(1)
Faculty of Life Sciences, Bar-Ilan University, Ramat-Gan, Israel
Keywords
Multiparametric approachMinimally invasive techniquesTissue compartmentsMicrocirculatory blood flowHemoglobin oxygenationTissue pO2 Ionic homeostasisLaser Doppler flowmetry5.1 Introduction
Evaluation of brain physiological functions, in real time, is performed using various methodological approaches that could be classified according to the invasiveness of the technologies used, as shown in Fig. 5.1 [1]. The lists of techniques presented in the figure are the major ones used, and additional new methodologies may not be included, or may appear in another category as well. The noninvasive group includes most of the techniques that are used in patients on a daily basis. In this approach, the sensors may touch the skin but do not penetrate the skull. This group is subdivided into the mapping/imaging option or the local measurement approach. In the second group, the minimally invasive methods, the sensors may be located epidural (below the bone) or subdural touching the pia mater. In the third approach, invasive methods, the sensor or the probe is inserted into the tissue itself, creating a new microenvironment and and a small amount of damage. Each available technology or device provides information on a limited number of parameters (one or two). We introduce here the term “physiological mapping” as presented in the minimally invasive group, named MPA or the multiparametric approach or multiparametric assembly.
Fig. 5.1
Schematic presentation of the various techniques available for monitoring of patients or experimental animals. Techniques are classified according to invasiveness of the monitoring. (© American Scientific Publishers, reprinted with permission [1])
The MPA was developed mainly to study the brain, but the same MPA concept and technology were used in other organs such as heart and kidney. The aim of the MPA is to provide real-time data describing the relationship between hemodynamic, metabolic, ionic, and electrical activities in the cerebral cortex. Normal brain mitochondrial function is a precondition for the performance of all other brain functions. Therefore, a short introduction on brain energy metabolism is presented here. The aim of this chapter is to demonstrate the historical development of the technology used in monitoring the brain and other organs functions under various pathophysiological conditions. We present the stages of the development and typical results obtained in our laboratory. It is important to note that in our monitoring system all the probes were placed on the surface of the brain and never penetrated the tissue itself. Most of the references cited in this chapter were published by our group.
5.2 Brain Energy Metabolism
The functional capacity of the brain is related to its ability to perform its work. It is possible to assess this ability through the knowledge of changes in the oxygen balance, that is, the ratio between oxygen supply and oxygen demand in the brain [1]. Healthy brain cells perform various types of activities (right side of Fig. 5.2; energy demand). The energy is derived through several complex enzyme systems, in which oxygen is the ultimate electron acceptor. The electron transfers down the respiratory chain result in the production of ATP. Concomitantly with electron transport, the respiratory chain components switch between reduced and oxidized states, each of which has different spectroscopic properties. The formation of the pyrophosphate bonds depends on the sufficiency of sugar and oxygen functions, whose inadequacy can, ultimately, lead to death. Because most of the energy consumed by tissues is dependent on the availability of oxygen, the terms “energy” and “oxygen” are used here synonymously. In a normal healthy brain, the ratio or balance between oxygen supply and oxygen demand is positive and reflects brain cell functional capacity to do work. That is, the supply mechanism of blood flow and the oxygenated blood circulation is able to provide the spectroscopic properties of the respiratory chain components that are unique to their redox status and used as internal markers of the state of oxidative phosphorylation.
Fig. 5.2
Schematic presentation of the brain energy balance concept linked to energy supply and demand. The mitochondrial NADH redox state represents the balance between oxygen supply and demand. (© American Scientific Publishers, reprinted with permission [1])
In excitable tissues, such as brain or muscle tissue, as well as in other cells, the activity of Na–K-ATPase is very sensitive to alterations in ionic homeostasis. An increase in extracellular potassium ion concentration, K+, will stimulate pumping activity to bring the extracellular K+ back to normal levels, that is, the 3-mM range. The activation of Na–K–ATPase increases the hydrolysis of ATP, and thus the mitochondria phosphorylate the ADP molecules that are released. The accelerated activity of the mitochondria will be accompanied by a more oxidized state, more oxygen delivery to the cells, blood flow, and blood supply. This coupling between energy consumption and energy production is maintained so long as the O2 supply is well regulated.
