UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Peripheral chemoresponsiveness and exercise induced arterial hypoxemia in highly trained endurance athletes Cooper, Trevor Kenneth 1993-02-20

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
831-ubc_1994-0022.pdf [ 1.3MB ]
Metadata
JSON: 831-1.0077240.json
JSON-LD: 831-1.0077240-ld.json
RDF/XML (Pretty): 831-1.0077240-rdf.xml
RDF/JSON: 831-1.0077240-rdf.json
Turtle: 831-1.0077240-turtle.txt
N-Triples: 831-1.0077240-rdf-ntriples.txt
Original Record: 831-1.0077240-source.json
Full Text
831-1.0077240-fulltext.txt
Citation
831-1.0077240.ris

Full Text

Peripheral Chemoresponsiveness and Exercise Induced Arterial Hypoxemiain Highly Trained Endurance Athletes.byTrevor Kenneth CooperBachelor of Physical EducationA THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMaster of ScienceinTHE FACULTY OF GRADUATE STUDIESSchool of Human KineticsWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIANovember, 1993(c) Trevor Kenneth Cooper, 1993In presenting this thesis in partial fulfillment of the requirements for an advanceddegree at The University of British Columbia, I agree that the Library shall makeit freely available for reference and study. I further agree that permission forextensive copying of this thesis for scholarly purposes may be granted by theHead of my Department or by his or her representatives. It is understood thatcopying or publication of this thesis for financial gain shall not be allowedwithout my written permission.School of Human KineticsThe University of British Columbia2075 Wesbrook MallVancouver, CanadaV6T 1W5Thursday, November 25, 1993ABSTRACTTo determine whether highly trained endurance athletes (HT) who developexercise induced arterial hypoxemia (EIH) also demonstrate reduced peripheralchemoresponsiveness (PC) during exercise, twelve (N=12) HT male cyclists wereselected for study. Basic pulmonary function data(FEV1= 4.69 ± 0.66 L, PVC = 6.12 ±0.82 L,FEV1/FVC = 0.77 ± 0.08,FEFmax= 10.52 ± 1.57 L•sec’ , and MVV = 194 ±21 L.min’) were obtained on all subjects. Subjects exercised on a cycle ergometer toexhaustion to determine their maximal aerobic capacity(“O2max= 5.08 ± 0.32 Lmin1,66.6 ± 4.7 mL.min1•kg),and ventilatory threshold(VO2TH= 3.29 ± 0.12 Lmin1,44.3 ± 4.2 mL.minl.kg).Oxygen saturation of arterial hemoglobin(SaO2max)wasmonitored with an ear oximeter (Hewlett-Packard, 47201A), to determine whethersubjects exhibited EIH(SaO2max91%) during the maximal cycle ergometer test.Subjects withSaO2max 93%were placed in the normal saturation group (NOS,Sa°2max= 93.4 ± 0.4 %) while subjects whoseSaO2max91% were placed in the lowsaturation group (LOS,5a°2max= 89.9 ± 0.9 %). Ventilatory responses to hypercapnic(13% C02,21%02,66% N2) and hyperoxic (100%02)gas mixtures were determined atrest, and during exercise on a cycle ergometer at approximately 25%VO2max,50%VO2max, V02Th.Hypercapnic peripheral chemoresponsiveness was lower in LOSsubjects than NOS subjects and increased in both groups from rest to 50 %VO2max.Hyperoxic peripheral chemoresponsiveness was not different in LOS and NOS subjectsand did not change with exercise. Pre-stimulusSaO2fell significantly during exercise inall subjects with LOS having lowerSaO2than NOS atVO2during the hypercapnicchemoresponse tests only. No evidence for a relationship between pre-stimulusSaO2andeither hypercapnic or hyperoxic peripheral chemoresponsiveness was found. The results11of this study provide information which may help explain variations in the ventilatoryresponse to exercise in athletes. Additionally, data from this study suggest a role ofaltered ventilatory control in highly trained endurance athletes who do and do notdemonstrate exercise induced arterial hypoxemia.111TABLE OF CONTENTSAbstract iiTable of Contents ivList of Tables viiList of Figures ixList of Abbreviations and Symbols XiAcknowledgment xvIntroduction 1Methods 6Subjects 6Resting Pulmonary Function Tests 6Maximal Cycle Ergometer Test 7Chemoresponse Tests 8Apparatus 9Hypercapnic Chemoresponse 11Hyperoxic Chemoresponse 11Saturation 12Statistical Analysis 12Results 14Subject Selection 14Anthropometric Data 16Resting Pulmonary Function Data 16Maximal Cycle Ergometry 17Exercise Workloads 19Hypercapnic Peripheral Chemoresponse 21Pre-Stimulus Arterial Hemoglobin Saturation 25Hypercapnic Peripheral Chemoresponse and Pre-Stimulus ArterialHemoglobin Saturation 26ivHyperoxic Chemoresponse.27Pre-Stimulus Arterial Hemoglobin Saturation 30Hyperoxic Peripheral Chemoresponse and Pre-Stimulus ArterialHemoglobin Saturation 32Discussion 33Resting Pulmonary Function, Maximal Exercise Tests, and Subject Selection.... 34Measurement of Peripheral Chemoresponses 35Hypercapnic Peripheral Chemoresponse 36Hyperoxic Peripheral Chemoresponse 36Effect of Training Status on Peripheral Chemoresponsiveness 37Effect of Exercise on Peripheral Chemoresponsiveness 38Peripheral Chemoresponsiveness and Exercise Induced Arterial Hypoxemia 39Peripheral Chemoresponsiveness and Pre-Stimulus Arterial HemoglobinSaturation 40Implications for Ventilatory Control 40Summary 41Bibliography 43Appendix A 48The Pulmonary System: a Limiting Factor In Exercise Performance 48Normal, Healthy, Untrained Individuals 48Highly Trained, Endurance Athletes 49Exercise Induced Arterial Hypoxemia 50Definition of Exercise Induced Arterial Hypoxemia 50Causes of Exercise Induced Arterial Hypoxemia 50Ventilation-Perfusion Heterogeneity and Veno-Arterial Shunt 51Diffusion Disequilibrium 51Relative Hypoventilation 52VMechanical Limitation of Ventilation and Respiratory MuscleFatigue.52Ventilatory Control 53Control Mechanisms 53Central Control of Ventilation 54Supra-Pontine Control 54Central Chemoreception 55Spinal Motor Neuron Interaction 55Peripheral Control of Ventilation 55Peripheral Mechanoreceptors 56Peripheral Chemoreceptors 56Aortic Body Chemoreceptors 56Carotid Body Chemoreceptors 56Appendix B 58Appendix C 83Appendix D 94viLIST OF TABLESTable 1 Age, height, mass and body surface area of subjects, group data 16Table 2 Resting Pulmonary Function, group data 17Table 3VO2max,peak power output, andSaO2maxof subjects, groupdata 18Table 4VO2TH,and power output atVO2,group data 18Table 5 Power outputs maintained during chemoresponse tests, group data 20Table 6 Age, height, mass, and body surface area, individual subject data 58Table 7 Pulmonary function, individual subject data 59Table 8VO2max, peak power output, andSaO2max, individual subjectdata 60Table 9VO2TH,and power output at VO2Ti-j, individual data 61Table 10 Workloads during chemoresponse tests, individual subject data 62Table 11 Hypercapnic peripheral chemoresponse, individual subject data 63Table 12 Hypercapnic peripheral chemoresponse, group data 64Table 13 Pre-hypercapnic peripheral chemoresponseSaO2,individualsubject data 65Table 14 Pre-hypercapnic peripheral chemoresponse SaO2, group data 66Table 15 Hyperoxic peripheral chemoresponse, group data 67Table 16 Hyperoxic peripheral chemoresponse, individual subject data 68Table 17 Pre-hyperoxic peripheral chemoresponseSaO2,group data 69Table 18 Pre-hyperoxic peripheral chemoresponseSaO2,individual subjectdata 70Table 19 RMANOVA, Hypercapnic Chemoresponse 71Table 20 Polynomial Contrasts, Hypercapnic Chemoresponse 72Table 21 RMANOVA, Pre-Hypercapnic Chemoresponse SaO2 73Table 22 Polynomial Contrasts, Pre-Hypercapnic ChemoresonseSaO2 74Table 23 RMANOVA, Hyperoxic Chemoresponse 75viiTable 24 Polynomial Contrasts, Hyperoxic Chemoresponse76Table 25 RMANOVA, Pre-Hyperoxic ChemoresponseSaO2 77Table 26 Polynomial Contrasts, Pre-Hyperoxic ChemoresponseSaO2 78Table 27 Linear Regression, Hypercapnic Chemoresponse and PreHypercapnic ChemoresponseSaO2, NOS subjects 79Table 28 Linear Regression, Hypercapnic Chemoresponse and PreHypercapnic ChemoresponseSaO2,LOS subjects 80Table 29 Linear Regression, Hyperoxic Chemoresponse and Pre-HyperoxicChemoresponseSaO2,NOS subjects 81Table 30 Linear Regression, Hyperoxic Chemoresponse and Pre-HyperoxicChemoresponseSaO2,LOS subjects 82Table 31 Age, height, mass, and body surface area, individual subject data 94Table 32 Age, height, mass, and body surface area, group data 96Table 33VO2max, peak power output, and SaO2max, individual subjectdata 97Table 34VO2max, peak power output, andSaO2max,group data 99Table 35VO2TH,and power output at/O2m,individual subject data 100Table 36VO2TH, and power output at VO2TH, group data 102vrnLIST OF FIGURESFigure la PCO2, P02, and Vj of single hypercapnic peripheralchemoresponse 21Figure lbPETCO2,ETO2,and Vj for a single hypercapnic peripheralchemoresponse 23Figure 2 Hypercapnic peripheral chemoresponse at various exerciseintensities, group data 24Figure 3Pre-C02 response SaO2 at variousexercise intensities, group data 25Figure 4aPCO2, P02, and VI of a single hyperoxicperipheralchemoresponse 27Figure 4bPETCO2,ETO2,and Vj for a single hyperoxic peripheralchemoresponse 29Figure 5 Hyperoxic peripheral chemoresponse at various exerciseintensities, group data 30Figure 6 Pre-02 responseSaO2at various exercise iiitensities, group data 31Figure 7 Power output at various exercise intensities, group data 83Figure 8 Power output at various exercise intensities, NOS subjects 84Figure 9 Power output at various exercise intensities, LOS subjects 85Figure 10 Hypercapnic peripheral chemoresponse at various exerciseintensities, NOS subjects 86Figure 11 Hypercapnic peripheral chemoresponse at various exerciseintensities, LOS subjects 87Figure 12Pre-C02 response SaO2 atvarious exercise intensities, NOSsubjects 88Figure 13Pre-C02responseSaO2at various exercise intensities, LOSsubjects 89Figure 14 Hyperoxic peripheral chemoresponse at various exerciseintensities, NOS subjects 90ixFigure 15 Hyperoxic peripheral chemoresponse at various exerciseintensities, LOS subjects 91Figure 16Pre-02responseSaO2at various exercise intensities, NOSsubjects 92Figure 17 Pre-O2 responseSaO2at various exercise intensities, LOSsubjects 93xLIST OF ABBREVIATIONS AND SYMBOLS(A-a)D02 Alveolar-arterial oxygen difference.[H+]a Arterial hydrogen ion concentration.EIH Exercise induced arterial hypoxemia.F(x,y)group Omnibus F ratio for RMANOVA group main effect with x numeratordegrees of freedom, and y denominator degrees of freedom.F(x,y)thalsOmnibus F ratio for RMANOVA trials main effect with x numeratordegrees of freedom, andy denominator degrees of freedom.F(x,y)intOmnibus F ratio for RMANOVA trials x group interaction with xnumerator degrees of freedom, andy denominatordegrees of freedom.F(1,y)linOmnibus F ratio for linear polynomial contrast with 1 numerator degreeof freedom, andy denominator degrees of freedom.F(1,y)linmtOmnibus F ratio for group interaction of linear polynomial contrast with1 numerator degree of freedom, and y denominator degrees of freedom.F(1,y)quad Omnibus F ratio for quadratic polynomial contrast with 1 numeratordegree of freedom, and y denominator degrees of freedom.F(1,y)quad mtOmnibus F ratio for group interaction of quadratic polynomial contrastwith 1 numerator degree of freedom, andy denominatordegrees offreedom.fRRespiratory frequency.FVC Forced vital capacity.xiFEV1 Forced expired volume in first second.FEFmaxMaximal expiratory flow rate.GDiC Line of general direction of change.HT Highly trained endurance athletes.[KiaArterial potassium ion concentration.LPO Low power output, 50% of maximal aerobic power.MPO Moderate power output, aerobic power at ventilatory threshold.MVV Maximal voluntary ventilation.PACO2Alveolar partial pressure of carbon dioxide.PAO2Alveolar partial pressure of oxygen.PaCO2Arterial partial pressure of carbon dioxide.a°2Arterial partial pressure of oxygen.PC Peripheral chemoresponsiveness.PCO2 Partial pressure of carbon dioxide.PETCO2End-tidal partial pressure of carbon dioxide.PETO2 End-tidal partial pressure of oxygen.pH Negative logarithm of hydrogen ion concentration.pHaNegative logarithm of hydrogen ion concentration in arterial blood.xliP02 Partial pressure of oxygen.QcPerfusion.RMANOVA Repeated-measures analysis of variance.SaO2Oxygen saturation of arterial hemoglobin.SaO2maxMinimal oxygen saturation of arterial hemoglobin during maximal cycleergometer test.VA Alveolar ventilation.VA:QCVentilation-perfusion ratio.VCO2Rate of carbon dioxide elimination.VE Volume of air expired.VE Expired ventilation, or volume of air expired per minute.VE/VCO2Ventilatory equivalent for carbon dioxide elimination.VE/V02Ventilatory equivalent for oxygen consumption.Vj Volume of air inspired.Vj Inspired ventilation, or volume of air inspired per minute.VLPO Very low power output, 25% of maximal aerobic power.V02 Rate of oxygen uptake.VO2maxMaximal rate of oxygen uptake.x1uVO2Rate of oxygen consumption at the ventilatory threshold.xivACKNOWLEDGMENTI would like to thank all the students, faculty, friends and relatives who encouraged meduring my studies, especially during the data collection of “the thesis that wouldn’t die”.Specifically:Dr. Don McKenzie, my advisor. Don taught me how to do “real” research, andgave me encouragement and support, both emotional andfinancial when I needed it. He exposed me to some of thebetter things in life (Surf on!), gave me memories I willnever forget (Cape North, Tuk, etc....), and occasionally ledme astray (with the help of Howie and James)!Diana Jespersen, my lab buddy. Diana was primary source of emotional“enlightenment” in the lab and was never, well almostnever, short of smiles and laughs that made even the worstmoments seem like they weren’t all that bad. She alsotaught me everything I know about equipment.Dr. Ken Coutts, whose unparalleled problem solving ability and handinesswith duct tape (MacGyver) frequently solved problemseven Diana gave up on. Ken also taught me how to be coolunder pressure when everything in the lab breaks down in asequential serenade to Murphy.Pat, Dave, Tiz, Maria and Jim,my lab-mates and fellow combatants who humored me bywatching me “go through the motions”, who all managed toavoid the traps I fell into, and finished their theses beforeme even though I was the first