University of Southern California Dissertations and Theses

Human Skeletal Muscle Oxygenation And Perfusion: Non-Invasive Measurement By Near-Infrared Spectroscopy

(USC Thesis Other)

Human Skeletal Muscle Oxygenation And Perfusion: Non-Invasive Measurement By Near-Infrared Spectroscopy

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 HUMAN SKELETAL M USCLE OXYGENATION AND PERFUSION: NO N-INVASIVE M EASUREM ENT BY N EA R -IN FR A R E D SPECTROSCOPY by Laura Marcu A Thesis Presented to the FACULTY OF THE SCHOOL OF ENGINEERING UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE IN BIOMEDICAL ENGINEERING May 1995 Copyright 1995 Laura Marcu UMI Number: 1378424 UMI Microform 1378424 Copyright 1996, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 This thesis, written by Laura Marcu under the guidance of her Faculty Committee and approved by all its members, has been presented to and accepted by the School of Engineering in partial fulfillment of the re quirements for the degree of ....................Master..pf..Scj.ence ...............in „B i omed i.ca 1 _ _ Enc[i.neerinq D a te 04/20/95......................... Faculty Com m ittee f / c d ./k /. H C nairm ai ACKNOWLEDGEMENTS The experiments were conducted at the University of Southern California’ s Department of Biomedical Engineering, under the guidance of Dr. Jean-Michel Maarek, to whom I am grateful for his assistance, suggestions and criticism. I would also like to thank Dr. Vasilis Marmarelis for his support, and Dr. Sandra Howell for introducing me to the interesting field of physiology. I express my gratitude to all the colleagues, friends, and undergraduate students who volunteered as sub jects for the experiments. CONTENT ACKNOWLEDGEMENTS ii LIST OF FIGURES v LIST OF TABLES vii PREFACE viii 1. PRINCIPLES OF NEAR-INFRARED SPECTROSCOPY 1 1.1. Tissue optical properties 2 1.1.1. Absorption in tissue 3 1.1.2. Scattering in tissue 8 1.1.3. Refractive index 8 1.14. Propagation of light in tissue 9 1.2. Optical spectroscopy measurements 12 1.3. Clinical applications 15 2. TISSUE OXYGENATIO-PHYSIOLOGICAL ASPECTS 17 2.1. General aspects 17 2 .2. Muscle oxygenation 21 3. EXPERIMENTAL METHODS 23 3.1. Instrumentation 23 3.2. Experimental protocol 25 4. DATA ANALYSIS 27 4.1. Filtering 27 4.2. Near— Infrared spectroscopy algorithm 28 4.3. Fourier analysis 31 4.4. Statistical analysis 31 4.5. Oxygen consumption estimation 32 iii 5. EXPERIMENTAL RESULTS 33 5.1. First protocol observations 33 5.2. Second protocol observations 36 5.3. Oxygen consumption 39 6 . LUMPED PARAMETER MODEL 70 6.1. Model design 70 6.2 . Simulations results 74 7. DISCUSSION 78 8 . CONCLUSIONS 82 REFERENCES 84 iv LIST OF FIG URES F ig u re 1.1: Optical absorption spectra of oxy and deoxy 5 hemoglobin. F ig u re 1.2: Optical absorption spectra of cytochrome a,a3. 6 F ig u re 1.3: Pure water absorption spectra. 7 F ig u re 1.4: Photon migration diagram. 9 F ig u re 1.5: Steady state measurements. 13 F ig u re 1.6: Time resolved measurements. 14 F ig u re 2.1: Model of the respiratory system of mammals. 17 F igu re 2.2: Dissociation curve for Hb and Mb. 19 F igu re 2.3: 0 2 consumption rate of various organs. 22 F igu re 3.1: Experimental setup. 24 F ig u re 5.1: Filtering effect on optical signal (static measure- 49 ments). F ig u re 5.2: Filtering effect on optical signal (venous pressure 40 variation— 1 cycle/min). F ig u re 5.3: Filtering effect on optical signal (venous pressure 41 variation— 2 cycle/min). F ig u re 5.4: HbO, HbR, Hbt relative concentration value changes 42 during venous occlusion. F ig u re 5.5: Averaged muscle HbO response to venous occlusion. 43 F ig u re 5.6: Averaged muscle HbR response to venous occlusion. 4 4 F ig u re 5.7: Averaged muscle Hbt response to venous occlusion. 45 V F igu re 5.8: HbO, HbR, Hbt relative concentration value changes 46 during arterial occlusion. F igu re 5.9: Averaged muscle HbO response to arterial occlusion. 47 F igu re 5.10:Averaged muscle HbR response to arterial occlu- 43 sion. F igu re 5.11:Averaged muscle Hbt response to arterial occlusion. 49 F igu re 5.12: Oscillations of HbO, HbR and Hbt (20 mm Hg, 50 1 cycle/ min frequency. F igu re 5.13: Oscillations of HbO, HbR and Hbt (40 mm Hg, 51 1 cycle/ min frequency F igu re 5.14: Oscillations of HbO, HbR and Hbt (60 mm Hg, 52 1 cycle/ min frequency. F igu re 5.15: Oscillations of HbO, HbR and Hbt (20 mm Hg, 53 2 cycle/ min frequency. F igu re 5.16: Oscillations of HbO, HbR and Hbt (40 mm Hg, 5 4 2 cycle/ min frequency. F igu re 5.17: Oscillations of HbO, HbR and Hbt (60 mm Hg, 55 2 cycle/ min frequency. F igu re 5.18: Oscillations of HbO signal. 56 F igu re 5.19: Oscillations of HbR signal 57 F igu re 5.20: Oscillations of Hbt signal. 58 F igu re 5.21: Phase delay: Hbo and cuff pressure oscillations. 59 F igu re 5.22: Phase delay: Hbt and cuff pressure oscillations. 60 F igu re 5.23: Phase delay: HbR and cuff pressure oscillations. 61 F igu re 5.24: HbO, HbR, Hbt concentration. Fibers optic 62 geometry: 1.2 cm distance, 60° angle. F igu re 5.25: HbO, HbR, Hbt concentration v. Fibers optic 63 geometry: 3 cm distance, 90° angle. vi F ig u re 5.26: Averaged muscle HbO, HbR and Hbt response to 50 mm Hg. F ig u re 5.27: Averaged muscle HbO, HbR and Hbt response to 100 mm Hg. F ig u re 5.28: Averaged muscle HbO response to 150 mm Hg occlusion. F ig u re 5.29: Averaged muscle HbR response to 150 mm Hg occlusion. F ig u re 5.30: Conversion rate HbO-HbR (200 mm Hg). F ig u re 5.31: Conversion rate HbO-HbR (150 mm Hg). F ig u re 5.32: Skeletal muscle oxygen consumption rate. LIST OF TABLES T able 1: Simulation parameters. PREFACE Several technologies have been used over the past few years for the evaluation of regional oxygenation mechanisms. Among those, optical methods, and more specifically Near Infrared Spectroscopy (NIRS) have been identified as convenient tools to estimate regional oxidative metabolism noninvasively. NIRS has been developed experimentally and clinically for the non-invasive monitoring of changes in brain oxygenation by Jbbsis-Vander Vliet in 1977. He demonstrated the feasibility of in vivo NIRS monitoring of changes in the oxygenation state of hemoglobin and the oxidation level of cytochrome a,a3 in the brain. Over the last decade this method has gradually evolved and the information extracted from absorption and/or scattering properties of tissues has been applied for therapeutic dosimetry, diagnostic spectroscopy and optical imaging. NIRS is based on the following properties: - Very few biological component absorb light in the near-infrared range. Among those hemoglobin/myoglobin and cytochromes are the only detectable absorbers reacting to situations of hypoxia/anoxia and ischemia. - Near-infrared light penetrates skin and bone deeply. - NIRS gives the possibility to separately assess blood and cytochrome status. This allows a greater insight in the causative factors which lead to a deficient O2 supply. NIRS requires relatively inexpensive instrumentation by comparison to other noninvasive techniques for intracellular monitoring of oxidative metabolism such as magnetic resonance spectroscopy, positron emission tomography, NADH fluorimetry or somatosensory evoked potentials. The relative homogeneous muscle structure offers the opportunity for a facile evaluation of tissue oxygen gradients. Therefore, NIRS has been intensively applied for evaluation of changes in muscle oxygenation and saturation during ischemia [1--6] and venous occlusion [1], muscle oxygen utilization in patients with peripheral vascular diseases [7], mitochondrial myopathy [8] and heart failure [9] and in healthy subjects during exercise [10,11]. In present study, forearm skeletal muscle of healthy human subjects is investigated by NIRS in the following physiological conditions: - occlusion of venous outflow at different levels; - occlusion of arterial inflow (ischemia); - oscillatory variation of venous pressure at different mean pressure levels and different frequencies. The objectives of the experiments are to noninvasively: - determine changes in skeletal muscle oxygenation during ischemia and during changes in muscle blood volume caused by changes in venous pressure; - identify the mean pressure level where ischemia starts being induced; - evaluate the ability of NIRS to assess mechanical characteristics (hemodynamics) of the human forearm skeletal muscle vasculature; - determine human skeletal muscle oxygen consumption. Experimental observations are compared with estimations derived with the help of a lumped parameter model of the forearm vascular bed. In the first chapter I present background information about near-infrared spectroscopy, including historical background, optical properties of tissue, propagation of light in tissue, optical spectroscopy measurements, and clinical applications. Basic concepts of absorption, scattering and photon migration are included. The second chapter presents background information on tissue oxygenation mechanisms including skeletal muscle particularities. The main steps in the respiratory chain of mammals, transport of oxygen and oxygen consumption principles are pointed out. The experimental methods including instrumentation, devices, and experimental protocol for each kind of experiment are described in chapter three. Chapter four presents the data analysis methods. This includes processing of optical data, NIRS algorithm, Fourier analysis of relative changes in hemoglobin content, statistical analysis and oxygen consumption estimation method. The experimental results are shown in chapter five. This comprise filtering effect on optical data, changes of oxygenated, deoxygenated and total hemoglobin concentrations during various conditions (different occlusion levels, venous pressure variation at different mean cuff pressure levels and frequencies, fibers optics geometry) and muscle oxygen consumption under hypoxic conditions. Chapter six presents the lumped parameter model used in computer simulations to identify the mechanical characteristics which best describe the experimental results. Chapters seven and eight present the data interpretation and summarize the main findings of this work. 1. PR IN C IPLES OF NEAR-INFRARED SPECTROSCOPY The optical study of tissues began with the spectroscopic studies in 1935, when Glenn Millikan proposed a "metabolic microscope" (a dual wavelength spectrophotometer using green and red light) by which he observed the metabolic demand as expressed by the deoxygenation of hemoglobin and myoglobin in the cat soleus muscle during functional activity (tetanic contraction and ischemia) [12]. He also devised a very sensitive dual wavelength spectrophotometer for oximetry in the ear that was the precursor of pulse oximetry. Later on, he recognized the problem of light scattering and its wavelength dependence with green and red light and also observed that near-infrared light would be better than other wavelengths due to its higher penetration in tissue. Modern techniques were pioneered in the 1950’s by Chance, who developed differential spectroscopy to measure changes in light absorption by cytochromes and other pigments in turbid media and tissues [13]. NIRS was first attempted by Fran Jobsis-Vander Vliet in 1977, who demonstrated the existence of a "optical window" for transmission of near-infrared light in biological tissues [14]. Near-infrared spectroscopy is based on the difference in optical absorption spectra among different tissue conditions. The changes in absorption spectra can be quantified using the principles of reflectance and transmittance measurements. The spectral characteristics of remitted light (i.e., diffusely reflected from or transmitted through the tissue) are the results of a complex interplay of the: - absorption and scattering properties of tissue; - distribution of chromophores and scattering components; - source-tissue-detector geometry. NIRS can be used to investigate the metabolic, physiologic and structural status of the tissues. 