteristic aortic impedance, indicating that this anesthetic affects the mechanical properties of the aorta. Propofol had no effect on arterial wave reflection patterns. The results indicate that propofol reduces left ventricular afterload via decreases in peripheral resistance and increases in arterial compliance.
Key words: Anesthetics, intravenous: propofol. Heart: left ventricular afterload. Hemodynamics: aortic blood flow; aortic pressure. Signal processing: coherence function; power spectrum analysis.
INDUCTION or maintenance of anesthesia with propofol frequently is associated with hypotension. [1-9]Reductions in systemic arterial blood pressure produced by propofol have been attributed to a combination of venous [4-6,10-14]and arterial vasodilation [3,4,7,13,15-18]and direct depression of myocardial contractility. [11,16,19-23]Propofol-induced decreases in systemic vascular resistance most often have been used to describe reductions in left ventricular afterload caused by this intravenous anesthetic. However, systemic vascular resistance inadequately characterizes afterload, because this calculated index does not consider the viscoelastic and frequency-dependent properties of the arterial wall, the dynamic phasic nature of arterial pressure and flow, or the potential effects of wave reflection occurring within the vascular tree. Arterial mechanical properties are more thoroughly described using aortic input impedance (Zin), defined by the complex ratio of aortic blood pressure and flow and described with modulus and phase spectra in the frequency (omega) domain. [24-26].
Aortic input impedance often is interpreted through a simple electrical analog known as the three-element Windkessel because frequency-dependence makes many features of the Zin(omega) spectrum difficult to quantify. The Windkessel model consists of a resistor (characteristic aortic impedance [Zc]) in series with a parallel combination of another resistor (R; total arterial resistance) and a capacitor (C; total arterial compliance). Characteristic aortic impedance is determined by the resistance of the aorta and the compliance of this vessel. The total hydraulic resistance is represented by R (according to the law of Poiseuille) of the entire arterial vasculature. The energy storage element of the arterial system is denoted as C and is defined primarily by the mechanical properties of the aorta itself. The interaction of these arterial properties with the mechanical characteristics of the left ventricle helps determine overall cardiovascular performance. The three-element Windkessel closely approximates Zin(omega) in a wide variety of physiologic conditions. [27,29]We demonstrated recently that halothane and isoflurane produce differential effects on Zin(omega) quantified with Windkessel parameters, indicating that this experimental technique also can be used to describe changes in frequency-dependent indices of left ventricular afterload in the presence of anesthetics. The effects of propofol on specific arterial resistance and compliance variables have not been described and should be examined to provide a more complete understanding of the action of this intravenous anesthetic on the arterial circulation. This investigation was undertaken to characterize the effects of propofol on aortic input impedance and to quantify alterations in afterload produced by this agent using the three-element Windkessel model.
All experimental procedures and protocols used in this investigation were reviewed and approved by the Animal Care Committee of the Medical College of Wisconsin. All procedures conformed to the Guiding Principles in the Care and Use of Animals of the American Physiologic Society and were performed in accordance with the Guide for the Care and Use of Laboratory Animals (Department of Health, Education, and Welfare--Department of Health and Human Services publication [NIH] 85-23, revised 1985).
Surgical implantation of instruments has been described in detail previously. [30,31]Briefly, under general anesthesia and using aseptic technique, conditioned mongrel dogs (weighing 26+/-2 kg) underwent a left thoracotomy, and heparin-filled catheters were placed in the proximal descending thoracic aorta and the right atrium for measurement of aortic pressure and fluid or drug administration, respectively. An ultrasonic transit-time flow probe (Transonic Systems, Ithaca, NY) was positioned around the ascending thoracic aorta for measurement of continuous aortic flow. Typical aortic pressure and blood flow waveforms are depicted in Figure 1. A pair of miniature ultrasonic segment length transducers (5 MHz) were implanted in the left ventricular subendocardium for measurement of changes in regional contractile function (percent segment shortening). A high-fidelity micromanometer (P7; Konigsberg Instruments, Pasadena, CA) was positioned in the left ventricle for measurement of continuous left ventricular pressure and the maximum rate of increase in that pressure (dP/dtmax). A heparin-filled catheter was inserted directly into the left atrial appendage, and the left ventricular micromanometer was cross-calibrated in vivo against pressures measured via arterial and left atrial catheters (P50pressure transducer, Gould Instruments, Oxnard, CA). All instrumentation was secured, tunneled between the scapulae, and exteriorized via several small incisions. The pericardium was left widely open, the chest wall closed in layers, and the pneumothorax evacuated by a chest tube. Each dog was fitted with a jacket (Alice King Chatham, Los Angeles, CA) to prevent damage to the instruments and catheters that were housed in an aluminum box within the jacket pocket.
