INFORMATION TO USERS

This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the qu()lity is heavily dependent upon the quality of the original submitted.

The following e}(planation of techniques is provided to help you understand markings or patterns which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity.

2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during e}(posure and thus cause a biurred image. You will find a good image of the page in the adjacent frame.

3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again - beginning below the first row and continuing on until compl£~1.

4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced.

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received.

){ero){ University Micromms 300 North Zoeb Road Ann Arbor, Michigan 4B106 75-17,118

DRESSENDORFER p Rudolph H., Jr., 1943­ COMPARISCN OF CARDIORESPIRATORY RESPONSES TO GRADED UPRIGIff EXERCISE IN AIR AND WATER. University of Hawaii, Ph.D., 1974 Physiology

Xero}{ University Microfilms, Ann Arbor, Michigan 48106 COMPARISON OF CARDIOq~SPIRATORY RESPONSES TO GRADED UPRIb~T EXERCISE IN AIR AND WATER

A DISSERTATION SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAII IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN PHYSIOLOGY DECEMBER 1974

By Rudolph H. Dressendorfer, Jr.

Dissertation Committee: Martin D. Rayner, Chairman Vincent DeFeo Suk Ki Hong Terence O. Moore Richard M. Smith Robert A. Tracy iii

ABSTRACT

The objective of this investigation was to determine the effect of head-out water immersion per se on the cardiorespiratory response to graded upright cycling exercise. The use of a standard Monark bicycle ergometer for cycling underwater enabled the subjects to perform the same exercise in air and water. The energetics of this innovated underwater ergometer are evaluated. The submaximal to maximal exercise responses of 7 trained male subjects cycling in 22-25 C air and 30 C water, and running on a treadmill were compared. Four athletes within this group were also tested in 25 and 35 C water. Oxygen uptake (V02), heart rate (fh), cardiac stroke volume (Qs; determined by impedance cardiography), expiratory minute volume (~E)' respiratory frequency

(fR)' and rectal temperature (Tre) were measured during rest and exer­ cise bouts of 5 min duration. Work rates were adjusted to elicit approximately 20, 35, 50, 70 and 100% of maximum oxygen uptake (~02max).

Total immersion time was 50 min. The V0 2max (N=7) for cycling was 3.29 L/min in air and 3.18 L/min in 30 C water. The 3% decrease in

V02max during im~ersion was not statistically significant. Treadmill Vo 2max (N=7) was 3.67 L/min. The values for V02max in the athletes were 3.88 L/min (cycling in air), 3.75 L/min (cycling in 25, 30, and 35 C water) and 4.47 L/min (running). During submaximal exercise fh in 30 C water was 5-7 bpm slower than in air while 0E and fR were unchanged. During maximal exercise, however, peak fh and VE were sig­ nificantly lower in 30 C water by 10 bpm and 16 L/min, respectively. Oxygen transport was maintained by increases in Qs and the fraction of iv oxygen removed from the inspired air which compensated for the reduc­ tions in fh and VE. V0 2max was unchanged by the use of SCUBA in 30 C water in spite of further reductions in VE and maximal voluntary ven­ tilation. V02max and submaxima1 VE were unaffected by water temperature (Tw). Tre , fh, and calculated cardiac output (6), however, increased with Tw at all levels of V02. Peak 6 values obtained in the 4 athletes were 24.3 L/min (cycling in air), 23.1, 25.2, and 27.1 L/min (cycling in 25, 30, and 35 C water, respectively), and 26.8 L/min (running). The distribution of the elevated Qduring cycling in 30 and 35 C water versus 25 C water must have been to the skin or other organs with low

02 extraction since v02max was unchanged. Blood lactate was higher and pyruvate lower following maximal exercise in 35 C compared to 25 C immersion suggesting a greater anaerobic work component in the warm water. The above findings show that aerobic power is not significantly altered by head-out immersion in 25 C to 35 C water. Moreover, V02max during cycling in water appears to be limited by factors other than pulmonary ventilation or the pumping capacity of the heart. In addition

Tw should be considered when making cardiorespiratory measurements in water. v

TABLE OF CONTENTS Page

ABSTRACT ... iii LIST OF TABLES vii LIST OF FIGURES viii CHAPTER 1 GENERAL INTRODUCTION 1 Background ..... 1 State of the Problem 5 The Purpose ..... 6 The Hypothesis .... 7 CHAPTER 2 MATERIAL AND GENERAL METHODS 9 Subjects ...... 9 Methodology . 10 Experimental Protocol 21 Statistics . 25 List of Abbreviations. 27 CHAPTER 3 THE MONARCH BICYCLE ERGOMETER AS A SUITABLE UNDERWATER EXERCISER 30 Introduction 30 Methods ...... 31 Results . 32 Discussion ..... 36 CHAPTER 4 COMPARISON OF CARDIORESPIRATORY RESPONSES TO GRADED CYCLING EXERCISE IN AIR AND WATER 40 Introduction 40 Methods ...... 42 Resu1 ts ...... 44 Discussion . 59 CHAPTER 5 EFFECT OF SCUBA ON PULMONARY VENTILATION DURING SUBMAXIMAL AND MAXIMAL CYCLING EXERCISE IN WATER 64 Introduction 64 Methods .. 66 Results . 68 Discussion . 74 vi

Page

CHAPTER 6 THE EFFECT OF WATER TEMPERATURE ON CARDIORESPIRATORY AND METABOLIC RESPONSES TO MAXIMAL EXERCISE . 82 Introduction 82 Methods ..... 85 Results . 87 Discussion . 91 CHAPTER 7 THE EFFECTS OF BODY POSITION AND WATER IMMERSION ON STROKE VOLUME 97 Introduction 97 Methods ...... 99 Results . 104 Discussion . 107 CHAPTER 8 CARDIORESPIRATORY RESPONSES TO GRADED CYCLING EXERCISE IN AIR AND IN 25 C, 30 C, AND 35 C WATER ... . 112 Introduction · . . .. 112 Methods ... · ... . 115 Resu1 ts ...... 121 Discussion . . . .. 146 CHAPTER 9 GENERAL SUMMARY AND CONCLUSIONS 150 APPENDIX A CHARACTERISTICS OF SUBJECTS · 156 APPENDIX B INDIVIDUAL DATA COMPARING RESTING AND EXERCISE RESPONSES IN AIR AND 30 C WATER . 160 APPENDIX C INDIVIDUAL DATA COMPARING THE USE OF COLLINS VALVE OR SCUBA DURING REST AND EXERCISE IN 30 C WATER ...... 162 APPENDIX D INDIVIDUAL DATA COMPARING MEASUREMENTS MADE DURING REST AND EXERCISE IN AIR, WARM WATER (35 C) AND COLD WATER (25 C) 166 APPENDIX E INDIVIDUAL DATA COMPARING CARDIORESPIRATORY RESPONSES TO REST AND EXERCISE FOR CYCLING IN AIR, 25, 30, AND 35 C WATER, AND TREADMILL RUNNING 168 BIBLI OGRAPHY ...... 172 vii

LIST OF TABLES Table Page

1 Reliability of Quantitative Determinations 19 2 Reproducibility of Repeated Determinations on Separate Days ...... 20 3 Description of Subjects 43 4 Comparison of Mean Cardiorespiratory and Rectal Temperature Responses to Rest and Submaxima1 Upright Cycling in Air and 30 C Water ..... 46

5 Comparison of Mean Cardiorespiratory and Rectal Temperature Responses to Maximal Upright Cycling in Air and 30 C Water ...... 48

6 Maximum Voluntary Ventilation in Air, 30 C Water, and in Water Using SCUBA . 49 7 Effect of SCUBA on the Cardiorespiratory Response to Rest and Exercise in 30 C Water . 69

8 Description of Subjects 86

9 Effect of Water Temperature on Cardiorespiratory and Metabolic Parameters During Rest, Submaxima1 Exercise, and Maximum Effort Exercise 88 10 Description of Subjects 100 11 Effect of Body Position and Immersion on Heart Rate and Stroke Volume .. 105 12 Description of Subjects 116 13 Mean Cardiorespiratory Responses to Rest, Sub­ maximal, and Maximum Exercise for Upright Cycling in Air, 25, 30, 35 C Water, and Treadmill Running 122 14 Regression Equations for Cardiac Output During Exercise Found in Present Study Compared with Published Values . 131 15 Percentage Change in Maximal Cardiorespiratory Responses for Various Conditions of Exercise . 143 viii

LIST OF FIGURES Figure Page

1 Cut-Away Diagram of Immersion Tank 12 2 Experimental Protocol 22

3 Oxygen Up~ake as a Function of Aerobic Work Rate . 23

4 Oxygen Uptake as a Function of Pedal Frequency for "No-Load" Cycling in Water and Air .. 33 5 Regression Equation of Oxygen Uptake as a Function of Pedal Frequency . 34

6 Oxygen Uptake as a Function of Work Load During Cycling in Air and 30 C Water . 35 7 Heart Rate and Rectal Temperature as a Function of Submaximal Oxygen Uptake ...... 50 8 Regression of Heart Rate in Air on %V02 max 51

9 Regre~sion of Heart Rate in 30 C Water on % V0 2 max ...... 52 10 Oxygen Pulse as a Function of % V0 2 max 53 11 Expiratory Minute Volume and Breathing Frequency as a Function of Submaximal Oxygen Uptake 54 12 Regression of Expiratory Minute Volume in Air on Oxygen Upta ke ...... 55

13 Regression of Expiratory Minute Volume in 30 C Water on Oxygen Uptake ...... 56

14 Effect of Immersion on the Ventilatory Response to Exercise in a Typical Subject ... 57

15 Comparison of Cardiorespiratory Responses to Maximal Exercise in 25 C Air and 30 C Water 58

16 Effect of SCUBA on Pulmonary Ventilation During Submaximal Exercise ...... 71 17 Effect of SCUBA on the ventilatory Response to Submaximal Exercise in a Typical Subject ... 72 ix

Figure Page

"18 Comparison of Cardiorespiratory Responses to Exercise while Breathing from Collins Valve or SCUBA ...... 73 19 Effect of SCUBA on Alveolar Ventilation and Breathing Frequency . 75 20 Effect of SCUBA on the Submaximal Cardio­ respiratory Responses of a Typical Subject 76 21 Peak Heart Rate and Alveolar Ventilation as a Function of V0 2 max during 4 Conditions of Exercise . 80 22 Effect of Immersion and Water Temperature on Cardiorespiratory Responses to Maximal Exercise 93 23 Impedance Cardiograph Recording at Rest 102 24 Reciprocal Changes in Heart Rate and Stroke Volume at Rest . 106 25 Impedance Cardiograph Recording Immediately Following Exercise . 118 26 Abrupt Slowing of Heart Rate after Exercise in Cold Water . 120 27 Example of Impedance Cardiograph Recording Following Maximal Exercise . 124 28 Heart Rate as a Function of Oxygen Uptake During Running and Cycling in Air . · . · ·· 126 29 Heart Rate as a Function of Oxygen Uptake During Cycling in 30 C Water. . . · .. · · · ·· 127 30 Heart Rate as a Function of Oxygen Uptake During Cycling in 25 C Water. . . ·· · ··· 128 31 Heart Rate as a Function of Oxygen Uptake During Cycling in 35 C Water. .. ·· · · 129 32 Cardiac Output as a Function of Oxygen Uptake During Treadmill Running . 132 33 Cardiac Output as a Function of Oxygen Uptake During Cycling in Air . 133 x

Figure Page

34 Cardiac Output as a Function of Oxygen Uptake During Cycling in 25 C Water . 134 35 Cardiac Output as a Function of Oxygen Uptake During Cycling in 30 C Water . 135 36 Cardiac Output as a Function of Oxygen Uptake During Cycling in 35 C Water ..... 136

37 Stroke Volume as a Function of Oxygen Uptake for Five Conditions of Exercise . 137

38 Stroke Volume as a Function of Oxygen Uptake .. 138

39 Expiratory Minute Volume as a Function of Submaximal Oxygen Uptake ...... 140 40 Rectal Temperature as a Function of Exercise Intensity and Ambient Condition ...... 141 41 Cardiovascular Responses to Maximum Effort Running and Cycling . 144

(Appendix C)

1 Cardiorespiratory Responses to Various Conditions of Maximal Exercise . 165 CHAPTER 1 GENERAL INTRODUCTION

Background

"In what terms do we describe organs of the body; in terms of rest regarding exercise as a variant~ or in terms of its capacity to do work? It seems that the most uniform measure of an organ is in terms of the maximum which it can accompZish. " Sir Joseph Barcroft, 1934

In order to describe how successfully a system operates and is regulated under stress conditions it is necessary to know its full­ scale capacity. Under the assumption that maximum oxygen uptake is an objective criterion of cardiorespiratory performance (Taylor et aZ.~ 1955), physiologists have used this measurement to specify the limits of the respiratory and cardiovascular systems and their capability to respond to exercise. This assumption was also used in the experiments to be described in order to assess the effect of water immersion on

the cardiorespiratory re~ponse to exercise. Although the technique for measuring maximum oxygen uptake was developed 50 years ago (Hill et aZ.~ 1924) the basic principle remains unchanged. When the measurement is obtained under properly standardized conditions which assure a) sufficient work duration and intensity, b) engagement of about 50% or more of the total muscle mass, c) the upright working position and d) independence from motivation or skill of the subject, then maximum oxygen uptake is a highly reproducible measurement of the maximal aerobic metabolic rate (Astrand, 1952; Taylor et aZ.~ 1963). The reproducibility of the measurement has a 2 coefficient of reliability of 0.95 and repeated determinations on separate days differ by only 2-4% (Rowell, 1962). It has been shown that maximum oxygen uptake is unaffected or only slightly reduced (2-4%) by a variety of physical stresses: dehydration of 5% of body weight (Saltin, 1964), acute loss of 14% (500 ml) of blood volume (Rowell, 1962), high air temperature (Rowell et al.~ 1969), sometimes high body temperature (Rowell et al.~ 1969), pyrogen-induced fever (Grimby, 1962), acute starvation (Henschel et al.~ 1954), preg­ nancy (Dressendorfer, unpublished data) and 3 hours of hard, prolonged exercise (Saltin and Stenberg, 1964). Since these stresses may markedly affect cardiodynamics, the central circulation and pulmonary ventila­ tion, the relative constancy of maximum oxygen uptake suggests that some sort of emergency reaction occurs during a maximal effort of up to five minutes that allows compensatory mechanisms to restore the cardiorespi­ ratory capacity. The question remains, however, whether the physical stress encountered during water immersion is sufficient to affect maximum oxygen uptake. H. Rahn (personal communication) has called the physiology of submersion "a new physiology" because all organ functions are affected by the simple act of head-out water immersion. The two primary stresses are weightlessness and the unequal hydrostatic pressure distribution across the chest wall, but thermal, motor and sensory problems are also encountered. Many studies have emphasized the importance of body position and the type of exercise used to elicit maximum oxygen uptake. Uphill treadmill walking and running are generally regarded as the measures of "true" maximum oxygen uptake, but cycling on a bicycle ergometer is 3 often used. The value for maximum oxygen uptake on a bicycle ergometer, however, may range 1-19% lower than that obtained by uphill treadmill runni,~g (Astrand and Saltin, 1961; Chase et aZ.~ 1966; Faulkner et aZ.~ 1971, Hermansen and Saltin, 1969). Body position also affects the measurement. It is well-documented that maximunl oxygen uptake during work in the supine posture is only about 85% of that for the same exercise in the upright position (Astrand and Saltin, 1961; Stenberg et aZ.~ 1967). Whether upright bicycling or uphill treadmill exercise is used, it is clear that maximum oxygen uptake is only maximal for a specified type of work. Strict comparability of the measurement during experimental manipulation of the environmental condition therefore requires the same type of exercise be used in all cases. This leads to the problem of comparing exercise in air with that in water. Prior to beginning the present series of experiments in the summer of 1972 it had been reported that maximum oxygen uptake during swimming was 5-13% lower than that running or cycling (Astrand and Saltin, 1961; Dixon and Faulkner, 1971; McArdle et aZ.~ 1971). The lower maximum oxygen uptake during swimming was attributed to differences in body position, active muscle mass involved, swimming skill, breathing pattern, swimming style, and the effect of water immersion. In these studies it was not possible to factor out the effect of water immersion per se on the cardiorespi­ ratory response to exercise because of the different types of work. Hence, it was not possible to conclude whether water immersion or other factors caused the observed reduction in maximum oxygen uptake. Magel and Faulkner (1967), however, reported no difference in maximum oxygen uptake between swimming and running. An interesting finding in their 4 study was that maximum oxygen uptake was maintained during swimming in the face of a significantly lower (p<.OOl) heart rate and expiratory minute volume. The lowering of these two parameters was also a con­ sistent finding in the other studies cited. Craig and Dvorak (1969) and Moore et al. (1970) used ergometers which could be submersed and compared the cardiorespiratory responses in air and water using the same type of exercise. In both studies oxygen uptake ranged from resting values to about 2 liters per minute. Their findings were in mutual agreement and showed that heart rate for a given submaximal oxygen uptake was approximately 4 to 30 beats per minute slower in water. Colder water potentiated a slowing of the exercise heart rate. There was no difference in expiratory minute volume between exercise in air or water. Lally (1973) confirmed the findings of Craig and Dvorak, and Moore, et al. Unfortunately in all three studies maximum oxygen uptake was not determined. Either the work rate was too light or the total working muscle mass was too small and precluded the attainment of full cardiorespiratory strain. Al­ though these studies did not answer the question concerning the effect of water immersion on maximum oxygen uptake they did show that water temperature could markedly effect exercise heart rate. In making the comparison between exercise in air and water it might be more informative to express various results as a percentage of maximum oxygen uptake rather than in absolute values of oxygen uptake or work load. For example, heart rate, pulmonary ventilation, blood pressure and regional blood flow are best related to the percentage of maximum oxygen uptake utilized. This permits the 5 assessment of the relative strain imposed and is therefore especially important when individuals with widely different aerobic power are compared (Astrand and Rodah1, 1970).

Statement --of the Problem To date, researchers interested in the aerobic power of swimmers have compared the maximum oxygen uptake measured while swimming with the value obtained by the same subjects running or cycling. While maximum oxygen uptake has been found to be the same or more often lower for swimming, peak heart rate and pulmonary ventilation have been reported to be significantly lower in swimming regardless of whether aerobic power is reduced. This is an interesting finding because heart rate and expiratory minute volume are important compo- nents of the oxygen transporting system. It would seem that reductions in these two parameters must be compensated by increases in stroke volume and/or arteriovenous oxygen difference and pulmonary oxygen ex- traction in order for oxygen uptake to remain unchanged. Whether the observed cardiorespiratory differences constitute adaptations specific to swimming or whether they are general effects of water immersion cannot be determined from swimming studies because of the different type of exercise used in air for comparison. The extent to which body position, the use of different muscles and thermal factors modify the cardiorespiratory response to maximal swimming must be considered. The starting point of the present investigation was to find a type of exercise that would elicit an individual's maximum oxygen uptake in water which also would be suitable for use in air. Any alterations in 6 maximum oxygen uptake or its related cardiorespiratory components during water immersion could not be attributed to differences in pos­ ture or active muscle mass. Water immersion and/or water temperature would have to be responsible for the observed changes. To separate these t",o influences water temperature could be easily varied and its effect evaluated. Using this approach it may be possible to determine the independent effect of immersion on the capacity of the oxygen transport system.

The Purpose

The purpose of this investigation was to compare the metabolic, cardiorespiratory and body temperature responses to upright, graded exercise in air and water. Of particular interest was whether maximum oxygen uptake is altered by head-out water immersion as compared to air when the same type of exercise is used. More specifically the purpose may be summarized as follows:

1. To develop an exercise machine suitable for eliciting maximum oxygen uptake in both air and water.

2. To measure the heart rate and ventilatory responses to submaximal and maximal exercise in air-and water.

3. To determine the ventilatory changes that occur with the use of SCUBA during mild to maximal exercise in water.

4. To evaluate the effect of v.Jater temperature on maximum oxygen uptake and its related cardiorespiratory parameters. 7

5. To determine the extent to which cardiac output during rest and exercise is altered by water immersion and water temperature.

6. To determine if possible what oxygen transporting factors might limit maximum oxygen uptake in water.

The Hypothesis

It is hypothesized that maximum oxygen uptake is unaltered by water' immersion in spite of significant reductions that may occur in heart rate and expiratory minute volume. In other words oxygen trans­ porting capacity must be maintained during maximal exercise in water by compensatory increases in stroke volume and the amount of oxygen extracted from the inspired air. Supporting this hypothesis is the finding that endurance training results in a reduced heart rate at a given oxygen uptake throughout the range of exercise tolerance (Holmgren et aZ., 1960; Hartley et aZ., 1969). These investigators observed that cardiac stroke volume in­ creases after training and compensates for the lower heart rate so that submaximal cardiac output is mostly unchanged. Perhaps a similar com­ pensation occurs acutely with water immersion. An increased stroke volume has in fact already been observed during rest and exercise in water (Dixon and Faulkner, 1971; Arborelius et aZ.~ 1972). Hence the lower heart rate in water may not indicate a reduction in cardiac output. Magel and Faulkner (1967) found maximum oxygen uptake was unchanged during swimming as compared to running, but expiratory minute volume was significantly reduced. This reduction was not found, however, during submaxima1 swimming (McArdle et aZ., 1971) which suggests that water 8 immersion may only affect the expiratory minute volume at high venti­ latory flow rates. Perhaps water immersion reduces the extent to which hyperventilation occurs at exercise intensities that elicit a high per­ centage of the maximum oxygen uptake. The mechanism for this reduction could be due to the greater work of breathing during immersion (Hong

et aZ. 3 1969). It is also hypothesized that water temperature will have a direct influence on the exercise heart rate. The assumption made here is that water temperature affects body temperature to the extent that heart rate not only reflects the subject's metabolic state but also his thermal status. A higher submaximal heart rate in warm water for example might indicate a greater cutaneous blood flow for head dissipation. Further­ more the greater thermal drive in warm water may result in a higher heart rate during maximal exercise. If an increase in "max imal" cardiac output also occurs, how would this affect maximum oxygen uptake? One might expect a greater blood flow to the active muscles to increase maximum oxygen uptake. However, if an increase in cardiac output does occur in warm water and the total increment is directed to the skin, maximum oxygen uptake would not be expected to change. Given the findings that peak cardiac output is higher in warm water but maximum oxygen uptake is unchanged, the pumping capacity of the heart would not appear to be a factor limiting aerobic power. Such an outcome would be very interesting since it would lend to the theory that peripheral factors such as muscle blood flow or metabolism are more important in the limitation of maximum oxygen uptake. 9

CHAPTER 2 MATERIAL AND GENERAL METHODS

Subjects

A general description of the seven male subjects who participated in these experiments is given in Appendix A. Those subjects who were involved in each separate study are described according to their updated anthropometric characteristics and maximal cardiorespiratory responses to treadmill exercise in the appropriate chapter. They ranged in age from 24 to 31 years and had a wide range of treadmill maximal oxygen uptake (Vo2max), from 2.51 to 4.89 L/min and 38 to 70 ml/kg-min. All subjects were associated with the School of Medicine, University of Hawaii at Manoa. Their physical activity habits varied from recreational exercises to competitive distance running (see Appendix A). None could be classified as truly sedentary. Subject OM, a middle distance runner, placed 6th in the 1972 United States Olympic Trials mile run. Subjects BR and RS were smokers and were the least active. Except for DM and LW, they had participated as subjects in a physical conditioning study and had undergone at least two maximal treadmill tests one month prior to beginning these experiments.

It was determined by duplicate tests that V02max did not change during the course of a single study. In fact, only DB managed to improve his value significantly over the two-year period, and that increase came in the first six months. Generally speaking it can be said that the subjects maintained a steady level of physical condition­ ing throughout the testing period. 10

They were knowledgeable as to the conduct of the experiments and were judged highly motivated. The subjects were recommended to get plenty of rest the day before and to eat a light meal on the morning of a testing session, but no attempt was made to control such pre-test variables. They wore gym shorts and track flats for conditions in air. In water they wore a nylon swim suit and a 4-kilogram weight belt.

Methodology

Testing conditions. Testing was conducted in an air-conditioned laboY'atory with subjects breathing room air or, in SCUBA experiments, compressed air. The inspired fraction of oxygen (FI02) in both cases ranged from 20.84 to 20.95%, with a mean ± S.E. of 20.90 ± 0.01% (N=27). The inspired fraction of carbon dioxide (FIC02) ranged from 0.01 to 0.05%, with a mean of 0.03 ± 0.002% (N=27). Room air temperature was 22 to 25 C. Barometric pressure was 758 to 766 mm Hg. Relative humidity was 40 to 65%. For immersion experiments water temperature was 25, 30, or 35 C with an average standard error of ± 0.2 C. Most of the testing sessions were held between 0900 and 1200 hours.

Mode of exercise. Treadmill exercise was performed on a Quinton Treadmill, model SQ-18. Speed varied from 80 to 200 m/min and grade from 0 to 14°, depending on the ability of the subject and the desired work level. A Monarch mechanically-braked ergometer was used for cycling in air. Work load was adjusted by changing the tension of the friction belt on the flywheel (Astrand, 1960). Individual work load settings were determined by the desired relative work load, assuming 11 a constant gross mechanical efficiency of 21%. Pedal frequency in air was 60 rpm for submaximal exercise. Since pedal frequencies below 60 rpm and above 70 rpm have been shown to influence the value obtained for V02max (Hel"mansen and Saltin, 1969) the rpm for maximal cycling in air was matched (± 5 rpm) to that used in water. A similar ergometer was used in water. The calibration and ener­ getics of this underwater exerciser are discussed in Chapter 3. Body position for cycZing exercise was the same for both air and water. Seat height was adjusted so that the knee was extended about 165 0 in the down position of the pedal. The subject cycled in an upright sitting position. However, his hands were placed on the footrests rather than the handle­ bars (Fig. 1). This posture is very similar to the sprinting position taken by a bicycle racer. The legs are vertical, but due to the lower position of the hands the trunk is nearly horizontal. This posture is quite comfortable and subjectively does not restrict respiratory or general bicycling movements. It was used in order to simulate the upper body position of a surface swimmer. For immersion experiments the water level was adjusted so that the subject was submersed to the sternal notch. However, due to the nearly horizontal position of the trunk, the upper part of the back and some­ times the shoulders were above water. The subject1s chin was always in the water. The position in water was, therefore, a modified type of head-out immersion. The hydrostatic pressure on the thorax was probably similar to that for surface swimming and less than that for the conven­ tional position of head-out immersion (Hong et aZ.~ 1969). FIGURE 1 Cut-Away Diagram of Immersion Tank. This figure shows the subject1s exercising position in water, which was the same for air. Note subject1s hands are on footrests, causing the upper body to be some~Jhat horizontal. The friction belt was removed from the underwater ergometer. N 13

Respiratory measurements. The subject breathed through a Collins Triple-J valve which has high-velocity, low-resistance characteristics. The inspiratory and expiratory resistance of this valve have been determined to be zero cm H20 at 112 Ljmin (Bartlett et aZ.~ 1972). Since the valve floated on the water during immersion experiments, a 40 cm length of corrugated plastic tubing (3.8 cm internal diameter) was fitted to the inlet side of the valve to prevent the subject from inspiring water. In order to make the two conditions equal, the hose was also used for cycling experiments in air. Expired gas passed through the one-way breathing valve and into an 85 cm length of the corrugated plastic tubing. Two methods for collecting the expired gas were used. The first method (referred to here as aliquot sampling) consisted of passing the expired gas from the tubing through a Parkinson-Cowan, CD-4, high-velocity, low resistance flowmeter where the minute volume was read by eye and verified from the polygraph readout of a linear potentiometer fitted on the meter. The gas then flowed into a 5.5 liter baffled mixing chamber which was seated on the outlet side of the flowmeter. Gas temperature was sensed by a Yellow Springs thermistor probe situated in the mixing chamber near the junction between the meter and the chamber. An aliquot sample of the minute volume was continuously withdrawn from the mixing chamber at 1 liter per minute and pumped into a 2 liter butyl rubber bag via a Beckman microcatheter sample pump. The rest of the expired gas passed out of the chamber and mixed with the room air. The smallest diameter of the external airway was 3.8 cm, which was the internal diameter of the tubing. The gas contents of the rubber bag were analyzed within 14

15 min for FE021 and FEC02 using a Beckman E-2 oxygen analyzer and a Beckman LB-l carbon dioxide analyzer, respectively. The second method consisted of collecting the entire expired minute volume as described by Daniels (1971) in 150 liter meteorological balloons. The fractions of 02 and C02 in the balloon were then deter­ mined by sampling 1 liter of the entire mixed gas volume using a Beckman OM-ll oxygen analyzer and the Beckman LB-l carbon dioxide analyzer. The gas volume and temperature were measured in a balanced 350 liter Tissot gasometer. Comparisons between the two methods for sampling and analyzing the expired gases differed by no more than ± 0.05% for either 02 or C02. Simultaneous measurements taken during rest and steady state exercise showed no significant difference in the calculation of VE' V0 ' or VC02' 2 Respiratory frequency was recorded continuously on a Beckman 4-channel polygraph by means of a small thermister placed in the breath­ ing valve. The rate was counted for the entire minute of gas collection.

Equations used for calculating respiratory ~ exchange. Oxygen uptake (V02) was determined using the open circuit method according to the principles outlined by Consolazio, Johnson, and Pecora (1963). The following equation (referred to here as the Haldane equation) was used to calculate V02:

lRefer to List of Abbreviations in this chapter for definition of symbols. 15 where FEN2 was assumed to be equal to 1.00 -(FE02 + FEC02) and FIN2 equal to 1.00 - (FI02 + FIC02) .