1.1. T issu es o p tic a l p ro p erties In tissues as in any other turbid media, light is scattered or absorbed due to inhomogeneities and absorption characteristics of the medium. Two analytical approaches have been used for the study of light propagation in tissue [15,16]: 1. Radiative transfer theory, in which light propagation is represented by transport of individual photons (i.e., transport of power through turbid media) which may be either elastically scattered or locally absorbed in tissue according to the linear coefficients scattering (ps) and absorption (jia). 2 2. Electromagnetic wave theory, where scattering and absorption processes are associated with microscopic spatial variations in the dielectric properties of tissue. This approach takes into account the statistical nature of the medium, statistical moments of the wave and all diffraction effects. In the evolution of light propagation in tissue models, the first method was found to be more practical to apply in spite of its heuristic nature. The absorption and scattering properties of tissues play a crucial role in the study of light propagation though tissue. The coefficients |ia and |is are expressed in units of cm'1 and represent the rates of radiant energy loss -d§l dL due to absorption and scatter per incremental unit photon path length dL in the tissue. Typically, in the so-called "therapeutical or optical window", [600 to 1300 nm], the total tissue absorption coefficient is in the range 10" 2 to 10'1 cm'1, depending on blood content and pigmentation. The scattering coefficient is in the 102 to 103 cm" 1 range, decreasing roughly inversely with wavelength [16]. 1.1.1. A b sorp tion in tissu e The total optical absorption in tissue is the sum of the absorption due to specific chromophores, either endogenous or exogenous. The absorption 3 coefficient, |ia, for each absorber is expressed by: Va = e • c Where e represent the extinction coefficient (cm2/mmol) and C is the concentration of the absorber (mmol/cm3). In the near— infrared region the endogenous chromophores are typically heme proteins such as hemoglobin, whose absorption characteristics depend on the oxygenation state, myoglobin and cytochrome a,a3 in both reduced and oxidized form (figures 1.1— 1.3.), melanin, and bibirubin. The exogenous chromophores are usually the photodynamic sensitizers (i.e. porphyrin-like structure) which are used during photodynamic therapy. Hemoglobin molecule is composed of four hemoglobin monomers, which can bind one O2 molecule each. The absorption coefficients ko and H r of oxyhemoglobin (HbO) and deoxyhemoglobin (HbR) of hemolyzed blood are shown in figure 1. The absorption of HbO and HbR are equal at the wavelengths X = 548, 568, 587, and 805 nm. These wavelengths are called isosbestic points. The second important absorber is the enzyme cytochrome a,a3 (Cyt), which is normally attached to the inner mitochondrial membrane and is an active 4 elem ent (i.e., catalyses the reduction of 0 2 to water) on the terminal part of the respiratory chain. Cytochrome a,a^ has absorption peaks at 563, 554, 550 and 605 nm in the reduced condition while in the oxidized condition, the absorbance at these wavelength is less (figure 1.2 a). The copper atoms confer on cytochrome a,a^ the capacity to absorb light in the near— infrared in the range 800— 850 nm with a maximum at approximately 830 nm (figure 1.2 b) [18,19]. 100-3 HbO HbR 10- o E u u > o H L O 0 3 0.4 0 5 0.6 0.7 0.8 0.9 Wovaiength i / t m ) F ig u re 1.1: Optical absorption spectrum (cm2/mole) of oxyhemoglobin (HbO), and deoxyhemoglobin (HbR) absorption coefficients in cm"1. From [15]. 5 jttwbaaeelA) Reduced 600 700 800 900 500 1000 Wavekapb (am) a Abeorfauce (A) 0-3 0-2 0-1 - 1000 800 900 700 W m k a p t i n i F ig u re 1.2 : Optical absorption spectra (cm2/mmole) for cytochrome a,a3: visible and near IR (a), and only near IR (b). From [18]. The amount of water contained in mammals soft tissue is in range of 70-- 80%. Therefore, the optical properties of tissues are dominated by water 6 absorption when the latter is significant [20], The pure water absorption spectra is shown in figure 1.3. A peak attenuation spectra in near-infrared range may be seen at 965 nm. This peak absorption has been used for pathlength estimation in tissues [21], e c o o 3 C • < -2 . 2.0 1 . 0 0.2 X (p m ) F ig u re 1.3: Pure water attenuation (absorption) spectra in m'1. From [15]. In tissues, the hydration shell of a protein was described as a layer of water interacting with the macromolecule (i.e., bound water), which has different thermodynamic, kinetic, and spectroscopic properties from those observed for "bulk" water [22]. In literature, those observations were mainly made for Raman spectroscopy. Only recently, this aspect has been started to be taken in the account in NIRS. This is a new subject of investigation regarding NIRS. 7 1.1.2. S c a tter in g in tissu e Optical propagation in biological media is dominated by scattering (|is) due to: 1) discontinuities in the refractive index at the microscopic level (i.e. inhomogeneities of cellular structure); 2) particle sizes which are of the order of the optical wavelength [16,23]. The detailed dependence of scattering on tissue architecture is not well understood, and most descriptions of scattering properties are empirical. Except for striated muscle which has regularly aligned fibers, the differential scattering cross section in soft tissue is independent of the incoming photon direction and could be described by the angle, 0 (angle of photon deflection after a single scattering event), between the incoming and outgoing photons in the scattering plane. The scattering angular distribution is characterized by the anisotropy factor g = <cos0> (average cosine of the scattering angle), which is typically in the range 0.7 to 0.95 for soft tissue in the optical window [16]. 1.1.3. R efractive in d ex For soft tissue the refractive index, n, lies in the range 1.3 to 1.5 [16]. Two surface effects are particularly important in reflectance measurements: 8 - The refractive indices at the surface gives rise to specular reflection. Specular reflectance yields photons which have not "sampled" the tissue interior. - Internal reflection of photons propagating within the tissue and which strike the tissu e-air boundary at an oblique angle. Total internal reflection can occur, which reduces the "escape" of photons from the tissue as observable diffuse reflectance or transmittance. 1.1.4. P ro p a g a tio n o f lig h t in tissu e The propagation of light in tissue or scattering media is described in terms of "particle" or "photon migration" in the radiative transfer approximation [24]. A schematic diagram of photon migration is shown in figure 1.4. Light in Light out F ig u re 1.4: Photon migration diagram. From [12]. The total pathlength, L, traveled by photons when migrating from source to detector is the scattering length times the number of scattering events. In the absence of scatterers, the total path length traveled by photons before detection is the geometric distance between source and detector. As the number of scatterers increases, the total pathlength, L, becomes greater than the geometrical distance between source and detector. In addition, since the change in direction at each scattering event is random, there is a distribution of total pathlengths traveled by photons that successfully migrate from source to detector. In the presence of absorbers, the probability that photons survive scattering events decreases. The probability of detecting photons that traveled on certain pathlength in the absence of absorber is also changed [24]. Commonly used parameters to describe photon migration in tissue are: - scattering albedo: a = ps/|ia; - total attenuation coefficient: |it = |ia + IV > - free path: mfp = l/|it (the average distance traveled by a photon before a scattering or absorption event); - transport scattering coefficient: |is' = ps(l-g), typically in the 10 -100 cm'1 range; - penetration depth 8: « - 1 - , 1 v-'ff J 3 | i > , t | i , ( i - e ) ) expressed in cm, where \ieff represent the effective attenuation 10 coefficient. The penetration depth increases rapidly for wavelengths bigger than 600--700 nm due to decreasing hemoglobin absorption. Above this threshold, 5 increases slowly, except for a small dip at around 965 nm caused by a peak of absorption of water [23]. - diffusion coefficient: D = 82 According to the diffusion approximation the fluence rate (effective concentration of photons at position r and time t) in a tissue may be obtained by solving the general diffusion equation [24]: ^ r’ ^ ~ D v2§ + (r> 0 = S (r, t) where D is the diffusion coefficient and S a source term. The propagation of light in tissue can be characterized using Monte Carlo (random walk) simulation or when possible by analytical solution of the diffusion equation. The amount of light propagating through and remitted by tissues is strongly affected by scattering and does not simply depend on the absorption spectrum of tissue chromophores. The ability of near-infrared light to penetrate deeply in tissue in spite of the very short mean free path between interactions and the high remittance of light scattered out of the tissue after li deep penetration are due to the following observations: - most interactions are scattering events rather than absorption events (high albedo coefficient); - scattering interactions are highly forward-directed (high anisotropy factor) such that photon continue to penetrate tissue after multiple scattering events. 1.2. O ptical sp ectro sco p y m easu rem en ts The main goal of noninvasive optical absorption spectroscopy is to characterize the changes which take place in tissue by quantifying the absorption spectra in the wavelength range of interest. The accurate measurements in tissue of these absorption spectra are limited by the fact that there is a distribution of optical path length due to multiple scattering. Two methods have been used to separate the effect of absorption and scattering: 1) steady— state and 2) time— resolved measurements of diffusely reflected (R) or transmitted (T) light by tissue. a) S tead y sta te m ea su rem en ts (figure 1.5) imply a continuous (cw) irradiation or pulsed irradiation where the pulse length is long compared with the photon propagation time in the tissue volume [15]. In this case the 12 absorption coefficient is directly related to the incident ( [I0) and transmitted (7) light intensity by the Lambert-Beer law: l n ( y ) = eCL Wilson et al. [25] has demonstrated that, for an optically homogeneous and semi— infinite medium measurements of the radial (p) dependence of the local diffuse reflectance, R(p), can be used to obtain both pa and |is'. a o T a o o - 1 F ig u re 1.5: Steady state measurements: a) of the total remittance as either reflectance R and/or transmittance T. b) of the local spatial distribution of remittance R(p) along the tissue surface. From [23]. This method allows measurements in real time. It can be used for investigation of dynamic processes in tissue. The required instrumentation is relatively inexpensive. The main limitations of the method are: the measurements are sensitive to boundary conditions at the tissue surface; 13 the effects of tissue heterogeneity and of finite tissue volume are not known and may lead to significant errors in the derived optical properties; the necessity of acquiring a large data set, insensitivity to small changes in the optical coefficients [15]. b) T im e r e so lv e d m ea su rem en ts is the alternative approach, where for constant speed of light in tissue, c, (i.e., constant volum e-averaged tissue refractive index, n), the photon "time-of-flight," t, is directly related to the optical pathlength: L = ct. This method can be used in both the time domain and the frequency domain (figure 1.6 ). T im e -R e so lv e d S p ectro sco p y F reouency- R esolved Spectroscopy m c • » c >* e a S o u rce D etector Source 1 t 1 D etector R eflectan ce T ran sm itta n ce b F ig u re 1.6: Time resolved measurements, a) Detecting light a distance p away from a source which emits an impulse of light (tim e-resolved) and a steady state light with sinusoidally modulated intensity (frequency-resolved), b) Reflectance and transmittance geometries of the source and detector. From [24]. 