Electrical Resistances of Interstitial and Microvascular Space as Determinants of the Extracellular Electrical Field and Velocity of Propagation in Ventricular Myocardium
Background In myocardial ventricular tissue, extracellular electrical resistance (ro) is an important determinant of propagation velocity (Θ) and the magnitude of the extracellular bipolar electrogram (ΔVo). The extracellular space is composed of two compartments, the vascular space and the interstitial space. To assess the electrical equivalent of this compartmentation in the ventricular myocardium and its effect on ro, Θ, and ΔVo, electrical cable analysis was performed in an arterially perfused rabbit papillary muscle.
Methods and Results Vascular resistivity was changed from 75 to 86 to 143 and to 221 Ω/cm by variation of hematocrit in the perfusate from 0% to 10% to 40% and to 60%. As a means to vary the volume of the interstitial space and with this as its resistivity, colloid osmotic pressure (COP) in the perfusate was changed from 9 to 36 and to 94 mm Hg by altering the dextran concentration in the perfusate from 10 to 40 to 80 g/L. Decreasing COP had a marked effect on ro (56% decrease), ΔVo (decrease from 61 to 42 mV), Θ (increase from 48 to 59 cm/s), and the diameter of the muscle fiber (increase of 12%). If COP was increased from 36 to 94 mm Hg, ro (by 35%) and ΔVo (from 62 to 75 mV) increased; Θ and diameter showed no significant changes. In contrast, alterations of intravascular electrical resistivity in a range from 75 to 221 Ω/cm did not induce any significant changes in ro, ΔVo, Θ, and diameter of the preparations.
Conclusions We conclude from our data that (1) the microvascular tree in ventricular myocardium is electrically insulated to a large degree from the interstitial space and that (2) electrical current flow in the extracellular space during excitation is confined to the narrow, anisotropic interstitial space.
Myocardial ventricular tissue is composed of discrete cells interconnected by low-resistance gap junctions1 2 and an extracellular space that hosts the dense microvascular network.3 Morphometric measurements indicate that the extracellular compartment occupies approximately 20% to 25% of tissue volume3 4 and that within this compartment, the microvessels occupy a large volume fraction. Only approximately 5% to 10% of total tissue volume is taken by the interstitial space, which appears as narrow, irregular clefts in histological sections.5 6 7
Functionally, myocardial tissue can be represented by a “reaction-diffusion system” in which an impulse (reaction), generated by ionic current flow through activated membrane channels, propagates (diffuses) through an electrical medium. The electrical properties of this medium are a matter of ongoing investigation, and inhomogeneities in electrical resistance are likely to be present in both the intracellular and extracellular compartments. In the intracellular compartment, inhomogeneities of propagation due to the anisotropic cellular architecture occur during propagation in both transverse and longitudinal directions.8 9 10 11 Many of the simulations used for the analysis of propagation in normal and pathological settings involve a so-called single-domain model of cardiac tissue where intracellular and extracellular compartment resistances are added to form a single resistor. Theoretical12 and experimental13 studies have stressed the importance of the resistive properties of the extracellular space, ie, of the bidomain nature of cardiac tissue. Comparison of the amplitude of the extracellular wave front with the action potential in dog hearts14 as well as linear cable analysis in isolated arterially perfused tissue13 15 have shown that the lumped electrical resistances of the extracellular and intracellular compartments are of approximately equal magnitude. Therefore, changes in extracellular and intracellular resistivities are predicted to affect propagation velocity to an equal degree.
The resistive properties of the intracellular compartment have been analyzed in a large number of studies, mostly involving the effect of gap junctional resistance on electrical interaction between cells. By contrast, no information is available at present on the relative contributions of the components of the extracellular compartments, the interstitial and the vascular compartments, to the electrical extracellular resistance. Accordingly, it was the purpose of this work to assess the multicompartment nature of extracellular myocardial resistance. To this aim, we used the arterially perfused rabbit papillary muscle preparation,13 16 which allows for the simultaneous measurements of the electrical resistances in extracellular and intracellular compartments, conduction velocity, and the amplitude of the extracellular electrical field. Moreover, both the resistive properties of the intravascular and the interstitial spaces can be modified in this preparation independently via changes in the composition of the perfusate.
