| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of Pediatrics, Anesthesiology, and Bioengineering, University of Washington, Seattle, Washington 98195
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Critical intracellular myocardial oxygen tension was determined by optical spectroscopic measurement of myoglobin oxygen saturation in crystalloid-perfused guinea pig hearts. Accurate end-point determinations of the maximally oxygenated and deoxygenated myoglobin were made. Hearts were subjected to a steady decrease in perfusate oxygen tension while left ventricular developed pressure, maximal left ventricular dP/dt, myocardial oxygen consumption, lactate release, and adenosine release were measured as indices of myocardial function. Intracellular myoglobin was found to be only 72% saturated under baseline conditions with an arterial oxygen tension of >600 mmHg at 37°C. Baseline intracellular oxygen tension was 6.3 mmHg. Myocardial oxygen consumption was decreased by 10% when the oxygen tension fell to 5.7 mmHg, and cardiac contraction decreased 10% when oxygen tension was 4.1 mmHg. Adenosine release and, finally, lactate release began to increase at sequentially lower oxygen tensions. The present results indicate that the buffer-perfused guinea pig heart at 37°C has an intracellular oxygen tension just above the threshold for impaired function.
myocardial oxygen consumption; myoglobin oxygen saturation; optical spectroscopy; Langendorff; guinea pig
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
MYOCARDIAL FUNCTION IS DEPENDENT on a constant supply of oxygen delivery from the coronary circulation to the myocytes. An inadequate supply of oxygen to the heart results in rapid deterioration of function. Thus knowledge of the critical intracellular oxygen tension necessary for normal function is of interest. Intracellular myocardial oxygen tension represents the ultimate result of all factors involved in the balance between oxygen supply and oxygen demand. Knowledge of this variable is essential to understanding the energetic status of the myocardium. With the use of the known relationship between oxygen tension and myoglobin saturation (14), intracellular oxygen tension can be calculated for tissues containing myoglobin. Thus the relationship between intracellular oxygen tension and various markers of cellular and organ function can be determined, elucidating the value of the critical oxygen tension.
Cardiac myoglobin oxygen saturation has been determined optically in isolated myocytes (5, 18, 19), in buffer-perfused hearts (6, 9, 10, 17, 21), and in vivo (2, 16). Cardiac deoxymyoglobin has also been determined with nuclear magnetic resonance in buffer-perfused hearts (8, 20) and in vivo (3). All of these previous measurements of cardiac intracellular myoglobin saturation, whether by optical or nuclear magnetic resonance methods, rely on estimates of the fully saturated and fully unsaturated end points in individual hearts. In other words, results are dependent on the fraction of total labile signal obtained during the experiment. The same is true in the present study; however, extra procedures were used to obtain the highly saturated and unsaturated myoglobin end points. The present study was undertaken to determine the relationship between intracellular oxygen tension and cellular and whole organ function in the isolated buffer-perfused guinea pig heart. A reflectance optical spectroscopic approach was used to determine the intracellular oxygen tension from myoglobin oxygen saturation measurements.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Heart Perfusion
Heart preparation. All experiments were performed in accordance with the University of Washington Animal Care Committee regulations. Adult guinea pigs (750-1,000 g) of either sex were injected intraperitoneally with 1,000 units of heparin, and 1 h later were anesthetized by an intraperitoneal injection of pentobarbital (100-125 mg/kg). Hearts were rapidly excised, immersed in ice-cold buffer, and the aorta cannulated for perfusion in the Langendorff manner. A modified Krebs-Henseliet buffer was used for perfusion with the following concentrations (in mM): 118 NaCl, 4.7 KCl, 1.2 MgSO4, 1.2 KH2PO4, 25 NaHCO3, 11 glucose, 1.75 CaCl2, 2.0 pyruvate, and 5 U/l insulin. The pulmonary artery was cannulated to obtain coronary venous measurements.
Perfusion system.
A constant coronary perfusion pressure of 80 cmH2O was
maintained by continuous overflow from an upper buffer reservoir, and temperature at 37°C was maintained by water-jacketed glassware for
buffer reservoirs, gas exchangers, and a heart chamber. Although perfusion of the heart was maintained by constant pressure, oxygen content of the perfusate was controlled by two computer-driven pumps
with variable flow. A dual gas-exchange system allowed for a linear
transition from perfusion with oxygen to nitrogen-equilibrated buffer.
Two glass gas exchangers (13) were used to equilibrate the
buffer with either 95% O2-5% CO2 or 95%
N2-5% CO2. Buffer was recirculated
continuously between each reservoir and its gas exchanger throughout
the experiments, and bypass circuits were utilized to keep the buffer
from stagnating at any point in the perfusion system. Buffer that
passed through the heart was discarded. Figure
1 shows a schematic diagram of the
perfusion apparatus.
|
Oxygen tension measurement. Arterial oxygen tension was continuously measured in a 37°C water-jacketed stirred chamber (volume 0.9 ml) located just proximal to the aorta with a Clark oxygen electrode (dual oxygen analyzer model 203; Instech). The chamber was stirred to minimize the effect of flow on the oxygen tension measurement. A similar water-jacketed and stirred chamber was used to measure the oxygen tension in coronary venous buffer from the cannulated pulmonary artery. Oxygen electrodes were readied for each experiment by rechloriding the electrodes, wetting the electrode tip with half-saturated potassium chloride solution, and placement of new 0.001-in. polyethylene membrane. Electrodes were calibrated to 150 mmHg in water equilibrated with room air at 37°C.