The usefulness of this equation is that the inspiratory minute volume need not be measured. However, its validity is based on the assumption that:

which is called the Haldane transformation (Haldane, 1912). This assumption has been questioned by Dudka et aZ. (1971) and by Cissik et aZ. (1972). However, recent experiments by Wagner et aZ. (1973), Luft et aZ. (1973), and Wilmore and Costi11 (1973) have re-confirmed the adequacy of the Haldane transformation for estimating respiratory gas exchange during various conditions of rest and exercise. ~C02 The equation also corrects for R, i.e., the ratio: / V02 . The values for FI02 and FIC02 were determined by repeated measurements using the Scho1ander method (Scho1ander, 1947) and are 0.2090 and 0.0003, respectively. The volume of carbon dioxide eliminated (VC02) was calculated from:

assuming that the volume of inspired carbon dioxide is very small and can be neglected. Reduction of VE from ATPS to STPD was calculated assuming complete saturation with water vapor at the measured temperature. The actual body temperature (Tre) was not used in the conversion from ATPS and STPS 16

since the range for Tre was 35.8 to 38.1 C, and the correction would be small. Alveolar ventilation was calculated from:

where the physiological dead space volume (VO) in ml was assumed to be equal to the subject1s height in em plus the dead space of the respira­ tory valve (see Chapter 5). Measured values for valve dead space were 315 ml (Collins Trip1e-J), 80 ml (SCUBA), and 75 ml (Oanie1s).

Respiratory calibrations. Reference gases from cylinders of mixed gases of known concentration (Scholander analysis) were used to cali­ brate the Beckman E-2 and LB-1 gas analyzers before and after each experiment. The analyzers were remarkably stable and on only a few occasions were "drift" corrections necessary. Calibration of the Beckman OM-ll oxygen analyzer was accomplished by simply adjusting the instrument to read 20.9% 02 while sampling room air. The accuracy of the electronic analyzers was routinely verified by paired analysis with the Scholander micro-gas analyzer. The Parkinson-Cowan gasmeter was calibrated by filling the Tissot gasometer with room air and emptying it into the gasmeter using pul­ satile flow rates of 25, 50, and 100 L/min. Ten determinations at each rate were made on several occasions. The gasmeter correction factor was constant at 0.981 for the flow rates tested.

Cardiovascular measurements. Heart rate was determined from a continuous EKG record obtained from precordial surface leads by counting 17

the number of R-waves over the full minute of expired gas collection. Immediately following maximal exercise the paper speed was increased to 25 mm/sec in order to observe the EKG waveform. No obvious irreg­ ularities in the EKG were observed. Cardiac output was calculated from a beat-to-beat estimate of stroke volume using impedance cardiography and a beat-to-beat measure­ ment of the R-R interval obtained from the EKG. This method is described in Chapter 7. Arterio-venous oxygen difference was calculated from the Fick equation:

Venous blood lactate and pyruvate were determined by enzymatic analysis. The method is described in Chapter 6.

Body temperature. Rectal temperature was monitored on a Yellow Springs tele-thermometer with an expanded scale using a calibrated thermistor probe (No. 402) which was inserted in the rectum to a depth of 12-15 em. All subjects used the same probe. Rectal temperature was measured during the period of expired gas collection and at 5 min post­ maximal exercise. It was not measured during treadmill exercise.

Arlthropometric measurements. Physical data include age, height, body weight, Dubois body surface area, percent fat, mean skinfold thick­ ness, vital capacity, and residual volume. This information is presented in Appendix A and in the tables which describe the subjects. 18

Body weight was measured to the nearest 0.05 Kg using a Fairbanks­ Morse platform scale with the subject wearing a nylon tank suit. Re­ ported weights are for the day of treadmill testing. Percent fat was determined by the skinfold technique (Sloan and Weir, 1970) and by

hydrostatic weighing (Brozek et al. 3 1963). Mean skinfold thickness was determined from the unweighted mean of 10 sites: head, chest, back, stomach, upper arm, lower arm, hand, thigh, calf, and foot. Vital capacity was measured with a 9-liter Collins spirometer. Residual

volume was estimated by a three-breath method (Rahn et aZ. 3 1949).

Reliability of measurements. It was especially important for the instrumentation to be reliable on a day-to-day basis because of the com­ parative nature of these investigations. Therefore, the reliability of a measurement was judged from its reproducibility in a series of trials on separate days. The reliability of gas analysis was determined by analyzing a reference gas from a cylinder containing oxygen and carbon dioxide in physiological concentrations. The reliability of volumetric analysis was determined by periodically checking the calibration of the Parkinson-Cowan gasmeter against the Tissot gasometer. The Tissot was checked for leaks, level, and counterbalance each test day. The meteo­ rological balloons were routinely checked for leaks. Subject variability was checked by the test-retest method on deter­ minations for maximum oxygen uptake, and resting heart rate, stroke volume, and cardiac output. The results of these precautions are shown in Table 1, and for the least variable subject, JM, in Table 2. TABLE 1 RELIABILITY OF QUANTITATIVE DETERMINATIONS

Measurement Method Principle Number ± S.L 1. 02 concentration a. Scholander analyzer chemical absorption 27 0.02 vol% b. Beckman E-2 paramagnetism 15 0.05 vol% c. Beckman OM-ll polarography 13 0.02 vol% 2. C02 concentration a. Scholander analyzer chemical absorption 27 0.03 vol% b. Beckman LB-l infra-red absorption 28 0.03 vol% 3. gas volume a. 350-1 iter Tissot volumetric spirometer b. Parkinson-Cowan gas- volumetric 30 0.1 L/min at meter 100 L/min 4. maximum oxygen a. Douglas bag method open-circuit spirometry 43 1-2% uptake b. aliquot sampling open-circuit spirometry 17 1-2% 5. heart rate surface electrodes electrocardiography 6. stroke volume impedance cardiography transthoracic impedance 21 4 ml (resting)

--' 1.0 20 TABLE 2 REPRODUCIBILITY OF REPEATED DETERMINATIONS ON SEPARATE DAYS

Measurement Condition Number ± S.E. l. Heart rate sitting at rest 4 2 bpm (24 C air)

2. stroke volume II 4 3 ml

3. cardiac output II 4 O. 1 L/min 4. maximum 02 uptake cycling 7 O. 1 L/min (22-25 C air)

5. peak heart rate II 7 bpm

6. peak expiratory II 7 3.8 L/min minute volume 21

Experimental Protocol

The experimental protocol is shown in Fig. 2. The exercise regimen is a modified Taylor protocol for determining maximum oxygen uptake

(Taylor et aZ. 3 1955). The basic design is the same, i.e., a discon­ tinuous series of increasing work loads. However, there are four modi­ fications. The Taylor protocol utilizes a 15-min warm-up, 3-min exercise bouts, fixed work loads that have an oxygen-cost near or above the pre­ dicted maximum oxygen uptake, and prolonged rest periods between each work load. Since it was desired to make submaximal steady-state measure­ ments, the following modifications were made: 1) 4 submaximal work loads were substituted for the warm-up, 2) 5-min exercise bouts were used, 3) work loads were adjusted to elicit 20, 35, 50, 70, and 100% of each subject1s predicted maximum oxygen uptake, and 4) there were 2-min rest periods between each load. Gas collections were made from minutes 3 to 4 and 4 to 5 in order to assure that a metabolic steady-state had been reached. The relative work loads were adjusted using the relationships between oxygen uptake and the index of work rate for the particular exercise as shown in Fig. 3. This technique assumes that all subjects have the same degree of mechanical efficiency for a given type of exercise. This assumption was found to be within acceptable limits for cycling, both in air and water. In other words, the skill factor is small in stationary cycling exercise. However, in treadmill exercise the oxygen cost for a given grade and speed varies considerably between individuals. The coefficient of variation may range from 8 to 15 percent x x 100

70

x « :::: 50 (\,j o .> 35 o~ 20 i I

- '--- o 5 7 12 14 19 2\ 26--28 33----35 40

tim e, min

FIGURE 2 Experimental Protocol. Exercise work rate was adjusted to produce a given % V02 max, as shown in the rectangular columns. IIX" indicates a l-min gas collection period. Following rest, work and recovery periods were 5-min N and 2-min, respectively. N 23

Bicycle Ergometer (Air) Bicycle Ergomete r (Water)

V , -max o2 L/min V. at­VI 0a 450

W,(Kpm/min) W,(rpm)

Treadmill

-max Vo , 2 ml/Kg·min

W, (speed, 9 ra de)

FIGURE 3 Oxygen Uptake as a Function of Aerobic Work Rate. Upper left figure shows the linear relationship for cycling in air. The slope is approximately lL per 450 kpm/min. Upper right figure shows curvilinear relationship for cycling in water with the friction belt removed. Power output in this case is indexed as rpm. ~02 is roughly proportional to rpm 3. Lower figure shows linear relationship for treadmill exercise. In this case.V02 is proportional to the speed (v) and grade (e). V02 max is represented as the point on the V-axis where V02 does not increase with W. 24

(I. Astrand, 1960; Shepard, 1969). Grade and speed settings were mostly, therefore, adjusted by trial-and-error. In general, the recommendations made by the Minnesota group for administering maximal exercise tests were followed (Taylor et aZ.~ 1963; Taylor et aZ.~ 1969). Accurate prediction of the maximal work load required prior submaximal testing and some knowledge of the subject's endurance performance capacity. The maximal work load is important because it is known to affect the rate of anaerobic metabolism (Karlsson, 1971) and consequently lactic acid production, pulmonary ventilation, and the time of exhaustion. In this study the subject1s treadmill

V02max was known before the experiments began. Since the treadmill V02max is the same or usually higher than for cycling, the former was used to predict the correct load setting to elicit the latter. The final work load chosen for cycling experiments \'Jas actually IIsupra­

11 maxima1 , i.e., it had an estimated oxygen cost of about 105% of the subject's V02max. The subject could not, therefore, reach a metabolic steady-state at this load because of the high anaerobic component. All subjects were, however, able to complete the 5-min work period. In some cases the work rate was observed to decline as the subject ap~roached exhaustion. After each test the subject was questioned as to whether a maximal effort was achieved. Ninety percent of the time the answer was positive, and the subject's physical appearance was consistent with this. Ten percent of the tests had to be re-administered. The Taylor criterion for absolute plateauing (Taylor et aZ.~ 1955) of oxygen uptake between the 4th and 5th minutes was observed in about 40% of the tests. 25

The 002 during the 4th minute was within 0.15 L/min of the final value in about 95% of the tests. A plateauing of heart rate or expired minute volume was rarely observed. Only full minute gas collections were made. Total experimental time averaged 50 minutes, which includes the 5-min recovery period following the final work load. Water exposure times at the various water temperatures were not significantly different.

Statistics

The arithmetic mean (x), standard deviation (SO), and the standard error of the mean (SE) were calculateo according to ordinary statistical methods (Snedecor and Cochran, 1967). Statistical significance was tested on intra-individual differences of paired observations by use of Student's "t"-test. The conventional level of statistical significance, p<0.05, was used and is denominated in the text by an asterisk (*). Linear regression analysis was used to describe one variable (y) as a function of another (x). The regression equation of y and x is:

y = y + b (x - x) = a + bX where x and y denote the mean of the independent and the dependent variables; "a" is a constant sometimes called the y-intercept, and "b" is the regression coefficient or slope obtained by:

L (x-x) (y-y) b = L (x-x)2

Curvilinear regression analysis was performed by an IBM computer utilizing a stepwise regression program (BMO 02R, University of California 26

1971, based upon Efroymson, 1960) on the polynomial equation:

where Co' Cl ' etc., are regression coefficients. For correlation analysis the coefficient of correlation (r) was obtained by:

The accuracy of a single determination of the various methods was calculated by 10 or more separate analyses of the same sample. The error of the methods was estimated from the standard error of the mean of these separate analyses. 27

------List of Abbreviations

Symbol Dimension Definition

Respiration a Arterial (subscript) Alveolar (subscript) ml/100 ml Arteriovenous oxygen content difference breaths/min Breathing frequency Fraction of expired gas Fraction of inspired gas L/min Maximum voluntary ventilation (free choice of respiratory' frequency) Fraction of oxygen removed from the inspired air

R Respiratory gas exchange ratio

RV L Residual lung volume v Venous v- Mixed venous L/min Alveolar ventilation

L Vital capacity

L Respiratory dead space volume L/min Expiratory minute volume Ventilatory equivalent for oxygen L/min Submaximal oxygen uptake 28

---List of Abbreviations (Continued)

Symbols Dimensions Definition

L/min or Maximal oxygen uptake ml/kgomin % Percent of maximal oxygen uptake

L Tidal Volume

Circulation

BP mmHg Blood pressure EKG Electrocardiogram f h beats/min Heart rate 02 pulse ml/beat Oxyg2n pulse L/min Cardiac output

Qs ml Stroke volume TPR mmHg/ml/min Total peripheral vascular resistance

Other Terms

BTPS Body temperature and pressure, saturated with vapor BSA Dubois body surface area B~l (or Wt) kg Body weight

% fat % Percent of BW as fat DB gm/cc Body density [La] mg % Lactate concentration [pva] mg % Pyruvate concentration 29

---List of Abbreviations (Continued)

Symbols Dimensions Definition

SCUBA Self-contained underwater breathing apparatus STPD Standard temperature and pressure, dry gas T C Temperature (centigrade) C Ambient temperature

C Rectal temperature

C Water temperature kpm/min or watt Work rate 30

CHAPTER 3 THE MONARK BICYCLE ERGOMETER AS A SUITABLE UNDERWATER EXERCISER

Introduction

Bicycle ergometers, particularly the Monarch (Varberg, Sweden), are widely used to impose known work loads on exercising subjects. The power output is expressed in kpm/min and is the product of the linear velocity of the flywheel circumference and the torque acting on the flywheel braking system (von Dobe1n, 1954). Although transmission frictional losses do occur, they are small and constant. Wyndham et aZ. (1966) suggest a net efficiency of 25%, however, the coefficient of variation may be as high as 5 percent. At first glance it seemed the Monarch ergometer might provide the ideal type of exercise in water which could easily be duplicated in air. It was anticipated that the gross efficiency of using the ergometer in water would be reduced due to the drag forces of water acting to brake the motion of the legs, pedals, chain, and flywheel. But since Craig and Dvorak (1969) and Moore et aZ. (1970) using ergometers with an external transmission had shown no reduction in overall mechanical efficiency during immersion in 30 and 35 C water, it could be assumed that immersion per se does not affect the subject's internal efficiency, which has been defined as "physiologica1 efficiency" by Whipp and Wasserman (1969). Furthermore, differences in gross efficiency can be overcome by expressing the observed parameters, such as heart rate, at a given V02 or at a given % V02 max. The main concern was whether the 31

immersed ergometer would be suitable for light to maximal exercise. The purpose of this study was to calibrate the ergometer in terms of oxygen uptake versus an index of work output.

Methods

A description of each of the six subjects is presented in Table 3. The cardiorespiratory data shown refer to maximal effort treadmill walking. Initially the ergometer was simply immersed in a fiberg1assed wooden tank, 152 cm long, 64 cm wide, and 146 cm deep (see Fig. 1). Preliminary testing revealed that setting the load and counting pedal revolutions with the ergometer underwater were difficult and impractical. Also the breaking forces of the water Oil the "man -to-friction be1t" transmission system at a pedal frequency of 60 rpm were much greater than estimated. The following modifications were, therefore, made: 1) the ergometer was fitted with a magnetic reed switch which was wired directly to the Beckman recorder in order to count pedal frequency, 2) standard auto­ motive grease nipples were mounted on the crank and flywheel to provide easy regreasing of the bearings, 3~ lead weights were fastened to the rear stand to prevent the ergometer from moving during heavy work, and most important, 4) the friction belt was completely removed. For the calibration experiment the subject breathed from a Collins Trip1e-J valve or from SCUBA. For the latter, a standard 71 ft3 com­ pressed air bottle was mounted on the inside wall of the tank. The U.S. Divers 2-stage, double-hose regulator was positioned in the water at the point that just prevented free-flow. One minute expired gas collections 32 were taken using the aliquot sampling system previously described. After entering the water the subject sat quietly for 5 min after which a 2 min resting gas collection was taken. Keeping pace with a metronome, he then proceeded to pedal as close as possible to 20, 40, 50, 60, 70 and/or maximum rpm for 5 min each, with 2 min recovery intervals. The water temperature was 30 ± 0.2 C. Immersion time was approximately 52 min. The subjects were also tested for "no-load" cycling in air at 20, 40, 50, and 60 rpm.

Results

The results of the \102 versus rpm relationship for "no load ll cycling in air and water are shown in Fig. 4. Some subjects were tested mare than once. Only three subjects (DB, RD, JM) were able to pedal at 70 rpm for the full 5 min and two (RD, JM) at 80 rpm. In total, 49 data points were obtained for Collins valve breathing and 38 for SCUBA breathing. A polynomial regression analysis of the following form was applied to both sets of data individually.

where Co Cl Cn are regression coefficients and f is pedal frequency in rpm. Essentially the same equation was obtained for each set, indicating SCUBA breathing does not significantly alter the V0 2­ rpm relatiunship. The data were consequently pooled. The combined data points and best fit equation are shown in Fig. 5. The best fit (r ~ 0.996) resulted when n=3. The coefficients (± SE) are: 33

5

00 o 4 i' og 5 <1'0 0.. r- ~ c: "E °0 "- 3 o ...J

-<\I "~

(l) ~ o 0 a. 2 " Collins valve :::J o SCUBA c: (l) Ol >. o )( 0 ~ o : <1' \"no load" cycling in air

91. ~- o~ _--- ... - a - _el------F(gI--

o 10 20 30 40 50 60 70 80

Pedal Frequency (f) • Rev/min

FIGURE 4 Oxygen Uptake as a Function of Pedal Frequency for "No­ Load" Cycling in Hater and Air. The friction belt \'Jas removed in both cases. Greater braking force of water becomes apparent at 30 rpm and above. The use of SCUBA does not affect the relationship. 34

OF UNDERWATER CYCLING 5 OXYGEN COST

. 2 Vo = .274 + .002025 (f) - .000059 (f) 2 +.000008 (f)3 • •. 4 4l • ., .. (r=.996) • 0 a- t- . U) : 0 , ..-.. . c . 'E • J3 0

uS ,p ~ . « I- a.. ~ .. z2 ·f w ••4. (!) co·•• r • 0 X 0 ., .,~.. • • ..~. 0 cIJ II 0 ,~

. , i 0 10 20 30 40 50 60 70 80 PEDAL FREQUENCY, (f) rev/min

FIGURE 5 Regression Eq~ati0n of Oxygen Uptake as a Function of Pedal Frequency. R~sting ~02 plus f3 term explains 99% of the variance in V0 2· 3.0

./B 0-E-\ , . " " ,. " ,. ./" ~~" " ,. " " ,. , " ,. " " ,. 2.0 -j1O;:>'/ ,. I " ,. ,. ,. " ,. " " '" ~~,. " " " V ) 0' " ,. '" " 02 fj ,," ,. " ,. '" L/min " " ~~,. ,. '" " ,," ~"0 <.". ~.s-,." " ,." 0<'/ 1.0-1 ,.,.,,?",. o RS o BR " ,. ,. / ,. ,. 11 LW

I I 300 600 900 W ,Kpm/min

. FIGURE 6 Oxygen U~take as.a Function of Work Load During Cycling in Air and 30 C Water. Line.AB. relates VO Z and Wat rest and exercise (60 rpm) in air. Broken lines represent the VOZ-W relationshlP.in water at a given rpm. Point A is resting V02 in air or water. Point B is the mean VO Z max for these subjects. Point C i? the V02 for "no-load" cycling in air w at 60 rpm and ln water at 30 rpm. Point D is the VO, max ln water. Point E is the ()"l oxygen cost of the load setting during maximal exerClse in water. 36 3 Co = + .274 ± 0.80 x 10- 4 C1 = + .002025 ± 0.13 x 10- 7 C2 - - .000059 ± 0.13 x 10- 12 C3 = + .000008 ± 0.86 x 10-

In addition, data were obtained with three subjects cycling in air (N=12) and water (N=22) during which the friction belt was on and the power output of the ergometer observed. The results for cycling in air are shown in Fig. 6 along with the theoretical relationship between

V02 and Win water.

Discussion

The significance of each order of the f term and its coefficient in the regression equation has been discussed by Morlock and Dressen­ dorfer (1974). In brief, each f term theoretically represents the oxygen uptake or metabolic power required to perform a given type of work. The fO term approximates the oxygen uptake observed for resting metabolism. The fl term is probably the power required to overcome external friction. And the f2 and f3 terms are thought to represent the power required to overcome fluid drag forces, where f2 is pro­ portional to viscous drag and f3 is proportional to inertial drag. Because the value of the f1 and f2 terms are approximately equal but opposite in sign, their individual importance is cancelled from the equation. In fact, a standard stepwise polynomial regression analysis o showed that 99% of the variance in V02 can be explained by the COf and C3f3 terms alone. In this regard, it is of interest to note that Whitt (1971) found the power requirement of bicycling on the road to also be 37 a cubic function of speed. Most of the inertial drag is probably caused by the flywheel. Due to the 52:14 gear ratio of the ergometer, it is spinning 3.7 times faster than the subject is pedaling. It is, therefore, postulated that the inertial drag on the flywheel is the cause of most of the oxygen cost of underwater cycling when using a Monark-type ergometer. This hypothesis would explain the small inter-subject variation from the regression line since flywheel drag depends only on rpm. Denison et aZ. (1972) used a similar ergometer for underwater exercise. They reported an oxygen uptake of approximately 0.75 L/min for unloaded cycling at 60 rpm in water. The V02/rpm relationship found in the present study would predict a V0 2 of 2.0 L/min for the same fre­ quency. The flywheel on the Monarch ergometer, however, is about 32 cm greater in circumference than the one used by Denison. 2 The Monarch flywheel consequently travels the same distance in 49 revolutions as does the Denison flywheel in 60. The predicted V0 2 for 49 rpm is 1.2 L/min. Furthermore, inertial fluid drag includes the concept of frontal area normal to the relative velocity of the moving fluid (Dailey and Harleman, 1966). Some provision must, therefore, be made to account for the differences in either total surface area or thickness between the two flywheels. In either case the ratio of the Monarch flywheel to that used by Denison is about 1.5. Dividing 1.2 L/min by 1.5 yields a V0 of 0.8 L/min which is close to that observed by Denison et aZ. (1972). 2

2The flywheel diameter is 42 cm for the one used by Denison and 52 cm for the Monarch. 38

The greater size of the Monarch flywheel, therefore, appears to account for the observed discrepancy. The results shown in Fig. 6 are particularly interesting because they permit a discussion of the theoretical efficiency of underwater

cycling. Line AB represents the observed relationship between V0 2 and Wfound in three subjects during cycling in air. The slope of the line. gives a net efficiency of 22% which agrees with the findings of I. Astrand (1960). The broken lines depict the theoretical V0 2/W rela- tionship for cycling in water at the various rpm shown. These lines were derived in the following manner: 1) the observed Wduring under­ water cycling was first multiplied by 0.9 to correct for the buoyant force of water acting on the 1 kg pendulum weight of the ergometer,3

2) since the V0 2 of underwater cycling is affected by both Wand rpm, the V02 due to the Walone was estimated by subtracting the V0 2 for the observed rpm (calculated from the regression. equation for "no-load" cycling shown in Fig. 5) yielding a ~V02' which represents the oxygen cost for the observed W, 3) the ~V02 for cycling in air was determined using the same methods as in (2) (note point C on the graph represents "no-load" cycling in air at 60 rpm), 4) the slope of ~V02/W was found by linear regression analysis to be the same in water as air, suggesting that the internal or "physiological efficiency" (~Jhipp and Wasserman,

1969) is not impaired by immersion, 5) the V0 2/W lines for underwater cycling were, therefore, drawn parallel to AB, with the y-intercept equal to the V02 required for "no-load" cycling at the particular rpm.

3Bouyant force = volume of pendulum weight x density of water at 30 C. Correction factor (i.e., 0.9) = weight of pendulum in air--bouyant force. 39

The theoretical gross efficiency of underwater work on a stationary ergometer consequently depends on the rate that the various parts of the drive train move through the water. At any oxygen uptake of 2 L/min the gross efficiency of underwater cycling is 16, 13, 8, and 1% at 30, 40, 50, and 60 rpm, respectively, compared to 19% in air. Thus, useful or external work appears "to decline rapidly in water when cyclic limb move­ ments exceed about 30-40 rpm. In swimming the maximal rate of stroke frequency may be lower than cycling. McArdle et aZ. (1971) reported that none of their varsity swimmers (mean V0 2max of 3.36 L/min) could swim for longer than 2 min at 50 strokes/min. The finding that SCUBA breathing does not significantly affect the

V02/rpm relationship is not surprising since the oxygen cost of breath­ ing is reportedly small compared to the total V0 2 of exercise (Otis,. 1964). This matter will be treated in a greater detail in Chapter 5. In summary, the braking forces of water acting alone are enough to produce a convenient relationship between V0 2 and ur.loaded cycling rpm. When the Monark ergometer is used in w~ter with the friction belt re­ moved, the external work performed by the subject is not known in con­ ventional units of power, but it can L~ accurate1y indexed by pedal frequency. The ergometer, therefore, appears to meet the requirements for comparing graded exercise in water with that in air. 40

CHAPTER 4 COMPARISON OF CARDIORESPIRATORY RESPONSES TO GRADED CYCLING EXERCISE IN AIR AND WATER

Introduction

The effect of water immersion per se on the maximal capacity of the oxygen transporting system has not previously been determined. Several studies have compared swimming with treadmill or cycling exercise and found a lower ~02max for swimming. There are a number of possible explanations for the difference in maximum aerobic power in swimming versus running or cycling. Differences occur in the size of the exercising muscle mass, body position, and degree of swimming skill. These differences, however, may preclude an objective deter­ mination of the effect of immersion. In addition, the reasons why the findings in the literature are not always in agreement may be explained by differences. in the type of exercise used for comparison . Mean V0 2max during swimming has been reported as 1, 5, 6, 11, and 19% lower than for running by Magel and Faulkner (1967), Dixon. and Faulkner (1971), Holmer et aZ. (1974), Holmer (1972), and Astrand and Saltin. (1961), respectively. McArdle et al. (1967) found swimming V0 2max was 10% lower than that for uphill walking. Compared to cycling, Astrand and Saltin (1961) found a 13% lower V02max for swimming whereas Holmer (1972) found it was only 2% lower. Of the six studies cited which compared swimming to treadmill exercise there were a total of 55 subjects. After adjusting for the unequal sample size of these six experiments, the overall mean V0 2max was calculated to be 7% lower for 41

swimming; i.e., 3.99 L/min vs. 4.30 L/min for treadmill work. There is general agreement of findings in the above studies that

the lower swimming V0 2max is attended by significant t'eductions in peak heart rate, expiratory minute volume, and respiratory exchange coeffi­

cient. Holmer et al. (1974) reported that the 6% decrease in v02max (0.28 L/min) was attended by an 8% decrease in maximal expiratory minute volume (30 L/min). Magel and Faulkner (1967) found only a 1%

reduction (0.06 L/min) in V0 2max for swimming compared to running, but peak heart rate was 6% lower (12 bpm) and maximal expiratory minute volume 17% lower (22 L/min). It appears from these findings that heart rate and expiratory minute volume are lowered disproportionately to the

reduction in v02max. McArdle et al. (1971) studied submaximal swimming and walking and found heart rate averaged 4-11 bpm lower in swimming at any given level of oxygen uptake. Statistical significance, on the other hand, \'Jas demonstrated only at work levels requiring an oxygen uptake of 2.24 L/min (67% of swimming v02max) or higher. This agrees with the findings of Craig and Dvorak (1969) and Moore et al. (1970j, and suggests a true immersion effect which is independent of the type of exercise. The question arises as to whether maximal heart rate and expiratory minute volume are actually suppressed during swimming or whether other factors become limiting before full card"iorespiratory strain is ex­ perienced. If the latter case is true, the term "pea kll is more appro­ priate than maximal in that there still may be some reserve. The purpose of the present study was to compare the cardiorespira­ tory responses of subjects exercising on the Monark ergometer in both 42 air and water. The primary objective was to determine the maximal responses for each condition. Secondary objectives were to measure heart rate and expired minute volume at equal submaxima1 oxygen uptakes and to express the findings in terms of % ~02 max. A water temperature of 30 C was chosen in order to compare the results with those in the literature (Craig and Dvorak, 1969; Moore et aZ., 1970), and also be­ cause it was found in pilot experiments that exercising subjects maintained rectal temperatures of 37 ± 0.5 C. The aim was to avoid the complications of either a cold or heat stress. It was not planned for 30 C water to be the thermal equivalent of 25 C air.

Methods

The six male subjects who participated in this study are described in Table 3, (Chapter 3). Data were collected on subject DB only for maximal exercise. These subjects were familiar with the cycling exercise in water and had completed at least one maximal-effort cycling test during head-out immersion prior to the study. The subject wore a 4 kg weight belt for immersion experiments. Oxygen uptake was measured using the aliquot sampling system. Other parameters were measured with methods previously described. Maximal voluntary ventilation (MVVF) was determined using a 15 sec test, the subject having a free choice of respiratory frequency. MVVF was measured with the subject in the exercising position, 5 min following the end of maximal exercise. The same gas collection equipment was used as in exercise testing. The best of three trials was recorded as MVVF· TABLE 3 DESCRIPTION OF SUBJECTS . b . Age Ht vJt V.C. V0 2max fh VE Initials yrs cm kg %Fata L,BTPS L/min ml/kgxmin bpm L/min BTPS DB 24 180 77.5 21 6.1 4.22 54 182 166.9 RD 29 178 71.9 9 6.8 4.76 66 171 177.3 JM 26 184 70.7 11 5.5 3.68 52 168 134.2 BR 29 170 55.4 16 4. 1 2.51 38 172 98.8 RS 29 172 66.2 12 6.5 2.73 41 171 100.4 LW 24 164 58.2 13 4.2 2.64 45 191 100.5

Mean 27 175 68.3 14 5.5 3.42 49 176 129.7 askinfold technique, after Sloan and Weir (1970) btreadmill exercise

~ w 44

Water temperature was 30 ± 0.1 C. Total immersion time was

50 ± 2 min. The modified Taylor protocol described in Chapter 2 was followed.