Our results suggest that the microvascular tree in ventricular myocardium is electrically insulated from the interstitial space to a large degree. Consequently, extracellular electrical current flow during excitation is mostly confined to the narrow, interstitial clefts.
Preparation and Perfusion of Papillary Muscles
The method for the preparation of the isolated, arterially perfused rabbit papillary muscle has been described previously in detail.13 16 Rabbits of either sex weighing 2 to 3 kg were anticoagulated with heparin (200 IU/kg IV) and anesthetized with pentothal (50 mg/kg IV). The hearts were rapidly excised, placed in cold (4°C) Tyrode’s solution, and transported to a dissection tray. The atria, the left ventricular wall, and the nonperfused part of the right ventricle were removed. The tissue was mounted with the left ventricular septal surface on a wax platform that contained the ground electrode (Fig 1⇓). The septal artery (diameter, 150 to 200 μm) was cannulated and perfused with the following solution (mmol/L): Na+ 149, K+ 4.5, Ca2+ 1.8, Mg2+ 0.49, Cl− 133, HPO42−/H2PO4− 0.4, HCO3− 25, glucose 20 and insulin 1 IU/L, heparin 400 IU/L, albumin 4 g/L, and dextran (Mr, 70 000) 40 g/L. In some experiments, 10% newborn calf serum (BioWhittaker) was added to the perfusate (see “Results”). The total ischemic time measured from the removal of the heart to cannulation was less than 4.5 minutes.
After cannulation, the preparation was placed in the recording chamber, maintained at 37°C, and perfused with the solution previously described and washed bovine erythrocytes (hematocrit, 40%). A roller pump (Ismatec Instruments) was used to maintain a perfusion pressure of 35 to 45 mm Hg by adjustment of the perfusion flow rate (80 to 100 mL/min per 100 g tissue). Perfusion pressure in the septal artery was measured with a pressure transducer (P23 ID, Gould). This perfusion pressure is normal for arteries with a diameter of 120 to 160 μm because in the rabbit, about 40% to 50% of the peripheral coronary resistance is located in vessels with a diameter >150 μm.17 The blood perfusate was equilibrated with a mixture of N2, O2, and CO2 in a membrane gas exchanger. The relative amounts of N2, O2, and CO2 were adjusted to yield a pH of 7.35 to 7.45 and a Po2 of 120 to 160 mm Hg in the perfusate. The pH in the perfusate was continuously monitored throughout the experiment, and in addition, controlled at the beginning and at the end of each experiment by means of a blood gas analyzer (AVL 993S, AVL). Stainless steel tubing between the membrane gas exchanger and the recording chamber prevented diffusional gas losses. The preparation was surrounded by a humidified artificial gaseous atmosphere composed of 75% N2, 20% O2, and 5% CO2.13
Determination of the Amplitude of the Extracellular Field, Propagation Velocity, and Resistances of the Extracellular and Intracellular Spaces
The application of linear cable analysis to cylindrically shaped papillary muscles and the validation of the method have been described previously in detail.13 16 In brief, presentation of a cardiac muscle cylinder as a linear electrical cable assumes (1) a constant cross-sectional shape (cylindrical or elliptical), (2) homogeneous distribution of intracellular and extracellular resistivities along the cable axis and in the direction of the fiber radius, and (3) uniform propagation of the excitation wave front. In a uniform cable, no gradients of potential exist within a cross-sectional area during excitation or flow of subthreshold current. In comparison to the complex extracellular and intracellular architecture, cable analysis will yield a value for the lumped extracellular and intracellular resistances (see “Discussion”).
Separation of the extracellular resistance from the intracellular resistance necessitates two measurements18 : (1) application of a subthreshold current pulse Is between an extracellular electrode located at the tip of the muscle and a corresponding electrode located at the interventricular septum. The linear component of the resulting extracellular voltage profile along the muscle reflects current flow through the longitudinal tissue resistance, rt (Ω/cm), which consists of ro (extracellular longitudinal resistance in Ω/cm) and ri (intracellular longitudinal resistance in Ω/cm) in parallel. The longitudinal whole tissue resistance, rt, is obtained from the subthreshold voltage, Vs, recorded between two extracellular electrodes (Fig 1⇑), the interelectrode distance, Δx, and the subthreshold current pulse strength, Is:
(2) During excitation and propagation of an active wave front, local current within the wave front flows through a resistance network, which, in a linear cable, consists of ro and ri in series. The amplitude of transmembrane action potential (ΔVm) and the amplitude of the bipolar extracellular electrogram (ΔVo) determine the ratio (q) of the extracellular (ro) and the intracellular (ri) longitudinal electrical resistance. ΔVo is a measure for the amplitude of the extracellular wave front voltage and an important determinant of the amplitude of the ECG.
