Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of the diaphragm in task failure during inspiratory resistive loading in the rabbit Osborne, Salma 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1994-953756.pdf [ 2.59MB ]
Metadata
JSON: 831-1.0088028.json
JSON-LD: 831-1.0088028-ld.json
RDF/XML (Pretty): 831-1.0088028-rdf.xml
RDF/JSON: 831-1.0088028-rdf.json
Turtle: 831-1.0088028-turtle.txt
N-Triples: 831-1.0088028-rdf-ntriples.txt
Original Record: 831-1.0088028-source.json
Full Text
831-1.0088028-fulltext.txt
Citation
831-1.0088028.ris

Full Text

THE ROLE OF THE DIAPHRAGM IN TASK FAILURE DURING INSPIRATORY RESISTIVE LOADING IN THE RABBIT.  by SALLY SALMA OSBORNE B.Sc., University of British Columbia (1983) M.Sc., University of British Columbia (1988)  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE STUDIES (Department of Experimental Medicine)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA  May 1994 (c) Sally Salma Osborne  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written  permission.  (Signature)  Department of  £peri eiIal  The University of British Columbia Vancouver, Canada Date  DE.6 (2)88)  c3 j z 4  I-  -.  9V  iied c Ii  L  11  ABSTRACT  In experimental animal models, fatigue of the diaphragm has been implicated as the predominant determinant of hypercapnic ventilatory failure and ultimately as the cause of task failure of inspiratory muscles during inspiratory resistive breathing. The purpose of this study was to examine the effects of increased inspiratory resistive loads on diaphragm function in the anesthetized rabbit model to test three hypotheses: first, that task failure results from a decrease in neural activation; second, that task failure results from a decrease in neuromuscular transmission to the diaphragm; and third, that the development of hypoventilation and hypercapnia precede task failure. We assessed central motor output and neuromuscular transmission to the diaphragm by continuous monitoring of phrenic nerve activity and electromyogram activity of the costal diaphragm during both sustainable and exhaustive inspiratory resistive loads. We found a linear relationship between the severity of the target inspiratory airway pressure achieved with resistive loading and the indices of motor output to the diaphragm and activity of this muscle. Central motor output to the diaphragm remained  elevated throughout resistive  loading even at task failure.  Neuromuscular transmission, as assessed by evoked compound potentials of the diaphragm, remained intact throughout inspiratory resistive loading including at task failure. The activity of the diaphragm remained elevated and coupled to central motor output throughout resistive loading, including at task failure.  111  Hence, task failure did not result from either a decrease in neural activation nor from a decrease in neuromuscular transmission to the diaphragm. We found that despite substantial increases in inspiratory effort, rabbits hypoventilated during both sustainable and exhaustive loads. Therefore, hypercapnia typically accompanied inspiratory resistive loading.  Furthermore, we found that the  elevated levels of arterial Pc0 2 associated with prolonged loading alone, suppressed central drive to the diaphragm through a time-dependent reduction in breathing frequency. We observed task failure only during intense loading at target pressure close to the maximum strength of the rabbit diaphragm. The activity of the inspiratory muscles (parasternal intercostal and diaphragm) remained elevated and coupled despite severe arterial hypoxemia and hypercapnia during task failure. In contrast, a susbstantial decay in expiratory muscle activity and in abdominal pressure swings preceded task failure.  In  conclusion, neural activation and impulse propagation to the diaphragm were maintained during inspiratory resistive loading even at task failure. Task failure followed a loss in abdominal muscle assist to the diaphragm.  iv TABLE OF CONTENTS  Abstract  ii  Table of Contents  iv  List of Figures  vi  List of Tables  iix  List of Abbreviations  ix  Acknowledgements  xii  Dedication  xiii  Chapter I: General Introduction Skeletal Muscle Fatigue: Definitions Etiology of Fatigue Respiratory Muscle Fatigue: A Definition Diaphragm Function and Fatigue in Response to Inspiratory Flow Resistive Loads: Human studies Diaphragm Function and Fatigue in Response to Inspiratory Flow Resistive Loads: Animal Studies Objectives Specific Aims Significance References Chapter II: Diaphragm Activation and Phrenic Nerve Impulse Propagation During Prolonged Inspiratory Resistive Loading Introduction Methods Results The effect of prolonged inspiratory resistive loads on activation of the diaphragm The effect of severe inspiratory resistive loads on activation and neuromuscular transmission to the  1 3 4 7 8 12 16 17 18 26  33 33 35 43 43  V  diaphragm  .  Discussion Evaluation of Methods Neuromuscular Transmission to the Diaphragm Constant Diaphragm Activation during Prolonged Loading Prolonged Inspiratory Resistive Loading and Indices of Diaphragm Fatigue Task Failure of the Respiratory Muscles during Severe Inspiratory Resistive Loading References Appendix I  Chapter III: Ventilation During Prolonged Hypercapnia in the Anesthetized Rabbit  52 64 64 66 72 73 77 80 85  86 86 89 91  Introduction Methods Results Steady state response to hypercapnia: 30 minute exposure to hypercarbia Ventilatory response to prolonged hypercapnia: 3 hour exposure to hypercarbia Discussion References  95 108 114  Chapter IV: Respiratory Muscle Activity and Task Failure During Severe Inspiratory Resistive Loading  117  Introduction Methods Results Discussion References Chapter V: Summary and Conclusions References  91  117 118 123 133 140 146 154  vi LIST OF FIGURES  Figure 1:  Mechanical characteristics of skeletal muscle  Figure 2:  Schematic diagrams of the human and the rabbit diaphragm.  Figure 3:  A schematic diagram of the chest wall illustrating the primary actions of the diaphragm  19 .  21  23  Figure 4:  Pathway involved in skeletal muscle tension development  Figure 5:  A schematic diagram of the experimental setup  39  Figure 6:  Peak electrical activity of the phrenic nerve, the costa! diaphragm and transdiaphragmatic pressure swings during prolonged inspiratory resistive loads  45  Peak inspiratory airway opening pressure and transdiaphragmatic pressure swings during prolonged inspiratory resistive loads  47  Changes in arterial pH, pCO 2 and ventilation during prolonged inspiratory resistive loads  49  Peak electrical activity of the phrenic nerve and the costa! diaphragm during severe loads  53  Figure 10: The relationship between peak inspiratory pressure and indices of activity and drive to the diaphragm  56  Figure 11: Changes in arterial blood gases and negative inspiratory pressures recorded at the airway opening during severe inspiratory resistive loads  58  Figure 12: The change in minute ventilation during severe inspiratory resistive loading  61  Figure 13: Representative evoked diaphragm compound action potentials during severe inspiratory resistive loading  70  Figure 7:  Figure 8:  Figure 9:  .  .  .  25  vii Figure 14: The relationship between peak inspiratory pressure and transdiaphragmatic pressure swings during brief inspiratory resistive loading  75  Figure 15: Minute ventilation and arterial PC0 2 levels during prolonged exposure to FICO2= 0.10 in hyperoxic rabbits  97  Figure 16: Comparison of arterial Pco 2 levels and minute activity of the diaphragm in rabbits exposed to prolonged hypercarbia and rabbits exposed to prolonged inspiratory resistive loading  102  Figure 17: Comparison of activity of the diaphragm and breathing frequency in rabbits exposed to prolonged hypercarbia and rabbits exposed to prolonged inspiratory resistive loading  .  .  .  104  Figure 18: Comparison of arterial Pc0 2 levels and the minute ventilation in rabbits exposed to prolonged hypercarbia and rabbits exposed to prolonged inspiratory resistive loading  106  Figure 19: Electromyographic activity of the costal diaphragm, parasternal intercostal and transversus abdominis during severe inspiratory resistive loading  126  Figure 20: Sample tracing of arterial blood pressure, airway opening pressure, abdominal pressure, airflow and phasic-moving average of costal diaphragm, parasternal intercostal and transversus abdominis during baseline and severe inspiratory resistive loading  129  Figure 21: Sample tracing of the activity of inspiratory muscles during the final minutes leading to respiratory arrest  131  vii’ LIST OF TABLES  Table I:  Table II:  Table III:  Table IV:  Table V:  Table VI:  Table VII:  Respiratory variables at baseline and during prolonged inspiratory resistive loading in the anesthetised rabbit  51  Respiratory variables at baseline and during severe inspiratory resistive loading in the anesthetised rabbit  63  Arterial blood gases and ventilatory variables during the initial 30 minutes of exposure to 10% inspired C02 in hyperoxic anesthetized rabbits  93  Mean values for respiratory times, mean inspiratory flow rate (VT/TI) and duty cycle (TI/TrOT) in anesthetized rabbits exposed to 10% inspired C02 for the initial 30 minutes  94  Arterial blood gases and changes in ventilatoiy variables produced by 3 hours of exposure to 10% inspired C02 in anesthetized rabbits  99  Mean values for respiratory times, mean inspiratory flow rate (VT/TI) and duty cycle (TI/TrOT) in anesthetized rabbits exposed to 10% inspired C02 for 3 hours  100  Respiratory variables, arterial blood gases and pH at baseline and during severe inspiratory resistive loading in the anesthetized rabbit  125  ix LIST OF ABBREVIATIONS AND SYMBOLS  Aapp  Area of apposition  Bf  Breathing frequency  CSF  Cerebrospinal fluid  CNS  Central nervous system  cm H20  Centimeters of water  E  Expiratory  EELV  End expiratory lung volume  EMGdi  The moving average of the diaphragm electromyogram  EMGps  The moving average of the parasternal electromyogram  EMGta  The moving average of the transversus abdominis electromyogram  ENGdi  The moving average of the diaphragm electroneurogram  Fc  Centroid frequency of the EMG power spectrum  FRC  Functional residual capacity  H  Hydrogen cation  3 HC0  Bicarbonate anion  I  Inspiratory  LM.  Intramuscular  I.V.  Intravenous  2 Mg  Magnesium cation  mm  Minutes  x ml  Milliliters  mm Hg  Millimeters of mercury  mV  Millivolts  M wave  Evoked compound muscle action potential Phosphate ion  Pab  Abdominal pressure  PACO2  Alveolar partial pressure of carbon dioxide  PaCO2  Arterial partial pressure of carbon dioxide  Pao  Airway pressure  Pdi  Transdiaphragmatic pressure  Pa02  Arterial partial pressure of oxygen  Pes  Esophageal pressure  pH  Negative logarithm of hydrogen ion concentration  Ppl  pleural pressure  ptp  Peak to peak  QB  Quiet breathing  sec  seconds  S.E.M.  Standard error of the mean  TE  Expiratory duration  TI  Inspiratory duration  TI/TrOT  Duty cycle  ‘PrOT  Duration of a breath or total respiratory time  xi Airflow VE  Minute ventilation  VT  Tidal volume  VT/TI  Mean inspiratory flow rate  xl’  ACKNOWLEDGEMENTS  I am indebted to my supervisor Dr. Jeremy Road for his support and patience throughout the course of my studies. I sincerely appreciate Dr. John Ledsome’s valuable advice as my interim advisor during Dr. Road’s absence and grateful for his wisdom and integrity. I am equally grateful to Dr. Bill Milsom whose enthusiasm for the study of control of breathing encouraged me to continue further with my graduate studies.  He has invested much time in my post  graduate work and his unconditional support is most appreciated. Many thanks to Dr. Darlene Reid for her advice, useful resources and unending encouragement. Thanks to Dr. Angelo Belcastro for introducing me to the techniques in his laboratory and biochemical correlates of muscle fatigue. Special thanks and appreciation for the expert technical assistance of Michael Boyd, Miguel Pachenko and the enthusiastic assistance of the animal care unit staff at the University Hospital, U.B.C.; their high professional standards made the protocols feasible.  xlii  DEDICATION  To my children, Simone Kimberly and Corbin Elliot Osborne whose sunny personalities were a source of strength and inspiration.  To my husband, David Nelson Osborne who surpassed the standards for equal partnership.  In memory of Rene Theophile Hyacinth Laennec and his prescient wisdom: “Nothing hurts the progress of science more than to divert terms from their customary meaning without sufficient reason and to create bad new ones”.  1 I: General Introduction  Life depends on the ability of respiratory muscles to overcome resistive and elastic forces and to produce phasic contractions continuously. Normal breathing is accomplished easily due in good part to the enormous reserve capacity of the respiratory muscles. Functionally, respiratory muscles are skeletal muscles and as such share common mechanical characteristics with other skeletal muscles (Figure 1) and can adapt to long-term functional alterations (Farkas and Roussos, 1983; Farkas, 1991).  In the mammalian species, the diaphragm is considered to be the principal muscle for inspiration during quiet breathing. separates the thoracic and abdominal cavities.  This dome-shaped structure The muscle fibres of the  diaphragm originate from three different regions: 1) the sternal region arising from the posterior aspect of the xiphoid process; 2) the costal region arising from the lower six ribs and the costal cartilages and 3) the crural region arising from the upper three lumbar vertebrae. These fibres ascend and radiate inwards inserting into the central tendon, a thin but strong aponeurosis (Leak, 1979; Figure 2).  During quiet inspiration contraction of the diaphragm decreases pleural pressure, expands lung volume and increases abdominal pressure pushing the ribcage  2 upward and outwards and leading to protrusion of the anterior abdominal wall (Figure 3).  Other primary muscles of inspiration include the scalene and  parasternal intercostal muscles. Quiet expiration is thought to be passive and mainly due to the elastic recoil of the lungs.  During augmented breathing activity, more of the respiratory muscle mass is recruited to achieve greater ventilation. The accessory muscles of inspiration such as the sternocleidomastoid, external intercostals, serratus posterior superior and pectoralis major may be activated. Furthermore, expiration becomes active with phasic contractions of the primary expiratory muscles in the ventral abdominal wall as well as the ribcage internal intercostals.  In theory, the respiratory muscles could fail to generate adequate force contracting against respiratory loads that require too great an effort for too long a period of time. In 1977, Macklem and Roussos proposed that respiratory muscle fatigue might underlie respiratory failure observed with several clinical conditions such as atrophy of respiratory muscles with prolonged mechanical ventilation, neuromuscular weakness, chronic obstructive pulmonary disease or with increased inspiratory loads.  To date, it has been difficult to confirm  respiratory muscle fatigue in patients progressing slowly to chronic respiratory failure (Begin and Grassino, 1991; Grassino and Clanton, 1991; Rochester, 1991). It has long been accepted that acute inspiratory resistive loading can  3 induce fatigue of the diaphragm both in man (Roussos and Macklem, 1977; Aubier Ct aL, 1981; Moxham et aL 1981, 1982; Cohen et al. 1982; Grassino and Macklem, 1984; Bellemare and Bigland-Ritchie, 1987 and Yan et aL 1992) and in experimental animal models (Aldrich, 1985, 1987, 1988, 1991; Alexandrovna and Isaev, 1990; Bazzy and Haddad, 1984; Bazzy and Donnelly, 1993; Mayock et aL 1987, 1991; Oliven et aL 1988) although confirmation for this hypothesis has been elusive.  Skeletal Muscle Fatigue: Definitions  In 1979, Edwards defined skeletal muscle fatigue in mechanistic terms as a decrease in force generation due to either a decrease in neural impulse propagation known as “neural fatigue” or due to the failure of the contractile apparatus known as “muscle fatigue” (Edwards, 1979).  Since then, various  additional factors have been considered and the definition of muscle fatigue has been simplified by Hultman and Sjoholm (1986) as “the failure to maintain an expected force or power output”.  Figure 4 shows the pathway involved in muscular contraction. In theory, fatigue can be brought about by impaired function of any one or combination of alterations in steps involved in the pathway from the CNS to the contractile apparatus. Therefore, depending on its origin, skeletal muscle fatigue can be  4 broadly categorized in three groups: 1)  Central fatigue, 2) failure of  neuromuscular transmission and 3) peripheral fatigue.  Etiology of Fatigue  First, the cause of fatigue may be “central” due to a decrease in the frequency or intensity of neural output to the skeletal muscle from the central nervous system. This may be conscious due to a lack of motivation or an inability to tolerate the discomfort associated with the fatiguing stimulus.  It may be  unconscious due to protective feedback mechanisms operating through inhibitory afferent input to the central nervous system (for a review see Enoka and Stuart, 1992). Skeletal muscle afferent input during increased work may come in form of the proprioceptive information from the Golgi tendon organs or the muscle spindles.  Alternatively,  Group Ill and IV afferent fibres may decrease  motoneuron output in response to stretch or the metabolic state of the muscle. Hence, afferent input from the muscle may serve as an important feedback link between the working muscle and excitatory motor output from the CNS and/or spinal reflex activity during exhaustive work.  Secondly, fatigue of skeletal muscle might also result from a decrement in neuromuscular transmission (Naess and Storm-Mathesen, 1955).  Whereas  neuromuscular transmission failure is undoubtedly the major cause of muscle  5 fatigue in some pathologic conditions such as botulism, Lambert-Eaton myasthenic syndrome and myasthenia gravis, its role in the response of skeletal muscle to voluntary contractions against exhaustive loads remains controversial. Recently this issue has been reexamined in animal models of respiratory muscle fatigue. Two studies have considered the importance of declining neuromuscular transmission to the diaphragm under severe inspiratory resistive loads (Aldrich, 1987, 1991; Bazzy and Donnelly, 1993).  Finally, fatigue of skeletal muscle may be peripheral (myogenic) in origin. The loci for this fatigue are distal to the neuromuscular junction including the sarcolemma, the transverse-tubule, the sarcoplasmic reticulum and the regulatory and contractile proteins. There is little evidence that propagation of action potentials across the sarcolemma or t-tubules is blocked except in the artificial situation  of  continuous  high-frequency  electrical  stimulation  where  transmembrane flux of water and local electrolytes, specifically potassium across the t-tubule, result in decreased contractile force (Westerbiad et aL 1991). The major focus of current research on skeletal muscle fatigue looks beyond excitability of the cell membrane to the various processes involved in coupling of sarcolemmal action potential and cross bridge interaction resulting in muscle contraction (excitation-contraction coupling).  Metabolic consequences of sustained muscular work have long been considered  6 as correlates of peripheral fatigue by virtue of their effect(s) on excitation contraction coupling and/or cross bridge interaction. In the past 25 years, two hypotheses [the exhaustion (or depletion) hypothesis and the accumulation (build up of metabolites) hypothesis] have been proposed as metabolic causes of fatigue. Depleted metabolites include ATP, PCr and glycogen. Accumulated metabolites include H, P, lactate, free intracellular Ca 2 and Mg 2 (for reviews see Green, 1987; Kirdendall, 1990; Vollestad and Sejersted, 1988; Enoka and Stuart, 1992 and Westerblad et aL, 1991). The extent to which any of these metabolites contribute to fatigue of the muscle will depend on the intensity of work, the amount of time elapsed since the onset of activity, and the specific metabolic profile of the muscle or muscle fibre. Owing to the specificity of the metabolic profile of muscle fibres, the current approach in determining the role of myogenic factors in fatigue is a reductionist one. Typically, studies consider the levels, compartmental content, rates of production/utilization, and mechanism of action of these metabolites, in vitro, in single intact or skinned fibre preparations. Typically, these studies employ artificial electrical stimulation regimens to induce fatigue and to assess responsivity (Westerblad et aL, 1991).  A coherent interpretation of the literature on the etiology of skeletal muscle fatigue is complicated by the diversity of definitions of muscle fatigue, the variety of indices used to define muscle fatigue and the numerous paradigms used to study it. There is general acceptance that the specific mechanism causing fatigue  7 of a skeletal muscle will depend on the nature of the fatiguing stimulus (Enoka and Stuart, 1992). This point is of considerable importance when examining the various breathing maneuvers employed to trigger respiratory muscle fatigue.  