All dogs received systemic analgesics (fentanyl) as required after surgery. Dogs were treated with intramuscular antibiotics (40 mg *symbol* kg sup -1 cephalothin and 4.5 mg *symbol* kg sup -1 gentamicin) and were allowed to recover a minimum of 7 days before experimentation. Dogs were trained to stand quietly in an animal sling during hemodynamic monitoring. Segment length signals were monitored with an ultrasonic amplifier (Crystal Biotech, Hopkinton, MA). End-systolic and end-diastolic segment lengths were determined at maximum negative left ventricular dP/dt and just before the onset of left ventricular isovolumic contraction, respectively. Percent segment shortening was calculated using the equation:percent segment shortening = (end-diastolic segment length-end-systolic segment length) *symbol* 100 *symbol* end-diastolic segment length sup -1. Hemodynamic data were continuously recorded on a polygraph (model 7758A; Hewlett-Packard, San Francisco, CA) and digitized by a computer interfaced with an analog to digital converter.
Calculation of Aortic Input Impedance Spectra
Aortic blood pressure and blood flow waveforms were transformed from the time to frequency domain using spectral analysis to determine Zin. [27,32]The aortic input impedance spectrum was displayed by plotting the magnitude and phase of Zinas a function of frequency (omega). The Zin(omega) modulus was used to describe the ratio of the magnitude of pressure to the magnitude of flow at each point in the frequency domain. The Zin(omega) phase was used to describe the difference between the phase angles of flow and pressure at each frequency. Typical Zin(omega) magnitude and phase spectra are illustrated in Figure 2.
Aortic input impedance spectra were determined from digitized, steady-state aortic blood pressure and aortic blood flow waveforms as described previously. Briefly, data files consisting of 4,096 points were sampled at 100 Hz and divided into five 2,048 point bins with 1,536 point overlap. A Hamming window was applied to each bin to reduce side lobe leakage. The autopower spectrum of the aortic blood pressure [P sub pp (omega)] and aortic blood flow [Ppf(omega)] and cross power spectrum between aortic pressure and blood flow waveforms [Ppf(omega)] were determined using a Welch periodogram technique. [34,35]The aortic input impedance [Zin(omega)] was calculated as a function of frequency (omega) using the formula: Zin(omega) = Ppp(omega) *symbol* [Ppf(omega)] sup -1. The calculated Zin(omega) spectra were corrected for the phase responses of the aortic flow probe and the aortic pressure transducer as described previously. Correlation of aortic pressure and flow waves at each frequency of the input impedance spectrum was determined using the magnitude squared coherence (MSC), where MSC(omega) = [left vertical bar] [Ppf(omega)] [right vertical bar]2*symbol* [Ppp(omega) *symbol* Pff(omega)] sup -1. Input impedance data with magnitude squared coherence values < 0.8 were discarded as outlined previously. .
The characteristic aortic impedance (Zc) was determined from the aortic input impedance spectra as the mean of the magnitude of Z sub in (omega) ([left vertical bar] Zin(omega) [right vertical bar]) between 2 and 15 Hz. [27,36,37]Total arterial resistance (R) was calculated as the difference between the value of [left vertical bar] Z sub in (omega) [right vertical bar] at zero frequency and Zc. [left vertical bar] Zin(omega) [right vertical bar] at zero frequency is equal to systemic vascular resistance determined as the ratio of mean arterial pressure and mean aortic blood flow (0000i.e., cardiac output). The C component of the Windkessel model was calculated using the equation. Equation 1where Ad= the area under the diastolic portion of the arterial pressure curve above mean venous pressure (assumed to be zero mmHg); MAP = mean aortic pressure; MAQ = mean aortic blood flow; Pes= end-systolic aortic pressure; and Ped= end-diastolic aortic pressure. The diastolic period for compliance calculation defined as the time between the dichrotic notch and minimal aortic pressure. The value of C was determined from the average of 5 consecutive beats for each intervention.