Results

The individual results of this study are presented in Appendix B. Submaxima1 data was not obtained for subject DB. Mean data, therefore, represents a sample size of 5, except in Fig. 15 where the results for maximal exercise also include DB. Significant differences (p<.05) be­ tween air and water values are marked by an asterisk. In general, the selected submaximal exercise intensities for each subject closely approximated the desired relative work loads of 20, 35,

50, and 70% V02max. Mean submaximal responses for rectal temperature and heart rate are shown in Table 4 and Fig. 7. Rectal temperature was significantly lower in water at relative work loads of 35% V02max and above. Heart rate was lower in water at any given oxygen uptake, but the difference (5 to 7 bpm) was significant only at 70% v02max (2.1 L/min), or higher. The individual data for heart rate versus % V0 2max are shown in Figs. 8 and 9. A typical linear relationship was found for both conditions. The slopes of the regression lines are nearly equal, but the y-intercept is lower in water, indicating a parallel shift to the right. Oxygen pulse versus %V02max is shown in Fig. 10. Differences in the two curves are not significant, indicating that the product of stroke volume and arteriovenous oxygen difference increases in a similar manner in both air and water. 45

Mean ventilatory responses are shown in Table 4 and Fig. 11. There is a trend for R to be lower in water at all work levels. No significant differences were found for submaxima1 expiratory minute volume or breathing frequency. Individual data for ~E versus ~02 are shown in Figs. 12 and 13. Regression analysis revealed nearly identical relationships in air and water when ~02 is less than 75% ~02max. Above that level, ~E rises at a significantly faster rate in air (Fig. 13). Ventilatory responses during submaxima1 exercise are shown in Fig. 14 for a subject whose responses were representative of the group. Mean maximal cardiorespiratory responses are shown in Table 5 (N=5), and Fig. 15 (N=6). ~02max was not significantly changed by immersion in spite of significant reductions in peak heart rate and expiratory minute volume. Although there was an overall decrease in V0 2max in water of 4%, three subjects had slightly higher values in water and, therefore, the difference of 0.14 Ljmin is not significant. Subject DB had a decline of 0.46 Ljmin (12%) in V0 2max, which was verified by a retest. Significantly lower values in water were found for peak heart rate (6%), peak expiratory minute volume (10%), C02 elimination (11%), and R (8%). No difference was found for breathing frequency, 02 pulse, or the ven­ tilatory equivalent for oxygen. The results of the tests for MVVF are shown in Table 6. (The sub­ jects were later tested using SCUBA and those results will be presented in Chapter 5.) A significant decrease in MVVF (11%) was found in water. The number of respiratory cycles during the 15 sec test, however, was not different. 46 TABLE 4 COMPARISON OF MEAN CARDIORESPIRATORY AND RECTAL TEMPERATURE RESPONSES TO REST AND SUBMAXIMAL UPRIGHT CYCLING IN AIR AND 30 C WATER (N=5) . V0 2 . fh Tre L/min, STPD %V0 2 max bpm °C Air Water Air Water Air Water Air Water X .31 .30 10 11 64 59 36.9 36.85 SE .03 .02 .8 1.9 5.2 6. 1 .05 .06

X .66 .66 22 22 78 72 36.85 36.65 SE .06 .10 .9 .9 3.7 4.5 .03 .09

X 1.08 1.16 35 38 91 90 36.9 36.65* SE .10 .16 2. 1 1.9 4.3 2.6 .04 .08

X 1. 51 1.58 49 52 106 105 37.0 36.8* SE .16 .24 1.8 1.5 4.4 2.2 .07 .06

X 2.26 2.13 72 70 139 124* 37.25 36.90* SE .34 .32 1.9 1.2 3.3 4.5 .04 .03

*p<0.05 47

TABLE 4 (Continued) COMPARISON OF MEAN CARDIORESPIRATORY AND RECTAL TEMPERATURE RESPONSES TO REST AND SUBMAXIMAL UPRIGHT CYCLING IN AIR AND 30 C WATER (N=5) . . VC02 VE fR L/min, STPD R L/min, BTPS br/min Air Water Air vJater Air Water Air Water X .25 .25 .83 .84 12.0 10.9 12 11. 5 SE .02 .03 .04 .04 .8 .5 .9 1.6

X .54 .49 .81 .74* 20.9 19.9 18 16.5 SE .05 .02 .01 .03 1.0 2.0 .6 2.2

X .88 .89 .82 .77 31.2 31.2 21 20 SE .07 .13 .01 .03 1.3 3.9 1.6 3.5

X 1.29 1.31 .86 .83 42.7 45.7 22.5 25 SE .12 .19 .02 .02 2.8 5. 1 1.1 4. 1

X 2.13 1. 91 .94 .89 67.6 64.5 27.5 29.5 SE .33 .29 .02 .01 7.3 6.7 1.8 3.4 48

TABLE 5 COMPARISON OF MEAN CARDIORESPIRATORY AND RECTAL TEMPERATURE RESPONSES TO MAXIMAL UPRIGHT CYCLING IN AIR AND 30 C WATER (N=5) . V02 max fh °2/pu1se Tre L/min, STPD bpm m1/b °C Air Water Air Water Air Water Air Water X 3.12 3.04 172 165* 18.2 18.5 37.7 37.25* SE .40 .45 1 2 2.3 2.8 . 12 .08

. . VC02 VE .. L/min, STPD R L/min, BTPS VE/V02 fR Air Water Air Water Air Water Air Water Air Water X 3.44 3.06* 1.11 1.01* 135.8 123.7 43.3 41.0 48.5 50 SE .43 .45 .03 .01 21.0 18.6 2.4 1.9 2.8 4.5 49 TABLE 6 MAXIMUM VOLUNTARY VENTILATION IN AIR, 30 C WATER, AND IN WATER USING SCUBA

Air Water Wa ter(SCUBA) Subject MVVF fR MVVF fR MVVF fR

DB 241 156 205 148 144 144 RD 295 84 272 96 257 56 JM 171 44 146 38 124 28 BR 186 76 177 104 154 64 RS 194 188 186 184 168 140 LW 193 140 145 152 150 156

Mean 213 115 189* 120 160* 98*

± SE 19 22 19 21 17 22 50

31.2

36.8 f Tre • °c 36.4

36.0

140

120

100 f h , b Imin 80

60

@ 25° C Air

0 40 o 30 C Wotor

20

0 0.5 1.0 1.5 2.0 2.5

FIGURE 7 Heart Rate and Rectal Temperature as a Function of Sub-· maximal Oxygen Uptake. These are mean values fer 5 subjects. On the horizontal scale, left to right, points represent approximately 10 (resting), 20, 35, 70 and 100 % V02 max. Differences in Tre are significant at and ~bove 35%, and fh in water is significantly lower at 70% V02 max. Vertical bars indicate ± 1 SEM. 51

175

150

125 f h , bprn 100

75

fh =1.20(% V02m.a x) + 50.8 r=.98

i IIIII i III o 10 20 30 40 50 60 70 80 90 100

% V02 rna >(

FIGURE 8 Regression of Heart Rate in Air on % V0 2 max. Individual data points of 5 subjects are shown. 52

175 II) Water / / / / / / 150 / / 0/ / / air~ 125 / ~ / /0 / f h , / bpm 100 / .& /0 0 / V e 0/ 75 0/ 00/ / f =J.I7(%V max)+45.7 / h 02 / r= .98 50

I I I I i I I I I I o 10 20 30 40 50 60 70 80 90 100

0/0 \102 ma x

FIGURE 9 Regression of Heart Rate in 30 C Water on % V0 2 max. Individual data points of 5 subjects are shown. Regres­ sion line for water is shifted downward 5-7 bpm. 20

-o 15 '1 Cl> .D ...... E ~ Q) (/) ::3 a...... 10 () Cyc lingin air oC\J o Cycling in 30°C water

5 1"" III -I II .- J II 10 20 30 40 50 60 70 80 90 100

% V02 max

FIGURE 10 Oxygen Pulse as a Function of %V0 max. Differences are not significant. 2

01 (.oJ 54

30

,. f R 20 br/min 10

o

VE BTPS. L Imin 40

e 25° C Air 20 0 30° C Water

10

0 0.5 1.0 1.6 2.0 2.5 VOr.' L I min

FIGURE 11 Expiratory Minute Volume and Breathing Frequency as a Function of Submaximal Oxygen Uptake. These are mean values for the same 5 subjects represented in fig. 7. Differences between air and water values are not significant. 55

200 25°C Air

175 > 75 % V max: . o2. V II 47.9 (V )- 26.3 E 02 r a .B9 150

125 \IE o L/min, 100 BTPS

75

50

25

o 1.0 2.0 3.0 4.0 VOrl L/min, STPD

FIGURE 12 Regression of Expiratory Minute Volume in Air on Oxygen Uptake. Individual data points of 5 subjects are. shown. Lower line and equation are for points below 75% V02 max. Upper line and equation are for all points shown. The different slopes indicate a relative hyperventilation occurs above 75% V02 max. 56

200 30°C Water

175 > 75 % V02 max: VE .. 40.6 (VOe) -10.8 r ll .90 150 .. 125 \IE

L/min, 100 STPS

75 o

50 < 75% Vozmox: "E :: 26.4 (Vo,) + 3.7 25 r= .96

o 1.0 2.0 3.0 4.0

,\;0 2 L/rr\ln ,ST PO

FIGURE 13 Regression of Expiratory Minute Volume in 30 C Water on Oxygen Uptake. See explanatory note in fig. 12. Broken line indicates regression in air for all points shown in fig .. 12. Regression equation for exercise in water below 75% V0 2 max is the same as that for air. The shallower slope for water above 75% ~02 max suggests less hyper­ ventilation occurs in immerslon. 57

SUBJ.1 RD

o

o

_03

~o---o- Q. vT:J I'~~

70 60

50 o25°C AIR \IE 40 o30°C WATER

30 20

\0 60% VOZmoll OL-_-l-__J-_-l-__.l-_.....u:._-,,,--_ 0.5 1.0 1.5 2.0 2.5 3.0 OXYGEN UPTAKE, L/min

FIGURE 14 Effect of Immersion on the Ventilatory Response to Exercise in a Typical Subject. Except for R, the parameters shown are not affected by immersion. This subject had a V02 max of 4.4 L/min. 58

0.0 200,.------.. Air 0 Air a- llD 3.23t O.35 190 xa173!.2 t- !!2 4.0 c: 100 c: 'E E 170 "- "- ..J p>.1 ~160

)( 0 E Water Wator N ll.. 3.09 ~ Ra 163!: 3 .~ 0.37 2..0 no 4.0 5.0 140 160 100 170 100 190 200 VO!J.. max, fh , b /min L/min (STPD)

Cii 220 90 a- Air o Air t- 1l"145~19 lt a 51 OJ 200 00 c: 180 70 E c: 0 "- 160 E 60 ..J 0 "- p<.05 ... )( ..0 60 0 E ....rr. 40 p>.4 w Water Water .> xa l31 t l7 R-56

100 120 140 160 leo 200 30 40 DO 60 70 eo DO

\IE, L /min (BTPS) f R I br/min

FIGURE 15 Comparison of Cardiorespiratory Responses to r~aximal Exercise in 25 C Air and 30 C Water. The filled circles represent individual data for 6 subjects. Values in air are projected on the vertical axis and those in water on the horizontal axis. The diagonal line is the line of identity. 59

For the most part, the subjects preferred maximal cycling in water. They attributed this preference to less heat stress and dyspnea. The subjective degree of leg fatigue was the same for both conditions. Two subjects (DB and BR) who had higher percent body fat, experienced diffi­ culty in staying seated on the bicycle during maximal cycling in water. Three subjects (DB, BR, and RS) were unable to maintain their prescribed maximal pedal frequency in water during the final minute, but did not experience the typical sensation of physical exhaustion.

Discussion

Maximum oxygen uptake for treadmill walking was about 6% higher than for cycling in air and 10% higher than for cycling in water. These . findings are in agreement with those reported by Astrand and Sa1tin (1961). The results of submaxima1 cycling in water show a lower heart rate, but no change in expiratory minute volume or breathing frequency. These findings support the data presented by Craig and Dvorak (1969), Moore et ale (1970), and McArdle et ale (1971). However, as was pointed out by McArdle and his colleagues, submaxima1 exercise in water does not always tell the whole story. In their study, and in this one, heart rate was not significantly lower until about 70% V02max was achieved. And although submaxima1 expiratory minute volume was not different, it was significantly reduced at V0 2max. In the present study VE and fh were reduced even in those three subjects who had the same V02max in water as in air. This confirms the findings of Magel and Faulkner (1967) and suggests that immersion in 60

30 C water either reduces the maximal attainable ~E and fh, or other factors compensate to maintain oxygen transport before they become limiting. Hermansen and Saltin (1969) observed that during maximal exercise, oxygen uptake sometimes reaches a plateau before ~E and fh. If this is also true for underwater exercise, the lowering of VE and fh at ~02max is probably due to the augmentation of other oxygen transport­ ing mechanisms, such as Qs or 02F. According to the Fick equation, for the value of ~02 to remain un­ changed, a lowering of heart rate must be compensated by an equal increase in the product of stroke volume and arterio-venous oxygen difference. Arborelius et aZ. (1972) using a dye-dilution technique found a 35% increase in stroke volume at rest during heat-out immersion in 35 C water. They postulated that the increase is due to a shift of 0.7 L of blood from the dependent regions of the body into the intra­ thoracic vascular bed and the heart. In other words, the greater stroke volume resulted from a greater diastolic filling of the heart by means of the Frank-Starling mechanism. Dixon and Faulkner (1971) estimated cardiac output during tethered swimming and running. Using a C02 rebreathing technique they found the 6.5% lowering of peak heart rate was nearly offset by a 5% increase in stroke volume, so that cardiac output was only slightly lower (2%) for swimming as compared to running. It is noteworthy to mention that swimming V02max was still reduced 5% because (a-v)02 declined 3.5%. The difference in V0 2max, however, was not significant. 61

It is also possible that the required cardiac output is smaller for exercise in 30 C water. Faulkner (1968) suggested that the lower heart rate in swimming may indicate less heat stress; i.e., a smaller fraction of the cardiac output distributed to the skin would permit a greater flow for metabolic needs. Thus the reduced fh might simply mean a smaller Qis required.

According to the Haldane equation, for the value of V0 2 to remain unchanged a lowering of expired minute volume must be compensated by an equal increase in the amount of oxygen removed from the inspired air. A greater oxygen extraction at the lung, with immersion, could be explained by an increased pulmonary diffusing capacity (Guyatt et al.~ 1965; Cerney and Reddan, 1973), or a better ventilation-perfusion balance (Arborelius et al.~ 1972), or slower breathing. However, in the present study breathing frequency was unchanged, and Cerney and Reddan (1973) have concluded that the passive increase in pulmonary capillary blood volume that occurs at rest during immersion is probably overridden during heavy exercise by active vasoconstriction. Since MVVF was also reduced in water, there may be other reasons, besides alterations in diffusion and perfusion, to explain the lower VEe

Agostoni et ale (1966) and Hong et al. (1969) have shown that respiratory mechanics are affected by submersion to the neck. Marked decreases in expiratory reserve volume and small but significant decreases in vital capacity occur, which are attributed to the craniad displacement of the diaphragm and an increase in the central blood volLme. Moreover, there is an increased airways flow resistance and total work of breathing by as much as 60%. The hydrostatic force on the abdomen and chest wall was 62

probably smaller in the present study because of the modified position of immersion (Fig. 1). Neverthe1ess~ the 11% decrease in MVVF and 10% decrease in peak VE may have been due to the hydrostatic force counter­ acting the force of the inspiratory muscles and the increased airways resistance. An increase in the work of breathing would also account for the proportional decrease in VE with respect to the decrease in MVVF; i.e.~ there was no change in the breathing reserve ratio (VE/MVV). A lower R value in water was found throughout exercise. This find­ ing is the only distinguishing ventilatory difference between air and water below 70% V0 2max (Fig. 15). The anaerobic threshold is typically aZ.~ found around 60-70% V0 2max (Wasserman et 1973). During steady-state exercise below 60% V02max R must be equal to the metabolic gas exchange ratio. The lower R in water without a significant reduction in ~02max, therefore~ indicates that C02 is being stored. The lower Rand VE at maximal exercise in water suggests a smaller compensation for metabolic acidosis. This could be due to an upward shift in the anaerobic thres­ ho1d~ or a smaller change in acid-base balance, or a reduced sensitivity to acute acidosis in general. A final speculation is that during re­ sistance breathing there may be a trade-off between C02 retention and the extra respiratory work to maintain adequate alveolar ventilation and C02 elimination. The reduced VE, however, should not be considered a hypoventi1atory response until an increase in arterial C02 is demon­ strated. The greater decline in Tre in water from rest to the end of the first exercise period accounted for most of the difference in Tre up to 70% V02max. During rest the water was unstirred, so convective heat 63

loss probably increased markedly at the onset of exercise. At 70 and

100% V02max heat storage was greater in air. To what extent the lower Tre contributed to the lower fh in water can not be determined from these data since only one water temperature was used. The subjective responses were ve\'y interesting. Compared to air there appeared to be a mismatch between the overall subjective stress and leg fatigue during maximal cycling in water. In other words, the subjects felt their legs "gave out" before they were "out-of-breath." Their perceptions seem to fit the data. In summary, the maximum oxygen uptake for cycling in air and cycling in water were not significantly different. However, three subjects had

V02max in water reduced by an average of 10%. Whether this reduction was caused by physiological or physical factors due to immersion is equivocal. The lack of gravitational assist in underwater cycling may be a disadvantage, but this is a physical effect and it should be dis­ tinguished from physiological changes occuring with immersion. Hoes et al. (1968) pointed out that because of inertia, the peak force applied by the subject on the pedal and crank in the downstroke of cycling may be twice the actual resistive load. If this is also true in underwater cycling, the reduction in body weight which is due to the buoyant force of water may be a serious handicap against increasing power output. The physical work capacity of an underwater cyclist is, therefore, affected not only by V02max and rpm (Fig. 6) but also by his mechanical advantage in applying force. 64

CHAPTER 5 EFFECT OF SCUBA ON PULMONARY VENTILATION DURING SUBMAXIMAL AND MAXIMAL CYCLING EXERCISE IN WATER

Introduction

As discussed in the previous chapter, the ventilatory response to exercise below 75% V02max was essentially unaltered by head-out water immersion (Fig. 13). Expiratory minute volume during maximal exercise, on the other hand, was significantly reduced by 10 percent (Fig. 15). The implication of this reduction was considered in terms of altered respiratory mechanics due to compression of the chest and abdomen by the hydrostatic force of water. As a result of head-out immersion, intrapulmonary pressure becomes much less than ambient hydrostatic pressure on the thorax, lung volume decreases, airway resistance increases, and the total work of breathing increases (Hong et aZ.~ 1969). However, the added work of breathing apparently was not great enough to affect ventilation at flow rates requiring less than 75% vo2max. Another interesting observation about the ventilatory response to exercise was made during the course of calibrating the Monarch under­ water ergometer using SCUBA. Although the oxygen uptake-pedal frequency relationship was unchanged by the use of SCUBA (Fig. 4), reductions in expiratory minute volume were observed at all rpm, compared to using the Triple-J valve. Increased external airway resistance is known to reduce pulmonary ventilation during steady-state exercise (Cerretelli et aZ., 1969). Since the exhaust hose from the SCUBA valve was connected to the 65 same expired gas airway as the Trip1e-J valve, it was assumed that the increased inspiratory resistance of SCUBA was probably responsible for the reduction in ~E' The U.S. Divers 2-stage regulator used in the calibration study, is a demand-type regulator designed to supply air to the diver as he starts to inhale. In the second-stage of the regulator, a pressure- sensing diaphragm responds to the negative inspiratory pressure, and air is supplied until the pressure at the diaphragm returns to ambient; i.e., the differential pressure becomes zero. With the use of aU-tube water manometer, it was determined that a side pressure at the U.S. Divers double-hose mouthpiece of about 5 cm H20 was required to open the diaphragm. This agrees with the value for inspiratory resistance of 5.5 cm H20, at a flow rate of 30-112 L/min, reported by Christianson (1971). The inspiratory resistance of the Trip1e-J valve is given as 0.5 cm H20 at 112 L/min in the Warren E. Collins Co. catalog. The SCUBA resistance is, therefore, about 11 times greater. The purpose of this study was to measure the effect of SCUBA on the ventilatory response to graded exercise in water. The primary objective was to determine if the external airway resistance of SCUBA, when added to the increased internal airway resistance caused by head-out water immersion, would effect V0 2max. Loomis et aZ. (1972) reported a reduc­ tion of about 22% in ~02max during swimming compared to cycling when their subjects breathed from SCUBA for both conditions. They suggested the decrease in V02max was presumably due to the different types of exercise. Since a 35% reduction in VE was also observed, another rea­ sonable explanation is that the combined effect of SCUBA and water 66

immersion on breathing resistance may have contributed to the lowering

of ~02max while swimming. By comparing V0 2max during underwater cycling with and without SCUBA, it would be possible to make a determination on this question.

Methods

The same six subjects (Table 3) that were described in the two preceding chapters also participated in this study. They were familiar with the use of SCUBA. Subjects RD, JM, and BR were considered skilled SCUBA divers. A standard 71 cu. ft. compressed air bottle was mounted on the inside wall of the immersion tank. The air bottle was moved up or down in the water so that the SCUBA regulator was positioned at the po"int that just stopped free flow. At this point the regulator was about half-way immersed. The aliquot sampling system was used for expired gas collec­ tion and analysis. At rest and any given exercise level the subject alternately breathed compressed air from SCUBA or room air using the Collins Triple-J valve. The entire experiment was conducted in one session for each subject while he was immersed in the head-out position previously described. Since some subjects had expressed ventilatory discomfort when using

SCUBA at relative work loads above 70% V02max, the experimental protocol was slightly revised. The subject entered the water and rested quietly for five minutes. He then exercised for 7 min at work rates requiring approximately 15, 30, and 60% v02max. At rest and during exercise two gas collections were made. The first collection was made between 67

minutes 4 and 5; then while the subject continued to cycle at the same rpm, the investigator switched respiratory valves and a second collec­ tion was made from minute 6 to 7. The order of the valves was alter­ nated. After completing the 60% ~02max load, the subject rested for 3-5 min and then exercised at maximum rpm for 5 min using SCUBA only. The ~02max and related peak cardiorespiratory values reported in this chapter for the subject using the Triple-J valve were taken from the previous study (Chapter 4). Approximately 5 min after the end of maxi­ mal exercise the subject completed 2 or 3 tests for MVVF using the pro­ cedure previously discussed. Respiratory valve dead space was determined by measuring the volume of water required to fill the space between the intake and exhaust valve3, and the mouthpiece. Physiological dead space in ml was assumed to be equal to the subjects height in cm (Shephard, 1959). This esti­ mation is a slight variation from Radford's rule of thumb that the dead space in ml equals the body weight in pounds (Radford, 1955). With the present subjects the "height method" produced a mean VD of 175 ml, whereas the "weight method" would predict 150 ml (Table 3). Insofar as VD increases with exercise (Kagawa and Kerr, 1970), the higher value was preferred. It should be pointed out, however, that since the same value for Vo was used for both Collins valve and SCUBA breathing conditions, the comparison between the two is virtually unaffected by its absolute magnitude.

Water temperature and exposure time averaged 30.3 ± 0.1 C and 51 min, respectively. 68

Results

The individual results are presented in Appendix C. Submaxima1 data were not obtained for subject DB. The mean data shown in Table 7 and Figs. 16 and 19 represent a sample size of 5. The mean values obtained for maximal exercise (Fig. 18) include DB. In general, the subjects maintained the same pedal frequency for both gas collections at a given exercise intensity. Mean values in % ~02max were approximately 10, 15, 30, 60, and 100, as desired. The differences in ~02' % ~02max, VC02' R, Tre , and fh (except at the 30% work level) observed during submaximal exercise were small and non­ significant. Although there was no difference at rest, VE was signi­ ficantly lower during submaxima1 and maximal exercise when using SCUBA (Table 7, Figs. 16 and 18). Breathing frequency was significantly lower using SCUBA at 30% v02max and above. Calculated tidal volume was sig­ nificantly larger at the 60% level using SCUBA, but it was not statis­ tically different otherwise. The cardiorespiratory responses shown in Fig. 17 for one subject are generally typical for the group. Differences in the amount of oxygen removed from the inspired air (02F) were signi­ ficant throughout exercise and were inversely related to VEe Linear regression analysis of the dependency of VE on V0 ' from rest to 100% 2 V0 2max, when using SCUBA or the Trip1e-J valve gave the following results: 1J for Trip1e-J: VE = 39.6 (V02) -4.8, r = .97; 2J for SCUBA: ~E = 34.7 (~02) -4.7, r = .95.

There was no difference in V0 2max or peak heart rate using SCUBA (Fig. 18). Peak expiratory minute volume and breathing frequency were, however, reduced 24 L/min (35%) and 16 br/min (29%) during maximal 69 TABLE 7 EFFECT OF SCUBA ON THE CARDIORESPIRATORY RESPONSE TO REST AND EXERCISE IN 30 C WATER. (SUBJECTS ALTERNATELY BREATHED FROM COLLINS TRIPLE-J VALVE OR SCUBA) . . T V0 2 %V0 2 max fh re L/min STPD bpm °C 3-J SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA

~ .26 .25 10 9 56 55 37.2 37.05 SE .02 .02 .9 1.7 6.2 6.4 .06 .07

X .40 .38 15 14 60 59 36.6 36.65 SE .04 .04 2.5 2.6 7.6 6.8 .20 . 18

X .84 .73 29 27 80 76* 36.6 36.6 SE .08 .06 4.3 4.4 4.9 5.4 .21 .19

X 1.80 1.77 63 62 116 112 36.6 36.6 SE .11 .09 5.8 5.9 7. 1 6.1 24 .25

~ 3.04 2.98 100 100 165 161 37.25 37.15 SE .45 .43 0 0 2 3 .08 .11 70

TABLE 7 (Continued) EFFECT OF SCUBA ON THE CARDIORESPIRATORY RESPONSE TO REST AND EXERCISE IN 30 C WATER. (SUBJECTS ALTERNATELY BREATHED FROM COLLINS TRIPLE-J VALVE OR SCUBA) . . VC02 VE f R VT L/min STPD R L/min BTPS br/min L/br 3-J SCUBA 3-J SCUBA 3-5 SCUBA 3-J SCUBA 3-J SCUBA X .22 .22 .86 .86 10.5 10. 1 9.5 7 1.4 1.5 SE .02 .03 .04 .06 1.4 2.9 2.0 1.5 .2 .3

X .42 .38 .77 .79 15. 1 11.8* 9.5 10 1.6 1.7 SE . 14 . 12 .03 .04 3.7 2.6 3.0 3.3 .4 .6

X .69 .62 .78 .78 24.6 18.5* 16 11.5* 1.6 1.8 SE .12 .11 .04 .03 3.1 2.8 2.2 2.3 .2 .4

~ 1. 52 1.45 .89 .86 53.2 41.1* 27 18* 2. 1 2.4* SE . 17 .14 .08 .03 6.4 6. 1 3.9 4.2 .2 .3

X 3.06 3.32* 1. 01 1. 12* 123.7 106.8 50 38.5* 2.6 3.0 SE .45 .45 .01 .02 18.6 17.3 4.5 7.4 .4 .4 71

30 0 C Water

50

40

\IE' 30 L/min (8T PS)

20

o Collins triple-Jvalve J o SCUBA valve 10

o 0.5 1.0 1.5

FIGURE 16 Effect of SCUBA on Pulmonary Ventilation During Sub­ maximal Exercise. These are mean values for 5 subjects. From the left, paired data repres~nt approximately 10 (resting), 15, 30, and 60% V0 2 max. The subject alter­ nately breathed from SCUBA or Collins valve while main­ taining a steady-state at each level. Differences during exercise are statistically significant. 72

Subj. R.D.

fh 60 ]60 4 ~ R ·ro'Sf ~ '7 0'0 0 6'0 \---'0 F e 0 °2 ,4' :5·0 /

V lHI T .., e-...... ~ 1-13 o-~ 1·0 20 20 16 fR 10 6 1 60

40

30 VE 20

10

0 0·5 1·0 1·5 2:0 OXYGEN UPTAKE, L/mm

FIGURE 17 Effect of SCUBA on the ventilatory Response to Sul?maximal Exercise in a Typical Subject. IQ this subject (V02 max = 4.2 L/min using SCUBA) the lower VE during SCUBA breathing became more apparent as exercise intensity increased. This was due to a slower rate of breathing. The lower VE was compensated by a higher oxygen extraction (02F). 73

5.0r-:--:-::--~----:;---.." 200t~-:':":""--:------... Collinl valve Colllnl valvo c XII 3.09 t. 0.37 19 ii1l163~ 3 'E ...... 4.0 18 -I c )( .- 170 o E E 3.0 p>.1 ...... 160 p>.05 ... .0 .~ SCUBA SCUBA j{ a 3.04 ~ 1l.1I159!4 0.35 13Q6£---&.-.L-...l--.L.--I_..L-....I 2.0 3.0 4.0 5.0 140 1150 160 110 IGQ 190200

Voem qx I L /min fhl b/min

190,.------..D..-...,. 90r------~ o 170 o 80 CoilinG volvo C ollino valvo 150 1l. m I31! 17 70 a 56!7 c x 'E 130 6 ...... -I p<.05 p<.05 SCUBA SCUBA RlII071 14 l'l1l40 t 6

70 90 110 130 150 170 190 30 40 60 60 70 00 00 "E aTPS ' L 1min fA • br Imin

. FIGURE 18 Comparison of Cardiorespiratory Responses to Exercise while Breathing from Collins Valve or SCUBA. See fig. 15 for explanatory note. Subjects were immersed to the neck in 30 C water for both conditions. 74 exercise with SCUBA. Subjects DB and BR had reductions in ~E of 56 and

40 L/min, respectively, without a change in V0 2max. The results for the test of maximum voluntary ventilation are shown in Table 6 (Chapter 4). MVVF using SCUBA was significantly reduced by 12% compared to using the Trip1e-J valve in water. The reduction in MVVF with SCUBA was due to a significantly lower breathing frequency. Respiratory valve dead space was 315 m1 for the Trip1e-J valve and 80 m1 for SCUBA. Generally the subjective response to breathing from SCUBA during maximal exercise was that it felt easier than expected. The sense of dyspnea, however, was slightly greater with SCUBA. Three subjects com­ plained of moderate chest pains during normal breathing following the experiment, which persisted for about 2-3 days.

Discussion

The results indicate the use of SCUBA during exercise primarily affects the ventilatory response. According to L.E. Farhi (personal communication) an inspiratory resistance of 5.5 cm H20 is much too small to produce the observed reductions in expiratory minute volume. Since respiratory valve dead space is also known to affect ventilation

(Bartlett et aZ. 3 1972), an attempt was made to correct for the 4-times greater dead space in the Collins valve by estimating the alveolar ven­ tilation (VA). The results of this calculation are shown in Fig. 19 for the group, and for an individual whose responses were typical of the group in Fig. 20. The 5% reduction in the estimated VA was greatest at

V02max using SCUBA, but this was not statistically different from the 75

30°C Water 30

f R , 20 br/min 10-

o

40

30

VA, L/min 20 (BT PS) o Coil ins triple-J valve

o SCUBA valve 10

I I o 0.5 1.0 1.5

FIGURE 19 Effect of SCUBA on Alveolar Ventilation and Breathing Frequency. Differences between the minute volumes shown in fig. 16 and those shown here represent the estimated dead space ventilation. The slow~r breathing frequency using SCUBA (p<.05 at 30 and 60% V02 max) Illay be attributed to its smaller valve dead space and greater inspiratory resistance. 76

SUBJECT: R.S

120 110

fh I 100 b/mln 90 80 10 25 20 fn , 16 br/mln 10 6

50

40

VE 6TPB I L/min 30

20

0 Coiline valvo 10 0 SCUBA

0.5 1.0 ~o Vo!?, L Imin 74% VOl L"lQII

FIGURE 20 Effect of SCUBA on the Submaximal Cardiorespiratory Responses of a Typical Subject. Broken lines represent the extimated alveolar ventilation. Note that heart rate and oxygen uptake were the same for both breathing conditions. 77

Collins valve. Loomis et aZ. (1972) substituted a Collins Triple-J valve for the SCUBA valve on a 2-stage, double hose regulator similar to the one used in the present study. They found the 30% decrease in VE obtained during exhaustive exercise using full SCUBA was reduced to about 4% using the substituted Collins valve with the regulator. Their results and those found in the present study indicate that most of the reduction in VE during exercise using the SCUBA mouthpiece is due to its smaller dead space volume than the Collins valve. The smaller dead space would also tend to explain the higher 02F with SCUBA. A lower 02F indicates a higher FE02. The quantity of room air in the breathing valve following inspiration should be equal to the dead space volume. With each expi­ ration, that quantity would be pushed through the collection system and dilute the expired gas concentrations. The greater valve dead space would, therefore, account for the higher measured FE02 and lower cal­ culated 02F found with Collins valve breathing. In addition a larger dead space volume would also cause a greater rebreathing of expired gas and perhaps C02 retention. This may have contributed to the higher VE using the Collins valve. Finally the slower breathing frequency with SCUBA was also a factor leading to its relatively higher 02F. The results of the MVVF test probably best indicate the separate effects of immersion and SCUBA resistance on breathing capacity. Re­ calling from Table 6, the average reduction in MVV was 11% for water in~ersion and an additional 12% for SCUBA. The combined effect of immersion and SCUBA was a decreasp Jf 30% from the MVV value in air using the Collins valve. Comparison of MVV results with those reported 78 in the literature is hazardous because of the a1inear pressure-flow characteristics of the systems used for imposing breathing resistances. Nevertheless, Demedts and Anthonisen (1973) found that an inspiratory resistance of about 7.5 cm H20 at 120 L/min reduced maximum exercise ventilation 12% and 15 sec MVV, 11 percent. This compares favorably with the reductions observed for the same parameters during water immersion. Thus the added resistance to breathing due to immersion may have been close to 7.5 cm H20 during maximal exercise, in which VE was 131 L/min. Hong et aZ. (1969) found an average increase in total intra­ pulmonary pressure of 16 cm H20 during submersion to the neck. This va"1 ue represents the integrated hydrostatic pressure on the re1 axed chest wall. In the present study the position of the thorax was approx­ imately horizontal so that the depth to which the chest was immersed, and therefore the hydrostatic pressure on the chest wall, was probably less than that found by Hong et aZ. (1969). Flook and Kelman (1973) found a 29% decrease in MVV using an inspiratory resistance of about 15 cm H20 at 100 L/min. The reduction is close to that observed for immersion plus SCUBA. Hence, the resistance of SCUBA alone may have been approximately 7.5 cm H20, or about the same as for water immersion. This value compares favorably with that reported by Christianson (1971). In the present series, the average peak expiratory minute volume was about 70% of the 15-sec MVV, and this figure was independent of the experimental condition. Demedts and Anthonisen (1973) reported the same finding using various breathing resistances. They suggested that since MVV and maximum exercise ventilation are both chiefly limited by dynamic airway compression during expiration, the effect of added 79

resistance on peak ~E might be predicted from the depression of MVV measured with and without that resistance. The present findings would tend to support their hypothesis. The lower breathing frequency with SCUBA may be explained by the marked lengthening of inspiration that occurs with added inspiratory resistance (Flook and Kelman, 1973). It is interesting to note that Craig et at. (1970) found exhaustion occurred during exercise when expiratory time reached some minimal value which was characteristic for each subject. The higher peak R observed using SCUBA suggests that hyperventilation was greater in spite of a lower breathing frequency. In summary, the use of SCUBA during a 5 min maximal exercise test did not affect ventilation to the extent that aerobic work capacity or v0 max was altered. The observed 24 Llmin decrease in peak ~E could be 2 explained primarily by a reduction in dead space ventilation with SCUBA of 18 L/min. Peak VA was consequently not significantly different. The finding reported by Loomis et al. (1972) that ~02max was lower during SCUBA swimming than for SCUBA cycling in air is probably not due to an effect of SCUBA plus immersion per see Although ~02max with SCUBA is not significantly altered, its practical use during very heavy exercise is questioned on the grounds that it may cause excessive res- piratory distress. . Peak VA and fh are shown in Fig. 21 with the corresponding values for V0 2max obtained during the four conditions of maximal exercise discussed in Chapter 4 and this chapter. Individual data are presented in Appendix C, Fig. 1. The reductions in ~A and fh during the three cycling conditions appear to be a linear function of V02max. The 80

-Cycling- Air f 120 Cf) a.. Walldng enl- 110 Water c E ...... 100 --l "« ~ 90

180

E 170 0- .0

'-+-..c 160

~/"'"--...---,...----,.,---.-.----.-.- 3.0 3.1 3.2 3.3 3.4

Vomax,L/min 2

FIGURE 21 Peak Heart Rate and Alveolar Ventilation as a Function of ~02 max During 4 Conditions of Exercise. Mean values of 6 subjects are shown. Two types of exercise are repre­ sented, treadmill walking and cycling. Cycling was per­ formed in 25 C air, 30 C water, and 30 C water using SCUBA. Subjects were immersed to the neck for experiments in water. 81

reverse also could be true; i.e., the decrease in Vo 2max may result from a reduction in VA and fh. Statistical treatment of the data,

however, indicated that the change in Vo 2max during cycling in air and water was not siynificant whereas the reductions in VA and fh were significant. Compensations for these reductions must, therefore, occur in other oxygen transport factors such as stroke volume and oxygen extraction. This is especially evident in the comparison of cycling in air versus treadmill walking where a significant decline in Vo 2max was observed with no change in fh and even a significant increase in VA. A determination on this dilema can possibly be made by altering the water temperature to see if peak fh and VA can be manipulated without a corresponding change in V0 2max. 82

CHAPTER 6 THE EFFECT OF WATER TEMPERATURE ON CARDIORESPIRATORY AND METABOLIC RESPONSES TO MAXIMAL EXERCISE

The effect of water temperature on the cardiorespiratory response to maximal exercise has surprisingly been the subject of very little research. In 1965 the National Collegiate Athletic Association set the narrow limits of 24.4 to 25.6 C for swimming pool water temperature during official, NCAA-sanctioned competition (Costill, 1966). Appar­ ently this decision was based on empirical findings that performance was best at these temperatures. Soon after, Costill (1966) reported that ~02max, measured during a 3 min bout of tethered swimming, was the same in 18, 25, and 32 C water. Furthermore, he found no difference in peak heart rate or rectal temperature. Costill 's conclusion was that these water temperatures do not affect the cardiorespiratory responses to maximal exercise when the work duration and exposure time are short. Costill et al. (1967) later presented similar results for 20 min of submaximal swimming (~02 = 3.0 L/min) in 17, 27, and 33 C water. Mean exercise heart rates were 146, 145, and 148 bpm, respectively. Their results are in disagreement with the later findings of Craig and Dvorak (1969) and Moore et al. (1970) who reported that submaximal exercise heart rate was inversely related to water temperature, ranging from 10-26 bpm lower in 25-16 C compared to 30 or 35 C water. In contrast to Costill's results (1966) Nadel et al. (1974) found that compared to 33 C, ~02max during swimming was reduced about 6% in 26 C and 10% in 18 C water. An important difference, however, was that their subjects 83

were immersed from 40 to 100 min prior to maximum swin~ing and the reductions in ~02max may have been related to the degree of cold stress. Heart rate during submaximal swimming was approximately 15-35 bpm slower at a given oxygen uptake in 26 and 18 C versus 33 C. What is the mechanism by which water temperature Inight alter ~02max? Rowell (1974) characterized ~02max as the cardiovascular state in which about 80-85% of the cardiac output perfuses at least 50% of the total skeletal muscle mass while the blood flow to other tissues is un­ changed or reduced. Thus, if muscle blood flow was reduced by some thermal effect of high or low water temperature, this might explain

alterations in ~02max. From the cold stress viewpoint, Nadel et ale (1974) suggested that local muscle fatigue, perhaps due to low muscle temperature, rather than cardiorespiratory distress caused the 15% re­ duction in ~02max observed in two of their three subjects. Both of these subjects had esophage1 temperatures below 36 C during maximal swims in 18 C water. It is also possible that cold stress may increase adrenergic tone even in active muscles and thereby reduce blood flow and oxygen delivery. In warm water, especially during heavy exercise, heat stress nlight alter the pattern of vasomotor outflow so that a greater fraction of the cardiac output is distributed to the skin. Since skin has low metabolic activity, ~02max would fall proportional to the fall in arteriovenous 02 difference. Thermoregulatory mechanisms may, therefore, cause a reduction in muscle blood flow, and affect oxygen uptake. Cold water immersion would tend to favor reduced periph­ eral circulation to prevent heat loss and warm water immersion would tend to favor increased cutaneous blood flow to facilitate heat loss. 84

The purpose of this experiment was to determine the effect of short term (20 min) immersion in warm (35 C) and moderately cold (25 C) water on ~02max and its related cardiorespiratory parameters. Basic­ ally this was a study of the effects of high and low mean skin tempera­ tures on the cardiovascular and thermal responses to exercise, since it has been shown that skin temperature is essentially "c1amped ll to within 1.0 C of stirred water temperature (Brown and Brenge1mann, 1970;

Nadel et al. 3 1974). Peripheral tissue temperature changes involve considerable heat exchange as approximately 50% of the body mass is within 2.5 em of the surface (Carlson and Hsieh, 1970). Using water­ perfused suits, Rowell et al. (1969b) studied human cardiovascular adjustments to rapid changes in skin temperature during exercise. Suddenly raising skin temperature caused heart rate to accelerate while right atrial pressure, central blood volume, and stroke volume de­ creased. The fall in stroke volume prevented cardiac output from increasing in proportion to heart rate. Abruptly lowering skin tem­ perature caused a dramatic reversal of these changes: cutaneous vaso- and venoconstriction rapidly shifted blood centrally so that central blood volume and stroke volume were restored. Faulkner et al. (1971) observed progressive increases in heart rate and arteriovenous oxygen difference and decreases in stroke volume and cardiac output during 10 min of maximum cycling exercise in air. Maintenance of low skin temperature, however, can keep the cutaneous veins constricted and abolish the "drift" in heart rate, stroke volume, etc., normally seen in prolonged, heavy exercise (Rowell, 1974). These findings suggest that peripheral pooling of the vascular volume in cutaneous 85 veins is a cause of cardiovascular drift. Peripheral pooling under­ standably depends on the thermoregulatory status of the body, partic­ ularly skin temperature. The influence of skin temperature on cutaneous venomotor state, however, may be due strictly to local effects of temperature on veins (Vanhoutte and Shepherd, 1970). Water temperature, therefore, not only affects the thermal gradient from core to skin and thereby influences the amount of peripheral blood flow, it may also have direct effects on the blood vessels themselves.

Methods

A general description of the three subjects along with their cardiorespiratory responses to maximal treadmill running is shown in Table 8. Oxygen uptake was measured using the collection system described by Daniels (1971), with the exception that a Collins Trip1e-J breathing valve was used. Except for blood data, the methods used for all other measurements are described in Chapter 2. Three to 5 m1 blood samples were drawn anaerobically in a syringe from a superficial forearm vein at rest and approximately 2 and 5 min post-maximal exercise. Three ml of the whole blood was transferred directly into a chilled test tube of perchloric acid and deproteinized within one minute of collection. The precipitate was spun down at 0 C in a refrigerated centrifuge. The supernatant was treated and analyzed enzymatically for lactate and pyruvate concentrations according to the well-described techniques available from Sigma Chemical Co. (St. Louis, Mo.). Duplicate determinations were made on each sample. The highest 86 TABLE 8 DESCRIPTION OF SUBJECTS

. b Subject Age Ht Wt %Fata V02max fh VE La yr cm kg L/min ml /Kg ·mi n bpm L/min mg% BTPS

DB 24 180 75.5 15 4.46 59 191 190.6 100 RD 30 178 70.0 9 4.80 69 168 174.0 64 JB 27 184 67.8 10 3.89 57 179 136.6 80

Mean 27 181 71.1 11 4.38 62 179 167.1 81 askinfold technique, after Sloan and Weir (1970) btreadmi11 running 87 post-exercise value is reported. The standard error of the mean on five determinations of the same resting blood sample was 0.2 mg %for lactate and 0.006 mg %for pyruvate. The remainder of the withdrawn blood was packed in ice and analyzed for pH within 10 min by using a microelectrode (Radiometer, Copenhagen). The pH electrode was cali­ brated with standard solutions provided by the manufacturer. The experimental protocol was markedly different from previous experiments. A blood sample was taken after the subject rested quietly for 5 min in air. He then entered the immersion tank and immediately began cycling at 20 rpm to stir the water. This rate does not appre­ ciably raise oxygen uptake from rest (Fig. 5). A two minute "resting" data collection was made. from minutes 3 to 5. The subject then exer- cised for 5 min at 50% V0 2max and a collection was made during the 5th minute. Without stopping he continued to exercise for 3 min at approx­ imately 75% V0 2max and for 5 min at 100% V02max. Gas collections were made during minutes 3-4 and 4-5 of maximal exercise. At the end of maximal exercise the subject recovered in the water for 1 min and then got out for the post-exercise blood sampling and 5 min post-exercise measurement of rectal temperature and heart rate. The procedure for the air control experiment was the same.

Total immersion time was 20 min. Water temperatures were 25.0 ±

0.1 C and 35.6 ± 0.5 C.

Results

Individual data are presented in Appendix 0 and mean data in Table 9. TABLE 9 EFFECT OF WATER TEMPERATURE ON CARDIORESPIRATORY AND METABOLIC PARAMETERS DURING REST, SUBMAXIMAL EXERCISE, AND MAXIMUM EFFORT EXERCISE. (Tre VALUE IN PARENTHESIS. . REFERS TO 5 MIN .POST-EXERCISE MEASUREMENT) %V02max V0 2 VC02 R VE f h f R Tre La Pva pH AIR 23.0 ± 0.2 C: X 9 .32 .26 .79 12.2 57 11 37.2 6.3 .72 7.37 SE 0.88 .03 .03 .03 1.7 5.8 1.5 .07 .44 .05 .00 X 54 2.08 1.78 .86 58.3 121 21 37.3 SE 1.5 .11 .10 .03 1.8 5.2 1.5 .09 X 100 3.86 4.65 1. 21 162.7 172 51 37.7{38.0) 87.3 3.00 7.17 SE 0 .15 .19 .03 17.5 .88 6.5 .17{.13) 9.9 .76 .02 WATER 25.0 ± 0.1 C: X 8 .32 .24 .75 10.6* 51 12 37.1 8.6* .71 7.37 SE .88 .03 .03 .04 1.6 7.4 2.0 .09 .35 .05 .02 X 53 2.02 1.65 .82 55.4 102* 22 37.0 SE 1.8 .14 .14 .09 4.5 6.6 2. 1 . 17 X 100 3.78 4. 15 1.09 143.9 153 46 36.9 (37 .1) 86.3 2.65 7.17 SE 0 .16 .03 .04 9.9 5.8 2.7 .15{.18) 4.7 . 14 .03 ~~ATER 35.6 ± 0.5 C: X 8 .31 .24 .79 11.0* 57 13 37.0 8.1 .58 7.37 SE 1.0 .03 .03 .05 1.6 11.7 2.3 .07 1.2 .03 .02 ~ 55 2.04 1.66 .82 57.5 117 22 37.1 SE 1.2 .07 .05 .05 .93 2.7 1.5 .06 X 100 3.72 4.42 1.19 162.6 172 54 37.5{37.8) 103.4 2.32 7.21 SE 0 .20 .22 .01 15. 1 3.5 4.7 .12{.12) 8.3 .23 .02

co co 89

Using the same protocol as described above, V02max during treadmill running was an average of 4% greater than the mean shown for these sub­ jects in Table 3. Cycling in air mean V02max was, however, 3% lower than found earlier (Appendix B).

Resting. V02 and VC02 were the same for all conditions. VE was significantly lower at both water temperatures. VT was reduced about 350 ml/br in water since fR was 1 and 2 br/min higher in 25 C and 35 C water respectively. fh was the same in 35 C, but it was 6 bpm lower in 25 C water. Tre was 0.1 C lower after 5 min immersion at both water temperatures. There were minor differences in resting blood data, al­ though lactate was significantly higher prior to 25 C immersion.

Submaximal exercise. The initial exercise level was 53-55% ~02 V0 2max. was the same for all conditions. VC02 and R were slightly lower in water. VE02 and fR were approximately the same for all condi­ tions. fh was significantly reduced by 19 bpm in 25 C water compared to air and by 15 bpm compared to 35 C water. 02 pulse reflected the same relationship. Tre increased 0.1 C in air and 35 C water over resting values, but declined 0.1 C in 25 C water.

Maximal exercise. Compared to air, V02max was 2% and 3.5% lower in 25 C and 35 C water, respectively. (Recalling the data shown in Appendix B, ~02max in 30 C water was 3.5% lower than in air for the same subjects.) Maximal rpm was the same for both water temperatures.

Air ~. 35 f water. ~E' fh and the increase in Tre were exactly the same. R, fR and post-exercise pH were approximately the same. fh at 5 min post-exercise was 92 bpm in air and 91 bpm after 35 C water 90 immersion. Venous lactate was 18% higher and pyruvate 23% lower after maximal exercise in 35 C water. pH was about the same.

~ 25 water vs. 35 e water. Although v02max was slightly higher in 25 C water, ve02 was reduced 6%. R was, therefore, 8% lower. VE was reduced 19 L/min (12%) and fh was 19 bpm (11%) lower in 25 e water. fR was 8 br/min lower in 25 e water, but VT was about equal. Tre did not rise in 25 e water. There was, however, a gain of 0.2 e post­ exercise. fh at 5 min post-exercise in 25 C water was 77 bpm, or 15% lower than after exercise in air or 35 C water. Of special interest was the finding that subject RD reached V02max in 25 e water with a peak fh of only 144 bpm, or 27 bpm lower than in air.

Subjective responses. The initial sensation of entering 25 e water was described as one of sharp pain coming from the skin, but this lasted for only a few seconds. The subjects felt cold and occasionally shivered throughout the 5 min period while pedaling at 20 rpm. At the

50% V02max load they felt slightly cold, but not really uncomfortable. At the highest exercise level they experienced thermal comfort, which was an unusual feeling for that work rate. Immersion in 35 C water initially produced a weak thermal sensation of being too warm. Pedal- ing at 20 rpm, however, was associated with occasional chills, but not frank shivering. Exercise at 50% v02max felt warmer than the same work rate in air. All three subjects expressed extreme dislike for maximal exercise in 35 C water due to heat stress. The subjects felt almost fully recovered within 5 min after 25 C immersion. The feeling of fatigue persisted for 30 min to 2 hours after 35 e immersion. 91

Discussion

The increase in treadmill V02max of 6% over previous testing for subjects DB and JM reflects, in part, their greater participation in running training during the 6 months time interval. Part of the in­ crease may also be due to the different protocol used. The protocol in the present study resembles the Bruce protocol (Bruce et aZ.~ 1973). It has been determined that the Bruce protocol may yield a significantly lower (6.5%) V0 2max for the same subjects than the Taylor protocol (Froelicher et aZ.~ 1974). The Bruce-type protocol was chosen for this study because continuous exercise minimizes the oxygen deficit and, therefore, lactic acid production, that occurs at the onset of interval work (Karlsson, 1971).

Resting data. The lower VE in water agrees with the findings reported in Chapter 4 for resting in 30 C water compared to air. The reduction is due to a smaller tidal volume. Since expiratory reserve volume is known to decrease with submersion, the smaller VT may be related to the mechanics of breathing at a smaller functional residual capacity. The reduction in fh in 25 C water is of the same magnitude observed for these subjects in 30 C water. The percent changes from air control are about the same found by Craig and Dvorak (1969) for 25 C and 30 C water. Lack of change in fh during 35 C immersion agrees with the findings of Craig and Dvorak (1969) and Arborelius et aZ. (1972). It is not known why there is a certain lactate concentration in the blood during resting conditions. In any case, lactate values in resting blood probably do not indicate muscle hypoxia. 92

Submaximal exercise data. Compared to air, ventilatory data at both water temperatures agree with findings previously reported for 30 C water. The outstanding feature during submaximal exercise is the significantly lower fh in 25 C vs. 35 C water. This is in disagreement with the data of Costill et aZ. (1967), but confirms the findings of Nadel et aZ. (1974) and Moore et aZ. (1970). Since the tachycardia of exercise was not reduced in 35 C water compared to air, immersion per se is not responsible for lowering heart rate. The fact that Tre did not increase in 25 C water and the supposition that mean skin tempera­ ture is low, tempts one to propose that peripheral vasoconstriction elicited by the colder water initiates a baroreflex which causes heart rate to decrease. Denison et aZ. (1972) found heart rate was 5-13 bpm higher in 35 C water than in air at an equal ~02 which disagrees with the present study. Unfortunately they did not, however, measure body temperature.

Maximal exercise data. Lack of a difference in V02max at the two water temperatures supports the findings of Costill (1966), but con­ flicts with the 6% lower value in 26 C vs. 33 C water reported by Nadel et aZ. (1974). The differences in immersion time, however, may prevent a valid comparison. The effects of immersion and water temperature on . . V02max, peak fh, VE, and lactate concentration are shown in Fig. 22. The large reduction in fh for 25 C compared to 35 C water apparently does not affect muscle blood flow enough to alter V02max. Although the extent to cutaneous vasodilation and venous pooling in 35 C water is not known, sufficient muscle blood flow must have been produced to 93

105

La 95 mg%

85

15~ VE L/min 145

3·90 V02maX3'80 Umin 3'70 AIR

FIGURE 22 Effect of Immersion and Water Temperature on Cardiorespi­ ratory Responses to Maximal Exercise. These are mean values for 3 subjects. Exercise was of the same intensity and duration at both water temperatures. Tre was 37.7, 36.9, and 37.5 C for air, 25 C and 35 C water, respectively. Blood lactate was 20% higher in 35 C water than 25 C water. 94

maintain ~02max in that condition also. As with total muscle blood flow, no quantitative measurements of total skin blood (CBF) during exercise have been made. Rowell (1974) estimated CBF from changes in cardiac output and blood flow to major vascular beds during heating. He suggests that "max imal" CBF in resting man may be 8 L/min, or higher; but during maximal exercise the skin becomes relatively vasoconstricted. Estimates of "max imal" CBF during exercise under hyperthermic conditions range between 2-4 L/min. Diversion of this much blood away from the active muscles probably could not be compensated by an increased oxygen extraction and V02 would fall. Since V02max was the same in 35 C as 25 C water, the assumed 2-4 L/min of CBF in 35 C could have been supplied by a greater cardiac output. If stroke volume was the same in both water tempera­ tures, estimated cardiac output would have been on the order of 2 to 3 L/min higher in 35 C water. Since maximal rpm was the same in both water temperatures, total mechanical work output was probably equivalent also. Considering the unchanged v02max it is, therefore, difficult to explain the higher blood lactate in 35 C water compared to 25 C. According to Huckabee (1958) in aerobic metabolism the lactate accumulation in blood reflects the rate at which pyruvate is being mobilized. Under anaerobic condi­ tions lactic acid is generated to replenish the hydrogen accepting enzyme nicotinamide adenine dinucleotide. To separate the mass action production of lactate from its anaerobic pathway, Huckabee introduced the "excess lactate" hypothesis. Excess lactate (XL) is the fraction of change in lactate concentration that could be expected due to an 95

oxygen lack. It can be calculated from:

XL = Ln -Pn Lo / Po where Ln and Pn = the experimental lactate and pyruvate Lo and Po = the control lactate and pyruvate. In the present study XL does not provide any additional infor­ mation about the degree of oxygen lack during maximal work in 25 C vs. 35 C water as compared to using lactate values alone. In other words the value for Pn Lo/Po is the same for both conditions. It is inter- esting to note, however, that lactate was higher and pyruvate lower following maximal exercise in 35 C water. This may account for the greater VE and VC02 in the warm water. Interestingly although the lactate level in blood was the same following maximal exercise in air and 25 C water, XL was 13% greater in air. Perhaps this explains, . in part, the 13% greater VE in air. The failure for Tre to rise during maximal exercise in 25 C water is in keeping with the subjects' sensation of thermal comfort. The rapid recovery of heart rate after exercise in 25 C water seems related to the smaller residual thermal drive, i.e., post-exercise rectal temperature. In summary, it seems clear from these findings that although

V0 2max is unaltered by short term exposure to moderately cold and warm water, the cardiorespiratory responses to maximal exercise are markedly affected by water temperature. Moreover, the results in 25 C water suggest that high values for heart rate and pulmonary 96 ventilation are not prerequisite for obtaining V02max. A critical determinant of Vo2max must be the distribution, as well as the magni­ tude, of cardiac output. A measure of the minute volume of the heart, however, would provide part of the missing information. 97

CHAPTER 7 THE EFFECTS OF BODY POSITION AND WATER IMMERSION ON STROKE VOLUME

Introduction

The finding reported in Chapter 6 that peak heart rate was reduced by 19 bpm during maximal exercise in 25 C water compared to 35 C water without a change in ~02max indicates central hemodynamics are altered by water temperature. According to the Fick equation, for V02max to be the same when heart rate is significantly reduced, stroke volume and/or arteriovenous oxygen difference must be increased. The measurement of cardiac output should, therefore, improve our understanding of the effects of immersion and water temperature on the cardiovascular re- sponses to exercise. It is believed that only two studies to date have reported cardiac output during exercise in water. Both of these studies used the in­ direct technique of C02 rebreathing. Dixon and Faulkner (1971) found that V02max and Qmax were about the same for maximal swimming and run­ ning. A decrease of 12 bpm in heart rate during swimming was compen­ sated by a stroke volume increase of 8 ml/beat. Denison et al. (1972) compared the cardiorespiratory responses of 4 subjects during supine submaximal exercise in air and 35 C water. They found a 10% greater cardiac output at rest and at all levels of exercise in water, but this increase was due to a 10% higher heart rate; i.e., stroke volume did not change with respect to its value in air. The apparent conflicting results of these two studies concerning the effect of immersion on stroke volume perhaps can be explained by their differences in water 98 temperature4 and body position used for the control condition in air. This chapter will focus on the importance of body position. Stroke volume during supine exercise may increase by only 10-20%, and frequently there is little change from the resting value (Bevegard and Shepherd, 1967). This contrasts with the 40-60% increase that occurs from rest to exercise in the upright position (Bevegard and Shepherd, 1967). The failure of stroke volume to increase with immer­ sion in the study of Denison and his associates conflicts with the findings of Arborelius et ale (1972) who used a similar water tempera­ ture, but tested their subjects in the upright posture. The discrepancy could be explained on the assumption that the end-diastolic volume was already near maximal in the supine position in air. Additionally in the supine or prone position, hydrostatic transmural pressure differences in the blood vessels below the heart would be eliminated independently of immersion. In this regard the greater stroke volume found by Dixon and Faulkner during swimming versus running could also be due to a postural effect and not to immersion per see The purpose of the present study was to estimate the stroke volume alterations that occur with postural shifts and immersion. Postural shifts are known to elicit the full range of stroke volume changes. Stroke volume in the supine position may be 40% or more greater than standing (Wang et al.~ 1960). This is because of the diminished dia­ stolic filling of the ventricles in the standing position. The method of measurement chosen was a new non-invasive technique called impedance

4Dixon and Faulkner did not report the swimming pool water temperature, but it is unlikely that it would have been above 30 C. 99 cardiography. The principle of its operation is based on transthoracic electrical impedance changes that occur in synchrony with mechanical events of the cardiac cycle. The origin of the impedance changes, however, are not fully understood. The development and evaluation of the impedance cardiographic system to measure cardiac output has been reviewed by Kubicek et aZ. (1970). Comparisons of cardiac output using the impedance method and the dye-dilution method have resulted in mostly

good, but occasionally poor correlations (Kubicek et aZ. 3 1966; Smith

et aZ. 3 1969; Harley and Greenfield, 1968; Judy et aZ. 3 1967). Judy et aZ. (1967) found a correlation coefficient of 0.92 for cardiac output values measured in 11 dogs using impedance and electromagnetic flowmeter techniques simultaneously. The impedance method has apparently not been compared to a direct method of measuring cardiac output in exercising human subjects. As a student of Kubicek, Tracy (1971) used the imped­ ance method to estimate cardiac output during submaximal and maximal cycling and treadmill exercise in 6 athletes. He found the method yielded reproducible results which compared favorably with values in the literature obtained by direct methods. His high regard (personal commu­ nication) of the impedance method as a non-invasive technique prompted its use in the present study.

Methods

The 4 subjects who participated in this study are described in Table 10. They can be considered athletic subjects who were in excellent endurance fitness at the time of testing. 100 TABLE 10 DESCRIPTION OF SUBJECTS

Age Ht Dubois BSA Wt Subject yr cm m2 %Fata kg Air Waterb

DB 25 180 1.92 14 74.6 3. 1 RD 31 178 1.89 6 70.5 4. 1

DM 25 181 1.85 6 65.3 3.7 JM 28 184 1.90 11 67.7 3.6

Mean 27 181 1.89 9 69.5 3.6 ahydrostatic weighing technique, after Brozek et aZ. (1963) bat residual volume 101

Stroke volume and cardiac output estimates were obtained by the impedance method developed and described by Kubicek and his associates (1970). Of primary importance in the present investigation was whether the values obtained were consistent with those available in the 1itera- ture and were reproducible on a day-to-day basis. Each subject was, therefore, tested on at least two separate occasions. Measurements were made in the supine, sitting, and standing positions, and sitting immersed to the suprasternal notch in 30 C water. The subject was allowed 2-3 minutes to become adjusted to a new position before measure­ ments were taken. Measurements were made in all four positions using a random sequence on a given test day. The subject wore an impermeable plastic suit (described by Fox et aZ., 1968) during immersion to prevent the possibility of current shunting to the water which might affect the impedance measurement. 5 The impedance method was followed as described and illustrated in detail by Kubicek et aZ. (1970) and Tracy (1971). A typical resting impedance cardiograph record is shown in Fig. 23. Stroke volume (Qs) was calculated from the beat-to-beat recording of impedance changes using the equation (Kubicek, 1970):

5This precaution may have been unnecessary as R. M. Smith (personal communication) has found no difference in the mean impedance of immersed subjects with or without the plastic suit. 102

IN~'RUM!N'~ 11K ~(l11111 ~ I se'C"~ . 3 f.l 1~ 7 ~4' ~ • 1 :'1 I i \ i 1 I I lI I j Ii! . II I III: I,I j L :.. 1. i.~·JI·:~Ftlq:H-.::.1+:-I+F '-:r .....••• ! I:II!'I.-.'· ~.I:::\"" ~:r:t ~!I:;J. :-l:=jt• 1 .-1 ."-j-'" :.. :. I•:., ! : I., ., ,., I" I" "'Ij i· .. •. I" '-: I~~ntt!t'T:r _f-L d~;·-! =i-rf~.:,iJI ?\,.r-H,; -.l~-L +.. J,... : I... I I· .. •· .....1.,··1..: ... ,...... , '1 ... _.. :.. F' .. '. .. )'-.\.::' .1. I I ' i. i : :1 I ! I '.\" I, .! : L.;"

'.-~l-. .'J' r "rr"_~ 1 "'''! !., I' , I,·· .. ·J··""·I···T"I"~':'.. ·' ... \l-!"'j'CI"I'j"j"""-;-:-'~":.~I~";: r..:"...:.,....\.. II ·1 .!11-- ._: : J: '. I"... !. r+'" .. :.:;.L :::J:S -;c: ·-J--:f+ L\'--1.. ,; I. R_D • :. " ··1· : ,.' .. :: : .:I.,~. ;:"I':':l~':" .. ""i +J-t' ":I~&';';~ol''''''''''''I'' ~ ..1 ... + ,\",!""".. '!'l'··,": .... \' .. I.... i...... :-.:..: ..... +.-r .." , 1 I ...... :_c. y"+-'- t"ll-l" -~. '1-'... - ....~...... ;_. '-'.'~'~'~- . .'- -t.. '--:'~-1:'" 12;~F"';' ,::':' i I .. '11 I : ' ' I: r+:"'=h ::: j~.·",!:,:·!,·+·t-·· .., ."1.-,, , ,_ -1 ' 1 I·..'1··.··".1..,. t" .. ·I·~· . -;--: l.. ,...... ,. I I 1.\ l' !'. 1 I I ' . I "'! ' ·I~: , .... ,;:··t·, ...l,-+ ..':"" -- -"-'Et<'G'-' ._; :"j-' t· ,. :: ...j.-1'"1·I.·. ·..·I'.. :.· ..· -! ,. I·.. :!' ':"u. :,: ·1 ..•.: l L~. L .... 1 I' I' , :. .. ·IO~ I" I .' I:' I 1'1:'r~· :'1. . ':11 ..I.. '''. __ .1., I., 1 l' -- .' ,. ., '- D.ck.~ I ..n. ,I I -i=rSj:--:lr.--TI~rq r~l--·f·:]=-,rT.I+ll·~ q=FII+T~' ·:.·.JI··:':lkrrH:r~l U '~l~t:'L~';+"J~ ~. ;"~l.;" ++T:l~"~ L~i Ilk+".. P1kF-P(L.:=f1"~i-=f.':; ~ ~Jt. fV'f"P~W±1#!1MJ ''': I"...r! ~ l ~11~~~".!t-... r.. ·: ... ;:.·..:·.:·]"'·,.:'J'4~r~LT.··. i'.~. ...,. . ':11 fl' ,I .' I I M I· I . ~,.~~ .,. I r." ,.j ••:t; _ .. "N'lr1 . ..:--..., ·· _. -...... ,'j lli:-1··.. '· ... I.... _...... ,. .. .:""I" ",... ._.; .. 1,_.:.. '-' ,Oi ,,,.. . '+. :. _.... ·:".1__:"..1.. .. \...... I" I'" ,. ·jl"·1"'I.· ;..,.\_.. j "1"" ..... :. -I'::' :;:.' ·d·,. "'J..::: :.;.... ;.S. ':.'. ". ",' '.';: rt ~ .,~ -. -'·1-'1 r"\"r"i' ... I I.. 1 II"" .. :. ii': '''\''''1''; c :.'~. T .. -:: .:: o L.. ., , . .Hoarrtl Soundr"':'. ....,...I: 1..1...... " .. :..L. _", 11., 'I I ~· i i~·ir;jf ~"7J!lj1 :,,·it"· I: i ., ., ·II .. .j" .". i.. ··\·T T' " i' 1"1 I i· 'j'" ";1;"1 :!!i :i: TIii Tll if.1 ~lc~J:fJjS,.~I~ :-i:-::I:" .:= =~ =::Cl'~: ~=l~-- ~ L =+:: =: =:;? ~li iIi; i$ ~ Ij~ i~; ~ ~!~~I@~ t'. ·-,-1--,·-I·..:L, ,'" :.. ·1 ...... 1..·\··· .. · .•,. '" ... 1.. 1''': ....I· ,,', .,:. i'_ ·:"~:i·Hi ~11.Tt' ,.~ '-:, ': ~ . T---: .-:...... J....L.. . '--"!" '-J...L+..J, .... ". ~ ::: ;::; ::' ":; I';, 'ill i'i: I;,! r+': 1 ri-~ itTTTi-i-f"...... \ ." .1 .. 1. ... ,- "J'+'f'1-"':/.'....,,.,,,1'" 'r':;" ,',. W jli,;~;illi 'l"fto"' ll; 1m ;. "'1",1. : .I :. ..'.: : \' '1' ", !.. :, T ::.: .i,l ',4 ::I. I! till 4i :;4. ,pi J ::-; ~ "~, l , j I' I . I.,~.;' I nill ij~!!l' "', .'1' iTTi .. - ,--+ I I• ,..... '11i z .\...... (' ..... ·t·..'I'...... ,.., ''1i ·it- "Ii '!tl Ih 11 if +.. i"'M¥ -ii, I I : I 1 I. I !'. : i 1 : . ,,'::,' ,I'" '!~ .,i...mIi ~iLp:Ill;i~ll

..1 J ,I

FIGURE 23 Impedance Cardiograph Recording at Rest. Symbols are explained in the text. 103

where

= ventricular stroke volume.

p = electrical resistivity of blood at 100 kilohertz (an average value of 135 ohm-em was assumed in all cases).

L = the mean distance (em) between the two inner electrodes, measured in front and back.

Zo = the mean impedance (ohm) between the two inner electrodes.

= the first time derivative (ohm/sec) of the impedance change (6Z); it represents peak aortic flow and is measured from zero to the most negative point in the waveform. It is calculated by dividing the magnitude of the aZ/at by the positive amplitude of a calibration record. T = ventricular ejection time (sec); measured from zero to the most positive peak of aZ/at; its value is also equal to the interval (sec) between the first and second heart sound recording.

Heart rate (fh) was determined from the EKG record by dividing the chart paper speed (em/min) by the R-R interval (em). Measurements were taken from the recording using a micrometer with subdivisions of 4 x 10-4 em. Three to five continuous heart beats were measured and averaged. Cardiac output was calculated on a beat-to-beat basis from the Fick equation:

The subject was required to perform an apneic maneuver for optimal recordings because the impedance measurement is sensitive to lung volume changes. On a verbal count-down, 5 seconds prior to recording the subject exhaled normally to functional residual capacity and maintained 104 his glottis open until the investigator signaled him to breathe, or about 10 sec total. The recording of the impedance changes were made during this period of apnea. Three recordings were obtained in each body position on a given test day. Oxygen consumption and rectal temperature were not measured since they are similar in the upY'ight and supine postures (Bevegard and Shepherd, 1967) and immersion time was only about 10 min.

Results

The results are presented in Table 11. At least three recordings were made on each subject in each position on two separate days. The measurements were high reproducible as indicated by the small standard errors. The mean resting heart rates are remarkably low and are charac­ teristic of the training-bradycardia observed in endurance athletes (Astrand, 1956). Compared to the supine position, heart rate was 23%,

16%, and 2% higher for standing, sitting~ and immersed conditions, respectively. Supine heart rate was on the average 35% (24 bpm) lower than that reported for 20 active, but not endurance-trained males of comparable size (Bevegard et aZ.~ 1960; Wang et aZ.~ 1960; Bevegard et aZ.~ 1966). Supine stroke volume was on the average 50% (56 ml) greater than in those 20 subjects, but cardiac output was 0.3 L/min (4%) lower. The cardiac output fell 1.3 L/min in going from the supine to standing position which was of the same magnitude reported by Wang et aZ. (1960) in subjects with the same body surface area as those in this study. 105

TABLE 11 EFFECT OF BODY POSITION AND IMMERSION ON HEART RATE AND STROKE VOLUME

Subject Position Heart Rate Stroke Volume Cardiac Output bpm m1 Ljmin

DB standing 74 ± 5 72 ± 3 5.3 ± 0.6 sitting 64 ± 4 84 ± 3 5.3 ± 0.4 supine 56 ± 2 133 ± 8 7.4 ± 0.7 immersed 56 ± 1 138 ± 5 7.7 ± 0.2

RD standing 50 ± 1 117 ± 6 5.8 ± 0.4 sitting 47 ± 2 127 ± 5 6.0 ± 0.4 supine 42 ± 3 165 ± 4 7.0 ± 0.1 immersed 41 ± 1 191 ± 2 7.8 ± 0.3

OM standing 45 ± 4 146 ± 5 6.5 ± 0.7 sitting 49 ± 2 165 ± 3 8.2 ± 0.2 supine 40 ± 2 212 ± 7 8.4 ± 0.5 immersed 44 ± 2 194 ± 8 8.6 ± 0.4

JM standing 48 ± 3 126 ± 4 6.1 ± 0.2 sitting 45 ± 2 132 ± 3 5.9 ± 0.1 supine 39 ± 1 158 ± 1 6.2 ± O. 1 immersed 38 ± 1 155 ± 1 5.9 ± 0.0

Mean standing 54 ± 7 115± 16 5.9 ± 0.3 sitting 51 ± 4 127 ± 17t 6.3 ± 0.6 supine 44 ± 4* 167 ± 16* 7.2 ± 0.5 immersed 45 ± 4* 169 ± 14* 7.5 ± 0.6

*Significant1y different (P<0.05) from sitting tSignificantly different (P<0.05) from standing Qs 130 114 100 74 82 83

--V'---""'---- t I f ~ ~~I'--../- 42 48 55 h 64 67 68 ./ . Q 5.5 5.5 5.5 4.7 5.5 5.6

~T':i;• ." ;I,~~,l~.~ .' :t~I~!J"4'01Io~11 't 'T'1~ 11- -rJJ:'ll/hl~'rw"" •• "",,,"""t~I~""'I4,Q* Mtt-, .."IIt"'1~'r.a t '11&~'

Sub.ject J M,S itting in air

-

FIGURE 24 Reciprocal Changei in Heart Rate and Stroke Volume at Rest. This impedance cardiograph recording shows an atypical acceleration of the R-R interval during a 10 sec breath hold. Cardiac output on a beat-to-beat basis was ... C) essentially unchanged as the fall in stroke volume was proportional to the en increase in heart rate. The EKG is inverted. 107

Supine stroke volume was significantly higher than standing (31%) or sitting (24%), but the same as immersed. Due to the small number of subjects cardiac output was not different statistica'ily in any of the six possible combinations. It was, however, 19% greater sitting in water than sitting in air. Heart rate and stroke volume were nearly the same in the supine and immersed conditions. Generally, on a beat-to-beat basis, stroke volume and heart rate did not change appreciably during a single collection period. There were, however, a few notable exceptions when the R-R interval was sig­ nificantly different from one beat to the next. In three cases the heart rate slowed, and in one case (Fig. 24) the heart rate accelerated from beat-to-beat. These data were not used, since there obviously was not a cardiac steady-state; but as shown in Fig. 24, cardiac output was remarkably stable even in the face of a 47 ml reduction in stroke volume and a 26 bpm acceleration of heart rate. These reciprocal changes are characteristic of the artificially paced heart at rest and probably reflect a smaller end-diastolic volume due to a shorter· filling time (Ross et aZ.~ 1965).

Discussion

Stroke volume is defined as the volume of blood pumped from the heart with each beat. It is the difference between the end-diastolic volume and the end-systolic volume. The absolute magnitude in a given individual will depend on the size of his heart, the conditions that affect filling, and the conditions that affect emptying. The degree of diastolic filling has an anatomic limitation, but is otherwise 108

primarily determined by filling time and filling pressure. Filling time is a reciprocal function of heart rate, so a slower rate would favor a larger stroke volume. Filling pressure depends on the rate of venous return and central venous pressure. Conditions that increase central blood volume promote a greater filling pressure. Examples of such conditions are the supine position, peripheral vasoconstriction,

and apparently water immersion (Arborelius et aZ' 3 1972). The degree of systolic emptying is determined by preload, myocardial contractility, and after load. The greater the preload, or initial stretch of the ventricular myocardial fibers, the greater the stroke volume. This is called the Frank-Starling mechanism. Factors that influence the end­ diastolic volume also affect the preload. Among these are central blood volume, venous tone, intrathoracic pressure, venous compression by skeletal muscles, distensibility of the myocardium, atrial contrac­ tion, and body position. As suggested by Arborelius et aZ. (1972), the greater stroke volume observed during head-out immersion is probably the result of conditions that increase preload. Afterload is a direct function of the left ventricular pressure. The latter depends on ventricular size and arterial pressure. Accord­ ing to Braunwald (1974), afterload operates via a negative feedback loop to maintain arterial pressure equilibrium. For example, cold­ induced peripheral vasoconstriction increases central blood volume which favors a greater stroke volume and cardiac output. If both peripheral resistance and cardiac output rise, so will arterial pres­ sure. The increased arterial pressure brakes stroke volume by the afterload mechanism and decreases heart rate by the baroreceptor reflex. l®

Cardiac output is, therefore, lowered which allows arterial pressure to return toward its previous level. When preload and afterload are held constant, stroke volume is a function of the contractile state. This a property of cardiac muscle which determines the force of contraction or inotropic events of the heart. Contractility is increased by beta-adrenergic stimulation and decreased by vagal stimulation. The reflex slowing of the heart rate that is produced by baroreceptor stimulation also causes a negative inotropic effect. Supine cardiac output (7.2 L/min) as determined with the impedance method compared favorably with the values (x = 7.5 L/min) reported by

others using direct Fick and dye-dilution methods (Bevegard et al. 3

1960; Wang et al. 3 1960; Bevegard et al. 3 1966). Bevegard and his associates (1960, 1966) using a direct Fick determination found an 8-14% greater reduction in cardiac output from supine to sitting than the present findings. This was due primarily to a 6-16% greater fall in stroke volume in their studies. Perhaps the smaller reduction in cardiac output noted here is an effect of endurance training which leads to a better circulatory adaptation to postural stress. It is well known that inactivity can result in poor tilt table tolerance (Astrand and Rodahl, 1970). The subjects in the present study had a stroke index of 88 ml/m2 in the supine position compared to 54 ml/m2 for the 20 reference subjects previously mentioned. This large difference may be due in part to the greater endurance fitness of the present subjects. A large resting stroke volume and slow heart rate are characteristic of 110

endurance-trained athletes (Hermansen et aZ.~ 1970). Endurance train­ ing typically leads to a rise in stroke volume and a proportional decrease in heart rate (unchanged cardiac output) at a given submaxima1 oxygen uptake (Hartley et aZ., 1969). The stroke volume has been shown to increase with training without an observed increase in heart volume (Ekblom, 1969). This indicates that the end-systolic volume may be reduced following a training program. The mechanism for the training bradycardia in humans is not known. Lin and Horvath (1972), however, found that training bradycardia in rats resulted from a proportionally greater reduction in sympathetic tone than an increase in parasympa­ thetic tone. Stroke volume increased by 33% during immersion compared to sitting in air. This is the same percentage increase as found by Arborelius et aZ. (1972). Cardiac output, however, increased only 19% (1.2 L/min) compared to their 32% (1.8 L/min) increase. The difference is due to a larger reduction of heart rate during immersion, which may have been caused by the colder water temperature used in the present study. The values of heart rate and stroke volume in the supine position and during 30 C immersion were the same which suggests that central hemodynamics may be similar for these conditions. In summary, the use of the impedance cardiograph to estimate cardiac output was non-traumatic and well-accepted by the subjects. Values obtained for stroke volume with postural shifts and water immer­ sion compare very favorably with those reported by others using direct methods. Furthermore, the impedance method appears to operate satis­ factorily on a beat-to-beat basis when heart rate and stroke volume 111 are not in a steady-state (Fig. 24). The results of the present study suggest that the greater stroke volume during maximal swimming compared to running as reported by Dixon and Faulkner (1971) could be due ex­ clusively to the postural differences. In addition, the lack of increase in stroke volume during supine underwater exercise compared

to the same exercise in air (Denison et al. 3 1972) could also be ex­ plained in part by the assumption that stroke volume is already maximal in the supine position. 112

CHAPTER 8 CARDIORESPIRATORY RESPONSES TO GRADED CYCLING EXERCISE IN AIR AND IN 25 C, 30 C, AND 35 C WATER

Introduction

The importance of venous return in the regulation of cardiac output has been emphasized by Guyton (1968). Normally there are two principal factors that determine the rate at which blood returns to the heart from the peripheral circulation. These are 1) the degree of vasodilation in the peripheral vasculature, especially of veins; and 2) the ratio of blood volume to the momentary capacity of the circulatory system, i.e., its degree of filling, which Guyton refers to as mean systemic pressure. Either a decrease in peripheral resist­ ance or an increase in mean systemic pressure will increase venous return and, therefore, cardiac output. The role of the heart is con­ sidered as one of a force-feed pump. That is, the heart plays a permissive role in the regulation of cardiac output. It "permits" the cardiac output to be regulated according to the rates of arterial run­ off and venous return. Sugimoto et al. (1966) confirmed this permis­ siveness in resting dogs by showing that the range of heart rate at which cardiac output was maximal depended on the central venous filling pressure. As the filling or atrial pressure was increased the optimal heart rate for maximal cardiac output shifted to a higher rate. "Maxima'" cardiac output increased. In other words the decrease in stroke volume that resulted from insufficient filling time at high heart rates was prevented by a higher atrial pressure. The higher 113

atrial pressure must have produced a greater ventricular filling during diastole. The limitation in ventricular filling that occurs in resting dogs when the heart rate is artificially increased probably does not occur

during exercise (Horwitz et al' 3 1972). Presumably the muscle and respiratory pumping mechanisms by which returning blood is forced through systemic veins during exercise are major factors which enhance ventricular filling. Horwitz et al. (1972) found that left ventricular end-diastolic diameter and stroke volume were increased during strenuous treadmill exercise in dogs, indicating that filling time was not limit­ ing. Their findings suggest that the Frank-Starling mechanism was important in contributing to the greater stroke volume during exercise. Astrand et al. (1964) and Hermansen et al. (1970) using young, athletic subjects found no tendency for stroke volume to decrease during maximal cycling or treadmill exercise. Stroke volume either leveled-off at 40% V0 max (Astrand et al' 1964) or continued to increase up to 100% V0 max 2 3 2

(Hermansen et al' 3 1970). Would enhanced preloading contribute to an even greater exercise stroke volume? As suggested by Arborelius et al. (1972), water immer­ sion may be a convenient experimental model for altering the preload in human subjects. They found the transmural pressure gradiant of the right atrium increased by a mean of 13 mmHg with head-out immersion. There was also a corresponding 35% increase in the resting stroke volume and cardiac output. An analogy between these find"ings and those of Sugimoto et al. (1966) might be drawn and extended to exercise in water. If the higher right atrial pressure persists during exercise then a greater 114

cardiac output could be expected at a given heart rate. Does water temperature affect afterload? The findings presented in Chapter 6 suggest that water temperature has a marked effect on both submaxima1 and maximal heart rate. This effect is perhaps mediated by changes in after10ad since skin temperature is known to influence cutaneous vascular resistance which, in turn, may alter mean arterial pressure. Tachycardia is always a consequence of heat stress. Accord­ ing to Rowell (1974), venous pooling in the cutaneous circulation during body heating may be the underlying cause of the elevated heart rate whether it occurs at rest or in exercise. Warm water, of course, would potentiate venous pooling whereas cold water might prevent it. Under conditions where heavy exercise leads to hyperthermia the cardiovascular system is faced with the problem of how to meet the combined demands for oxygen transport to working muscle and heat trans­ port to skin. Important questions of priority are now raised because the two demands require simultaneous vasodilation in two major vascular beds, the skin and the active musculature. They essentially compete for the cardiac output. This raises the question of whether the cardiac output can adjust to the additional vasodilation in skin during exercise and still maintain the required arterial blood pressure. In water, however, the problem may not be so great. Because of the preloading effect with immersion the cardiac output may be able to increase accord­ ing to the needs of both metabolic and thermal demands. In other words, the assist given to the rate of venous return with water immersion may allow cardiac output to exceed its maximal permissive level in air. 115

The purpose of the following study was to estimate cardiac output during mild to maximal cycling exercise in air and in water at three different temperatures. Generally this study repeats the objectives of the experiments presented in Chapters 4 and 6 and includes an esti­ mation of cardiac output. Of particular importance in the present study was to determine whether alterations in peak heart rate caused by water temperature reflect similar changes in cardiac output. This is an im­ portant consideration because cardiac output is thought by many to be the major factor limiting maximum oxygen uptake (Rowell, 1974; Gollnick et aZ.~ 1972). If Vo2max is unchanged and cardiac output is signifi­ cantly lower in 25 C versus 35 C water, as was the case for heart rate, total blood flow per se would not be the limiting factor. In fact if peak cardiac output like heart rate is a direct function of water temperature, this would suggest that the cardiovascular system is not fully stressed during maximal exercise in colder water.

Methods

The four athletes who were described in Chapter 7 were also sub­ jects for the present investigation and are further characterized in Table 12. Mean skinfold thickness is the unweighted average of the

10 sites listed in Chapter 2. Vo 2max and Qrepresent the highest values obtained for treadmill running. The primary objective of this study was to assess the effects of water immersion and water temperature on the cardiodynamic responses to graded cycling exercise. Treadmill running was included, however, because the cardiovascular oxygen transport capacity, in terms of 116

TABLE 12 DESCRIPTION OF SUBJECTS

a . b Age Ht Wt MSF V0 2max Q Subject yr cm kg mm L/min m1/kg-min L/min

DB 26 180 73.3 10.2 4.49 61 24.1 RD 31 178 71.1 6.3 4.89 69 28.5 DM 25 181 65.9 5.8 4.59 70 30.5 JM 28 184 68.6 7.7 3.89 57 24. 1

Mean 28 181 69.7 7.5 4.47 64 26.8 aMean skinfold thickness bTreadmill running 117

V02max, stroke volume and cardiac output, is typically higher in run­ ning than for cycling. Maximal running, therefore, provides a better measure of cardiovascular capacity. The modified Taylor protocol as shown in Fig. 2 and described in Chapter 2 was followed. Oxygen uptake was determined using the Daniels gas collection system (Daniels, 1971) with a Collins Trip1e-J respira­ tory valve in all cases. Expiratory minute volume, heart rate and rectal temperature were measured as previously discussed. Resting determinations in water were made with the subject sitting quietly and the water unstirred. Cardiac output was calculated on a beat-to-beat estimation of stroke volume and the preceding R-R interval using the thoracic impe­ dance method (Fig. 25). Preliminary experiments revealed that it was not possible to perform the apneic maneuver (lung volume at FRC, open glottis) required for this method during heavy exercise. During hard cycling exercise it was difficult to restrain from "settingll the thorax. This is similar to doing a Valsa1va maneuver and it inter­ ferred with the measurement of thoracic impedance. The stroke voZume estimation was~ therefore~ made within 5 sec post-exercise~ as was previously done by Tracy (1971). Post-exercise measurements, however, do not represent steady-state conditions. Gilbert et aZ. (1967) and Davies et aZ. (1972) have described the oxygen uptake, heart rate, and cardiac output transients following the end of exercise. According to their data cardiac output does not change significantly (0 to -3%) within 5 sec of the end of heavy exercise. In the present study a second estimation of cardiac output was made at 10 sec post-maximal 118

ri~~~;;-r J~~~~~~~. ! f l

EKG

'I" ~,\! III l JIll lL I" I ['il """111'f'i~I'/Y''''1 jI~ff"'/#'jl il(~"11Y""""'1 i'l'l""ll V\'~'4'V Irl",vI~I..v Ij'!'"'vI\\-

Heart Sounds j fh= I 03 Os= 143 Q= 14.8 ~""~~ SUBJECT: JM. 25°C Water

FIGURE 25 Impedance Cardiograph Recording In@ediately Following Exercise. The recordings were generally of this quality after submaximal exercise. 119

exercise and in running, cycling in air and cycling in 35 C water it was not significantly different from the first 5 sec. In those three conditions post-exercise heart rate did not change from the exercise level within zero to 5 sec of recovery. Immediately following sub­ maximal exercise in 25 C and 30 C water, however, there was often a strong "off-response" for heart rate and the R-R interval sometimes increased significantly as shown in Fig. 26. Whenever this happened stroke volume would increase, presumably as the consequence of an increased fi1'ling time. These reciprocal changes were also noticed occasionally at rest (Fig. 24). As a result the calculated cardiac output did not change a great deal from beat-to-beat. The example shown in Fig. 26 represents one of the few cases when cardiac output did change from beat-to-beat and surprisingly it increased. In order to get a better estimate of exercise stroke volume in 25 C and 30 C water the calculated cardiac output was divided by the exercise heart rate. Since the heart rate during exercise was always higher than 0 to 5 sec post-exercise at these water temperatures, this procedure resulted in a lower calculated exercise stroke volume than that observed in the immediate post-exercise period. On the average this correction resulted in the calculated exercise stroke volume being about 5 ml lower than that measured. It is felt, however, that the lower stroke volume probably represents a better estimate for exercise. This procedure of extrapolating recovery stroke volume back to exercise was justified on the grounds that recovery cardiac output did not change within the 3 to 4 beats measured in spite of the falling heart rate. It 38860 t t i End exercise I lIllC 2 sec 3 soc

". JPost-Jex fh = 83 77 77 78 ~. lJ ~~\~~i~",/~~~.,...... A.----J1I __ EKG

exercise fh = 88 II I[ ! Ii Os= 174 186 ~ 193 193 ,~~1:\1~1J~li!MI~\IJ4\,~I,'I~jIM~\I~!I~,~\,--/,J~~~~ Q= 14.4 14.3 14.9 15.1 dZ/dt

h_~vJ"~N...A.J!'~~,..JJll~.,\.I~'~""""I\'" • 111~1~ ')_ .. -.. ,Ir r-~ -.. ,1/ I tJ' IU 'l"

Heart Sounds

Subj. R.D. 40% VA MAX, 6Z 2 25°C Water

FIGURE 26 Abrupt Slo\ving of Heart Rate after Exercise in Cold Water. It was not unusual for the R-R interval to increase rapidly in 25 C water following the end of submaximal exercise. Cardiac output on a beat-to-beat basis was typically maintained by a reciprocal increase ..... in stroke volume. Exercise 002 in this case was 1.7 L/min. N o 121

As in the previous study, the subjects wore a water impermeable plastic suit and 4 kg weight belt during immersion experiments. The use of this suit resulted in a greater resistance to cycling and conse­ quently caused a leftward shift in the V02/rpm curve shown in Fig. 5. The extent of this shift was small, however, and was corrected by re­ ducing the rpm slightly for a desired oxygen uptake. The insulation provided by the suit was not estimated, but since its thickness was only 0.16 mm it was probably small. Mean water temperatures were 24.9 ± 0.2 C, 29.9 ± 0.3 C, and

35.0 ± 0.1 C. Immersion time averaged 49 ± 3 min and was not signi­ ficantly different for the 3 water temperatures. The immersion time includes a 5 min post-exercise recovery period.

Results

Individual data are presented in Appendix E. Mean cardiorespira­ tory data are shown in Table 13. The subjects had no difficulty in performing the apneic maneuver after maximal exercise. An example of a good recording is shown in Fig. 27. Only three tests had to be repeated and these were due to faulty equipment operation. The wearing of the impedance electrodes around the chest and abdomen did not hinder respiratory movements. In general, the selected submaximal exercise intensities closely approximated the desired relative workloads of 20, 35, 50, and 70% V02max. TABLE 13 MEAN CARDIORESPIRATORY RESPONSES TO REST, SUBMAXIMAL, AND MAXIMUM EXERCISE FOR UPRIGHT CYCLING IN AIR, 25, 30, 35 C WATER, AND TREADMILL RUNNING . . %V0 max R T V0 2 2 VE re fh Q Qs (a-v)02 RUNNING .34 ± .03 8 ± 1 .79 ± .04 14.2 ± 1.3 - 61 ± 9 7.1 ± 0.9 122 ± 15 5.0 ± 0.6 .95 ± .06 21 ± 1 .80 ± .06 29,7 ± 1.8 - 78 ± 5 11.6±0.9 150 ± 14 8.3 ± 0.5 2.38 ± .11 53 ± 0 .87 ± .05 67.8 ± 2.3 - 123 ± 5 18.7 ± 1.4 151 ± 13 12.2 ± 0.9 3.38 ± .16 72 ± 3 .93 ± .05 100.0 ± 11. 1 - 155 ± 7 22.9 ± 1.2 149 ± 11 14.9±1.0 4.47 ± .21 100 ± 0 1. 11 ± .02 161.1 ± 10.7 - 180 ± 3 26.8 ± 1.6 149 ± i 0 16.8 ± 0.8 CYCLING IN AIR .35 ± .03 9 ± 1 .84 ± .06 15.8 ± 3.4 37.0 ± .04 56 ± 9 6.8 ± 0.4 127 ± 16 5.1 ± 0.5 .90 ± .07 24 ± 2 .88 ± .04 28.2 ± 3.3 36.95 ± .02 76 ± 10 9.7 ± 0.6 132 ± 14 9.3 ± 0.7 1.37 ± .06 35 ± 2 .91 ± .02 43.1 ± 3.4 37.05 ± .02 94 ± 9 12.2 ± 0.7 133 ± 13 11.2 ± 0.8 2.06 ± .11 53 ± 1 .92 ± .04 58.3 ± 5.0 37.2 ± .02 119 ± 5 16.1 ± 1.3 138 ± 13 12.9 ± 0.4 2.94 ± .13 76 ± 3 1.02 ± .01 91.8 ± 5.0 37.4 ± .04 146 ± 6 19.5 ± 0.9 133 ± 8 15.1 ± 0.4 3.88 ± . 16 100 ± 0 1.14 ± .02 152.3 ± 11.8 37.8 ± .07 171 ± 4 24.3 ± 0.3 142 ± 5 16.0 ± 0.5 CYCLING IN 25 C WATER .35 ± .01 9 ± 1 .79 ± .03 13. 1 ± 1.1 37.2 ± .08 49 ± 7 7.1 ± 0.4 152 ± 16 5.0 ± 0.4 .95 ± .06 25 ± 1 .78 ± .03 26.6 ± 1.6 37.0 ± .06 71 ± 6 10.0 ± 0.6 142 ± 5 9.6 ± 0.6 1.51 ± .07 40 ± 1 .82 ± .02 39.0 ± 2.5 36.9 1: .07 90 ± 3 14.0 ± 0.3 155 ± 6 10.7 ± 0.3 2.10 ± .10 55 ± 1 .87 ± .03 55.0 ± 3.3 36.8 ± .11 105 ± 4 15.7 ± 0.6 151 ± 9 12.7 ± 0.8 2.67 ± .04 71 ± 3 .93 ± .01 75.2 ± 2.6 36.9 ± .15 120 ± 5 18.4 ± 0.1 154 ± 5 14.5 ± 0.2 3.78 ± .15 100 ± 0 1.10 ± .03 144.4 ± 6.1 37.1 ± .18 157 ± 3 23.1 ± 0.4 147 ± 4 16.3 ± 0.5

..... N N TABLE 13 (Continued) MEAN CARDIORESPIRATORY RESPONSES TO REST, SUBMAXIMAL, AND MAXIMUM EXERCISE FOR UPRIGHT CYCLING IN AIR, 25, 30, 35 C WATER, AND TREADMILL RUNNING . . . % V0 max R T Q Q V0 2 2 VE re fh s (a-v)02 CYCLING IN 30 C WATER .29 ± .03 8 ± 1 .89 ± .05 12.1 ± 2.2 37.15 ± .04 52 ± 6 7.4 ± 0.7 153 ± 13 4.1 ± 0.5 .81 ± .13 21 ± 2 .83 ± .04 23.4 ± 2.9 37.0 ± .04 66 ± 8 10.2 ± 1.1 156 ± 6 7.8 ± 0.5 1.55 ± .08 41 ± 2 .85 ± .03 42.7 ± 2.3 36.9 ± .03 89 ± 6 14.0 ± 1.3 158 ± 12 11.2 ± 0.6 1.99 ± . 11 53 ± 4 .92 ± .05 55.2 ± 3.6 36.95 ± .24 104 ± 3 16.6 ± 0.6 161 ± 11 12.0 ± 0.9 2.58 + .17 69 ± 4 1.01 ± .06 75.0 ± 7.6 37. 1 ± .09 121 ± 8 19.6 ± 0.8 164 ± 12 13.1 ± 0.7 3.76 z .18 100 ± 0 1. 10 ± .02 132.4 ± 6.5 37.4 ± .10 158 ± 4 25.2 ± 0.6 159 ± 6 14.9 ± 0.6 CYCLING IN 35 C WATER .32 ± .01 9 ± 0 .82 ± .02 12.2 ± 0.3 37.25 ± .15 59 ± 4 8.3 ± 0.5 144 ± 12 3.9 ± 0.2 .85 ± .07 23 ± 1 .80 ± .01 25.0 ± 1.8 37.25 ± .12 77 ± 5 11.5 ± 1.0 147 ± 9 7.5 ± 0.8 1.44 ± .08 38 ± 1 .81 ± .01 36.0 ± 3.3 37.3 ± .13 101 ± 3 14.6 ± 0.6 145 ± 6 9.8 ± 0.2 2.04 ± .08 55 ± 1 .89 ± .03 56.0 ± 1.7 37.45 ± .11 123 ± 3 17.6 ± 0.6 144 ± 9 11.6 ± 0.4 2.52 ± .17 68 ± 3 .92 ± .02 74.6 ± 7.3 37.6 ± .08 139 ± 5 21 .1 ± 0.7 152 ± 6 11.9 ± 0.6 3.70 ± .15 100 ± 0 1.09 ± .01 145.7 ± 7.9 38.0 ± .12 170 ± 3 27.1 ± 0.3 160 ± 2 13.7 ± 0.7

N W fit t end exercise Isec 2 act; 3 sec )\JwvJ\~IJ~itJ~WUJJuuLJUuU1 dZ/dt

! I . I •I ~ I ! I I : • , I • ," I I. , I i A,_I .,'1 i. '\ j'. At") ,.)\~{"".i\...-~j .··1'\/,-,,1 / ../.....) ./'''-~J ./.•""--! / ...... -1 /...... J .r~ ...... J /1__,1 ,J"-J~J J-./'}..,A 1; r.rJ~-{ -~~.... 'r'/l1 t ../'; ('" 1 • Vii , V r V v ~U { • '/ ~I'- r ~\ ~r! ~,.1\ I 1 1 ,1\llll\~ ~ I,I!\\, 'llll,'lh1;\1, 11"i!11 ,:' I'" ,J.I.I "II\.,i, t ,11,1,,1, 1,1111 I,i'll II /r !' 1J\'i!! 'I JLI ,,1.1 ;,\ 'I!!I .~Ij:'l\i i!rl!~i:', ,;t\li~'~1] ~,4i"'i~:i ;~I'ii 1!'r~Ji'\IP ~rl ~nll ill iHJi~;i /J;'j':ili: 111; Ilj .ii" ,/t>: /! II )'!IFI11) ;:J't·ji I,}iij··'I1JIjld t\!;:llJj ;MJjj,i',r l,:: II ,rli/! 1\',' 11 ,J/ 1",'\; ", ,IV i, ", I H'eb;~ S~ ~~~s Ii JI I Ii I ," i'i I ,1 1 'rl" 'il,,~ !i'i

. end apnea begin apnea I

f' r r-" II ~~ -.l(\." //__--I-...... J ", '-.J ',,- '" /-,...1 \ r" ! \[v'./V\i\.....J~ J F '0. 'v ~ \J\) [\"'\NVlj\j SUBJEC T D. M. Maximal exercise, 35 °c water

FIGURE 27 Example of Impedance Cardiograph Recording Following Maximal Exercise. The subjects were required to perform an apneic maneuver immediately following exercise for 3 to 5 sec in order to obtain the impedance measurement. In this recording it can be seen where the --' N subject began his expiration to a relaxed lung volume. +:> 125

Results of Submaximal Exercise

Heart rate. The individual data, regression lines and best fit equation for fh vs. V02 for the 4 cycling conditions are shown in Figs. 28-31. The slopes of the regression lines are not significantly different. At a given submaximal V02 the mean value for fh was 0-8 bpm higher during cycling in air than for running (Fig. 28). This finding supports the results reported by Hermansen et al. (1970). fh was 9-11 bpm lower in 25 C and 30 C water than in air (Figs. 29,30) which con­ firms the findings described in Chapter 4.

The regression lines for fh vs. V02 in 25 C and 30 C were prac­ tically identical (Fig. 30). The submaximal heart rates reported by McArdle et al. (1971) for college swimmers swimming in 27-28 C water are 20-25 bpm higher at a V02 of 3.0 L/min than those found in the present study in 30 C water. This discrepancy supports the conclusion of Holmer et al. (1974a) that the subjects used by McArdle and his associates were not very fit swimmers. This may account for part of the reason why their V02max was 10% lower for swimming than treadmill walking. Cycling in 35 C water produced a significantly higher (13-20 bpm) fh than for 25 C water at a given submaximal V02 level (Fig. 31). This confirms previous results reported in Chapter 6. Compared to cycling in air fh was about 4% higher in 35 C water (Fig. 31) which is somewhat lower than the 10% elevation reported by Denison et al. (1972).

The coefficient of variation for fh vs. V02 ranged from 88 to 94%, indicating most of the variance in heart rate is accounted for by the different regression equations. 126

200 Cycling in Air • 175 fh = 31.6 "02 +50.4 o r =.94 150

125 f h bpm 100

75 o DB o RD A OM 0 " J til 50 A A

I I i I 1.0 2.0 3.0 4.0

V02 I L/min STPD

FIGURE 28 Heart Rate as a Function of Oxygen Uptake During Running and Cycling in Air. In this and the next 3' figures, the individual fh responses of the 4 experimental subjects are sh9wn versus V02 as the independent variable. For a given V02' fh was 0-8 bpm slower for running. 127

200 Cycling in 30°C Water

175 ", fh = 30.8 \;02 + 41.4 ~;" ;" r =.97 / ;" ;" ;" 0 ° 150 ", cycling in air""::l:+"'~ / / ;" / 125 ", ", f , ;"0 h ;" fY ", 0 bpm o / 100 ", ", ;" ;" ~ a ;" " ;" 75 Q DB ",0" RD III ° " II DM 0" " fA JM 50

I I I I 1.0 2.0 I. 3.0 4.0

V02 ,L/min STPD

FIGURE 29 Heart Rate as a Function of Oxygen Uptake During Cycling in 30 C Water. The parallel shift in the regression line for water represents a lowering of fh at a given VOZ of 10-12 bpm. This is slightly larger than the reductlon shown in fig. 9. 128

200 Cycling in 25°C Water

175 fh c 30.2 VOl. + 41.7 r =.95 150 ° cycling

f 125 h , bpm 100

75 DB RD OM JM

I I I I 1.0 2.0 j 3.0 4.0

V02 ,L/min STPD

FIGURE 30 Heart Rate as a Function of Oxygen Uptake During Cycling in 25 C Water. The fh-V02 response in 25 C was essentially identical to that in 30 C water. 129

200 Cycling in 35°C Water

175 fh = 32.7 VOz + 52.1 r =.97 ° 150

0 air

I) 125 f h , bpm 100 in 25°C water o

75 " DB ° RD 1I OM A J M 50

I II I I 1.0 2.0 3.0 4.0

V02 ' L/min STPD

FIGURE 31 Heart Rate as a Function of Oxygen Uptake During Cycling in 35 C Water. Regression lines for air and 25 C water are shown for comparison. The mean fh was 10 and 13 bpm higher at rest and maximal exercise, respectively, in 35 C versus in 25 C water. Predictions made from the corres­ ponding regression lines, how~ver, result in a somewhat greater difference at higher V02's. 130

Cardiac output. The individual data were used to calculate the

mean squares regression lines for 0 vs. V02 in each of the five experi­ mental conditions. These graphs are shown in Figs. 32-36. The differ­ ence between the slopes of any two regression lines was calculated from the residual mean squares and determined to be statistically non-signi­ ficant (Snedecor and Cochran, 1967). The regression lines are, there­ fore, essentially parallel. The regression lines for treadmill running and cycling in air compared favorably with the same relationship reported by others who used direct methods for measuring cardiac output (Table 14, Figs. 32, 33). The correlation coefficient for the regression line calculated from the impedance data ranged from 0.96 to 0.99 which means that 92 to 98% of the variance in Qis explained by the derived regression equation. The regression line for cycling in 25 C water is nearly identical to that for cycling in air (Fig. 34). Although the slopes remained the same, the intercept progressively increased with water temperature. The elevation of the regression line for cycling in 30 C water is signifi­ cantly greater than for cycling in 25 C water when V0 2 is above 35% V02max (Fig. 35). The same is true for 35 C water over 30 C water when V02 is above 50% V0 2max. Cardiac output was 9-11% higher cycling in 35 C water than in air which agrees with the findings of Denison et aZ. (1972). The comparison between 25 C and 35 C water is shown in Fig. 36.

Stroke volume. The relationship between Qs and V02 for the various conditions is shown in Fig. 37. The same relationship is shown in Fig. 38 where Qs was derived from the corresponding regression equations for

Qand fh at a given V02. TABLE 14 REGRESSION EQUATiONS FOR CARDIAC OUTPUT DURING EXERCISE FOUND IN PRESENT STUDY COMPARED WITH PUBLISHED VALUES

Regression Correlation Author Type of exercise Subjectsa Method Equation Coefficient Q= · Present study Treadmill 4 athletes impedance 4.74 Y02 + 6.45 .96 Ekblom &Hermansen, II 5 athletes dye-dilution 5.16 V0 + 5.40 .99 1968 2

II II · Hermansen et aZ.~ 13 athletes 5.0 V0 + 4.8 .95 1970 2 McDonough et al.~ II 16 coronary direct Fick 5.9 V0· + 3.45 .88 1974 patients 2 Reeves et aZ." II 10 normals II 5.92 V0· + 3.74 .97 1961 . 2 · Present study Upright cycling 4 athletes impedance 4.89 ~02 + 5.43 .98 Astrand et al. ~ II 12 athletes dye-dilution 4.35 V0 + 6.55 .96 1964 2 Hermansen et al.~ II 13 athletes II 5. 1 V0· + 4.8 .96 1970 2 Present Study II 4 athletes impedance 5.47 V0· + 6.79 .98 (35 C water) 2 Deni son et al. ~ Supine cycling 4 normals C02 rebreathing 6.90 V0 + 5.0 .97 (35 C water) 2 amale subjects only -" w -" 132

/I 1 30 1 I 1 1 1 1 1 1 Treadmill Exerci se 1 1 1 /I 1 25 1 1 1 6 01 • 0= 4.74 V + 6.45 1 02 0 r =.96 0 ... /I 20 1\

0, L/min 15

Rccvcf al 01,1961

10 • DB 0 RO /I OM , JIA

~A

5 :;;c i I I i i 1.0 2.0 30 40 5.0

' STPD \/02 L/min

FIGURE 32 Cardiac Output as a Function of Oxygen Uptake During Tread­ mill Running. In this and the following 4 graphs the iQdi­ vidual data of 4 subjects are plotted as a function of VOZ' The heavy regr~ssion line depicts the best fit relationshlp getween Q and V0 2. It begins at the resting V02 and ends at V0 2 max. The other regression lines were calculated from the raw data of previous studies which used a direct method for measuring cardiac output under similar conditions of exercise. 133

Cycling in Air

25

0= 4.89 "02 + 5.43 r = .98

20 o

0, L/min 15 o / 0 10 /~Astrond etol,I964 I o III~ Hermansen et 01,1970 /1 I 0 Cl DB Y ~ ~~ A J M

i I I I 1.0 2.0 3.0 4.0

V02 ,L/min S T P 0

FIGURE 33 Cardiac Output as a Function of Oxygen Uptake During Cycling in Air. The subjects used by Hermansen et al. (1970) were very similar in aerobic power to the present subjects. 134

Cycling in 25°C Water

25

Q=4.66 \/02 + 5.89 o ·. r =.99 "\

20

0, L/min 15

Gl 6. Ii 7~cycling in air ... ~ ... ~ 10 .­ .­ III DB .- o RD I CD .- I Ii Ii DM ;i A JM A

I I I I 1.0 2.0 3.0 40

V02 ,L/min ST PO

FIGURE 34 Cardiac Output as a Function of Oxygen Uptake During Cycling in 25 C Water. Compared to air, Qwas unchanged in 25 C water. 135

Cycling in 30°C Water

25

Q =5.08 "02 + 6.18 r =.97

20

° 0, ° L/min 15

i.

10 o DB ° RD to DM A J M

I I I I 1.0 2.0 3.0 4.0

V02 ' L/min STPD

FIGURE 35 Cardiac Output as a Function of Oxygen Uptake During Cycling in 30 C Water. Compared to air, the regression line for 30 C water was elevated 0.8 to 1.3 L/min. 136

Cycling in 35°C Water o

25

= 5.47 V0 + 6.79 Q 2 r = .98

.... 20 6' Denison et al,IS72-.z...,,/ Waier I I Q, I 9 L/min I I 15 1.11. o

10 o DB o RD A OM A JM

5 I I I I 1.0 2.0 3.0 40

V I L/min STPD o2

FIGURE 36 Cardiac Output as a Function of Oxygen Uptake During Cycling in 35 C Water. Compared to 25 C, the regression line was elevated 1.5 to 4.0 L/min. The subjects of Deni~on et aZ. (1972) did supine exercise in 35 C water and Q was estimated using.a C02 rebreathing technique. 137

160

150 OS. ml

140 e---/ 130 )( Running o Cycling in air A Cycling in 25°C Wator o II II 30°C II'

" II II 350C II

120 L-.--,r--.--.....----,--...... ---..,..----.---r--- 1.0 2.0 3.0 4.0

V02 ' L/MIN

FIGURE 37 Stroke Volume as a Function of Oxygen Uptake for Five Conditions of Exercise. Individual points represent the mean value of 4 subjects. Points at the far left were determined at rest in the same position as used. for exer­ cise. 138

160

150

Q '. 140 S' ml

)(/ 130 o CYCLING, 30·C WATER A 35·C l:i. 25 ce X RUNNING, 25·C AIR

o CYCLING, U " 120

f , i i i 1.0 2.0 3.0 4.0

V02 , L/mm

FIGURE 38 Stroke Volume as a Function of Oxygen Uptake. The points shown in this figure were calculated from the corresponding regression equations f9r 0 and fh as a function of a given V02 value; i.e., Qs = Q/fh· 139

A marked increase (23%) in Os was observed on transition from rest to exercise in the treadmill series. This increase has also been ob­ served by Hermansen et aZ. (1970) and to a greater extent by Wang et aZ. (1960). The increase in Os was much smaller from rest to exercise for cycling. In the cold water, however, Os fell 10 ml/beat at the lightest work rate. There was a tendency for Qs to increase gradually during submaximal cycling exercise, but the overall change was less than 15 ml/beat. Submaximal Qs was not statistically different in the various water temperatures. Above 50% 002max Os was significantly greater in 30 C water than in air. During submaximal exercise Os was on the average 13-26 m1/beat (10-19%) higher in water than air. The increase of about 15 ml/beat in 35 C water compared to air conflicts with the findings of Denison et aZ. (1972) who reported no difference.

Expiratory minute volume. Regression lines for 0E vs. V02 where V02 is less than 80% V02max during cycling are shown in Fig. 39. The differences in VE for the various conditions of exercise are small and not significant. R values parallel the differences in VE'

Rectal temperature. Tre responses are shown in Fig. 40. Above 35% V02max Tre is significantly lower in 25 C and 30 C water than in either 35 C water or air. The difference between 35 C water and air is due mainly to the lower initial Tre in air. The rate of change is about the same. Tre was directly related to water temperature as was pre­ viously discussed in Chapter 6. 140

100

75

l'J AIR V = 289 V +35, r=.98 . E .o 2 25 o 30° C Water ~E=28.9~02+0.3, r=.99 A 35 C II VE=28.0V02+O.5, rs.98 A 25 C II VE =26.4 V0 2+ 1.8, r a .99

/

O...... ---,r------r----,----r--- 1.0 2.0 30 4.0

FIGURE 39 Expiratory Minute Volume as a Function of Submaximal Oxygen Uptake. These are the regression lines determined from the individual data for cycling exercise less than 80% VOZ max. Slopes and intercepts are not statistically different for any two conditions. sao

37.5 . /_ / /0 . .--.~.~. _/ Tre , .~ /~- / °c .~<~~. ~7. / tf__~.~ III 2SoC AIR fJ. 2SoC WATER o 30°C A 3SoC

:!G.5 ~

• I i I ~v\;---...... ,--- o 1.0 2.0 3.0 40 5' poll mal.

V02 • L/min

FIGURE 40 Rectal Temperature as a Function of Exercise Intensity and Ambient Condition. These are mean values of 4 subjects taken at the end of 5-min of rest and graded cycling exercise. The point at the far right represents a post-maximal exercise measurement...... ~..... 142

Results of Maximal Exercise

Comparisons of the maxi~dl cardiorespiratory responses are shown in Table 15 and summarized in Fig. 41.

Maximum oxygen uptake. V0 2max was 0.59 L/min (13%) lower for cycling in air versus treadmill running. This is higher than the per­ centage difference reported by some (Astrand and Saltin, 1961; Hermansen and Sa1tin, 1969; Hermansen et al., 1970) but lower than others (Chase et al., 1966). Compared to cycling in air V0 2max was 2.5%, 3%, and 4.5% lower in 25 C, 30 C, and 35 C water, respectively. Most of the differ­ ence was due to subject DB whose V0 2max in water was on the average 11.5% lower than in air. Subjects JM and DM recorded the same or slightly higher average V02max in water than air. In general, V02max was about the same at all water temperatures although there was a tend­ ency for it to be lower in 35 C water. These data agree very closely with the findings reported in Chapter 6.

Heart rate. Peak fh was the same for cycling in air and cycling in 35 C water which supports previous findings. fh was also the same for 25 C and 30 C water. The difference in fh between 35 C and 25 C was 13 bpm and accounts for one-half of the difference in cardiac output. . Cardiac output. Q was 9% lower cycling compared to running which is about the same as reported by Hermansen et al. (1970). The lower Q was due to 5% lower values for both fh and Qs during maximal cycling. Qwas 1.2 L/min (5%) lower in 25 C water than in air. It was, however, 0.9 L/min (4%) and 2.8 L/min (11.5%) higher in 30 C and 35 C 143

TABLE 15 PERCENTAGE CHANGE IN MAXIMAL CARDIORESPIRATORY RESPONSE FOR VARIOUS CONDITIONS OF EXERCISE . . . Condition V02 max Q f h Qs (a-v)02 VE R

Running ~ eye ; ng ;:;:> -13%* -9% -5% -5% -5% -5% 3% alr Cycling in -3% 4% -8% 12% -7% -13% -4% water 30 C 1% -8% -1% -8% 10%* 9%* 0 25 C -2% 17%** 8%** 9%* -16%** 1% -1% 35 C

* p<.05 ** p<.005 144

160

155

150 Os 145 ml 140 :::~ 160

160 27

26 o 25 L/mln 24

23 4·6 4·4

4·3

4·2

Vo2mall .4·1 L/min 4·0 3·9 3·0

3·7

Running AIR

FIGURE 41 Cardiovascular Responses to Maximum Effort Running and Cycling. Cycling exercise was performed in 25 C air and in three different water temperatures. 145 water than in air. 0 was, therefore, directly proportional to water temperature, being 4.0 L/min (17%) higher in 35 C than 25 C water.

Stroke volume. Qs was about the same in 30 C and 35 C water. The difference of 1.9 L/min (7%) in 0 was, therefore, due to a lower fh of 12 bpm in 30 C water. Compared to maximal cycling in air Qs was ele­ vated 3.5%, 12%, and 13% in 25 C, 30 C, and 35 C water, respectively.

Qs fell 5-7 m1/beat from its value at 70% V0 2max during maximal exercise in 30 C and 25 C water, but it increased 8 and 9 m1/beat in 35 C water and air, respectively. The difference of 10 m1/beat in 30 C water versus running compares favorably with the findings of Dixon and Faulkner (1971).

Expiratory minute volume. VE was 12 L/min (9%) lower in 30 C than

25 C water even though V02max was the same. Ventilatory equivalents were about the same for running and cycling in 30 C water, and for cycling in air and 35 C water. Breathing efficiency in terms of the ventilatory equivalent was highest in 30 C water and lowest in 35 C water. R values were not different.

Rectal temperature. Tre rose 0.2, 0.3, and 0.4 C during the 5 min of maximal cycling in 25, 30, and 35 C water, respectively. The change in Tre was similar in air and 35 C water. Tre continued to rise 0.2-0.3 C during the 5 min post-exercise period. Tre was signifi­ cantly greater in 35 C compared to 30 C water. Sweating was observed at all three water temperatures, however, it was minimal in 25 C. Un­ fortunately no attempt was made to measure the amount of sweat that collected in the feet of the plastic suit. 146

Discussion

The impedance technique for estimating cardiac output gave not only reproducible values but also produced results which compare fav­ orably in absolute terms with those in the literature. Since the method is non-traumatic, resting values for fh and Qs are more likely to be undisturbed. The low resting fh may, therefore, explain in part the high resting values for Qs in the two air series. The higher initial Qs might account for its smaller change in the transition from rest to exercise. The apneic maneuver apparently did not affect O. This suggests that breath holding without Val salva for only 5 sec post- exercise does not evoke a depression of left-ventricular performance in man. The value for peak Qhas been found to be highly correlated with

V02max (Rowell, 1974). The predicted value of Qwas, therefore, cal­ culated from regression equations derived from data obtained on athletes . using the dye-dilution method. For upright cycling exercise Astrand et al. (1964) found, 0 = 4.28 V0 2max + 6.9, (r = 0.87). Using this equation for the present subjects· V0 2max, the predicted value of Qis 23.5 Ljmin compared to the observed value of 24.3, or about 3% lower.

Ekblom and Hermansen (1968) found, Q = 8.14 V02max - 9.1, (r = 0.98) for treadmill running. The predicted value of 0 is 27.3 Ljmin compared to the observed value of 26.8, or about 2% higher. It therefore seems reasonable to conclude that the peak 0 values obtained using the im­ pedance method are also consistent with those in the literature.

The 13% higher V02max during running compared to cycling was probably due to the subjects' greater emphasis on running training. 147

D.M. for example is a national caliber distance runner who does very little cycling. His treadmill V02max was 25% higher than on the ergometer. The higher peak fh and Qs in running were equally respon­ sible for the higher ~ of 2.5 L/min. Hermansen et al. (1970), however, found the 6% reduction in cycling V0 2max was due to a 6% reduction in Qmax which in turn resulted from a 5% decrease in stroke volume, since fh was about the same. Their finding has contributed to the popular belief today that stroke volume is the major determinant of V02max. On the other hand, Miyamura and Honda (1972) found no difference in Qs, but v02max for cycling was still 10% lower than for running due to 5% reductions in fh and (a-v)02. Thus, the present data agree in part with both studies. The lower peak fh in cycling versus running is particularly interesting because it suggests that the capacity of the heart was not reached before leg fatigue developed. The cardiodynamic aspects of maximal cycling in the various water temperatures are perhaps the most exciting data in this study. Although

V0 2max was not different, Qwas highly dependent on water temperature. The increased thermal demand for blood flow in 35 C water was apparently responsible for the 4.0 L/min increase in Qcompared to 25 C water. The distribution of the elevated Qwas probably to the skin or other organs with low oxygen extraction since the calculated (a-v)02 fell in propor­ tion to the increase in Q. It is unlikely the splanchnic or renal blood flows were increased because it has been demonstrated by Rowell et al. (1964) and Grimby (1965) that these regions are actively vasoconstricted during maximal exercise. It is also possible that part of the greater Qmay have perfused inactive muscle or non-nutritive channels in active 148

muscle (Rowell et aZ' 3 1970). The value of 4 L/min is consistent with the highest estimate of cutaneous blood flow during maximal exercise predicted by Rowell (1974). Since both fh and Qs increased during maximal exercise in 35 C compared to 25 C water, the central circulation was probably not limited in the colder water. There may have been a decrease in after­ load in 35 C water due to peripheral vasodilation and cutaneous venous pooling which would have enhanced the permissive level of the cardiac output. In addition body temperature, or more specifically heart tem­ perature, may also affect the size of Qs independent of the inotropic effects of tachycardia since Qs was greater in 30 C compared to 25 C water. This greater myocardial contractility could be a result of in­ creased sympathetic stimulation and/or local temperature. In any case the value of Q per se does not limit ~02max in water. Furthermore, owas apparently not "maximal" during cycling in air since it was higher in both 30 C and 35 C water. In addition Q was slightly higher in 35 C than treadmill running although V0 2max was 17% lower. These findings suggest that V02max is limited by the type of exercise, the muscle mass involved, and the metabolic capacity of the active muscles rather than by the pumping capacity of the heart.

If the Qvs. V02 and fh vs. V02 relationships are truly linear, Qs vs. .V02 must. be hyperbolic (Fig. 38). However, in a few individual cases, Q vs. V02 was observed to be more hyperbolic in shape than linear. Nevertheless, in submaximal cycling. exercise in water there was a tend- ency for Qs to increase with V02 which suggests that adrenergic stimula- tion may decrease end-systolic volume even when the heart is preloaded 149 by water immersion. This is consistent with the increase in Os sometimes observed during supine exercise (Bevegard et aZ., 1960). Peak ~E was 13% lower in 30 C water than in air which agrees with . previous findings reported in Chapter 4. VE was higher in the two extreme water temperatures and only 5% lower than in air. The equal value for ~E in 25 C as in 35 C disagrees with previous findings (Chapter 6) and suggests that lactate formation may have been the same. R values indicate the degree of hyperventilation was not changed by water immersion or water temperature.

In summary, V02max was ahout 3% lower in water than air, but this was not significant statistically. It was not altered by changing water temperature from 25 C to 35 C. Exercise in 25 C and 30 C water was accompanied by a lower heart rate and a larger stroke volume at a given V02 than in air. Peak heart rate was reduced 12-14 bpm in 25 and 30 C water compared to air or 35 C water. This suggests that the skin may have been vasoconstricted in 30 C water even during exhausting exercise. Cardiac output increased with water temperature which was due primarily to a greater heart rate. Maximum oxygen uptake was not related to peak cardiac output during maximal cycling exercise in water. The thermal demand to increase skin blood flow may have been met by in­ creasing cardiac output rather than by redistribution. The permissive level of maximal cardiac output was presumably enhanced by a better venous return in water. Skin and core temperatures probably have a multiplier effect on increasing cardiac output at various water temper­ atures. These findings suggest that maximum oxygen uptake is determined by metabolic limitations (diffusion, 02 utilization) and the size of the active muscle tnass rather than by circulatory factors. 150

CHAPTER 9 GENERAL DISCUSSION AND SUMMARY

The theory that maximum oxygen uptake, as determined from pulmonary gas exchange, indicates the maximal rate of aerobic metabolism was an underlying premise of this investigation. Since aerobiosis must depend on oxygen transport, V0 2max should also be a useful measure of the functional capacities of the respiratory and cardiovascular systems.

In the present study V02max was used to determine the effects of water immersion on the oxygen transporting capacity of the combined cardio­ respiratory system. It was argued that in order to get a true under­ standing of the effects of immersion per se on oxygen transport, the same exercise must be performed in water as in air. Studies which compare swimming versus running or cycling are forced to consider other differences such as body position, degree of skill required, and muscle mass utilized. These differences complicate an interpretation of the observed cardiorespiratory changes that occur with immersion.

Values in the literature for V02max while swimming range from 1 to 19 percent lower than for treadmill running and average about

7 percent lower. Compared to cycling in air, swimming v02max has been found to be 2 to 13 percent lower. It is apparent that different types of exercise may elicit a different "peak \/02." Therefore no definite explanation can be given with regard to immersion for the reason why swimming V02max is lower than that for exercise in air. The present investigation used the same upright cycling exercise in air and water. The use of a standard Monarch bicycle ergometer for 151 underwater exercise enabled this comparison. Subjects were immersed to the neck while cycling in water and were apparently able to engage the same muscles as used for cycling in air. When the friction belt was removed from the flywheel a convenient relationship was found between the oxygen cost of cycling in water and the frequency of ped­ aling (Fig. 5). This empirically derived relationship was used to set the desired relative work rates which were 20, 35, 50, 70 and 100%

V0 2max. It was theorized that the energy expenditure required when cycling underwater is due primarily to the drag forces of water which brake the movement of the flywheel. With the friction belt in place external work could be observed. It was found that the gross efficiency of underwater cycling declines exponentially when pedal frequency exceeds 30 rpm (Fig. 6). The buoyant force of water leads to a loss of body weight and this also impedes the performance of underwater work. The submaxima1 to maximal exercise responses of 5 trained male subjects were measured while cycling in 22-25 C air and 30 C water, and also during treadmill running. Of these seven, 4 subjects were also tested in 25 and 35 C water. Immersion time for experiments in water averaged 50 min. In total 26 tests were administered for cycling in air, 35 in water at various temperatures and 15 for treadmill running.

Heart rate. Exercise fh was consistently lower in 30 C water com­ pared to air. The regression lines relating fh to V02 for both conditions were parallel, but fh for a given V02 was 5-7 bpm lower in water (Figs. 9, 29). There was a tendency for the difference in fh to become even . greater at V02max. Lowering the water temperature to 25 C did not 152

significantly change fh with respect to its value in 30 C (Fig. 30), but raising it to 35 C resulted in fh values similar to those in air (Fig. 31). Peak fh during maximal exercise was 13-19 bpm lower in 25 C water compared to 35 C water or air (Figs. 22,41). These findings indicate that the heart rate response to exercise in water is sensitive to water temperature and changes in fh with immersion are not neces­ sarily due to immersion per se. The mechanism for the slowing of fh may be due to a baroreflex initiated by vasoconstriction in response to the lowering of mean skin temperature. In favor of this hypothesis was the finding that bradycardia occurred in 25 C and 30 C water, but not in 35 C water.

PlIlmona~y ventilation. Submaximal VE and fR at a given Q02 were unchanged by immersion and were also independent of water temperature (Figs. 11,39). The use of SCUBA, however, caused significant reductions in both submaximal and maximal VE and f R (Figs. 16,18). The lower VE was probably a consequence of the smaller dead space volume in the SCUBA mouthpiece because the estimated alveolar ventilation using SCUBA was not different from using the Collins valve (Fig. 19). Peak ~E during maximal exercise was consistently lower in 30 C water than in air (Fig. 15, Table 13). This reduction was attributed to the greater work of breathing during immersion. The finding that maximum voluntary ventilation was significantly reduced during immersion supports this conclusion (Table 6). Water temperature was also found to affect peak VE being 19 Ljmin lower in 25 C compared to 35 C without a change in

V02max (Fig. 22). The mechanism for this thermal effect is unknown, 153

but it probably relates to changes which occur in the central circula­ tion during hyperthermia.

Maximum oxygen uptake. The mean V02max for cycling was 3.29 Llmin in air and 3.18 Llmin in 30 C water. The 3% lower value in water was not significantly different (p>.05). Peak values for fh and VE were, however, significantly lower in 30 C water by 10 bpm and 16 Llmin, respectively. The V02max for treadmill running was 3.67 Llmin or 10% higher than for cycling in air. Peak fh and VE were also significantly higher for running. ~02max was independent of water temperature (Figs. 22,41) although fh and VE were both significantly changed as previously discussed. These findings provide strong evidence that oxygen transport during maximum exercise is not reduced by water immersion. Increases in stroke volume and the amount of oxygen extracted at the lung compensated for the observed reductions in fh and VEe The lower V02max for swimming which has been reported by investigators comparing swimming to exercise in air is probably not the result of immersion per se, but rather due to differences in body position and the muscles utilized.

Cardiac stroke volume. Qs was estimated using the non-invasive thoracic impedance technique. Relative changes observed in Qs with postural shifts and water immersion were consistent with values reported by others using direct methods. Compared to sitting in air Qs increased 33% when the subjects sat in 30 C water immersed to the neck. The mag- nitude of Qs during immersion was the same as for the supine position (Table 11). The increase in Qs due to immersion was attributed to a greater diastolic filling of the heart as suggested by Arborelius 154 et aZ. (1972). During submaximal exercise Qs was about 10-25% higher in water than air. Submaximal Qs was not statistically different in the 3 water temperatures. During maximal exercise Qs was 12% higher in 30 and 35 C water than in air, and only 4% higher in 25 C water. Qs during maximal running was 7 ml/beat (5%) higher than for maximal cycling in air, but about 10 ml/beat (7%) lower than for maximal cycling in 30 and 35 C water (Fig. 37).

Cardiac output. The values obtained for Qat a given V0 2 using the impedance technique agree in absolute terms with the findings reported by others using direct methods (Figs. 32,33; Table 14). Qwas linearly related to V0 2 for all conditions. Although the slope of the Qvs. V02 regression line remained the same, the intercept progressively increased with water temperature. Qtherefore increased with Tw at all levels of V0 2. Compared to cycling in air 6 was the same throughout exercise in 25 C water (Fig. 34). In 30 and 35 C water, however, Qwas elevated at a given V0 with respect to that in air (Figs. 35,36). During maximal 2 . exercise in 35 C water Q was 2.8 L/min higher than in air although

V0 2max was 0.18 L/min (5%) lower. Although fh was 10 bpm slower and V02max was 0.77 L/min (17%) lower, peak Qin 35 C water was slightly larger than for running. Organs with low 02 extraction such as. skin, therefore, must have received a larger fraction of the total Q in 35 C water. This finding is contrary to the belief that Qis the major factor limiting V0 2max. It would appear from these results that either the pumping capacity of the heart is greater than the ability of the active muscles to receive it, or the extraction of oxygen at the cellular 155

level is the rate limiting factor determining V02max. In conclusion the present investigation attempted to isolate the effect of immersion on maximum oxygen uptake by comparing the cardio­ respiratory responses. of subjects performing the same cycling exercise in air and water. V0 2max was not affected by immersion in 25, 30, and 35 C water for periods up to one hour. However, differences did occur in cardiac and ventilatory parameters. Significant decreases in heart rate and expiratory minute volume during maximal exercise in 25 and 30 C water were compensated by increases in cardiac stroke volume and the fraction of oxygen removed from the inspired air. Exercise in 35 C water was associated with an elevated cardiac output at all levels of oxygen uptake. 156 Appendix A

Characteristics of Subjects

David Baker Age: 24-26 Height: 180 cm Weight: 73.3 - 77.5 kg % Fat: 14 - 21 Vital capacity: 6.1 L, BTPS Residual volume: 1.8 L, BTPS Occupation: Graduate student in physiology Physical activity: Endurance-type training consisting of 2-4 hours of swimming and 2-3 hours of jogging per week. Occasionally competed in long distance swimming. Tests of maximum oxygen uptake: Ljmin, STPD

Bicycle ergometer, N=6; range 3.67 - 3.83, mean 3.76 ± .03 Treadmill running, N=3; 4.22, 4.46, 4.49 Underwater cycling, N=8; range 3.30 - 3.59, mean 3.40 ± .04

Rudy Dressendorfer Age: 29-31 Height: 178 cm Weight: 68.2 - 71.9 kg %Fat: 6 - 9 Vital capacity: 6.8 L, BTPS Residual volume: 1.4 L, BTPS Occupation: Graduate student in physiology Physical activity: Competitive distance runner, 6-8 hours of running per week, plus regular racing. Tests of maximum oxygen uptake: Ljmin, STPD

Bicycle ergometer, N=5; range 4.15 - 4.43, mean 4.32 ± .05 Treadmill running, N=3; 4.76, 4.80, 4.89 Underwater cycling, N=9; range 4.07 - 4.42, mean 4.22 ± .04 157

Duncan Macdonald Age: 25 Height: 181 cm Weight: 63.3 - 65.9 kg %Fat: 7 Vital capacity: 6.0 L, BTPS Residual volume: 1.3 L, BTPS Occupation: First year medical student Physical activity: Competitive distance runner, 2-6 hours of running per week. Achilles heel injury prevented regular racing. In the previous year ranked 7th best American miler by Track and Field News. Tests of maximum oxygen uptake: L/min STPD Bicycle ergometer, N=2; 3.67, 3.65 L/min Treadmill running, N=2; 4.59, 4.53 L/min Underwater cycling, N=3; 3.59, 3.69, 3.70 L/min

James Morlock Age: 26-28 Height: 184 cm Weight: 67.7 - 70.7 kg % Fat: 10 - 11 Vital capacity: 5.5 L, BTPS Residual volume: 1.4 L, BTPS Occupation: Graduate student in physiology Physical activity: Endurarce-type training consisting of 1-2 hours of bicycling, 0 to 1 hour of jogging and 1/2 ­ 1 hour of swimming weekly. Recreational surfing 10-12 hours per week. Tests of maximum oxygen uptake: L/min, STPD

Bicycle ergometer, N=7; range 3.66 - 3.75, mean 3.71 ± .01 Treadmill running, N=2; 3.89, 3.89 Underwater cycling, N=9; range 3.54 - 3.78, mean 3.71 ± .03 158

Benjamen Respicio Age: 29 Height: 170 cm Weight: 65.4 - 67.0 kg % Fat: 16 Vital capacity: 4.1 L, BTPS Occupation: Machinist for Department of Physiology Physical activity: 1/2-1 hour jogging per week, occasional SCUBA diving. Tests of maximum oxygen uptake: L/min, STPD Bicycle ergometer, N=2; 2.25, 2.21 Treadmill walking, N=2; 2.47, 2.51 Underwater cycling, N=2; 2.05, 2.03

Richard Smith Age: 29 Height: 172 cm Weight: 64.1 - 66.2 kg % Fat: 12 Vital capacity: 6.5 L, BTPS Occupation: Assistant Professor of Physiology Physical activity: Occasional swimming and hiking for recreation on weekends. Tests of maximum oxygen uptake: L/min, STPD Bicycle ergometer, N=2; 2.60, 2.63 Treadmill walking, N=2; 2.67, 2.73 Underwater cycling, N=2; 2.32, 2.29 159

Louis Wong Age: 24 Height: 164 cm Weight: 57.2 - 58.3 kg % Fat: 13 Vital capacity: 4.2 L, BTPS Occupation: 2nd year medical student Physical activity: 1-3 hours of recreational tennis per week. Occasional bicycling Tests of maximum oxygen uptake: L/min, STPD Bicycle ergometer, N=2; 2.63, 2.64 Treadmill walking, N=l; 2.59 Underwater cycling, N=2; 2.66, 2.60 160

Appendix B

Individual Data Comparing Resting and Exercise Responses in Air and 30 C Water

02 uptake, heart rate, and rectal temperature . T V02 . fh re bpm °C L/min, STPD %V0 2 max Subject Air Water Air Water Air Water Air Water RD .36 .24 8 5 48 43 37.05 36.95 JM .39 .34 9 9 56 46 36.8 36.95 BR .28 .27 12 13 74 68 36.8 36.65 RS .22 .37 8 16 69 72 36.95 36.95 LW .30 .26 11 10 74 68 36.95 36.8 RD .81 1.01 18 23 65 63 36.8 36.5 JM .82 .71 22 19 76 60 36.8 36.8 BR .52 .42 23 20 83 77 36.9 36.6 RS .65 .56 24 24 82 80 36.95 36.95 LW .55 .61 21 23 86 82 36.75 36.45 RD 1. 32 1.51 30 34 78 80 36.9 36.45 JM 1.23 1.58 33 42 86 90 36.8 36.65 BR .75 .77 33 37 93 94 36.95 36.55 RS 1.07 1.02 41 44 96 95 37.0 36.95 LW 1.05 .93 40 35 103 90 36.8 36.6 RD 1.99 2.38 46 54 96 102 36.95 36.65 JM 1. 74 1. 78 47 48 97 100 37.2 36.80 BR 1.06 1.01 47 50 110 104 36.85 36.65 RS 1.34 1.28 51 55 106 106 37.1 37.0 LW 1.44 1.44 55 54 120 113 36.85 36.8 RD 3.31 3.18 75 72 138 121 37.2 36.95 JM 2.80 2.51 75 67 134 113 37.4 37.05 BR 1.47 1.39 65 68 138 125 37.15 36.7 RS 1.90 1.69 73 73 140 120 37.3 37.0 LW 1.84 1.86 70 70 143 140 37.15 36.85 DB 3.81 3.35 100 100 181 152 37.8 37.3 RD 4.38 4.42 100 100 171 164 38.0 37.4 JM 3.74 3.75 100 100 172 162 37.75 37.45 BR 2.25 2.05 100 100 169 163 37.45 37.05 RS 2.60 2.32 100 100 171 163 37.8 37.1 LW 2.63 2.66 100 100 175 173 37.3 37.15 161

Appendix B (Continued)

Individual Data Comparing Resting and Exercise Responses in Air and 30 C Water

C02 production, gas exchange ratio expiratory minute volume, and breathing frequency . . VC02 VE fR L/min, STPD R L/min, BTPS br/min Subject Air Water Air Water Air Water Air Water RD .25 .17 .71 .71 10.8 9.9 9.5 7.5 JM .33 .30 .85 .88 14.4 13.5 15 13.5 BR .25 .23 .89 .86 12.3 10.2 12 12 RS .20 .34 .91 .92 10.1 11.2 12.5 8 LW .23 .22 .77 .85 12.6 10.5 11.5 15.5 RD .64 .68 .79 .68 23.2 25.2 16 15 JM .67 .58 .82 .82 23.7 23.2 18 20 BR .44 .32 .85 .75 19.2 14.7 18.5 17.5 RS .48 .38 .77 .69 19.0 16.4 18.5 9 LW .45 .47 .81 .77 19.4 19.8 20 21.5 RD 1.01 1.06 .77 .70 32.8 36.6 16.5 16.5 JM 1.00 1.32 .81 .84 33.3 43.5 19 24 BR .63 .63 .83 .82 26.1 27.3 16 26.5 RS .91 .72 .85 .71 32.6 21.9 22.5 8 LW .86 .72 .82 .77 31.1 27.8 22 25 RD 1.63 1.87 .82 .78 48.0 60.6 19.5 22 JM 1.49 1.60 .86 .90 49.6 47.6 22 21 BR .99 .86 .93 .85 34.7 36.3 22.5 30.5 RS 1.11 1.00 .83 .78 38.6 32.6 23 13.5 LW 1.25 1.22 .87 .85 42.8 51.3 26.5 37 RD 3.18 2.80 .96 .88 90.3 84.3 27 28.5 JM 2.66 2.36 .95 .94 79.2 71.5 21 28 BR 1.47 1.21 1.00 .87 52.7 50.9 30 28 RS 1.67 1. 51 .88 .89 59.6 48.2 27 20.5 LW 1.67 1.66 .91 .89 56.4 67.7 31.5 41.5 DB 4.27 3.45 1.12 1.03 190.7 167.5 63 86 RD 4.66 4.47 1.06 1.01 209.1 193.2 56 50 JM 4.30 3.79 1.15 1.01 154.4 131 .5 52 43.5 BR 2.70 2.15 1.20 1.05 108.9 94.5 47 62 RS 2.91 2.30 1.12 .99 113.6 95.9 39 37 LW 2.63 2.61 1.00 .98 92.8 103.4 48 57 162 Appendix C

Individual Data Comparing the Use of Collins Valve (3-J) or SCUBA During Rest and Exercise in 30 C Water

02 uptake, heartrate, and rectal temperature . T V0 2 %V02 milX fh re L/min STPD bpm °C Subject 3-J SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA RD .22 .26 5 6 45 41 37.4 37.0 JM .29 .21 8 6 39 39 37.2 37.2 BR .24 .22 12 11 62 60 37. 1 37.1 RS .31 .34 13 15 72 73 37. 1 37.1 LW .23 .22 9 8 64 60 37.1 36.8 RD .38 .37 9 9 43 46 36.7 36.9 JM .35 .26 9 7 40 40 37.0 36.9 BR .30 .35 15 17 66 63 37.0 36.9 RS .51 .49 22 21 77 76 36.0 36.0 LW .47 .44 18 16 72 68 36.3 36.5 RD .78 .76 18 18 66 62 36.5 36.4 JM .94 .67 26 18 70 63 36.7 36.8 BR .66 .64 32 32 88 86 36.7 36.7 RS 1.10 .94 44 41 90 86 37.1 37.1 LW .72 .65 27 24 84 81 35.8 36.0 RD 1.84 1.86 42 44 92 90 36.4 36.4 JM 2.15 2.01 58 53 113 112 36.9 36.9 BR 1.48 1.50 71 74 118 116 36.6 36.6 RS 1.69 1.69 73 74 119 116 37.3 37.2 LW 1.83 1. 77 69 64 136 127 35.85 35.7 DB 3.35 3.32 100 100 152 149 37.3 37.0 RD 4.42 4.21 100 100 164 153 37.4 37.0 JM 3.75 3.78 100 100 162 162 37.45 37.15 BR 2.05 2.03 100 100 163 157 37.05 36.8 RS 2.32 2.29 100 100 163 160 37.1 37.4 LW 2.66 2.60 100 100 173 172 37.15 37.3 Appendix C (Continued) Individual Data Comparing the Use of Collins Valve (3-J) or SCUBA During Rest and Exercise in 30 C Water C02 production, gas exchange ratio, expiratory minute volume, breathing frequency, and tidal volume . . VC02 VE fR VT Ljmin STPD R Ljmin BTPS brjmin Ljbr Subject 3-J SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA RD .18 .21 .82 .81 8.7 6.4 4.5 5 1.9 1.3 JM .21 .15 .72 .72 8.3 6.4 6 3.5 1.4 1.8 BR .21 .22 .88 1.00 9.5 9.2 8 8 1.2 1.2 RS .30 .34 .95 1.00 16.0 21.3 13.5 6 1.6 2.7 LW .21 .17 .91 .79 9.9 7.0 14.5 12 0.7 0.6 RD .29 .29 .78 .79 10.4 8.7 6 6.5 1.7 1.3 JM .25 .18 .73 .71 10.9 6.5 3 10 1.1 0.6 BR .21 .26 .71 .75 8.9 9.3 8 6 1.1 1.6 RS .97 .86 .88 .92 29.2 21.0 9.5 5 3.1 4.2 LH .36 .33 .76 .76 16.0 13.3 20.5 23 0.8 0.6 RD .56 .55 .72 .75 20.7 16.5 13 8.5 1.6 1.9 JM .76 .50 .80 .75 28.8 14.3 20.5 9 1.4 1.6 BR .47 .47 .72 .73 16.8 14.6 13 9 1.3 1.6 RS 1.12 1.07 .93 .89 34.5 29.3 14 9.5 2.5 3.1 LW .54 .51 .75 .79 22.4 17.9 19 20.5 1.2 0.9

--' '"w Appendix C (Continued) Individual Data Comparing the Use of Collins Valve (3-J) or SCUBA During Rest and Exercise in 30 C Water C02 production, gas exchange ratio, expiratory minute volume, breathing frequency, and tidal volume . VC02 VE fR Vr L/min STPD R L/min BTPS brim; n L/br Subject - 3-5 SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA 3-J SCUBA RD 1.45 1.47 .81 .79 50.1 37.5 22 15 2.3 2.5 Jr~ 1.92 1.69 .90 .84 60.8 44.7 27 18.5 2.3 2.4 BR .90 .89 .82 .82 32.2 21.0 22 10.5 1.5 2.0 RS 1.59 1.55 .94 .92 52.5 43.7 21 13.5 2.5 3.2 LW 1. 76 1.63 .96 .92 70.5 58.7 42 34.5 1.7 1.7 DB 3.45 3.52 1.03 1.06 167.5 109.3 86 47 1.9 2.3 RD 4.47 4.61 1.01 1.10 193.2 162.1 50 44 3.9 3.7 JM 3.79 4.11 1.01 1.09 131.5 116.2 43.5 41.5 3.0 2.8 BR 2.15 2.21 1.05 1.09 94.5 54.7 62 22 1.5 2.5 RS 2.30 2.61 .99 1.14 95.9 97.5 37 23 2.6 4.2 LW 2.61 3.04 .98 1.17 103.4 103.6 57 62 1.8 1.7

--' en -+::> 165

\ 5.0..------, 190

180 c::: °e 170 .a...... 160

)( ::150 () E - 140 N 2.0 o 130 .> L:;;T~m--;;!:CA----:C:t-VI-~C':-:"wa Tm Cw '

Ci) ~ 170 co 'c; 150 c::: E 1'50 E ...... ~ ...... -: IIO ..;~~ ~~~~. .a )( ~~~-----* E 90 " w 70 ~ ~ x 50 T C C CWe m A w

FIG. 1 (Appendix C). CARDIORESPIRATORY REXPONSES TO VARIOUS CONDITIONS OF MAXIMAL EXERCISE. Individual values for 4 conditions of exercise are shown. Tm refers to treadmill walking. CA is cycling in air. Cw is cycling in 30 C water. And Cws is cycling in 30 C water breathing from SCUBA. The broken line connects the mean values. This figure and Fig. 21 in the text were prepared from the same set of data. Appendix 0 Individual Data Comparing Measurements Made During Rest and Exercise in Air, Warm Water (35 C) and Cold Water (25 C)

23.0 ± 0.2 C AIR . Subject R T La Pva pH % V0 2 max V0 2 VE fh fR re DB 10 .38 .84 15.5 69 13 37.1 7.0 .72 7.37 RD 7 .27 .78 10.0 51 8 37.3 6.5 .80 7.38 JM 9 .32 .75 11.2 52 11 37.3 5.5 .64 7.37 DB 56 2.06 .91 60.7 131 24 37.1 RD 55 2.28 .83 54.7 114 19 37.4 Jr~ 51 1.89 .83 59.5 117 20 37.3 DB 100 3.70 1.26 191.0 172 64 37.4(37.7) 69 3.8 7.20 RD 100 4.15 1.21 166.4 171 45 37.9{38.1) 103 1.5 7.16 JM 100 3.73 1.16 130.8 174 44 37.9{38.1) 90 3.8 7.15

25.0 ± 0.1 C WATER DB 10 .36 .83 13.9 66 16 37.0 9.2 .76 7.35 RD 8 .32 .73 8.9 42 9 37.1 8.0 .75 7.40 JM 7 .27 .69 9. 1 46 12 37.3 8.5 .61 7.36 DB 54 1.94 .99 62.9 115 26 36.8 RD 56 2.30 .67 55.9 99 19 36.8 JM 50 1.82 .81 47.5 93 21 37.3 DB 100 3.59 1.16 148.2 152 49 36.8{36.8) 81 2.5 7.17 RD 100 4.09 1.03 158.5 144 41 36.7{37.0) 82 2.5 7.21 J~1 100 3.67 1.08 124.9 164 49 37.2{37.4) 96 2.9 7.12

--' 0'1 0'1 Appendix 0 (Continued) Individual Data Comparing Measurements Made During Rest and Exercise in Air, Warm Water (35 C) and Cold Water (25 C)

35.6 ± 0.5 C WATER . . Subject f T La Pva pH %V0 2 max V0 2 R VE fh R re DB 10 .36 .81 13.9 80 17 36.9 7.2 .54 7.39 RD 7 .30 .69 8.6 44 9 37.1 10.5 .56 7.40 JM 7 .26 .86 10.4 46 12 37.1 6.5 .64 7.33 DB 57 2.01 .85 59.2 122 24 37.0 RD 53 2.17 .72 56.0 113 23 37.2 JM 55 1.93 .88 57.2 115 19 37.2 DB 100 3.51 1.18 160.9 177 63 37.3(37.6) 89 2.2 7.22 RD 100 4.12 1.18 189.6 165 49 37.7(38.0) 117 2.0 7.24 J~1 100 3.54 1.21 137.4 173 49 37.6(37.9) 104 2.8 7.18

.... 0'\ ...... 168

Appendix E Individual Data Comparing Cardiorespiratory Responses to Rest and Exercise for Cycling in Air, 25, 30, and 35 C Water, and Treadmill Running. SUBJECT: DB . . . V0 max R V ftl Q Q (a-v)02 V0 2 % 2 E Tre s RUNNING .29 6 .70 12.4 83 6.4 77 4.6 1.07 24 .67 31.6 90 11.9 132 9.0 2.39 53 .75 69.8 138 17.3 125 13.8 3.83 85 .87 127.8 176 22.6 128 16.9 4.49 100 1.01 173.8 190 24.1 127 18.6