14 In the time domain the optical response of the tissue to a short input pulse (<50 ps) gives the time of flight distribution, which is related to the average photon pathlength. In the frequency domain, the intensity of the light source is modulated at high frequency (107 to 10 9 Hz), and the signal is detected by a phase sensitive detector [16]. The advantages of tim es— resolved methods are: the measurements are relatively insensitive to boundary conditions (influence the magnitude of the signal rather than the time or frequency dependence); by adjusting the time range or frequency of the measurements, it is possible to perform absolute quantitative spectroscopy even in limited tissue volumes; spatially resolved measurements are not required (the emitting and detecting probes can be placed close to each other). The disadvantage consists in relatively expensive instrumentation. 1.3 C lin ical ap p lica tio n s The clinical applications of NIRS are based on extracting information from the absorption and/or scattering properties of tissues, which are then related to the function or structure of the tissue. 15 a) In d ia g n o stic sp ectroscop y, measurements of endogenous chromophores are used for: monitoring changes in regional tissue oxygenation and blood flow (brain, muscle, internal organs); characterization of tumors and breast tissue; measurements of endogenous cutaneous pigments; endoscopic moni toring of blood content and oxygenation of internal tissues (i.e.,mucosa of the gastro-intestinal tract); calibration of other optical diagnostic techniques (in vivo fluorescence spectroscopy or microscopy), to correct for the intrinsic tis sue absorption and scattering. b) O ptical im a g in g is based on tim e-resolved spectroscopy and facilitate detection of tumors and cerebral injury or visualization of blood vessels, tis sues slices, skin disorders and other tissues under the skin layers [16,26,27,28], Two types of imaging techniques are used: - Direct, in which measured properties of the light field itself consti tutes the image. Those include time— of— flight or time gating measure ments, which have been used for imaging of breast tissue [26] and scan of rat body [27]. - Reconstructive, in which optical information is used to generate a quantitative map of some optical property of the tissues (scattering or absorption coefficient) [28]. 16 2. TISSU E OXYGENATION-PHYSIOLOGICAL ASPE C T S 2.1. G en eral a sp ects In mammals, the oxygen (O2) plays a key role in energy metabolism of active tissue. The cessation of O2 supply results in loss of organ function within seconds or minutes, therefore a continual feed of adequate amount of O2 to tissue is the most vital task for living organism. Oxygen is taken up from the atmosphere into the lung mainly by gas bulk flow, and is transported from there to the O2 consuming tissues by circulation of blood. A model of the respiratory system of mammals is given by Weibel [17] (figure 2.1). 1 0 0 0 , FLOW _ PRESSURE RATE — GRADIENT * CONDUCTANCE i < PE* Pi V 0 j (A ) = ( P l- P E ) o j * G(A) V ^ A - B ls (P A -P b )o j * G (A -B ) in F A ' Pb- g BLOOD S V 0 j (B-C)= (P b -R B o 2 * G (B -C ) V 0 2 ( C ) = ( F t - ? ) o , ■ G (C ) SINK F ig u re 2.1: Model of the respiratory system of mammals. The O2 flow rate through each of the sequential steps is the product of a pressure difference and a conductance. From [17]. 17 Several aspects must be pointed out: a) The flow of O2 through this system is determined by the rate at which O2 is consumed by mitochondria in the process of making ATP. b) The transfer of O2 through any element of the respiratory system is given by the product of a diffusion conductance G, by the difference between O2 partial pressure inside and outside the element: Mo 2 = G[P 0 2( i n ) - P 0 2(out)] According to Fick's first law of diffusion, the partial pressure gradient acts as the driving force for diffusive flux. Therefore the above equation can be written [17,29]: M » , = 4 [ I V in )-'Vo u t > ] Where K = 0l D% is the Krogh's diffusion coefficient (a = solubility coefficient, Dr = diffusivity coefficient), S is the cross section of diffusion layer, and Ax represent the layer thickness. c) Oxygen diffuses through the cell membrane and is transported in blood plasma by free diffusion and by convection, which are proportional with the concentration gradients. 4. Inside the red blood cell and red muscle cell, oxygen reacts chemically with hemoglobin or myoglobin and is transported by both free and hemoglobin/myoglobin— facilitated diffusion. Hemoglobin (Hb) and myoglobin (Mb) are the main 0 2 carriers. The degree of binding to a carrier C is expressed by the saturation S': where [C--O2] is the concentration of the carrier— 0 2 compound and [Cp] is the total carrier concentration. The 0 2 dissociation curve for Hb and Mb is presented in figure 2.2 [16,29]. PQ ^ at which a saturation of 50% is achieved is call P50. 100-j- 9 0“ 8 0- - 7 0 " * 6 0“ £ 5 0 “ Mb Hb Oxygsn tension, torr F ig u re 2.2: Dissociation curve for Hb and Mb. From [29]. 19 5. The blood which flows at a rate Qg will carry along an O2 flow: V o 2 = Q b ' C o 2 Therefore O2 delivery by the blood and the oxygen uptake V02 depend on blood flow (Qg) and the difference in O2 content (Cn ) between arterial (a) 2 and venous (v) blood [17]. Vo2 = Qb t^o2(a) VQ represents the most important physiological 0 2 sink and is given by the oxidative metabolic requirement of the tissues. Hypoxia as well as a low cardiac output generates diverse physiologic responses which are function of the sensitivity of different organ systems to oxygen deprivation. Usually the local reaction is to increase blood flow and O2 extraction. Since 0 2 extraction equals the ratio of Vq2 to Qb, when Q decreases, O2 extraction must increase to maintain Vq^. Consequently, high 0 2 extraction or low mixed venous saturation are indicators of an imbalance between Vq^ and Qb [13]. On the other hand, the disorders that impair oxygen delivery to tissues (hypoxia, anemia, hypovolemia, hyperdynamic sepsis) involve circulatory compensation which distributes a limited 0 2 supply to tissues 20 with high requirement for oxygen. The supply to the brain and heart is conserved at the detriment of skeletal muscle, kidney and splanchnic organs. 2.2. M u scle o x y g en a tio n Several characteristics make skeletal muscle distinct from the other types of tissue: - Muscle fibers have a preferential direction along which the muscle contracts, and blood capillaries tend to be oriented along the fibers. Capillary anastomoses which are not aligned with muscle fibers and the heterogeneity of capillary hemodynamics disturb that arangement [30]. - Myoglobin stores O2 and facilitates the O2 diffusion in deeper parts of the muscle fibers. In a heavily working muscle fiber, myolobin reduces the P n drop between sarcolemma and fiber center. 2 - Skeletal muscle varies its O2 requirement from the resting state is the greatest variability in energy turnover from the resting state to maximal activity. As shown in figure 2.3 its Vq^ reaches the second largest absolute value of all organs and may vary between rest (0.2 ml O2 100 g^min'1) and maximum performance (16 ml O2 100 g'1 miri1 for electrical stimulation) by a factor of 80, thus covering a range that is much larger that in any other tissue [29]. skeletal muscle brain cortex kidney F ig u re 2.3: O2 consumption rate of various organs. Lightly shaded portions of columns represent variability. From [31].] 22 3. EXPERIM ENTAL M ETHODS 3.1. In stru m en ta tio n The experimental setup used for this study consists of: a NIRS system and a pressure generator ensemble (figure 3.1). The NIRS system was similar to those described by others authors [2,8,91- Pulses of near-infrared light with wavelengths 775, 810, 862 and 904 nm generated by four diode lasers (Laser Diode Inc.) were transmitted to the surface of the forearm by means of 4 factory-coupled optic fibers grouped in a bundle. Each diode laser was pulsed in sequence at a rate of 450 Hz and with a pulse duration of 60 ns. A detecting fiber connected to photomultiplier (Hamamatsu, R936) was used to collect diffusely reflected light from the tissue. Emitting and detecting fibers were fixed by a mechanical support which maintained constant the fibers geometry. The experiments were made for three geometrical arrangements: 1.5 cm distance and 90° angle; 1.2 cm distance and 60° angle; and 3 cm distance and 90° angle. The photomultiplier output current was processed by means of a high bandwidth transresistance amplifier and a sample-and-hold device for on-line computer sampling (1 laser pulse/sec) and storage of the optical signals. 23 P ulse O xim eter D iode lasers Sam ple and hold Am plifier PMT P iston Pump COMPUTER Converter W aveform G enerator F igure 3.1: Experimental setup The second part, pressure generator ensemble incorporates a piston pump driven by a stepper motor which was used to inflate in a periodic fashion a sphygmomanometer cuff placed around the right upper arm and to induce sinusoidal oscillations of the venous pressure (frequency 1— 2 cycles/min; amplitude, 10— 15 mm Hg). The oscillations frequency was set by a waveform generator. A pressure transducer connected to the pressure cuff were used to measure the pressure variations. 3.2. E xp erim en tal p rotocol Two groups of six healthy adults each (4 males, 2 females; 20-38 years old) participated in the experiment. The first group followed the protocol: a) venous occlusion, b) arterial occlusion (ischemia) c) oscillatory variation of venous pressure. During this protocol the geometrical parameters were 1.5 cm distance and 90° angle between emitting and detecting fibers. For the second group, measurements were made at three levels of occlusion (1.2 cm distance and 60° angle probes geometry). On two subjects, data were acquired at the same levels of occlusion, but different geometrical parameters (3 cm distance, 90° angle). The subjects were comfortably seated during the measurement period. Resting arterial pressure was measured at the beginning of each experiment. Arterial hemoglobin saturation was 25 monitored with a pulse oximeter sensor attached to the finger. In the first protocol, the optical signals were recorded continuously while the cuff* was inflated to 50 mm Hg (venous occlusion) and 200 mm Hg (arterial occlusion) pressure. The time sequence was of 1 min baseline 7 min occlusion and 7 min recovery. In the following three maneuvers, the cuff pressure was raised to 20, 40, or 60 mm Hg. Periodic oscillations of the cuff* pressure (1 cycle/min) were initiated 30 sec after the cuff pressure rise and continued for 14 min. A similar set of 3 venous occlusion maneuvers was repeated while the oscillation frequency was 2 cycles/min. A 5 min waiting period was allowed between each measurement period. The second protocol was directed toward quantifying the NIRS response of skeletal muscle during impaired venous outflow or arterial inflow. The optical signals were registered continuously while the cuff pressure was elevated at 50, 100, and 150 mm Hg. The time sequence was of 3 min baseline, 7 min occlusion, and 7 min release. Also, a 5 min period was used between each measurement. 