Under conditions in which the oxygen supply or delivery is limited, after a stroke or heart attack, the energy supplier, that is, the mitochondria, will not be able to produce the amount of ATP needed. As a result, energy-demanding processes will be restricted. The net effect of the imbalance between energy demand and supply will be manifested by a decrease in the tissue’s ability to do work, which can lead to development of various pathological states. The left side of Fig. 5.2 presents the various parameters that could be monitored in addition to mitochondrial function as representative of oxygen supply. To evaluate brain oxygen balance, it is necessary to measure parameters that represent the oxygen supply and demand at the same time. Mitochondrial NADH redox state represents the supply as well as the balance. Therefore, it is necessary to measure more parameters in addition to mitochondrial NADH. Figure 5.3 [1] shows that our approach aims to monitor, in real time, a small volume of the cerebral cortex containing all the tissue elements that are parts of a typical functioning brain.
Fig. 5.3
Schematic presentation of the concept “brain physiological mapping.” All the presented techniques were developed and used in our laboratory. (© American Scientific Publishers, reprinted with permission [1])
We are interested in the microenvironment of the brain containing neurons, glia, synapses, and the microcirculatory elements (small arterioles and capillaries). The various parameters and the technology developed are presented in Fig. 5.3. During the development process we pursued the goal of being minimally invasive in terms of penetration to the cortical tissue itself. It was obvious that the various probes could not monitor the same volume of tissue because of the size of each probe used. Therefore, we attempted to minimize the diameter of the various probes located in the MPA that had a 5- to 6-mm contact area with the cerebral cortex. In most of the perturbations used, such as global ischemia, anoxia, hypoxia, or hemorrhage, most of the areas in the cortex respond in the same way. We have tested this concept by the development of the multisite monitoring of NADH and other parameters as was presented in Chap. 6. The initial step in the development of the MPA was the establishment of the fiber-optic-based NADH monitoring system in 1972 when the first UV transmitting optical fibers appeared (in Table 5.1). It was a continuation of the long-term usage of old devices for NADH monitoring in vivo where the animal was located in an optic-based rigid device. The connection of the brain to the fluorometer via optical fibers enabled us to monitor, for the first time, the brain of unanesthetized animals. The initial data on the use of this technology appeared in two papers [2, 3].
Table 5.1
Milestones in the development of brain multiparametric monitoring of NADH fluorescence and other physiological parameters in vivo by Mayevsky et al.
Protocol | Year | Discovery/activity | Author(s) |
|---|---|---|---|
1 | 1973 | First fiber-optic-based fluorometer–reflectometer used in the brain of an unanesthetized animal; monitoring of NADH and electrocortical activity (ECoG) | |
2 | 1974+1977 | Simultaneous monitoring of NADH in vivo together with extracellular K+ (microelectrode and surface electrode) and ECoG | |
3 | 1980 | Monitoring of brain NADH together with tissue pO2 and ECoG | [17] |
4 | 1982 | First multiparametric assembly for NADH, extracellular K+, H+, DC steady potential, and ECoG | [9] |
5 | 1983 | Monitoring of NADH, pO2, extracellular K+, DC, and ECoG inside hyperbaric chamber | [19] |
6 | 1990–1992 | Simultaneous real-time monitoring of brain NADH, HbO2, ECoG, DC potential, extracellular K+, and Ca2+ | |
7 | 1995 | Simultaneous monitoring of brain NADH, CBF, ECoG, DC potential, extracellular K+, Ca+2, H+ | [24] |
8 | 1996 | Multiparametric monitoring of neurosurgical patients | [42] |
9 | 1997 | Monitoring of brain NADH, CBF, DC potential, extracellular K+,Ca2+ together with high-energy phosphates by 31P-NMR spectroscopy | [27] |
10 | 2000 | Monitoring of mechanism of CSD propagation | [30] |
11 | 2001 | Multiparametric monitoring under ICP elevation | [33] |
12 | 2003 | Multiparametric monitoring of rats under traumatic brain injury | [36] |
The list in Table 5.1 is organized in chronological order. All details of the technological aspects and animal preparation appear in the original relevant publications; therefore, a short description of the technology relevant to each parameter appears in the initial part of the methods section. In the results section, typical responses to various types of perturbations are presented together with the technology used in the specific study. Our approach was to develop a new upgraded version of the monitoring system and present initial preliminary results. The next step was to run a large, well-designed study on a few groups of animals, and the data were quantitated and analyzed for statistical significance.
To save space we are presenting here only typical results collected during the developmental stage.
5.3 Methods
5.3.1 NADH Monitoring
NADH can be measured by utilizing its absorption spectrum in the UV range, as well as by the blue fluorescence spectrum under UV illumination. All details regarding the monitoring of NADH appear in Chap. 4.