CYCLING IN AIR .33 8 .88 13.8 36.9 81 6.5 80 5.0 1.09 29 .94 33.2 36.95 105 9.7 92 11.2 1.53 40 .95 48.8 37. 1 120 11.8 98 13.0 1.95 52 .97 62.8 37.25 133 13.8 104 14.1 2.67 71 1.02 84.8 37.35 160 18.4 115 14.5 3.83 100 1.18 176.4 37.7 182 23.7 130 16.1

CYCLING IN 25 C WATER .35 10 .89 16.3 37.25 70 8.1 116 4.3 .85 24 .81 25.1 37.1 86 10.9 127 7.8 1.43 41 .87 41.9 37.0 99 13.9 140 10.3 1.88 54 .93 55.6 37.0 116 14.7 127 12.8 2.65 77 .90 79.8 37.15 132 18.8 142 14.1 3.46 100 1.12 146.4 37.55 159 21.9 138 15.8

CYCLING IN 30 C WATER .30 9 .94 13.2 37.1 66 8.6 130 3.5 .70 21 .84 22.9 36.95 70 10.8 154 6.5 1.47 44 .94 44.7 36.9 102 13.5 133 10.9 1.83 55 .93 55.5 37.0 109 15.3 140 12.0 2.22 66 .94 67.5 37.15 116 17.2 148 12.9 3.36 100 1.07 123.9 37.40 152 24.1 159 13.9

CYCLING IN 35 C WATER .30 9 .81 12.8 37.35 65 8.4 129 3.6 .80 24 .79 25.5 37.2 91 12. 1 133 6.6 1. 31 39 .80 35.0 37.25 105 13.8 131 9.5 1.82 54 .87 51.8 37.35 128 16.0 125 11.4 2.21 66 .87 66.2 37.5 139 19.2 138 11.5 3.35 100 1.10 146.2 37.75 169 27.0 160 12.4 169

Appendix E (Continued)

SUBJECT: RD . . . . R T Q Q (a-v)02 ~ %V0 2 max VE re fh s RUNNING .36 7 .76 13.9 70 9.7 138 3.7 .89 18 .73 26.5 69 12.8 185 7.0 2.55 52 .83 67.0 116 21.4 184 11.9 3.43 70 .82 86.1 142 23.8 168 14.4 4.89 100 1.14 175.1 177 28.5 161 17.5

CYCLING IN AIR .32 7 .67 10.6 37.1 54 7.9 146 4.0 .87 20 .78 23.2 37.0 72 11.3 157 7.7 1.37 31 .87 36.2 37.05 89 14.2 160 9.6 2.37 54 .93 65.0 37.15 115 19.5 170 12.2 3.26 75 1.04 100.8 37.4 139 21.1 152 15.4 4.36 100 1.13 167.0 38.0 165 25.0 152 17.4

CYCLING IN 25 C WATER .36 9 .78 11.9 37.4 44 7.2 164 5.0 1.01 24 .79 27.5 37.1 70 10.3 148 9.8 1.68 40 .83 43.2 36.8 88 14.7 167 11.4 2.37 56 .89 63.2 36.6 102 17.5 172 13.5 2.65 63 .95 71.4 36.6 108 18.1 168 14.6 4.19 100 1.17 160.9 36.75 151 23.5 156 17.8

CYCLING IN 30 C WATER .24 6 .75 8.4 37.2 58 8.3 143 2.9 1.13 27 .79 30.8 37.05 86 12.7 148 8.9 1.67 39 .83 46.9 36.9 90 16.6 184 10. 1 1.92 45 .87 55.9 36.8 97 18. 1 187 10.6 2.59 61 .92 74.2 36.9 107 20.9 195 12.4 4.24 100 1.08 151 .7 37.2 157 25.7 164 16.5

CYCLING IN 35 C WATER .35 9 .77 11.6 37.4 58 9.8 169 3.6 .95 23 .80 25.4 37.4 81 14.0 172 6.8 1.61 39 .86 41.8 37.4 105 16.4 156 9.8 2.18 53 .88 58.1 37.5 117 18.5 158 11.8 2.69 66 .93 81.5 37.65 132 21.2 161 12.7 4.07 100 1.12 166.7 38.0 163 26.2 161 15.5 170

Appendix E (Continued)

SUBJECT: DM . . . . V0 max R Tre fh Q Q (a-v)02 ~ % 2 VE s RUNNING .41 9 .83 17.9 45 6.0 133 6.8 1.01 22 .85 33.8 80 12.6 158 8.0 1.30 28 .93 38.7 82 13.1 160 9.9 2.49 54 .94 72.5 125 20.7 165 12.0 3.09 67 .96 78.7 152 25.3 166 12.2 4.59 100 1.20 149.1 178 30.5 171 15.1

CYCLING IN AIR .44 12 .97 25.8 37.05 48 7.0 146 6.2 .82 24 .93 34.5 36.9 67 9.3 139 8.8 1.24 34 .91 48.9 37.0 87 11.9 137 10.4 1.88 51 .79 43.5 37.2 107 14.6 136 12.9 2.77 75 .99 81.5 37.35 134 17.4 130 15.9 3.67 100 1.08 125.3 37.7 164 24.2 148 15.2

CYCLING IN 25 C WATER .31 8 .75 11.6 37.0 36 6.9 192 4.4 .86 23 .70 23.2 36.85 56 8.2 146 10.5 1. 35 36 .76 32. 1 36.7 85 13.4 158 10. 1 2.01 54 .77 47.0 36.65 98 15.3 156 13. 1 2.77 75 .92 79.7 36.65 119 18.4 155 15.0 3.70 100 1.04 135.4 36.95 155 23.1 149 16.0

CYCLING IN 30 C WATER .37 10 .96 17.9 37.1 46 7.6 191 4.9 .86 23 .73 23.1 36.95 58 10.1 174 8.5 1.71 46 .79 43.0 36.85 92 15.6 172 11.0 1.89 51 .84 45.9 36.9 100 17.3 173 10.9 2.51 68 .99 61.8 37.05 119 20.5 172 12.2 3.69 100 1.09 125.7 37.35 155 26.6 172 13.9

CYCLING IN 35 C WATER .29 8 .85 12.3 36.8 47 7.5 159 3.8 .68 19 .84 20.3 36.9 69 10.2 141 6.7 1. 31 36 .77 27.2 37.0 91 13.8 152 9.5 2.02 56 .96 54.7 37.2 119 18.8 158 10.7 2.35 62 .94 59.2 37.5 133 21.8 164 10.3 3.59 100 1.06 129.1 37.9 168 27.7 165 13.0 171

Appendix E (Continued)

SUBJECT: JM . . . % V0 max R V T fh Q Qs (a-v)02 V0 2 2 E re RUNNING .31 8 .88 12.6 45 6.2 139 5.0 .82 21 .94 27. 1 72 9.1 i26 9. 1 2.07 53 .97 61.8 115 15.4 134 13.4 3.18 82 1.06 107.5 150 19.8 132 16.0 3.89 100 1.10 132.6 175 24.1 137 16.1

CYCLING IN AIR .31 9 .83 13. 1 37.0 42 5.8 137 5.4 .81 22 .86 21.8 37.0 61 8.5 139 9.5 1.33 36 .93 38.4 37.05 80 11.1 138 12.0 2.06 56 .99 61.9 37.2 119 16.8 141 12.3 3.05 83 1.04 100.0 37.5 154 21.0 136 14.5 3.66 100 1.19 140.5 37.9 174 24.2 139 15.1

CYCLING IN 25 C WATER .38 10 .75 12.5 37.2 45 6.2 137 6.2 1.09 29 .81 30.7 37.0 73 10.7 147 10.2 1.56 41 .82 38.9 37.0 90 14.0 156 11.1 2.13 57 .89 54.4 37.0 104 15.4 148 13.8 2.60 69 .95 70.1 37.15 122 18.3 150 14.2 3.76 100 1.08 134.8 37.35 164 23.8 145 15.8

CYCLING IN 30 C WATER .24 6 .91 9.0 37.25 38 5.3 147 5.0 .54 15 .85 16.7 37.1 50 7.3 147 7.4 1.36 36 .85 36.1 37.0 74 10.5 142 12.9 2.30 62 1.05 63.5 37.15 110 15.9 145 14.5 3.02 81 1.19 96.7 37.35 143 20.0 140 15. 1 3.73 100 1.15 128.4 37.7 170 24.3 143 15.3

CYCLING IN 35 C WATER .34 9 .86 12.0 37.45 65 7.6 117 4.5 .98 26 .76 28.8 37.45 69 9.8 142 10.1 1.52 40 .82 40.0 37.6 102 14.5 142 10.5 2.14 57 .87 59.6 37.7 129 17.2 133 12.4 2.92 77 .94 91.4 37.85 154 22.3 145 13. 1 3.77 100 1.07 140.8 38.3 178 27.3 154 13.8 172

BIBLIOGRAPHY

Agostoni, E., G. Gurtner, G. Torri, and H. Rahn (1966). Respiratory mechanics during submersion ahd negative pressure breathing. ~. ~. Physio1. 21: 251-258. Arbore1ius, M., Jr., U. I. Balldin, B. Lilja, and C. E. G. Lundgren (1972). Hemodynamic changes in man during immersion with the head above water. Aerospace Med. 43: 592-598. Astrand, I. (1960). Aerobic work capacity )n men and women with special reference to age. Acta Physio1 Scand 49, Wuppl. 169: 11-92. Astrand, P.-O. (1952). Experimental Studies of Physical Working Capacity in Relation to Sex and Age. Copenhagen: Munksgaard. (1956). Human fitness with special reference to sex and age. Physiol. Revs. 36: 307-335. (1960). Work tests with the bicycle ergometer. Monark-Crescent AB. Varberg, Sweden. ______~~~ and B. Saltin (1961). Maximal oxygen uptake and heart rate in various types of muscular activity. ~.~. Physiol. 16: 977-981. ______~~ ' T. E. Cuddy, B. Saltin, and J. Stenberg (1964). Cardiac output during submaximal and maximal work. ~.~. Physiol. 19: 268-274. ___~~ , B. Ekblom, R. Messin, B. Saltin, and J. Stenberg (1965). Intra-arterial blood pressure during exercise with different muscle groups. ~.~. Physiol. 20: 253-256. _~~--;-;--.-- and K. ROGdhl (1970). Textbook of Work Physiology. New York: McGraw-Hill. Barcroft, H. (1934). Lectures in the Architecture of Physiological Function. New York, Cambridge University Press, p. 368. Bartlett, H. L., J. L. Hodgson, and J. Kollias (1972). Effect of respiratory valve deadspace on pulmonary ventilation at rest and during exercise. Med. Sci. Sports 4: 132-137. Bevegard, S., A. Holmgren, and B. Jonsson (1960). The effect of body position on the circulation at rest and during exercise, with special reference to the influence on stroke volume. Acta Physiol. Scand. 49: 279-298. --...- 173

_____~~~' U. Freyschuss, and T. Strandell (1966). Circulatory adaptation to arm and leg exercise in supine and sitting position. ~.~. Physiol. 21: 37-46, 1966. ______~~_- and J. T. Shepherd (1967). Regulation of the circulation during exercise in man. Physiol. Revs. 47: 178-213. Braunwald, E. (1974). Regulation of the circulation. New~. J. Med. 290: 1124-1129. Brown, A. C. and G. L. Brengelman (1970). The interactions of peripheral and central inputs in the temperature regulation system. In: Physiological and Behavioral Tem erature Regulation, J. D. Hardy, A. P. Gagge, and J. A. J. Stolwijk eds.). Springfield; Charles C. Thomas, pp. 684-702. Brozek, J., F. Grande, J. T. Anderson, and A. Keys (1963). Densito­ metric analysis of body composition: revision of some quantitative assumptions. Ann. rr.Y. Acad. Sci. 101: 113-140. Bruce, R. A., F. Kusumi, and D. Hosmer (1973). Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am. Heart J. 85: 546-562. Carlson, L. D. and A. C. L. Hsich (1971). Control of Energy Exchange. N.Y.: MacMillan, p. 44. Cerney, F. and W. Reddan (1973). Effects of water immersion on pulmonary gas exchange at rest and during exercise. Sci. Med. Sports (abstract), 5: 54-55. Cerretelli, P., R. S. Sikand, and L. E. Farhi (1969). Effect of increased air-way resistance on ventilation and gas exchange during exercise. ~.~. Physiol. 27: 597-600. Chase, G. A., C. Grave, and L. B. Rowell (1966). Independence of changes in functional and performance capacities attending prolonged bed rest. Aerospace Med. 37: 1232-1238. Christianson, R. A. (1971). Regulator performance test results: breathing resistance. Under Sea Industries, Inc. Engineering Bulletin No. 701. Cissik, J. H., R. E. Johnson, and B. E. Hertig (1972). Production of gaseous nitrogen during human steady-state exercise. Aerospace Med. 43: 1245-1250, 1972. Consolazio, C. F., R. E. Johnson, L. J. Pecora (1963). Physiological Measurements of Metabolic Functions in man. New York: McGraw-Hill. 174

Costnl, D. L. (1966). Effects of water temperature on aerobic working capacity. Res. Quart. 39: 67-73. ____~--~--, P. J. Cahill, and D. Eddy (1967). Metabolic responses to submaxima1 exercise in three water temperatures. ~.~. Physio1. 22: 628-632. Craig, A. B., Jr., and M. Dvorak (1969). Comparison of exercise in air and in water of different temperatures. Med. Sci. Sports 1: 124-130. Craig, F. N., W. V. Blevins, and E. G. Cummings (1970). Exhausting work limited by external resistance and inhalation of carbon dioxide. ~.~. Physio1. 29: 847-851. Daily, J. W. and D. R. F. Harleman (1966). Fluid Dynamics Addison-Wesley Publishing Co., Inc.; Reading, Mass. pp. 170-171. Daniels, J. (1971). Portable respiratory gas collection equipment. ~. ~. Physio1. 31: 164-167. Davies, C. T. M., P. E. DiPrampero, and P. Cerrete11i (1972). Kinetics of cardiac output and respiratory gas exchange during exercise and recovery. ~.~. Physio1. 32: 618-625. Demedts, M. and N. R. Anthonisen (1973). Effects of increased external airway resistance during steady-state exercise. ~. ~. Physiol. 35: 361-366. Denison, D. M., P. D. Wagner, G. L. Kingaby, and J. B. West (1972). Cardiorespiratory responses to exercise in air and underwater. ~. ~. Physiol. 33: 426-430. Dixon, R. W. and J. A. Faulkner (1971). Cardiac outputs during maximum effort running and swimming. ~.~. Physio1. 30: 653-656. Dressendorfer, R. (1974). Body composition maximum oxygen uptake, and endurance performance during pregnancy: case study. In prep. Efroymson, M. A. (1960). Multiple regression analysis. In: Mathematical Methods for Di~ita1 Computers, edited by A. Ralston and H. S. Wilf. N.Y.: Wiley. Ekb1om, B. (1969). Effect of physical training on oxygen transport system in man. Acta. Physio1. Scand. Supp1. 328. and L. Hermansen (1968). Cardiac output in athletes. ---:;---;;- ~. ~. Physiol. 25: 619-625. 175

Faulkner, J. A. (1968). Physiology of swimming and diving. In: Exercise Physiology, edited by H. B. Falls. N.Y.: Academic Press, 1968. ______~~------' D. E. Roberts, R. L. Elk, and J. Conway (1971). Cardiovascular responses to submaximum and maximum effort cycling and running. ~.~. Physiol. 30: 457-461. Flook, V. and G. R. Kelman (1973). Submaximal exercise with increased inspiratory resistance to breathing. J. ~_. Physiol. 35: 379- 384. - Froelicher, V. F. Jr., H. Brammell, G. Davis, I. Noguera, A. Stewart, M. C. Lancaster (1974). A comparison of three maximal treadmill exercise protocols. ~.~. Physiol. 36: 720-725. Gilbert, R., J. H. Auchincloss, Jr., and G. H. Baule (1967). Metabolic and circulatory adjustments to unsteady-state exercise. ~. ~. Physiol. 22: 905-912. Gollnick, P. D., R. B. Armstrong, C. W. Saubert, IV, K. Piehl, and B. Saltin (1972). Enzyme activity and fiber composition in skeletal muscle of untrained and trainen men. ~.~. Physiol. 33: 312-319. Grimby, G. (1962). Exercise in man during pyrogen-induced fever. Scand. ~. Lab. Clin. Invest. 14, Supple 67: 1-112. (1965). Renal clearances during prolonged supine exercise -----,..-~ at different loads. ~.~. Physiol. 20: 1294-1298. Guyatt, A. R., F. Newman, F. F. Cinkotai, J. I. Palmer, and M. L. Thompson (1965). Pulmonary diffusing capacity in man during immersion in water. ~.~. Physiol. 20: 878-881. Guyton, A. C. (1968). Regulation of cardiac output. Anesthesiology 29: 314-326. Haldane, J. S. (1912). Methods of Air Analysis. London: Griffin. Harley, A. and J. C. Greenfield, Jr. (1968). Determination of cardiac output in man by means of impedance plethymography. Aerospace Med. 39: 248-252. . Hartley, L. H., G. Grimby, A. Ki1bom, N. J. Nilsson, I. Astrand, J. Ekblom and B. Saltin (1969). Physical training in sedentary middle-aged and older men. III. Cardiac output and gas exchange at submaxima1 and maximal exercise. Scand. J. Clin. Lab. Invest. 24: 335-344. - -- -- 176

Henshe1, A., H. L. Taylor, and A. Keys (1954). Performance capacity in acute starvation with hard work. ~'~' Physiol. 6: 624-633. Hermansen, L. and B. Saltin (1969). Oxygen uptake during maximal treadmill and bicycle exercise. ~'~' Physio1. 26: 31-37. ______~~---, B. Ekblom, and B. Saltin (1970). Cardiac output during submaximal and maximal treadmill and bicycle exercise. ~' ~' Physiol. 29: 82-86. Hill, A. V., C. N. H. Long, and H. Lupton (1924). Muscular exercise, lactic acid, and the supply and utilization of oxygen. VII and VIII. Muscular exercise and oxygen uptake. Proc. Roy. Soc. London. Sere B. 97: 155-176. Hoes, M. J. A. J. M., R. A. Binkhorst, A. E. M. C. Smeekes-Kuy1, and A. C. A. Vissers (1968). Measurement of forces exerted on pedal and crank during work on a bicycle ergometer at different loads. Int. I. Angew. Physio1. 26: 33-42. Holmer, I. (1972). Oxygen uptake during swimming in man. ~.~. Physio1. 33: 504-509.

------:---: , A. Lundin, and B. O. Ericksson (1974a). Maximum oxygen uptake during swimming and running by elite swimmers. ~.~. Physio1. 36: 711-714. , E. M. Stein, B. Sa1tin, B. Ekb10m, P.-O. Astrand (1974b). ---:-:----:-Hemodynamic and respiratory responses compared in swimming and running. ~.~. Physio1. 37: 49-54. Holmgren, A., F. Mossfe1dt, T. Sjostrand, and G. Strom (1960). Effect of training on work capacity, total hemoglobin, blood volume, heart volume, and pu1serate in recumbant and upright positions. Acta Physiol. Scand. 50: 72-83. Hong, S. K., P. Cerrete11i, J. C. Cruz, and H. Rahn (1969). ~' ~. Physio1. 27: 535-538. Horwitz, L. D., J. M. Atkins, and S. J. Leshin (1972). Role of the Frank-Starling mechanism in exercise. Circ. Res. 31: 868-875. Johnson, J. M., M. Niederberger, L. B. Rowell, M. M. Eisman, and G. L. Brenge1mann (1973). Competition between cutaneous vasodilator and vasoconstrictor reflexes in man. ~.~. Physio1. 35: 798-803. Judy, W. V., F. M. Langley, K. D. McCowen, D. M. Stinnett, L. E. Baker, and P. C. Johnson (1967). Comparative evaluation of the thoracic impedance and isotope dilution methods for measuring cardiac output. Aerospace Med. 40: 532-536. 177

Kagawa, J. and D. Kerr (1970). Effects of brief graded exercise on specific airway conductance in normal subjects. ~.~. Physiol. 28: 138-144. Karlsson, J. (1971). Lactate and phosphagen concentrations in working muscle of man with special reference to oxygen deficit at the onset of work. Acta Physiol. Scand. Suppl. 358. Kubicek, W. G., J. N. Karnegis, R. P. Patterson, D. A. Witsoe, and R. H. Mattson (1966). Development and evaluation of an impedance cardiac output system. Aerospace Med. 37: 1208-1212. , D. A. Witsoe, R. P.Patterson, and A. H. L. From ------~(J~u~1-y~1~,-·1969 to December 31, 1970). Development and evaluation of an impedance cardiographic system to measure cardiac output ard other cardiac parameters. NASA Report Contract No. NAS 9-4500. Lally, D. A. (1973). "Cardio-respiratory responses to exercise in divers. 1I (Unpublished Ph.D. Thesis) University of Hawaii, 1973. Lin, Y.-C. and S. M. Horvath (1972). Autonomic nervous control of cardiac frequency in the exercise-trained rat. ~.~. Physiol. 33: 796-799. Loomis, J. L., W. C. Nicholas, L. Bartlett, P. Carroll, and E. R. Buskirk (1972). Flow control valve for expired gas collection from scuba-equipped swimmers. ~.~. Physiol. 32: 869-871 Luft, U. C., L. G. Myhre, and J. A. Loeppky (1973). Validity of Haldane calculation for estimating respiratory gas exchange. ~. ~. Physiol. 34: 864-865. Lundvall, J. (1972). Tissue hyperosmolarity as a mediator of vasodi­ lation and transcapillary fluid flux in exercising muscle. Acta Physiol. Scand. Suppl. 379. , S. Mellander, H. Westling, and T. White (1972). Fluid ---;----;:--transfer between blood and tissues during exercise. Acta Physiol. Scand. 85: 258-269. Magel, J. R. and J. A. Faulkner (1967). Maximum oxygen uptakes of college swimmers. ~.~. Physiol. 22: 929-938. McArdle, W. D., R. M. Glaser, and J. R. Magel (1971). Metabolic and cardiorespiratory response during free swimming and treadmill walking. ~.~. Physiol. 30: 733-738. McDonough, J. R., R. A. Danielson, R. E. Wills, D. L. Vine (1974). Maximal cardiac output during exercise in patients with coronary artery disease. Amer.~. Cardiology 33: 23-29. 178

Miyamura, M. and Y. Honda (1972). Oxygen intake and cardiac output during maximal treadmill and bicycle exercise. ~.~. Physiol. 32: 185-188. Moore, T. 0., E. M. Bernauer, G. Seto, Y. S. Park, S. K. Hong, and E. M. Hayashi (1970). Effect of immersion at diffe~ent water temperatures on graded exercise performance in man. Aerospace Med. 41: 1404-1408. Morlock, J. F. and R. H. Dressendorfer (1974). Modification of the Quinton-Monarch ergometer for underwater use. Undersea Med. Res. In press. Nadel, E. R., I. Holmer, U. Bergh, P.-O. Astrand, and J. A. J. Stolwijk (1974). Energy exchanges of swimming man. ~.~. Physiol. 36: 465-471. Otis, A. B. (1964). The work of breathing. In "Handbook of Physiology, Sec. 3 Respiration, Vol. 1, W. O. Fenn and H. Rahn (Eds.) American Physiological Society, Washington, D.C. Radford, E. (1955). Ventilation standards for use in artificial respiration. ~.~. Physiol. 7: 451-460. Rahn, H., W. O. Fenn, and A. B. Otis (1949). Daily variations of vital capacity, residual air, and expiratory reserve volume including a study of the residual air method. ~. ~. Physiol. 1: 725-736. Reeves, J. T., R. F. Grover, S. G. Blount, Jr., and G. F. Filley (1961). Cardiac output response to standing and treadmill walking. ~.~. Physiol. 16: 283-288. Ross, J., Jr., J. W. Linhart, and E. Braunwald (1965). Effects of changing heart rate in man by electrical stimulation of the right atrium: studies at rest, during exercise, and with isoproterenol. Circulation 32: 549-556.

Rowell, L. B. (1962). "Factors Affecting the Prediction of the Maximal Oxygen Intake from Measurements Made During Submaximal Work with Observations Related to Factors Which May Limit Maximal Oxygen Intake." (Ph.D. Thesis). University of Minnesota, Minneapolis. (1974). Human cardiovascular adjustments to exercise ----;--:,..,--and thermal stress. Physiol. Revs. 54: 75-159. , J. R. Blackmon, and R. A. Bruce (1964). Indocyanine -----'"green clearance and estimated hepatic blood flow during mild to maximal exercise in upright man. ~. Clin. Invest. 43: 1677-1690. 179

, G. L. Brenge1mann, J. A. Murray, K. K. Kraning, II, -----a-nd~F~.~Kusumi (1969a). Human metabolic responses to hyperthermia during mild to maximal exercise. ~.~. Physio1. 26: 395-402. , J. A. Murray, G. L. Brenge1man, and K. K. Kraning, II. (l969b). '1uman cardiovascular adjustments to rapid changes in skin t~~mperature during exei'cise. Circ. Res. 24: 711-724. , G. L. Brenge1mann, J. R. Blackmon, and J. A. Murray ----~(~1=97=0~)-.- Redistribution of blood flow during sustained high skin temperature in resting man. ~.~. Physio1. 28: 415-420. Sa1tin~ B. (1964). Circulatory response to submaximal and maximal exercise after thermal dehydration. ~.~. Physiol. 19: 1125-1132.

----- and J. Stenberg (1964). Circulatory response to prolonged severe exercise. ~.~. Physiol. 19: 833-838. , A. P. Gaage, U. Bergh, and J. A. J. Sto1wijk (1972). ---;:;--.--Body temperatures and sweating during exhaustive exercise. ~. ~. Physio1. 32: 635-643. Scho1ander, P. F. (1947). Analyzer for accurate estimation of respiratory gases in one-half cubic centimeter samples. -J. ----Biol. Chern. 167: 235-250. Shephard, R. J. (1959). Partitiona1 respirometry in human subjects. ~. ~. Physio1. 13: 357-367. (1969a). A nomogram to calculate the oxygen cost of ------;----:- running at slow speeds. ~. Sports Merl. Phys. Fitness 9: 10-16. (1969b). Endurance Fi tness. Toronto: Un'j vers i ty of ---=To-r-o-n-=-to---;::P-res s. Sloan, A. W. and J. B. de V. Weir (1970). Nomograms )or prediction of body density and total body fat from skinfo1d measurements. ~. ~. Physio1. 28: 221-222. Smith, J. J., V. T. Weidmeier, F. E. Tristani, and K. E. Cooper (1969). Measurement of cardiac output during body tilt using the impedance cardiograph. -Fed. --Proc. 28: 643. Smyke, E. (ed). (1964). liThe 1965 official national collegiate atn1etic association swimming guide." New York: The National Collegiate Athletic Bureau. Snedecor, G. W. and W. G. Cochran (1967). Statistical Methods. Ames: Univ. of Iowa Press. 180

Stenberg, J., P.-O. Astrand, B. Ekblom, J. Royce, and B. Saltin (1967). Hemodynamic response to work with different muscle groups, sitting and supine. ~.~. Physiol. 22: 61-70. Sugimoto, T., K. Sagawa, and A. C. Guyton (1966). Effect of tachycardia on cardiac output duri~g normal and increased venous return. Am.~. Physiol. 211: 288-292. Taylor~ H. L., E. Buskirk, and A. Henschel (1955). Maximal oxygen intake as an objective measure of cardio-respiratory performance. ~. ~. Physiol. 8: 73-80. ______~~~, Y. Wang, L. Rowell, and G. B10mqvist (1963). The standardization and interpretation of submaximal and maximal tests of working capacity. Pediatrics 32: 703-722. ______, W. Haskell, and S. M. Fox (1969). Exercise tests: a summary. In: Measurement in Exercise Electrocardiograph~, edited by H. Blackburn. Springfield, Ill.: Thomas. Tracy, R. A. (1971). "Case studies of the Effect of Systematic Training on Maximal Oxygen Consumption, Myocardial and Circulatory Function and Running Performance among Selected Middle D"istance Runners." (Unpublished Ph.D. Thesis) University of Minnesota. Vanhoutte, P. M. and J. T. Shepherd (1970). Effect of temperature on reactivity of isolated cutaneous veins of the dog. Am. ~. Physiol. 218: 187-190. von Dobeln, W. (1954). A simple bicycle ergometer. ~. ~. Physio1. 7: 222-224. Wagner, J. A., S. M. Horvath, T. E. Dahms, and S. Reed (1973). Validation of open-circuit method for the determination of oxygen consumption. ~.~. Physio1. 34: 859-863. Wang, Y., R. J. Marshall, and J. T. Shepherd (1960). The effect of changes in posture and of graded exercise on stroke volume in man. J. Clin. Invest. 39: 1051-1061. Wasserman, K., B. J. Whipp, S. N. Koyal, and W. L. Beaver (1973). ~. ~£l. Physiol. 35: 236-243. Whipp, B. J. and K. Wasserman (1969). Efficiency of muscular work. ~. B£El. Physio1. 26: 644-648. Wilmore, J. H. and D. L. Costi11 (1973). Adequacy 9f the Haldane transformation in the computation of exercise V02 in man. ~. ~. 'physiol. 35: 85-89. 181

Wyndham, C. H., N. B. Strydom, W. P. Leary, and C. G. ~Jilliams (1966). Studies of the maximum capacity of men for physical effort. Intern. I. Angew. Physio1. 22: 285-310. Wyss, C. R., G. L. Brenge1mann, J. M. Johnson, L. B. Rowell, and M. Niederberger (1974). Control of skin blood flow, sweating, and heart rate: role of skin vs. core temperature. ~. ~. Physio1. 36: 726-733.