The Windkessel model allows for quantification of characteristics of Zin(omega) in physical terms (R, Zc, and C) but does not strictly describe the influence of arterial wave reflections. Variables that predict changes in the timing of arterial wave reflections, the frequency of the first minimum of [left vertical bar] Zin(omega) [right vertical bar] (Fmin) and the zero phase intercept of Zin(omega) (Ftheta), were directly calculated at each intervention from the Zin(omega) modulus and phase spectra, respectively. In addition, the arterial wave reflection factor (Delta Z/Zc), defined as the ratio of the difference between the first minimum and following maximum of Zin(omega) and Zc, was determined from the Zin(omega) modulus spectrum. This arterial wave reflection factor is proportional to the magnitude of the reflected waves.
All dogs (n = 8) were fasted overnight and received 500 ml 0.9% saline before experimentation. Intravenous fluids (0.9% saline) were continued at 3 ml *symbol* kg sup -1 *symbol* h sup -1 for the duration of each experiment. The instrumentation was calibrated and baseline systemic hemodynamics were recorded in the conscious state. Continuous aortic blood pressure and aortic blood flow waveforms were recorded for later generation and analysis of Zin(omega). Anesthesia was induced with 5 mg *symbol* kg sup -1 intravenous propofol. After tracheal intubation, anesthesia was maintained with propofol infusions at 25, 50, or 100 mg *symbol* kg sup -1 *symbol* h sup -1 administered in a random manner. Lungs were mechanically ventilated with an air and oxygen (25%) mixture. Arterial blood gases were maintained at conscious levels by adjustment of air and oxygen concentrations and respiratory rates throughout the experiment. Systemic hemodynamics and aortic pressure and flow waveforms were recorded after 15 min of equilibration at each propofol infusion rate. The infusion rate of propofol was then changed and data were recorded after a similar period of equilibration.
Statistical analysis of data in the conscious state and during each anesthetic intervention was performed by analysis of variance with repeated measures followed by application of the Student's t test with Duncan's correction for multiplicity. Changes between interventions were considered statistically significant when the P value was less than 0.05. All data were expressed as mean+/-SEM.
Propofol was associated with a significant (P < 0.05) increase in heart rate and dose-dependent decreases in systolic, diastolic, and mean arterial pressure and left ventricular systolic pressure (Table 1). Decreases in left ventricular end-diastolic pressure, cardiac output, stroke volume, and systemic vascular resistance also were observed during administration of propofol. Declines in left ventricular peak positive dP/dtmaxand percent segment shortening occurred, consistent with a negative inotropic effect. Total arterial resistance was decreased by propofol (3.05+/-0.20 during control to 2.29+/-0.18 dynes *symbol* second *symbol* centimeter sup -5 *symbol* 103at the highest dose; Figure 3). A dose-related increase in C was observed (0.53 +/-0.04 during control to 1.15+/-0.17 ml *symbol* mmHg sup -1 at the highest dose; Figure 3). This propofol-induced increase in C was accompanied by an increase in characteristic aortic impedance (Zc; 1.49+/-0.15 during control to 2.20+/-0.20 dynes *symbol* second *symbol* centimeter sup-5 *symbol* 102at the highest dose; Figure 3).
The Ftheta, the first minimum of the magnitude of the Z sub in (omega) spectrum (Fmin), and the arterial wave reflection factor (Delta Z/Zc) were unchanged with the administration of propofol. These findings indicate that the timing of arterial wave reflection and the magnitude of the reflected waves were unaffected by this intravenous anesthetic.