5.3.2 Microcirculatory Blood Flow
To measure the total blood flow (TBF) from the same cortical area as the MPA location in real time, we used the laser Doppler flowmeter (LDF) technique [4–6]. The LDF measures relative flow changes, and readings have been shown to correlate with the relative changes in cerebral blood flow (CBF) measured by the two other quantitative approaches [6]. The principle of the LDF is to utilize the Doppler shift, namely, the frequency change that light undergoes when reflected by moving red blood cells. A beam of low-power light or diode laser light is transmitted by an optical fiber to the tissue. After the multiple scattering of the light, another optical fiber picks up the reflected light that is recorded by a photodetector. The run signal is analyzed by a complicated algorithm developed by the manufacturers, and the results are presented in percentage of a full scale (0–100 %), thereby providing arbitrary relative flow values. The change in the total back-scattered light is an indirect measure of the blood volume in the sampled tissue. To quantify and normalize CBF values, we defined the reading value after death as 0 CBF. The 100 % value was defined as percent CBF read on the LDF scale during the control period.
5.3.3 Oxygen Electrodes
The electrodes were constructed inside 1.6-m (OD) polyethylene tubing (PE-160). Platinum wire (25 μm in diameter) was sealed in glass by a flame, and an insulated lead wire was attached to the other end of the platinum wire. Two assemblies, along with Teflon-coated 250-μm silver wire (for use as a reference), were pulled into the PE tubing so that the glass-sealed ends of the platinum were flush with the end of the PE tubing. The Teflon on the end of the silver wire was removed, and this bare section could extend beyond the edge of the PE tubing. The electrode zero response was tested in saline bubbled with N2, and its linearity was tested in saline bubbled with different mixtures of O2 and N2. A cellulose diacetate membrane was put on the tip by dipping the electrode in 5 % cellulose diacetate solution.
5.3.4 Ion-Selective Electrodes and DC Potential
To monitor the extracellular levels of K+, Ca2+ and H+, we used specially designed mini-electrodes made by World Precision Instruments (WPI; Sarasota, FL, USA). A flexible tubing made of polyvinyl chloride was sealed at one end with a membrane sensitive to a specific ion. The tube was filled with the appropriate solution and connected to an electrode holder with a salt bridge between the membrane and an Ag-AgCl pellet located inside the holder. The interface between the polyvinyl chloride tubing and the holder was glued with epoxy, and such electrodes were usable for a few weeks. The sensitivity of the electrodes to the specific ion was close to the Nernstian value, namely, 50–60 mV/decade for K+ and H+ or 25–30 mV/decade for Ca2+.
DC potential was measured concentrically around the ion-selective electrodes. Each electrode had a saline bath around its perimeter, and an Ag-AgCl electrode (WPI) was connected to it.
5.3.5 Reference Electrode
An Ag-AgCl electrode connected via a saline bridge to the neck area of the animal was used as the reference electrode. Polyethylene tubing stuffed with a cotton string that expanded when wet was inserted into the electrode holder (WPI) and glued with 5-min epoxy.
5.3.6 Electrocorticography (ECoG)
Spontaneous electrical activity of the brain surface was measured by two polished stainless steel rods or silver wires inserted into the MPA.
5.3.7 Temperature Measurements
Brain temperature was measured with a thermistor probe) Yellow Springs Instruments) located inside the MPA and connected to a thermometer.
5.3.8 Data Collection and Analysis
Relative TBF
The LDF provided arbitrary units (blood flow and volume) that were calibrated in relative terms compared with the normoxic brain that served as the control (0–100 % range).
NADH Redox State
Measurements of NADH provided a method for evaluating the mitochondrial redox state. The normoxic NADH level is considered to be 100 %, and the changes caused by the treatment given were calculated in comparison with the normoxic level. Mayevsky [7] has shown in the past that the NADH redox state correlates well with brain functional activity. The origin of the NADH signal in our system is mainly mitochondrial [8], whereas the cytoplasmic NADH contribution to the signal is negligible. Nevertheless, because the NADH is not measured in a purified form, we prefer to label the trace in the records as corrected fluorescence (CF).
Multiparametric Calculation
The MPA provides analog signals of electrical activities and NADH redox states as well as extracellular concentrations of K+, H+, and Ca2+. The mV values measured by the electrodes were transformed to mM values by using the standard calibration technique [9] and the principles of the Nernst equation. All calculations were performed via special software programs.