2 6 4. DATA ANALYSIS Data analysis included the following steps: - Filtering of raw optical data, which aimed to removing experimental noise. - Four wavelength NIRS algorithm in association with least-square minimization, which deconvolved the oxygenated (HbO), deoxygenated (HbR), and Cytochrome aa3 (Cyt) values from the optical signals. - Fourier analysis, applied to the concentrations of HbO, HbR, Hbt, and Cyt to compute their amplitude and phase response relative to cuff pressure variation. - Statistical analysis. - Oxygen consumption computation. 4.1. F ilter in g High frequency noise from the optical signals was filtered out by a low pass filter (6 poles butterworth filter). The filter cutoff frequency was changed for different protocol measurements. A value of 0.02 Hz was used for static 27 conditions, 0.05 Hz and 0.07 Hz for dynamic conditions, i.e., oscillatory variation of the venous pressure at 1 cycle/min and 2 cycles/min, respectively. The coefficients of the butterworth filter were used in a recursive digital system with zero phase shift given by the following equation: N N g = Y b f - Y a g o m A-j n* m — n Aj n P m - n n - - N n = 1 Where a sample of the filter output (gm) is given in terms of past (second sum) as well as future (first sum) samples of the input. 4.2. N e a r -in fr a r e d sp ectro sco p y algorith m As shown in Chapter 1, the near-infrared absorption spectra of HbO, HbR and Cyt are broad and overlap extensively. The main role of NIRS algorithm is to express changes in the relative quantities of HbO, HbR and Cyt to each other in correspondence to total absorption changes at each several wavelength which are used during investigation. A four wavelength algorithm was used to convert the variation in optical signals into changes in the amounts of HbO, HbR, and Cyt. Near-infrared 28 algorithms use experimentally determinated extinction coefficients for HbO, HbR, and Cyt at different wavelengths. The extinction coefficients used in this study were reported by Wray and coworkers [21]. The relationship between absorption and concentration is given by Lambert- Beer law: 70 OD = l o g y = zCL Optical density (OD) measurements at times tj and t£ yield: AOD = log I, -lo g I, = e(C ,-C ,)L l2 * i where 1. represent the remitted light at certain time 1, and I. at time 2. l\ c2 The difference (C2-Cj) gives the variation in concentration between tj and t£. The total optical density difference for each wavelength can be written as: A 0 D X1 = [ a l ACHbO \ 1 + b l ACHbR \ 1 + Cl ACC y n i ) ' L AODk2 = ^a2ACHbQx^ + b2ACHbRx^ + c2ACCytx^j - L AODX3 = [ a 3ACHbQx^ + b3ACHf)Rx^ + c3ACCytx^ y L A OD \a = [ a ^ C Hb0x^ + bAACHbRx^ + cAACCytx^ y L where a, b, c and d are the extinction coefficients and AC the changes in 29 concentration of each absorption component. Taking in consideration the above observation the system can be written in a matrix form as follow: A OD^ AODX: A OD\, A OD\, b j C j a 2 ^2 C2 a 3 b 3 C3 b4 c4 AC HbO AC JIbR AC Cyt J • L or ODx = A - Ch L Where QD^represent optical densities values vector, A is extinction coefficients matrix, is concentration values vector and L is the optical pathlength. L was approximated according to the formula: L = B ■ I Where I is the geometrical distance between emitting and detecting fibers and B is a pathlength factor for tissue which takes in the account the scattering effect. The value of B for skeletal muscle was estimated between 3.6 and 5.0 [7,10]. Solving the system by least squares minimization a - yields: 30 Ch L = 1 AT ■0D % Where M; are the measured values and Hi represent the fitted values. The variations in total hemoglobin Hbt were computed from the relative concentration values of HbO and HbR: ACHbt ~ ACHbO + ACHbR 4.3. F o u rier a n a ly sis Fourier analysis was used on HbO, HbR, HbT concentration values acquired during oscillatory variation of venous pressure for both frequencies and for all mean cuff pressure levels. The amplitude and phase response relative to cuff pressure oscillation were computed. 4.4. S ta tistic a l a n a ly sis Concentration changes with respect to baseline at different times and for all mean cuff pressures were compared for both groups of subjects. The changes in amplitude and phase response with increased mean cuff pressure were 31 also analyzed. The methods used for analysis were ANOVA with repeated measures and Tukey post-hoc test. 4.5. O xygen co n su m p tio n estim a tio n The oxygen consumption Vn was computed using the method presented by Cheatle [7] and De Blasi [10], which assumes that the conversion rate from oxygenated (HbO) to deoxygenated (HbR) hemoglobin is proportional to V0 : Vo2 0 6 (AC HbO ~ A C HbR) ' L The effective Vq^ was determinated by regression analysis applied on the conversion HbO-HbR curve and taking in the account the molecular ratio between hemoglobin and oxygen (1:4). 32 5. EXPERIMENTAL RESULTS The subjects arterial blood pressure was in the normal range: 110-135/65-85 mm Hg. The pulse rate was between 60 and 80 pulses/min. The filtering effect on optical signals is shown in figures 5.1— 5.3. Figure 5.1 is typical for optical signals acquired in static conditions. Figures 5.2 and 5.3 are for those measurements obtained during oscillatory variation of venous pressure, 1 cycle/min and 2 cycles/min, respectively. 5.1. F ir st p ro to co l ob serv a tio n s A typical variation of HbO, HbR and Hbt concentration values (pmol/100 g tissue) during venous occlusion is shown in figure 5.4. All 3 signals, show an increase of the relative concentrations during occlusion followed by return to baseline values during the recovery period. Figures 5.5 shows the average muscle HbO response (mean +/-SE) during venous occlusion for all six sub jects. A statistically significant increase of the HbO signal from 0 to 1.2 |imol/ 100 g tissue was observed during the entire venous occlusion maneuver. The HbO increase was indicative of engorgement of the forearm microvasculature 33 with arterial blood. The HbO signal returned to baseline shortly after defla tion of the sphygmomanometer cuff. A similar behavior was observed for the HbR and Hbt optical signals (figures 5.6 and 5.7) indicating that venous blood also accumulated in the forearm vascular bed. Figure 5.8 shows the characteristic changes of HbO, HbR and Hbt concentra tion values during arterial occlusion (200 mm Hg). After occlusion, a simulta neously increase of HbR and decrease of HbO concentrations were observed. Figure 5.9 displays average muscle HbO response (mean +/-SE) during arte rial occlusion. A statistically significant decrease of the HbO concentration, ranging from 0 to 1.3 p.mol/100 g tissue, was observed during the entire ai'te- rial occlusion maneuver. The decrease was linear until time 3 (approximative 3.5 min after occlusion was induced). The rate of decrease was markedly reduced between times 3 and 5 (no statistical difference). The HbO decrease was suggestive of muscle tissue deoxygenation during ischemic conditions. An abruptly increase and overshoot (1 pmol/lOO g tissue above baseline) was observed immediately after release. Times 6 and 7 show a significant increase over the baseline which reflected a large inflow of oxygenated blood at the beginning of recovery period. In opposition to HbO response, the HbR average concentration (figure 5.10) continuously increased (0-1.3 gmol/100 g tissue) immediately after arterial occlusion. The increase reaches a plateau after time 3. No significant change was observed between time 3 and 5. This behav ior is complementary to HbO response and confirms the tissue deoxygenation. A decrease of concentration under the baseline values was recorded when the cuff pressure was released. Figure 5.11 shows the averaged Hbt concentra tion values trend. No significant variation from baseline were observed dur ing the occlusion. An significant increase of Hbt concentration (about 0.8 q.mol/100 g tissue) was recorded only shortly after cuff release. This also indi cated an increase of blood flow in forearm vasculature. The typical variation of HbO, HbR and Hbt concentrations during oscillatory variation of venous pressure at 20 mm Hg mean cuff pressure and 1 cycle/min is shown in figure 5.12. All 3 hemoglobin concentration signals followed rela tively closely the pressure signal oscillations. Similar patterns were obtained when the mean cuff pressure was increased to 40 mm Hg (figure 5.13) and 60 mm Hg (figure 5.14). However, a decrease of the oscillation amplitude and an increase of phase the phase delay between hemoglobin and pressure signals were observed with increased mean pressure. Comparable trends for the HbO, HbR, Hbt concentrations were observed when the venous pressure oscillations were 2 cycles/min. Characteristic patterns are display in figure 5.15 for 20 mm Hg, figure 5.16 for 40 mm Hg and figure 5.17 for 60 mm Hg mean cuff pressure. Figure 5.18 presents the average amplitude of the HbO oscillations as a func tion of mean cuff pressure. For oscillation frequencies of 1 cycle/min and 2 35 cycles/min, a significant decrease of the HbO oscillations was observed as the mean pressure was increased. Additionally, the oscillations of the HbO signal observed at 1 cycle/min were in general larger than those observed at 2 cycles/min. Similar trends were observed on the reduced hemoglobin and total hemoglobin signals (figure 5.19 and 5.20). These observations indicate that increasing the average pressure in the forearm vascular bed reduced the effect of outflow pressure variations on vascular volume variations. Oscillations of the HbO and Hbt signals were delayed in phase compared with the cuff pressure oscillations (figures 5.21 and 5.22). When the oscilla tion frequency was 1 cycle/min, the phase delay significantly increased with mean cuff pressure, ranging from about -20 degrees at 20 mm Hg to about -60 degrees at 60 mm Hg for both signals. The phase relationship between fore arm hemoglobin content and oscillatory cuff pressure was less well defined when the pressure oscillations occurred at 2 cycles/min. Oscillations in reduced hemoglobin HbR (figure 5.23) appeared delayed in phase compared to the oscillations of HbO and Hbt. 5.2. S econ d p ro to co l o b serv a tio n s Figure 5.24 shows the HbO, HbR and Hbt concentrations during occlusion at 50 mm Hg, 100 mm Hg and 150 mm Hg, and for emitting-detecting fibers 36 geometry of 1.2 cm distance and 60° angle. When the cuff pressure was ele vated to 50 and 100 mm Hg an increase of the concentrations for all 3 compo nents was observed, representative of accumulation of both venous and arterial blood in the forearm vasculature. However the HbR and Hbt concen tration changes were larger at 100 mm Hg than at 50 mm Hg. The occlusion at 150 mm Hg was characterized by increase of HbR to 1.2 |imol/100 g tissue and decrease of HbO with 2.5 (imol/100 g tissue. This behavior indicated that the hypoxic conditions were reached in forearm. Figure 5.25 shows the variation of HbO, HbR and Hbt for the same subject with a different fiber geometry (3 cm distance and 90° angle). The measure ments show similar trends of concentration variation with slightly larger val ues of hemoglobin concentration in tissue for HbO, HbR, Hbt during 50 and 100 mm Hg occlusion and for HbR during 150 mm Hg occlusion. The HbO concentration was similar during 150 mm Hg occlusion.This behavior demon strated that both fiber geometry leaded to comparable results. The averaged HbO, HbR and Hbt responses during 50 mm Hg (venous occlu sion) for all six subjects show significant increase during the entire occlusion maneuver. The values over the intervals 1— 2 min, 2-3 min and 3-4 min after occlusion (times 1, 2 and 3), 2— 3 min and 1— 2 min before cuff pressure release (times 4 and 5), 1— 2 min and 2— 3 min after cuff pressure release (times 6 and 7) are presented in figure 5.26. Among the times 1— 5 no significant change in 37 HbO (0.7 (imol/100 g tissue) and Hbt (1.5 pmol/100 g tissue) concentrations was observed. However, a small increase of HbR concentration was recorded. During 100 mm Hg occlusion average HbO, HbR and Hbt responses (figure 5.27) also showed a significant increase. A relatively constant concentration of HbO (about 1.3 pmol/100 g tissue) was maintained in muscle during occlu sion (times 1— 5). HbR and Hbt concentration displayed a significant increase from time 1 to time 5. This demonstrated an accumulation of deoxygenated blood mainly due to increased concentration of HbR in forearm muscle. Figure 5.28 shows average muscle HbO response during 150 mm Hg occlu sion. A statistically significant decrease (2 (imol/100 g tissue) of the HbO con centration was observed. The decrease was linear during entire occlusion (time 1 to 5). The HbO decrease indicated muscle tissue deoxygenation. An increase was observed immediately after release. Times 6 show a significant increase over the baseline value, however time 7 suggested a return to base line shortly after cuff pressure release. HbR average concentration (figure 5.29) show a continuous increase immediately after elevation of cuff pressure. A significant continuous increase was recorded from time 1 (0.6 (imol/100 g tissue) to time 5 (1.2 (imol/100 g tissue). A decrease of concentration under the baseline values was recorded when the cuff pressure was released. No sig nificant variation of Hbt concentration relative to baseline was observed dur ing occlusion. 