Physiological monitoring. Two 0.003-in. diameter silver wires were placed in the right ventricular epicardium for electrocardiographic monitoring and pacing was achieved by a pacing wire placed in the right ventricular cavity. The electrocardiogram (ECG) signal was continuously monitored from the right ventricular epicardial wires via a ECG amplifier (model ISO-DAM-C, World Precision Instruments). Pacing was achieved using a software algorithm written in LabVIEW (National Instruments), and pacing was set between 180 and 240 beats/min with a 2.5-ms pulse duration and 500 mV amplitude.
The left atrial appendage was opened, and a 4-mm diameter water-filled latex balloon was placed in the left ventricle after excision of the mitral valve leaflets. Left ventricular pressure was continuously measured with a 5.5-Fr catheter-tip pressure transducer (model SPR-783; Millar) inside the balloon and the signal was amplified with a pressure amplifier (model 8805B; Hewlett-Packard). The volume of the balloon was adjusted via the catheter lumen at the beginning of each experiment to give a left ventricular diastolic pressure of 5 to 10 mmHg. Left ventricular systolic and diastolic pressures were selected for each heartbeat when an optical spectrum was acquired by using a computer algorithm. Ventricular function was determined by left ventricular developed pressure (systolic-diastolic) and by maximal left ventricular dP/dt (dP/dtmax) computed by a standard routine in LabVIEW with averaging over each second. Coronary flow was measured with an ultrasonic flow probe (Transonics) in the aortic inflow tubing just above the coronary arteries. The flowmeter was calibrated by timed volume collections of buffer. Myocardial temperature was measured by a myocardial temperature sensor placed in the right ventricle (Mon-a-therm; Mallinckrodt Medical) using a thermocouple-to-analog converter (model TAC80B-T; Omega). All physiological data were acquired by a 12-bit analog-to-digital converter at a sampling rate of 1 kHz (model AT-MIO-16E-10; National Instruments) and displayed on a desktop personal computer (model P133; Gateway) using software developed in LabVIEW. Mean values were determined by averaging over 1-s intervals and recorded for the time period corresponding to each acquired optical spectrum.Spectral Acquisition
Instrumentation. Optical reflectance spectra were acquired using a custom-designed, bifurcated, coaxial optical fiber bundle to carry illuminating and reflected light. Illumination from a constant intensity quartz-tungsten-halogen white light source (model 66184; Oriel Instruments) was passed through a 1.0-in. water filter and electromechanical shutter (model 76995; Oriel Instruments) before transmission to the optical probe to decrease tissue heating. The probe consisted of a 0.75-mm-thick illuminating ring of fibers surrounding a central core of detecting fibers 1.75 mm in diameter in a bull's eye configuration. The detecting and illuminating fibers were separated by 1.25 mm, resulting in an estimated average depth of tissue penetration of 0.6 mm. The probe was placed in contact with the free wall of the left ventricle. Constant light intensity was ensured by a photofeedback system (model 68850; Oriel Instruments). Spectra from 450 to 950 nm were acquired via a diffraction spectrograph (model 100S; American Holographics) with a 512-pixel photodiode array (model C4350; Hamamatsu), using a 50-ms exposure time. Spectral data were converted into digital form using a 16-bit analog-to-digital converter (model AT-MIO-16X; National Instruments) and entered into a separate desktop computer (66 MHz, Gateway), where results were acquired and displayed by software written in Microsoft C/C++. Spectral acquisition was gated to diastole for each eighth heartbeat, to acquire data at approximately 2-s intervals to avoid excessively large data files.
End-point determinations. Maximal myoglobin oxygen saturation was produced experimentally in the heart by perfusion with oxygenated buffer and by infusing adenosine to maximally vasodilate the coronary arteries and potassium to arrest the heart and thus lower myocardial oxygen consumption. The excess flow and diminished oxygen consumption has the effect of augmenting the cytosolic oxygen content and thus increasing the myoglobin oxygen saturation. This was accomplished by infusing a solution containing 10 mM adenosine and 150 mM KCl in buffer into the perfusion system just proximal to the flow probe. The infusion rate was maintained at ~10% of the coronary flow so that a final perfusate concentration of 1.0 mM adenosine and 15 mM K+ was achieved in the perfusate entering the heart. Maximal deoxygenation was produced by infusion of sodium dithionite (Na2S2O4) at the end of each experiment. A solution of 0.5 M sodium dithionite in buffer was infused into the perfusate as described above so that a final concentration of 50 mM was achieved. Sodium dithionite is a reducing agent that donates one or more electrons; it rapidly and completely deoxygenates aqueous solutions to which it is added. Spectra were acquired during both infusions.