Real-time Data Acquisition
Because of the large number of parameters measured and the need for the logarithmic transformation of the data from the K+, Ca2+, and H+ electrodes, it was necessary to use a computerized system for data acquisition and storage. Furthermore, to record all the analog signals (at least 16 channels), it was more appropriate to use a computer display instead of paper recording for continuous collection of information.
5.3.9 Animal Preparation for Monitoring
All experimental protocols were approved by the institutional animal care committee under the instruction of the National Institutes of Health. The experimental procedures were detailed previously [10, 11]. To demonstrate the performance of the various monitoring devices, we used male rats and male Mongolian gerbils. The preparation of the animals for monitoring was presented in Chap. 4.
5.4 Results and Discussion
In this section we present and discuss the stages of the monitoring system development and the data collected by these systems. The protocol numbers in this section are in the same order as they appear in Table 5.1.
5.4.1 Fiber-Optic-Based Fluorometer and EEG
During the initial step, we developed a fiber-optic-based fluorometer/reflectometer for monitoring of mitochondrial NADH in anesthetized or unanesthetized animals. Various types of fiber-optic probes were developed over the years. To keep a direct and constant contact between the brain and the probe, special design holders were developed. The first fiber-optic probe was described in 1973 [3, 11], and the other types were presented in our review paper [12]. After the end of the operation, the probe was inserted into the holder that was cemented to the skull.
Figure 5.4 shows the typical two responses obtained in a rat experiment [2]. As can be seen, we monitored two signals by the fluorometer, namely, NADH fluorescence at 450 nm and total back-scattered light at the excitation wavelength (366 nm), called reflectance. In addition to NADH we monitored EEG activity by placing two stainless steel electrodes on the brain surface. In these preliminary studies the increase in the fluorescence and reflectance signals was in the down direction as compared to the upward direction used in other records in the book.
Fig. 5.4
a Typical response of rat brain and systemic blood pressure (MAP) to a nitrogen cycle. Ref reflection, Flu fluorescence, MAP mean arterial pressure, SB stop breathing, SN stop nitrogen, SBS start breathing spontaneously. b Response of brain to application of KCl (0.4 M) on dura surface under hyperoxia (100 % O2) inducing cortical spreading depression. (© Reprinted with kind permission of Springer Science + Business Media [2])
In Fig. 5.4a, the complete elimination of oxygen led to a large increase in the NADH and the EEG signal disappeared very quickly. In this animal the change in the reflectance was very small as compared to most of the rats. As soon as the rat started to breathe air, the NADH level recovered very rapidly. In Fig 5.4b, the effect of brain activation induced by cortical spreading depression (CSD) is shown. In this response, blood volume changes concomitantly with the mitochondrial NADH. The NADH redox state was shifted to a more oxidized state by the increase in ATP demand. The CSD developed only in the stimulated hemisphere, as seen in the EEG signal, whereas the contralateral hemisphere served as a control. The same monitoring system was used in other studies [13].
5.4.2 Addition of K+ Monitoring
The next step in the development of the MPA was to measure extracellular potassium levels representing the oxygen demand processes (Fig. 5.5). In the initial stage we used the microelectrode for potassium available in the early 1970s. The results of such an experiment [14] are shown in Fig. 5.5c. We combined the measurement of NADH, ECoG, DC potential, and extracellular K+ from the same hemisphere [15]. This electrode was inserted into the cerebral cortex while the NADH was placed on the surface. The leakage of potassium stimulated the energy metabolism recorded as the oxidation of NADH (state 4 to state 3 transition).
Fig. 5.5
a Representation of a combined electrode for extracellular K+, DC steady potential, and electrocorticogram (ECoG) leads from the brain surface. b Recording made using the combined surface electrode during cortical spreading depression. (© American Physiological Society, reprinted with permission [16].) c Effect of cortical spreading depression (elicited by 0.6 M KCI) on extracellular potassium, DC potential, ECoG, and NADH fluorescence of exposed rat brain cortex. Potassium and DC steady potential were measured by microelectrodes. (© Reprinted with permission from Elsevier [14])
To minimize the invasiveness of the potassium electrode, we developed a surface electrode for potassium located on the brain in the same configuration as the NADH probe. The technical details of the preparation of the surface potassium electrode were provided by Crowe et al. [16]. The experimental procedure is shown in Fig. 5.5a and the typical response to CSD in Fig. 5.5b. To measure the extracellular potassium in the rat brain, it was necessary to remove the dura mater because diffusion of the ion through the dura was very slow and undetectable [16]. Responses to CSD were recorded by the two types of electrodes used.