38 5.3. O xygen co n su m p tio n The oxygen consumption rate derived from the data acquired during ischemic conditions, 200 mm Hg occlusion (first protocol) and 150 mm Hg occlusion (second protocol) are presented in the following. Figure 5.30 shows the averaged changes of conversion rate from oxygenated to deoxygenated hemoglobin (AHbO— AHbR) during occlusion at 200 mm Hg. A fast desaturation rate was observed immediately after occlusion. After cuff release a rapid recovery of Hbt and a slower recovery in oxygen content occurred. A similar behavior was observed during 150 mm Hg occlusion (fig ure 5.31). The oxygen consumption rate for consecutive sequence of 20 sec postocclusion are presented in figure 5.32. During the first 20 sec after occlu sion the oxygen consumption was 4.03 pmol/min/100 g tisssue at 200 mm Hg and 3.6 (imol/min/100 g tisssue at 150 mm Hg. An decrease of oxygen con sumption was observed after the first minute of occlusion, which indicated muscle oxygen depletion. 39 O p t i c a l s i g n a l - n o o s c il la t io n s 2800 S 2 60 0 2 40 0 ,5*2200 2000 “ 5 1800 c c 1600 200 4 0 0 6 0 0 Tim e (seconds) 8 0 0 1000 1200 Filtered optical signal 2600 2 60 0 2 2400 2000 1800 1600 200 4 0 0 6 0 0 Tim e (seconds) 8 0 0 1000 1200 F igu re 5.1: Filtering effect (bottom) on optical signal (upper) obtained in static m easurem ents-no oscillation. Optical signal - F -1 cycle/min 1800 £ 1600 % 1000 □ T 800 200 400 6 0 0 600 Tim e 1000 Tim e (seconds) 1200 1400 1600 1800 Filtered optical signal 1800 1600 .5* 1400 © 1 2 0 0 - 1000 200 400 600 600 Tim e 1000 Tim e (seconds) 1200 1400 1600 1800 F igu re 5.2: Filtering effect (bottom) on optical signal (upper) obtained in dynamic measurements— venous pressure variation at 1 cycle/min. O p tic a l s ig n a l - F - 2 c y c le / m i n 2 40 0 5*2 0 0 0 1800 1600 8 0 0 1000 Tim e (seconds) Filtered optical signal 1200 1600 1800 2 15 0 2100 g 2 05 0 1950 ® 1900 1850 6 00 200 400 800 Time 1000 Tim e (seconds) 1200 1400 1600 1800 F igu re 5.3: Filtering effect (bottom) on optical signal (upper) obtained in dynamic measurements— venous pressure variation at 2 cycle/min. Concentration change (pmol/100 g tissu e) Venous occlusion - 50 mmHg LEGEND HbO HbR Hbt Pressure 100 200 300 400 500 600 700 800 900 Time(seconds) F ig u re 5.4: HbO, HbR, Hbt relative concentration value changes during venous occlusion (50 mm Hg). HbO - Venous occlusion 0 3 Z 3 C /3 C /3 V -■ 0 3 O O 2.57 1.71 O £■ 0.64 — O _ Q X ■0.21 | 1 2 II 0 u ii 0.0 3.0 6.0 9.0 Time (min) 12.0 F ig u re 5.5: Averaged muscle HbO response (mean +/-SE) to 50 mm Hg venous occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1 -2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 2— 3 min before release; 4: average over the interval 1— 2 min before release; 5: average over the interval 1— 2 min after release; 6 : average over the interval 2— 3 min after release. 43 Hbr - Venous occlusion 1.71 1.07 O ) o o 0.32 -1.07 2 0 4 6 8 10 12 Time (min) F igu re 5.6:Averaged muscle HbR response (mean +/-SE) to 50 mm Hg venous occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1 -2 min after occlusion; 2: average over the interval 2— 3 min after occlusion; 3: average over the interval 2— 3 min before release; 4: average over the interval 1— 2 min before release; 5: average over the interval 1— 2 min after release; 6 : average over the interval 2 -3 min after release. 44 Hbt - Venous occlusion 3.21 -1.07 0.0 3.0 6.0 9.0 Time (min) 12.0 F ig u re 5.7: Averaged muscle Hbt response (mean +/-SE) to 50 mm Hg venous occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1 -2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 2 -3 min before release; 4: average over the interval 1— 2 min before release; 5: average over the interval 1 -2 min after release; 6 : average over the interval 2 -3 min after release. 45 Arterial occlusion - 200 mmHg LEGEND HbO HbR Hbt Pressure c-4 100 200 300 400 500 600 700 800 900 Time(seconds) F igu re 5.8: HbO, HbR, Hbt relative concentration value changes during arterial occlusion (200 mm Hg). 46 HbO - Arterial occlusion — . 1.71 C D O c n .CD 1.07 H D) g 0.21 4 O -0.64 0 J O 1 -1.28 - -2.14 4 5 D * 4 6 8 10 12 14 16 Time (min) F igu re 5.9:Averaged muscle HbO response (mean +/-SE) to 200 mm Hg arterial occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1 -2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3 -4 min after occlusion; 4: average over the interval 2 -3 min before release; 5: average over the interval 1— 2 min before release; 6: average over the interval 1 -2 min after release; 7: average over the interval 2 -3 min after release. 4 7 HbR - Arterial occlusion 1 .9 2 CD S 0.85 0.21 - 0.21 -0.85 4 6 8 10 12 14 16 2 0 Time (min) F igu re 5.10: Averaged muscle HbR response (mean +/-SE) to 200 mm Hg arterial occlusion. Line is representative of cuff pres sure trace. 0 : baseline; 1: average over the interval 1— 2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3— 4 min after occlusion; 4: average over the interval 2-3 min before release; 5: average over the interval 1 -2 min before release; 6: average over the interval 1 -2 min after release; 7: average over the interval 2— 3 min after release. 48 Hbt - Arterial occlusion ^ 1-07 C D C O ■ B 0.42 D> O O O E 1 2 3 - 0.21 - -0.85 - -1.50 II II r 1 i 1 i 1 i 1 i 1 i 1 i 1 i 0 2 4 6 8 10 12 14 16 Time (min) F igu re 5.11: Averaged muscle Hbt response (mean +/-SE) to 200 mm Hg arterial occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1--2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3 -4 min after occlusion; 4: average over the interval 2-3 min before release; 5: average over the interval 1— 2 min before release; 6: average over the interval 1— 2 min after release; 7: average over the interval 2— 3 min after release. 4 9 HbO o scilla tio n s t o o o 0.1 CO o > £ - 0.1 8 - 0.2 c ( 100 200 3 0 0 4 0 0 5 0 0 T im e (se c o n d s ) a 6 0 0 7 0 0 8 0 0 9 0 0 HbR oscillation s t o S 0 .0 5 T O C - 0 .0 5 100 200 3 0 0 4 0 0 5 0 0 T im e (se c o n d s) 7 0 0 6 0 0 8 0 0 9 0 0 b Hbt o scilla tio n s t o T O is -0 .2 100 200 3 0 0 4 0 0 Tim e (se c o n d s ) C 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 Figure 5.12: Oscillations of HbO (a),HbR (b) and Hbt (c) signals (continuous line) for 20 mm Hg mean cuff pressure and 1 cycle/ min frequency (dotted line). 50 HbO oscillations 0.1 O ) cn - 0.1 - 0.2 100 200 300 400 Time (seconds) a 500 600 700 800 9 0 0 HbR oscillations ■B 0.05 - 0.05 - 0.1 100 200 300 400 Time (seconds) 500 600 700 800 900 O b Hbt o scilla tio n s cn 0.1 cn c - 0.1 ■ g - 0.2 8 - 0 .3 100 200 3 0 0 4 0 0 5 0 0 T im e (se c o n d s ) C 6 0 0 7 0 0 8 0 0 9 0 0 Figure 5.13: Oscillations of HbO (a),Ht)R (b) and Hbt (c) signals (continuous line) for 40 mm Hg mean cuff pressure and 1 cycle/ min frequency (dotted line). Concentration changes (jjmoi/10 0 g tissue) Concentration changes (pmol/10 0 g tissue) Concentration changes (jjmol/10 0 g tissue) HbO oscillations 0.1 100 200 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 T im e (se c o n d s) a HbR o scilla tio n s 0.1 100 200 3 0 0 4 0 0 T im e 5 0 0 T im e (se c o n d s ) 6 0 0 7 0 0 8 0 0 9 0 0 b Hbt o scilla tio n s 0.1 100 200 3 0 0 4 0 0 T im e (se c o n d s ) 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 Figure 5.14: Oscillations of HbO (a), HbR (b) and Hbt (c) signals (continuous line) for 60 mm Hg mean cuff pressure and 1 cycle/ min frequency (dotted line). Concentration changes (pmol/100 g tissue) Concentration changes (pmoi/100 g tissue) Concentration changes (jjm ol/100 g tissue) H bO oscillations 0.5 0.5 0 100 200 300 400 500 600 700 a o o Time (seconds) a HbR o scilla tio n s 0 .5 100 200 3 0 0 4 0 0 Tim e (s e c o n d s ) 5 0 0 7 0 0 6 0 0 8 0 0 b Hbt oscilla tio n s 0 .5 100 5 0 0 200 3 0 0 4 0 0 6 0 0 7 0 0 8 0 0 T im e (se c o n d s ) Figure 5.15: Oscillations of HbO (a),HbR (b) and Hbt (c) signals (continuous line) for 20 mm Hg mean cuff pressure and 2 cycle/ min frequency (dotted line). H bO oscilla tio n s 1.5 o > o E 0 .5 O) £ - 6 - 0 . 5 8 - 1 . 5 100 200 3 0 0 4 0 0 T im e (se c o n d s ) a 5 0 0 6 0 0 7 0 0 8 0 0 HbR oscillations - 0.5 100 200 300 400 Time (seconds) 500 600 700 800 b Hbt o scilla tio n s cn o O - 1 S -2 100 200 3 0 0 4 0 0 T im e (s e c o n d s ) C 5 0 0 6 0 0 7 0 0 8 0 0 Figure 5.16: Oscillations of HbO (a), HbR (b) and Hbt (c) signals (continuous line) for 40 mm Hg mean cuff pressure and 2 cycle/ min frequency (dotted line). 54 Concentration changes (pmol/10 0 g tissue) Concentration changes (pmol/10 0 g tissue) Concentration changes (}jmol/10 0 g tissue) HbO oscillations 1.5 0 .5 200 100 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 T im e (s e c o n d s ) a H bR o scilla tio n s 2 1 0 1 •3 200 3 0 0 0 100 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 T im e ( s e c o n d s ) b Hbt o sc illa tio n s 2 1 0 ■ 2 3 0 0 0 100 200 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 T im e (s e c o n d s ) Figure 5.17: Oscillations of HbO (a), HbR (b) and Hbt (c) signals (continuous line) for 60 nun Hg mean cuff pressure and 2 cycle/ min frequency (dotted line). HbO amplitude response a > •o 1 3 0 20 40 60 80 Mean cuff pressure F ig u re 5.18: Oscillations of HbO signal as a function of mean cuff pressure and oscillation frequency of 1 cycle/min (dotted bar) and 2 cycles/min (filled bar). co n . C D 0.1 D C _ Q X HbR amplitude response C D ■ D = 3 • 4— * Q. E C O C D > 0 .3 0.2 0.0 20 40 60 Mean cuff pressure 80 F igu re 5.19: Oscillations of HbR signal as a function of mean cuff pressure and oscillation frequency of 1 cycle/min (dotted bar) and 2 cycles/min (filled bar). 57 Hbt amplitude response 0 .8 ---------------------------------------------------------------------------------- 0 20 40 60 80 Mean cuff pressure F igu re 5.20: Oscillations of Hbt signal as a function of mean cuff pressure and oscillation frequency of 1 cycle/min (dotted bar) and 2 cycles/min (filled bar). 58 Phase response (degrees) o.o HbO phase response - 20.0 - -40.0 - -60.0 - -80.0 20 40 60 Mean cuff pressure 80 F igu re 5.21: Phase delay between Hbo and cuff pressure oscilla tions as a function of mean cuff pressure for oscillation frequency equal to 1 cycle/min (dotted bar) and 2 cycles/min (filled bar) Phase delay (degrees) HbR phase response -20 -40 -60 - -80 - 20 40 60 Mean cuff pressure 80 F igu re 5.22: Phase delay between Hbt and cuff pressure oscilla tions as a function of mean cuff pressure for oscillation frequency equal to 1 cycle/min (dotted bar) and 2 cycles/min (filled bar) Phase delay (degrees) o.o -30.0 -60.0 - -90.0 Hbt phase response 20 40 60 Mean cuff pressure 80 F ig u re 5.23: Phase delay between HbR and cuff pressure oscilla tions as a function of mean cuff pressure for oscillation frequency equal to 1 cycle/min (dotted bar) and 2 cycles/min (filled bar) O cclu sio n - 5 0 m m H g LEGEND HbO ■ HbR Hbt Pressure o > 1.5 cn c 8 - 0 .5 100 200 3 0 0 T im e (se c o n d s) a 4 0 0 5 0 0 6 0 0 O cclu sio n - 1 0 0 m m H g O ) 3- 2 < D O 200 4 0 0 6 0 0 T im e (se c o n d s) b 8 0 0 1000 1200 O cclu sio n - 1 5 0 m m H g cn < “ - 1 cn • -5-2 T o - 3 200 4 0 0 6 0 0 T im e (s e c o n d s) C 8 0 0 1000 1200 F ig u re 5.24: HbO, HbR, Hbt relative concentration value changes during 50 mm Hg occlusion (a), 100 mm Hg occlusion (b), and 150 mm Hg occlusion (c). Fibers optic geometry: 1.2 cm distance, 60° angle. 