to start. If they hadn’t brokeneverything I would have finished first!and finally...Sue Hopkins, my “soul” mate. Sue was the only person in the lab I couldregularly argue with and actually learn something in theend as she was usually right. Many of Sue’s ideas havesubconsciously and mysteriously appeared in my research(Actually I think she used to plant thoughts into mysubconscious when I was sleeping and then....) and as aresult I owe most of “this” to her. NOT ALL.. .but most.Thanks again everyone...see you at the beach!xvINTRODUCTIONA number of authors (11, 20, 39-41, 43) have reported a decrease in partialpressure of oxygen in arterial blood and/or desaturation of hemoglobin during intenseexercise in highly trained endurance athletes. This phenomenon has been called exerciseinduced arterial hypoxemia and its incidence has been reported to be between 40 (42) and52 percent (39) in highly trained male endurance athletes. The occurrence of exerciseinduced arterial hypoxemia in highly trained athletes is reproducible with at least onegroup (39) finding a testlretest correlation of 0.95, p<.O5The most likely causes of exercise induced arterial hypoxemia are: alveolarventilation(VA) toperfusion(Qc)heterogeneity(VA:Qc ),veno-arterial shunt,diffusion disequilibrium, and the absence of adequate compensatory hyperventilation (10,20, 49). In normal individuals and those highly trained athletes who do not demonstrateexercise induced arterial hypoxemia,‘/A:Cheterogeneity and veno-arterial shuntaccount for the widening of the alveolar-arterial02difference from less than 5 torr at restto approximately 30 torr during heavy exercise(9). Athletes with high cardiac outputs(33 L•min1)and maximal pulmonary blood volumes(1.6 L) are likely to have veryshort red blood cell pulmonary transit times (10, 19). In addition, the absence of adequateexercise hyperpnea would result in a lowering of alveolarP02on a breath by breathbasis. The rate of equilibration of02would then decrease on a breath by breath basis dueto a lowering of the02driving pressure across the gas exchange barrier. This results in asubsequent fall in arterialP02accompanied by the maintenance of a constant alveolararterial02difference. Another mechanism which may be related to the high cardiacoutputs and maximal pulmonary blood volumes seen in the highly trained athletes whoexhibit exercise induced arterial hypoxemia, is pulmonary edema. There may be a1decrease in diffusion of02across the alveolar membrane which is caused by an increasein the diffusion distance associated with the extravascular lung water accompanyingpulmonary edema (33, 44, 45, 52, 53). Thus, shortened red blood cell pulmonary transittime and pulmonary edema may result in a significant diffusion disequilibrium at the endof the pulmonary capillary. These mechanisms, together with ventilation-perfusionheterogeneity and relative hypoventilation, are likely to explain the additional wideningof the alveolar-arterial02difference to approximately 45 torr in highly trained athleteswho develop exercise induced arterial hypoxemia (9).Conflicting data have been reported regarding the level of compensatoryhyperventilation associated with heavy exercise and the influence it may have on exerciseinduced arterial hypoxemia. Relative hypoventilation has been associated withhypoxemia during exercise at 75 - 90 %VO2max(11) while another group (20) did notfind evidence for such a hypoventilation accompanying a similar level of exerciseinduced arterial hypoxemia during exercise at essentiallyVO2max.The differencebetween methodologies in these studies could explain the disagreement in their results. Atmaximal exercise intensities the accumulated ventilatory drive associated with thedemand forC02elimination may become so strong that hypoventilation is not presentunder these circumstances when it may have been present at lower exercise intensities. Ina subsequent study (19), a relationship was found betweenPa02and ventilatoryequivalent forC02 (‘JEIVCO2)supporting the hypothesis that hypoventilation plays arole in exercise induced arterial hypoxemia.The ultimate goal of the respiratory system is the maintenance of bloodhomeostasis with respect toa°2,partial pressure of carbon dioxide (C02) in arterialblood(PaCO2),and hydrogen ion concentration in arterial blood ([Hja). Studies2involving carotid body resection in human subjects (17, 18, 50, 56) have shown that theperipheral chemoreceptors play an important role in the control of the acute ventilatoryresponse to hypoxia and hypercapnia. The responsiveness of the peripheralchemoreceptors or the integration of their feedback in the brain stem are responsible, inpart, for the inter-individual differences in the ventilatory response to changes ina°2,PaCO2, [Hia,exercise (9), and possibly potassium (37). Studies investigating theventilatory responses of trained and untrained individuals have reported conflictingresults. One study reported the chemoresponsiveness of individuals to increase withtraining (25), while other authors have reported that highly trained athletes have similar(29) and lower (5, 54) chemoresponsiveness than untrained individuals. Studiescomparing the chemoresponsiveness of humans and animals at rest and exercise have alsoreported conflicting results. One group (31) found a correlation betweenVEandhypercapnic peripheral chemoresponse in man while another (1) did not find such arelationship in the cat. It is clear that chemoresponsiveness is highly variable betweenindividuals and it is this variability that could explain some of the differences in theresults of these studies. However, like the incidence of exercise induced arterialhypoxemia, chemoresponsiveness at rest and mild exercise has been relativelyreproducible, with reported mean coefficients of variation (V) ranging from 23 ± 15%(46) to 25±6%(32).It is interesting to note that reductions ina°2and/orSaO2have been found atexercise intensities below maximum and near the ventilatory threshold (11, 19). Arelative hypoventilation, possibly mediated through reduced peripheralchemoresponsiveness could explain the development of exercise induced arterialhypoxemia at these workloads in some athletes. For this reason this study was designed to3investigate the peripheral chemoresponsiveness of highly trained athletes, who do and donot develop exercise induced arterial hypoxemia, during very light, light, and moderateexercise. The relationship between peripheral chemoresponsiveness and the developmentof exercise induced arterial hypoxemia during exercise was also examined. Our generalresearch questions were:1. Is there a significant difference in peripheral chemoresponsivenessbetween rest, very light, light, and moderate exercise in subjects who doand do not demonstrate exercise induced arterial hypoxemia?2. Is there a significant difference in the pattern of change of peripheralchemoresponsiveness, across exercise levels, between subjects who do anddo not demonstrate exercise induced arterial hypoxemia?3. Is there a significant decrease inSaO2from rest to very light, light, andmoderate exercise in subjects who do and do not demonstrate exerciseinduced arterial hypoxemia?4. Is there a significant difference in the pattern of change ofSaO2, acrossexercise levels, between subjects who do and do not demonstrate exerciseinduced arterial hypoxemia?5. Is there a relationship between peripheral chemoresponsiveness and SaO2measured at each exercise level in subjects who do and do not demonstrateexercise induced arterial hypoxemia.Our hypotheses with regards to the research questions we were interested in were:1. There is a reduction in peripheral chemoresponsiveness from rest to verylight, light, and moderate exercise in subjects who do and do notdemonstrate exercise induced arterial hypoxemia.42. Subjects who demonstrate exercise induced arterial hypoxemia have lowerperipheral chemoresponsiveness than subjects who do not demonstrateexercise induced arterial hypoxemia at all exercise levels.3. There is a reduction inSaO2from rest to very light, light, and moderateexercise in subjects who do and do not demonstrate exercise inducedarterial hypoxemia.4. Subjects who do not demonstrate exercise induced arterial hypoxemia willdemonstrate an initial fall inSaO2from rest to light and perhaps very lightexercise but will not demonstrate a further fall inSaO2at moderateexercise intensities. Subjects who do demonstrate exercise induced arterialhypoxemia will demonstrate a fall inSa°2from rest to light, very light,and moderate exercise.6. Regression analysis ofSaO2and peripheral chemoresponsiveness,measured at each exercise level in subjects who do and do not demonstrateexercise induced arterial hypoxemia, will have a positive correlationcoefficient.5METHODSSUBJECTSHighly trained male cyclists were recruited through personal contact or throughadvertisements in the Cycling British Columbia monthly newsletter and gave informedconsent prior to participation in any experiments. Prior to participation in the study,subjects answered questions that increased the likelihood that they would satisfy theinclusion criteria which were: 1) normal pulmonary function with no history ofpulmonary disease and 2)<lO2max 5.0L•min1or 60.0 mL•min1•kg.RESTING PULMONARY FUNCTION TESTSPulmonary function was tested using a MedGraphics CPXID system equippedwith pulmonary function software. The CPX/D system was calibrated prior to eachtesting session by withdrawing and injecting a known volume (5 x 3.00 Liters) throughthe pneumotach (MedGraphics, disposable) at various flow rates.Each subject performed a minimum of three flow:volume maneuvers in order toobtain a reproducible measurement (according to ATS standards) of forced vital capacity(FVC), forced expiratory volume in one second(FEV1), ratio of forcedexpiratoryvolume in one second to forced vital capacity(FEV1/FVC)and maximal forcedexpiratory flow rate(FEFmax).In addition, subjects performed at least two maximalvoluntary ventilation (MVV) maneuvers. Data obtained were compared to normativevalues generated by the MGC CPXID software. If subject values for FVC,FEVi0,FEV1Q!FVC, FEFmax or MVV fell below the normal range, then subjects were excludedfrom further study.6MAXIMAL CYCLE ERGOMETER TESTSubjects completed a progressive intensity test to volitional fatigue on anelectronically braked cycle ergometer (Mijnhardt, KEM-3). Subjects used their ownpedals and cycling shoes on the cycle ergometer and adjustments were made to the saddleand handlebars to approximate their normal riding position. Prior to testing subjectswarmed-up either by riding at a slow pace to the testing facility on their bicycle or bypedaling on the cycle ergometer until they felt ready to begin. During the progressiveintensity test subjects pedaled at their preferred pedal frequency (between 30 and 100revolutions•min1)at a steadily increasing work rate (30 Watts•min’ ramp) from zeroload until they discontinued the test, or their pedaling frequency fell below 30revolutions•min-1.During the test subjects breathed through a two-way, non-rebreathingvalve (Hans-Rudolph, #2700B). The inspired gas volume (Vacumetrics, #17 150 air flowmeter), and expired gas02(Applied Electrochemistry, Oxygen Sensor N-22M andOxygen Analyzer S-3A11) andCO2(Beckman, LB-2) contents were monitored andrecorded by a personal computer for analysis (Rayfield system). Every fifteen seconds thecomputer system calculated and displayed the expired minute ventilation(‘/E)and therates of oxygen consumption(V02)and carbon dioxide elimination(VO2).After thetest was completed theVO2at the ventilatory threshold(VO2TH)was determinedaccording to a previously documented computer technique (6). In summary, for eachsubject, a third order polynomial curve was fitted to a plot of‘CO2versus VO2 usingleast-squares regression. In addition, a straight line indicating the general direction ofchange (GDiC) was fitted between the endpoints of the polynomial curve fit. Beginningat the lowest measuredV02the computer calculated the distance between the predictedvalue of the polynomial curve fit and the GDiC perpendicular to the GDiC. This7calculation was repeated at 10 mLV02intervals with the largest difference (Dmax)producing theVO2for that subject. Work rates at approximately 25% ofVO2max,50% ofVO2maxandVO2THwere determined from linear least squares regressionanalysis ofVO2and power output. Percent saturation of hemoglobin in arterial blood(SaO2)was measured with an ear oximeter (Hewlett-Packard, 47201A) and recorded onanother personal computer (1 Hz) for later analysis. Prior to placement of the ear sensor atopical vasodilator cream (Finalgon®, Boehringer/Ingetheim) was applied to the pinna ofthe ear to enhance perfusion. TheSaO2data obtained during the progressive intensity testwas smoothed as a 30 second moving mean with the lowest value chosen asSaO2max.The level ofSaO2max chosen for inclusion into the LOS group (SaO2max91.0 %) wasbased on the definition of exercise induced arterial hypoxemia reported previously (39).The level ofSaO2maxchosen for inclusion into the NOS group(SaO2max93.0 %) waschosen to differentiate the NOS subjects as much as possible from subjects in the LOSgroup. Subjects satisfying these criteria who could not be assigned to either the exerciseinduced arterial hypoxemia group (LOS) or the normal group (NOS) based upon resultsin the maximal cycle ergometer test were excluded from further study.CHEMORESPONSE TESTSVentilatory responses to hypercapnia and hyperoxia were determined at rest andwhile the subjects exercised at approximately 25% ofVO2max(very low power output,VLPO), 50% ofVO2max(low power output, LPO) andVO2(moderate power output,MPO). The chemoresponse trials at rest and each exercise intensity were performed ondifferent days at least 24 hours after training. Subjects reported to the lab at least twohours after eating or drinking caffeine. During the resting determinations, subjects8remained in a supine position on a cot. Prior to exercise determinations, subjects