Experimental Protocol
After hearts had reached a functional steady state (~20 min), baseline venous effluent samples were obtained for lactate and adenosine measurement. Hearts demonstrating a baseline systolic pressure
80 mmHg and a developed maximal dP/dt of 1,000 mmHg/s were deemed acceptable preparations. A steady linear decrease in
buffer oxygen content was begun by simultaneously ramping up the flow
of deoxygenated buffer and ramping down the flow of oxygenated buffer
using a pump control routine written in LabVIEW. A constant perfusion
pressure was maintained by ensuring that the total flow from the two
pumps exceeded the coronary flow at all times, with excess flow
overflowing from the upper reservoir. Desaturation was performed over
20 min, and venous effluent samples were obtained every 2 min for
lactate and adenosine measurements. Baseline values for each measured
variable were determined by averaging the values corresponding to the
50 spectra acquired just before beginning the desaturation.
| |
ANALYSIS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Myoglobin saturation. Myoglobin oxygen saturation was determined by partial least squares analysis using reference spectra obtained from in vitro solutions of myoglobin and cytochrome c in scattering media as described in earlier work (15, 16). Briefly, the near-infrared wavelength region from 600 to 850 nm was used for these analyses. Spectra were preprocessed by taking second derivatives with respect to wavelength to reduce baseline offsets. The partial least squares analysis is based on interpretation of unknown spectra using spectra obtained under controlled conditions (calibration spectra). From the in vitro calibration spectra, the partial least-squares analysis can be used to predict other test spectra under known conditions to determine the predictability and sensitivity of the method. Partial least- squares analysis can be used to accurately predict concentrations of an analyte of interest (in this case, the saturation of myoglobin) from complex spectra taken from solutions (or tissue) in which multiple absorbing species with overlapping spectral features exist. The accuracy of the prediction depends, in part, on how well the calibration spectra represent the sample of interest. Previous studies have demonstrated successful prediction of myoglobin saturation using this method (15, 16, 17).
Two sets of 300 spectra each were generated and used for calibration and test sets. These spectra were obtained from solutions of cytochrome c and myoglobin in various concentrations of Intralipid, as previously described (15). Cytochrome c has spectral features that overlap those of myoglobin, although its concentration in heart tissue is less than that of myoglobin. Inclusion of cytochrome c in the calibration spectral data set improves the prediction accuracy. Subsequently, the calibration spectra were used to predict the myoglobin saturation from unknown reflectance spectra from the heart. Resultant fractional saturation values were scaled between 0 and 0.995 using end points determined experimentally as described above. A maximal myoglobin fractional saturation of 0.995 was chosen because this is the saturation of myoglobin for an oxygen tension of 452 mmHg, which is the average maximal venous oxygen tension determined experimentally. This value represents the maximal plausible intracellular oxygen tension that will produce conservative maximal values in the calculated critical oxygen tensions (DISCUSSION).Prediction threshold. The two sets of spectra obtained for the calibration and test set included components of myoglobin and cytochrome c. Each composite spectrum included weighted amounts of oxymyoglobin and deoxymyoglobin spectra such that the total myoglobin spectral contribution remained unity. The fraction of oxymyoglobin (or %saturation) in each composite spectrum was thus known. Partial least-squares analysis was used with one set of spectra (calibration set) to determine a set of calibration coefficients from which the myoglobin oxygen saturation of the other set of spectra (test set) could be predicted. Because the myoglobin oxygen saturation was known for all spectra, the predicted saturation could be compared with the known saturation. Thus for the in vitro solutions, a detection threshold could be determined.
Intracellular oxygen tension.
Intracellular oxygen tension was determined from the myoglobin oxygen
saturation measured from the optical spectra taken from the heart.
Intracellular oxygen tension was determined from the saturation values
using the Hill equation and the relationship between the
PO2 at which myoglobin is half-saturated with
O2 (P50) and temperature. The Hill equation in
simplified form when the Hill exponent is unity (14) is
|
(1) |
|
(2) |
Myocardial oxygen consumption.
Myocardial oxygen consumption was calculated from the arteriovenous
oxygen content difference multiplied by flow, using Eq. 3
|
(3) |
O2 is myocardial
oxygen consumption in milliliters per minute per gram, F is the flow
per gram of heart weight (wet), and
PaO2 and
1 · atm1 (1) at
37°C. Because coronary venous oxygen tension was measured downstream
from the heart, measured venous oxygen tension values lagged slightly.
However, the pulmonary artery catheter and mixing chamber have a volume
of 1 ml. At a flow of 15 ml/min (mean baseline flow) this corresponds
to a delay of 4 s. During deoxygenation, flow increased, thus
decreasing the lag time.
Adenosine analysis. Coronary venous effluent samples were divided into 2 aliquots for adenosine and lactate analysis. Approximately 1-ml aliquots were placed in test tubes containing 25 µl adenosine deaminase inhibitor erythro-9-(2-hydroxy-3-nonyl)adenine (10 µM) at specified time points. Samples were heated in a boiling water bath for 1 min to halt enzyme activity and then kept on ice or frozen until analyzed. Before analysis, samples were centrifuged at 5,000 g for 10 min to remove precipitated salts.