5.4.3 NADH and pO2 Measurements
Because mitochondrial NADH is sensitive to intracellular levels of oxygen, we had developed a MPA that contained a surface oxygen electrode in combination with NADH and ECoG (Fig. 5.6a). We used the term electrocorticography (ECoG) when the electrodes were in direct contact with the cerebral cortex. It is important to note that oxygen electrode readings are averaging the oxygen level in the vascular, extracellular, and intracellular compartments [17].
Fig. 5.6
a Schematic presentation of location of various probes on surface of gerbil brain. The large cannula contained light guide for NADH measurement and pO2 and ECoG electrodes. Cannula for KCl application was located 2–3 mm anterior to large cannula. b, c Effects of anoxia (b) and carotid arteries occlusion (c) on metabolic and electrical activity of the gerbil brain. (© John Wiley and Sons, reprinted with permission [17])
Typical results of this MPA are presented in Fig. 5.6b, c. The response to anoxic episode, shown in Fig. 5.6b, is typical to the lack of oxygen, namely, a sharp drop in pO2 and a large increase in NADH. At the recovery phase, the NADH recovered immediately to the pre-anoxic level whereas the pO2 showed an overshoot pattern. One minute after breathing air, a secondary response to the anoxia was recorded in all signals (vertical line). The ECoG shows a depression of the amplitude as observed after initiation of cortical spreading depression (CSD). The NADH became oxidized whereas the pO2 showed a tri-phasic response.
The pO2 was inversely related to the reflected light changes, namely, that when the reflectance showed an increase, the pO2 trace showed a parallel decrease; and when the reflectance showed a large decrease (increased blood volume), the pO2 showed three- to fourfold increase. A gradual recovery of the oxygen signal to the normoxic level was recorded in parallel to the gradual increase in the reflected light.
We concluded that the secondary event is the typical response to CSD, as described previously [8] after an anoxic cycle.
Figure 5.6c presents the effects of unilateral or bilateral carotid artery occlusion. In this animal the occlusion of the contralateral carotid artery (Locc) had a very small effect on NADH or pO2. Occlusion of the right carotid (while the left one was closed) induced a large fall in pO2 and an increase in NADH with similar kinetics. The recovery after bilateral recirculation was rapid, without any significant overshoot in the pO2. In the second ischemia episode, when nitrogen was applied to the ischemic brain, a large overshoot in the pO2 was recorded after recirculation. The large increase in pO2 was parallel to the large decrease in reflectance with the same kinetics toward the normoxic level. The NADH shows an oxidation cycle after recirculation caused by CSD, as can be seen in the ECoG trace.
5.4.4 The First Multiparametric Monitoring System
Our aim was to provide a new holding system allowing an easy implantation of surface probes in various combinations on the brain of small mammals and to reevaluate the potential of such surface electrode assemblies in tracing physiological and pathological events [9]. At this stage, we added to the basic assembly system the measurements of extracellular tissue pH and K+, DC potential, and local temperature, in addition to the other parameters, including NADH fluorescence, reflected light, and electrocorticogram (ECoG). Ideally, the multiprobe assembly designed (Fig. 5.7) for our project had to be rugged enough to withstand routine use by nonspecialists and yet be miniaturized enough to fit the limited surface available for implantation on rat and gerbil skulls (about 6–7 mm in diameter on each hemisphere). For ischemic studies, our interest was focused on monitoring mean values over large areas, suggesting the use of larger electrodes. In contrast, following propagated events such as spreading depression would have required the tip of the multiprobe assembly to be concentrated in a tiny space, perceiving the same phase of the wave, unless each sensor could be located with respect to the wavefront by an auxiliary signal. The second option was retained by recording the wave of the DC potential concentrically to each sensor. In addition, the complete assembly had to provide adequate protection and shielding of the high impedance ion-sensitive electrodes (K+ and pH) and a stable enough reference junction for use in unanesthetized animals. With K+ surface electrodes, concentric DC potential has been proposed as the best approximation for the DC potential component to be subtracted from the sensor signal [16]. In this study we also attempted to check the accuracy of the method by recording DC potential differentially between a central barrel and the peripheral slit and to reevaluate its usefulness in correcting surface K+ and pH measurements under various conditions.