adjustedthe cycle ergometer then warmed-up for 5-10 minutes at approximately 50% of theirexercise work rate. During all determinations subjects listened to music from a radio orwith earphones.ApparatusThe same apparatus was used for the hypercapnic and hyperoxic chemoresponsetests. The subjects breathed through the pneumotach of the CPXJD system which wasconnected to a differential pressure transducer (MedGraphics, disposable; Validyne,DP250). The flow measurement system of the CPX/D was calibrated by withdrawing andinjecting a known volume through the pneumotach at various flow rates. A sample ofinspired and expired gas was continuously taken from the CPX/D pneumotach and wasanalyzed by the fast response CPXID gas analyzers (Medical Graphics Corporation, 02 -zirconia fuel cell,C02- infrared absorption). The CPX/D gas analyzers were calibratedwith test gases of known composition. This enabled the approximate monitoring of V02,VCO2,andVEon a breath-by-breath basis as well as an accurate record of end-tidal 02andC02pressures(PETO2andPETCO2)during the steady-state period preceding thechemoresponse tests. TheV02,VCO2,andVE dataobtained and displayed in real timeon the CPX/D system were not accurate due to the variations in inspired and expired gascomposition that accompanied the chemoresponse tests. This was primarily due to theeffect of gas density on flow measurements but was also a factor of the inability of theCPXID software to accommodate the input from the gas analyzers during and afterchemoresponse trials (see Figures la and 5a). The raw analyzer outputs of the CPXIDanalyzers were channeled through an AID board to a personal computer where the9appropriate offset and scale factors were applied to the raw signals enabling the real timedisplay ofP02andPCO2during the chemoresponse tests. The CPXID flow signal wasused to confirm the timing of the flow signal on the inspiratory side of the breathingcircuit (see below). The distal end of the CPXID pneumotach was connected to a two-way non-rebreathing valve (Hans-Rudolph, #2700B). The inspiratory port ofthe nonrebreathing valve was connected to the outlet of a 4-way t-type valve (Hans-Rudolph,).Three of the inlets of the 4-way valve were connected to three ports of a bag-in-boxsystem in the following manner. One of the inlets of the 4-way valve was connected to aport that was continuous with the interior of the airtight plexiglass box (volume 250liters). The second inlet was connected to a port that was continuous with one 60 literDouglas bag inside the box that was partially filled with the hyperoxic gas mixture. Thethird inlet was connected to a port that was continuous with a second 60 liter Douglas baginside the box that was partially filled with the hypercapnic gas mixture. The fourth portof the bag-in-box system was connected to the pneumotach and differential pressuretransducer of a Medical Graphics MGC/2001 metabolic cart (Hans-Rudolph, #3 800;Validyne, DP250). The flow measurement system of the MGC/2001 was calibrated byrepeatedly injecting a known volume through the pneumotach at various flow rates. Theflow signal of the MGC/2001 system was channeled through the A/D board and into thepersonal computer where the appropriate offset and scale factors were applied, the signalwas integrated and the resulting inspired volume (Vi) signal was displayedin real time.Using this system, Vj was accurately determined by measuring inspired air of constantgas composition and density during the chemoresponse trials. SaO2 wasmonitoredthroughout the testing period (Hewlett-Packard, 47201A Oximeter). The oximeter wascalibrated according to the published instructions immediately before the chemoresponse10testing began and again halfway through the chemoresponse determinations. TheSaO2signal was channeled through the AID board and into a personal computer whereappropriate offset and scale factors were applied andSaO2was displayed in real time.Every effort was made to prevent the subjects from being aware of changes in inspiredgas composition during all determinations. Subjects were allowed to adjust to theapparatus for five to ten minutes and when theirVE,ETO2,and Pj’J’C0 hadstabilized, data collection commenced.Hypercapnic ChemoresponseThe technique used for the determination of the hypercapnic peripheralchemoresponse was modified from that previously reported by another author (32). Thegas mixture used was approximately 13%C02,21%02and 66% N2. After a minimumof 30 seconds of pre-stimulus data had been collected, subjects were switched to thehypercapnic gas mixture using the 4-way valve for one breath and then immediately backto room air. Data collection continued for approximately 60 seconds. A period of three tofive minutes separated each of a minimum of five repeated trials.The hypercapnic chemoresponse for each subject was determined in the followingmanner. The control ‘/j was calculated as the mean ‘/j of the five breaths immediatelypreceding the stimulus breath. The stimulus ‘7J was chosen as the highest single breath‘/j recorded within 20 seconds of the stimulus breath. The control PETCO2 wascalculated as the mean value during the 30 second pre-stimulus period. The trial responsewas calculated as the ratio of the difference between the control Vi and the stimulus Vj tothe difference between the controlP1fC02and the PETCO2 of the stimulus breath. Theindividual subject response was calculated as the mean of the five trial responses.11Hyperoxic ChemoresponseThe technique used for the determination of the hyperoxic peripheralchemoresponse was modified from that previously reported by another author (47). Thegas mixture used was 100% 02. After a minimum of 30 seconds of pre-stimulus data hadbeen collected, subjects were switched to the hyperoxic gas using the 4-way valve.During resting studies subjects breathed the hyperoxic gas for 20 seconds (usually threebreaths). During exercise studies subjects breathed the hyperoxic gas for the samenumber of breaths as they had in the resting studies. Once the period of hyperoxic gasbreathing was completed subjects again inspired room air. Data collection continued forapproximately 60 seconds. A period of three to five minutes separated each of aminimum of five repeated trials.The hyperoxic chemoresponse for each subject was determined in the followingmanner. The control ‘/j was calculated as the mean over the 30 second, pre-stimulusperiod. After hyperoxic breathing had begun ‘1j data was smoothed using a three breathmoving mean. An individual subject response was calculated as the ratio of the lowestthree breath ‘/j value averaged across trials to the average control period Vj value.SaturationFor both the hypercapnic and hyperoxic responses the pre-stimulus SaO2 wasdetermined in the same manner. For each trial theSaO2was determined to be the meanvalue of the fifteen seconds immediately preceding the stimulus breath or breaths. ThesubjectSaO2was calculated as the mean of the five trialSaO2values.12STATISTICAL ANALYSISThe Students’ T-test was used to compare group means of the descriptive subjectdata. The hypercapnic and hyperoxic responses as well as the pre-stimulusSaO2valueswere analyzed as 2 x 4 RMANOVA’s. In addition, the correlation between both sets ofperipheral chemoresponse data and their associated pre-stimulusSa02was determinedwith linear regression analysis. The level of significance for all statistical comparisonswas set at p = 0.05.13RESULTSSUBJECT SELECTIONA total of 36 male cyclists were recruited for the study through advertisements inthe Cycling British Columbia newsletter, and through word of mouth from previoussubjects. Of the initial 36 subjects, 2 were excluded from further study because theirresults in the pulmonary function tests were substantially below predicted, 10 wereexcluded because theirVO2maxwas less than 5.00 L•min’ or 60.0 mL•min1.kg, and12 were excluded because either they did not satisfy theSaO2maxcriteria or the were notneeded as the group they qualified for was complete. Using the criteria,SaO2max91.0%, the first six subjects accepted into the study qualified for the lowSaO2max(LOS)group. The reported incidence of exercise induced arterial hypoxemia of 52 % in highlytrained endurance athletes (39, 42) makes this a highly unlikely occurrence. However, atleast three of the subjects selected for the LOS group had been examined previously forexercise induced arterial hypoxemia in our lab and had tested positive. Based upon thereported incidence of exercise induced arterial hypoxemia in highly trained enduranceathletes only twelve more subjects should have been required to complete the selection ofsubjects for the NOS group. This, however, was not the case. The upper limit of theSaO2maxcriteria,SaO2m94.0 %, was originally chosen in an attempt to create aslarge a separation between subjects in the two groups as possible. Using this criteria toselect subjects for the normalSaO2max(NOS) group resulted in only two subjectsqualifying for that group out of the next fifteen subjects tested. A number of the subjectsnot qualifying for the NOS group did qualify for the LOS group but were not studiedfurther as that group was complete. The remaining subjects hadSaO2maxvalues which14fell between the upper and lower limits of theSaO2maxselection criteria. The upper limitof theSaO2maxcriteria was then lowered toSaO2max93.0 % (1 SD below the meandecrease inSa°2found in normal subjects during exercise (39)) and three more subjectswere found for the NOS group as well as one additional subject for the LOS group. Theadditional LOS subject was accepted because he demonstrated the highest degree ofexercise induced arterial hypoxemia of all subjects tested. Initially it was thought that theapparent low number of subjects qualifying for the LOS group was indicative of anincidence of exercise induced arterial hypoxemia that was much higher than previouslyreported. However, examination of data from all subjects indicated an incidence ofexercise induced arterial hypoxemia in subjects withVO2max 5.00L•min1or 60.0mL•min’ •kg1 of 31 % (8 subjects out of 26) which is substantially lower than valuesreported by Powers et al. (39) and is slightly lower than values reported in more recentstudies (41, 42) even though the subjects in this study had lowerVO2maxvalues.Interestingly, two subjects that did not satisfy the fitness selection criteria (VO2max =4.69 ± 0.19 L•min’, 53.8 ± 3.5mL.minLkg)did demonstrate mild exercise inducedarterial hypoxemia(SaO2max= 90.3 ± 0.1 %). Individual subject descriptive data andVO2max,peak power,SaO2maxandVO2TH datafor subjects not qualifying forcomplete study can be found in Appendix D.Of the subjects selected for complete study, six were competitive road cyclists(LOS, three; NOS, three), three were competitive off-road cyclists (LOS, one; NOS, two),one was a competitive duathlete (LOS), one was a recreational triathiete (LOS), and onewas a recreational athlete who had previously competed as an Olympic class oarsman(LOS).15ANTHROPOMETRIC DATASubjects in both groups were similar in height, weight, and body surface area(BSA) (Table 1). Individual subject values for all descriptive variables can be found inTable 6.Table 1 Age, height, mass and body surface area of subjects, group data.AGE HEIGHT MASS BSAGROUP(yrs) (cm) (kg)(m2)LOS (n=7) 28.9 ± 8.1 180.1 ± 5.9 74.9 ± 7.3 1.94 ± 0.10NOS (n=5) 25.4 ± 5.3 185.3 ± 5.0 79.0 ± 2.5 2.03 ± 0.06Values are means ± SD. LOS, low oxygen saturation; NOS, normal oxygen saturation.RESTING PULMONARY FUNCTION DATAResting pulmonary function data were compared with normative values generatedby the testing apparatus. Subjects in both groups had similar values for FVC, FEVi,FEVi/FVC,and MVV. Subjects in the LOS group had significantly higher FEFmaxvalues than NOS subjects(tio= 2.24,p= 0.025). Individual subject data can be found inTable 7. All subjects demonstrated normal or supra-normal pulmonary function. Thehighest values for any pulmonary function variable, when compared to individualpredicted values, were found in theFEFmax datawith values ranging as high as 152% ofpredicted.16Table 2 Resting Pulmonary Function, group data.FVC FEV1 FEV1/FVCFEFmaxMVVGROUP (L) (L) (L•sec’) (L.min’)(% Pred.) (% Pred.) (% Pred.) (% Pred.)LOS (n=7) 5.76 ± 0.78 4.51 ± 0.65 0.79 ± 0.09 11.25 ± 1.44*197 ± 16104± 12 98±7 132± 14 109± 10NOS (n=5) 6.60 ± 0.63 4.95 ± 0.65 0.75 ± 0.07 9.49 ± 1.18 188±27107±4 97±6 107±11 104±12Values are means ± SD. LOS, low oxygen saturation; NOS, normal oxygen saturation.Value indicated by(*)is significantly higher than NOS(p<0.05).MAXIMAL CYCLE ERGOMETRYThere were no significant differences between groups inVO2max,or peak poweroutput (Table 3). Results from theVO2maxtest for all subjects are listed in Table 8. ThemeanVO2maxof all subjects was 5.08 ± 0.32 L•min’ or 66.6 ± 4.7 mL•min1•kg,indicating the high training status of the subjects. Values forVO2maxfor subjects in bothgroups are comparable to data reported in previous studies (20, 41) but are lower, onaverage, than those reported by Dempsey et al. (11).There was substantial overlap in the range ofVO2maxfor subjects in each group(LOS, 59.6 - 74.5; NOS, 62.6 - 69.9 mL.min1.kg).As a result there was not asignificant relationship betweenVO2maxandSaO2maxin these subjects (r = 0.138,F(1,10)reg= 0.193,p= 0.669). This remained the case when subjects were separated intogroups (LOS, r = 0.392,F(1 1O)reg= 0.909,p= 0.384; NOS, r = 0.592, F(1,10)reg =171.62 1,p= 0.293). These results are contrary to those reported previously by other authors(39, 42).Table 3VO2max, peak power output, andSaO2maxof subjects, group data.VO2maxVO2maxPeak PowerSaO2maxGROUP(L.minl) (mL.min1.kgi) (Watts) (%)LOS (n=7) 4.96 ± 0.26 66.8 ± 6.0 458 ±31 89.9 ± 0.9*NOS (n=5) 5.24 ± 0.34 66.4 ± 2.9 453 ±23 93.4 ± 0.4Values are means ± SD.SaO2mis lowest arterial oxygen saturation during themaximal cycle ergometer test. Value denoted by(*)is significantly lower than NOS (p <0.05).TheVO2THof subjects in the LOS group was higher than in the NOS