The adenosine in each sample was separated isocratically at 0.5 ml/min on a Beckman model 126 HPLC with a C-18 column (5 µm, LiChrospher 100 RP-18; Hewlett-Packard) using a mobile phase of 8 mM KPO4 containing 3.5% methanol. Adenosine was detected using a Beckman model 206 single wavelength detector at 260 nm. Sample adenosine was identified and quantified by comparison of retention time and peak to adenosine standards (7).Lactate analysis. Aliquots of 0.5 ml of effluent were placed in heparinized lithium/sodium fluoride tubes (Brinkman Instruments), and placed on ice. Lactate samples were analyzed immediately after the completion of the desaturation protocol on a lactate analyzer (Yellow Springs Instruments).
Statistics and curve fitting. Physiological values were determined from the waveforms recorded by the LabVIEW program for the same time points that the optical spectra were acquired. Values for each variable sampled were averaged over 1-s time periods. Curve fitting was performed using standard routines from Origin (Microcal). Curve fitting was performed for comparison within and between data sets, using empiric fitting. Where sufficient data were present, curve fitting was performed for each heart. Data from each heart were fit using the empiric relationships as described below. The resultant fit curves were described by 1,000 points and averaged point by point. Averaged curves are displayed along with the original data. Equations for the averaged curves were determined by fitting the averaged curve points using the same fitting routine as in the original curve fitting. For lactate and adenosine analyses, fewer data points were available. In these cases, curve fitting was performed using all data from all seven hearts.
Left ventricular developed pressure, dP/dtmax, and myocardial oxygen consumption were normalized with baseline measurements given a value of one for each experiment for comparison, and then curve fit to intracellular oxygen tension using the sigmoid Bolzman equation of general form
![y=(A<SUB>1</SUB>−A<SUB>2</SUB>)/[1+e<SUP>(x−x<SUB>0</SUB>)/dx</SUP>]+A<SUB>2</SUB>](/content/vol281/issue6/fulltext/H2463/img005.gif)
|
(4) |
|
(5) |
|
(6) |
|
(7) |
|
(8) |
|
(9) |
,
y 1.
Critical oxygen tension. Critical intracellular oxygen tensions were determined for both a 10 and 20% change in function from baseline for each variable. These definitions of critical intracellular oxygen tension are arbitrary but allow for comparison with previous studies. For left ventricular developed pressure, left ventricular dP/dtmax, and myocardial oxygen consumption, an individual critical intracellular oxygen tension was determined from each heart, using an average of the three data points closest to the 90 or 80% function value (one point above and one point below). Because hundreds of experimental data points were available for analysis (Fig. 6), a local estimate of the critical oxygen tension values was used in preference to an estimate based on an arbitrary curve fit. Individual critical values for adenosine and lactate were determined by curve fitting all data from each heart, because fewer data points were available for analysis. Values are expressed as means ± SD unless otherwise indicated.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Myoglobin Saturation
Partial least-squares analysis was used to predict the myoglobin oxygen saturation for the 300 test spectra using 300 distinct calibration spectra. Figure 2 shows the predicted myoglobin fractional saturation of the test spectra using the partial least squares method. A high correlation with the known saturation values can be seen. The standard deviation of the residuals is 0.023, suggesting a detection threshold of ~2.3% for the myoglobin saturation measurements.
|
Baseline conditions are shown in Table 1,
and represent the mean value ± SD for each variable from the
seven hearts studied. Coronary flow increased during deoxygenation to a
maximum value of 14.3 ± 2.7 ml · min1 · g1. A similarly
increased flow (13.7 ± 3.5 ml · min1 · g1) was noted
during adenosine and KCl infusion used to determine the maximum
myoglobin oxygenation.
|
Figure 3 shows the myoglobin oxygen
saturation for a single deoxygenation experiment using the end points
defined as above. Average baseline myoglobin saturation during
perfusion with 95% O2-5% CO2 equilibrated
buffer at the start of the experimental protocol was only 72 ± 7 (n = 7). Myoglobin saturation was found to decrease
almost immediately as the perfusate oxygen tension was lowered (Figs. 3
and 4).
|
|
Figure 4 shows myoglobin saturation as a function of arterial and venous oxygen tension. A Bolzman sigmoidal equation (Eq. 4) was used for fitting myoglobin saturation to arterial oxygen tension, and a hyperbolic function (Eq. 5) was used in fitting myoglobin saturation to venous oxygen tension.
Intracellular Oxygen Tension
Initial intracellular oxygen tension was 6.3 ± 2.6 mmHg. Intracellular oxygen tension decreased as expected with decreasing arterial oxygen tension and thus decreasing venous oxygen tension as shown in Fig. 5, A and B. Halting the deoxygenation during a control experiment resulted in a prompt (<5 s) leveling off of intracellular oxygen tension, demonstrating that the deoxygenation ramp was sufficiently slow.
|
A second-order polynomial curve fit (Eq. 7) represents the relationship between arterial and intracellular oxygen tension as shown in Fig. 5A (r2 = 0.72). There is a linear relationship between venous and intracellular oxygen tension (r2 = 0.96), except below a venous oxygen tension of ~15 mmHg where intracellular oxygen tension falls steeply (Fig. 5B). The mechanism underlying the drop in intracellular oxygen tension at low venous oxygen tension is unclear.