The intracellular and the extracellular longitudinal resistances (ri and ro) can be calculated from Equations 1 and 2 using q and rt.
The relation between longitudinal resistance, rx, of a given tissue compartment x (in Ω/cm) and the specific resistance or resistivity, Rx, of this compartment (in Ω/cm) is given by
where Ax (cm2) is the area of this compartment in a cross section.
Longitudinal propagation velocity (Θ) is calculated from the interelectrode distance (Δx) and the conduction time (tc) measured as the time interval between the steepest part in the deflection and inflection of the bipolar extracellular electrogram.
Extracellular electrodes were made of polyethylene tubing (diameter, 45 μm) backfilled with 150 mmol/L NaCl and contained a fine silk thread. Electrical contact between the electrode and the muscle was made only by the electrolyte-to-silk bridge. In such a way, mechanical damage of the preparation by the electrodes was prevented and DC stability was obtained.
The transmembrane action potential (ΔVm) was recorded with a conventional intracellular floating microelectrode (Fig 1⇑). The intracellular and extracellular microelectrodes were connected to high-input impedance amplifiers (OPA 128 JM, Burr-Brown). The signals were amplified in custom-built differential amplifiers and sampled on an analog-to-digital converter data acquisition board (NB-MIO-16L, National Instruments) installed on a personal computer (Macintosh IIfx, Apple Computer Inc). The sampling rate for the signals was 25.0 kHz with 12-bit resolution. Off-line data analysis was performed using data analysis software (igor, Wavemetrics) on a Macintosh IIfx computer. In addition, action potentials, perfusion pressure, and contractile parameters were monitored on a strip chart recorder (Linear Corder Mark IV, Watanabe Instruments Corp). Subthreshold constant current strength, Is, was measured in the feedback loop of an operational amplifier connected between the preparation and ground.
The diameters of the preparations and the interelectrode distances were measured on video images acquired with a CCD video camera (Panasonic wv-BL202, Matsushita Co Ltd) and digitized on a frame grabber board (DT 2255, Data Translation Inc) installed on the personal computer. To analyze the exact fiber dimensions and interelectrode distances, images were acquired at a fixed interval during diastole. Off-line image analysis was done with an image processing and analysis software (image, NIH).
Experimental Protocol: Variation of Electrical Resistivity of the Intravascular Space and the Interstitial Space
It is well known that the specific resistance (resistivity) of blood is a function of red blood cell content.19 20 This property was used to produce marked and selective changes of the electrical resistance of the intravascular space and to assess this effect on the changes of extracellular and intracellular myocardial longitudinal resistance, on the magnitude of the electrical extracellular field (ΔVo), and on propagation velocity (Θ). This rationale assumes that the change in hematocrit does not affect intravascular space volume and resistance. To our knowledge, no such effects have been described. The resistivity of the perfusates was measured in every experiment using a laboratory conductivity meter (PW 9505, Philips) at a temperature of 25°C. The dependence of perfusate resistivity on the red blood cell content is shown in Fig 2⇓. In absence of erythrocytes, the resistivity of the perfusate was 74 Ω/cm. This compared with 221 Ω/cm with a hematocrit of 60% (threefold increase).
As a means to vary the volume of the interstitial space, we changed the colloid osmotic pressure (COP) of the perfusate and controlled the resulting volume change of the preparation by measuring the diameter of the whole papillary muscle. COP was measured with a collodion bag technique (Sartorius collodion bags).21 The effect of changing dextran concentration on perfusate resistivity was small and not significant. As shown from a single experiment, COP changed from 9 mm Hg with 10 g/L dextran to 94 mm Hg with 80 g/L dextran (Fig 2⇑). In previous work it was shown that physiological extracellular electrical properties are obtained with a perfusion pressure of 35 to 45 mm Hg and with a dextran concentration of 40 g/L (COP, 36 mm Hg) in the presence of a minimal amount of albumin.13 In such conditions, the values obtained for extracellular resistance, extracellular wave front voltage, and propagation velocity are very close to those observed in whole hearts in vivo.14 Therefore, these conditions were selected for control. In each experiment, one control period of 30 to 60 minutes was followed by a test period (20 to 40 minutes) and a subsequent control period (30 minutes). In some hearts, a second test period followed a by a further control period was performed.