62 O cc lu sio n - 5 0 m m H g LEGEND HbO - HbR Hbt Pressure O ) 6 0 0 200 3 0 0 T im e (s e c o n d s) a 4 0 0 5 0 0 100 O cclu sio n - 1 0 0 m m H g 8 cn o 6 < u O ) I 2 0 2 * — » c <D O c •2 1200 4 0 0 6 0 0 T im e (se c o n d s) 8 0 0 1000 0 200 b O cclu sio n - 1 5 0 m m H g cn o cn -2 y -4 1200 4 0 0 6 0 0 T im e (se c o n d s) C 8 0 0 1000 200 F ig u re 5.25: HbO, HbR, Hbt relative concentration value changes during 50 mm Hg occlusion (a), 100 mm Hg occlusion (b), and 150 mm Hg occlusion (c). Fibers optic geometry: 3 cm distance, 90° angle. 63 C D 1 3 C /3 C /3 2.14 +-» 0 3 O 1.50 O O 107 Hb - 50 mmHg occlusion 0.42 _ Q X oc X - 0.00 o _ Q X _ -0.64 f v o Legend I I Hbt V HbR + HbO 6 r 1 I I I I I I I I I I I I I I [— 2 4 6 8 10 12 14 16 Time (min) F igu re 5.26: Averaged muscle HbO, HbR and Hbt response (mean +/-SE) to 50 mm Hg venous occlusion. Line is representa tive of cuff pressure trace. 0: baseline; 1: average over the inter val 1 -2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3 -4 min after occlu sion; 4: average over the interval 2 -3 min before release; 5: aver age over the interval 1 -2 min before release; 6 : average over the interval 1 -2 min after release; 7: average over the interval 2 -3 min after release. 64 < x > Hb -100 mmHg occlusion O T 4.28 V -» cn O 3.21 H O E 2.35 1.28 - S i J 021 JD ■ Xi -0.85 O JO X I t— I — r L e g e n d + HbO 6 7 i — I — i — I — i — I — r T 4 6 8 10 12 14 16 Time (min) F igu re 5.27: Averaged muscle HbO, HbR and Hbt response (mean +/-SE) to 100 mm Hg venous occlusion. Line is representa tive of cuff pressure trace. 0: baseline; 1: average over the inter val 1— 2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3— 4 min after occlu sion; 4: average over the interval 2— 3 min before release; 5: aver age over the interval 1— 2 min before release; 6 : average over the interval 1— 2 min after release; 7: average over the interval 2— 3 min after release. HbO -150 mmHg occlusion C D 3 C O C O O ) o o 1.52 -0.43 - O e ^ 3 - o 40 X -2.14 ' I ' I ' I 1 I 1 I 1 I 1 I 1 , 0 2 4 6 8 10 12 14 16 Time (min) F ig u re 5.28: Averaged muscle HbO response (mean +/-SE) to 150 mm Hg occlusion. Line is representative of cuff pressure trace. 0: baseline; l:average over the interval 1— 2 min after occlusion; 2: average over the interval 2— 3 min after occlusion; 3: average over the interval 3 -4 min after occlusion; 4: average over the interval 2— 3 min before release; 5: average over the interval 1— 2 min before release; 6: average over the interval 1— 2 min after release; 7: average over the interval 2— 3 min after release. 66 HbR -150 mmHg occlusion 1.71 1.07 O ) o o 0.64 -0.64 4 0 2 6 8 10 12 14 16 Time (min) F igu re 5.29: Averaged muscle HbR response (mean +/-SE) to 150 mm Hg occlusion. Line is representative of cuff pressure trace. 0: baseline; 1: average over the interval 1— 2 min after occlusion; 2: average over the interval 2 -3 min after occlusion; 3: average over the interval 3 -4 min after occlusion; 4: average over the interval 2— 3 min before release; 5: average over the interval 1 -2 min before release; 6: average over the interval 1— 2 min after release; 7: average over the interval 2— 3 min after release. 67 D eoxygenation rate 200 mm H g occlusion -2 - 8 - 3 - 300 400 500 600 700 Time (seconds) 0 1 0 0 200 800 — Hbt — HbO-HbR| F ig u re 5.30: Conversion rate from oxygenated to deoxygenated hemoglobin (200 mm H g occlusion). Deoxygenation rate 150 mm Hg occlusion ,9-3 100 200 Time (seconds) — HbO-HbR] Hbt F ig u re 5.31: Conversion rate from oxygenated to deoxygenated hemoglobin (150 mm Hg occlusion). Oxygen consumption 4.5 % 3.5 -■ • ' S 3 “ q ... o o o 5 2.5 - ■ 0.5 80 100 120 140 160 180 20 40 60 Time (seconds) -H -150 mm Hg 2 0 0 mm Hg | F ig u re 5.32: Skeletal muscle oxygen consumption rate. 69 6. LUM PED PARAMETER MODEL 6.1. M odel d esig n A lumped parameter model based on the model described by Braakman and coworkers [32] was designed and used in computer simulation to corroborate the experimental results. The forearm vasculature was modeled by sequence of resistive and compliant elements in which the main arterial resistance is located at the arteriolar level and the compliant elements are found in the vein and venules. Figure 6.1. shows the adopted model where Pa and Pv are the arterial and venous pressure, respectively. Pm is the tissue (muscle) microvasculature pressure and C represent the compliance of vein and venules. Ra and Rv symbolize the resistance of arterial and venous sides of the vasculature, respectively. P, m " R 7 F igu re 6.1: Lumped parameter model. 70 The differential equation that describes the system was derived from mass balance equation and was written in the following form: By rearranging the above expression we obtain: dP R + R P R m , T 3 a T v a v v a _ r\ d t m R R C “ R R C a y a V A sinusoidal driving force (frequency f) was used to simulate the oscillatory variation of the venous pressure. Therefore, the venous pressure was written as a sum of a steady state component, Pvl, and a dynamic component, Pv2: Pv = PV i + P V 2 sin(27u/) Consequently, Pm has also two components: one for static (Pmi), and other for dynamic (Pm2) conditions: P = P + Pm TO TO | TO 2 Hence, the analytical solution of the above mentioned differential equation was given by the sum of steady state solution: and transient solution: P v 9R a ( R a + R v ) r 2 i c / C R a R v P m 2 “ “ ~ 2 ^ 2 p „ 2 - 2 p 2 p 2 - p 2 R + R C0S ( 27CA ) ~ s i n ( 2 t c A ) R„ + R,r + 2 R R , + 4 k r C R„R„ L lxa + a v In agreement with the observations regarding blood flow and oxygen diffu sion in tissues (Chapter 2), the variation of Hbt (variation in blood volume) is proportional to the variation of Pm. The phase delay ( < j > ) between venous pres sure oscillation and blood volume variation is given by: < ) ) = atan f 2 7 u / C R a R v ' V R a + R v j and the amplitude (A) is expressed by: P v , R a ( R a + R v ) A = ( h * + E j + 2 R a R v + 4 ^ 2 / 2 C 2 r X ) “ S(|> The experimental results showed a simultaneous decrease of oscillations amplitude and increase of phase delay with increase mean cuff pressure level. The purpose of simulations was to identify the behavior of microvascu lature resistance and compliance which lead to analogous results. The simulations were run for different resistance and compliance values which are shown in Table 1. The simulations baseline was established by using the estimated values for compliance and resistance published by 72 Braakman et al, [32]. The Ra (i.e., about 129.1 kPa s cm'3 100 g, 1 kPa = 7.5 mm Hg) value comprises the arterial vessels, arterioles, capillaries and venules resistances. Rv (i.e., about 24.5 kPa s cm-3 100 g) represents the veins resistance. The compliance C (i.e., about 1.609 cm3 kPa 100 g'1) incorpo rates the arterioles and veins compliance. Table 1: S im u lation p aram eters S im u la tio n no. C ase no. C [cm3 kPa 100 g-1] Ra [kPa s cm-3 lOOg] Rv [kPa s cm-3 lOOg] 0 (Baseline value - B) 1.609 x 1.00 129.1 x 1.00 24.5 x 1.00 1 1 B x 1.10 B x 0.90 B x 0.90 C increase 2 B x 1.20 B x 0.80 B x 0.80 R decrease 3 B x 1.30 B x 0.70 B x 0.70 4 B x 1.40 B x 0.60 B x 0.60 2 5 B x 1.10 B x 0.95 B x 0.95 C increase 6 B x 1.20 B x 0.90 B x 0.90 R decrease 7 B x 1.30 B x 0.85 B x 0.85 8 B x 1.40 B x 0.80 B x 0.0 3 9 B x 0.90 B x 0.90 B x 0.90 C decrease 10 B x 0.80 B x 0.80 B x 0.80 R decrease 11 B x 0.70 B x 0.70 B x 0.70 12 B x 0.60 B x 0.60 B x 0.60 4 13 B x 0.90 B x 1.10 B x 1.10 C decrease 14 B x 0.80 B x 1.20 B x 1.20 R increase 15 B x 0.70 B x 1.30 B x 1.30 16 B x 0.60 B x 1.40 B x 1.40 5 17 B x 1.10 B x 1.10 B x 1.10 C increase 18 B x 1.15 B x 1.15 B x 1.15 R increase 19 B x 1.20 B x 1.20 B x 1.20 20 B x 1.25 B x 1.25 B x 1.25 73 6 .2. S im u la tio n s resu lts Figure 6.2 (Simulation 1) shows the phase and amplitude variations as a result of increased compliance in range 10-40% from the baseline and decreased resistance (both Ra and Rv) in the same range. In such conditions the phase delay increases while the amplitude decreases. Simulation 1 C increase / R d e c re a se 5.8E-05 g) 7 3 .5 CD 5.6E-05 - 5.4E-05 c 72.5 o Q. 5.2E-05 " D --5E-05 -4.8E-05 nj 71.5 4.6E-05 Case amplitude | i- p h ase F igu re 6.2: Phase and amplitude model amplitude response (cases 1— 4). C value increase (10— 40%) from baseline value (case 0). Ra and Rv decrease (10-40%) from baseline value (case 0). The second simulation was realized in the same conditions as the first one (C increase, Ra and Rv decrease), except that the Ra and Rv values decreased only in the range 5— 20%. This change lead to opposite results compared to 74 the first simulation: phase delay increase and amplitude decrease (figure 6.3). Simulation 2 C increase / R d ec re a se 4 .7 E -0 5 -4 .6 E -0 5 E 75.5 - o > CD “ D X 7 5 - -4.5E-05 q . --4.4E-05 0 " O 13 - 4.3E-05 ^ 8* 74.5 - 4 .2 E -0 5 4 .1 E -0 5 73.5 Case amplitude I p h ase F igu re 6.3: Phase and amplitude model amplitude response (cases 5-8). C value increase (10-40%) from baseline value (case 0). Ra and Rv decrease (5— 20%) from baseline value (case 0). During the third simulation the C, Ra, Rv values were decreased by 10— 40% from baseline. Figure 6.4 shows that, under conditions, the phase increased and the amplitude decreased. Simulation 4 was run for decrease of the compliance in the range 10— 40% and an increase of the resistances by 10— 40%. The simulation results (figure 6.5) shows a similar behavior of phase and amplitude as those obtained for simulation 1 and 3: increase of the phase and decrease of the amplitude. 75 Simulation 3 C d e c re a se / R d ec rea se 0 .0 0 0 1 6 •0.00014 S 70 -- o> a > 3 ^ to 65 -- - - 0.00012 - - 0.0001 £-60 • - 8E-05 co 55 ■ • • 6E-05 4E-05 C ase p h ase amplitude | < n c o CL V) Q ) u Q > * D Z J 2 = CL E < F ig u re 6.4: Phase and amplitude model amplitude response (cases 9— 10). C value decrease (10— 40%) from baseline value (case 0). Ra and Rv decrease (10-40%) from baseline value (case 0). Simulation 4 C d ec rea se / R in crease 5.8E-05 < j > 73.5 a > a > 2 , 73 5.6E-05 5.4E-05 c 72.5 5.2E-05 5E-05 ( I S 71.5 4.8E-05 4.6E-05 0 13 14 15 16 CD 0 1 c o CL C O CD k _ C D T 5 3 Q. E < C ase p h ase • amplitude F ig u re 6.5: Phase and amplitude model amplitude response (cases 13-16). C value decrease (10-40%) from baseline value (case 0). Ra and Rv increase (10— 40%) from baseline value (case 0). 76 An increase of the compliance and resistances in the range 10— 40% lead to an increase of the phase delay and a decrease of the amplitude (figure 6.6). This behavior is similar to that obtained in cases 5 to 8 (simulation 2). Simulation 5 C in crease / R increase 80 5E-05 --4.5E-05 a? 78 ■ - 2 , 58 7 7 - - - -4E-05 0 .76 3.5E-05 ■ “ 75 -- 3E-05 2.5E-05 0 17 18 19 20 Case p h a se -x - amplitude | F ig u re 6 .6 : Phase and amplitude model amplitude response (cases 17-20). C value increase (10-40%) from baseline value (case 0). Ra and Rv increase (10-40%) from baseline value (case 0). 