Respirato,y Muscle Fatigue: A Definition  The definition of respiratory muscle fatigue was recently considered at a National Heart, Lung and Blood Institute Workshop (1990).  The current  definition of respiratory muscle fatigue is based on a consensus arrived at this workshop which involved primarily chest physicians.  This new definition  describes respiratory muscle fatigue in functional terms as a condition in which there is a loss in the capacity for developing force and or velocity of a muscle in response to a load and which is reversible by rest. Furthermore, the traditional definition of skeletal muscle fatigue as a failure to generate an expected force was redefined as “task failure”.  Introducing the concept of reversibility in this new definition of fatigue permits one to contrast fatigue from muscle weakness. Further, this new definition has broadened the concept of respiratory muscle fatigue significantly. An implication of this definition is that fatigue may be observed with respect to a task even though the force required for its completion can still be generated. For example, the capacity of the diaphragm in vivo to develop force is estimated by the  8 pressure generated across the diaphragm (Pdi). Animal models of diaphragm fatigue have demonstrated a reversible loss in the capacity to develop Pdi in response to tetanic electrical stimulation following inspiratory resistive loaded breathing. However, the capability of the diaphragm to generate pressure during spontaneous breaths against these loads known as tidal Pdi swings, is maintained for periods up to five hours after one observes a decrement in response to tetanic electrical stimulation (Mayock, et aL, 1991; see Chapter II as well). According to the current definition of respiratory muscle fatigue proposed at the National Heart Lung and Blood Institute workshop (1990), these preparations (Mayock et at., 1987, 1991; Aldrich 1990), qualify as models of diaphragm fatigue.  Diaphragm Function and Fatigue in Response to Inspiratoiy Flow Resistive Loads: Human studies  The diaphragm is the primary muscle of inspiration. Accordingly, there has been much focus on its response to artificial external loading.  Specifically, the  response to flow resistive loads which are defined by the resistance of porous discs placed externally at the inspiratory inlet of a breathing circuit have been studied.  The human ability to tolerate high inspiratory flow resistive loads is  well documented (Freedman and Campell, 1970). It has been shown that the increase in oxygen consumption associated with inspiratory resistive loading is  9 proportional to the integral of mouth pressure. In turn, oxygen consumption can be modified through pattern of breathing (Jones et aL, 1985). Although the immediate response of awake human subjects to inspiratory resistive loading is extremely variable (Axen and Haas, 1979), the steady state response to sustainable inspiratory loads is typically characterized by an increase in inspiratory time, a decrease in breathing frequency and a reduced mean inspiratory flow rate (Axen et aL, 1983; Tm Hof et aL 1986; Jones et aL, 1985). Furthermore, relatively large inspiratory loads can result in an increase in expiratory time, elevated arterial Pco 2 levels (Tm Hof et aL, 1986) and reduced arterial P0 2 levels (Eastwood et aL, 1994). The specific pattern of breathing employed by conscious humans may delay fatigue. Load perception and other factors may influence this phenomenon (Clague et aL, 1992 and Eastwood et aL, 1994).  Therefore, it is necessary that the complex nature of the potential  interplay between load perception and behavioral responses be emphasized when interpreting ventilatory responses to loaded breathing in conscious humans.  Inspiratory pressure swings across the diaphragm (Pdi) decrease against relatively severe target inspiratory flow resistive loads (60-80% Pdimax) in conscious humans. This loss in target pressure generation is thought to reflect a decrement in diaphragm force generation and hence diaphragm fatigue in the conscious human. Although breathing against such high inspiratory resistance is hard work and not particularly pleasant (Mead, 1979), it has been argued that fatigue under  10 these circumstances is not purely motivational. Other indices of fatigue are also present. These indices include a drop in the frequency-Pdi curve (commonly referred  to  as  the  force-frequency  curve),  alterations  in  diaphragm  electromyogram power spectrum (high-low ratio or centroid frequency), and twitch occlusion (Aubier et aL, 1981, 1985; Moxham et aL, 1981, 1982; Bellemare and Bigland-Ritchie, 1987).  Any conclusions arising from the above findings must be tentative as the underlying mechanisms involved in changes in these indices are unclear and their presence precedes rather than being causative of task (pressure) failure. A decrease in the frequency-pressure curve of the diaphragm after inspiratory resistive loading suggests alteration of those processes involved in force generation which are distal to the CNS. For example, a decrease in pressure output of the diaphragm in response to high frequency electrical stimulation of the phrenic nerve is thought to reflect either a decrease in phrenic nerve impulse propagation, neuromuscular junction failure, or loss of action potential propagation across the sarcolemma or t-tubule (high frequency fatigue see above: skeletal muscle etiology). It is regarded as uncertain whether severe inspiratory resistive loading leads to high frequency fatigue in humans (Moxham, et aL, 1981 cf Aubier, et aL, 1981).  There is some agreement that a decrease in pressure output of the diaphragm  11 occurs in response to low frequency stimulation (Moxham, et aL, 1981, 1982; Aubier, 1981). Although low frequency fatigue is thought to reflect fatigue of myogenic origin, the cellular mechanisms remain to be elucidated.  Similarly, the mechanisms underlying shifts in the diaphragm EMG power spectrum are unknown. A shift in the diaphragm EMG power spectrum is evaluated by a change in it’s centroid frequency (Fc) or a change in the ratio of high frequency power to low frequency power (H/L ratio). It has been shown that Fc is more sensitive to shifts in the diaphragmatic EMG power spectrum. Additionally, changes in the Fc have been detected during levels of voluntary hyperpnea that can be readily sustained (Sieck et aL, 1985). Consequently, the specificity of using power spectral changes of diaphragm EMG in predicting fatigue has been questioned (Sieck et aL, 1985).  The degree to which the diaphragm is activated by the CNS during inspiratory resistive loading has been assessed by the twitch occlusion technique.  This  technique suggests that half of the reduction in diaphragm pressure generation is due to central fatigue in awake human subjects (Bellemare and Bigland Ritchie, 1987). Recent data show that this technique could be insensitive to the contribution of central mechanisms of fatigue to pressure failure due to the phenomenon of twitch potentiation (Mador et aL, 1994; Wragg et aL, 1993).  12 Diaphragm Function and Fatigue in Response to Inspiratoiy Flow Resistive Loads: Animal Studies  Several studies in animals breathing against inspiratory resistive loads attempt to define the role of diaphragm fatigue in respiratory failure. Bazzy and Haddad (1984) were the first to show a decrease in diaphragm activity (EMG) and pressure output (Pdi) in response to intense inspiratory resistive loads in a study on three chronically instrumented awake sheep. These loads were sustainable for prolonged periods (up to 3 hours) but eventually resulted in a decrease in diaphragm activity (EMG) followed by a drop in diaphragm pressure output and respiratory acidosis (Bazzy and Haddad, 1984). In a subsequent study using the same preparation, Bazzy and Donnelly (1992) show that phrenic activity (ENGdi) increased at a time when diaphragm muscle activity (EMGdi) and pressure (Pdi) were maintained. Based on these observations and the decrease seen in the evoked compound potentials of the diaphragm (M-waves) in two sheep, the authors concluded that neuromuscular transmission fatigue contributed to failure of the diaphragm to generate inspiratory pressure (Bazzy and Donnelly, 1992).  In anesthetized rabbits exposed to approximately one hour of inspiratory resistive loading, Aldrich (1987) arrived at a similar conclusion based on two observations. Firstly, there was a substantial decrease in diaphragm M-wave amplitude and area with inspiratory resistive loading. Secondly, inspiratory  13 resistive loading reduced the frequency-pressure response elicited by stimulation of the phrenic nerve but did not affect the frequency-pressure response elicited by direct stimulation of the diaphragm.  In contrast, prolonged (6 hours)  inspiratory resistive loads of similar magnitude in the anesthetized piglet did not affect the M-wave despite a reduction in the frequency-pressure response of the diaphragm to phrenic nerve stimulation (Mayock et aL, 1987, 1991).  An  examination of neuromuscular response to inspiratory muscle challenge under a variety of loads has not been done.  In other species, in both awake and anesthetized preparations, there is evidence that central fatigue can play a significant role in respiratory failure associated with inspiratory resistive loaded breathing. In the unanesthetized infant monkey, intense inspiratory resistive loading resulted in respiratory acidosis and maintained levels of peak tidal airway pressure and diaphragm EMG (Watchko et aL, 1988). Although the activity of each diaphragm contraction (EMG) was not reduced, there was a decrease in diaphragm activity per minute due to reduced frequency of breathing. These findings are consistent with the notion that an optimization of inspiratory work occurs in response to loaded breathing and is achieved through a decrease in central rhythm.  Central modulation of respiratory activity during inspiratory resistive loading is not limited to the processes involved in rhythm generation.  For example,  14 Scardella and associates found no change in frequency of breathing in response to prolonged (2.5 hour) intense inspiratory resistive loading (Scardella et aL, 1986) in awake goats.  Instead they observed a decrease in tidal volume  associated with an increase in immunoreactive beta endorphin in the cerebrospinal fluid (Scardella, Santiago and Edelman, 1989). This increase in endogenous opioids in the CNS produced differential inhibition of respiratory muscle electrical activity (Scardella et aL, 1990). Furthermore, lactic acidosis of respiratory muscles is another feature of intense inspiratory resistive loading in this model (Petrozinno et aL, 1992). To account for these findings in the awake goat, Scardella and Petrozzino have developed a conceptual model involving both facilitatory and inhibitory pathways activated by lactic acid stimulation of Group III and IV afferents (Petrozzino et aL, 1993). These findings are also consistent with the notion that central fatigue may indeed play a protective or adaptive role in delaying the onset of myogenic fatigue in response to intense inspiratory resistive loads.  The possibility that peripheral (myogenic) fatigue of the diaphragm may eventually occur if the inspiratory resistive loads are sufficiently intense and/or prolonged has been examined in a handful of studies. In the anesthetized piglet, submaximal contractions of the diaphragm under moderate inspiratory resistive loads causes a decrease in the maximal force generating capacity of the diaphragm (Mayock, et aL 1991). However, there is no change in the levels of  15 ATP, phosphocreatine, lactate or glycogen in the diaphragm after prolonged loading (6 hours) in this preparation. Similarly, the anesthetized rabbit breathing against incremental increases in inspiratory threshold load fails to show any evidence of diaphragm glycogen depletion or lactate accumulation even at maximal loads that result in respiratory arrest (Ferguson et at, 1990). Other indices of myogenic fatigue have not been measured during inspiratory resistive loading in animal models.  It is clear that breathing against an inspiratory resistive load reduces those indices for which a reduction would imply that central and neuromuscular transmission fatigue of the diaphragm contribute to respiratory failure. The effects of inspiratory resistive loading appear to depend on the intensity of the load and its duration.  What remains to be determined from the studies  conducted to date is the extent to which central and neuromuscular transmission fatigue of the diaphragm contributes to the failure of the diaphragm to generate an expected inspiratory pressure (task failure).  Furthermore, in all animal  species where inspiratory resistive loading has been employed to trigger fatigue of the diaphragm, the target inspiratory load has been sufficiently intense to result in hypoventilation and hypercapnic ventilatory failure. The relationship between hypercapnic ventilatory failure, diaphragm fatigue and task failure during inspiratory resistive loading remains unclear.  16 Objectives  In the studies discussed below, the anesthetized rabbit model of inspiratory resistive loading was chosen to examine the effects of increased inspiratory resistive loads on diaphragm function. The rabbit is a suitable model to assess diaphragm function since the rabbit diaphragm has a similar proportion of fatiguable fibres as man (Green et aL, 1984) and therefore has the potential to demonstrate fatigue. The diaphragm of other species, for example the dog, is composed entirely of type I and ha fibres which are fatigue resistant (Green et aL, 1984).  Additionally, it has been shown that anesthetized rabbits will breathe spontaneously against loads that result in a decrease in the maximal forcegenerating capacity of the diaphragm (Aldrich, 1985, 1987). Since fatigue can result from failure of one or more elements of a closed-loop system, examining diaphragm function in a preparation that breathes spontaneously under anesthesia offers the opportunity to distinguish the extent to which each of these elements contributes to diaphragm failure independent of the conscious factors which might contribute to fatigue.  Accordingly, the major objective of these studies was to test the following hypotheses:  17 1. The failure of the respiratory system to maintain a target inspiratory pressure (task failure) results from a decrease in central motor output to the diaphragm. 2. Task failure results from a decrease in neuromuscular transmission to the diaphragm. 3. The development of hypoventilation and hypercapnia precede task failure.  Specific Aims  The specific aims of the studies discussed below were to:  1. Assess central motor output and neuromuscular transmission to the diaphragm during periods of prolonged sustainable and exhaustive inspiratory resistive loading.  2. Determine the changes in breathing pattern (volume, frequency, and timing) in response to inspiratory resistive loading of different intensities and durations, ranging from sustainable to exhaustive loads.  3. Determine whether the respiratory muscles could be loaded to produce task failure.  18 4. Determine the relationship between task failure and hypercapnic ventilatory failure during inspiratory resistive loading.  Significance  These experiments will provide insight into the in vivo function of the normal diaphragm under inspiratory resistive loading and the control of ventilation under these conditions. Since the results of these studies are derived in the anesthetized state, they will provide information on mechanisms involved in respiratory failure during inspiratory resistive loading independent of conscious factors.  19  Figure 1: Mechanical characteristics of skeletal muscle:  Panel A: The active length-tension relationship of skeletal muscle (adapted from Lieber, 1992). The tension generated in skeletal muscle is a direct function of the magnitude of overlap between actin (insert solid lines) and myosin (insert crossed line) filaments. Maximal tension is produced at optimal length. Lengths shorter or greater than the optimal result in less filament overlap and lower tension. Optimal length of the diaphragm is reported at a lung volume below end expiratory lung volume (EELV) at the end of a normal quiet breath (FRC).  Panel B:  An increase in EELV can place the diaphragm in a position of  mechanical disadvantage. In contrast, a decrease in EELV can be mechanically advantageous in increasing the potential for diaphragm excursion.  20  A 125  -  C 0 C  100  -  I— C) C 0 U, I  75  -  E E x  50  -  0  C U, C,  25  -  U, 0  0 1.0  I  I  I  I  1 .5  2.0  2.5  3.0  3.5  Sarcomere Length (urn)  B  Diaphragm position at: End Expiration End Inspiration <  1 EELV  >.Diaphragm Excursion  21  Figure 2:  Schematic diagrams of the human diaphragm in situ (top panel) and  the peritoneal surface of the diaphragm of the rabbit (bottom panel). Adapted from Farkas, 1991 and Leak, 1979.  a  23  Figure 3. A schematic diagram of the chest wall showing the rib cage and abdominal compartments. The area of apposition (Aapp) is the region of the diaphragm apposed against the lower ribcage. Contraction and shortening of the diaphragm leads to progressive “peeling away” of this muscle from the ribcage at the Aapp. This down ward “piston-like” action of the diaphragm increases abdominal pressure and decreases pleural pressure. The increase in abdominal pressure in turn acts as a fulcrum for movement of the rib cage by two mechanisms: 1) Insertional action: the force applied at the origin of diaphragm on the ribcage in the direction of muscle fibre insertion (central tendon) rotating the ribs upward (pump handle motion) and outward (bucket handle motion) 2) Appositional action: a distending force on the lower rib cage at the Aapp moving the ribs outward. The decrease in pleural pressure displaces the upper ribcage inward. [Adapted from De Troyer & Loring, 1986]  ZLI  Rib Cage Dome-----,.  A app.  Abdomen  25  Brain Central fatigue  Spinal Cord  I Neuromuscular Fatigue  Peripheral Nerve  Neuromuscular Junction  I Sarcolemma Transverse Tubules Peripheral  Calcium Release  (Myogenic)  Crossbridge Formation  Fatigue  Contraction Tension  Figure 4.  Pathway involved in skeletal muscle tension (force) development  (adapted from Kirdendall, 1990).  26 References  1. Aldrich, T.K. (1987). Transmission fatigue of the rabbit diaphragm. Respir PhysioL 69: 307-3 19. 2. Aldrich, T. K. (1991). Central and transmission fatigue. Seminars in Respiratoiy Medicine 12: 322-330.  3. Aldrich, T. K. (1988). Central fatigue of the rabbit diaphragm. Lung 166: 233-241. 4. Aleksandrovna, N.P. and G.G. Isaev (1990). Central and peripheral components of the fatigue of respiratory muscles in inspiratory resistive load in cats. Sechenov Physiological Journal of the US.S.R. 76: 658-666. 5. Aubier, M., D., Murciano, Y. Lecocguic, N. Virres and R. Pariente (1985). Bilateral phrenic stimulation; a simple technique to assess diaphragmatic fatigue in humans. J. AppL PhysioL 58: 58-64. 6.  Aubier, M., G. Farkas, A. de Troyer, R. Mozes and C. Roussos (1981). Detection of diaphragmatic fatigue in man by phrenic stimulation. J AppL Physiol. 50: 538-544.  7. Axen, K. and S.S. Haas (1979). Range of first-breath ventilatory responses to added mechanical loads in naive men. J AppL PhysioL 46: 743-75 1. 8. Axen, K., S.S. Hass, R. Hass, D. Guadino and A. Haas (1983). Ventilatory adjustments during sustained mechanical loading in conscious humans. J. AppL PhysioL 55: 1211-1218.  27 9. Bazzy A. R. and D. F. Donnelly (1993). Diaphragmatic failure during loaded breathing: role of neuromuscular transmission. J AppL PhysioL 74: 16791683. 10. Bazzy A.R. and 0.0. Haddad (1984).  Diaphragmatic fatigue in  unanesthetized adult sheep. J AppL PhysioL 57: 182-190. 11. Bellemare F. and B. Bigland-Ritchie (1987). Central components of diaphragmatic fatigue assessed by phrenic nerve stimulation. J AppL PhysioL 62: 1307-1316. 12. Begin, P. and A. Grassino (1991). Inspiratory muscle dysfunction and chronic hypercapnia in chronic obstructive pulmonary disease. Am. Rev. Respir Dis. 143: 950-9 12. 13. Clague, J.E., J. Carter, M.G. Pearson and P.M.A. Calverley (1992). Effort sensation, chemoresponsiveness, and breathing pattern during inspiratory resistive loading. J. AppL PhysioL 73: 440-445. 14. Cohen, C.A., 0. Zagelbaum, D. Gross, C. Roussos and P.T. Macklem (1982). Clinical manifestations of inspiratory muscle fatigue. Am. J Med. 73: 308-316. 15. Daubenspeck, J.A. (1994). Invited editorial on “Discrimination of transiently applied mechanical loads: breathing versus pulling. J AppL PhysioL 76: 3-4. 16. De Troyer, A. and S.H. Loring (1986). Actions of the respiratory muscles. In: Mead J, Macklem P.T. (eds.). Handbook of Physiology, Accessory muscles. Washington, D.C., American Physiological Society, p. 447.  28 17. Eastwood, P.R., D. Hiliman and K.E. Finucane (1994). Ventilatory responses to inspiratory threshold loading and role of muscle fatigue in task failure. .L AppL PhysioL 76: 185-195. 18. Enoka, R.M. and D.G. Stuart (1992). Neurobiology of muscle fatigue. J AppL PhysioL 72: 1631-1648. 19. Farkas, G.A., and C. Roussos (1983). Diaphragm in emphysematous hamsters: Sarcomere adaptability. J AppL PhysioL 54: 1635-40. 