Systemic vascular resistance, calculated as the ratio of mean arterial pressure and mean arterial flow (cardiac output), is the most frequently used estimate of left ventricular afterload. This index of afterload provides a useful intuitive picture of the arterial resistance to left ventricular ejection, however, this parameter alone does not provide a complete description of afterload. Systemic vascular resistance fails to account for the dynamic, phasic nature of arterial pressure and blood flow, ignores the viscoelastic properties of the arterial wall, does not consider the potential effects of arterial wave reflections, and cannot be used to parametrically quantify changes in afterload induced by pharmacologic agents or cardiovascular disease. [25,40,41]Zin(omega) incorporates the frequency-dependent and viscoelastic properties of and the wave reflection characteristics occurring in the arterial circulation and provides a more complete experimental description of left ventricular afterload. Unfortunately, many of the features of Z sub in (omega) are difficult to evaluate in a physiologically meaningful way because of the frequency-dependence inherent to the model. The three-element Windkessel has been used as a simplified model of Zinto facilitate quantitative analysis of Zin(omega) spectrum. This electrical analog displays most of the characteristics of Zin(omega) in the frequency domain. The Windkessel describes three variables which are properties of the arterial system (Zc, R, and C). Zin(omega) can be determined from these variables as a function of frequency : Equation 1.
In the current investigation, Windkessel parameters were used to quantify Zin(omega) spectra in the conscious state and during propofol anesthesia. The results confirm and extend the findings of previous studies demonstrating that propofol-induced decreases in systemic vascular resistance contribute to declines in mean arterial pressure in experimental animals and humans. [3,4,7,13,15-18,23]Decreases in systemic vascular resistance were accompanied by declines in R, indicating that propofol causes vasodilation by affecting resistance vessels. Decreases in R caused a reduction in arterial pressure despite propofol-induced increases in Zc. Changes in Zcare determined by the viscoelastic properties of the aortic wall and the dimensions of this great vessel. [25,44]Typically, R is an order of magnitude greater than Zc, consistent with the concept of the aorta as a low-resistance, high-compliance conduit and the arterioles as resistance vessels. The increases in Zcobserved during the administration of propofol most likely resulted from dose-related decreases in intra-aortic pressure, which infers corresponding decreases in aortic diameter. These actions would be expected to cause increases in the hydraulic resistance of the aorta and increases in Zc. In the absence of simultaneous reductions in R and increases in C produced by propofol, an increase in Zcmay theoretically lead to less efficient coupling of the left ventricle with the arterial system and contribute to wasted left ventricular energy. .
Propofol caused dose-dependent increases in C, indicating that this agent affects arterial compliance as well as resistance. The vast majority of C is determined by aortic compliance. [28,46]The elastic properties of the proximal aorta allow the efficient storage of left ventricular ejection energy generated during systole and the effective diastolic redistribution of this energy to the arterioles and capillaries. The rectifying properties of the combination of the aortic compliance and the aortic valve maintain diastolic arterial pressure and enhance coronary perfusion. Total arterial compliance is determined by the interrelation between collagen, elastin, and vascular smooth muscle in the arterial wall and is inversely related to intra-arterial pressure. [36,45]However, experimental evidence has suggested that the relationship between compliance and pressure is relatively flat over the range of mean pressures observed in this investigation. [38,47,48]A previous investigation from our laboratory using this same model found only a small increase in C with halothane and isoflurane. In contrast, equihypotensive doses of sodium nitroprusside increased the slope of the pressure-compliance relationship, confirming that this arterial vasodilator causes direct increases in C by affecting aortic mechanical properties. [33,36,49]In the current investigation, propofol increased the slope of the pressure-compliance relationship (Figure 4) over a range of arterial pressures similar to those observed in our previous study with volatile anesthetics and sodium nitroprusside. A test for parallelism was performed to compare the slopes of the compliance versus pressure relationship for propofol to those of halothane, isoflurane and sodium nitroprusside. The slope of the propofol compliance versus pressure relationship (-2.34 x 10 sup -3 ml *symbol* mmHg sup -2) was significantly greater than that of halothane (-1.43 x 10 sup -3 ml *symbol* mmHg sup -2, t = 2.32, P < 0.05) and isoflurane (-1.41 x 10 sup -3 ml *symbol* mmHg sup -2, t = 2.33, P < 0.05), but equal to that of sodium nitroprusside (-3.70 x 10 sup -3 ml *symbol* mmHg sup -2, t = 0.07, P > 0.05). These findings indicate that propofol-induced increases in C are not entirely dependent on reductions in arterial pressure and suggest that this intravenous anesthetic decreases left ventricular afterload by modifying the mechanical characteristics of the aorta.