Ventricular Function
Left ventricular developed pressure decreased with decreasing intracellular oxygen tension. Figure 6A shows data from all of the desaturation experiments, with empiric curve fitting using the Bolzman sigmoidal function. To facilitate comparison between experiments, pressures for each experiment were normalized between 0 and 1. Critical oxygen tensions determined from individual values for both a 10 and 20% decrease in function are shown in Table 2.
|
|
Left ventricular dP/dtmax was determined for the heartbeats corresponding to each acquired spectrum. The mean baseline maximal dP/dtmax was 1,250 ± 150 mmHg/s before the start of the deoxygenation. Figure 6B shows normalized dP/dtmax as a function of intracellular oxygen tension. Critical values for both a 10 and 20% decrease in function are given in Table 2.
Control experiments have been previously performed under equivalent perfusion conditions without hypoxia. Heart systolic pressure and dP/dtmax showed no significant change after 2 h of unperturbed oxygenated perfusion, although myoglobin saturation was found to decrease by 2% (17).
Myocardial Oxygen Consumption
The relationship between normalized myocardial oxygen consumption and intracellular oxygen tension is shown in Fig. 6C. Myocardial oxygen consumption decreased steadily as intracellular oxygen tension decreased. Some clustering of the data can be seen at high intracellular oxygen tension; however, the underlying mechanism is unclear.Adenosine Production
Venous effluent adenosine concentration was also determined as a function of intracellular oxygen tension. These results are illustrated in Fig. 7A. An exponential function was used to fit these data.
|
Lactate Production
Venous lactate increased from a baseline value of ~1.2 mM as oxygen tension was decreased. Lactate production as a function of intracellular oxygen tension increased in an exponential manner with decreasing oxygen tension. Figure 7B shows lactate concentration in the venous effluent as a function of intracellular oxygen tension.Critical Oxygen Tension
For purposes of comparison, complements of the fit curves for adenosine and lactate concentrations were plotted as a function of intracellular oxygen tension. The resultant curves are shown with the other variables in Fig. 8. From this figure it can be appreciated that each of the measures of myocardial function appears to be relatively independent of intracellular oxygen tension over the range of oxygen tensions from 6 to 12 mmHg. At oxygen tensions below this range, myocardial function changes more steeply as the intracellular oxygen tension approaches zero. With decreasing intracellular oxygen tension, myocardial oxygen consumption declined first, followed in order by left ventricular developed pressure, left ventricular dP/dtmax followed by an increase in adenosine release and lactate release.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The most striking finding in the present study is that cardiac myoglobin is only 72% saturated with an intracellular oxygen tension of 6.3 mmHg in hearts perfused with buffer having an oxygen tension of over 600 mmHg. This is in contrast to the usual assumption that myoglobin is fully saturated when an isolated heart is perfused with buffer equilibrated with 95% oxygen. An intracellular oxygen tension this low is surprising and demonstrates a large drop from the vascular oxygen tension. In a normal blood-perfused heart with an arterial oxygen tension of 90 mmHg, the maximum theoretical vascular-to-cellular oxygen tension difference is 90 mmHg. However, in the present buffer-perfused experiments, the comparable number is ~637 (643-646, Table 1) mmHg. Why this is so remains unclear.
Recent advances in both optical and magnetic resonance spectroscopic techniques have made determination of myocardial intracellular critical oxygen tension possible. However, both techniques depend on a correct measurement of myoglobin oxygen saturation. Specifically, these spectral analyses depend on having well-defined, fully saturated (oxygenated), and fully unsaturated (deoxygenated) myoglobin end points. Precise determination of these end points has proved to be elusive. Magnetic resonance approaches are also hindered by an uncertain detection threshold for deoxygenated myoglobin. The present study differs from prior reports by procedures to optimize these critical end points.
End-Point Determinations
Determinations of the fully saturated and fully unsaturated end points are critical for accurate estimations of myoglobin oxygen saturation. Myoglobin has been commonly assumed to be fully saturated in the crystalloid-perfused heart with an arterial buffer oxygen tension of ~600 mmHg (4, 8). However, some investigators have indicated that myoglobin is not fully saturated (6, 10). Because both optical and magnetic resonance spectral approaches are dependent on the myoglobin saturation end points, incorrect assumptions about these end points result in erroneous estimations of intracellular oxygen tension.If the baseline intracellular oxygen tension were assumed to be equal to the baseline coronary venous oxygen tension of 105 mmHg, then the myoglobin oxygen saturation would have been 0.987. As the present results demonstrate (Figs. 3 and 4, Table 1), the measured baseline myoglobin oxygen saturation was 72 ± 7% corresponding to an intracellular oxygen tension of 6.3 ± 2.6 mmHg. Further supporting the observation that myoglobin is not fully saturated at baseline is the relationship between myoglobin saturation (and hence intracellular oxygen tension) and arterial buffer oxygen tension. In all cases, there was an immediate decrease in myoglobin saturation as arterial oxygen tension was decreased. If fully oxygenated buffer provided an excess of oxygen, an initial plateau would have been expected in the relationship between myoglobin saturation and arterial oxygen tension, but this was not observed (Fig. 4A).