20. Farkas, G. A. (1991). Functional characteristics of the respiratory muscles. Semmars in Respir. Med. 12: 247-257. 21. Ferguson G.T., C.G. Irvin, and R.M. Cherniack (1990). Relationship of diaphragm glycogen, lactate, and function to respiratory failure. Am. Rev. Resp. Dis. 141: 926-932. 22. Freedman, S. and E.J.M. Campbell (1970). The ability of normal subjects to tolerate added inspiratory loads. Respir. PhysioL 10: 213-235. 23. Grassino, A. and T. Clanton (1991). Respiratory muscle fatigue. Sem. Respir. Med. 12: 305-321. 24. Green, H.J. Neuromuscular aspects of fatigue (1987). Can. 1. Spt. Sci, 12: 7S-18S. 25. Green, H.J., Rechman, H. and D. Pette (1984). Inter- and intra-species comparisons of fibre type distribution and of succinate dehydrogenase activity in type I, ha and JIb fibres of mammalian diaphragms. Histochemistiy 81: 67-73.  29 26. Huitman, E. and H. Sjoholm (1986). Biochemical causes of fatigue. In: Human Muscle Power, N.L. Jones, N. M. Cortney, and A. J. McComas (Eds.) Champaign, IL: Human Kinetics pp. 215-235. 27. Im Hof, V., P. West and M. Younes (1986). Steady state response of normal subjects to inspiratory resistive load. .1 AppL PhysioL 60: 1471-81. 28. Jones G.L., K.J. Killian, E. Summers and N.L. Jones (1985). Inspiratory muscle forces and endurance in maximum resistive loading. J AppL PhysioL 58: 1608-1615. 29. Kirkendall, D. T. (1990). Mechanisms of peripheral fatigue. Med. Sci. Sports Exerc. 22: 444-449. 30. Leak, L.V. (1979). Gross and ultrastructural morphologic features of the diaphragm. Am. Rev. Respir Dis. 119:3-21. 31. Lieber, R. L. Skeletal Muscle Structure and Function. Williams & Wilkins, Baltimore, MD, p.55, 1992. 32. Loring, S.H. and A. De Troyer (1985). Actions of the respiratory muscles. In Lung Biology in Health and Disease. The Thorax, Part A, Vol. 29 (Ed: C. Lenfant), Marcel Dekker Inc., N.Y., pp. 327-349. 33. Macklem, P.T. and C. Roussos (1977). Respiratory muscle fatigue: a case of respiratory failure? Clin. ScL MoL Med. 53: 419-22, 1977. 34.  Mador, J.M., U.J. Maglang and T.J. Kufel (1994). Twitch following voluntary diaphragmatic contraction. Am. J Resp. Crit. Care Med. 149: 739-743.  30 35. Mayock, D.E., R. J. Badura, J.F. Watchko, T.A. Standaert and D.E. Woodrum (1987). Response to resistive loading in the newborn piglet. Pediatr. Res. 21:121-125. 36. Mayock, D.E., T.A. Standaert, T.D. Murphy and D.E. Woodrum (1991). Diaphragmatic force and substrate response to resistive loaded breathing in the piglet. J AppL PhysioL 70: 70-76. 37. Mead, J. (1979). Responses to loaded breathing A critique and a synthesis. .  BuLL Eur. PhysiopathoL Respii 15: 61-71. 38. Moxham, J., A.J.R. Morris, S.G., Spiro, R.H.T., Edwards and M. Green (1981). Contractile properties and fatigue of the diaphragm in man. Thorax 36: 164-168. 39. Moxham, J., R.H.T. Edwards, M. Aubier, A. Dc Troyer, G. Farkas, P.T. Macklem and C. Roussos (1982).  Changes in EMG power spectrum  (high-low ratio) with force fatigue in humans. J AppL PhysioL 53: 10941099. 40. National Heart Lung and Blood Institute Workshop Summary. Respiratory Muscle Fatigue (1990). Am. Rev. Respir Dis. 142: 474-480. 41. Naess, K. and A. Storm-Mathesen (1955). Fatigue and sustained tetanic contractions. Acta PhysioL Scand. 34: 35 1-366. 42. Oliven, A., S. Lohda, M.A. Adams, B Simhjai and S.G. Kelsen (1988). Effect of fatiguing resistive loads on the level and pattern of respiratory activity in awake goats. Respir PhysioL 73: 3 11-324.  31 43. Petrozzino, J.J., A.T. Scardella, T.V. Santiago and N. H. Edelman (1992). Dichioroacetate blocks endogenous opioid effects during inspiratory flowresistive loading. J AppL PhysioL 72: 590-596. 44. Petrozzino, J.J., A.T. Scardella, N.H. Edelman and T.V. Santiago (1993). Respiratory muscle acidosis stimulates endogenous opioids during inspiratory loading. Am. Rev. Respu Dis. 147: 607-6 15. 45. Rochester, D.F. (1991). Respiratory muscle weakness, pattern of breathing, and CO 2 retention in chronic obstructive pulmonary disease. Am. Rev. Respu Dis. 143: 901-903. 46. Scardella, A.T., R.A. Parisi, D.K. Phair, T.V. Santiago and N.H. Edelman (1986). The role of endogenous opioids in the ventilator)’ response to acute flow resistive loads. Am. Rev. Respir Dis. 133: 26-31. 47. Scardella, A.T., T.V. Santiago and N. H. Edelman (1989). Naloxone alters the early response to an inspiratory flow-resistive load. J AppL PhysioL 67: 1747-1753. 48. Scardella, A.T., JJ. Petrozinno, M. Mandel, N.H. Edelman and T.V. Santiago (1990). Endogenous opioid effects on abdominal muscle activity during inspiratory loading. J AppL PhysioL 69: 1104-1109. 49. Sieck, G.C., A. Mazar and M. J. Belman (1985). Changes in diaphragmatic EMG spectra during hyperpneic loads. Respii PhysioL 61: 137-152. 50. Vollestad N.K. and O.M. Sejersted (1988). Biochemical correlates of fatigue. Europ. J. AppL PhysioL 57: 336-347.  32 51. Westerbiad, H.J.A., 3. Lee, J. Lannnergren and D.G. Allen (1991). Cellular mechanisms of fatigue in skeletal muscle. Am. J PhysioL 261: C195-C209. 52. Wragg, S., C. Hamnegard, J. Road, J. Goidsone, M. Green and 3. Moxham (1993). Twitch Pdi depends on contractile history. Am. Rev. Respu Dis. 147: A699 (Abstract). 53. Yan, S., T. Similowski, A.P. Gauthier, P.T. Macklem, and F. Bellemare (1992). Effect of fatigue on diaphragmatic function at different lung volumes. J AppL PhysioL 72: 1064-1067.  33 II: Diaphragm Activation and Phrenic Nerve Impulse Propagation During  Prolonged Inspiratory Resistive Loading  Introduction  Experimental animals subjected to inspiratory flow resistive loads breathe spontaneously against loads sufficiently intense to cause a decrease in alveolar ventilation and result in hypercapnic ventilatory failure. Several investigators have suggested that diaphragm fatigue is an important, if not predominant, determinant of hypercapnic ventilatory failure in these studies (Aldrich and Appel, 1985; Aldrich 1987, 1988, 1991; Alexandrovna and Isaev, 1990; Bazzy and Haddad, 1984; Bazzy and Donnelly, 1993; Mayock et aL 1987, 1991; Oliven et aL 1988).  Failure of processes proximal to the diaphragm during inspiratory resistive loading has been hypothesized in studies which show that indices of fatigue at more distal sites [such as diaphragm pressure response to supramaximal tetanic stimulation of the phrenic nerve (frequency-pressure curve), diaphragm electromyogram centroid frequency and the concentrations of glycogen and lactate in the diaphragm] remain unchanged (Aldrich, 1988; Watchko et aL, 1988; Ferguson et aL, 1990). In the rabbit and sheep models of inspiratory resistive loading, electromyographic evidence for neuromuscular transmission failure has  34 been documented (Aldrich, 1987; Bazzy and Donnelly, 1993). However, Mayock  et aL found no electromyographic evidence for neuromuscular transmission failure (Mayock et aL, 1991; 1992) in the piglet with a similar fatiguing load.  More proximally, the activity of the phrenic nerve (ENGdi), an index of central drive to the diaphragm during inspiratory resistive loading, has recently been examined in the cat and sheep. According to these studies, there is no decrease in neural drive to the diaphragm at a time when a drop in pressure generation of the diaphragm (Pdi) is observed (task failure). However, decreases in ENGdi are observed in the period following task failure or when higher loads are applied (Bazzy and Donnelly, 1993; Aleksandrovna and Isaev, 1990).  The role of central and neuromuscular fatigue during inspiratory resistive loading therefore remains controversial. We set out to answer two questions in a series of studies on rabbits subjected to inspiratory resistive loads of varying intensities and durations 1) Does neural activation of the diaphragm decline with inspiratory resistive loading?  2) Does neuromuscular transmission to the  diaphragm decrease during inspiratory resistive loading? We describe the effects of four inspiratory resistive loads on ventilation and diaphragm function in the anesthetized rabbit. Loads 1 and 2 were applied for a four hour period. Load 1 was less and Load 2 was equivalent to loads previously reported to be sufficient to cause diaphragm fatigue in the rabbit (Aldrich, 1985, 1987). Loads  35 3 and 4 were greater in intensity and they are applied in a stepwise fashion until task failure. Our studies show that both neural activation and neuromuscular transmission to the diaphragm remain intact even under extreme loading conditions when the respiratory muscles failed as pressure generator during tidal ventilation.  Methods  Animals and groups.  Thirty three, male, New Zealand White rabbits were  obtained from Geo-Bat Rabbitries (Abbotsford, B.C.) and cared for according to the principles outlined by the Canadian Council for Animal Care at the Animal Resource Unit facility at the University Hospital (U.B.C). Experimental protocols received ethical approval from the University of British Columbia Animal Care Committee.  Preparation.  The rabbits were anesthetized with intramuscular injection of  Ketamine (Ketavet, Parke-Davis, 30 mg/kg) and a sedative, Xylazine (Rompun, Bayer, 7 mg/kg) in the hind limb. Anesthesia was maintained throughout the studies by supplementing half the initial dose every 30-40 minutes.  Rectal  temperature was continuously monitored and maintained between 37.5-39°C with a heating pad. Saline was infused via a marginal ear vein to maintain blood pressure.  36 A schematic diagram of the experimental setup is shown in Figure 5. Rabbits were placed in the supine position and the trachea was cannulated (I.D. = 4 mm stainless steel L shaped tubing) and connected to a heated pneumotachograph (Fleisch # 00) in series with a miniature two way non-rebreathing valve (Hans Rudolph no. 2814; 2.5 ml dead space). Pressure across the pneumotachograph was measured with a differential pressure transducer (± 2 cm H20, Validyne MP 45) and a carrier preamplifier (Gould model 13-44615-35). The carrier output was electronically integrated (Gould integrator amplifier 13-4615-70) to record tidal volume (calibrated for a range of 5-30 ml). An adjustable needle valve was placed at the inspiratory port of the non-rebreathing valve to apply flow resistive loads. Supplemental oxygen was provided at the inspired port throughout the experiments and adjusted to prevent hypoxia when possible. The left carotid artery was cannulated with polyethylene tubing (PE 150, Clay Adams, Parsippany, N.J.) to measure blood pressure and to sample blood for blood gas and pH analysis (Model 168 pH/blood gas analyzer, Corning Medical, Medfield, MA).  The left phrenic rootlet (C4) was exposed at the neck. The nerve was cleaned then secured in contact with bipolar platinum electrodes and immersed in a pool of mineral oil. Whole phrenic nerve discharge activity recorded from the intact nerve was amplified (Grass P5 series AC preampliers; Grass Instruments Co., Quincy, MA), filtered (100 and 10,000 Hz low and high frequency cutoffs) and  37 whole wave rectified.  The moving average (time constant = 0.1 sec) was  computed using a filter with Paynter response (EMG signal processor, Raytech Instruments, Vancouver, Canada). The phrenic nerve discharge activity was monitored on an oscilloscope (Gould DST 1421) and the moving average (ENGdi) amplified (Gould medium gain DC preamplifier model 13-4615-10) and recorded on an 8 channel chart recorder (Gould model 8188-812, Cleveland, OH).  To record the diaphragm electromyogram and evoked compound diaphragm potentials (M-waves), a midline upper abdominal incision was made and the uninsulated tips of two multi-stranded stainless steel fine wires (Cooner wire #AS 631) were sutured 1 cm apart into the left costal hemi-diaphragm midway between the costal margin and central tendon.  The moving average of the  electromyogram (EMGdi) was processed as described above for the phrenic nerve signal and continuously recorded on the chart paper. M- waves were elicited during expiration at zero flow. Single shocks (0.2 msec duration) were delivered to the left phrenic nerve via a Grass constant current stimulator (model S48 equipped with a PSIU6E stimulus isolation unit) and the M-waves were displayed on the oscilloscope. The current stimulus was increased until maximum M-wave amplitude was observed and then set at five times this threshold during the experiment.  38 Transdiaphragmatic pressure (Pdi) was measured using two air filled ballooncatheters assembled from polyethylene tubing (PE 200 Clay Adams, Parsippany, N.J.) and 5 cm long latex balloons (A&E Corp., Farmingdale, NJ.).  One  balloon catheter was placed through the abdominal incision underneath the diaphragm (Pab) and a second balloon-catheter was placed transorally into the esophagus (Pes).  The catheters were placed across a differential pressure  transducer (Validyne, Northridge, CA) to measure Pdi as Pab-Pes.  The  esophageal catheter was positioned where the greatest negative peak pressure could be obtained during tidal inspiration. Airway pressure (Pao) was measured at the tracheal tube using a differential pressure transducer (Validyne, Northridge, CA).  To ensure an adequate translation of pleural pressure to  esophageal pressure, both airway pressure and esophageal pressure were recorded during an occluded inspiratory effort at end expiratory lung volume. A pressure difference of less than 10% in the two pressures was deemed acceptable.  39  Figure 5. A Schematic Diagram of the Experimental Setup.  LID  I.V Saline  Arterial Blood Gases & pH Arterial Blood Pressure  closed during  Phrenic Ne,ve Stimulation Phrenw Neive Recording (ENGdI)  EMGdi  Body Temperature  41 Protocol A: The effect of prolonged mspiratoiy resistive loads on activation of the diaphragm.  After a 20 minute period of stabilization, baseline measurements were made. Resistive loading was applied at end expiration in two groups of seven rabbits. This was accomplished by adjusting the inspiratory needle valve to increase swings in Pdi to approximately 1.5 (Load 1) or 5 times baseline values (Load 2). Inspiratory resistive loads were maintained for a period of 4 hours.  The  following variables were measured hourly. Arterial blood (0.3 ml) was sampled in heparinized syringes and arterial blood gases and pH measured within one minute of sampling, corrected for rectal temperature.  Arterial bicarbonate  concentration was calculated based on predicted equations. Pressures generated by the respiratory system (Pao, Pdi), ENGd1, EMGdi, airflow, tidal volume, arterial blood pressure were recorded continuously. The mean value for tidal volume, breathing frequency, peak inspiratory pressures (Pao, Pdi) and peak ENGdi and EMGdi were calculated hourly during a representative 60 second interval. Duration of inspiration (TI) and expiration (TE) were determined from the electromyographic recordings. TI was defined from the EMGdi as the time from initial rise to the point where a rapid decline was first observed. TE spanned the time from the rapid decline in the EMGdi to its next initial rise. In a separate sham control group (n = 6), rabbits were allowed to breathe spontaneously for 4 hours, without any resistive load, under the same anesthetic, surgical and variable measurement regimen as the loaded group.  42 Protocol B: The effect of severe incremental inspiratoiy resistive loads on activation and neuromuscular transmission to the diaphragm.  Another group of rabbits (n =13) were subjected, in a stepwise fashion, to two intense inspiratory resistive loads sufficient to produce additional 10-15 cm H20 increments in airway inspiratory pressure (Loads 3 and 4). Following baseline measurements, arterial blood samples, peak ENGdi, EMGdi and respiratory variables were analyzed every 10 minutes. The integrity of neuromuscular transmission to the diaphragm was determined by evoked diaphragm potentials (M-waves) obtained in duplicate every 10 minutes. Baseline measures of M wave amplitude (peak to peak, ptp) were made both before and immediately following the imposition of inspiratory resistive loads. The baseline M wave was stored on the oscilloscope and compared to the M waves monitored during the loaded period. To ensure that the stimulus was maximal, M-wave thresholds were confirmed after each set of M-wave recordings.  Data Analysis.  All variables were statistically compared before and after inspiratory loading using a paired Student’s t-test. A Friedman’s ANOVA (repeated measures) was used to detect any significant changes during the loaded period, followed by a Wilcoxon signed rank test to determine which values were significantly different  43 from each other (Systat 5.02). Statistical significance was defined as P <0.05. All values represent mean  ±  S.E.M.  Results  The inspiratory resistive loads are defined in this study in terms of mean target peak airway inspiratory pressures (Pao). These target loads and the equivalent resistances are tabulated in Appendix I along with values for loads reported in the literature in other animal studies of inspiratory resistive loading.  Protocol A: The effect of prolonged uzspiratoiy resistive loads on activation of the diaphragm  Following application of inspiratory resistive loads, ENGdi increased gradually and reached a steady state level within 10-15 minutes. This level of diaphragm activation was maintained throughout the 4 hour loading period and was directly proportional to the severity of the load. Parallel changes in transdiaphragmatic pressure swings (Pdi) and diaphragm phasic inspiratory activity (EMGdi) were observed during the 4 hour period (Figure 6). The ENGdi:EMGd1 ratio did not change during the loaded period with either load (Table I). Figure 7 shows the change in transdiaphragmatic pressure swings (Pdi) for the target inspiratory pressures (Pao) achieved in the two groups of rabbits exposed to prolonged  44 inspiratory resistive loads and in the sham control group.  Within the first hour breathing against Loads 1 and 2, ventilation (<rE) was found to drop 15% and 40% respectively relative to baseline (Figure 8). remainder of the loaded period, ventilation  (‘TE)  For the  was maintained (Figure 8).  Breathing against Load 1, the initial drop in VE resulted entirely from a decrease in breathing frequency (44± 5 versus 34± 4 min.’, p <0.05). The frequency of breathing (Bf) decreased due to an increase in Ti (0.69± 0.08 versus 0.84± 0.05 sec. p < 0.05). There was no significant change in tidal volume or expiratory time. In contrast, breathing against Load 2, the initial drop in ‘CTE resulted from a decrease in Bf (44± 5 versus 31± 3, p <0.05) and a decrease in VT (Table I). The frequency of breathing decreased due to an increase in both inspiratory and expiratory times (Table I).  During loaded breathing there was a gradual decrease in both pH and a gradual increase in PaCO2 (Figure 8). Pa02 was maintained above 100 mm Hg and no significant change in arterial bicarbonate levels was detected throughout the 4 hours of loaded breathing (Table I). In the sham control group, no significant changes in respiratory variables were observed (Figures 6 and 7). Blood gas analysis showed only a decrease in pH after 3 hours of anesthesia (Figure 8).  45  Figure 6:  Peak electrical activity of the phrenic nerve (ENGdi),  the costal  diaphragm (EMGdI) and transdiaphragmatic pressure swings (Pdi) at baseline (B) and during prolonged inspiratory resistive loads in anesthetized rabbits.  (. load 1, n=7; • load 2, n=7)  Six sham control rabbits  (0)  breathing spontaneously  without imposition of a resistive load under the same anesthetic regime.  *  =  Significantly different from baseline (B), p < 0.05.  (I)  C  D 0  0  0  ci CD ci  0  0  (-  I  *  *  I I I  I  *  *  *  I  I  -i N) NJ CJ 0(31001001001  PdI (cm H20)  cJ  1”)-  uJ  0 I  0 0  —i  EMG  I  N.) 0 0  (% I  C’4 0 0  *  *  0 0  -  baseline) 01  0 0  —  -  QJ  0 I  0 0  —i  ENC  I  0 0  N)  (% I  0 0  C.J  *  *  I  0 0  -  *  baseline)  *  0 0  01  47  Figure 7.  Peak inspiratory airway pressure (Pao) and transdiaphragmatic  pressure swings (Pdi) at baseline (B) and during prolonged inspiratory resistive loads (.load 1, n=7; • load 2, n=7) in anesthetized rabbits. Six sham control rabbits  (0)  breathing spontaneously without imposition of a resistive load under  the same anesthetic regimen.  *  =  Significantly different from baseline (B), p <0.05.  35 30 0  * *  *  25 20  E C)  15  -o  10  *  *  3  4  0  5 0 B  1  2  Loaded period (hours)  —35 —s  0 I  *  —30  *  *  *  *  *  2  3  4  *  E—20 C-) 0 C  a 1 0  *  —5 0 B  1  Loaded period (hours)  49  Figure 8: Changes in arterial pH, pCO 2 and ventilation inspiratory resistive loads  (.  rabbits. Six control rabbits  (0)  ((rE)  during prolonged  Load 1, n = 7; • Load 2, n = 7) in anesthetized breathing spontaneously without imposition of a  resistive load.  *  t  =  Significantly different from baseline (B), p < 0.05.  =  Significantly different from the first hour of loaded breathing, p < 0.05.  0,  0 C  0 ci  CD  -c  0 U UCD U-  r\J  -i  I  I  I  I  I  I  I  00000000  4o1o)-JcYJcD0  VE (%baseline)  I  I  00  -  (,1  -  -  r’J  -  U,  I  I  -I-  \  F•i*  I  I  I  I  I  V 1* I  I  I  0000000000  O(D0-N)  POCO2 (mm Hg)  U,  b _L  I  pH  *  I  *  CJ  51 Table I: Mean (± S.E.M.) values for respiratory variables at baseline and during Load 2 (n=7).  