Although not accounted for by the Windkessel model, patterns of arterial wave reflection also were examined using the magnitude and phase components of Zin(omega) determined in the conscious state and during propofol anesthesia. Reflected waves occur at branching sites in the arterial circulation because the characteristic impedance of a proximal trunk does not necessarily equal the combined characteristic impedances of the distal branches. This mismatch causes some of the forward energy of left ventricular ejection to be reflected back toward the heart. Oscillations of the Zin(omega) modulus spectrum at higher frequencies have been shown to be directly proportional to the magnitude of reflected waves. The frequency of the Fminof the Zin(omega) modulus and the Fthetaof the Zin(omega) phase inversely correlate with the distance to the major reflecting site (the average sum of all reflecting sites in relation to the aortic root). In the current investigation, Delta Z/Zc, Fmin, and Fthetawere unchanged by propofol, indicating that this intravenous anesthetic has no effect on arterial oscillatory properties despite concomitant decreases in R and increases in Zcand C produced by this drug.
The current results must be interpreted within the constraints of several possible limitations. Arterial pressure waveforms measured with a chronically implanted fluid-filled catheter were used to calculate Zin(omega). Although the magnitude and phase of Zin(omega) were appropriately corrected using established methods, a high-fidelity micromanometer placed at the aortic root may have provided a better frequency response. The distance between the pressure and flow transducers may have introduced an error in the Zin(omega) phase spectra and the determination of Fthetadespite appropriate adjustment of the magnitude and phase Zin(omega) for the distance between these instruments by verified techniques. The Zin(omega) modulus spectra were somewhat less continuous in propofol-anesthetized dogs than those obtained in the conscious state (Figure 2) because some frequencies between the fundamental and corresponding harmonics were excluded on the basis of magnitude squared coherence criteria. This relative discontinuity in Zin(omega) may have introduced an error in the calculation of Zc, Delta Z/Zc, and Fmin. Generation of random heart rates by cardiac pacing during anesthesia would have provided a greater number of fundamental and harmonic frequencies, resulting in more continuous Zin(omega) spectra during propofol anesthesia. However, this spectral discontinuity resembles spectra generated with standard Fourier series analysis, an established method for evaluating aortic or pulmonary input impedance and wave reflection properties under a variety of physiologic conditions. [25,45].
The doses of propofol used in this investigation were chosen to produce reliable anesthesia in all dogs. The 5 mg *symbol* kg sup -1 bolus dose followed by infusions at 25, 50, and 100 mg *symbol* kg sup -1 *symbol* h sup -1 caused reductions in mean arterial pressure that were similar to those observed during the administration of halothane, isoflurane, and sodium nitroprusside in a previous study performed at this laboratory. Although plasma concentrations of propofol were not measured in this investigation, a previous study demonstrated that infusions of 20 and 40 *symbol* mg kg sup -1 *symbol* h sup -1 produced plasma concentrations of 2-13 micro gram *symbol* kg sup -1 in dogs, which are within the anesthetic range in humans. Thus, the 25 and 50 mg *symbol* kg sup -1 *symbol* h sup -1 infusions of propofol used in the current investigation may correlate with clinically relevant propofol concentrations. However, because plasma concentrations of propofol were not specifically obtained, direct comparison of the hemodynamic effects of this agent between the chronically instrumented canine model and humans can be inferred only indirectly.
In summary, the current results demonstrate the effects of propofol on left ventricular afterload quantified using the Windkessel model of Z in (omega) are complex. Propofol affects both arteriolar tone (decreases in R) and aortic mechanical properties (increases in Zcand C) in chronically instrumented dogs. However, propofol does not alter arterial wave reflection patterns determined from Zin(omega). Propofol decreases arterial pressure at least partially through reductions in arterial resistance and increases in arterial compliance.
The authors thank John Tessmer and Dave Schwabe, for technical assistance, and Angela Barnes, for preparation of the manuscript.