groupalthough the difference did not reach statistical significance(tio= 1.79 1,p= 0.052)(Table 4). As a result LOS subjects had higher power outputs at their VO2mthan NOSsubjects(tio= 3.346,p= 0.004). Individual subject values for VO2are listed in Table9.Table 4VO2TH,and power output atVO2TH,group data.Group VO2TH VO2 Power at VO2TH(L.min’) (mL.min1.kg’) (Watts)LOS (n=7) 3.26 ± 0.14 46.0 ± 4.7 300 ± 26*NOS (n=5) 3.31±0.11 42.0± 1.8 259 ± 9Values are means ± SD. Value denoted by(*)is significantly higher than NOS(p<0.05).18EXERCISE WORKLOADSThe workloads for the chemoresponse trials were selected for two reasons: toextend the measurement of peripheral chemoresponsiveness during exercise to the highestintensities possible, hoping to elicit a hypoxemic response in the LOS subjects, and tochoose as the highest workload an intensity that would allow completion of the peripheralchemoresponse data collection. Thus, the highest workload that could be practically usedcorresponded to the power output atVO2TH.Since this was thought to average around 75% ofVO2maxin highly trained endurance cyclists the absolute workloads of 25 % and50 % ofVO2maxandVO2THwere chosen for the lower exercise levels. This would haveresulted in a roughly linear increase in power output from the very low power output tothe moderate power output exercise levels. The power outputs in both LOS and NOSincreased in a linear fashion from the very low power output to moderate power outputexercise levels(F(1,1o)lin= 846.883,p<0.001). However, not only was the rate ofincrease in power output in the NOS group lower than that of the LOS group (F(1,1 0)linmt= 14.756,p0.003), but the pattern of change in the increase in power output wasdifferent in the two groups(F(3,30)jnt= 12.858,p<0.001) (Figure 7). Specifically,subjects in both groups had similar power outputs at the very low power output and lowpower output exercise levels, but the NOS subjects had a substantially lower poweroutput at moderate power output exercise level than the LOS subjects (Table 5). This wasbecause theVO2THof LOS subjects was higher than NOS subjects and as a result themoderate power output workload, which was approximately the ventilatory thresholdworkload, was higher in the LOS subjects. Consequently, the two lower workloads wereeffectively absolute workloads while the highest workload was a relative workload. The19workloads derived by regression analysis for each subject, based upon their individualV02/poweroutput relationship, are listed in Table 10.Table 5 Power outputs maintained during chemoresponse tests, group data.Exercise LevelGROUP VLPO LPO MPOLOS(n=7) 70±11 194±19 272±28NOS(n=5) 77±13 199±10 232±11Units are Watts. Values are means ± SD. LOS, low oxygen saturation; NOS, normaloxygen saturation. VLPO, very low power output; LPO, low power output; MPO,moderate power output.20HYPERCAPNIC PERIPHERAL CHEMORESPONSEThe results of a typical hypercapnic peripheral chemoresponse trial can be seen inFigure la and Figure lb.Figure laPCO2 , P02 , and Vj of single hypercapnic peripheralchemoresponse.80-60-I I I I I1801401201001009080‘ 70> 605040I I I I I I-10 -5 0 5 10 15 20Time (sec)Subject, PT. Exercise level, MPO. Traces are offset and scaled data sampled at 20 Hz.Stimulus breath occurs at t=0.The response shown in Figure la and Figure lb is of an LOS subject at themoderate power output exercise intensity. The offset and scaled analyzer output traces are21shown in Figure la and the breath-by-breath values derived from the offset and scaledtraces are shown in Figure lb. In these figures the switching of inspired gas from roomair to the test gas occurs at t=tO. Prior to the stimulus breath are shown five breaths used tocalculate the pre-stimulus control Vj, and the pre-stimulus control PETCO2. Followingthe stimulus breath are shown the breaths occurring in the next twenty seconds, fromwhich the breath with the highest VJ was chosen as the response breath. The responsebreath in this particular trial occurred two breaths after the stimulus breath (less than 5seconds after the stimulus). This was the case in most trials with the response breathbeing the second or third breath after the stimulus breath, the time decreasing withincreasing exercise intensity. At lower exercise levels, or at rest, the Vj trace shows anobvious increase in slope following the stimulus breath. At the higher exercise levels thisslope increase is less visible in the raw Vj trace and is only seen in the breath-by-breathplot of Vj.22Figure lbPETCO2,ETO2,and Vj for a single hypercapnic peripheral120E116114‘ 1121 6014012010080chemoresponse.Exercise level, MPO. Subject, PT. Plot shows values extracted from raw data shown inFigure la. Vj calculated from Vj andRon a breath-by-breath basis.50 -9-400000O00000000— I I I I- 000 0-000 0 0 00 0 0 0 00— I I I I- 00 0,00 0• 00 ,000,000,I I I —-10 -5 0 5 10 15 20Time (sec)23EE-CCMCUFigure 2 Hypercapnic peripheral chemoresponse at various exercise intensities,group data.2-1.5:1—0.5 -0-I I IRest VLPO MPOValues are means ± SD. Open circles, LOS; Open squares, NOS. VLPO, very low poweroutput; LPO, low power output; MPO, moderate power output.The hypercapnic peripheral chemoresponse was significantly higher in NOS thanin LOS(F(1,10)group= 13.652,p0.004) and it increased significantly in both groups(F(3,30)trjals = 10.446,p<0.001) from rest to exercise (Figure 2). In addition, whilehypercapnic peripheral chemoresponse as a function of exercise level had a significantlinear component in both groups(F(1,1O)ljn = 13.413,p= 0.004), there was also asignificant plateauing of hypercapnic peripheral chemoresponse near the higherLPO24workloads in both groups(F(1,10)quad= 8.197,p= 0.017). Individual subject, averaged,hypercapnic chemoresponses are listed in Table 11.Figure 3Pre-C02responseSaO2at various exercise intensities, group data.Values are means ± SD. Open circles, LOS; Open squares, NOS.output; LPO, low power output; MPO, moderate power output.VLPO, very low powerPre-Stimulus Arterial Hemoglobin SaturationThe pre-stimulusSa02during hypercapnic peripheral chemoresponse trials wasnot different in LOS and NOS averaged across exercise levels(F(1,10)group= 0.019,p =0.894), however a significant fall inSaO2was seen in both groups(F(3,30)trials= 15.12,10099 -98 -97 -96 -95 -94Rest VLPO LPO MPO25p<0.0001). The fall inSaO2in the LOS group was larger than that of the NOS group(F(3,30)jnt= 3.2 13,p= 0.037) and it occurred mostly from the low power output tomoderate power output exercise intensities. The fall inSaO2in both groups was almostentirely a linear function of increasing exercise intensity (F(1, 10)lin= 34.299,p<0.000 1) with the more pronounced decrease in the LOS group at the moderate poweroutput exercise intensity(F(J.,10)quadmt= 6.596,p= 0.028) accounting for anydeviation from linearity.Hypercapnic Peripheral Chemoresponse and Pre-Stimulus Arterial HemoglobinSaturationThere was not a significant linear relationship between hypercapnic peripheralchemoresponsiveness and pre-stimulusSaO2in either group of subjects (LOS, r2 =0.115,p= 0.078; NOS, r2 =0.097,p= 0.181).26HYPEROXIC CHEMORESPONSEThe results of a typical hyperoxic peripheral chemoresponse trial can be seen inFigure 4a and Figure 4b.Figure 4aPCO2, P02,and Vj of a single hyperoxic peripheral chemoresponse.60-800 -;irTime (sec)Exercise level, MPO; subject, PT. Plot shows offset and scaled data sampled at 20 Hz. Vjcalculated in real-time from inspiratory flow rate.The response shown in Figure 5a and Figure 5b is of an LOS subject at themoderate power output exercise intensity. The offset and scaled analyzer output traces are27shown in Figure 5a and the breath-by-breath values derived from the offset and scaledtraces are shown in Figure 5b. In these figures the switching of inspired gas from roomair to the test gas occurs at t=O. Prior to the stimulus breath are shown a number ofbreaths from the 30 second pre-stimulus control period used to calculate the pre-stimuluscontrol Vi. Following the stimulus breath are shown the breaths occurring in the nextthirty seconds, from which the breath with the lowestViis chosen as the response breath.At lower exercise levels, or at rest, the Vj trace shows an obvious decrease in slopefollowing the stimulus breath. At the higher exercise levels this slope decrease is lessvisible in the V1 trace and is only seen in the breath-by-breath plot of VT.28Figure 4bPETCO2,ETO2,and V1 for a single hyperoxic peripheralchemoresponse.-E 55-E0:popp00Co°°°,°- 80060O-040O- 00 0-0200-0000000— I I I I140 -120 -pOp100-00 U0000001 0 00U-00060-I I I I I I-10 -5 0 5 10 15 20 25 30Time (see)Exercise level, MPO; subject, PT. Plot shows values extracted from offset and scaled datashown in Figure 5a. VI calculated from Vj andfRon a breath-by-breath basis.The hyperoxic chemoresponse was not significantly different either betweengroups or across exercise levels(F(1,10)group= 0.000,p= 0.988, F(3,30)trials = 2.152,p= 0.115), nor was there a significant interaction effect(F(3,30)int1.224,p= 0.3 18)(Figure 5). Individual subject, averaged, hyperoxic chemoresponses are listed in Table16.29Figure 5 Hyperoxic peripheral chemoresponse at various exercise intensities,group data.I I I IRest VLPO LPO MPOValues are means ± SD; open circles, LOS; open squares, NOS. VLPO, very low poweroutput; LPO, low power output; MPO, moderate power output.Pre-Stimulus Arterial Hemoglobin SaturationThe pre-stimulusSaO2during hyperoxic peripheral chemoresponse trials was notdifferent in LOS and NOS(F(1,10)group = 0.966,p= 0.349) averaged across exerciselevels, however a significant fall inSaO2was seen in both groups(F(3,30)trials= 18.769,p<.0001) (Figure 6). The fall inSaO2in the LOS group was not larger than that of theNOS group(F(3,30)int = 0.456, p = 0.7 15). The fall inSaO2in both groups was almost30 -25 -20 -15 -10 -530Cti)Centirely a linear function of increasing exercise intensity (F(1, 10)lin= 36.276,p<.0001)with a significant inverted sigmoid function(F(1,10)cubjc= 11.828,p= 0.006)accounting for any deviation from linearity. The pre-stimulusSaO2data for all subjectsduring hyperoxic peripheral chemoresponse trials are shown in Table 18.Figure 6Pre-02responseSaO2 atvarious exercise intensities, group data.100 -99 -98 -97 -96 -95 -94-Values are means ± SD; open circles, LOS; open squares, NOS. VLPO, very low poweroutput; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO31Hyperoxic Peripheral Chemoresponse and Pre-Stimulus Arterial HemoglobinSaturationThere was not a significant linear relationship between hyperoxic peripheralchemoresponse and pre-stimulusSaO2in either group of subjects (LOS, r2 = 0.015,p =0.540; NOS, r2 = 0.002,p= 0.846).32DISCUSSIONExercise induced arterial hypoxemia occurs in 40 to 50 % of highly trainedendurance athletes (39, 42) and has a detrimental effect on bothVO2maxand exerciseperformance (27, 42). The primary mechanisms suggested to be responsible for exerciseinduced arterial hypoxemia are veno-arterial shunt, diffusion disequilibrium secondary toincreased pulmonary transit time or pulmonary edema,VA:Qcheterogeneity (11, 19, 49)and relative hypoventilation (11, 30). A substantial amount of research has beenundertaken to elucidate the proportional importance of these mechanisms with somewhatcontradictory results. Much of the work investigating the relative importance ofhypoventilation in exercise induced arterial hypoxemia has involved only description orcomparison of the ventilatory responses of the subjects who do and do not demonstrateexercise induced arterial hypoxemia. No studies have examined the ventilatory controlmechanism directly and related any observations made to the incidence of exerciseinduced arterial hypoxemia. As a result some authors have excluded the hypoventilationmechanism as playing a major role in the development of exercise induced arterialhypoxemia (42).This study represents the first attempt to measure the peripheralchemoresponsiveness to hypercapnia and hyperoxia, at rest and during very light tomoderate exercise, in highly trained endurance athletes. The peripheralchemoresponsiveness of these highly trained endurance athletes is compared with resultsobtained from previous studies on trained and untrained individuals at rest and duringexercise. In addition, the relationship between peripheral chemoresponsiveness andexercise induced arterial hypoxemia is investigated by comparing the peripheral33chemoresponsiveness of highly trained endurance athletes who do and do notdemonstrate exercise induced arterial hypoxemia.RESTING PULMONARY FUNCTION, MAXIMAL EXERCISE TESTS, AND SUBJECTSELECTIONThe subjects in this study all demonstrated pulmonary function values withinpredicted normal ranges. Interestingly, it was common for subjects to be significantlyhigher than predicted onFEFmaxand MVV, while at the same time values for FEV1 andFEV1/FVC were only normal or slightly sub-normal. It is possible that the veryhighFEFmaxvalues, as high as 152 % of predicted, were simply related to the superiordevelopment of the thoracic musculature in these athletes. However, many subjects whohad very highFEFmaxvalues also had mid-expiratory flow rates that were substantiallybelow predicted. All subjects reached at least 90 % of their MVV during the incrementalcycle ergometer test.Values forVO2maxfor subjects in this study are comparable to data reported insome previous studies (20, 41) but are lower, on average, than those reported byDempsey et al. in 1984 (11). Contrary to previous reports (39), no relationship was foundbetweenVO2max and SaO2max in the subjects in this study either as awhole or whendivided into groups based on the presence of exercise induced arterial hypoxemia. Theincidence of exercise induced arterial hypoxemia in the highly trained endurance athleteswho completed this study was 58 %. This result overestimates the actual incidence ofexercise induced arterial hypoxemia observed in all subjects completing the maximalcycle ergometer test as most subjects completing that test were not studied furtherbecause: 1) they did not pass the pulmonary function criteria, 2) they did not pass the34VO2maxcriteria, or 3) they qualified for the LOS group but that group was already full.The incidence of exercise induced arterial hypoxemia in all subjects meeting theVO2maxcriteria for inclusion in the study was only 38 %, which agrees with lower incidencesreported (41). There were two subjects who did not meet theVO2maxcriteria that diddemonstrate exercise induced arterial hypoxemia. They would have been categorized asmoderately trained in the previous study (39) that reported no cases of exercise inducedarterial hypoxemia in subjects of similarVO2max.Although subjects in both groups demonstrated similar levels of aerobic power atmaximal exercise, the aerobic power at the ventilatory threshold of the LOS subjects washigher than the NOS subjects. As a result the moderate power output exercise intensitywas a relative work load while the very low power output and low power output exerciseintensities were both relative and absolute work loads in these subjects. Comparison ofthe exercise intensities attained in this study with similar studies of peripheralchemoresponsiveness (30, 32, 21, 22, 47, 51) confirm that not only were higher absolutepower outputs attained with these highly trained endurance athletes but higher relativeexercise intensities were attained as well.MEASUREMENT OF PERIPHERAL CHEMORESPONSESComputerization of the techniques of McClean et al. (1988) and Stockley (1978)made the collection and analysis of data during both resting and the three exercisedeterminations of peripheral chemoresponsiveness to hypercapnia and hyperoxiapossible. The automation and computerization of the data collection and analysis alsopermitted more reliable determination of valid responses than in the previous studies35since errors in the calculation of ‘1j from the slope of a chart recording of Vj wereeliminated.Hypercapnic Peripheral ChemoresponseAt the time that the study was undertaken the technique described by McClean etal. (1988) was the only technique that enabled measurement of the peripheralchemoresponse to hypercapnia during exercise. During the hypercapnic peripheralchemoresponse trials it was impossible to keep the subjects completely unaware of thebeginning ofC02breathing. At an inspired fraction of 13 % the level ofC02in the testgas was sufficiently high to be tasted by most subjects. A sharp increase in the peak andmean expiratory flow rates during the second and third breaths after the stimulus breath,in conjunction with reports from some subjects of a perceived need to exhale forcefully,confirmed that the administration of the stimulus was not completely blind. However, anumber of factors suggest that the influence of these cortical perceptions on theperipheral chemoresponse to hypercapnia, although unknown, is probably small: 1) theperipheral chemoresponse to hypercapnia was determined from‘1jand notVEand thecortical perceptions would not be likely to increase the inspiratory drive, 2) increases inthe inspiratory flow rate were much smaller in proportion to those in expiratory flow rate,and 3) subjects were naive about which test gas they were receiving in what doses duringany given chemoresponse trial.The use ofPETCO2to quantify the stimulus delivered to the carotid bodychemoreceptor is based upon the assumption that there is a small and constant arterial toalveolarC02difference. This has not been documented in the literature and could be aconfounding factor, especially in subjects who likely suffer from some diffusionlimitation of02.The fact that the exercise intensities used did not elicit exercise induced36arterial hypoxemia in any of the subjects suggests that if a significant arterial to alveolarC02difference can exist it was not likely to be present in these studies.Hyperoxic Peripheral ChemoresponseUnlike during the hypercapnic peripheral chemoresponse tests, subjects wereunaware when measurement of hyperoxic peripheral chemoresponsiveness took place.The test gas, although it was dry and was inspired for between 5 and 20 seconds, did notincrease or decrease any sensations of dyspnea at rest or during exercise. Also, with theaddition of on-line display of inspired and expiredP02andPCO2 tothe method ofStockley (47) evidence of hypoventilation in response to the hyperoxic stimulus couldeasily be detected facilitating the identification of valid responses.The lack of isocapnic conditions during the hyperoxic chemoresponse tests resultsin a variable underestimation of the decrease in ventilation in response to hyperoxia atrest and during exercise. Thus caution is required when comparing data from this studywith other studies where isocapnia was maintained.EFFECT OF TRAINING STATUS ON PERIPHERAL CHEMORESPONSIVENESSThe hypercapnic peripheral chemoresponsiveness of the highly trained enduranceathletes in this study was higher at rest (mean ± SD, 0.54 ± 0.30 L•minl.mmHg1)thanvalues reported for healthy untrained males (mean ± SD, 0.38 ± 0.14 L.min4.mmHg(32). This is in agreement with data indicating that peripheral chemosensitivity tohypercapnia is higher in trained versus untrained individuals (25) but is contrary toreports of others documenting a decrease in chemosensitivity to hypercapnia inindividuals following prolonged exercise training (35). The results of the second studywere obtained using aC02rebreathing technique in which both central and peripheral37chemoreceptive mechanisms are active. The results of the study by Miyamura and Ishida(35) are not in conflict with data from this study if, while peripheral chemoresponsivenessincreases with training, central chemoresponsiveness decreases with training, the netresult being an overall increase in chemoresponsiveness to hypercapnia.The resting hyperoxic peripheral chemoresponsiveness of subjects in this study(mean ± SD, 20.0 ± 8.1 %) is similar to previously reported values in a group of athletesof similar aerobic capacity (mean ± SD, 22 ± 2 %) (31) and is slightly higher than a groupof healthy, normal subjects (mean ± SE (n=35), 16.2 ± 2.6 %) (47). The differencebetween the data in this study and the data of Stockley (1978) is not significant (t45 =0.8 13,p= 0.2 10) and would suggest that there is no difference between trained anduntrained individuals in the resting contribution to normoxic of hypoxic sensitivity.This seems intuitive because at rest both trained and untrained individuals have the sameSaO2anda°2.EFFECT OF EXERCISE ON PERIPHERAL CHEMORESPONSIVENESSSubjects in this study demonstrated an increase in peripheralchemoresponsiveness to hypercapnia associated with exercise supporting previous reportsof an augmentation of peripheral chemoresponsiveness to hypercapnia during exercise(21, 22). Further, this data indicates that on transition from very light to moderateexercise the rate of increase in peripheral chemosensitivity decreases with increasingexercise intensity. It would seem that as exercise intensity increases approaching andexceeding the ventilatory threshold peripheral chemosensitivity to hypercapnia plays aless important role in the development of exercise hyperpnea.38As others have reported (31, 47) this study did not demonstrate a statisticallysignificant change in hyperoxic peripheral chemoresponsiveness with increasing exerciseintensity. A previous report (51) indicated that hyperoxic chemoresponsiveness wasaugmented by exercise, however, the observations in that study were based on anextrapolation ofVE toan infinitely higha°2and are questionable in light of the morerecent work by the same group (31).PERIPHERAL CHEMORESPONSIVENESS AND EXERCiSE INDUCED ARTERIALHYPOXEMIAIn the present study the peripheral chemosensitivity to hypercapnia was lower insubjects who demonstrated exercise induced arterial hypoxemia. This effect cannot beattributed to a difference in trained status because subjects in both groups had similarVO2max.Differences between the LOS and NOS subjects at the two low exerciseintensities are not likely to be simple exercise effects because the subjects exercised at thesame relative and absolute metabolic rate. At the moderate power output exercise levelthe LOS subjects were exercising at a higher metabolic rate relative toVO2max,althoughboth groups were exercising at theirVO2,and the difference in hypercapnic peripheralchemoresponsiveness present at the low power output exercise intensities wasmaintained. The pattern of change in peripheral chemoresponsiveness to hypercapnia as afunction of increasing exercise intensity was the same in both LOS and NOS subjects.There seems to be a relationship between exercise induced arterial hypoxemia and thesensitivity of the peripheral chemoreceptors to hypercapnia.The absence of a difference in hyperoxic peripheral chemoresponsivenessbetween LOS and NOS subjects indicates that, at the exercise intensities studied, there is39no relationship between hyperoxic peripheral chemoresponsiveness and exercise inducedarterial hypoxemia. This is probably due to the fact that the moderate power outputexercise intensity was not high enough to induce arterial hypoxemia in the LOS subjectsand the baseline activation of the peripheral chemoreceptors to hypoxia was not differentfrom NOS subjects. A relationship between hyperoxic peripheral chemoresponsivenessand exercise induced arterial hypoxemia may become apparent at higher workloads butthis study has no data to support this hypothesis.PERIPHERAL CHEMORESPONSIVENESS AND PRE-STIMULUS ARTERIALHEMOGLOBIN SATURATIONNo evidence was found to support the hypothesis that peripheralchemoresponsiveness was related to exerciseSaO2.Although there was a difference inperipheral chemoresponsiveness to hypercapnia in LOS and NOS subjects there was nodifference between groups in pre-stimulusSaO2averaged across exercise intensities andperipheral chemoresponsiveness was not different in LOS and NOS subjects. Again,although subjects in the LOS group had lower pre-stimulusSa°2values at the moderatepower output exercise intensity, they did not demonstrate exercise induced arterialhypoxemia and this fact may have prevented the detection of a relationship between thesevariables.IMPLICATIONS FOR VENTILATOR Y CONTROLThe ventilatory control mechanism suggested by some authors (48) consists of afeed-forward component that directs the initial response of the ventilatory system toincreasedCO2load, and a feed-back component that is responsible for correcting errorsin the feed-forward portion of the control mechanism. Other authors have expanded this40model and suggest that the role of the peripheral chemoreceptors is to stabilizeventilation, effectively acting as a brake to the initial hyperventilation that accompaniesthe onset of exercise (3). Data from this study suggest that highly trained enduranceathletes, who have higher hypercapnic peripheral chemoresponsiveness than untrainedindividuals, control this initial increase in ventilation more effectively, maintainingtighter control onPaCO2, pHaanda°2.It would seem that highly trained enduranceathletes who demonstrate exercise induced arterial hypoxemia are less able to controltheir ventilation, they have lower hypercapnic peripheral chemoresponsiveness, undersimilar conditions and are more likely to experience wider variations inPaCO2,P11aanda°2than individuals of similar aerobic capacity who do not demonstrate exerciseinduced arterial hypoxemia.The mechanism of these changes in athletes with exercise induced arterialhypoxemia is unknown. It is possible that there are differences between these two groupsof subjects in the physiologic response of the carotid bodies themselves, in the integrationof afferent signals from the carotid bodies with other feed-back within the respiratorycenter or in the subsequent expression of this feed-back as a change in ventilation.SUMMARYThe results of this study provide information which may help explain variations inthe ventilatory response to exercise in athletes. Additionally, data from this study suggesta role of altered ventilatory control in highly trained endurance athletes who do and donot demonstrate exercise induced arterial hypoxemia. Further study is required toascertain the specific causes of the reduced hypercapnic peripheral chemoresponsiveness41in highly trained endurance athletes who demonstrate exercise induced arterialhypoxemia.42BIBLIOGRAPHY1. Aggarwal D., H.J. Milhom and L.Y. Lee. Role of the carotid chemoreceptors inthe hyperpnea of exercise in the cat. Respir Physiol. 26(2): 147-55, 1976.2. Brooks G.A. and T.D. Fahey. Exercise physiology: human bioenergetics and itsapplications. New York: Wiley, 1984.3. Brown D.R., H.V. Forster, A.S. Greene and T.F. Lowry. Breathing periodicity inintact and carotid body-denervated ponies during normoxia and chronic hypoxia.JAppi Physiol. 74(3):1073-82, 1993.4. Burger R.E., J.A. Estavillo, P. Kumar, et al. Effects of potassium, oxygen andcarbon dioxide on the steady-state discharge of cat carotid body chemoreceptors. JPhysiol (Lond). 401(519):519-31, 1988.5. Byrne-Quinn E., J.V. Weil, I.E. Sodal, et al. Ventilatory control in the athlete. JAppiPhysiol. 30(1):91-8, 1971.6. Cheng B., H. Kuipers, A.C. Snyder, et al. A new approach for the determinationof ventilatory and lactate thresholds. mt. J. Sports Med. 13(7):518-522, 1992.7. Chonan T., W. Hida, Y. Kikuchi, et al. Effects of elastic loading and exercise onpulmonary gas exchange in dogs. Tohoku JExp Med. 164(2):157-67, 1991.8. Coates G., H. O’Brodovich, A.L. Jefferies and G.W. Gray. Effects of exercise onlung lymph flow in sheep and goats during normoxia and hypoxia. J Clin Invest.74(1):133-41, 1984.9. Dempsey J.A. J.B. Wolffe memorial lecture. Is the lung built for exercise? MedSci Sports Exerc. 18(2):143-55, 1986.10. Dempsey J.A. and R.F. Fregosi. Adaptability of the pulmonary system tochanging metabolic requirements. Am J Cardiol. 55(10):59D-67D, 1985.11. Dempsey l.A., P.G. Hanson and K.S. Henderson. Exercise-induced arterialhypoxaemia in healthy human subjects at sea level. J Physiol (Lond).355(161):161-75, 1984.4312. Dempsey J.A. and B.D. Johnson. Demand vs. capacity in the healthy pulmonarysystem. Schweiz Z Sportmed. 40(2):55-64, 1992.13. Dempsey J.A., D.A. Pelligrino, D. Aggarwal and E.J. Olson. The brain’s role inexercise hyperpnea. Med Sd Sports. 1 1(2):213-20, 1979.14. Grimby G., B. Saltin and L. Withelmsen. Pulmonary flow-volume and pressure-volume relationship during submaximal and maximal exercise in young well-trained men. Bull Physiopathol Respir (Nancy). 7(1): 157-72, 1971.15. Hammond M.D., G.E. Gale, K.S. Kapitan, et al. Pulmonary gas exchange inhumans during exercise at sea level. JAppi Physiol. 60(5): 1590-8, 1986.16. Harrop G.A. The oxygen and carbon dioxide content of arterial and of venousblood in normal individuals and in patients with anemia and heart disease. J. Exp.Med. 30:241-257, 1919.17. Honda Y., S. Myojo, S. Hasegawa, et al. Decreased exercise hyperpnea in patientswith bilateral carotid chemoreceptor resection. JAppi Physiol. 46(5):908-12,1979.18. Honda Y., S. Watanabe, I. Hashizume, et al. Hypoxic chemosensitivity inasthmatic patients two decades after carotid body resection. JAppi Physiol.46(4):632-8, 1979.19. Hopkins S.R. Pulmonary diffusion limitation,VA:Qmismatch and pulmonarytransit time in highly trained athletes during maximal exercise. [Ph.D.]. Universityof British Columbia, Canada, 1993.20. Hopkins S.R. and D.C. McKenzie. Hypoxic ventilatory response and arterialdesaturation during heavy work. JAppi Physiol. 67(3):1 119-24, 1989.21. Jacobi M.S., C.P. Patil and K.B. Saunders. The transient ventilatory response tocarbon dioxide at rest and in exercise in man. Respir Physiol. 77(2):225-37, 1989.22. Jacobi M.S., C.P. Patil and K.B. Saunders. Transient, steady-state and rebreathingresponses to carbon dioxide in man, at rest and during light exercise. J Physiol(Lond). 41 1(85):85-96, 1989.23. Johnson A. and J.B. Lofstrom. A new method for studying the ventilatoryresponse in patients. Acta Anaesthesiol Scand. 34(6):440-6, 1990.4424. Johnson B.D., K.W. Saupe and J.A. Dempsey. Mechanical constraints on exercisehyperpnea in endurance athletes. JAppi Physiol. 73(3):874-86, 1992.25. Kelley M.A., M.D. Laufe, R.P. Miliman and D.D. Peterson. Ventilatory responseto hypercapnia before and after athletic training. Respir Physiol. 55(3):393-400,1984.26. Kelley M.A., R.J. Panettieri and A.V. Krupinski. Resting single-breath diffusingcapacity as a screening test for exercise-induced hypoxemia. Am J Med.80(5):807- 12, 1986.27. Koskolou M.D. and D.C. McKenzie. Arterial hypoxemia and performance duringintense exercise. Eur. J. Appi. Physiol. [in press], 1993.28. Lugliani R., B.J. Whipp, C. Seard and K. Wasserman. Effect of bilateral carotid-body resection on ventilatory control at rest and during exercise in man. N Engi JMed. 285(20):1105-11, 1971.29. Mahier D.A., E.D. Moritz and J. Loke. Ventilatory responses at rest and duringexercise in marathon runners. JAppi Physiol. 52(2):388-92, 1982.30. Martin B.J., K.E. Sparks, C.W. Zwillich and J.V. Weil. Low exercise ventilationin endurance athletes. MedSci Sports. 11(2):181-5, 1979.31. Martin B.J., J.V. Weil, K.E. Sparks, et al. Exercise ventilation correlatespositively with ventilatory chemoresponsiveness. JAppi Physiol. 45(4):557-64,1978.32. McClean P.A., E.A. Phillipson, D. Martinez and N. Zamel. Single breath of C02as a clinical test of the peripheral chemoreflex. JAppi Physiol. 64(1):84-9, 1988.33. Miles D.S., M.H. Cox, J.P. Bomze and R.W. Gotshall. Acute recovery profile oflung volumes and function after running 5 miles. J Sports Med Phys Fitness.31(2):243-8, 1991.34. Mines A.H. Respiratory physiology. (2nd ed.) New York: Raven Press, 1986.35. Miyamura M. and K. Ishida. Adaptive changes in hypercapnic ventilatoryresponse during training and detraining. Eur JAppi Physiol. 60(5):353-9, 1990.4536. Pan L.G., H.V. Forster, G.E. Bisgard, et al. Role of carotid chemoreceptorsandpulmonary vagal afferents during helium-oxygen breathing in ponies. JAppiPhysiol. 62(3):1020-7, 1987.37. Paterson D.J. Potassium and ventilation in exercise. JAppi Physiol.72(3):81 1-20,1992.38. Paterson D.J., P.A. Robbins and J. Conway. Changes in arterial plasmapotassiumand ventilation during exercise in man. Respir Physiol. 78(3):323-30, 1989.39. Powers S.K., S. Dodd, J. Lawler, et al. Incidence of exercise inducedhypoxemiain elite endurance athletes at sea level. Eur JAppi Physiol. 58(3):298-302, 1988.40. Powers S.K., S. Dodd, J. Woodyard, et al. Haemoglobin saturation duringincremental arm and leg exercise. Br J Sports Med. 18(3):212-6, 1984.41. Powers S.K., D. Martin, M. Cicale, et al. Exercise-induced hypoxemiain athletes:role of inadequate hyperventilation. Eur JAppi PhysioL 65(1):37-42, 1992.42. Powers S.K., D. Martin and S. Dodd. Exercise-induced hypoxaemia in eliteendurance athletes. Incidence, causes and impact on VO2max. Sports Med.16(1):14-22, 1993.43. Powers S.K. and J. Williams. Exercise-induced hypoxaemia inhighly trainedathletes. Sports Med. 4(1):46-53, 1987.44. Rasmussen B.S., P. Elkjaer and B. Juhi. Impaired pulmonary and cardiacfunctionafter maximal exercise. J Sports Sci. 6(3):219-28, 1988.45. Schaffartzik W., D.C. Poole, T. Derion, et al. VA:Q distribution during heavyexercise and recovery in humans: implications for pulmonary edema. JAppiPhysiol. 72(5):1657-67, 1992.46. Shaw R.A., S.A. Schonfeld and M.E. Whitcomb. Progressive and transienthypoxic ventilatory drive tests in healthy subjects. Am Rev Respir Dis. 126(1):37-40, 1982.47. Stockley R.A. The contribution of the reflex hypoxic drive to the hyperpnoea ofexercise. Respir Physiol. 35(1) :79-87, 1978.4648. Swanson G.D. Overview of ventilatory control during exercise. Med Sci Sports.1 1(2):221-6, 1979.49. Wagner P.D. Ventilation-perfusion matching during exercise. Chest. :192S-198S,1992.50. Wasserman K., B.J. Whipp, S.N. Koyal and M.G. Cleary. Effect of carotid bodyresection on ventilatory and acid-base control during exercise. JAppi Physiol.39(3):354-8, 1975.51. Weil J.V., E. Byrne-Quinn, I.E. Sodal, et al. Augmentation of chemosensitivityduring mild exercise in normal man. JAppi Physiol. 33(6):813-9, 1972.52. West J.B. and 0. Mathieu-Costello. Stress failure of pulmonary capillaries in theintensive care setting. Schweiz Med Wochenschr. 122(20):751-7, 1992.53. West J.B., K. Tsukimoto, 0. Mathieu-Costello and R. Prediletto. Stress failure inpulmonary capillaries. JAppi Physiol. 70(4):1731-42, 1991.54. West J.B. Respiratory physiology--the essentials. (3rd ed.) Baltimore: Williams &Wilkins, 1985:x, 183 p. : ill.;23 cm.55. Whipp B.J. The hyperpnea of dynamic muscular exercise. Exerc Sport Sd Rev.5(295):295-311, 1977.56. Whipp B.J. and K. Wasserman. Carotid bodies and ventilatory control dynamicsin man. Fed Proc. 39(9):2668-73, 1980.47APPENDIX AREVIEW OF LITERATURETHE PULMONARY SYSTEM: A LIMITING FACTOR IN EXERCISEPERFORMANCEThere are a number of factors which are commonly considered to limit athleticperformance in humans. However, which factor or combination of factors are mostimportant remains controversial. The controversy revolves around the likelihood that asone moves along the continuum of aerobic power from untrained normal individuals(VO2maxof approximately 40 mL.min1.kg)to highly trained endurance athletes(VO2maxof approximately 60 to 75 mLmin1.kg)which of these potentially limitingfactors becomes the most important for any specific sub-group of the populationundoubtedly varies.NORMAL, HEALTHY, UNTRAINED INDIVIDUALSIn healthy sedentary individuals, exercising at sea-level, the pulmonary systemshows near perfect regulation of alveolar gases, distribution of alveolar ventilation (VA)and perfusion(c)and diffusion equilibrium in the lung at rest and during exercise. Thecapacity of the pulmonary systems of these individuals to extract oxygen from theatmosphere exceeds that of their cardiovascular and metabolic systems to deliver and usethat oxygen (12). Maximal aerobic power(‘O2max)and therefore exercise performanceare limited by maximal stroke volume, cardiac output, skeletal muscle vascularity and/orthe oxidative capacity of the skeletal muscles (9) in these individuals . At sea level thepulmonary system is able to meet the demands placed on it for oxygen(02)extractionand carbon dioxide(C02)elimination during heavy exercise through a number of48mechanisms. Firstly, the hyperpnea of exercise ensures the maintenance of alveolaroxygen pressure(PAO2)above 110 mmHg maintaining arterial oxygen pressure(PaO2)near resting values (100 mmflg) and ensuring adequate elimination ofC02(2).Secondly, pulmonary capillary blood volume increases in a linear fashion up to 3 timesits resting value with the 4 to 5-fold increase in pulmonary blood flow. This maintains redblood cell transit times necessary for equilibration of blood in the pulmonary capillarieswith alveolar gas, a relatively uniform distribution of pulmonary blood flow andexpansion of the alveolar-capillary surface area, and relatively low pulmonary vascularresistance (9). Finally, the lymphatic system is capable of adequately draining thepulmonary interstitial space of pulmonary extra vascular water. This prevents thelengthening of alveolar-capillary diffusion distances (8).HIGHLY TRAiNED, END URANCE ATHLETESIn highly trained endurance athletes the physiological adaptations of thecardiovascular system and of the oxidative capacities of the skeletal muscles,accompanied by the limited scope for adaptation in the pulmonary system, may result inthe pulmonary system actually becoming the limiting factor in exercise performance (9,43). In fact, many highly trained endurance athletes demonstrate a significant andreproducible decrease ina°2or arterial hemoglobin saturation(SaO2)during moderateto intense exercise. It is not entirely clear what level ofSaO2is required to maintainperformance however some data suggests that measurable reductions in exerciseperformance begin to occur atSa°2’5of less than 90% (27).49EXERCISE INDUCED ARTERIAL HYPOXEMIAThe reduction rna°2accompanying exercise was first reported by Harrop in1919 (16) who observed a decrease inSaO2from 95.6% at rest to 85.5% immediatelyfollowing fifteen minutes of brisk exercise (heart rate = 140 beats•min’,R= 30breaths•min’). More recent reports of reductions ina°2and/orSaO2,a phenomenonnow commonly referred to as exercise induced arterial hypoxemia, are numerous (11, 20,39, 42, 43).DEFINITION OF EXERCISE INDUCED ARTERIAL HYPOXEMIAExercise induced arterial hypoxemia, although previously observed, was firstdefined by Powers et al. in 1988 (39) as a decrease inSaO2from a resting value ofapproximately 97% to a value less than or equal 91% during exercise. This definition of acritical level ofSaO2,allowed the authors to document the incidence and reproducibilityof exercise induced arterial hypoxemia in highly trained endurance athletes (52%, r=0.95,p<.O5). More recently the same authors have modified their definition of exercise inducedarterial hypoxemia based on direct measurement ofa°2in conjunction with measures ofSaO2(41). The incidence of exercise induced arterial hypoxemia is 40 % in highlytrained endurance athletes using the new definition of a decrease in SaO2 of 4 % belowrestingSaO2.CA USES OF EXERCISE IND UCED ARTERIAL HYPOXEMIAAlthough a significant amount of research is being conducted in an attempt toelucidate the mechanism underlying exercise induced arterial hypoxemia (7, 11, 20, 23,24, 26, 39-43, 49), the causes of exercise induced arterial hypoxemia are not clearlyunderstood and remain a topic of debate. A number of mechanisms have been identified50as likely contributors to the development of exercise induced arterial hypoxemia howevertheir relative importance is currently unknown.Ventilation-Perfusion Heterogeneity and Veno-Arterial ShuntThe distribution of alveolar ventilation (\o(V,\s())A)and pulmonary capillaryperfusion(Qc)orVA:Cthroughout the lung is not uniform and actually becomes lessuniform during heavy exercise (49). This‘/A:Cheterogeneity, accompanied by a left toright heart shunt, explains most of the widening (2.5- to 3-fold) of the alveolar to arterialP02 difference ((A-a)D02)in healthy, sedentary individuals (9). It should be noted thatalthough there is a widening of the(A-a)D02in healthy, sedentary individuals, theymaintain theira°2to within roughly 10 mmHg of resting values and therefore do notexhibit exercise induced arterial hypoxemia.Diffusion disequilibriumAbnormal widening of the (A-a)D02 or diffusion disequilibrium has beensuggested to be the other major cause of exercise induced arterial hypoxemia (15, 26, 42,43, 49). The additional widening of the(A-a)D02 beyond that seen in healthy,sedentaryindividuals is thought to be due to a decrease in red blood cell transit time below the levelnecessary for full equilibration of pulmonary blood with alveolar gas or a decrease in thediffusing capacity of the pulmonary system (12, 42, 49). The cause of this reduced transittime is an increase in pulmonary blood flow beyond the point at which pulmonarycapillary blood volume has reached its maximum morphological limits (9, 19). Anincreased diffusion distance, due to extremely high pulmonary capillary pressures andincreased plasma leakage into the interstitial spaces referred to as pulmonary edema, mayalso contribute further to the widening(A-a)D02due to (7, 9, 49).51Relative HypoventilationInadequate hyperventilation has been suggested as one of the major causes ofexercise induced arterial hypoxemia in highly trained endurance athletes. The lack of anappropriate hyperventilatory response to exercise causes an increased widening of the(A-a)D02 resulting in a drop ina°2(9, 49).Mechanical limitation of ventilation and respiratory muscle fatigueThe difference in maximal exercise ventilation corrected for metabolic ratebetween healthy, untrained individuals(/J/:/02= 19.0 ± 0.4,V]/Vco222.6 ± 0.7)and highly trained endurance athletes(VE/V02= 15.7 ± 0.2,VE/VCO2= 19.0 ± 0.7) atlow exercise intensities suggests that mechanical limitation of ventilation or respiratorymuscle fatigue might explain, at least in part, the lack of an appropriate hyperventilatoryresponse in athletes who develop exercise induced arterial hypoxemia (30). Thehypothesis regarding mechanical limitation of ventilation is supported by experiments inwhich the mechanical work of breathing was reduced in subjects breathing a mixture ofhelium (He) and02which resulted in an immediate and significant hyperventilation (11)as well as the observation that maximal volitional expiratory flow:volume limits may beexceeded at very high levels of exercise (14).Whatever the relative contributions of these mechanisms to the development ofexercise induced arterial hypoxemia in the highly trained, endurance athlete, it seemsclear that factors influencing the control of the ventilatory response to exercise must alsobe considered as playing a role in the development of exercise induced arterialhypoxemia.52VENTILATORY CONTROLDespite abundant scientific inquiry, the topic of ventilatory control is not fullyunderstood and remains controversial. Until recently (20), ventilatory control has notbeen directly studied as a potential contributing factor in exercise induced arterialhypoxemia although a number of investigations of exercise induced arterial hypoxemiahave included a description of the ventilatory response accompanying the exercisestimulusCONTROL MECHANISMSThe current body of data has led some researchers (48) to hypothesize a system ofventilatory regulation during exercise based upon the fact that in normal individuals,exercising at moderate intensities (ranging from rest to the anaerobic threshold) PaCO2,pH and P02 are regulated at essentially their resting values. One hypothesized structurefor the ventilatory controller combines feed-forward and feed-back mechanisms (48). Theactual physiological mechanisms that could contribute to the feed-forward response aresummarized below:1. arterial CO2 (pH) oscillations sensed by the carotid body,2. central neural stimulus to the respiratory center from the motor cortex,3. afferent input from exercising muscles to the respiratory center,4. venous return to the heart sensed by some unknown mechanism,5. a sensor responding to pulmonary blood flow, mixed venous C02,CO2 flux to thelung, or some other unknown humoral substance,6. an intrapulmonary chemoreceptor, also not yet identified, sensing mixed venousblood pH.53This feed-forward response would yield an exercise ventilation proportional to the CO2production, with any errors in this feed-forward response being corrected by the feedback response of the arterial chemoreceptors toPaCO2 (48).CENTRAL CONTROL OF VENTILATIONThe central nervous system plays an important role in the regulation of theventilatory response to exercise. The regulation is accomplished through the integrationof sensory input via three basic types of mechanisms; non-chemical input from suprapontine areas of the brain, chemical input from chemoreceptive regions of the medulla,and spinal motor neuron cross innervation (13).Supra-pontine ControlThe supra-pontine portion of the ventilatory control equation, plays an importantrole in the ventilatory response of humans during exercise. Its effect is obtainedthroughthe integration of three basic influences. The traditional voluntary influence of the highercenters has obvious and significant application to the control of breathing in athleticendeavors where the ventilatory musculature in also involved in the activity directly (13).The second influence of the higher centers is effected through the sensation of ventilatoryeffort. A number of receptors including, but not limited to, pulmonary and airway stretchreceptors, intercostal muscle spindles, and costo-vertebral joint receptors are located insuch a way that they provide feedback to the motor cortex regarding lung volume, ribcage distortion, upper airway resistance, development of muscle tension, and otherimportant determinants of ventilatory sensation (13). The third category of contributionsfrom higher levels of the CNS involves more direct influences of supra-pontinemechanisms on medullary output and the interaction of these inputs with the more54traditional chemoreceptor inputs to the same areas. These mechanisms are generallyclassified into two groups, cortical inputs (inhibitory) and diencephalic inputs (facilatory)however the activation of these influences is not clearly understood (13).Central ChemoreceptionThe majority of the resting ventilatory response of normal individuals at sea levelis mediated by the central chemoreceptors. Medullary chemoreceptive cells on the ventralsurface of the medulla are sensitive to changes in the pH of the medullary interstitial fluidand cerebrospinal fluid with decreases in pH stimulating ventilation (2). Due to thepresence of the blood-brain barrier, which is more permeable to CO2 than to H ions,these receptors are not influenced by arterial pH but rather byPaCO2.These receptors areresponsible for maintaining a ventilatory response adequate to maintain a resting PaCO2of about 40 to 45 mmHg.Spinal Motor Neuron InteractionThe influence of the spinal motor neurons on the control of ventilation is notclearly understood. The motor neurons do not generate rhythmic discharge but do showreciprocal inhibition. They receive input from a variety of sources includingmechanoreceptor afferents, phasic command signals and tonic decending influences.They could, in combination with the above mentioned mechanisms, conceivably generatean appropriate ventilatory response to exercise without inputs arising from outside thecentral nervous system (13).PERIPHERAL CONTROL OF VENTILATIONPeripheral influences on the ventilatory response to exercise come from primarilytwo sources, peripheral mechanoreceptors and peripheral chemoreceptors. The integration55of afferent signals from these receptors with others mentioned previously allow the closeregulation ofa°2,PaCO2andpHaobserved in normal individuals at sea level.Peripheral MechanoreceptorsA number of receptors have been identified that could play a role in the regulationof the ventilatory response to exercise. These receptors, some of which are skeletalmuscle spindles, Golgi tendon organs, and skeletal joint proprioceptors, send afferentsignals to the sensory cortex (2) and are thought to play a significant role in theneurogenic or phase 1 portion of the ventilatory response to exercise (55).Peripheral ChemoreceptorsPeripheral chemoreceptors are of basically two types, aortic body chemoreceptorsand carotid body chemoreceptors. Both receptors respond to changes in blood gastensions however their relative importance in the control of ventilation is quite different.Aortic Body ChenwreceptorsThe aortic bodies are located around the aortic arch and between the arch and thepulmonary artery and are therefore appropriately located to respond to changes in thechemical composition of arterial blood. They are stimulated by a decreased meana°2and by an increased meanPaCO2,the response to02being greater (34). Although theaortic body chemoreceptors are involved in respiratory regulation, the role of the carotidbody chemoreceptors is much greater.Carotid Body ChemoreceptorsThe carotid body chemoreceptors are located in the carotid bodies which arefound at the bifurcation of the common carotid arteries into the internal and externalcarotid arteries. Like the aortic bodies, they are stimulated by decreasingPa°2and byincreasingPaCO2,however they are also stimulated by a decrease in pHa (34) and56increasing blood potassium ([K]) levels (38). In addition, the carotid bodies respondquickly enough to be sensitive to the within-breath variations of pHa caused by withinbreath fluctuations inPACO2,possibly providing one of the more important signals usedto match ventilation to metabolic rate during exercise (34). That the carotid bodiesplayan important role in the control of exercise ventilation is clear (4, 17, 28, 36, 50).Inaddition, experiments involving 100%02breathing during exercise have indicated thatthe maximal reduction in ventilation during administration of 100% 02 (i.e., ‘silencingtthe carotid bodies) is greater during exercise than at rest suggesting enhancement ofcarotid body drive induced by exercise (55).The importance of the chemoreceptive drive to breath is clear, playing arole inboth the feed-forward and feed-back portions of the respiratory control mechanism,itsbriskness has profound effects on the ability to maintain homeostasis during moderate tointense exercise. Variations in carotid body drives have been suggested to play a role inthe control of exercise ventilation. Carotid drives have been reported toincrease (25) anddecrease (35) significantly with training and during exercise compared to rest (21, 22,51). For example, carotid body chemoresponsiveness to hypercapnia has been reported torange from 0.38 ± 0.14 L•min1.mmHg in healthy untrained individuals (32) to ashighas 2.15 ± 0.62 L•min1•mmHg’ in highly trained endurance athletes atrest (25). Thus,individual differences in variations in carotid body drives during exercise could explainsome of the variation in the ventilatory response to exercise and development of exerciseinduced arterial hypoxemia in highly trained endurance athletes.57(yrs)283127242446282533212022(cm)177183190184174179184178184191189174182± 6(kg)65.784.076.381.469.879.778.674.779.980.880.867.276.6 ± 6.0(m2)1.812.062.032.041.841.982.011.932.032.092.081.801.98 ±0.11APPENDIX BTABLESTable 6 Age, height, mass, and body surface area, individual subject data.AGE HEIGHT MASS BSASUBJECTTMPTDBGAHTBTRRSFMFDHMEJFMEAN±SD 27±758Table 7 Pulmonary function, individual subject data.FVCFEV1 FEVi/FVC FEFmMVVSUBJECT (L) (L) (L•sec’) (L•min)(% Pred.) (% Pred.) (% Pred.) (% Pred.)TM 5.21 3.72 0.71 9.40 18398 84 114 105PT 6.99 4.80 0.69 11.58 211122 101 133 115DB 6.31 5.39 0.85 12.07 20398 101 138 102GA 5.71 5.27 0.92 13.48 22195 105 152 114HT 6.20 4.25 0.69 9.70 183120 98 116 102BT 5.13 3.99 0.78 11.90 203104 99 146 128RR 6.39 4.62 0.72 8.10 184107 94 93 97SF 6.18 4.07 0.66 9.18 158112 88 107 85MF 5.91 4.91 0.83 9.25 192102 102 107 105DH 7.31 5.74 0.79 11.36 231109 103 125 111ME 7.20 5.40 0.75 9.55 177107 98 105 86IF 4.77 4.15 0.87 10.65 17892 95 128 98MEAN± 6.11±0.82 4.69±0.66 0.77±0.08 10.52±1.57 194±21SD 105±9 97±6 122±18 97±1159Table 8VO2max,peak power output, andSaO2max,individual subject data.VO2max VO2maxPeak PowerSaO2maxSUBJECT(L•min1) (mL•min1.kg) (Watts) (%)TM 4.92 74.9 470 90.0PT 5.25 62.5 490 90.4DB 5.21 68.3 479 90.0GA 4.85 59.6 430 91.0HT 5.14 73.6 470 89.6BT 4.88 61.5 465 90.2RR 5.15 65.5 435 93.4SF 4.88 65.3 435 93.3MF 5.00 62.6 455 93.2DH 5.65 69.9 450 94.2ME 5.54 68.6 490 93.1IF 4.50 67.0 400 88.0MEAN ± SD 5.08 ± 0.32 66.6 ± 4.7 456±27 91.4 ± 2.0SaO2mis the lowest arterial hemoglobin saturation during maximal cycle ergometertest.60Table 9VO2TH, and power output atVO2TH,individual data.‘‘O2ThVO2Power atVO2THSUBJECT(L•min’) (mL•min1.kg1) (Watts)TM 3.22 49.9 300PT 3.34 43.8 330DB 3.36 45.6 310GA 3.11 41.2 275HT 3.48 53.0 335BT 3.10 40.2 280RR 3.41 43.4 265SF 3.27 43.8 256MF 3.15 39.4 258DH 3.40 42.1 246ME 3.34 41.3 270JF 3.24 48.2 270MEAN±SD 3.29± 0.12 44.3 ±4.2 283 ±2961Table 10 Workloads during chemoresponse tests, individual subject data.Exercise LevelGROUP VLPO LPO MPOTM 70 190 260PT 80 220 280DB 60 180 310GA 70 180 260HT 90 220 300BT 60 200 270RR 60 200 230SF 80 200 230MF 70 190 220DH 80 190 230ME 95 215 250JF 60 170 225MEAN±SD 73± 12 196± 16 255±30Values are means ± SD. Units are Watts; LOS, low oxygen saturation; NOS, normaloxygen saturation. VLPO, very low power output; LPO, low power output; MPO,moderate power output.62Table 11 Hypercapnic peripheral chemoresponse, individual subject data.Exercise LevelSUBJECT Rest VLPO LPO MPOTM 0.24±0.19 0.86±0.33 0.74±0.19 0.76±0.31PT 0.41 ± 0.08 0.66 ± 0.20 1.06 ± 0.26 0.99 ± 0.23DB 0.60±0.27 0.93±0.37 1.14±0.38 0.54±0.11GA 0.49 ± 0.26 0.73 ± 0.15 0.87 ± 0.38 0.80 ± 0.20HT 0.22±0.10 0.44±0.27 0.64±0.18 0.77±0.29BT 1.13 ± 0.29 0.65 ± 0.21 0.67 ± 0.25 0.86 ± 0.31RR 0.60 ± 0.22 1.43 ± 0.53 1.75 ± 0.38 1.64 ± 0.38SF 0.68±0.17 0.87±0.16 0.69±0.26 0.98±0.32MF 0.99±0.47 1.45±0.61 1.13±0.23 1.20±0.33DH 0.54 ± 0.30 1.05 ± 0.49 1.57 ± 0.45 1.59 ± 0.79ME 0.37±0.31 1.16±0.66 1.24±0.33 0.72±0.50JF 0.16±0.20 0.53±0.19 1.00±0.40 1.17±0.32mean ± SD 0.54 ± 0.30 0.90 ± 0.33 1.04 ± 0.36 1.00 ± 0.34Values are means ± SD Units are L•min4•mmHg;LOS, low oxygen saturation; NOS,normal oxygen saturation. VLPO, very low power output; LPO, low power output; MPO,moderate power output.63Table 12 Hypercapnic peripheral chemoresponse, group data.Exercise LevelGROUP Rest VLPO LPO MPOLOS (n=7) 0.46 ± 0.33 0.69 ± 0.17 0.87 ± 0.20 0.84 ± 0.20NO5 (n=5) 0.64 ± 0.23 1.19 ± 0.25 1.28 ± 0.41 1.23 ± 0.39Values are means ± SD. Units are Lmin1•mmHg;LOS, low oxygen saturation; NOS,normal oxygen saturation. VLPO, very low power output; LPO, low power output; MPO,moderate power output.64Table 13 Pre-hypercapnic peripheral chemoresponse SaO2, individual subjectdata.Exercise LevelSUBJECT Rest VLPO LPOMPOTM 98.8±1.0 97.9±0.5 97.1±0.7 95.9±0.6PT 98.0±0.0 97.8±1.1 99.3±0.795.9±1.6DB 97.7 ± 1.3 97.7 ± 0.796.0± 0.7 95.0 ± 0.6GA 99.0±0.8 96.7±0.1 97.0±0.994.7± 1.0HT 97.0 ± 0.6 96.0 ± 0.6 96.6 ± 0.7 95.6± 0.9BT 96.6±0.8 97.7±0.7 96.6±0.596.0±0.8RR 99.1±0.9 97.4±0.9 97.2±0.896.3±1.7SF 96.7 ± 1.0 97.3 ± 0.9 96.2 ± 1.996.2 ± 0.9MF 96.5±0.5 97.0±0.8 96.0±0.896.3± 1.4DH 98.4 ± 0.9 97.3 ± 0.5 96.2± 0.7 96.4 ± 1.0ME 96.8±0.8 96.5±0.5 95.4±1.0 96.1±0.6JF 96.9± 1.3 97.6± 1.3 96.0± 1.693.6± 1.6MEAN ± SD 97.6 ± 1.0 97.2 ± 0.6 96.6 ± 1.0 95.7± 0.8Values are means ± SD. Units are %. VLPO, very low power output; LPO, lowpoweroutput; MPO, moderate power output. saturation. VLPO, verylow power output; LPO,low power output; MPO, moderate power output.65Table 14 Pre-hypercapnic peripheral chemoresponse SaO2, group data.Exercise LevelGROUP Rest VLPO LPO MPOLOS (n=7) 97.7 ± 0.9 97.4 ± 0.7 96.9 ± 1.1 95.2 ± 0.9NOS (n=5) 97.5 ± 1.2 97.1 ± 0.4 96.2 ± 0.7 96.3 ± 0.1Values are means ± SD. Units are %; LOS, low oxygen saturation; NOS, normaloxygensaturation. VLPO, very low power output; LPO, low power output; MPO, moderatepower output.66Table 15 Hyperoxic peripheral chemoresponse, group data.Exercise LevelGROUP Rest VLPO LPO MPOLOS (n=7) 18.6 ± 8.8 19.1 ± 6.8 16.2 ± 4.0 17.1 ± 5.2NOS (n=5) 22.1 ± 7.5 15.9 ± 6.5 16.2 ± 5.5 17.7 ± 5.5Values are means ± SD. Units are %; LOS, low oxygen saturation; NOS, normal oxygensaturation. VLPO, very low power output; LPO, low power output; MPO, moderatepower output.67Table 16 Hyperoxic peripheral chemoresponse, individual subject data.Exercise LevelSUBJECT Rest VLPO LPO MPOTM 11.0 16.4 11.5 17.5PT 28.8 25.2 19.4 26.0DB 12.4 9.2 11.4 16.3GA 9.2 11.3 13.5 10.7HT 16.7 24.7 20.6 12.5BT 31.5 24.0 17.4 15.2RR 21.8 8.2 12.0 7.3SF 25.0 25.5 19.0 17.2MF 9.8 13.3 8.8 17.2DH 29.9 18.7 21.7 21.7ME 24.0 14.0 19.3 19.3JF 20.3 23.0 19.6 21.2MEAN ± SD 20.0 ± 8.1 17.8 ± 6.6 16.2 ± 4.4 16.8 ± 5.1Units are %. VLPO, very low power output; LPO, low power output; MPO, moderatepower output.68Table 17 Pre-hyperoxic peripheral chemoresponseSaO2,group data.Exercise LevelGROUP Rest VLPO LPO MPOLOS (n=7) 97.7±0.9 97.3±0.5 96.1 ±0.8 95.6± 1.1NOS (n=5) 97.8 ± 0.8 97.6 ± 0.7 96.4 ± 0.6 96.3 ± 0.8Values are means ± SD. Units are %; LOS, low oxygen saturation; NOS, normal oxygensaturation. VLPO, very low power output; LPO, low power output; MPO, moderatepower output.69Table 18 Pre-hyperoxic peripheral chemoresponseSaO2,individual subjectdata.Exercise LevelSUBJECTTMPTDBGAHTBTRRSFMFDHMEJFMEAN±SDRest99.4± 1.497.0± 1.197.6±0.698.0±0.897.0± 1.396.7 ± 2.197.2± 1.396.9± 1.197.9± 1.797.9±0.998.9± 1.998.3± 1.297.7±0.8VLPO98.1 ± 1.096.8±0.896.7 ± 0.997.9±0.497.2±0.997.2±0.898.2± 1.397.2± 1.098.1±0.996.5 ± 0.697.8±0.596.9± 1.297.4±0.6LPO97.2±0.596.1 ±0.895.4± 1.196.5 ± 0.497.0 ± 0.695.8 ±0.996.1 ± 1.396.1 ± 1.197.5±1.296.3 ± 0.996.2±0.894.9± 1.696.3 ± 0.7MPO97.2±0.996.2± 1.095.0 ± 0.796.2 ± 0.696.3±0.794.3±1.496.1 ± 1.396.0 ± 0.595.4± 1.797.5±0.596.6± 1.094.2± 1.095.9± 1.0Values are means ± SD. Units are %. VLPO, very low power output; LPO, low poweroutput; MPO, moderate power output. saturation. VLPO, very low power output; LPO,low power output; MPO, moderate power output.70Table 19 RMANOVA, Hypercapnic Chemoresponse.UNIVARIATE AND MULTIVARIATE REPEATED MEASURES ANALYSISBETWEEN SUBJECTSSOURCE SS DF MS F PGROUP 1.563 1 1.563 13.652 0.004ERROR 1.145 10 0.115WITHIN SUBJECTSSOURCE SS DF MS F P G-G H-FLEVELS 2.012 3 0.67 1 10.446 0.000 0.001 0.000LEVELSX 0.172 3 0.057 0.894 0.456 0.426 0.451GROUPSERROR 1.926 30 0.064GREENHOUSE-GEISSER EPSILON: 0.6811HUYNH-FELDT EPSILON: 0.943371Table 20 Polynomial Contrasts, Hypercapnic Chemoresponse.SINGLE DEGREE OF FREEDOM POLYNOMIAL CONTRASTSPOLYNOMIAL TEST OF ORDER 1 (LINEAR)SS DF MS F1.469 1 1.469 13.413P0.004SOURCELEVELSLEVELSX 0.042 1 0.042 0.380 0.552GROUPSERROR 1.095 10 0.110POLYNOMIAL TEST OF ORDER 2 (QUADRATIC)SOURCE SS DF MS F PLEVELS 0.540 1 0.540 8.197 0.017LEVELSX 0.090 1 0.090 1.370 0.269GROUPSERROR 0.658 10 0.066POLYNOMIAL TEST OF ORDER 3 (CUBIC)SOURCE SS DF MS F PLEVELS 0.003 1 0.003 0.189 0.673LEVELSX 0.040 1 0.040 2.343 0.157GROUPSERROR 0.173 10 0.01772Table 21 RMANOVA, Pre-Hypercapnic ChemoresponseSaO2.UNIVARIATE AND MULTIVARIATE REPEATED MEASURES ANALYSISBETWEEN SUBJECTSSOURCE SS DF MS F PGROUP 0.024 1 0.024 0.019 0.894ERROR 13.092 10 1.309WITHIN SUBJECTSSOURCE SS DF MS F P G-G H-FLEVELS 23.103 3 7.701 15.120 0.000 0.000 0.000LEVELSX 4.909 3 1.636 3.213 0.037 0.044 0.037GROUPSERROR 15.280 30 0.509GREENHOUSE-GEISSER EPSILON: 0.8860HUYNH-FELDT EPSILON: 1.000073Table 22 Polynomial Contrasts, Pre-Hypercapnic ChemoresponseSaO2,SINGLE DEGREE OF FREEDOM POLYNOMIAL CONTRASTSPOLYNOMIAL TEST OF ORDER 1 (LINEAR)SOURCE SS DF MS F PLEVELS 22.548 1 22.548 34.299 0.000LEVELSX 1.488 1 1.488 2.264 0.163GROUPSERROR 6.574 10 0.657POLYNOMIAL TEST OF ORDER 2 (QUADRATIC)SOURCE SS DF MS F PLEVELS 0.550 1 0.550 1.555 0.241LEVELSX 2.333 1 2.333 6.596 0.028GROUPSERROR 3.537 10 0.354POLYNOMIAL TEST OF ORDER 3 (CUBIC)SOURCE SS DF MS F PLEVELS 0.005 1 0.005 0.010 0.922LEVELSX 1.088 1 1.088 2.105 0.177GROUPSERROR 5.169 10 0.51774Table 23 RMANOVA, Hyperoxic Chemoresponse.UNIVARIATE AND MULTIVARIATE REPEATED MEASURES ANALYSISBETWEEN SUBJECTSSOURCE SS DF MS F PGROUP 0.026 1 0.026 0.000 0.988ERROR 1081.887 10 108.189WITHIN SUBJECTSSOURCE SS DF MS F P G-G H-FLEVELS 117.389 3 39.130 2.152 0.115 0.130 0.115LEVELSX 66.757 3 22.252 1.224 0.318 0.317 0.318GROUPSERROR 545.572 30 18.186GREENHOUSE-GEISSER EPSILON: 0.8019HUYNH-FELDT EPSILON: 1.000075Table 24 Polynomial Contrasts, Hyperoxic Chemoresponse.SINGLE DEGREE OF FREEDOM POLYNOMIAL CONTRASTSPOLYNOMIAL TEST OF ORDER 1 (LINEAR)SS DF MS FSOURCE PLEVELS 83.122 1 83.122 2.968 0.116LEVELSX 11.933 1 11.933 0.426 0.529GROUPSERROR 280.023 10 28.002POLYNOMIAL TEST OF ORDER 2 (QUADRATIC)SOURCE SS DF MS F PLEVELS 34.114 1 34.114 2.267 0.163LEVELSX 28.392 1 28.392 1.887 0.200GROUPSERROR 150.498 10 15.050POLYNOMIAL TEST OF ORDER 3 (CUBIC)SOURCE SS DF MS FPLEVELS 0.153 1 0.153 0.013 0.911LEVELSX 26.432 1 26.432 2.297 0.161GROUPSERROR 115.051 10 11.50576Table 25 RMANOVA, Pre-Hyperoxic ChemoresponseSaO2.UNIVARIATE AND MULTIVARIATE REPEATED MEASURES ANALYSISBETWEEN SUBJECTSSOURCE SS DF MS F PGROUP 1.332 1 1.332 0.966 0.349ERROR 13.791 10 1.379WITHIN SUBJECTSSOURCE SS DF MS F P G-G H-FLEVELS 25.502 3 8.501 18.769 0.000 0.000 0.000LEVELSX 0.619 3 0.206 0.456 0.715 0.669 0.715GROUPSERROR 13.588 30 0.453GREENHOUSE-GEISSER EPSILON: 0.7780HUYNH-FELDT EPSILON: 1.000077Table 26 Polynomial Contrasts, Pre-Hyperoxic ChemoresponseSaO2.SINGLE DEGREE OF FREEDOM POLYNOMIAL CONTRASTSPOLYNOMIAL TEST OF ORDER 1 (LINEAR)SS DF MS F23.989 1 23.989 36.276P0.000SOURCELEVELSLEVELSX 0.552 1 0.552 0.835 0.382GROUPSERROR 6.613 10 0.661POLYNOMIAL TEST OF ORDER 2 (QUADRATIC)SOURCE SS DF MS F PLEVELS 0.001 1 0.001 0.002 0.967LEVELSX 0.011 1 0.011 0.019 0.892GROUPSERROR 5.696 10 0.570POLYNOMIAL TEST OF ORDER 3 (CUBIC)SOURCE SS DF MS F PLEVELS 1.512 1 1.512 11.828 0.006LEVELSX 0.056 1 0.056 0.439 0.523GROUPSERROR 1.278 10 0.12878Table 27 Linear Regression, Hypercapnic Chemoresponse and Pre-HypercapnicChemoresponseSaO2,NOS subjects.DEPENDANT VARIABLE IS SATURATIONN R R2 ADJ. R2 S.E.E.20 0.3 12 0.097 0.047 0.835REGRESSION COEFFICIENTSVARIABLE COEFF. STD STD TOLERANCE T P(2 TAIL)ERROR COEFCONSTANT 97.479 0.546 0.000 . 178.460 0.000RESP -0.660 0.474 -0.312 1.000 -1.392 0.181ANALYSIS OF VARIANCESOURCE SS DF MS F PREGRESSION 1.351 1 1.351 1.937 0.181RESIDUAL 12.554 18 0.69779Table 28 Linear Regression, Hypercapnic Chemoresponse and Pre-HypercapnicChemoresponseSaO2,LOS subjects.DEPENDANT VARIABLE IS SATURATIONN R R2 ADJ. R2 S.E.E.28 0.338 0.115 0.080 1.247REGRESSION COEFFICIENTSVARIABLE COEFF. STD STD TOLERANCE T P(2 TAIL)ERROR COEFCONSTANT 97.954 0.666 0.000 . 146.979 0.000RESP -1.596 0.870 -0.338 1.000 -1.834 0.078ANALYSIS OF VARIANCESOURCE SS DF MS F PREGRESSION 5.227 1 5.227 3.364 0.078RESIDUAL 40.400 26 1.55480Table 29 Linear Regression, Hyperoxic Chemoresponse and Pre-HyperoxicChemoresponseSaO2,NOS subjects.DEPENDANT VARIABLE IS SATURATIONN R R2 ADJ. R2 S.E.E.20 0.217 0.047 0.000 0.938REGRESSION COEFFICIENTSVARIABLE COEFF. STD STD TOLERANCE T P(2 TAIL)ERROR COEFCONSTANT 96.454 0.635 0.000 . 151.785 0.000RESP 0.032 0.034 0.217 1.000 0.944 0.358ANALYSIS OF VARIANCESOURCE SS DF MS F PREGRESSION 0.784 1 0.784 0.890 0.358RESIDUAL 15.848 18 0.88081Table 30 Linear Regression, Hyperoxic Chemoresponse and Pre-HyperoxicChemoresponseSaO2,LOS subjects.DEPENDANT VARIABLE IS SATURATIONN R R2 ADJ. R2 S.E.E.28 0.121 0.015 0.000 1.212REGRESSION COEFFICIENTSVARIABLE COEFF. STh STD TOLERANCE T P(2 TAIL)ERROR COEFCONSTANT 97.098 0.708 0.000 . 137.235 0.000RESP -0.023 0.038 -0.121 1.000 -0.6210.540ANALYSIS OF VARIANCESOURCE SS DF MS F PREGRESSION 0.566 1 0.566 0.385 0.540RESIDUAL 38.195 26 1.46982C1)0APPENDIX CFIGURESFigure 7 Power output at various exercise intensities, group data.350 -300 -_250 -200 -150 -100 -500-Values are means ± SD. Open circles, LOS, low oxygen saturation; Open squares, NOS,normal oxygen saturation. VLPO, very low power output; LPO, low power output; MPO,moderate power output.VLPO LPO MPO83Figure 8 Power output at various exercise intensities, NOS subjects.350300250200az150C100500VLPO, very low power output; LPO, low power output; MPO, moderate power output.p RRD SFp MF—N---- DHw MERest VLPO LPO MPO84Figure 9 Power output at various exercise intensities, LOS subjects.3503002502000150C100500p TMD PTp DB—NH-- GAw HTL BT—0---— JFVLPO, very low power output; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO85Figure 10 Hypercapnic peripheral chemoresponse at various exercise intensities,NOS subjects.2.0 -p RR1-1.UD SF- o MF12- —N-—DH1)rJ)08____wMEo0.4-c)0.0- IVLPO, very low power output; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO86Figure 11 Hypercapnic peripheral chemoresponse at various exercise intensities,LOS subjects.2.0 -p TM1.6- D PToDB12I I•IRest VLPO MPOVLPO, very low power output; LPO, low power output; MPO, moderate power output.LPO87Figure 12Pre-C02responseSaO2at various exercise intensities, NOS subjects.100 -p RR99D SF980 MF97-—-—DH96-8m ME95-94 -93 -VLPO, very low power output; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO88Figure 13 Pre-C02 responseSaO2at various exercise intensities, LOS subjects.100 -p TMo PT980 DB—s--- GAw HT96-BT—0--— JF95 -94 -93I I I IRest VLPO LPO MPOVLPO, very low power output; LPO, low power output; MPO, moderate power output.89Figure 14 Hyperoxic peripheral chemoresponse at various exercise intensities,NOS subjects.35 -p RR30D SF25___20:—is-—DH15W ME10 -5- I IRest MPOVLPO, very low power output; LPO, low power output; MPO, moderate power output.VLPO LPO90Figure 15 Hyperoxic peripheral chemoresponse at various exercise intensities,LOS subjects.35 -p TM30 D PTp DB25-—N--GAw HT20BTis-—Q-—JF10 -5- IVLPO, very low power output; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO91Figure 16Pre-02responseSaO2at various exercise intensities, NOS subjects.100 -99 -95 -94 -p RRD SFp MF—1SF— DHw ME93I I IRest VLPO LPO MPOVLPO, very low power output; LPO, low power output; MPO, moderate power output.98 -97 -96 -I92Pre-02responseSaO2at various exercise intensities, LOS subjects.p TMD PTp DB—u-— GAw HTBT—0— JFFigure 17100 -99 -95-94 -93-VLPO, very low power output; LPO, low power output; MPO, moderate power output.Rest VLPO LPO MPO93APPENDIX DTable 31EXCLUDED SUBJECT DATAAge, height, mass, and body surface area, individual subject data.Subject Age Height Mass BSA(yrs) (cm) (kg)(m2)AC 42 173 68.0 1.81AS 21 180 73.4 1.93BG 35 181 77.8 1.98CA 27 181 77.2 1.97CJ 26 172 61.0 1.72DL 32 191 94.6 2.25FH 35 176 68.3 1.83FM 25 181 77.6 1.98JG 34 191 94.0 2.23TV 23 181 78.6 1.99KR 25 187 88.2 2.14MF-W 22 181 65.0 1.84MF-J 23 178 68.1 1.85MS 24 183 81.0 2.03MT 25 180 70.2 1.89MW 28 172 65.8 1.78NG 28 180 77.6 1.97PK 26 185 84.5 2.09RH 25 186 80.7 2.06RM 29 183 78.9 2.01TC 27 185 85.3 2.1094TG 26 187 81.2 2.06TR 37 193 92.0 2.23TS 29 172 64.9 1.76mean±SD 28±5 182±6 77.2±9.5 1.98±0.15Values are means ± SD.95Table 32 Age, height, mass, and body surface area, group data.Age Height Mass BSAGROUP(yrs) (cm) (kg)(m2)UNFIT,NORMAL 30±6 182±7 79.7±9.7 2.01±0.16(n = 8)UNFIT, EN 29±7 187±6 87.5 ±9.2 2.13 ±0.14(n = 2)FIT,NORMAL 28±5 181±6 75.5±9.1 1.95±0.15(n = 11)FIT, EIH 24±3 180± 2 70.1± 6.3 1.89±0.07(n = 3)Values are means ± SD. Groups are UNFIT,(VO2max<5.00 L•min’ or 60.0 mL•minl.kg4);FIT,(VO2max5.00 L•min’ or 60.0 mL.min1.kgi); NORMAL,(SaO2max>91.0 %); EIH,(SaO2max91.0 %).96Table 33VO2max,peak power output, and lowest arterial hemoglobinsaturation during maximal cycle ergometer test, individual subjectdata.SubjectVO2max VO2maxPeak PowerSaO2max(Lmin1) (mL.min4.kg1) (Watts) (%)AC 3.89 57.2 381 92.4AS 4.72 64.3 500 93.0BG 4.14 53.2 389 95.5CA 5.15 66.7 450 90.0CJ 4.25 69.7 410 91.5DL 4.68 49.5 418 95.0FH 4.38 64.1 415 91.3FM 4.34 55.9 404 93.8JG 4.82 51.3 445 90.4JV 5.23 66.6 500 91.7KR 4.44 50.3 430 93.2MF-W 4.81 66.2 475 89.1MF-J 4.60 67.6 475 90.3MS 4.55 56.2 435 90.3MT 4.68 66.6 475 92.3MW 3.83 58.2 396 94.9NG 4.67 60.2 412 91.1PK 4.58 54.2 450 93.2RH 4.86 60.2 445 92.8RM 4.89 62.0 450 91.5TC 5.01 58.7 465 91.8TG 4.80 59.1 445 91.997TR 5.59 60.8 488 91.9TS 4.86 74.9 410 94.4mean ± SD 4.66 ± 0.40 60.6 ± 6.5 440±34 92.2 ± 1.7Values are means ± SD.98Table 34VO2max,peak power output, and lowest arterial hemoglobinsaturation during maximal cycle ergometer test, group data.GROUPVO2max(L.min’)4.34±0.34VO2max(mLmin’.kg’)54.7 ± 3.6Peak Power(Watts)414±26SaO2max(%)93.7± 1.3UNFIT, NORMAL(n = 8)UNFIT,EIH 4.69±0.19 53.8±3.5 440±7 90.4±0.1(n = 2)FIT, NORMAL 4.83 ± 0.37 64.4 ± 4.8 452 ± 36 92.1 ± 1.0(n = 11)FIT, EN 4.85 ± 0.28 66.8 ± 0.7 467 ± 14 89.8 ± 0.6(n = 3)Values are means ± SD. Groups are UNFIT,(VO2max<5.00 L•min1or 60.0 mL•min‘.kg4);FIT,(VO2max 5.00L•min1or 60.0 mL.min1.kg);NORMAL,(SaO2max>91.0 %); EIH,(SaO2max91.0 %).99Table 35VO2TH,and power output atVO2TH’individual subject data.VO2TH VO2Power atSubject (L•min’) (mL•min’•kg1) (Watts)AC 2.98 43.8 260AS 3.20 43.6 310BG 2.26 29.0 200CA 3.50 45.3 280CJ 2.73 44.8 250DL 3.23 34.1 250FH 2.74 40.1 238FM 3.03 39.0 275JG 2.98 31.7 255JV 3.50 44.5 325KR 3.10 35.2 270MF-W 2.60 40.0 270MF-J 2.93 43.1 280MS 3.08 38.0 275MT 2.84 34.7 280MW 2.81 42.7 280NG 3.00 39.3 234PK 2.96 35.0 265RH 3.35 41.4 280RM 2.90 36.8 245TC 3.30 38.7 290TG 3.12 38.4 275TR 3.84 41.8 310100TS 3.20 49.3 275mean ± SD 3.05 ± 0.33 39.6 ± 4.8 270±27Values are means ± SD.101Table 36 VO2TH , and power output at VO2TH, group data.VO2TH V02ThPower atVO2GROUP(L•min’) (mL•min1•kg) (Watts)UNFIT, NORMAL 2.94 ± 0.30 37.2 ± 4.8 259±26(n = 8)UNFIT, EN 3.03 ± 0.07 34.9 ± 4.5 265 ± 14(n = 2)FIT, NORMAL 3.15 ±0.15 41.4±4.1 276±31(n = 11)FIT, EIH 3.01 ± 0.46 42.8 ± 2.7 277±6(n = 3)Values are means ± SD. Groups are UNFIT,(VO2max<5.00 L•min’ or 60.0 mL•min‘.kg’); FIT,(VO2max5.00 L•min’ or 60.0 mL.min1•kg4);NORMAL,(SaO2max>91.0 %); EIH,(Sa°2max91.0 %).102

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.831.1-0077240/manifest

Comment

Related Items