Duration of Loading (hours) Baseline  1  2  4  VT (ml)  18(2)  15(1)*  15(1)*  13(1)*t  13(1)*t  Bf (min’)  44(5)  31(3)*  32(3)*  31(3)*  31(3)*  0.8 (0.3)  0.9 (0.4)  ENGdi/ EMGdi  1 (0.0)  16(3)*  VT/TI (mi/sec)  25(5)  TI/TrOT  0.50 (0.02)  0.47 (0.03)  Pa02 (mm Hg)  258 (42)  [HCO ] 3 (mM/L)  24 (1)  TI (sec)  0.7 (0.1)  TE (sec)  0.7 (0.1)  16(2)*  0.9 (0.4)  0.9 (0.4)  15(2)*  14(2)*t  0.47 (0.03)  0.44 (0.03)  0.45 (0.03)  239 (27)  214 (31)  206 (33)  197 (29)  24 (1)  27 (1)  28 (1)  26.4 (1.2)  1.0* (0.1)  1.0* (0.2)  0.9* (0.1)  1.0* (0.1)  1.0* (0.1)  1.1* (0.2)  1.2* (0.2)  1.2* (0.2)  *p <0.05 compared to baseline. tp <0.05 compared to value at other loaded periods.  52 Protocol B:  The effect of severe inspiratoiy resistive loads on activation and  neuromuscular transmission to the diaphragm.  Loads 3 and 4 resulted in a sustained four to six fold increases in both ENGd1 and EMGdi (Figure 9). There was no significant change in the ENGdi to EMGdI ratio and no significant reduction in the amplitude of the evoked compound diaphragm potentials observed during loading (Table II). Nor was there any significant change in the duration of M-waves. In fact, the compound action potentials were superimposable and no significant change in peak to peak amplitude was observed during loaded breathing (Figure 13, panel A).  Application of Loads 3 and 4 produced target inspiratory pressures exceeding -30 cm H20. The increase in resistance required to achieve these extreme target pressures was associated with arousal and swallowing.  Arousal due to the  sudden imposition of extreme inspiratory resistive loads was circumvented by a gradual increase of inspiratory resistance via the needle valve until the desired target peak Pao was obtained. Target loads were established within 10 minutes. Rabbits swallowed throughout load 3 and load 4. Swallowing was reflected as positive pressure swings in the esophageal pressure recordings. Accordingly, a continuous estimate of pleural pressure by recording esophageal pressure could not be obtained. For this reason, the peak inspiratory pressure generated by the respiratory system at tracheal opening (Pao) is shown as an index of mechanical  53  Figure 9: Peak electrical activity of the phrenic nerve (ENGdi) and the costal diaphragm (EMGd1) at baseline (B) and during severe loads. Failure to sustain inspiratory pressure, designated “F” on the abscissa, occurs 40-60 minutes after initiation of Load 4.  *  =  Significantly different from baseline, p < 0.05.  54  1 200  Load3  Load4  1100  V  •ENG  1000  a)  *  I  900  IEMG  800  (1)  600  C  500  >  400  C  300  *  *  *  20  30  40  *  200  100 0  B  10  50  10 20 30  Loaded period (minutes)  F  55 failure of the respiratory muscles. Pao increased with graded loading. There was a linear relationship between Pao and EMGdi and between Pao and ENGdi in both individual rabbits and among the groups of rabbits studied at different inspiratory resistive loads (Figure 10). The highest Pao was obtained with Load 4 and could be sustained for 30 minutes (Figure 11). Thereafter, Pao dropped significantly within 10 to 30 minutes (represented as time “F” on the abscissa, Figure 11). Pao decreased from 52± 1 to 42± 2 during this period indicating failure of the respiratory muscles to maintain a target force (task failure).  56  Figure 10: The relationship between peak inspiratory pressure (Pao) and indices of activity (top panel) and drive to (bottom panel) the diaphragm. Symbols represent the mean ± S.E.M. for the following groups: • Load 1 (n=7); • Load 2 (n=7); v Load 3 (n=8);  0  baseline (n=7); Load 4 (n=8).  Solid lines represent first order regression, r = 0.99. Dashed lines correspond to 95% confidence limits for the regression.  800 700 ci C  600  ci C’, Q  500 /  400 -o  300  0 200 100 0  I. . _i  I  I  I  I  I  I  0  10  20  30  40  50  60  Pao (cm H 0) 2 800  /  700  / /  ci C a) C’, C  600 /  500  /  400 300 -Q  z 200  / /  LU  / /  100  / /  0 II  / I  0  j  10  •  20  I  I  I  30  40  50  Pao (cm H 0) 2  •  60  58  Figure 11:  Changes in arterial blood gases (Pa02, PaCO2) and negative  inspiratory pressures recorded at the airway (Pao) at baseline (B) and during severe inspiratory resistive loads in anesthetized rabbits (n=8). Failure to sustain inspiratory pressure designated “F” on the abscissa occurs 40-60 minutes after initiation of Load 4.  *  t  =  =  Significantly different from baseline, p <0.05. Significantly different from the previous loaded period, p < 0.05.  S9  —65  Load3  200  Load 4  —60 V  —55 0 I  —45  E  —40  ‘3 0  ci  11  —50  —30  .4  —25  *  1 60  /1  1 20  60  *t I*f  —10 —5 0  1 00 80  —15  AI  B  J__L_L_  10  jL  0)  =  E E  L  40  E1  20 30 40 50 10 20 30 Loaded period (minutes)  0 C)  *Y  —20  p.m  1 40  I  —35  0  180  F  20 0  0  60 Severe loading resulted in a 40-45% decrease in minute ventilation (Figure 12) along with a decrease in mean inspiratory flow rate (VT/TI) and an increase in respiratory times (Table II). No change in inspiratory duration was seen with decreased inspiratory pressure. However, a significant increase in expiratory duration resulted in a decreased duty cycle (TI/Pr0T) (Table II). In addition, Load 4 led to significant acidemia (Table II), hypercapnia, and hypoxemia (Figure 11).  Introduction of the severe loads resulted in a pneumothorax in five rabbits. These animals were withdrawn from the study. A sudden change in breathing pattern and a large reduction in M-wave amplitude was observed in all rabbits that developed a pneumothorax. Post mortem examination was carried on all rabbits whose data was included in the study (n = 8) to confirm an intact pleura.  61  Figure 12:  The average decrease in minute ventilation  (‘7E)  during severe  inspiratory resistive loading as percent of baseline (B) value. Failure to sustain inspiratory pressure designated “F” on the abscissa occurs 40-60 minutes after initiation of Load 4.  *  =  Significantly different from baseline (B), p <0.05.  =  Significantly different from the previous loaded period, p <0.05.  2  100  d4  d3  V  80* T  C  *  *  * T  60  III  *  I  *  NT I  0  I  I  I  Ij  10 20 30 40 50  I  I  I  I  10 20 30 F  Loaded period (minutes)  I  63 Table II: Mean  (i  S.E.M.) values for respiratory variables at baseline and  during Load 4 (n = 8). Duration of Loading (minutes) Baseline  10  20  30  F  VT (ml)  18(2)  15(2)*  13(1)*  12(1)*  Bf (1/mm)  56(5)  38(6)*  38(5)*  39(4)*  31(4)*t  ENGd1/ EMGdi  1.0 (0.0)  1.2 (0.2)  1.2 (0.2)  1.4 (0.4)  1.7 (0.5)  VT/Ti (mi/sec)  31(3)  21(2)*  21(2)*  19(2)*  0.50 (0.02)  0.43* (0.02)  0.43* (0.02)  0.41* (0.02)  (0.02)  7.36 (0.01)  7.14* (0.03)  7.10* (0.03)  7.06* (0.03)  (0.04)  27 (0.4)  28 (0.6)  29 (0.7)  28 (0.7)  28 (0.7)  TI/TrOT  pH [HC0 ] 3 (mM/L) M wave (ptp-mV)  3.9 (0.4)  3.9 (0.5)  3.5 (0.4)  3.4 (0.4)  3.2 (0.4)  Ti (sec)  0.6 (0.0)  0.7* (0.1)  0.7* (0.1)  0.7* (0.1)  0.7* (0.1)  TE (sec)  0.6 (0.1)  1.2* (0.2)  1.1* (0.2)  1.1* (0.2)  1.6*4 (0.4)  *p <0.05 compared to baseline. <0.05 compared to value at other loaded periods. tp <0.05 compared to value at 30 minutes.  64 Discussion  This study demonstrates that despite hypoventilation and significant changes in blood gases, both activation of and neuromuscular transmission to the diaphragm are maintained throughout prolonged and severe inspiratory resistive loading even when the target inspiratory pressure is no longer sustained and a critical level of Pa02 is reached (Figure 11, load 4). Additionally, we find no significant change in the ratio between the activity of the phrenic nerve and that of the diaphragm (ENGdi:EMGdi) throughout loaded breathing at all intensities (Loads 1-4). These results provide additional support to the hypothesis that there is no significant change in neuromuscular transmission during both sustainable and exhaustive inspiratory resistive loads.  Evaluation of Methods  The study of anesthetized animals breathing spontaneously against inspiratory loads offers two distinct advantages in determining the locus (loci) of failure in experimental fatigue of the diaphragm. In the awake subject, apparent failure of the diaphragm as a pressure generator may, in fact, be due to loss of motivation of the subject unwilling to breathe against intolerable inspiratory loads. This behavioral response is predictable and has been documented in untrained subjects asked to breathe against inspiratory resistive loads. The same  65 loss of motivation may underlie performance decrements in the awake animal without the need to resort to the notion of fatigue as an explanation. Therefore, the first benefit of this protocol is that in the anesthetized preparation, it is assumed conscious factors such as motivational fatigue do not play a role in diaphragm performance. The second advantage is that spontaneous breathing against an inspiratory resistive load allows fatigue to develop without the need to impose a non-physiological pattern of diaphragm activation such as electrical stimulation.  Ketamine-xylazine anesthesia was chosen for several reasons. First, it has a wide margin of safety when administered intramuscularly. Secondly, it is one of the few anesthetics which does not depress spontaneous breathing. Finally, its use permits comparison of our study with results obtained in previous studies of inspiratory loading in the rabbit all of which have used the same anesthetic agent (Aldrich and Appel, 1985; Aldrich 1987, 1988; Ferguson Ct aL, 1990).  Ketamine alone produces a so-called “dissociative” anesthetic state that has been described as a functional and electrophysiological dissociation between the thalamo-neocortical and limbic systems. This unique state is characterized by catalepsy involving unconsciousness and somatic analgesia without muscular relaxation (White et aL, 1982). In combination with xylazine, a potent hypnotic with central muscular relaxant properties, ketamine provides adequate analgesia  66 and muscle relaxation for surgical procedures (Borkowksi et aL, 1990). Hypotension, hypercapnia and respiratory acidosis and hypoxia are characteristic cardiopulmonary effects of nonvolatile anesthetics in rabbits including ketamine xylazine anesthesia (Borkowksi et aL, 1990). We documented the effects of our anesthetic regime on ventilation, PaCO2 and arterial pH over a 4 hour period on rabbits (see control group, Figure 8). Prolonged anesthesia alone did not result in any significant change in ventilation or PaCO2 from baseline. Resting values of PaCO2 in the awake rabbits range from 25 to 37 mm Hg (Honda, 1968; Gauthier, 1973)  .  The range of baseline values of PaCO2 in our anesthetized  preparation are considerably higher (37-48 mm Hg) owing to the effects of hyperoxia and anesthesia (Honda, 1968; Borkowski et aL, 1990). Hypoxia and hypotension were purposely avoided (by design), in order to assess diaphragm fatigue independent of these factors.  After 3 hours of anesthesia a slight  decrease in arterial pH was observed. This change is likely due to the repetitive ketamine [hydrochloride] administration employed in our protocol.  Neuromuscular Transmission to the Diaphragm  The occurrence of neuromuscular transmission failure during maximal voluntary contractions in skeletal muscle remains open to debate. It has been suggested that conflicting findings may be due to methodological differences (Bigland Ritchie, 1987). Merton (1954) was first to show the integrity of neuromuscular  67 transmission during voluntary contractions in the adductor pollicis. He reported no decrease in the size of muscle compound action potentials (M-waves) evoked by single maximal shocks to the ulnar nerve during sustained maximal contraction of the adductor pollicis muscle despite a force decay to almost zero. Stephens and Taylor (1972) presented conflicting results using the first interosseous muscle. In a subsequent series of experiments, Bigland-Ritchie and coworkers reexamined the adductor pollicis and first interosseous muscles of the hand as well as other limb muscles and demonstrated that neuromuscular transmission is preserved during maximal voluntary contractions (see Thomas et aL, 1985) not withstanding force decay in the muscle.  In the diaphragm, neuromuscular transmission, as assessed by the M-wave, remains intact in humans following both severe inspiratory resistive breathing (Aubier et aL, 1981, 1985, Yan et aL, 1992) and maximal expulsive manoeuvres (McKenzie et aL, 1992) which result in a decrease in Pdi.  Similarly, during  prolonged inspiratory resistive breathing in the anaesthetized piglet (Mayock et aL 1991) the M-wave does not change despite a drop in the frequency-pressure curve of the diaphragm. In contrast to these results, Aldrich (1987) has shown a reversible 44% decrease in diaphragm M-wave amplitude in the anesthetized rabbit after a period of 58± 14 minutes of inspiratory resistive loading under relatively moderate loads (adjusted to an average Pao = 27 cm H20). Bazzy and Donnelly (1993) have reported similar changes in the diaphragm M-wave  68 amplitude of two awake sheep breathing against severe inspiratory resistive loads. These changes were accompanied by an increase in the ENGdi:EMGd1 ratio of the diaphragm prior to failure of the diaphragm to generate inspiratory pressure in the awake sheep (Bazzy and Donnely, 1993).  In our studies, we have demonstrated that anesthetized rabbits can generate similar inspiratory pressures (Load 2) as the pressures observed in the Aldrich (1987) study for 4 hours without a significant change in the ENGdi:EMGd1 ratio. Furthermore, we observed no significant change in the ENGdi:EMGd1 or M wave, two indices of the integrity of neuromuscular transmission to the diaphragm during two successive periods of sustained inspiratory pressure at even greater loads (Load 3 and Load 4).  Methodological differences may explain the conflict between our findings and those of Aldrich (1987) and Bazzy and Donnelly (1993). It has been shown that inspiratory resistive breathing causes a decrease in end expiratory lung volume (Mayock et aL, 1987; Oliven et aL, 1988). We were particularly wary of the possibility that the M-wave could be altered by such an artifact. Indeed, in our pilot studies, the M-wave amplitude decreased immediately upon loading but remained constant throughout loading. Sudden removal of the load resulted in an immediate return of the M wave amplitude to baseline levels (Figure 13, panel B). In our view, the shift in baseline M-wave amplitude upon loaded  69 breathing makes the M-wave obtained immediately upon loading a better measure of baseline M-wave. The shift in M-wave with loading probably results from changes in muscle length (Kim et aL, 1985) and/or a change in the volume conductivity of the surrounding tissues (see Brancatisano et aL, 1989).  It is  unlikely that changes in volume conductivity affected our recordings because the diaphragm electrodes were placed away from the area of apposition where recordings are vulnerable to such influences (Grassino, Whitelaw and Milic Emili, 1976). The most likely explanation for the immediate shift in the M-wave upon loading is a change in diaphragm length in response to severe inspiratory resistive loading.  In addition to a decrease in M-wave amplitude due to mechanical artifact such as a change in muscle length, other distinct limitations in the use of M-wave amplitude have been documented.  For example, the M-wave amplitude of  shocks delivered during contractions produced by tetanic simulation or sustained voluntarily can increase transiently early in tetany [pseudo-facilitation] (Hicks et aL, 1989). It is important to note, however, that changes associated with M-wave amplitude during sustained contractions do not preclude the comparison of M wave amplitude generated during periods of rest in an intermittently active muscle. In our study, we observed no significant change in the amplitude or area of diaphragm M-wave generated during expiration throughout loaded breathing.  70  Figure 13:  Panel A: Representative evoked diaphragm compound potentials (M-waves) at baseline and at time intervals during the severe inspiratory resistive loads. 13  =  Load3; LA  =  Load 4;  “F” = at time of pressure failure.  Panel B:  An example of shift in baseline evoked diaphragm  compound potential (M wave) upon loaded breathing.  .9.,  0.5 m V  2 msec  L4, 10 mm  50 mm  loaded  Load removed  72 Constant Diaphragm Activation during Prolonged Loading  Hypercapnia inevitably followed inspiratory resistive loading during breathing against all loads examined in our study. It is interesting that with prolonged loading (Loads 1 and 2), neural activation of the diaphragm (ENGdi) is maintained at a constant level despite rising PaCO2 (Figure 6). Clearly, maximal neural activation of the diaphragm is not reached with these loads since applying a greater load (Load 3) results in higher electrical activity of the phrenic nerve (ENGdi) and electromyographic activity of the diaphragm (EMGd1) (Figures 6 and 9). A reduction in responsiveness to increasing Pco 2 may be due entirely to central effects of prolonged (greater than one hour) exposure to high levels of PaCO2 alone. Alternatively, it has been hypothesized that the augmented activity of group III and IV muscle afferents during inspiratory resistive loading produce an inhibition of drive to respiratory muscles, possibly by the release of endogenous opioids (Petrozzino et aL 1992).  Such feedback inhibition may  account for the constant activation of the diaphragm seen during prolonged inspiratory resistive loading.  The effect of prolonged exposure to hypercapnia on activity of the diaphragm and ventilation is examined and described using this rabbit preparation in Chapter III.  73 Prolonged Inspiratoiy Resistive Loading and Indices of Diaphragm Fatigue  Currently, there are no universally accepted indices of diaphragm fatigue. Studies examining inspiratory muscle function in anesthetized animals during inspiratory resistive loading define fatigue as a decline in the frequency-pressure curve of the diaphragm in response to supramaximal tetanic stimulation of the phrenic nerves. However, it has been shown that the pressure generated by the diaphragm (Pdi) and its activity (EMGdI) are maintained during six hours of inspiratory resistive loading despite a decline in the frequency-pressure curve observed within the first hour of loaded breathing (Mayock et aL 1991).  Consequently, a drop in the frequency-pressure relationship of the diaphragm does not correlate with the activity (EMGd1) or the pressure it generates during spontaneous breathing. Similarly, in our study, target Pdi is maintained for 4 hours under target pressures (Load 2) which have been shown to result in a significant decrease in the frequency-pressure curve within one hour using this preparation (Aldrich, 1987).  Such a functional loss in maximal pressure  generating capacity of the diaphragm in response to electrical stimulation reflects failure of all motor units to respond to supramaximal stimuli. However, unlike epiphrenic electrical stimulation, phrenic motor unit recruitment is not uniform during spontaneous loaded breathing (Cairns and Road, 1993).  Therefore,  pressure loss during electrical stimulation does not necessarily require the  74 conclusion that force loss will arise in response to spontaneous loaded breathing. In support of this view, we have demonstrated that inspiratory pressures higher than those previously reported in the literature (e.g. Aldrich, 1985, 1987, 1988; Ferguson et aL, 1990) can be maintained in the anethetized rabbit for periods up to or greater than 30 minutes. Furthermore, failure to maintain target pressure is associated only with the most severe load (Figure 12, load 4). Target inspiratory pressures (Pao) achieved under this extreme load are within the reported range of maximum Pdi obtained during supramaximal stimulation of the phrenic nerves in rabbits with bound abdomens (55 ± 5 cm H20; Aldrich 1985, 1987, 1988, Ferguson et aL, 1990) and well above those previously achieved in the spontaneously breathing loaded animals prior to task failure (35± 2.6 cm H20; Ferguson et aL, 1990). Figure 14 shows the relationship between Pao and Pdi during brief loads in this preparation. It is clear that there is a good 1:1 relationship between Pao and Pdi at pressures between 10-35 cm H20. At higher loads, Pao overestimates Pdi by approximately 5 cm H20 indicating that performance seen under Load 4 reflects the extreme sustainable pressure of the diaphragm. In support of this view, Cairns and Road (1993) have shown that all phrenic motoneurons examined under inspiratory resistive loading are recruited by Pdi = 40 cm H20 and many fire at very high rates of activation (approaching 80 Hz).  75  Figure 14:  The relationship between peak inspiratory pressure (Pao) and  transdiaphragmatic pressure (Pdi) swings in two rabbits during brief inspiratory resistive loading. Airway pressures of -55 cm H20 measured during maximal load (Load 4) compare to Pdi swings of 50 cm H20. Symbols represent peak values during single inspirations.  7  60 -50 0 40 c\J  E -30 C) 0 C 0  .20  -10 0  cpl, Nov18  0 0 10  I  I  20 30 40 50 Pdi (cm H20)  60  -60 -50 0  Q8 40  0 C -20  -10 cp6, Nov18  0  0 0  i  10  20  I  I  I  I  30  40  50  60  Pdi (cm H20)  77 Task Failure of the Respirato,y Muscles during Severe Inspiratoiy Resistive Loading  Several factors may contribute to our subjects’ failure to maintain the target inspiratory pressure observed under Load 4.  First, hypercapnia has been  reported to attenuate diaphragm contractility in humans, dogs and piglets (Juan et aL, 1984; Schnader et aL, 1985; Watchko et aL, 1987). It is important to note  however that the effect of high levels of carbon dioxide on diaphragm contractility are assessed by the frequency-pressure technique. As mentioned above, this technique measures the maximal contractile response (pressure output) of the diaphragm to a supramaximal stimulus which generates an artificial pattern of muscle fibre recruitment. Whether the same mechnisms leading to loss of maximal contractility contribute to diaphragm task failure is open to debate. In the rabbit, in particular, it has been shown that short term C02 rebreathing does not affect diaphragm contractility (Aldrich and Appel, 1985). During each of the loads applied in this study, hypercapnia is severe and long term.  Consequently, any change in diaphragm contractility that would  contribute to the loss in pressure generation by this muscle should be manifest under all the loads examined and relatively early on. Since the activity of the diaphragm (EMGdi) and its ability to maintain pressure were maintained throughout loading for prolonged periods it seems unlikely that any change in the contractility of the diaphragm due to hypercapnia contributed to the loss in pressure generation.  78 Hypoxemia is a second factor that could result in respiratory muscle task failure or central nervous system depression, or both. As we were unable to measure Pdi during the extreme load at the time when inspiratory pressure (Pao) decreased, it is possible that the diaphragm or other respiratory muscles activated during the load, or both, displayed peripheral (myogenic) fatigue. One study has examined biochemical correlates of diaphragm fatigue during inspiratory resistive loading in the rabbit (Ferguson, Irvin and Cherniack, 1990). In this study, respiratory arrest induced by incremental inspiratory threshold loading resulted in the same Pa02 as measured in our study at task failure with no evidence of contractile fatigue of the diaphragm or alterations in diaphragm glycogen or lactate concentrations.  Furthermore, the role of hypoxia related  depression of central output to the respiratory muscles other than the diaphragm cannot be excluded since there seem to be differences in susceptibility of these outputs to hypoxia (Fregosi et aL, 1987; Neubauer, Ct aL, 1990). Hence, the loss of inspiratory pressure observed in our study could have been due to peripheral fatigue of the respiratory muscles or to hypoxic depression of output to the respiratory muscles.  In conclusion, although it would be difficult to predict to what extent the progressive and severe perturbations in arterial Pc0 , P0 2 2 and pH alone or interactively contribute to respiratory muscle task failure during extreme loads, the data in our studies clearly show that activation and neuromuscular  79 transmission to the diaphragm are maintained throughout both prolonged and severe inspiratory resistive loads in the anesthetized rabbit.  80 References  1. Aldrich T.K. and D. Appel (1985). Diaphragm fatigue induced by inspiratory resistive loading in spontaneously breathing rabbits. J AppL PhysioL 59: 1527-1532. 2. Aldrich, T.K. (1987). Transmission fatigue of the rabbit diaphragm. Respir PhysioL 69: 307-319. 3. Aldrich T.K. (1988). Central fatigue of the rabbit diaphragm. Lung 166: 233241. 4. Aleksandrovna, N.P. and G.G. Isaev (1990). Central and peripheral components of the fatigue of respiratory muscles in inspiratory resistive load in cats. Sechenov Physiological Journal of the US.S.R. 76: 658-666. 5. Aubier, M., D. Murciano, Y. Lecoeguic N. Viires and R. Pariente (1985). Bilateral phrenic stimulations: A simple technique to assess diaphragmatic fatigue in humans. J AppL Physiol. 58: 58-64. 6. Aubier M., G. Farkas, S. De Troyer, R. Mozes and C. Roussos (1981). Detection of diaphragmatic fatigue in man by phrenic stimulation. J AppL Physiol. 50: 538-544. 7. Bazzy, A.R. and D.F. Donnelly (1993). Diaphragmatic failure during loaded breathing: role of neuromuscular transmission. J AppL Physiol. 74: 16791683. 8.  Bazzy A.R. and G.G. Haddad (1984).  Diaphragmatic fatigue in  81 unanesthetized adult sheep. J AppL PhysioL 57: 182-190. 9. Bigland-Ritchie, B. (1987). Respiratory Muscles and their Neuromotor Control. New York, Liss, 1987, pp. 379-390. 10. Borkowksi, G.L., PJ. Danneman, G.B. Russell, and C.M. Lang (1990). An evaluation of three intravenous anesthetic regimens in New Zealand Rabbits. Lab. Animal ScL 40: 270-276. 11. Brancatisano, A., Kelly, S.M., Tully, A., Loring, S.H. and L.A. Engel (1989). Postural changes in spontaneous and evoked regional dipahragmatic activity in dogs. J. AppL PhysioL 66: 1699-1705. 12. Cairns A.C. and R.D. Road (1993). Phrenic motor axon firing rates during brief inspiratory resistive loads. Am. Rev. Resp. Dis. 147: A702. 13. Ferguson G.T., C.G. Irvin, and R.M. Cherniack (1990). Relationship of diaphragm glycogen, lactate, and function to respiratory failure. Am. Rev. Resp. Dis. 141: 926-932. 14. Fregosi, R.F., S.L. Knuth, D.K. Ward and D. Bartlett, Jr. (1987). Hypoxia inhibits abdominal expiratory nerve activity. J AppL PhysioL 63: 221-220. 15. Gauthier, H. (1973). Respiratory responses of the anesthetized rabbit to vagotomy and thoracic dorsal rhizotomy. Respir PhysioL 17: 238-247. 16. Grassino, A.E., Whitelaw, W.A. and J. Milic-Emili (1976). Influence of lung volume on electrode position on electromyography of the diaphragm. J AppL PhysioL 40: 97 1-975. 17. Hicks, A., J. Fenton, Garner, S. and A. J. McComas (1989). M wave  82 potentiation during and after muscle activity. J. Appi. Physiol. 66: 26061610. 18. Honda, Y. (1968). Ventilatory response to C02 during hypoxia and hyperoxia in awake and anesthetized rabbits. Respir PhysioL 5:279-287. 19. Juan, G., P. Calverley, C. Talamo, J. Schnader and C. Roussos (1984). Effect of carbon dioxide on diaphragmatic function in human beings. N. Eng. J Med. 310: 874-9. 20. Kim, M.J. S. Druz and J.T. Sharp (1985).  Effect of musce length on  electromyogram in a canine diaphragm strip prearation. J. App. PhysioL 58: 1602-1607. 21. Mayock, D.E., T.A. Standaert and D.E. Woodrum (1992).  Effect of  inspiratory resistive loaded breathing and hypoxemia on diaphragmatic function in the piglet. J AppL PhysioL 73: 1888-1893.  22. Mayock, D.E., R.J. Badura, J.F. Watchko, T.A. Standaert and D.E. Woodrum (1987). Response to resistive loading in the newborn piglet. Pediatr. Res. 21: 121-125. 23. Mayock D.E., T.A. Standaert, T.D. Murphy and D.E. Woodrum (1991). Diaphragmatic force and substrate response to resistive loaded breathing in the piglet. .L AppL Physiol. 70: 70-76. 24. Merton, P.A. Voluntary strength and fatigue (1954). J Physiol. 123: 553-564. 25. McKenzie, D.K., B. Bigland-Ritchie, R.B. Gorman and S.C. Gandevia (1992). Central and peripheral fatigue of human diaphragm and limb muscles as  83 assessed by twitch interpolation. J Physiol. 454: 643-656. 26. Neubauer, J.A., J.E. Melton and N.H. Edelman (1990). Modulation of respiration during brain hypoxia. J AppL PhysioL 68: 441-451. 27. Oliven, A., S. Lohda, M.E. Adams, B. Simhai and S.G. Kelsen (1988). Effect of fatiguing resistive loads on the level and pattern of respiratory activity in awake goats. Respir PhysioL 73: 3 11-324. 28. Petrozzino, JJ., A.T. Scardella, T.V. Santiago and N.H. Edelman (1992). Dichioroacetate blocks endogenous opioid effects during inspiratory flowresistive loading. J AppL Physiol. 72: 590-596. 29. Road, J.D., S. Osborne and Y. Wakai (1993). Delayed post-stimulus decrease of phrenic motoneuron output produced by phrenic nerve afferent stimulation. J AppL Physiol. 74: 68-72. 30. Schnader, J.Y., G. Juan, S. Howell, R. Fitzgerald, C. Roussos (1985). Arterial CO 2 partial pressure affects diaphragmatic function. J AppL Physiol. 58: 823-829. 31. Stephen, J.A. and A. Taylor (1972). Fatigue of maintained voluntary muscle contractions in man. J Physiol. 220: 1-18. 32. Thomas, C.K., J.J. Woods and B. Bigland-Ritchie (1985).  Impulse  propagation and muscle activation during long maximal voluntary contractions. .L AppL PhysioL 67: 183 5-1842. 33. Watchko, J.F., T.A. Standaert, and D.E. Woodrum (1987). Diaphragmatic function during hypercapnia: neonatal and developmental aspects. J AppL  84 PhysioL 62: 768-775. 34. Watchko, J.F., T.A. Standaert, D.E. Mayock, G. Twiggs and D.E. Woodrum (1988). Ventilatory failure during loaded breathing: the role of central neural drive. J AppL Physiol. 65: 249-255. 35. White, P.F., Way, W.L., and A.J. Trevor (1982). Ketamine-Its pharmacology and therapeutic uses. Anesthesiology 56: 119-136. 36. Yan, S, T. Similowski, A.P. Gauthier, P.T. Macklem, and F. Bellemare (1992). Effect of fatigue on diaphragmatic function at different lung volumes. J. AppL Physiot 72: 1064-1067.  85 APPENDIX I  Flow Resistances Reported in Animal Studies of Inspiratory Resistive Loading.  This thesis: Rabbit (measured at flow rate of 1 Litre/minute) Load 1:  900 cm H20/L/sec  Target Pao  =  8± 2 cm H20 (1.5 X Pdi  Load 2: 3780 cm H20/L/sec  Target Pao  =  27± 3 cm H20 (5.5 X Pdi  QI  Load 3: 6900 cm H20/L/sec  Target Pao  =  39± 3 cm H20 (8  QI  Load 4: 7500 cm H20/L/sec  Target Pao  =  53± 4 cm H20 (11 X Pdi Qi  X Pdi  QI  Aldrich (1985, 1987):  Rabbit  Mayock et aL (1987, 1991):  Piglet 0.65 cm H20/ml/sec  Watchko et aL (1988):  Infant monkey 0.24-0.42 cm H20/ml/sec  Bazzy and Haddad (1984)  Sheep  Target Pao = 27 cm H20  >  150 cm H20/L/sec  86  III: Ventilation During Prolonged Hypercapnia in the Anesthetized Rabbit  Introduction  As indicated in the preceding chapter, when a flow-resistive inspiratory load is added to the airway of the anesthetized rabbit, minute ventilation decreases immediately.  If the applied resistance is maintained, inspiratory efforts will  progressively increase. This response is reflected in the gradual rise in peak inspiratory pressures and tidal volume which result in increased ventilation over the next few minutes. In its final steady state, minute ventilation is lower than the level prior to loaded breathing in the anesthetized animal.  Mechanisms that offset added mechanical loads placed on the respiratory system are termed load compensation. In the anesthetized cat (Shannon and Zechman, 1972) and rabbit (Sant’Ambrogio and Widdicombe, 1965), neural reflexes which increase inspiratory alpha motoneuron activity are evoked during the first inspiratory effort after inspiratory resistive loading.  These immediate load  compensatory reflexes are mediated via thoracic dorsal roots and vagal afferents and are responsible for the first breath response to inspiratory resistive loads. Studies in which blood gases are kept constant by cross corporeal or extracorporeal gas exchange demonstrate that the transient progressive increase in inspiratory effort after the first inspiratory resistive loaded breath is due to the  87 early changes in blood gas tensions alone (Orthner and Yamamoto, 1974; Bruce et aL, 1974).  Thus, initially, load compensation is neurally mediated and  subsequently, load compensation involves the response of all respiratory muscle groups to changes in blood gas tensions.  Few studies have examined the effect of prolonged (i.e. longer than one hour) inspiratory resistive loading on ventilation in anesthetized animals.  In the  previous chapter, we have shown that during prolonged inspiratory resistive loading, a constant level of minute ventilation is established within 10-15 minutes of the onset of the load. During the following 3 hours, despite rising PaCO2, minute ventilation is maintained, as is the frequency (Bf) and intensity of motoneuron output (ENGdi) to the diaphragm under both Load 1 and Load 2. Hence, load compensation is incomplete and the respiratory muscles do not increase their work load in spite of progressive hypercapnic ventilatory failure. The constant level of diaphragm activation in our preparation despite a rising C02 stimulus could result from either competing inhibitory and excitatory feedback mechanisms to the respiratory centre intrinsic to loaded breathing or from the central effects of prolonged exposure to severe hypercapnia or both.  There are few published reports on central effects of prolonged exposure to hypercapnia without hypoxia or disease. A handful of studies have examined the effects on ventilation of prolonged (i.e. greater than 30 minutes) inhalation of  88 C02 (for a review see Dempsey and Forster, 1982).  Most of these reports  considered exposure of humans to relatively low levels of C02 (up to 5% inspired C02) for several days. In these studies, a slight reduction in ventilation between 3-10 days of exposure to increased levels of carbon dioxide is observed as there is a slightly reduced ventilatory response to inspired C02. Few reports have examined the effect of the first few hours of exposure to the very high levels of C02 that would produce elevated levels of arterial PC0 2 of the magnitude observed during the prolonged inspiratory resistive loading used in our preparation.  Bleich, Berkman and Schwartz (1964) measured the CSF bicarbonate response to sustained hypercapnia in awake dogs exposed to 12% inspired C02. In the first 30 minutes of exposure to elevated FICO2, PaCO2 rose from 36 to 80 mm Hg and continued to increase for 3 hours. Furthermore, both CSF and plasma bicarbonate responses to severe hypercapnia were rapid. For example, CSF bicarbonate increased by 2 mEq/L within the first 30 minutes of hypercapnic exposure and continued to increase a further 5 mEq/L over the next three hours. Although ventilation was not monitored in their study, it is likely that the decrease in hydrogen ion drive to central chemoreceptors due to bicarbonate buffering would cause a drop in ventilation during this period. Whether such rapid ventilatory acclimation to high levels of C02 could account for the progressive hypercapnic ventilatory failure in our preliminary study (Chapter II)  89 is unknown. Accordingly, the following study was designed to determine whether any significant time-dependent changes in ventilation would occur in our preparation solely as a result of prolonged hypercapnia. If so, then the constant level of drive to the diaphragm observed with prolonged inspiratory resistive loading, could be explained at least partially, by the effects of prolonged hypercapnia alone on central respiratory control.  Methods  Animals. Six male, New Zealand White rabbits weighing 3-3.5 kg were studied.  The source, care, surgical preparation and anesthetic regime utilized were identical to those described in Chapter II except for the following modifications. Following tracheostomy and carotid artery cannulation, the phrenic nerve was left intact in four rabbits. No recording of the whole nerve potentials was made in these animals.  In two rabbits, the whole phrenic nerve (rather than the  rootlet) was exposed, secured to bipolar platinum fine wire stimulating electrodes and bathed in mineral oil to record M-wave amplitudes, as described previously.  Protocol  After a 20 minute period of stabilization following surgery, baseline measurements of respiratory variables and blood gases were made in rabbits  90 breathing room air with supplemental oxygen. The rabbits were then exposed to an inspired gas mixture of 10% C02, 50% 02 balanced N2 (precision gas, Linde-Union Carbide) via a meterological bag in series with the inspiratory port for a period of 3 hours. The expiratory port of the two way non-rebreathing valve was exposed to room air. All measurements were obtained at 10, 30, 60, 120 and 180 minutes following hyperoxic hypercarbia and after 30 minutes of recovery from this gas mixture.  Arterial blood (0.3 ml) was sampled in  heparinized syringes and gases and pH measurements made within one minute of sampling.  Pressures generated by the respiratory system (Pao, Pdi), peak  moving average of the costal diaphragm electromyogram (EMGdi), airflow, tidal volume and arterial blood pressure were recorded continuously. The mean value for tidal volume, breathing frequency, peak negative inspiratory pressures (Pao, Pdi), TI, TE, VT/TI, TI/rrOT, EMGdi, and minute activity were calculated hourly during a representative 60 second interval. To compare the relative activation of the diaphragm during hypercarbia and inspiratory resistive breathing, peak EMGdi multiplied by breathing frequency (minute activity) was calculated. The durations of inspiration (TI) and expiration (TE) were determined from the EMOdi recordings by the procedure described in Chapter II.  Data Analysis. The initial response to hypercapnia was analyzed by comparing  measures taken during baseline and after 30 minutes from the onset of hypercarbia. A paired student’s t-test analysis was adopted. Comparison of  91 measurements taken during C02 exposure was made by Friedman ANOVA (repeated measures). A Wilcoxon signed rank test was used to determine which values differed with time during the high C02 exposure (Systat 5.02). Statistical significance was defined as P <0.05. All values in the text represent mean ± S.E.M.  Results  The results are described separately for the initial 30 minute exposure to hypercarbia and the prolonged exposure to 3 hours of hypercarbia.  Steady state response to hypercapnia: 30 minutes of exposure to hypercarbia.  Arterial blood gases and ventilatory variables at baseline and during the first 30 minutes of exposure to 10% inspired C02 are summarized in Table Ill. There was a two fold increase in minute ventilation in response to FICO2=0.10. This ventilatory response is within the range previously reported for anesthetized rabbits (Richardson and Widdcombe, 1969; Gautier, 1973). The rise in minute ventilation was due to an approximate two fold increase in VT and a 20% increase in breathing frequency (Bf). Respiratory times and related variables are summarized in Table IV. The increase in Bf was due to a reduction in TE in all the rabbits exposed to 10% inspired CO . In three of the six rabbits tested, Ti 2  92 decreased in response to hypercapnia as well. The increase in mean inspiratory flow rate (VT/Ti) with C02 breathing was mainly as a result of the increase in VT. On average, there was no significant change in duty cycle (TI/TroT) with hypercapnia. Hypercapnia resulted in increased central drive to the diaphragm as reflected by a two fold rise in the activity of the diaphragm (EMGdi) and a two fold rise in pressure swings generated by this muscle (Pdi) (Table III).  93 Table III: Mean values (± S.E.M.) for arterial blood gases and ventilatory variables during the initial 30 minutes of exposure to 10% inspired C02 in hyperoxic anesthetized rabbits (n = 6).  Duration (minutes) Baseline  10  30  47  65*  67*  (1)  (2)  (3)  Pa0 2 (mm Hg)  158 (17)  227 (14)  236 (14)  pH  7.35 (0.02)  7.24* (0.01)  7.24* (0.01)  HC0 3 (mM/L)  26 (1)  28* (1)  28* (1)  100 (0)  219* (20) 34*  236* (23) 37*  (3) 53*  (3) 53*  (3)  (3)  2 PaCO  VE  (%)  VT (ml)  19 (1)  Bf (min’)  44 (4)  EMGdi (%)  100 (0)  201* (29)  212* (21)  Pdi (cm H20)  5.2 (0.4)  8.8* (1.0)  8.5* (0.8)  *p <0.05 compared to baseline value.  94 Table W: Mean values (± S.E.M.) for respiratory times, mean inspiratory flow rate (VT/TI) and duty cycle (TI/Vr0T) in anesthetized rabbits (n = 6) exposed to 10% inspired C02 for the initial 30 minutes.  Duration (minutes) Baseline  10  30  Ti (sec)  0.6 (0.1)  0.5 (0.1)  0.5 (0.1)  TE (sec)  0.9 (0.1)  0.7*t (0.1)  (0.1)  TTOT (sec)  1.4 (0.2)  1.1* (0.1)  1.1* (0.1)  TI/TroT  0.40 (0.02)  0.42 (0.02)  VT/TI (mi/sec)  36 (3)  0.42 (0.01) 75* (8)  80* (8)  *p < 0.05 compared to baseline value. t p < 0.05 compared to other values obtained during 10% C02 exposure.  95 Ventilatoiy response to prolonged hypercapnia: 3 hours of exposure to hypercarbia.  Figure 15 shows ventilation in response to prolonged inhalation of 10% inspired C02 in anesthetized hyperoxic rabbits and the corresponding changes in PaCO2. PaCO2 rose relative to the steady state established within the first half hour of exposure to high C02. By 120 minutes of exposure to hypercarbia, minute ventilation fell as a result of a drop in Bf from 53 breaths/mm at 30 minutes to 44 breaths/mm  at 120 minutes (see Table V). This decrease in Bf resulted in  an average frequency of breathing equivalent to the baseline prior to hypercarbic exposure. In four of the six rabbits tested there was a decrease in VT from 40± 4 ml to 30± 3 ml during this period as well. In contrast, the elevated VT observed early in hypercapnia was either maintained or increased in the other two rabbits. Therefore, the average VT response of the group to prolonged hypercapnia was not significantly different over time (Table V).  Respiratory times increased variably by 120 minutes of hypercarbia. In two rabbits both TI and TE increased. In the remaining four rabbits, there was only an increase in TE with sustained hypercarbia. On average, duty cycle remained constant throughout sustained hypercarbia (Table VI). A drop in ventilatory drive late in hypercarbia was reflected by decreased mean inspiratory flow rate (VT/TI).  The average activity (EMGdi) and pressure generated by the  diaphragm (Pdi) did not change significantly during the the first 2 hours of  96 exposure to severe hypercarbia (Table V).  2 However, by 3 hours of CO  exposure, there was a significant drop in both EMGdi and Pdi.  We measured no change in the evoked compound action potential of the 2 in the two rabbits diaphragm during prolonged exposure to 10% inspired CO tested. The profile of M-waves were superimposable and the mean peak to peak amplitudes of the diaphragm M waves ranged from 5.2 ± 0.3 mV to 5.4 during prolonged hypercarbia.  i  0.6 mV  97  Figure 15:  2 levels during prolonged exposure to Top Panel: Arterial PCO FICO2=O.1O in hyperoxic anethetized rabbits (n=6).  Bottom Panel: Minute ventilation during prolonged exposure to FICO2=0.10 in hyperoxic anesthetized rabbits (n=6).  2 exposure. Arrows represent the onset of CO *  t  Significantly different from baseline (B), p < 0.5. =  Significantly different from 30 minute value, p <0.05.  c8  100 *1• 90 80 I  E E  /  60I  o  I 40 / 0  I//I  B  I  I  I  I  1030 60 120 Time (minutes)  I  I  180  * *  600  lit  500 400 C  9>  .  I’  300  200  iooØ B  II 1030 60 120 Time (minutes)  180  I  99 Table V: Arterial blood gases and changes in ventilatory variables produced by 3 hours of exposure to 10% inspired C02 in anesthetized rabbits (n = 6).  Duration (minutes) 120  10  30  60  47 (1)  65* (2)  67* (3)  77*t (1)  80*t (2)  89*t (3.0)  Pa0 2 (mm Hg)  158 (17)  236 (14)  245 (17)  238 (21)  243 (21)  pH  7.35 (0.02)  227 (14) 7.24* (0.01)  7.24* (0.01)  7.21*t (0.01)  7.17*t (0.01)  7.14*t (0.01)  HC0 3 (mM/L)  26 (1)  28* (1)  28* (1)  30* (1)  29* (1)  30* (0.5)  100 (0)  219* (20) 34*  236* (23) 37*  206* (23)  164*t (14)  151*t (16)  (3) 53* (3)  36* (3) 47* (4)  31* (2)  29* (2)  44*j.  42*t (4)  168* (15) 75*  Baseline  2 PaCO  VE  (%)  VT (ml)  19 (1)  Bf ) 1 (min  44 (4)  (3) 53* (3)  EMGdi (%)  100 (0)  201* (29)  212* (21)  193* (21)  Pdi (cm H20)  5.2 (0.4)  8.8* (1.0)  8.5* (0.8)  8.3* (0.8)  (4)  (0.8)  *p <0.05 compared to baseline value. tp <0.05 compared to other values obtained during 10% C02 exposure.  155*t (15) 6.9*t (0.7)  100 Table VI: Mean values (± S.E.) for respiratory times, mean inspiratory flow rate (VT/TI) and duty cycle (TI/TroT) in anesthetized rabbits (n = 6) exposed to 10% inspired C02 for 3 hours.  Duration (minutes  Baseline  10  30  60  120  180  Ti (sec)  0.6 (0.1)  0.5 (0.0)  0.5 (0.0)  0.5 (0.0)  0.6 (0.0)  0.6 (0.1)  TE (sec)  0.9 (0.1)  0.7*t (0.1)  0.7*t (0.1)  0.8 (0.1)  0.8 (0.1)  0.8 (0.1)  rrOT (sec)  1.4 (0.2)  1.1* (0.1)  1.1* (0.1)  1.3 (0.1)  1.44 (0.1)  1.4 (0.1)  TI/TroT  0.40 (0.02)  0.42 (0.01)  0.42 (0.02)  0.39 (0.02)  0.41 (0.01)  0.43 (0.02)  VT/Ti  36 (3)  75*  80* (8)  72* (6)  55*j.  50*t (5)  (mi/sec)  (8)  (5)  *p <0.05 compared to baseline value. <0.05 compared to other values obtained during 10% C02 exposure. <0.05 compared to values obtained during the first hour of 10% C02 exposure.  101 The data obtained during prolonged exposure to 10% C02 alone is compared to the degree of hypercapnia associated with prolonged inspiratory resistive loading (load 2, Chapter II) in Figures 16-18. Both stimuli result in similar levels of PaCO2, however there are distinct differences in breathing pattern in response to the two conditions. During loaded breathing the rabbits breathe slowly and  , breathing is deep and rapid. 2 shallowly whereas during inhalation of C0  102  Figure 16:  2 levels (top panel) and the Comparison of arterial PCO corresponding levels of minute activity (EMGdi X Bf) of the diaphragm (bottom panel) in rabbits exposed to prolonged hypercarbia (n = 6) and rabbits exposed to prolonged inspiratory resistive loading (n=7; data from Load 2, Chapter II).  *  =  Significantly different between the two groups, p < 0.05).  jO3  110  -  100Q)  I  I  ILoad 2  K\\NFICO2=0.10  90  E80 E  } +  0 C  °60  0  50 4Q-___ B  2  1  3  Time (hours) 120 >  *  +  .E  E  I ±  +  60  D<40 L  +  r  a .9  20  C  0  ———  B  1 2 Time (hours)  3  104  Figure 17.  Comparison of the peak electromyographic activity of the diaphragm (EMGdi) (top panel) and breathing frequency (Bf) (bottom panel) in rabbits exposed to prolonged hypercarbia (n = 6) and rabbits exposed to prolonged inspiratory resistive loading (n = 7; data from Load 2, Chapter II).  *  =  Significantly different between the two groups, p < 0.05).  ‘cc 400  I  ILoad 2  RNxNFICO2=0.10  + +  300 a) cri a  200 -D  CD  w100  -  0B  2  3  Time (hours)  60  a)  1  50  D  E Cl)  -c o a)  40 30 20  4-  QD  10  0  I  ----  B  1  2  Time (hours)  3  106  Figure 18.  2 levels (top panel) and the Comparison of arterial Pco corresponding levels of minute ventilation (bottom panel) in rabbits exposed to prolonged hypercarbia (n = 6) and rabbits exposed to prolonged inspiratory resistive loading (n=7; data from Load 2, Chapter II).  *  =  Significantly different between the two groups, p <0.05).  D’T 110  -  I  100-  ILoad 2  k\NFICO2=O.l 0  ++  90  E E  80-  (N  70  0  C—)  60 50 40-  -  B  -Th -  c  1  2  3  Time (hours) 650 600 550 500 450 400  Ii  350 300 250  +  200 150 100 50 0  -  —  B  —  —  —  1  —  2  Time (hours)  3  108 Discussion  This study demonstrates a significant time-dependent change in the regulation of ventilation during prolonged exposure to severe hypercapnia in the anesthetized hyperoxic rabbit. Specifically, there is a drop in breathing frequency (Bf) within two hours of exposure to severe hypercapnia followed by a reduction in activation of the diaphragm (EMGdi) by the third hour of C02 exposure (Table V).  Further work needs to be done to ascertain the mechanism(s)  underlying this phenomenon. We have shown that this effect is mediated initially through its action on central rhythm and subsequently by a decrease in the central motor output to the diaphragm per breath (EMGdi).  The top panels in figures 16 and 18 show the degree of hypercapnia associated with exposure to C02 alone compared with the degree of hypercapnia associated with prolonged inspiratory resistive loading. Although both stimuli result in similar levels of PaCO2, there are distinct differences in breathing pattern in response to the two conditions within the first hour of hypercapnia. Whereas Bf is elevated during C02 breathing, it is decreased during loaded breathing as a result of neurally mediated load compensatory mechanisms (Figure 17). Therefore, there is a greater level of diaphragmatic minute activity during C02 breathing compared to loaded breathing at 1 hour (Figure 16). In contrast, there is no significant difference between activation of the diaphragm per breath  109 (EMGdi) under the two conditions at similar levels of PaCO2 at 1 hour (Figure 17).  Consequently, the VT generated during C02 breathing (37± 3 ml) is  substantially greater compared to the volume generated against the flow resistance imposed by the inspiratory load (15± 1 ml). Hence, the increase in the mechanical time constant of the respiratory system during external inspiratory loading and the load compensating mechanisms operating during the first hour of loaded breathing result in a level of minute ventilation that is significantly less than predicted by the C02 stimulus alone (Figure 18).  While there are differences in breathing pattern during the first hour of exposure to C02 alone compared to loaded breathing, the drop in Bf with prolonged exposure (2 and 3 hours) to C02 is analogous to the initial load compensatory response to loaded breathing (Figures 16-17). It can be hypothesized that the reduction in Bf functions as an effective strategy to reduce the energetic cost of breathing and avert or forestall the development of task failure during conditions of increased ventilatory demand. Evoked compound action potentials of the diaphragm did not change significantly during prolonged exposure to C02 suggesting that there is no failure in neuromuscular transmission during this task. However, the fact that activation of the diaphragm (EMGdi) and hence the ability to generate pressure (Pdi) was found to have decreased significantly by the third hour of C02 breathing shows that prolonged hyperpnea associated with this ventilatory stimulus can result in a decrease in central motor output to the  110 diaphragm.  Reduction in Bf within two hours of C02 exposure may be due to 1) the effects of prolonged anesthesia; 2) a generalized depression of the CNS due to severe hypercapnia;  or 3) the central  effects of prolonged hypercapnia on  chemoreceptor activity. The data obtained in the control group during our initial studies (Chapter II) show that there is no change in Bf as a result of prolonged anesthesia (40± 5 at baseline versus 37± 4 after three hours of anesthesia). Therefore, it is unlikely that the reduction in Bf during prolonged C02 exposure is due to prolonged anesthesia.  Rising concentrations of C02 in the arterial blood have different effects on various anatomical structures within the brain.  According to Wyke and  Woodbury (see Wyke, 1963), there are three phases associated with the sequence of neurophysiological responses to increasing amounts of C02. The initial phase involves direct depression of cortical synaptic activity only and is achieved by inhalation of C02 in concentrations between 3.5 and 7 percent. The next phase involves a generalized reticulo-cortical activation which overcomes the initial direct cortical depression.  This stimulatory phase is increasingly evoked by  inhalation of C02 in concentrations between 5-20%  .  The final phase is  associated with narcosis or anesthesia, which leads to generalized depression of reticulo-cortical activity where inhaled C02 concentrations exceed 25 percent.  111 Therefore, it is unlikely that the decrease in breathing frequency observed after two hours of inhalation of 10% C02 is due to a generalized depressive effect of C02 on the CNS. Furthermore, it has been shown in the past that central inspiratory activity as indicated by phrenic motor discharge is tolerant of extensive hypercapnia in rabbits (up to PACO2 of 200 mm Hg; Kobayasi and Murata, 1979).  It has been demonstrated that in mammals central chemoreceptors situated beneath the ventrolateral medullary surface (Mitchell et aL, 1963; Schlaefke et aL, 1970) stimulate or inhibit ventilation in response to changes in extracellular fluid pCO2/[H9 or [WI alone (Bledsoe and Hornbein, 1981; Dempsey and Forster, 1982).  Recent reports suggest that other discrete areas within the  brainstem (in the vicinity of the nucleus tractus solitarius and locus coeruleus) may be involved in driving ventilation in response to this chemical stimulus as well (Coates, Li and Nattie, 1993; see Severinghaus, 1993 as well). mechanisms  by which  extracellular pCO2/[W]  is  sensed  by  The  central  chemoreceptors is not known.  It seems likely that the decrease in Bf following two hours of sustained hypercapnia is due to a depressed H stimulus at the central chemoreceptor sites. Rapid increases in cerebrospinal fluid bicarbonate concentration and brain tissue bicarbonate concentration are known to occur in response to sustained  112 hypercarbia, in several species including in the ketamine-xylazine anesthetized rabbit (Bleich et aL, 1964, Ponten, 1966; Messeter and Siesjo, 1971; Hasan and Kazemi, 1976; Ahmad, Berndt and Loeschke, 1976; Nattie and Giddings, 1988). The mechanism which permits bicarbonate ion [HCO ] to accumulate in the 3 CSF during hypercapnia is not known. However, since the CSF is an ionic solution low in protein and poorly buffered this increase must stem from either blood or brain tissue origin.  Ionic regulation of [HCO;] in the cerebral  extracellular fluids (CSF and brain interstitial fluid) during prolonged and especially severe hypercapnia is poorly understood (for a review see Fend, 1986). According to Nattie (1980), the time course of the continuous increase in CSF ] in response to severe sustained hypercapnia suggests 3 phases of 3 [HC0 regulation: 1) an immediate 5-20 minute response, 2) a slower response that plateaus over 1-3 hours and 3) an increase over hours to days which is concomitant with the increase in blood [HCO ]. It is assumed the third phase 3 reflects the slow exchange of ions between blood and CSF.  Exposure to high levels of C02 during the second phase, as described by Nattie (1980), can also result in alterations in brain metabolic byproducts which affect chemoreceptor activity directly or indirectly via other central respiratory neurons. These changes include a down regulation in brain excitatory amino acids, glutamate and aspartate, and increases in the inhibitory amino acid, GABA (Kazemi and Hoop, 1991). Irrespective of mechanisms that could potentially  113 mediate a reduction in chemoreception during the first few hours of severe hypercapnia, the question remains whether such change(s) are sufficient to produce a significant drop in ventilation. In a study on rats, Nattie (1980) suggests that alteration in central chemoreceptor activity during this period may result in decreased ventilation.  In conclusion, prolonged (greater than 1 hour) and severe hypercapnia alone can result in a decrease in minute ventilation in the anesthetized hyperoxic rabbit preparation. It seems that the levels of hypercapnia associated with prolonged moderate inspiratory resistive loading are sufficient to provide an inhibitory input to the central respiratory controller resulting in a suppressed level of drive to the diaphragm. Although activation of the diaphragm (EMGdi) is maintained within the first two hours of C02 exposure, Bf is reduced [rroT increases] and duty cycle decreases. Consequently, the breathing pattern associated with prolonged hypercapnia can potentially protect the respiratory muscles from overload. This mechanism may prevent excessive activation of the respiratory muscles when loading is severe.  Therefore, we infer that although load compensating  mechanisms are initiated at the onset and early in loaded breathing, they compete with changes in breathing pattern that result from the effects of prolonged hypercapnia that are load decompensating.  Surprisingly, this  decompensation develops within 1-2 hours of loading and is progressive.  114 References  1. Ahmad, H.R., J. Berndth and H. H. Loeschcke (1976). Bicarbonate exchange . 2 between blood, brain extracellualar fluid and brain cells at maintained PCO In: Acid Base Homoeostasis of the Brain Extracellular Fluid and The Respirator,’ Control System. H.H. Loeschcke (Ed.). Georg Thieme, Stuttgart, pp. 19-27. 2. Bledsoe, W.W. and T. Hornbein (1981). Central chemosensors and the regulation of their chemical environment. In: Regulation of Breathing, T. Hornbien (ed.). Lung Biology in Health and Disesase. New York, Dekker, pp. 347-428. 3. Bleich, H.L., P.M. Berkman and W.B. Schwartz (1964). The response of cerebrospinal fluid composition to sustained hypercapnia. J Clin. Invest. 43: 11-16. 4. Bruce, E.N., J.D. Smith and F.S. Grodins (1974). Chemical and reflex drives to breathing during resistance loading in cats.  ..  AppL PhysioL 37: 176-182.  5. Coates, E.L., H. Li and E.E. Nattie (1993). Widespread sites of brain stem ventilatory chemoreceptors. J AppL PhysioL 75: 5-14. 6. Dempsey J.A. and H.V. Forster (1982). Mediation of ventilatory adaptations. Physiological Reviews 62: 262-346. 7. Fend, V. (1986). Acid-base balance in cerebral fluids. In: Handbook of Physiology- The Respirator,’ System. Williams & Wilkins, Baltimore, MD, Vol  115 II, Part 1. Cherniack, N.S. and Widdicombe, J.G. pp. 115-139. 8. Hasan, F.M. and H. Kazemi (1976). Dual contribution theory of regulation of CSF HCO 3 in respiratory acidosis. J AppL PhysioL 40: 559-567. 9. Kazemi, H. and B. Hoop (1991). Glutamic acid and alpha-amino butyric acid neurotransmitters in central control of breathing. J AppL PhysioL 70 1-7. 10. Kobayasi, S. and K. Murata (1979). Phrenic activity during severe hypercapnia in vagotomized rabbits. J AppL PhysioL 47: 9 1-97. 11. Messter K. and B.K. Siesjo (1971). Regulation of the CSF pH in acute and sustained respiratory acidosis. Ada. PhysioL Scand. 83: 21-30. 12. Mitchell, R.A., H.H. Loeschcke, J.W. Severinghaus, B.W. Richardson and W. H. Maission (1963). Regions of respiratory chemosensitivity on the surface of the medulla. Ann. NY Acad. Sci. 109: 661-681. 13. Nattie. E.E. (1980). Brain and cerebrospinal fluid ionic composition and ventilation in acute hypercpania. Respiz PhysioL 40: 309-322. 14. Nattie, E.E. and B. Giddings (1988). Effects of amiloride and diethyl pyrocarbonate on CSF HC0 3 and ventilation in hypercapnia. J. AppL Physiol. 65: 242-248. 15. Orthner F.H. and W.S. Yamamoto (1974). Transient respiratory response to mechanical loads at fixed blood gas levels in rats. J AppL PhysioL 36: 280287. 16. Ponten, U. (1966). Consecutive acid-base changes in blood, brain tissue and cerebrospinal fluid during respiratory acidosis and alkalosis. Acta. NeuroL  116 Scand. 42: 455-471. 17. Sant’Ambrogio, G. and J.G. Widdicombe (1965). Respiratory reflexes acting on the diaphragm and inspiratory intercostal muscle of the rabbit. J PhysioL 180: 776-779. 18. Severinghaus, J.W. (1993). Invited editorial on “widespread sites of brain stem ventilatory chemoreceptors”. J AppL PhysioL 75: 3-4. 19. Schlefke, M. E., W.R. See and H.H Loeschcke (1970). Ventilatory response to alterations of hydrogen ion concentration in small areas of the ventral medullary surface.  Respir. PhysioL 10: 198-212.  20. Shannon, R. and F.W. Zechman (1972). The reflex and mechanical response of the inspiratory muscles to an increased airflow resistance. Respir. PhysioL 16: 5 1-79. 21. Wyke, B. (1963). The neurological basis of pH effects on brain function. In: Brain Function and Metabolic Disorders. Butterworth & Co., London, pp. 158-165.  117 W: Respiratory Muscle Activity and Task Failure During Severe Inspiratory  Resistive Loading.  Introduction  Fatigue of the diaphragm has been implicated as a contributing factor in ventilatory pump failure induced by severe inspiratory resistive loading (Aldrich 1985, 1987, 1988; Alexadrovna and Isaev, 1990; Bazzy and Haddad, 1984; Mayock, 1987, 1991). We have shown that neural activation, neuromuscular transmission and activity of the diaphragm are maintained in anesthetized rabbits under severe inspiratory resistive loads even when inspiratory pressure generation is reduced i.e. task failure occurs (Chapter II). Since we used peak inspiratory airway pressure (Pao), which reflects total respiratory muscle output as an index of muscle force, the specific contribution of respiratory muscles other than the diaphragm to force failure during extreme inspiratory resistive loading (Load 4) could not be assessed. Therefore, task failure may be a consequence of reduced function of extradiaphragmatic muscles.  In this study, we monitored the electromyographic activity of the other major inspiratory muscle, the parasternal intercostal (EMGps), because loss of inspiratory pressure may be due to failure of activation of this muscle. Alternatively, the expiratory muscles are thought to provide an important assist  118 to inspiration and hence failure of their activation may contribute to task failure during inspiratory resistive loading. Therefore, we measured electromyographic activity in one of the most active expiratory muscles, the transversus abdominis (EMGta) along with the inspiratory activity of the diaphragm (EMGdi) and the parasternal intercostal (EMGps) under extreme inspiratory loading until respiratory arrest.  Methods  Animals. Six New Zealand white rabbits were obtained from Geo-Bat Rabbitries  (Abbotsford, B.C.) and cared for according to the principles outlined by the Canadian Council for Animal Care at the Animal Resource Unit facility at the University Hospital (U.B.C). Experimental protocols received ethics approval from the University of British Columbia Animal Care Committee.  Preparation.  The rabbits (mean body weight 3.4 kg, range = 3.2-4 kg) were  anesthetized with i.m. injection of ketamine (Ketavet, Parke-Davis, 30 mg/kg) and a sedative, xylazine (Rompun, Bayer, 7 mg/kg). Anesthesia was maintained throughout the studies by supplementing half the initial dose every 30-40 minutes.  Rectal temperature was continuously monitored and maintained  between 38-39°C with a heating pad. Saline was infused via a marginal ear vein to maintain blood pressure.  Inspired air was supplemented with oxygen  119 throughout the study.  Rabbits were placed in the supine position and the trachea was cannulated and connected to a heated pneumotachograph (Fleisch # 00) in series with a miniature two way non-rebreathing valve (Hans Rudolph no. 2814; 2.5 ml dead space). Pressure across the pneumotachograph was measured with a differential pressure transducer (± 2 cm H20, Validyne MP-45) and a carrier preamplifier (Gould model 13-44615-35). The carrier output was electronically integrated (Gould integrator amplifier 13-4615-70) to record tidal volume (calibrated for a range of 5-30 ml). Airway pressure (Pao) was measured at the tracheal tube using a differential pressure transducer (± 80 cm H20 Validyne, Northridge, CA). An adjustable needle valve was placed at the inspiratory port of the non rebreathing valve to apply the flow resistive load. During resistive loading, 100% oxygen was provided at the inspired port from a meterological balloon. The left carotid artery was cannulated to measure blood pressure and to sample blood for blood gas and pH analysis (Model 168 pH/blood gas analyzer, Corning Medical, Medfield, MA).  To record the diaphragm electromyogram (EMGdi), a midline upper abdominal incision was made and the uninsulated tips of two multi-stranded stainless steel fine wires (Cooner wire #AS 631) were sutured 1 cm apart into the left costa! hemi-diaphragm midway between the costal margin and central tendon. Through  120 the same incision, a second pair of wires were sutured 1 cm apart in the transversus abdominis approximately 2 cm lateral to the linea alba and parallel to the transverse orientation of the muscle fibres to record activity of the abdominal muscles (EMGta).  An air filled balloon-catheter assembly was  secured underneath the dome of the right diaphragm above the liver and attached to a differential pressure transducer (± 56 cm H20 Validyne, Northridge, CA) to record abdominal pressure (Pab). The abdominal incision was then closed with sutures and surgical staples to restore the fascia, muscle and skin layers respectively. The sternum was exposed by a midline incision along the length of the 2nd to 5th rib and the insertion of the left pectoralis muscles dissected away to expose the parasternal muscles between the 3rd and 4th ribs. To record parasternal intercostal muscle activity, a pair of fine wires were sutured 1 cm apart in the parasternal muscles parallel to the muscle fibre orientation and the incision closed as described above.  The bipolar EMG signals were amplified (Grass P5 series AC preampliers; Grass Instruments Co., Quincy, MA; band pass 100 Hz-10 kHz), whole wave rectified and the moving averages (time constants = 100, 100, 200 msec; EMGps, EMGdi & EMGta respectively) computed using a four pole active filter with a Paynter response (EMG Signal Processor, Raytech Instruments, Vancouver, Canada). The moving averages of the EMG signals were further amplified (Gould medium gain DC preamplifier model 13-4615-10) and recorded on an 8 channel chart  121 recorder (Gould model 8188-812, Cleveland, OH).  Protocol  After a 20 minute period of stabilization, baseline measurements were made. Since there is no abdominal EMG activity during eupnea, a qualitative comparison of the level of activation in the different respiratory muscles from the baseline EMG values obtained during eupnea is not possible. Therefore, we defined maximum EMG activity during exposure to high inspired CO . 2  To  determine the maximum peak moving average EMG activities, rabbits were exposed to an inspired gas mixture of 9% C02, 50% 02 balanced N2 for 10 minutes from a meterological balloon in series with the inspired port. The peak moving average of the EMGs calculated after 10 minutes of breathing this gas mixture was designated in relative units as 100% activity. Rabbits were returned to breathing oxygen supplemented room air for 20-30 minutes until blood gases and minute ventilation returned to baseline. Inspiratory resistive loading was then applied at end expiration. The inspiratory needle valve was adjusted gradually within 5 minutes to a target peak inspiratory airway pressure (Pao) of approximately -55 cm H20. This load was maintained until respiratory arrest (cessation of breathing for  >  1 minute). Arterial blood (0.3 ml) was sampled  every 10 minutes in heparinized syringes and arterial blood gases and pH measured within one minute of sampling. Expiratory abdominal pressure swings  122 (Pab), peak inspiratory airway pressure (Pao), the peak moving average activity of respiratory muscles (EMGdi, EMGps, EMGta), airflow, tidal volume and arterial blood pressure were recorded continuously and their mean value calculated every 10 minutes during a representative 60 second interval. The duration of inspiration (TI) and expiration (TE) were determined from the EMGdi recordings. Ti was defined form the EMGd1 as the time from initial rise to the point where a rapid decline was first observed. TE spanned the time from the rapid decline in the EMGdi to its next initial rise.  Data Analysis.  All variables were statistically compared by Friedman’s ANOVA, repeated measures. For variables identified as significantly different, a Wilcoxon signed rank test was used to determine which values differed with time during loaded breathing (Systat 5.02). Statistical significance was defined as P < 0.05. All values in the text represent mean ± S.E.M.  Results  Responses to hyperoxic hypercarbic gas mixture.  There was an expected increase in both tidal volume (18± 1 ml at baseline to  123 28± 1 ml during C02 breathing, p <0.001) and breathing frequency (41± 4 breaths/mm at baseline to 58± 5 breaths/minute during C02 breathing p < 0.05) in response to breathing the hyperoxic hypercarbic gas mixture (PaCO2 = 69± 3 mm Hg, PaO2 = 239± 13 mm Hg) demonstrating that anesthesia had not abolished these responses to chemical stimulation. Increases in tidal volume, breathing frequency and inspiratory muscle activity (EMGd1, EMGps) had a gradual and early onset that reached steady state within approximately 5 minutes. The onset of abdominal muscle activity (EMGta) however had a latency of 2-3 minutes. The relative increase in the EMGdi and EMGps in response to hypercarbia was 2 fold reflecting the relative increase in minute ventilation (217± 21 mI/mm/kg to 488± 45 mi/mm/kg). These results are comparable to those obtained in Chapter III during steady state after 10 minutes of exposure to hyperoxic hypercarbia (FICO2 = 0.10) in a separate group of rabbits.  Response to extreme inspirato,y resistive loading.  Inspiratory resistive loading led to hypoventilation and severe alterations in blood gases in the anesthetized rabbit (Table VII) as described in Chapter II (Load 4)  .  The average target peak inspiratory pressure (Pao) of -58 ± 4 cm  H20 in this study was maintained for 20 minutes despite severe and progressive hypercapnia and moderate hypoxia.  124 Respiratory muscle activity increased substantially in all three muscles during loaded breathing. The increase in the activity of EMGdi, EMGps, EMGta during loaded breathing was three to four fold that observed during C02 stimulated breathing (Figure 19). The inspiratory muscle activity (EMGdi and EMGps) increased six fold relative to activity at baseline during unloaded breathing. There was no expiratory muscle activity (EMGta) observed at baseline during un loaded breathing  .  With the onset of phasic expiratory muscle activity during  loaded breathing, there was a parallel increase in abdominal pressure swings (Pab) during expiration. Phasic EMGta and expiratory pressure swings were maintained during the first 20 minutes of loaded breathing (Figure 19).  125 Table VII: Respiratory variables, arterial blood gases and pH at baseline and during extreme inspiratory resistive loading.  Duration of Loading (minutes)  VT (ml) Bf (1/mm) VE (mi/mm) VT/TI (ml/sec) TI/TrOT  Baseline 18(1) 47(6) 835(107) 36(4) 0.37 (0.03)  12(2)* 28(4)* 322(40)* 14(2)*  20 11(1)* 25(3)* 276(41)* 15(2)*  8(1)*t 21(2)*t 155(17)*t 12(1)*  0.37 (0.04)  0.26j (0.04)  0.20*1 (0.03) 0.7(0.1)*  Ti (sec) TE (sec) pH  0.5(0.1) 0.9(0.1) 7.39 (0.02)  0.9(0.1)* 1.7(0.3)* 7.09* (0.04)  0.8(0.1)* 2.2(0.3)* 7.01* (0.03)  2.8(0.5)*t 6.86* (0.06)  PaCO2 (mm Hg) Pa02 (mm Hg) ] (mmol/L) 3 [HC0  47(2) 157(30) 28(1)  119(16)*j 114(13)* 33(2)*  151(20)* 61(10)* 34(2)*  187(21)* 25(1)* 34(1)*  Values represent mean (± S.E.M.), n=6. <0.05 compared to baseline. <0.05 compared to value at 20 minutes load. p < 0.05 compared to other loaded periods.  126  Figure 19. Top panel: Peak moving average of the electromyogram from the costa! diaphragm (EMGdi), parasternal intercostal (EMGps) and transversus abdominis (EMGta) muscles at baseline and during loaded breathing. Failure to maintain inspiratory pressure, designated “F” on the abscissa occurs 30-40 minutes after loading. Bars represent mean ± S.E.M.  *  =  I  significantly different from other loaded periods, p <0.05.  significantly different from baseline, p <0.05.  Bottom panel: Peak inspiratory airway pressure, Pao (v) and expiratory swings in abdominal pressure, Pab  (.)  during  baseline and loaded breathing. Symbols represent mean ± S.E.M.  *  t  =  significantly different from baseline, p < 0.05.  significantly different from other loaded periods, p < 0.05.  1Z7  600  di  I  500  *  (ps *  ta  **  U,  *  -I  C D  400  *  C) >  a  300  C) L  C:,  200  t  lii  1 00 0  Baseline  10  20  F  Loaded period (mm utes) *  65 *  60 55 0 c’J  E C) a) D U) C,,  50 45  40 35 30  a)  25  0  20  L.  15 *  10  *  •1•  5 0  Baseline  10  20  F  Loaded period (minutes)  128 After 20-40 minutes of loaded breathing there was progressive drop in EMGta and a parallel fall in expiratory abdominal pressure swings within a 10 minute interval that was associated with a significant drop in Pao from -60 ± 4 cm H20 to -49 ± 5 cm H20 and in Pab from 9± 2 cm H20 to 3± 1 cm H20 (Figures 19, period F; Figure 20, panel C). In 4/6 rabbits EMG1a decayed to zero at this time. Inspiratory muscle activities (EMGdi and EMGps) were maintained during this period despite severe hypercapnia and hypoxia (Pa02=25± 1 mm Hg, Table VII). Following decreased inspiratory pressure output, ventilation continued for an average of 5 minutes until respiratory arrest. During this interval, breathing was characterized initially by periodic clusters during which peak inspiratory pressure (Pao) and activity of the inspiratory muscles (EMGdi, EMGps) were maintained followed by a parallel decay in peak EMGdI, EMGps, airflow and airway and blood pressures to zero (Figure 21).  129  Figure 20. Sample tracing of arterial blood pressure (BP), airway pressure (Pao), abdominal pressure (Pab), airflow  (7),  tidal volume (VT) and moving average of  the costal diaphragm (EMGdi), the parasternal intercostal (EMGps), and the transversus abdominis (EMGta) muscles during A) baseline, unloaded breathing B) loaded breathing C) loaded breathing at time of inspiratory pressure failure (task failure). Arrows represent zero.  130  B  A  C t4J4’  BP 50 mm Hg I  Pao 10 cm H 0 2  Pab  /  0 2 5 cm H  / EMGdi / EMGps  EMGta  VT 10 ml I  4•—  5 seconds  —+  EMF’V\ 44’\ 4 M  131  Figure 21. Sample tracing of arterial blood pressure (BP), airway pressure (Pao), airflow (r), moving average of costal diaphragm (EMGd1) and parasternal intercostal (EMGps) activity in the final minutes leading to respiratory arrest. Activity of transverus abdominis [not shown] decayed to zero in this preparation. Arrows represent zero.  H20  EMG  IOcm  5OmmHg  BP  I  :111111  H  .  i’,1 1_i IIiJj  <  ii  mrnute> -  I  flI  j fl  1[1  .  II  -  I  j fl  H I  —.  I’I/[  111  H,  __________  ())  133 Discussion  Response to inspirato,y resistive loading  Inspiratory flow resistive loading is used to examine the contribution of diaphragm fatigue to inspiratory muscle overload. In Chapter II, we demonstrate that neural activation (ENGdi), neuromuscular transmission (M-wave) and activity (EMGdi) of the diaphragm under inspiratory resistive loads of varying intensity and duration is maintained in the anesthetized rabbit. However, under extreme inspiratory resistive loading conditions in this animal model, there is a significant decrease in total respiratory muscle force output (Pao) prior to respiratory arrest and at a time when inspiratory duration Ti is unchanged. The major finding in this study is that this decrease in inspiratory pressure generation is associated with a loss of abdominal muscle activity (EMGta) and that the activity of both inspiratory muscles (EMGdi and EMGps) studied remains coupled and maintained until respiratory arrest.  The parasternal intercostals and the diaphragm are mechanically coupled (De Troyer and Sampson, 1982) and without their coordinated activity the upper ribcage would retract inwards during inspiration. Despite the maintenance of electromyographic activity in the two inspiratory muscles, we can not exclude the possibility that the two inspiratory muscles failed as pressure generators because  134 Pressure failed Pes during this severe load. or i Pd re asu me to le ab we were un ssure can not be Therefore, the drop in pre d. ge an ch un s wa TI en at a time wh due tion of Ti, and should be na mi ter e tur ma pre or n attributed to a prolongatio d tivity of the parasternal an ac e Th n. tio ac ntr co e scl to reduced respiratory mu rked time when there was a ma a at ed uc red t no s wa the diaphragm muscles pped y, airway pressure Pao dro all ion dit Ad ty. ivi act e scl decline in abdominal mu atory e task failure of the respir rib asc we e, for ere Th . ne when Pab began to decli ive loads, st severe inspiratory resist ain ag , res ssu pre ry ato pir system to generate ins atory l muscles rather than inspir na mi do ab the by ist ass m to a failure in diaphrag muscle fatigue. act with scles are stimulated to contr mu l na mi do ab t tha d he It has been establis worth et induced hyperpnea (Ains ic pn rca pe hy g rin du ty ivi phasic expiratory act et at, i et aL, 1989; Van Lunteren sak ka Ta 9; 198 n, lse Ke d aL, 1989; Oliven an eshold ry resistive, elastic and thr ato pir ex r de un d an 2) 1988; Wakai et aL, 199 1989; Road, 1989; Oliven Ct aL d an ers ev Le 3; 196 , op ish loaded breathing (B dominal portance of expiratory ab im the r, ve we Ho 9). 198 Oliven and Kelsen, derscored istive loading has been un res ry ato pir ins to e ns po muscle activity in res istivc our model, inspiratory res In 0). 199 9; 198 , aL et lla only recently (Scarde tment of hypercapnia. The recrui ve ssi gre pro ere sev th wi ed loading is associat 2 c1’ the excitatory effects of C0 of tue vir by ely lik st mo is abdominal muscles abdomir&l d Kirkwood, 1979) and an ton ain (B ns uro ne ry ato bulbospinal expir  135 motoneurons (Ledlie, Pack and Fishman, 1983).  Recruitment of abdominal muscles can increase the mechanical efficiency of the diaphragm during stimulated breathing by two mechanisms. The first mechanism involves a decrease in end expiratory lung volume (EELV) due to increased abdominal activity. Decreased EELV associated with the onset of abdominal activity has been demonstrated during inspiratory resistive and inspiratory threshold loading in both anesthetized and awake preparations (Mayock et aL, 1991; Oliven et aL, 1988; Martin, Aubier and Engel, 1982). A reduction in EELV in turn increases diaphragm length which allows for greater diaphragm shortening and hence increases the pressure generating capability of the diaphragm  (Road et aL, 1986; Road and Leevers, 1988).  Additionally,  abdominal muscle recruitment during stimulated breathing may assist the diaphragm through its relaxation at end expiration resulting in outward recoil of the chest wall that can contribute passively to inspiratory mechanical flow (Agostoni and Torn, 1967).  Recruitment of abdominal muscles during  inspiratory loading is not limited to studies examining anesthetized animal models. In fact, abdominal muscle activity during inspiratory loading is described in both awake goats and humans (Scardella et aL, 1990; Martin, Aubier and Engel, 1982).  In our study, expiratory muscle activity in all the abdominal  muscles is assumed to have decreased as expiratory Pab dropped in parallel with the drop in EMGta. We did not measure expiratory activity of the rib cage  136 muscles but expect that their activities were reduced as well since there is evidence that common bulbospinal neurons project to both rib cage expiratory muscles and the abdominal muscles alike (Road and Kirkwood, 1993).  To our knowledge the failure of abdominal muscle assist to diaphragm during inspiratory resistive loading has not been demonstrated previously.  The  mechanism(s) underlying a selective decrease in abdominal activity with severe loading remains speculative.  Several studies indicate that whereas hypoxic  stimulation of ventilation results in an increase in inspiratory muscle activity, expiratory muscle activation is either less, relative to hypercapnia (Brice et aL, 1990, Sears Berger and Philipson, 1982; Smith et aL, 1989), or inhibited (Fregosi, 1987). In other words, breathing becomes more of an inspiratory act during hypoxia. Such an inspiratory shift or selective inhibition of expiratory activity during hypoxia has been attributed to either 1) medullary hypocapnia secondary to hypoxia (Saupe et aL, 1992); 2)  preferential distribution of peripheral  chemoreceptor input to inspiratory pre motor neurons (Oyer Chae et aL, 1992) or 3) hypoxic depression of the brain stem (Fregosi et aL, 1987). In our preparation, selective inhibition of abdominal muscles during severe inspiratory resistive loading can not be due to central hypocapnia as progressive severe hypercapnia accompanies hypoxemia throughout inspiratory resistive loaded breathing.  The current literature suggests that depression of central neural  output during hypoxia is not uniform and there may well exist differences in the  137 vulnerability of respiratory motor outputs to central hypoxia (Neubauer, Melton and Edelman, 1990).  Given that hypoxemia is progressive and notably severe (PaO2  =  25± 1 mmHg)  by the time phasic expiratory abdominal activity (EMGta) and expiratory pressure swings (Pab) are reduced in our preparation, it seems likely that selective inhibition of the abdominal expiratory muscles reflects the non uniform vulnerability of respiratory nuclei to central hypoxia. In fact, the periodic cluster breathing (Biot breathing) observed shortly after the decrease in abdominal activity prior to respiratory arrest is characteristic of severe neural depression attributed to anoxia or brain stem damage (Plum and Brown, 1963; Webber and Speck, 1981). Furthermore, the transition from cluster breathing to respiratory arrest is marked by a parallel decrease in blood pressure and inspiratory muscle activity.  If the vulnerability of respiratory nuclei to central hypoxia is not  uniform then it provides the simplest explanation for the sequelae of decreased abdominal muscle activity, blood pressure and inspiratory muscle activity. According to Neubauer Ct aL (1990), if we consider central outputs other than those involved in control of respiration, it seems quite likely that the vulnerability of brainstem nuclei to hypoxia are non uniform. For example, single fibre recordings from the preganglionic cervical sympathetic nerve show separate populations of fibres that either increase or decrease their activity in response to brain hypoxia. Whole nerve recordings from the same nerve suggest  138 that brain hypoxia selectively depresses the phasic component and increases the tonic component of sympathetic discharge (Wasicko et aL, 1990). Therefore, although depression is a generalized response of the brain stem to CNS hypoxia, there may exist selective vulnerability within this network to hypoxia.  In a current conceptual model of respiratory muscle activity during inspiratory resistive loading, Petrozinno and co workers (1992) suggest that the response of each respiratory muscle will represent a balance between load-compensating reflexes and differential effects of endogenous opioids released during severe resistive breathing on individual respiratory muscles.  To what degree  endogenous opioids play a role in the differential response of inspiratory versus expiratory muscles to severe loads is difficult to ascertain because their levels in the anesthetized preparation have not been measured directly.  Responses to hyperoxic hypercarbic gas mixture. The magnitude of the ventilatory response to the average change in PaCO2 was within the reported range for anesthetized rabbits (Widdicombe, 1969). Although we did not examine the C02 threshold for expiratory muscle activity, we found expiratory muscle recruitment in response to 9% inspired C02 within a 2-3 minute delay in all rabbits. A similar delay in expiratory muscle activity to hypercapnia has been documented in the anesthetized cats (Bishop and Bachofen, 1972). It is important to note that the magnitude of EMGta in response to conventional C02 stimulated  139 breathing was significantly less than that observed with severe inspiratory resistive loading possibly owing to a lower C02 stimulus (PaCO2 = 69± 3 during C02 breathing versus PaCO2= 119± 16 after 10 minutes of loaded breathing). Additionally, inspiratory resistive loading results in greater electromyographic activity of the two inspiratory muscles due to an increase in inspiratory time (vagally mediated) and possibly from facilitatory afferent input(s) from chest wall muscle mechanoreceptors (Sant’Ambrogio and Widdicombe, 1965).  The discrepancy between the severity of the CO 2 stimulus during CO 2 stimulated breathing and inspiratory loaded breathing is considerable. In particular, PaCO2 reaches narcotic levels (187 mm Hg  =  26%) at the time of task failure.  Therefore, the central depressant effects of extreme CO 2 could potentially contribute to task failure as well.  In conclusion, anesthetized rabbits increase the phasic activity of both inspiratory and expiratory muscles in response to severe inspiratory resistive loads. Despite severe respiratory acidosis and hypoxemia, the phasic inspiratory activity of the costal diaphragm and parasternal muscles remains elevated throughout loading. The failure to generate inspiratory pressure (inspiratory task failure) results from a decay in abdominal muscle activity during expiration.  140 References  1. Agostoni, E. and G. Torn (1967). An analysis of the chest wall motions at high values of ventilation. Respii PhysioL 3: 3 18-332. 2. Ainsworth, D. M., C.A. Smith, S.W. Eicker, K. S. Henderson and J. A. Dempsey (1989). The effects of chemical versus locomotory stimuli on respiratory muscle activity in the awake dog. Respir PhysioL 78: 163-176. 3. Aldrich, T. K and D. Appel (1985). Diaphragm fatigue induced by inspiratory resistive loading in spontaneously breathing rabbits. J AppL Physiol 59: 1527-1532. 4. Aldrich, T.K. (1987). Transmission fatigue of the rabbit diaphragm. Resp. PhysioL 69: 307-3 19. 5. Aldrich T. K. (1988). Central fatigue of the rabbit diaphragm. Lung 166: 233-241. 6. Aleksandrovna, N.P. and G.G. Isaev (1990). Central and peripheral components of the fatigue of respiratory muscles in inspiratory resistive load in cats. Sechenov Physiological Journal of the US.S.R. 76: 658-666. 7. Arnold, J.S., M.A. Haxhiu, N.S. Cherniack and E. van Lunteren (1988). Transverse abdominis length changes during eupnea, hypercapnia and airway occlusion. J AppL PhysioL 64, 658-665. 8. Bainton, C.R. and P.A. Kirkwood (1979). The effect of carbon dioxide on the tonic and the rhythmic discharges of expiratory bulbospinal neurones.  141 PhysioL 196: 291-314. 9. Bazzy, A.R. and G.G. Haddad (1984). Diaphragmatic fatigue in unanesthetized adult sheep. J AppL PhysioL 57: 182-190. 10. Bishop, B. and H. Bachofen (1972). Comparison of neural control of diaphragm and abdominal muscle activities in the cat. J AppL PhysioL 32: 798-805. 11. Bishop B. (1963). Abdominal muscle and diaphragm activities and cavity pressures in pressure breathing. J AppL PhysioL 18: 37-42. 12. Brice, A.G., H.V. Forster, L.G. Pan, T.F. Lowery and C.L. Murphy (1990). Respiratory muscle electromyogram responses to acute hypoxia in awake ponies. Am. J PhysioL 68: 1024-1032. 13. De Troyer, A. and M.G. Sampson (1982). Activation of the parasternal intercostals during breathing efforts in human subjects. .1 AppL PhysioL 52: 524-529. 14. Fregosi, R.F., S.L. Knuth, D.K. Ward and D. Bartlett, Jr. (1987). Hypoxia inhibits abdominal expiratory nerve activity. J AppL Physiol. 63: 221-220. 15. Ledlie, J.F., A.I. Pack and A.P. Fishman (1983). Effects of hypercapnia and hypoxia on abdominal expiratory nerve activity. J AppL PhysioL 55: 1614-1622. 16. Leevers, A.M. and J.D. Road (1989). Mechanical response to hyperinflation of the two abdominal muscle layers. J AppL PhysioL 66: 2189-2195.  142 17. Martin, J., M. Aubier and L.A. Engel (1982). Effects of inspiratory loading on respiratory muscle activity during expiration. Am. Rev. Respu Dis.  125:  352-358. 18. Mayock D.E., R.J. Badura, J.F. Watchko, T.A. Standaert and D.E. Woodrum (1987). Response to resistive loading in the newborn piglet. Pediatr. Res. 21: 121-125. 19. Mayock, D.E., T.A. Standaert, T.D. Murphy and D.E. Woodrum (1991). Diaphragmatic force and substrate response to resistive loaded breathing in the piglet. .L AppL PhysioL 70: 70-76. 20. Neubauer, J.A., J.E. Melton and N.H. Edelman (1990). Modulation of respiration during brain hypoxia. J AppL PhysioL 68: 441-45 1. 21. Oliven A., E.C. Deal, Jr., S.G. Kelsen and N.S. Cherniack (1985). Effects of hypercapnia on inspiratory and expiratory muscle activity during expiration. J AppL PhysioL 59: 1560-1565.  22. Oliven, A., S. Lohda, M.E. Adams, B. Simhai and S.G. Kelsen (1988). Effect of fatiguing resistive loads on the level and pattern of respiratory activity in awake goats. Respu PhysioL 73: 3 11-324. 23. Oliven A. and S.G. Kelsen (1989). Effect of hypercapnia and PEEP on expiratory muscle EMG and shortening. J AppL PhysioL 66: 1408-1413. 24. Oyer Chae, L., J. Melton, J.A. Neubauer and N. H. Edelman (1992). Triagularis sterni and phrenic nerve responses to progressive brain hypoxia. J AppL PhysioL 72: 1522-1528.  143 25. Petrozzino J.J., A.T. Scardella, T.V. Santiago and N.H. Edelman (1992). Dichioroacetate blocks endogenous opioid effects during inspiratory flow resistive loading. i: AppL PhysioL 72: 590-596. 26. Plum F. and H.W. Brown (1963). The effect on respiration of central nervous system disease. Ann. N Y Acad. Sci 109: 915-930. 27. Richardson P.S. and J.G. Widdicombe (1969). The role of the vagus nerves in the ventilatory responses to hypercapnia and hypoxia in anesthetized and unanesthetized rabbits. Respu Physiol. 7:122-135. 28. Road, J.S. Newman, J.P. Derenne, and A. Grassino. (1986) In vivo lengthforce relationship of canine diaphragm. J AppL PhysioL 60: 63-70. 29. Road, J.D. and A.M. Leevers (1988). Effect of lung inflation on diaphragmatic shortening. J AppL PhysioL 65: 2383-2389. 30. Road, J.D. and P.A. Kirkwood (1993). Distribution of monosynaptic connections from expiratory bulbospinal neurons to motoneurons of different expiratory muscles in the cat. XXXII Congress of the I. UP.S. 141: 39, 1993 (Abstract). 31. Sant’Ambrogio, G. and J.G. Widdicombe (1965). Respiratory reflexes acting on the diaphragm and inspiratory intercostal muscle of the rabbit. J PhysioL 180: 776-779. 32. Saupe, K.W., C.A. Smith, S. Henderson and J.A. Dempsey (1992). Respiratory muscle recruitment during selective central and peripheral chemoreceptor stimulation in awake dogs. .1 PhysioL 448: 613- 631.  144 33. Scardella, A.T., T.V. Santiago and N.H. Edelman (1989). Naloxone alters the early response to an inspiratory flow-resistive load. J AppL PhysioL 67: 1747- 753. 34. Scardella, A.T., J.J. Petrozzino, M. Mandel, N.H. Edelman and T.V. Santiago (1990). Endogenous opioid effects on abdominal muscle activity during inspiratory loading. J AppL PhysioL 69: 1104-1109, 1990. 35. Sears, T.A., AJ. Berger and E.A. Philipson (1982). Reciprocal tonic activation of inspiratory and expiratory motoneurons by chemical drives. Nature 299: 728-730. 36. Smith, C.A., D.M. Ainsworth, K.S. Henderson and J.A. Dempsey (1989). Differential responses of expiratory muscles to chemical stimuli in awake dogs. J AppL PhysioL 66: 384-391. 37. Takasaki, Y., D. Orr, J. Popkin, X. Xie and T.D. Bradley (1989). Effect of hypercapnia and hypoxia on respiratory muscle activation in humans. .1 AppL PhysioL 67: 1776-1784. 38. Van Lunteren, E., M.A. Haxhiu, N.S. Cherniack and J.S. Arnold (1988). Ribcage and abdominal expiratory muscle responses to CO 2 and esophageal distension. .1. AppL PhysioL 64: 846-853. 39. Wakai, Y. M., M. Welsh, A.M. Leevers and J.D. Road (1992). The effect of continuous positive airway pressure and hypercapnia on expiratory muscle activity during wakefulness and sleep. J AppL PhysioL 72: 88 1-887. 40. Wasicko, MJ., J.E. Melton, J.A. Neubauer, N. Krawciw and N.H.  145 Edelman (1990). Cervical sympathetic and phrenic nerve responses to progresssive brain hypoxia. J AppL PhysioL 68: 53-58. 41. Webber, C.L. (Jr.) and D.F. Speck (1981). Experimental biot periodic breathing in cats effects of changes in Pi02 and PICO2. Respir. PhysioL 46: 327-344.  146 V: Summary and Conclusions  1.  The primary aim of this study was to assess central motor output and  neuromuscular transmission to the diaphragm during prolonged sustainable and exhaustive inspiratory resistive loads in the anesthetized rabbit. The experiments presented in Chapter II demonstrated that under all loads examined, including the exhaustive load that resulted in task failure, the central motor output to the diaphragm, as assessed by the phrenic electroneurogram, remained elevated throughout loaded breathing.  There was a linear relationship between the  severity of the target inspiratory pressure achieved with resistive loading and the indices of central motor output to the diaphragm and activity of this muscle. The electromyographic activity of the diaphragm was elevated throughout loaded breathing.  There was no significant change in the relation between central  motor output to the diaphragm and electromyographic activity of this muscle [ENG:EMG ratio]. Additionally, neuromuscular transmission to the diaphragm, as assessed by evoked compound potentials of the diaphragm, remained intact during all inspiratory resistive loads. Therefore, we have concluded that central motor output and neuromuscular transmission to the diaphragm do not contribute to task failure during inspiratory resistive loading in the anesthetized rabbit.  Methodological differences may explain the conflicting results between our  147 findings and those of others (Aldrich, 1987; Bazzy and Donnelly, 1993) in estimating the role of neuromuscular transmission in task failure during inspiratory resistive loading. These differences are described in detail in Chapter II. By examining inspiratory resistive loads of both greater duration and intensity than those previously reported we have constructed a clearer picture of the ventilator>’ response to inspiratory resistive loading.  2. The second aim of our study was to determine the changes in breathing pattern in response to inspiratory resistive loading of different intensities and durations, ranging from sustainable to exhaustive loads.  In Chapter II, we  reported that despite significant increases in inspiratory pressure swings during loaded breathing, ventilation was reduced compared to the baseline values obtained prior to loaded breathing.  Consequently, arterial carbon dioxide  tension rose during all loads examined in our study. Depending on the severity of the load, hypoventilation was due to changes in breathing pattern which suggest that an optimization of respiratory work develops early in response to loaded breathing and is achieved through a decrease in central rhythm. Furthermore, we demonstrated that a similar strategy arises upon prolonged exposure to elevated levels of arterial CO 2 which develop during prolonged inspiratory resistive loading (Chapter III). Hence, we suggest that although load compensating mechanisms are functional at the onset of inspiratory resistive loading, they compete with load decompensating mechanisms resulting from the  148 hypercapnia associated with prolonged loading.  The changes in breathing pattern under inspiratory resistive loads seemed mechanically appropriate. For example, it has been established that the most obvious effect of external resistance to air flow in humans is a reduction rate of flow and an increase in the time required for completion of the impeded phase (Zechman, Hall and Hull, 1957). In our rabbit model, during relatively minor loads (Load 1) ventilation was stabilized by a slight drop in breathing frequency due to an increase in inspiratory time.  Increased inspiratory time with  inspiratory resistive loading is consistent with a decrease in Hering-Breuer inspiratory inhibitory lung inflation input due to the decreased inspiratory air flow. Moderate inspiratory resistive loads (Load 2) led to a slight decrease in tidal volume and a significant drop in breathing frequency.  The drop in  breathing frequency resulted from increases in both the inspiratory and expiratory phase of breathing with moderate loading. Prolonging expiratory duration increases the time available to the inspiratory muscles to recover before the onset of the next inspiration and seems an effective mechanism to optimize the function of inspiratory muscles. During severe inspiratory resistive loads (Loads 3 and 4) this slow pattern of breathing was maintained until severe alterations in arterial blood gases and pH led to periodic cluster breathing and culminated shortly thereafter in respiratory arrest (Chapter IV).  149 Central modulation of respiratory activity achieved through a reduction in breathing frequency during inspiratory resistive breathing has been described previously in the awake infant monkey (Watchko et aL, 1988).  As this  observation was limited to an awake preparation, it had been suggested that such reductions in breathing frequency during inspiratory resistive loading reflect conscious motivational fatigue (Aldrich, 1991). In our study, we have extended these findings to the anesthetized preparation to show that this type of central response is not necessarily a result of conscious or behavioral factors. Furthermore, a decrease in central rhythmogenesis is not limited to breathing against exhaustive inspiratory resistive loads. Reduced frequency of breathing was typically seen with all inspiratory resistive loads and was sustained for prolonged periods as well. Additionally, we suggested that this form of central fatigue represents a means for maintaining respiratory muscle function during inspiratory resistive loading.  3. The third aim of our studies was to determine the target inspiratory resistive loads which lead to task failure. We were able to document task failure only during the most severe load (Load 4) at target pressures close to the maximum strength (Pdi max = 55± 9cm H20) previously documented for the rabbit diaphragm during supramaximal electrical stimulation with a bound abdomen (Aldrich and Appel, 1985; Aldrich, 1987, 1988, Ferguson et aL, 1990).  It is  important to clarify that previous studies had demonstrated a drop in the  150 frequency-pressure curve of the diaphragm within one hour at target pressures approximately one half the intensity of the target pressure that is associated with task failure in our studies (Aldrich and Appel, 1985; Aldrich 1987, 1988). We were able to demonstrate that the rabbit diaphragm is able to maintain transdiaphragmatic pressure for at least 4 hours undergoing inspiratory resistive loading of equivalent intensity (Load 2) without task failure. We were unable to determine esophageal pressure during loads of greater intensity due to repetitive swallowing associated with breathing against target inspiratory pressures of Pao = -30 cm H20. Therefore, initially, we could not determine to what extent the inability of the diaphragm or other respiratory muscles to generate inspiratory pressure contributed to task failure with target inspiratory resistive loads in excess of this value (Loads 3 and 4).  To assess the contribution of respiratory muscles to task failure during inspiratory resistive loading, we monitored the electromyographic activity of the parasternal intercostal and the transversus abdominis muscles and the diaphragm during extreme target inspiratory resistive loads (Pao = -58± 4 cm H20) until respiratory arrest (Chapter IV). We described the sequence of changes that occur from respiratory muscle task failure to respiratory arrest under severe inspiratory resistive loading as follows.  With extreme loading there was a  gradual decrease in Pa02 followed by a progressive decay in phasic expiratory abdominal muscle activity and expiratory abdominal pressure swings that led to  151 a significant decrease in Pao (i.e. task failure). The inspiratory activity of the parasternal intercostals and diaphragm were maintained elevated and coupled during this phase. Therefore, we attributed task failure during this phase to a decrease in abdominal muscle activation.  In the next phase, there was a  transition in breathing pattern to cluster breathing which corresponded with severe arterial hypoxemia (PaCO2 <25 mm Hg) followed by coinciding decreases in arterial blood pressure and inspiratory muscle activity of the diaphgram and parasternal intercostal muscles, airflow and inspiratory pressure leading to respiratory arrest. We proposed that the selective decrease in abdominal activity during exhaustive inspiratory resistive loading is most likely due to the differential inhibitory effect of progressive hypoxemia (Fregosi, 1987) which precedes the reduction in expiratory abdominal activity and expiratory pressure swings.  To our knowledge, this study is the first to show that the failure of abdominal muscle assist to the diaphragm during inspiratory resistive loading precedes task failure.  As inspiratory resistive loading was associated with progressive  hypercapnia in our model, we attributed the recruitment of abdominal muscles during this type of loading to the excitatory effect of CO 2 on expiratory bulbospinal neurons (Bainton and Kirkwood, 1979) and abdominal motoneurons (Ledlie, Pack and Fishman, 1983). We have underscored the importance of expiratory abdominal muscle activation in response to inspiratory resistive  152 loading and characterized the mechanisms by which their recruitment can increase the mechanical efficiency of the diaphragm (see Chapter IV). A recent study on conscious humans subjected to high inspiratory resistive loading demonstrates that the ventilatory response to this type of load is achieved partially through the mechanical assist provided to the diaphragm by the expiratory muscles (Yan, 1993). Our findings show that failure of expiratory muscle activity plays a more critical role in task failure than previously believed. Our observation that reduction in activity of the transversus abdominis accompanies inspiratory pressure failure indicates task failure is central in origin.  Although electromyographic evidence showed that activation of the inspiratory muscles was maintained during task failure under severe inspiratory resistive loading, myogenic fatigue of the inspiratory muscles during this period can not be conclusively ruled out.  We suspect that pressure loss would develop in our  model during supramaximal activation of the diaphragm (Aldrich, 1987) but do not believe that the response of the diaphragm to spontaneous activation against inspiratory resistive loading contributes to task failure. There is limited data on the indices of myogenic fatigue in the inspiratory muscles during inspiratory resistive loading. In contrast to earlier reports of a reduction in glycogen content of diaphragm during inspiratory resistive loading (Bazzy et aL, 1988), recent reports show no change in the levels of either glycogen, ATP, PCr or lactate in this muscle after severe and prolonged inspiratory resistive loading (Ferguson et  153  aL, 1990; Mayock et aL, 1991). Our findings suggest that task failure develops only when inspiratory muscle function is compromised by severe alterations in blood gases that undermine activation of respiratory muscles (central fatigue).  4. The fourth aim of our study was to determine the relationship between task failure and hypercapnic ventilatory failure during inspiratory resistive loading. We found that progressive hypercapnia was a salient feature of inspiratory resistive loading independent of the intensity of the loads examined. Surprisingly, central motor output to the diaphragm (ENGdi) remained constant during prolonged sustainable loads (Load 1 and 2) despite rising levels of arterial PaCO2.  We examined the possibility that prolonged exposure to severe  hypercapnia alone could suppress ventilatory drive to the diaphragm in our model by considering the time-dependent changes in ventilation during hypercapnia equivalent to the levels accompanying loaded breathing (Chapter III). We found a significant reduction in breathing frequency by two hours of  exposure to severe hypercapnia. We discussed several mechanisms that could potentially affect the regulation of CO 2 chemoreception during prolonged hypercapnia to produce a decrease in central rhythm. We concluded that the prolonged hypercapnia associated with inspiratory resistive loading suppresses drive to the diaphragm and results in load decompensation.  154 References  1. Aldrich T.K. and D. Appel (1985). Diaphragm fatigue induced by inspiratory resistive loading in spontaneously breathing rabbits. J AppL PhysioL 59: 1527-1532. 2. Aldrich, T.K. (1987). Transmission fatigue of the rabbit diaphragm. Respir PhysioL 69: 307-3 19. 3. Aldrich, T.K. (1988). Central fatigue of the rabbit diaphragm. Lung 166: 233-241. 4. Aldrich, T.K. (1991). Central and transmission fatigue. Seminars in Respiratoiy Medicine 12: 322-330.  5. Bainton, C.R. and P.A. Kirkwood (1979). The effect of carbon dioxide on the tonic and the rhythmic discharges of expiratory bulbospinal neurones. PhysioL 196: 291-314. 6. Bazzy, A.R., S.R. Akabas, A.P. Hays and G.G. Haddad (1988). Respiratory muscle response to load and glycogen content in type I and II fibres. Exp. NeuroL, 101: 17-28. 7. Bazzy A.R. and D.F. Donnelly (1993). Diaphragmatic failure during loaded breathing: role of neuromuscular transmission. J AppL PhysioL 74: 16791683. 8. Ferguson G.T., C.G. Irvin, and R.M. Cherniack (1990). Relationship of diaphragm glycogen, lactate, and function to respiratory failure. Am. Rev.  155 Respir Dis. 141: 926-932. 9. Fregosi, R.F., S.L. Knuth, D.K. Ward and D. Bartlett, Jr. (1987). Hypoxia inhibits abdominal expiratory nerve activity. J. AppL Physiol. 63: 221-220. 10. Ledlie, J.F., A.!. Pack and A.P. Fishman (1983). Effects of hypercapnia and hypoxia on abdominal expiratory nerve activity. J AppL PhysioL 55: 1614-1622. 11. Mayock, D.E., T.A. Standaert, T.D. Murphy and D.E. Woodrum (1991). Diaphragmatic force and substrate response to resistive loaded breathing ii the piglet. .L AppL PhysioL 70: 70-76. 12. Watchko, J.F., T.A. Standaert, D.E. Mayock, G. Twiggs and D.E. Woodrum (1988). Ventilatory failure during loaded breathing: the role of central neural drive. .L AppL PhysioL 65: 249-255. 13. Zechman, F., F.G. Hall and W.E. Hull (1957). Effects of graded resistance to tracheal airflow in man. J AppL PhysioL 10: 356-362. 14. Yan, S., P. Sliwinski, A.P. Gauthier, I. Lichros, S. Zakynthinos and P.T. Macklem (1993). Effect of global inspiratory muscle fatigue in ventilatory and respiratory muscle responses to CO . J AppL PhysioL 75: 137 1-1377. 2  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088028/manifest

Comment

Related Items