Statistical Analysis and Experimental Protcol
Statistical comparisons of values during control and test periods were made by ANOVA with a statistical analysis software package (statview, Abacus Concepts Inc). Results are given as mean±SD. Differences between groups were considered significant at P<.05. The Bonferroni correction was used for multiple comparisons.
Extracellular Resistance, Conduction Velocity, and Magnitude of Extracellular Wave Front Voltage
In the present work, tissue resistance (rt), propagation velocity (Θ), action potential amplitude, and the magnitude of the extracellular electrical field (ΔVo) were measured at fixed time intervals during the control and test periods. In order to obtain a high degree of accuracy, rt and Θ were calculated from voltage and conduction time profiles based on multiple measurements, as shown in Fig 3⇓. In Fig 3A⇓, superimposed digital images show the five consecutive locations of the apical extracellular electrode. The position of the extracellular electrode at the base of the papillary muscle remained unchanged during the measurements, which were obtained within 40 seconds. Fig 3B⇓ shows the superimposed original signals used to measure subthreshold voltage (Vs) and conduction time (tc) as a function of interelectrode distance (Δx). The profiles of subthreshold voltage and conduction time are depicted in Figs 3C⇓ and 3D⇓, respectively, and show a high degree of linearity. The slope on Fig 3C⇓ and the reciprocal slope on Fig 3D⇓ were taken to calculate rt (20 Ω/cm) and Θ (63 cm/s), and they served to validate the linear cable model in each individual experiment.13 The high degree of linearity of the extracellular subthreshold voltage profiles and of the longitudinal propagation profiles was consistent in all experiments. The mean coefficient of correlation, r2, was .998±.010 (n=214 measurements) for the voltage profiles and .996±.009 (n=214) for the propagation profiles, respectively.
During control and test conditions, there were no significant variations in the action potential amplitude (ΔVm) and extracellular wave front voltage (ΔVo) used to calculate the ratio of extracellular to intracellular resistance.
Changes of Colloid Osmotic Pressure in the Perfusate Affect Extracellular Resistance, Extracellular Wave Front Voltage, Propagation Velocity, and Diameter
The effects of decreasing COP from 36 mm Hg (40 g/L dextran) to 9 mm Hg (10 g/L dextran) on intracellular longitudinal resistance (ri), extracellular longitudinal resistance (ro), and fiber diameter in a single experiment are shown in Fig 4⇓. Decreasing COP in the perfusate induced swelling of the preparation, as shown by an increase of fiber diameter from 1.47 to 1.62 mm. Assuming an extracellular to intracellular space ratio of 1:3 (References 4 and 5) and an interstitial space volume of 5% of total tissue volume,7 this would correspond to an 85% increase of the extracellular compartment volume or to a 430% increase of the interstitial space volume. Parallel to the volume increase of the interstitial space, there was a marked decrease in ro to 53% of control, whereas ri did not change significantly. Both interstitial swelling and the change in ro were reversible upon the reversal of perfusion to control conditions.
The effects of the decrease in ro on the amplitude of the extracellular wave front voltage (ΔVo) and conduction velocity (Θ) are shown in Fig 5⇓. The reduction of COP produced a 31% decrease of ΔVo from 62 to 43 mV. Conduction velocity increased 41% from the control value of 46 cm/s to a maximum of 65 cm/s.
The results obtained from 23 experiments in 13 different hearts are listed in the Table⇓. Reducing COP from 36 to 9 mm Hg (n=9) decreased ro by 54% and increased fiber diameter by 12%. Assuming that the interstitial space occupies 5% of total tissue volume, this corresponds to a 120% increase of the interstitial space volume. The increase in interstitial space volume was associated with a 31% decrease of the extracellular wave front voltage (ΔVo) and a 31% increase in conduction velocity. When COP was increased from 36 to 93 mm Hg, the small increase in fiber diameter and decrease in conduction velocity were not significant. However, the extracellular longitudinal resistance increased by 35%, and the extracellular wave front voltage (ΔVo) increased by 22%.
The amplitude of the transmembrane action potential amounted to 104±2.3 mV during the control state and showed no significant variations among the three groups.
Alterations of Hematocrit in the Perfusate and Influence on the Extracellular Wave Front Amplitude, Propagation Velocity, and Diameter
The effect of an increase of the electrical resistivity of the intravascular space (from 143 to 221 Ω/cm) by increasing hematocrit from 40% to 60% on cable parameters is shown from a single experiment in Fig 6⇓. The results from 30 different perfusions in 15 hearts are summarized in the Table⇑. Hematocrit was varied in four steps from 60% to 40%, 10%, and 0%. This corresponded to resistivities in the perfusates of 221, 143, 86, and 74 Ω/cm. For the measurements in absence of red blood cells, 10% calf serum was added to the perfusate (see below). As shown in Fig 6⇓ and the Table⇑, changing the intravascular space resistance from 75 to 221 Ω/cm did not influence the electrical parameters significantly or produce a change in fiber diameter.
It has been shown in myocardial tissue and mesenterium that the physiological perfusion with crystalloid solution containing macromolecules for exertion of osmotic pressure requires a minimal amount of albumin22 or albumin and hemoglobin23 for normal capillary permeability and prevention of osmotic swelling. In our preparation, in absence of red blood cells, albumin (4 g/L) did not suffice to prevent rapid accumulation of interstitial water. By contrast, normal perfusion conditions were maintained with adding 10% calf serum to the perfusate. The effect of adding 10% calf serum to the perfusate is shown in Fig 7⇓. In the absence of calf serum, the perfusion with a solution devoid of red blood cells produced a rapid and partially irreversible increase of fiber diameter (swelling) and a marked decrease of ro to 20% of control. Accordingly, there was a decrease of the extracellular wave front voltage from 60 to 30 mV and a transient increase of velocity from 45 to 75 cm/s. In the presence of calf serum, neither ro, nor diameter, nor ΔVo, nor Θ changed significantly during perfusion in absence of red blood cells. Closely corresponding results were obtained in three other experiments.
In the present study we describe the contribution of the extracellular resistance to impulse propagation and passive electrical properties of the myocardium in the isolated, arterially perfused rabbit papillary muscle. Our results show that myocardial extracellular electrical resistance, conduction velocity, and the magnitude of the extracellular electrical field are sensitive to changes in the volume of the interstitial space. By contrast, alterations in the electrical resistivity of the intravascular space do not affect the electrical parameters.
The use of the so-called core conductor model of the papillary muscle assumes that all the intracellular and extracellular spaces distributed within a muscle’s cross section are assembled into single extracellular and intracellular compartments.12 13 In the presence of a gaseous atmosphere surrounding the muscle, electrical current flow during application of a subthreshold pulse or during excitation is confined to these two compartments, and no short circuit by a surrounding bulk solution is present. In such a case, both intracellular and extracellular longitudinal resistances can be obtained in a relatively simple way from the measurement of the amplitudes of extracellular and intracellular potentials during propagation and the extracellular voltage during subthreshold current flow (see “Methods”).12 13 16 The electrical shunting by a subendocardial space can be neglected because the radial width of this space, as determined by laser confocal microscopy during perfusion, amounts to only 6 to 8 μm (J. Fleischhauer, A.G. Kléber, unpublished observation). One advantage of this method is that its validity can be tested in each individual experiment. It was validated previously for the measurement of electrical tissue properties during normal perfusion and during acute ischemia.13 16 The very high degree of linearity of subthreshold current flow and propagation during the various experimental interventions demonstrate its applicability in this study. The values obtained for both intracellular and extracellular resistances can provide information only on the average or macroscopic electrical properties of the respective tissue compartments, however. In the intracellular compartment, small local inhomogeneities in the extracellular electrogram, probably due to cell borders or borders between fiber bundles, have been described.24 25 In the extracellular compartment, the vascular network as well as the interstitial clefts between single cells and between bundles6 26 of cells are highly nonhomogeneous. This inhomogeneity may affect the relation between compartment resistances and conduction velocity in the presence of a very high COP of the perfusate, as discussed below.
The local circuit current, generated at the wave front of propagation by the difference in membrane potential between excited and resting cells and flowing through the intracellular and extracellular compartments, is a major determinant of propagation velocity. Moreover, the electrical field created by flow of local circuit current through the resistance of the extracellular space forms the basis of the ECG.27 This local circuit current during excitation is determined by the electrical membrane properties as well as the complex electrical impedance formed by both the intracellular and the extracellular compartments. Many reports at the level of cell pairs28 and whole tissue18 have characterized the electrical properties of gap junctions and of the intracellular compartment. However, experimental reports on the effects of the extracellular space resistance on impulse propagation and the genesis of the ECG are scarce. In the isolated, arterially perfused papillary muscle, the extracellular longitudinal resistance is of about equal magnitude to the intracellular longitudinal resistance.13 15 16 Therefore, the extracellular voltage across a homogeneous wave front of excitation is about 50% of the amplitude of the transmembrane action potential. Measurements of the wave front amplitude in whole heart indicate that such a high value for extracellular resistance is physiological.14
In the present study, we assessed the roles of the vascular and the interstitial space on the extracellular resistance. Information on the relative contributions of these subcompartments to the extracellular resistance has not been reported previously. Reduction of COP of the perfusate resulted in an increase in fiber diameter that most likely was due to interstitial swelling. Morphometric analysis of cardiac tissue suggests that the extracellular compartment comprises 20% to 25% of total tissue volume,3 4 and about 5% (References 5 and 7) is interstitial space. In a study using confocal microscopy, the intravascular space was 20% (inner vessel diameter) to 25% (outer vessel diameter) of total tissue volume.29 Taking average values of these data, the observed increase of papillary muscle diameter of 12% (see Table⇑) would correspond to an increase of the total interstitial space volume by about 2- to 2.4-fold. An increase in intracellular volume concomitantly with the reduction in COP and with interstitial swelling is unlikely. This is because no changes in intracellular resistance (ri) were observed. In contrast to the change in COP, a small alteration of osmolality (by ±10% variation in extracellular [NaCl]) leads to a significant change in intracellular resistance, demonstrating that ri is a sensitive indicator of changes in intracellular volume (Yan Gan-Xin and A.G. Kléber, unpublished observation). As expected, the increase of interstitial fluid was associated with a marked decrease of the extracellular resistance, ro, a concomitant increase in propagation velocity, and a decrease of the extracellular wave front voltage. In the theoretical case, this change in conduction velocity is indirectly proportional to (ri+ro)1/2 during propagation along a linear cable.30 Comparison of the calculated change of Θ by 20% with the measured value of 23% indicates that the effect of the decrease of ro is closely predicted by linear cable theory.
In the case of the diminution of the extracellular space by an increase in COP in the perfusate, quantitative application of linear cable theory for the prediction of changes in velocity (Θ) from changes in ro does not seem possible. Experimentally, increasing COP to 90 mm Hg was followed by a marked increase of ro and the extracellular field magnitude Vo but only a small, nonsignificant decrease in fiber diameter and no change in Θ. At present, we have no straightforward explanation for the apparent discrepancy between the increases of ro and Vo and the unchanged Θ at very high osmotic pressure. One possible explanation could be related to the heterogeneous structure of the extracellular space.5 A small shrinkage of the interstitial space might first affect the very narrow intercellular clefts. A reduction in volume of these very narrow clefts might result in a extremely high cleft resistance, which would exclude them from participation in local current flow and propagation. Only the membrane parts adjacent to the larger clefts (with a larger volume and consequently lower resistance) would take part in propagation. A very similar explanation was given by Fozzard31 to explain measurements of cell membrane capacity (in series with cleft resistances) and propagation in Purkinje fibers. In all, our results suggest that the changes in propagation velocity are not linearly related to (ro+ro)1/2 (Reference 30) within the whole range of ro tested.
An unexpected and new finding was that a 3-fold increase in the electrical resistance of the perfusate did not affect ro, Vo, or Θ. The vascular space has been reported to occupy about 16% of cardiac volume in the ventricle, or about 85% of the extracellular space.7 Consequently, a marked change in resistance of the major extracellular compartment should have easily been detected. In the simplest possible model of electrical current flow, the intravascular (rvasc) and the interstitial (rint) resistances are located in parallel (1/ro=1/rint+1/rvasc). With about equal interstitial and intravascular resistivities (70 Ω/cm, absence of red blood cells) and for an intravascular to interstitial space ratio of 5:1 (Reference 7), the resistance of the intravascular space is about 5 times smaller than the interstitial space resistance (see “Methods”). Increasing the intravascular space resistivity to 220 Ω/cm (Table⇑) would have changed the intravascular to interstitial resistance ratio from 0.2 to 0.63 and would have increased ro by 2.3-fold, that is, to an extent detectable by our method. The discussion about the absence of this effect is speculative. On one hand, an argument for a relatively small effect of hematocrit on ro might reside in the different rheological behavior of blood in the microcirculation versus the macrocirculation. In vessels with a diameter of >0.5 mm, blood flows as a homogeneous solution, whereas a dissociation of plasma from red blood cells occurs in the microcirculation with a decrease of the effective viscosity (Fahraeus-Lindquist effect).32 Accordingly, the change in effective electrical resistivity of blood with hematocrit might be smaller than anticipated from the measurements of blood resistivity in vitro. On the other hand, an argument speaking in favor of an effect of red blood cells on intravascular electrical resistance in capillaries relates to the apposition of the red blood cells to capillary walls. Since an average red blood cell is of larger diameter than an average myocardial capillary, one might argue that the close contact of the (deformed) red blood cell wall with the endothelium during capillary passage might form an electrical seal to current flow. The reason that the change in intravascular resistivity with hematocrit was not at all reflected in a change of ro can be best explained by an insulating effect of the capillary endothelium, however. As an important finding, extracellular electrical resistance, extracellular wave front voltage, and conduction velocity remained constant even when blood cells were completely removed from the perfusate. This indicates that the endothelial layer of the microvasculature in ventricular myocardium represents an electrical seal to current flow during excitation. The high series resistance provided by the endothelial barrier would then mask the effect of the changes of hematocrit. Our results therefore suggest that local current flow during excitation and repolarization will be confined to the interstitial space.
As shown previously in heart muscle and mesentery, prevention of interstitial swelling during arterial perfusion requires the presence of a minimal amount of albumin and/or red blood cells, in addition to a normal COP.22 23 Most likely, albumin determines vascular permeability by electrostatic apposition to subendothelial collagen, thereby sealing postulated “pores” in the microvasculature.22 In the present experiments, interstitial swelling occurred if erythrocytes were withdrawn from the perfusate even in the presence of 4 g bovine albumin per liter and a normal COP. Normal steady-state perfusion conditions in the absence of red blood cells (normal electrical parameters, no change in fiber diameter) were achieved by adding newborn calf serum to the perfusate, however. The explanation for this effect remains unknown.
An extrapolation of our results to the interpretation of the ECG in conditions of changed hematocrit or changed COP is complicated by the fact that changing blood composition appears to affect the ECG not only by modifying the intracardiac source but also by changing the complex resistances between the heart and the body surface. Thus, several clinical reports have shown that increasing hematocrit decreased QRS amplitude, and decreasing hematocrit had the opposite effect.33 34 This phenomenon, however, is explained by the so-called Brody effect,35 that is, by extracardiac resistances changing with hematocrit. A direct influence of blood protein concentration was demonstrated by Heaf,36 who showed a positive correlation between blood albumin concentration and QRS amplitude upon albumin infusion. This effect might well have been caused by the change of the intracardiac electrical field described in this study.
This work was supported by the Swiss National Science Foundation and the Swiss Heart Foundation. We thank Dr S. Rohr for his help with some of the experiments and H.U. Schweizer, J. Burkhalter, and D. de Limoges for technical assistance.
- Received December 20, 1994.
- Revision received January 11, 1995.
- Accepted January 16, 1995.
- Copyright © 1995 by American Heart Association
Weidmann S. The electrical constants of Purkinje fibres. J Physiol (Lond). 1952;118:348-360.
Polimeni PI. Extracellular space and ionic distribution in rat ventricle. Am J Physiol. 1974;227:676-683.
Frank JS, Langer GA. The myocardial interstitium: its structure and its role in ionic exchange. J Cell Biol. 1974;60:586-601.
Sommer JR, Scherer B. Geometry of cell and bundle appositions in cardiac muscle: light microscopy. Am J Physiol. 1985;248:H792-H803.
Van den Bos GC, de Jang-Strakova Z. Morphometric measurement of interstitial space in Langendorff-perfused rat hearts. J Physiol (Lond). 1982;328:69P. Abstract.
Spach MS, Heidlage FJ. A multidimensional model of cellular effects on the spread of electrotonic currents and on propagating action potentials. Crit Rev Biomed Eng. 1993;20:141-169.
Saffitz JE, Kanter HL, Green KG, Tolley TK, Beyer EC. Tissue-specific determinants of anisotropic conduction velocity in canine atrial and ventricular myocardium. Circ Res. 1994;74:1065-1070.
Fast VG, Kleber AG. Microscopic conduction in cultured strands of neonatal rat heart cells measured with voltage-sensitive dyes. Circ Res. 1993;73:914-925.
Fast VG, Kléber AG. Anisotropic conduction in monolayers of neonatal rat heart cells cultured on collagen. Circ Res. 1994;75:591-595.
Roberts DE, Hersh LT, Scher AM. Influence of cardiac fiber orientation on wave front voltage, conduction velocity, and tissue resistivity in the dog. Circ Res. 1979;44:701-712.
Kléber AG, Yan G-X. The effect of the finite extracellular resistance on impulse conduction in the ventricular myocardium. In: Sideman S, Beyar R, Kléber AG, eds. Cardiac Electrophysiology, Circulation, and Transport. Boston/Dordrecht/London: Kluwer Academic Publishers; 1991.
Kléber AG, Riegger CB, Janse MJ. Electrical uncoupling and increase of extracellular resistance after induction of ischemia in isolated, arterially perfused rabbit papillary muscle. Circ Res. 1987;61:271-279.
Chilian WM, Eastham CL, Marcus ML. Microvascular distribution of coronary vascular resistnace in the beating left ventricle. Am J Physiol. 1986;251:H779-H788.
Hill DW, Thompson FD. The effect of haematocrit on the resistivity of human blood at 37°C and 100 kHz. Med Biol Eng. 1975;182-186.
von Muralt A. Messung des kolloidosmotischen Druckes. In: von Muralt A, ed. Einführung in die Praktische Physiologie. Berlin: Springer-Verlag; 1943.
Huxley VH, Curry FE. Albumin modulation of capillary permeability: test of an adsorption mechanism. Am J Physiol. 1985;248:H264-H273.
McDonagh PF. Both protein and blood cells reduce coronary microvascular permeability to macromolecules. Am J Physiol. 1983;245:H698-H706.
Spach MS, Dolber PC, Heidlage JF. Resolution of Discontinuous Versus Continuous Propagation: Microscopic Mapping of the Derivatives of Extracellular Potential Waveforms. Orlando, Fla: WB Saunders Co; 1990.
Sommer JR, Jennings RB. Ultrastructure of cardiac muscle. In: Fozzard HA, ed. The Heart and Cardiovascular System. New York, NY: Raven Press, Publishers; 1986.
Plonsey R, Fleming DG. Bioelectric Phenomena. New York, NY: McGraw-Hill Book Co; 1969.
Weingart R. Cell-to-cell coupling in cardiac tissue. Adv Physiol Sci. 1980;8:59-68.
Fleischhauer J, Lipp P, Roberts N, Niggli E, Lehmann L, Kléber AG. In vivo quantification of myocardial vasculature by laser scanning confocal microscopy and stereology. Circulation. 1994;90(suppl I):I-265. Abstract.
Tasaki I, Hagiwara S. Capacity of muscle fiber membrane. Am J Physiol. 1957;188:423-429.
Fahraeus R, Lindqvist T. The viscosity of the blood in narrow capillary tubes. Am J Physiol. 1931;96:562-568.
Rosenthal A, Restieaux NJ, Feig SA. Influence of acute variations in hematocrit on the QRS complex of the frank electrogram. Circulation. 1971;44:456-465.
Oreto G, Luzza F, Donato A, Satullo G, Calabro MP, Consolo A, Arrigo F. Electrocardiographic changes associated with haematocrit variations. Eur Heart J. 1992;13:634-637.
Brody DA. A theoretical analysis of intracavitary blood mass influence on the heart-lead relationship. Circ Res. 1956;4:731-738.