A reliable, fully deoxygenated end point is also needed for both optical and magnetic resonance methods. With in vivo studies coronary artery occlusion has been used to obtain a deoxygenated end point (2, 3, 16). The method for determining the deoxygenated end point in the buffer-perfused heart in previous studies (4, 8) is not explicitly stated but presumably, the endpoint was achieved by stopping the inflow. Stopping coronary flow to obtain the desaturated myoglobin end point depends on continued myocardial oxygen consumption to deplete the oxygen tension down to zero. Because myocardial oxygen consumption begins to decline when the intracellular oxygen tension is above 5 mmHg (Table 2), the dithionite treatment used in the present experiments may give a more reliable end point than stopping coronary flow.
In determining the myoglobin saturation values, an assumption must be made regarding the maximal intracellular oxygen tension attainable during infusion of adenosine and KCl. In METHODS it is argued that the maximal plausible intracellular oxygen tension under these conditions is that of the venous oxygen tension (452 mmHg). Thus the myoglobin saturation values determined were scaled between 0 and 0.995 (or essentially 0 and 1). This assumption results in conservative estimates of the critical intracellular oxygen tensions. If the maximal intracellular oxygen tension is, in fact, lower than the venous oxygen tension, than the calculations presented here overestimate the myoglobin saturation, and thus overestimate the intracellular oxygen tension. If the true maximal intracellular oxygen tension is only 20% of the venous oxygen tension (90 mmHg), then the maximal myoglobin saturation with adenosine and KCl infusion is 0.974. If the myoglobin saturations determined are scaled between 0 and 0.974, then the values presented in Table 2 would be decreased by a maximum of 9%. For example, the critical intracellular oxygen tension for myocardial oxygen tension at 90% maximal would change from 5.7, as reported in Table 2, to 5.2 mmHg.
The present findings demonstrate that in the crystalloid-perfused heart, myoglobin saturation (and hence intracellular oxygen tension) is directly related to the perfusate oxygen tension. From Figs. 4 and 5 it can be appreciated that as arterial oxygen tension decreases, myoglobin saturation and intracellular oxygen tension likewise decrease, even at high arterial oxygen tensions. The change in intracellular oxygen tension with venous oxygen tension can be described as a linear relationship, especially at venous oxygen tensions above ~25 mmHg.
Detection Thresholds
The detection threshold for oxymyoglobin/myoglobin saturation with the present optical spectroscopic method is about 2%, as illustrated by Fig. 2. The detection threshold for deoxymyoglobin by nuclear magnetic resonance has not been published. Various reports assume a 10% (12) or 20% (11) deoxymyoglobin detection threshold for magnetic resonance. If the baseline conditions of the present study are comparable to the baseline conditions of the magnetic resonance experiments where no deoxymyoglobin was detected in buffer-perfused hearts at 25°C (8) or 37°C (4), then the present results imply that the magnetic resonance detection threshold is >20%, because baseline myoglobin saturation was 72% (and thus 28% deoxymyoglobin) in the present study.Critical Intracellular Oxygen Tension
Critical intracellular oxygen tension can be defined by various indices of myocardial function, and can be arbitrarily defined for various degrees of change in these functions from baseline. There is no a priori reason to believe that a single critical intracellular oxygen tension would describe all indices of myocardial function. In the current study, several indices of myocardial function were chosen to represent both global organ and cellular level function. Two levels of change in function from baseline were chosen to calculate a critical threshold for each index of function. Both 10 and 20% changes in function from baseline have been used in earlier reports, although these percent changes are used arbitrarily. Global organ function was represented in the current study by measurements of left ventricular developed pressure and dP/dtmax. Cellular function was represented by myocardial oxygen consumption, adenosine, and lactate release.Although myoglobin saturation in the buffer-perfused heart appears to be limited by oxygen delivery at the full range of arterial oxygen tensions studied, ventricular and cellular function are not impaired until intracellular oxygen tension falls below an oxygen tension of ~5 mmHg. Using the rationale described above, critical intracellular oxygen tensions were determined to range from 2.8 to 5.7 to mmHg using a 10% change in function from baseline. With the threshold set at a 20% change in function, values ranged from 1.2 to 4.1 mmHg as shown in Table 2.
Comparison with Reported Values
Previous work using a magnetic resonance spectral approach has determined critical oxygen tensions defined as a 20% change in function for the isolated perfused rat heart at 25°C ranging from 0.8 to 4.4 mmHg for various functional indices (8). In particular, a critical oxygen tension of 4.0 mmHg for the rate-pressure product (RPP), and 2.1 mmHg for myocardial oxygen consumption was reported. In the present study, left ventricular developed pressure is proportional to RPP, because hearts in the present study were paced at a constant rate. In the present study, at 37°C the critical oxygen tension at 80% for left ventricular developed pressure was 3.1 mmHg, which is lower than the value of 4.0 mmHg reported in the magnetic resonance study as reported by Kreutzer and Jue (8). There is a greater difference between the critical oxygen tension of 2.1 mmHg reported for myocardial oxygen consumption in the magnetic resonance study and 4.1 mmHg found in the present study. However, Kreutzer and Jue (8) used a myoglobin P50 of 1.5 mmHg that is about double the P50 value of 0.78 mmHg at 25°C found in a recent study (14).How much of the difference between the present results and those from the magnetic resonance experiments is due to errors in the measurements vs. differences in temperature and P50 is unclear. Certainly, both oxygen delivery and oxygen consumption may be different for hearts at different temperatures. Oxygen solubility of crystalloid buffer is lower at 37°C than at 25°C, but the difference is only ~16% (1). Thus only some of the difference in myoglobin saturation can be explained on the basis of oxygen solubility.