77 7. DISCUSSION The present study was undertaken to determine whether NIR spectroscopy can be used to characterize noninvasively the vascular bed of the forearm. Measurements of changes in oxidized, reduced and total hemoglobin concen trations were performed in human skeletal muscle in both static (different levels of occlusion) and dynamic (occlusion+oscillatory variation of venous pressure) conditions. Fourier transform analysis was applied to characterize the amplitude and phase of relative hemoglobin content signals relative to cuff pressure oscillations. A regression analysis was used to quantify the oxy gen consumption rate during ischemic conditions. Variations in optical absorption were converted in hemoglobin content and oxygenation changes based on absorbance characteristics for oxygenated and reduced hemoglobin. Hemoglobin and myoglobin have similar optical absorp tion characteristics which prevent the differentiation of the 2 oxygen carriers by optical means. However, myoglobin is fully oxygenated except when tissue PQ is very low and its oxygenation status remained constant in the condi tions of venous occlusion. During arterial occlusion the signal is the result of both haemoglobin and myoglobin. 78 Venous occlusion (50 mm Hg) resulted in a significant increase of both oxi dized and reduced hemoglobin contents in the vascular bed of the forearm. The increase was maintained throughout the period of occlusion and reflected the accumulation of both arterial and capillary/venous blood in the forearm microvasculature. The increase in vascular blood volume could have resulted from microvascular distention and recruitment of normally unperfused seg ments of the capillary bed. In resting conditions, a fraction of skeletal muscle capillaries are non functional which could have been opened by the elevation of the venous pressure. Arterial occlusion (200 mm Hg) as well as the occlusion at 150 mm Hg gener ated a significant decrease of oxygenated hemoglobin and increase of deoxy genated hemoglobin. This behavior indicated that sudden forearm ischemia is associated with a very rapid depletion of muscle tissue O2 store that become maximal after minimum 4— 5 min of arterial occlusion. A longer occlusion time is required to reach maximal values at 150 mm Hg occlusion. During the recovery period tissue oxygenation and also total hemoglobin content increased rapidly to a level above the baseline. More than 4 min postocclusion period are required to return to normal conditions. This time is shorter (2 min) when ischemia was induced at 150 mm Hg. When pressure oscillations were superimposed to a steady elevation of the cuff pressure (venous occlusion), the optical signals reflected sinusoidal varia 79 tions added to a deflection similar to that observed during constant elevation of the venous pressure. For the total hemoglobin as well as oxy and deoxyhe moglobin signals the amplitude of the oscillations decreased when the mean cuff pressure was increased. An increase of HbO an Hbt phase delay relative to the cuff pressure signal was generated by increase mean cuff pressure. The elevation of mean cuff pressure generates a pressurization of the venous vas cular segments which leads to decrease of compliance and resistance of indi vidual veins and venules. In such conditions, one would expect an increase of blood volume oscillations associated with phasic variations of the venous pressure. However, the opposite observation in the present study is consis tent with recruitment of additional segments of the vasculature. This behav ior was demonstrated by the lumped parameter model. The computer simulations showed that a decrease in the blood volume oscillations in the capacitive segment was possible when the increase in compliance associated with recruitment was larger than the decrease in resistance or when both the compliance and resistance increased. In the simulation, these variations in mechanical properties also resulted in an increase of the phase delay between venous pressure and Hbt signals which was in agreement with our experi mental observations. The impaired venous outflow and arterial inflow induced by elevation of mean cuff pressure at 50 and 100 mm Hg was characterized by accumulation of both venous and arterial blood in tissue microvasculature. This demon- 80 strated a continuous oxygenation of tissue even when the arterial blood was partially occluded. The hypoxic conditions occurred only when the mean cuff pressure (150 mm Hg) was considerably above the mean arterial pressure (120 mm Hg). Consistent results were obtained under the same occlusion levels but differ ent emitting— detecting fibers geometry. Under the conditions of this experi ment, this indicated that the variation of optical signals (i.e., changes in hemoglobin concentration) could be mainly attributed to changes in muscle hemoglobin content and less to changes in skin and subcutaneous fat hemo globin content. When blood flow was interrupted the hemoglobin desaturation rate was pro portional to muscle oxygen consumption. The regression analysis applied on different time sequence during deoxygenation process showed a decrease of oxygen consumption after one minute postocclusion. The findings of a mean Vn for human skeletal muscle of 4.03 pmo/min/100 g tissue (200 mm Hg occlusion) and 3.60 pmol/min/100 g tissue (150 mm Hg) are comparable to the ones reported by De Blasi (3.07-7.58 pmo/min/100 g tissue) [10] and Cheatle (5.1 pmo/min/100 g tissue) [7], Those values characterize the first 40 sec of occlusion when myoglobin O2 content remains stable due to its P50 (Chapter 2 ). 81 8. CONCLUSIONS Energy metabolism of active tissues requires permanent availability of oxy gen. Near Infrared Spectroscopy has been identified as an useful tool for non- invasive investigations of regional tissue oxygenation. This study underlines the usefulness of NIRS for noninvasive monitoring skeletal muscle vascular properties, function and oxygen consumption. The original contributions of this work is in the noninvasive characterization of skeletal muscle hemody namics features. The study has demonstrated that: - Venous occlusion is characterized by accumulation of both arterial and venous blood in the muscle microvasculature. - Arterial occlusion rapidly depletes the oxygen stores of human skele tal muscle. - Muscle hypoxia is induced only when the occlusion level is signifi cantly above arterial pressure. - The decrease of blood volume oscillations as well as the increase of phase delay relative to venous pressure variation are due to recruit ment of additional segments of vasculature. - Skeletal muscle oxygen consumption can be estimated noninvasively. 82 The simplicity of steady state near-infrared spectroscopy is attractive and leads to accurate results when the distribution of pathlength is constant. However, uncertainty about the real pathlength and its variation limits the accuracy of steady state Near Infrared Spectroscopy. Therefore a combination of steady state and time-resolved Near Infrared Spectroscopy gives a better estimation of hemoglobin concentration changes. Future research will deter mine the optimal algorithm for computation of oxy and deoxy— hemoglobin concentrations in various clinical situations. 83 R EFERENCES 1. N. B. Hampson, and C. A. Piantadosi, “Near infrared monitoring of human skeletal muscle oxygenation during forearm ischemia”, J. Appl. Physiol, 64:2449-2457, 1988. 2. K. Sahlin, “Non-Invasive Measurements of 0 2 Availability in Human Skeletal Muscle with Near-Infrared Spectroscopy”, Int. J. Sports Med., 13:S157-S160, 1992. 3. M. Ferrari et al.,’ ’ Time— resolved spectroscopy of human forearm”, J. Photo- chem. Photobiol., B 16:141— 153, 1992. 4. B. Chance, “Time resolved spectroscopic (TRS) and continous wave spectro scopic (CWS) studies of Photon migration in human arms and limbs”, Oxygen Transport to Tissue XII, Plenum Press, New York, pp:21-31, 1991. 5. R. A. De Blasi et al., “Muscle oxygenation by fast near infrared spectros copy (NIRS) in ischemic forearm”, Oxygen Transport to Tissue XIII, Plenum Press, New York, pp: 163— 172, 1992. 6 . B. Chance et al., “Time— resolved spectroscopy of Hemoglobin and Myoglo bin in Resting and Ischemic Muscle, Anal. Biochem., 174:698-707, 1988. 7. T. R. Cheatle et al., “Near-infrared spectroscopy in peripheral vascular dis ease”, £r. J. Surg., 78:405-408, 1991. 8 . E. Sobolewski et al., “Near infrared reflectance spectroscopy of mitochon drial myopathy”, Neurology, 40(Sl):645--648. 9. J. Wilson et al., “Noninvasive Detection of Skeletal Muscle Underperfusion With Near-Infrared Spectroscopy in Patients With Heart Failure”, Circula tion, 80:1668-1674, 1989. 10. R. A. De Blasi et al., “Noninvasive measurements of forearm oxygen con sumption during exercise by near infrared spectroscopy”, Oxygen Transport to Tissue XV, Plenum Press, New York, pp:685— 692, 1994. 84 11. B. Chance et.al., “Recovery from exercice— induced desaturation in the quadriceps muscle of elite competitive rowers”, Am. J. Physiol., 262:C766— C775, 1992. 12. B. Chance, “Curent state of methodology on hemoglobin oximetry in tis sues”, Oxygen Transport to Tissue XV, Plenum Press, New York, pp:23~32, 1994. 13. C. A. Piantadosi, Near Infrared Spectroscopy: "Principles and Application to Noninvasive Assessment of Tissue Oxygenation", J. Crit. Care, 4:308-318, 1989. 14. F. F. Jobsis, “Non— invasive near infrared spectroscopy of cerebral and myocardial oxygen sufficiency and circulatory parameters”, Science, 198:1264-1267, 1977. 15. A. Ishimaru,”Wave Propagation and Scattering in Random Media”, Aca demic Press, Inc., London, 1978. 16. B. C. Wilson et al., “Tim e-Dependent Optical Spectroscopy and Imaging for Biomedical Applications”, Proceedings of the IEEE, 80:918— 930, 1992. 17. E. R. Weibel, “The pathway for oxygen Structure and Function in the Mammalian Respiratory System”, Harvard University Press, 1994. 18. P. A. Rea et al, “N on-invasive optical methods for the study of cerebral metabolism in the human newborn: a technique for the future”, J. Med. Eng. T>ch., 9:160-166, 1985. 19. F. F. Jobsis et.al., “Near— infrared monitoring of cerebral oxygen suffi ciency”, Neurol. Res., 10:7-17, 1988. 20. J. T. Walsh Jr. and J. P. Cummings,” Effect of the Dynamic Optical Prop erties of Water on Midinfrared Laser Ablation”, Lasers Surg. Med., 15:295- 305, 1994. 21. S. Wray et.al., "Characterization of the near infrared absorption spectra of cytochrome aa3 and hemoglobin for the non-invasive monitoring of cerebral oxygenation", Biochim. Bioph. Acta, 933:184— 192, 1988. 22. B. A. Bolton and J. R. Scherer, “Raman Spectra and Water Absorption of Bovine Serum Albumin”, J. Phys. Chem., 93:7635— 7640, 1989 23. B. C. Wilson and S. L. Jacques, “Optical Reflectance and Transmittance of Tissues: Principles and Applications”, IEEE J. Quantum Electronics, 26:2186-2199, 1990. 85 24. E. M. Sevick et al., “Quantitation of Time and Frequency Resolved Optical Spectra for the Determination of Tissue Oxygenation”, Anal. Biochem., 195:330-351, 1991. 25. B. C. Wilson, T. J. Farrell, and M. S. Patterson, “An optical fiber-based diffuse reflectance spectrophotometer for noninvasive investigation of photo dynamic sensitizers in vivo”,Proc. SPIE, 6:219— 232, 1990. 26. R. Berg,0. Jarlman, and S. Svanberg, “Medical transillumination imaging using short-pulse diode lasers”, Appl. Opt., 32:574-579, 1993. 27. D. A. Benaron and D. K. Stevenson, “Optical Time— of— Flight and Absor bance Imaging of Biological Media”, Science, 259:1463— 1466, 1993. 28. J. Singer et al., “Image reconstruction of the interior of bodies that diffuse radiation”, Science, 248:990-993, 1990 29. K. Groebe and G. Thews, “Basic mechanisms of diffusive and diffusion- related oxygen transport in biological systems: A review”, Oxygen Transport to Tissue XIV, Plenum Press, New York, pp:21-33, 1992. 30. A. S. Popel, “Theory of oxygen transport to tissue”, Crit. Rev. Biomed. Eng., 17:257-321, 1989. 31. K. Groebe, “O2 transport in skeletal muscle: Developement of concepts and current state”, Oxygen Transport to Tissue XV, Plenum Press, New York, pp: 15-22, 1994. 32. R. Braakman, P. Sipkema, and N. Westerhof, "A Dynamic Lumped Model for Skeletal Muscle Circulation", Ann. Biomed. Eng., 17:593-616, 1989. 86