In summary, the present results do not confirm the assumption that cardiac myoglobin is fully saturated when perfused with buffer having an oxygen tension over 600 mmHg at 37°C. Thus oxygen delivery from the perfusate limits myoglobin saturation in the buffer-perfused heart. Baseline myoglobin saturation was 72% and intracellular oxygen tension was 6.3 mmHg. Myocardial oxygen consumption was decreased by 10% when the oxygen tension fell to 5.7 mmHg, and cardiac contraction decreased 10% when oxygen tension was 4.1 mmHg. Adenosine release and finally, lactate release began to increase at sequentially lower oxygen tensions. The present results indicate that the buffer-perfused guinea pig heart at 37°C has an intracellular oxygen tension just above the threshold for impaired function.
| |
ACKNOWLEDGEMENTS |
|---|
The author thanks John David Tune and Mark W. Gorman for help with the adenosine analyses, and Eric O. Feigl for a critical reading of the manuscript. Wayne A. Ciesielski provided expert technical assistance in all phases of the research.
| |
FOOTNOTES |
|---|
This work was supported by American Heart Association Grant 9960231z.
Address for reprint requests and other correspondence: K. A. Schenkman, Anesthesia and Critical Care, CH-05, Children's Hospital and Regional Medical Center, 4800 Sand Point Way NE, Seattle, WA 98105 (E-mail: kschen{at}chmc.org).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 24 November 2000; accepted in final form 6 August 2001.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
ANALYSIS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Altman, PL,
and
Dittmer DS.
Respiration and Circulation. Bethesda, MD: Federation of American Societies for Experimental Biology, 1971, p. 18.
2.
Arai, AE,
Kasserra CE,
Territo PR,
Gandjbakhche AH,
and
Balaban RS.
Myocardial oxygenation in vivo: optical spectroscopy of cytoplasmic myoglobin and mitochondrial cytochromes.
Am J Physiol Heart Circ Physiol
277:
H683-H697,
1999
3.
Bache, RJ,
Zhang J,
Murakami Y,
Zhang Y,
Cho YK,
Merkle H,
Gong GF, AH,
and
Ugurbil K.
Myocardial oxygenation at high workstates in hearts with left ventricular hypertrophy.
Cardiovasc Res
42:
616-626,
1999
4.
Chung, Y,
and
Jue T.
Regulation of respiration in myocardium in the transient and steady state.
Am J Physiol Heart Circ Physiol
277:
H1410-H1417,
1999
5.
Gayeski, TEJ,
and
Honig CR.
Intracellular PO2 in individual cardiac myocytes in dogs, cats, rabbits, ferrets, and rats.
Am J Physiol Heart Circ Physiol
260:
H522-H531,
1991
6.
Heineman, FW,
Kupriyanov VV,
Marshall R,
Fralix TA,
and
Balaban RS.
Myocardial oxygenation in the isolated working rabbit heart as a function of work.
Am J Physiol Heart Circ Physiol
262:
H255-H267,
1992
7.
Herrmann, SC,
and
Feigl EO.
Subtraction method for the high-performance liquid chromatographic measurement of plasma adenosine.
J Chromatogr A
574:
247-253,
1992.
8.
Kreutzer, U,
and
Jue T.
Critical intracellular O2 in myocardium as determined by 1H nuclear magnetic resonance signal of myoglobin.
Am J Physiol Heart Circ Physiol
268:
H1675-H1681,
1995
9.
Leisey, JR,
Scott DA,
Grotyohann LW,
and
Scaduto RC, Jr.
Quantitation of myoglobin saturation in the perfused heart using myoglobin as an optical inner filter.
Am J Physiol Heart Circ Physiol
267:
H645-H653,
1994
10.
Makino, N,
Kanaide H,
Yoshimura R,
and
Nakamura M.
Myoglobin oxygenation remains constant during the cardiac cycle.
Am J Physiol Heart Circ Physiol
245:
H237-H243,
1983.
11.
Mole, PA,
Chung Y,
Tran TK,
Sailasuta N,
Hurd R,
and
Jue T.
Myoglobin desaturation with exercise intensity in human gastrocnemius muscle.
Am J Physiol Regulatory Integrative Comp Physiol
277:
R173-R180,
1999
12.
Murakami, Y,
Zhang Y,
Cho YK,
Mansoor AM,
Chung JK,
Chu C,
Francis G,
Ugurbil K,
Bache RJ,
From AH,
Jerosch-Herold M,
Wilke N,
and
Zhang J.
Myocardial oxygenation during high work states in hearts with postinfarction remodeling.
Circulation
99:
942-948,
1999
13.
Neely, JR,
and
Rovetto MJ.
Techniques for perfusing isolated rat hearts.
Methods Enzymol
39:
43-60,
1975[Medline].
14.
Schenkman, KA,
Marble DR,
Burns DH,
and
Feigl EO.
Myoglobin oxygen dissociation by multiwavelength spectroscopy.
J Appl Physiol
82:
86-92,
1997
15.
Schenkman, KA,
Marble DR,
Feigl EO,
and
Burns DH.
Near-Infrared spectroscopic measurement of myoglobin oxygen saturation in the presence of hemoglobin using partial least-squares analysis.
Appl Spectroscopy
53:
325-331,
1999.
16.
Schenkman, KA,
Marble DR,
Feigl EO,
and
Burns DH.
Optical spectroscopic method for in vivo measurement of cardiac myoglobin oxygen saturation.
Appl Spectroscopy
53:
332-338,
1999.
17.
Schenkman, KA,
and
Yan S.
Propofol impairment of mitochondrial respiration in isolated perfused guinea pig hearts determined by reflectance spectroscopy.
Crit Care Med
28:
172-177,
2000[ISI][Medline].
18.
Takahashi, E,
and
Doi K.
Visualization of oxygen level inside a single cardiac myocyte.
Am J Physiol Heart Circ Physiol
268:
H2561-H2568,
1995
19.
Takahashi, E,
Sato K,
Endoh H,
Xu ZL,
and
Doi K.
Direct observation of radial intracellular PO2 gradients in a single cardiomyocyte of the rat.
Am J Physiol Heart Circ Physiol
275:
H225-H233,
1998
20.
Zhang, J,
Murakami YZ, Y,
Cho YK,
Ye Y,
Gong G,
Bache RJ,
Ugurbil K,
and
From AH.
Oxygen delivery does not limit cardiac performance during high work states.
Am J Physiol Heart Circ Physiol
277:
H50-H57,
1999
21.
Zundorf, J,
Tauschek D,
Frank K,
Ito K,
Nioka S,
Kessler M,
and
Chance B.
Monitoring of redox-state of respiratory enzymes and myoglobin oxygenation in the working rat heart in normoxia and oxygen deficiency.
Adv Exp Med Biol
317:
583-592,
1992[Medline].
This article has been cited by other articles:
|
|
 |
|
  E. Takahashi, T. Takano, Y. Nomura, S. Okano, O. Nakajima, and M. Sato In vivo oxygen imaging using green fluorescent protein Am J Physiol Cell Physiol, October 1, 2006; 291(4): C781 - C787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Chung, S.-J. Huang, A. Glabe, and T. Jue Implication of CO inactivation on myoglobin function Am J Physiol Cell Physiol, June 1, 2006; 290(6): C1616 - C1624. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Widen and C. J. Barclay ATP splitting by half the cross-bridges can explain the twitch energetics of mouse papillary muscle J. Physiol., May 15, 2006; 573(1): 5 - 15. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Chung Oxygen reperfusion is limited in the postischemic hypertrophic myocardium Am J Physiol Heart Circ Physiol, May 1, 2006; 290(5): H2075 - H2084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Marcinek, K. A. Schenkman, W. A. Ciesielski, D. Lee, and K. E. Conley Reduced mitochondrial coupling in vivo alters cellular energetics in aged mouse skeletal muscle J. Physiol., December 1, 2005; 569(2): 467 - 473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Marin and K. G. Franchini Reduced oxygen supply explains the negative force-frequency relation and the positive inotropic effect of adenosine in buffer-perfused hearts Am J Physiol Heart Circ Physiol, July 1, 2005; 289(1): H131 - H136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Ejike, L. S. L. Arakaki, D. A. Beard, W. A. Ciesielski, E. O. Feigl, and K. A. Schenkman Myocardial oxygenation and adenosine release in isolated guinea pig hearts during changes in contractility Am J Physiol Heart Circ Physiol, May 1, 2005; 288(5): H2062 - H2067. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Matsumoto, H. Tachibana, T. Asano, M. Takemoto, Y. Ogasawara, K. Umetani, and F. Kajiya Pattern differences between distributions of microregional myocardial flows in crystalloid- and blood-perfused rat hearts Am J Physiol Heart Circ Physiol, April 1, 2004; 286(4): H1331 - H1338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Marcinek, K. A. Schenkman, W. A. Ciesielski, and K. E. Conley Mitochondrial coupling in vivo in mouse skeletal muscle Am J Physiol Cell Physiol, February 1, 2004; 286(2): C457 - C463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Schenkman, D. A. Beard, W. A. Ciesielski, and E. O. Feigl Comparison of buffer and red blood cell perfusion of guinea pig heart oxygenation Am J Physiol Heart Circ Physiol, November 1, 2003; 285(5): H1819 - H1825. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Beard, K. A. Schenkman, and E. O. Feigl Myocardial oxygenation in isolated hearts predicted by an anatomically realistic microvascular transport model Am J Physiol Heart Circ Physiol, November 1, 2003; 285(5): H1826 - H1836. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