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Processing And Visualization In Functional Magnetic Resonance Imaging (Fmri) Of The Human Brain

PDF

Auditory brainstem responses (ABR): quality estimation of auditory brainstem responsses by means of various techniques

PDF

Design of a portable infrared spectrometer: application to the noninvasive measurement of glucose

PDF

Estimation of upper airway dynamics using neck inductive plethysmography

PDF

Auditory brainstem responses (ABR): variable effects of click polarity on auditory brainstem response, analyses of narrow-band ABR's, explanations

PDF

Pulse oximetry failure rates

PDF

Decomposition Of Neuronal Function Using Nonlinear Systems Analysis

PDF

Flourine-19 NMR probe design for noninvasive tumoral pharmacokinetics

PDF

Computation and validation of circulating blood volume with the indocyanine green dilution technique

PDF

Comparison of evacuation and compression for cough assist

PDF

An ultraviolet raman, fluorescence and time-resolved spectroscopic investigation of glucose and other carbohydrates

PDF

A multimodal screen reader for the visually impaired

PDF

Respiratory system impedance at the resting breathing frequency range

PDF

Three-dimensional functional mapping of the human visual cortex using magnetic resonance imaging

PDF

Propofol Effects On Eeg And Levels Of Sedation

PDF

Cross-correlation methods for quantification of nonlinear input-output transformations of enural systems using a Poisson random test input

PDF

Functional water MR spectroscopy of stimulated visual cortex using single voxel

PDF

Fault tolerant characteristics of artificial neural network electronic hardware

PDF

A user interface for the ADAPT II pharmacokinetic/pharmacodynamic systems analysis software under Windows 2000

PDF

A fluorescence microscopy study of quantum dots as fluorescent probes for brain tumor diagnosis

Asset Metadata

Creator Marcu, Laura (author)

Core Title Human Skeletal Muscle Oxygenation And Perfusion: Non-Invasive Measurement By Near-Infrared Spectroscopy

Degree Master of Science

Degree Program Biomedical Engineering

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag engineering, biomedical,health sciences, radiology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Maarek, Jean-Michel (committee chair), Howell, Sandra (committee member), Marmarelis, Vasilis A. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-8980

Unique identifier UC11357703

Identifier 1378424.pdf (filename),usctheses-c18-8980 (legacy record id)

Legacy Identifier 1378424-0.pdf

Dmrecord 8980

Document Type Thesis

Rights Marcu, Laura

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA