UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prediction of triathon performance from ventilatory threshold measurements Langill, Robert H. 1993-12-31

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

Item Metadata


ubc_1993_spring_langill_robert.pdf [ 3.13MB ]
JSON: 1.0077181.json
JSON-LD: 1.0077181+ld.json
RDF/XML (Pretty): 1.0077181.xml
RDF/JSON: 1.0077181+rdf.json
Turtle: 1.0077181+rdf-turtle.txt
N-Triples: 1.0077181+rdf-ntriples.txt
Original Record: 1.0077181 +original-record.json
Full Text

Full Text

PREDICTION OF TRIATHLON PERFORMANCE FROM VENTILATORY THRESHOLD MEASUREMENTS  by  ROBERT HENRY LANGILL B.Sc., University of British Columbia, 1989.  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHYSICAL EDUCATION in THE FACULTY OF GRADUATE STUDIES School of Physical Education and Recreation  We accept this thesis as conforming to the required standards  THE UNIVERSITY OF BRITISH COLUMBIA March, 1993. © Robert H. Langill, 1993.  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of Physical Education The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  ii ABSTRACT  The purpose of this study was to predict Ironman Triathlon (2.4 mile swim, 112 mile cycle, 26.2 mile run) performance times from ventilatory threshold (TVENT) measurements of swimming, cycling, and running. Ten trained triathletes (mean age = 29.7yrs., ht = 179.8cm, wt = 76.8kg, bodyfat = 11.4%) performed progressive intensity tests for treadmill running, cycle ergometry, and tethered swimming. The excess CO 2 elimination curve was used to determine TVENT in each component sport with the resulting estimated times of 64.2, 380.0, 174.5, 672.8 minutes for swimming, cycling, running, and overall time respectively. Individual estimates were then compared to actual segment and overall times to produce the following linear regression equations for predicting actual from estimated time (in minutes): actual swim = 1.15 * estimated swim - 6.75 actual cycle = 0.22 * estimated cycle + 262.6 actual run = 3.03 * estimated run - 267.1 actual overall = (-3.58 * est. swim) + (-0.10 * est. cycle) + (3.76 * est. run)+ 291.35  Significant correlations of r = 0.83, 0.70, 0.76, and 0.89 were calculated between swim, cycle, run, and overall estimated versus actual times respectively. Thus, between 49 and 69% of the variance in actual time is explained by TVENT for that component sport. Also, 78% of the total variability was accounted for by the TvENT estimation when the three sports were combined. These findings suggest that while TvENT is able to account for a significant proportion of triathlon performance time other factors such as fatigue, dehydration, terrain, heat, etc. are confounding the overall prediction.  iii TABLE OF CONTENTS  ABSTRACT ^  ii  TABLE OF CONTENTS ^  iii  LIST OF TABLES ^ ACKNOWLEDGEMENT ^  vi  CHAPTER 1: INTRODUCTION TO THE PROBLEM ^ 1.1 Introduction to the Problem ^ 1.2 Statement of the Problem ^ 1.2.1 Subproblems ^ 1.3 Definitions ^ 1.4 Delimitations ^ 1.5 Limitations ^ 1.6 General Hypothesis ^ 1.6.1 Secondary hypotheses ^ 1.7 Rationale ^ 1.8 Significance of the Study ^  1 1 4 4 4 5 5 6 6 7 8  CHAPTER 2: LITERATURE REVIEW ^ 10 2.1 Introduction^ 10 2.1.1 Single Endurance Sports ^ 10 2.1.2 Triathlon ^ 11 2.2 Single vs. Multiple Mode Sport ^ 12 2.2.1 Specificity of training ^ 12 2.2.2 Intrinsic and Extrinsic Factors ^ 14 2.2.3 Cumulative Effects ^ 16 2.3 Physiological Determinants of Endurance Performance ^ 16 2.3.1 Maximal Aerobic Power ^ 17 2.3.1a Swimming ^ 18 2.3.1b Cycling ^ 19 2.3.1c Running ^ 19 2.3.1d Triathlon ^ 19 2.3.1e Concerns ^ 20 2.3.2 Economy of Motion ^ 21 2.3.2a Swimming ^ 21 2.3.2b Cycling ^ 21 2.3.2c Running ^ 22 2.3.2d Triathlon ^ 22 2.3.2e Concerns ^ 23  iv 2.3.3 Anaerobic Threshold ^ 2.3.3a Swimming ^ 2.3.3b Cycling ^ 2.3.3c Running ^ 2.3.3d Triathlon ^ 2.3.3e Concerns ^ 2.4 Conclusions ^  24 24 25 25 26 27 27  CHAPTER 3: METHODS AND PROCEDURES ^ 3.1 Subjects ^ 3.2 Testing Procedures ^ 3.3 Testing Protocols^ 3.4 Experimental Design and Data Analysis ^  29 29 29 30 34  CHAPTER 4: RESULTS AND DISCUSSION ^ 4.1 Results ^ 4.2 Discussion ^  35 35 44  CHAPTER 5: SUMMARY AND CONCLUSIONS ^ 5.1 Summary ^ 5.2 Conclusions ^ 5.3 Recommendations ^  56 56 57 58  BIBLIOGRAPHY^  59  APPENDIX A: Sample calculation for estimation of the swim component ^ 71 APPENDIX B: Sample calculation for estimation of the cycle component ^ 73 APPENDIX C: Sample calculation for estimation of the run component ^ 76 APPENDIX D: Individual data for estimated and actual swim,cycle,and run ^ 77 APPENDIX E: Mechanisms of lactate and ventilatory threshold ^ 80  v LIST OF TABLES  1. Descriptive data for all subjects ^  35  2. Weekly training mileage and previous triathlon experience for all participants ^  36  3. Heart rate and lk;4 2 measured at TVENT and expressed as a percentage of their maximal value ^  38  4. Estimated and actual swimming times and paces for the Ironman Triathlon ^  40  5. Equations used in the determination of the cycling portion of the lronman Triathlon^  40  6. Estimated and actual cycling times and paces for the Ironman Triathlon ^ 41 7. Estimated and actual run times and paces for the lronman Triathlon ^ 42 8. Conversion factors from estimated to actual for each component and overall time in the Ironman Triathlon ^  43  vi  ACKNOWLEDGEMENT  Firstly, I would like to thank my committee members; Dr. Taunton, Dr. Coutts, and Dr. Vaughan for their contributions towards this project. Particularly, I would like to thank very much my committee chair Dr. Rhodes for his guidance in not only this thesis but my entire course of study towards my Master's Degree. My greatest admiration is extended to the subjects who participated in this study. The time and energy they put into this project and training for a sport of this nature is a testimony to the type of people they are. Finally, I would also like to thank my parents, Gwen and Clarence, and my brother John, for their support and their ability to listen to me complain even when they probably did not understand exactly what I was complaining about.  1 CHAPTER 1  1.1 Introduction to the Problem  Over the years many attempts have been made to describe the factors which limit the ability to perform prolonged endurance exercise. Studies have attempted to identify the physiological variables that are the most important determinants of endurance performance. Among the key variables that have been shown to be significantly related to performance of endurance activities are: maximal oxygen uptake (VO2M of VO2  (%VO2M  ),  ),  economy of motion, fractional utilization  anaerobic threshold, and fuel supply (Sjodin and  Svedenhag, 1985). Early research (Owles, 1930; Harrison and Pilcher, 1930) supported the existence of critical levels of work intensity above which there was accumulation of blood lactate with accompanying increases in CO 2 excretion and ventilation. This critical intensity was termed the "anaerobic threshold" and defined the workrate at which the tissues oxygen supply first fell below demand and excess energy needs were supported by anaerobic metabolism (Wasserman and Mcllroy, 1964). In theory, below this point work may be performed for indefinite durations since the energy requirements are being supplied predominantly through unlimited aerobic energy sources, while waste products are being adequately removed (Anderson and Rhodes, 1989). In order to best define the anaerobic threshold point both noninvasive and invasive variables such as VE  ,  VE /V02 VCO2 excess CO 2 , and blood lactate have all been monitored with ,  ,  varying degrees of success. Particular emphasis has been placed on the refinement of noninvasive respiratory variables in order to minimize the need for invasive procedures. This has allowed researchers to use progressive workload  2 increases to bring about nonlinear responses in respiratory exchange variables to noninvasively define anaerobic threshold. While the nomenclature and mechanisms inherent in the anaerobic threshold concept are still being challenged, there is widespread support particularly in the performance data. The threshold appears to represent a key parameter defining the ability to maintain high intensity exercise (Whipp et al., 1981) and a critical intensity level above which endurance performance is severely limited (Rhodes and McKenzie, 1984). It is suggested that exercise intensity at anaerobic threshold represents a maximal steady state level that an athlete can sustain for an extended period without accumulating lactate. Anaerobic threshold related variables have been shown to have a close relationship to endurance performance that is good or better than other physiological variables including 1%0 2 (Tanaka et al., 1981; Davis et al., 1979). Several performance studies have documented the value of anaerobic threshold as a determinant of endurance activities. Studies of long distance running produced correlations of the order of 0.88 to 0.99 for distances ranging from 3.2 km to the marathon (Conconi et al., 1982; Farrell et al., 1979; Powers et al., 1983; Sjodin and Schele, 1982; Williams and Nute, 1983). Rhodes and McKenzie (1984) extended the concept of anaerobic threshold by instead of just correlating it with performance used treadmill running velocity at the anaerobic threshold to estimate actual marathon time with an r=0.94. While results have been strongest in running, other sports have shown some support. Miller et al. (1985) found anaerobic threshold to have a strong relationship (r=0.93) to 15 km time trial cycling performance. As well, Loat and Rhodes (1991) found the anaerobic threshold, as defined by excess CO 2 , to be a workload that could be sustained for one hour on a cycle ergometer. All of these represent situations  3 where anaerobic threshold has been correlated with some success to performance in single sport endurance contests. The triathlon is an endurance contest in which participants compete consecutively in three sports, usually swimming, cycling, and running. The endurance triathlon has really only existed since 1978 when the Waikiki Rough Water Swim (3.9 km), the Around Oahu Bike Race (180.2 km), and the Honolulu Marathon (42.2 km) were combined into the Hawaii Ironman Triathlon (O'Toole et al., 1989). The three sport performance dictates that the successful triathlete is one who has the ability to perform each sequential event at optimal pace without creating fatigue that will hinder performance in the next event. While the physiological bases for success in the triathlon remain to be clarified it will obviously include the ability to maintain minimal alterations in homeostasis of cardiovascular, haemodynamic, thermal and metabolic function for long periods of time (O'Toole et al., 1989). The triathlete will seek to perform at a level optimizing potential for the duration of the event. Studies have demonstrated evidence that in single sport contests (particularly strong in running) anaerobic threshold can be used to represent this point of optimal pace for endurance sports. Since the triathlon is an endurance contest by nature it seems reasonable that those factors which contribute to success in single event endurance contests should also contribute to triathlon success. Thus, the concept of anaerobic threshold should be able to be applied here in order to characterize performance just as in other endurance contests. At this point very few studies have looked at using ventilatory threshold as an indicator of success in triathloning.  4 1.2 Statement of the Problem The purpose of this investigation was to determine if triathlon performance times can be predicted from ventilatory threshold measurements of swimming, cycling, and running.  1.2.1 Subproblems 1) to determine if the velocity at ventilatory threshold during running, cycling, and swimming can estimate race pace (time) for that component of the triathlon.  2) to determine if each component can be used together to predict overall triathlon time.  1.3 Definitions 1) Ventilatory threshold - (TvENT) the point where the aerobic energy response is of insufficient magnitude to supply the tissues energy requirement and there is an increased reliance on anaerobic processes with an accompanying abrupt increase in excess CO 2 .  2) Maximal Steady State - the highest intensity work may be performed at for theoretically indefinite durations.  3) Excess CO 2 - nonmetabolic CO2 (EXCO 2 ) formed as a result of the hydrogen ions of lactic acid being buffered by bicarbonate in the following reactions: HLa + NaHCO3 = NaLa + H2 CO3 = CO 2 + H2 O (Wasserman et al., 1973)  5 The calculation of excess CO 2 will be based on the formula of Volkov et al. (1975) where: ExCO 2 = VCO2 - (RQrest * tiO2 )  1.4 Delimitations  This study was delimited by: 1) a sample of triathletes from the Vancouver area between the ages of 18 and 35 years with a minimum of two triathlons experience.  2) a respiratory gas sampling rate set at 15 second intervals.  3) the methodology applied to determine velocity at TvENT for swimming, cycling, and running.  1.5 Limitations  This study was limited by: 1) the data collection capabilities of the Beckman Metabolic Measurement Cart and the Hewlett Packard Data Acquisition system interfaced with it.  2) the individuals metabolic response to the exercise protocols.  3) race day conditions (terrain, weather, equipment problems, etc.)  6 1.6 General Hypothesis Predicted time for the triathlon will correlate highly with actual triathlon performance time. Specifically, a highly significant relationship will exist between:  1) treadmill run velocity at TVENT and the run section of the triathlon.  2) tethered swim velocity at TVENT and the swim section of the triathlon.  3) cycle ergometry velocity at TVENT and the cycle section of the triathlon.  1.6.1 Secondary hypotheses 1) a significant difference will exist between  VO2mAx measured for swimming,  cycling, and running such that: V02 MAX run >  VO2 MAX cycle > V02 MA  swim  2) TVENT measured for swimming, cycling, and running will demonstrate the  following:  a) TVENT absolute (expressed as a V02 ) will show a significant difference for  each component as follows: TVENT run > TVENT cycle TVENT - VENT swim b) TVENT relative (expressed as a %1102mAx ) will demonstrate a  nonsignificant difference between each component sport when expressed as a percentage of maximum.  7  3) the relative heart rate at TvENT (expressed as a %HRmAx ) will also show a nonsignificant difference between the three component sports of the triathlon.  1.7 Rationale Velocity at TvENT has been used with success (r=0.94) in running by Rhodes and McKenzie (1984). Farrell et al. (1979) observed that runners set a race pace which closely approximates the running velocity at which lactate begins to accumulate in the plasma. These observations both support the hypothesis of a velocity which optimizes pace and minimizes fatigue (maximal steady state). Since the physiological basis for success is similar in other endurance sports it seems likely that this concept could be extended. The relationship between l%02mAx in swimming, cycling, and running is documented in the literature. Mean VO2mAx values have been reported as follows:  treadmill run^cycle ergometry^tethered swim 60.5^ 57.9^52.5^Kohrt et al. (1987a) 57.4^ 54.4^46.8^Kohrt et al. (1987b)  Thus, cycling maximums represent about 4% less and swimming maximums represent about 15% less than those achieved running on the treadmill. From these results it is expected that the absolute TvENT in each component will also show the same pattern. However, when TvENT is expressed in relative terms it is expected that there will no longer be differences. This is supported by the work of Schneider et al. (1990) where a nonsignificant difference was found between  8 TvENT's expressed as a % kinfAx for cycle ergometry and treadmill running.  Relative heart rate at TvENT is also expected to show this same nonsignificant difference for each component sport. Roalstad et al. (1987) reported during the Hawaii Ironman Triathlon that subjects maintained average heart rates of approximately 75% of their maximal heart rates during the bike and run portions of the race. They also observed that the better finishing times were found in those triathletes where heart rates fluctuated least over the course of the event.  1.8 Significance of the Study If indeed ventilatory threshold defines a "critical intensity", then exercise at workloads greater than this should result in a progressive metabolic acidosis that becomes rate-limiting. Exercise time to exhaustion (duration) should have an inverse relationship to the amount exercise workrate (intensity) exceeds TVENT. It is important to determine if the protocols employed in this experiment  actually do define a maximal steady state at which an athlete can perform. Most endurance athletes regardless of endurance ability wish to perform for prolonged periods of time without factors such as lactate accumulation limiting performance (maximal steady state). If it can be shown that TVENT does characterize this point of maximal performance it will have profound implications for the endurance athlete. It could be used in exercise prescription to select a training intensity which elicits maximal aerobic performance. There could also be application to evaluation of training over time based on TVENT changes. This study is also significant in that it takes the concepts of TVENT to a performance level. Thresholds are often stated to be indicative of maximal endurance ability without actual validation. This study validates its statements about the TVENT by actually testing the theories in an endurance setting  9 (triathlon). The use of threshold concepts outside of the lab setting is the truest test of these measures. An endurance triathlon will definitely test the ability of TvENT to characterize maximal steady state during prolonged sport.  10 CHAPTER 2  2.0 Literature Review  2.1 Introduction  Endurance sports have been a very popular area of research for physiologists in the past few decades. This is undoubtedly due to the opportunity to observe a variety of factors under highly demanding conditions and thereby gain a better understanding of human performance potential. Much of this research in endurance capacity is designed so that a better idea of those factors contributing to success in endurance contests will emerge. As such, this would help the physiologist to develop methods for predicting performance. The notion of predicting performance success is very attractive, but is a complicated and multifaceted question encompassing areas of physiology, biomechanics, morphology, psychological profile, etc. (Toussaint and Beek, 1992). A new series of questions emerge with the advent of the multiple sport contest of triathlon, particularly with respect to the relationships between the component sports. Research, to date, is conflicting but several lines of evidence suggest a similar physiological basis that may allow prediction as has been attempted in single contest endurance sports. This review will concentrate on the physiological determinants of endurance performance in the emerging research area of triathlon by examining current findings in this sport and its component sports of swimming, cycling, and running.  2.1.1 Single Endurance Sports  There is little information concerning the aerobic capacity of competitive endurance cyclists and swimmers, in contrast to the wealth of data on endurance  11 runners. While swimmers tend to train comparable distance to endurance athletes (Costill et al., 1991) the majority of competitive swim events tend to be sprint oriented contests lasting less than 3 minutes (Smith et al., 1984). Consequently, most of the research has focussed on maximal performance as opposed to sustained performance. By virtue of the man-machine interaction (Kyle and Mastropaolo, 1978), cycling research has been confined more to biomechanical considerations of cadence, seat height, gear ratios, etc. It is unquestionably running where the majority of performance research has been done. This is likely the result of cycling and swimming having much greater biomechanical components while running is more physiological (Sjodin and Svedenhag, 1985). Lack of research should not be taken to mean a lack of evidence for the ability to determine endurance performance success in these sports. While often there are few direct correlation studies, the sports exhibit characteristics supporting the likelihood of these concepts.  2.1.2 Triathlon While many sports have developed ultraendurance contests, triathlon is a sport created specifically to test endurance ability (O'Toole and Douglas, 1989). As a contest the Ironman triathlon, with its 2.4 mile swim, 112 mile cycle and 26.2 mile run, is synonymous with ultraendurance performance. There has been an exponential rise in popularity from only 15 starters in the first Ironman (1978) to 1275 starters in 1988 (selected from over 20,000 applicants) (O'Toole and Douglas, 1989). This rise in popularity has promoted an increased interest within the scientific community, now focussed on examining the physiological consequences of multisport training and racing. O'Toole et al. (1989) state that success in a triathlon depends upon the ability of the triathlete to perform each  12 of the sequential events at optimal pace without creating fatigue that will hinder performance in the next event. A heterogeneous group of Ironman triathletes maintained average heart rates of approximately 75% of maximum during the bike and run portions (Roalstad et al., 1987). Those with heart rates fluctuating least over the course of the event had the best finishing times suggesting that there is an optimal, sustainable pace that the triathlete will want to approximate to be successful. This suggests the key determinants of endurance performance, characterizing multisport endurance, may be similar to single sport endurance.  2.2 Single vs. Multiple Mode Sport  2.2.1 Specificity of training  It is the above statement of O'Toole et al. (1989) regarding "not creating fatigue that will hinder the next sequential event" that is a key factor differentiating multiple from single endurance sports. This suggests a new series of considerations that must be made to apply performance research to triathlon. The theory of training specificity suggests that adaptations to training are specific to the mode of training (Kohrt et al., 1987a). An activity is required to tax the metabolic pathways of choice (aerobic vs. anaerobic) and must also stress (or recruit) specific muscle groups and neural pathways used in the activity (Kohrt et al., 1989). Single sports necessitate adaptation to only one mode while the triathlete will need to reflect, to some degree, the specificity seen in each component sport. Clausen et al. (1973) postulated that there were both central adaptations (resulting from nonspecific training) and peripheral adaptations (resulting from training specific to the muscle groups used). Muscles used in swimming are different from those used in cycling and running, and while there is some overlap of muscle usage in cycling and running the range of motion,  13 recruitment patterns, length of muscles, type of contraction (concentric vs. eccentric) and speed of contraction are different (O'Toole et al., 1989). Thus, while all component sports are aerobic in nature they utilize different muscle groups (or the same muscles differently) to perform the necessary movement patterns of swimming, cycling, and running. Cycle and swim training result in improvements specific to that activity (Kohrt et al., 1987a) and produce more of a localized muscle stress and peripheral adaptations in the muscles. Magel et al. (1978) states exercise using large muscle groups (eg: running) stresses the oxygen transport mechanisms, while exercise involving smaller muscle groups (eg: swimming) tends to stress oxygen utilization. Much of the specificity of swimming is thought to relate to the reduction in active muscle mass coupled with the reduction of antigravity work while in water, and the horizontal exercise position. Central to multisport training is the question of whether there is an ability to transfer benefits from one activity to another with the concurrent ("crosstraining") training done by triathletes. Both swim training (Magel et al., 1978) and cycle training (Pechar et al., 1974; Town and Sinning, 1982) have failed to demonstrate improvements in running ability. As well, run training has failed to demonstrate improvements in swim ability (McArdle et al., 1978). Running to improve cycling is not as definitive with some studies (Pechar et al., 1974; Town and Sinning, 1982) showing an improvement indicative of a generalized effect from run training, although this improvement tended to be quite small, suggesting only a small generalizable training effect. The work of Kohrt et al. (1987b) is the only study demonstrating a possible crosstraining effect by reducing training in cycling and swimming while maintaining run levels. At the end of a three month training phase, there was a reduction in both running  14 and cycling 1%0 2mAx , while swimming was maintained. It is therefore possible that swim 1%02mAx was maintained by a generalized training effect from the other two component sports.  2.2.2 Intrinsic and Extrinsic Factors  While physiological variables govern performance to some degree, there are undoubtedly other factors that must be considered to be reducing the ability of the triathlete to maintain a constant pace. Many of the factors are present in single endurance sports, but by virtue of the increased duration these factors are more definitive "controllers of pace" in ultraendurance triathlons. To gain an understanding of the importance of these factors we only need to look at the reasons people require medical attention at the Hawaii Ironman. The primarily diagnosed reasons are dehydration (52%), exhaustion (20%), trauma (13%), heat cramps (6%) and electrolyte imbalance (%5) (Laird, 1989). Undoubtedly, many participants will show combinations of these, not single problems. The Ironman distance triathlon, by virtue of its competition time, is directly related to many of these medical problems experienced. Climate, season of the year, and topography of the course provide extrinsic factors that can magnify the medical problems. From the point of view of reduction in pace, there are few problems with the swim unless it is in a body of water where waves, current, and cold water temperature are prevalent. The most often cited problem in the bike portion is cramping due to the time on the bike (Laird, 1989). Some athletes have also cited mild dehydration at this point (Farber et al., 1991). It is during the run where problems start to become severe. The cumulative effects of time, distance, and heat begin to be manifested in dehydration, exhaustion, and electrolyte imbalances (Laird, 1989).  15 In events over four hours electrolyte imbalance and dehydration are suggested to be important factors in race performance (Hiller, 1989). van Rensenburg, et al. (1986) reported a decrease in body weight of 4.5% in a triathlon with mean finish time of 11.45 hours. Costill and Miller (1980) noted that work performance decreases and physiological indices of stress increase when even low levels of dehydration are combined with heat. This effect is magnified if there is a concomitant salt depletion. Heat cramps are often associated with this hyponatremia (Hiller, 1989). Costill and Miller (1980) estimated a 5-7% total body sodium chloride deficit in an endurance athlete sustaining a 5.8% body weight loss. Even with a modest sweat rate, combined with the heat of the Hawaii Ironman, there could be a loss of 36g of salt from the total body sodium stores of about 125g (Hiller, 1989). The solution considering that 70% of hyponatremic athletes are also dehydrated is to ensure a recommended supplemental sodium intake combined with programmed hydration (Hiller, 1989). The exact protocols remain to be investigated. Available evidence on substrate utilization during prolonged exercise supports an increase in fat metabolism (O'Toole et al., 1989). Free fatty acid increases have been reported of between 2.9 and 4.6 times prerace values in ultraendurance triathlons (Holly et al., 1986; Van Rensenburg et al., 1986). The actual use and regulation of substrates appears to be similar to other endurance activities, although there is evidence ultraendurance athletes competing for 9-16 hours may find standard rates of carbohydrate replacement to be insufficient to sustain this duration of exercise (Applegate, 1989). As well, the bike portion of the triathlon combines liquid and solid carbohydrate feedings in order to replace energy. Often carbohydrate replacement, particularly solid food, is stopped about one hour prior to the start of the run (O'Toole et al., 1989).  16 2.2.3 Cumulative Effects While these intrinsic factors will play a role in performance the cumulative effects of performing three endurance events in their own right sequentially will bring about changes in body physiology that must be dealt with. The study of Kreider et al. (1988) compared single sport performances to equivalent component sport performances in a triathlon. One of the main differences observed in the cycling was the increase in core temperature during swimming causing thermoregulatory and cardiovascular responses to occur earlier in triathlon cycling. These responses were similar to changes found in the latter stages of the control cycling session. The subjects were unable to maintain control cycling work rates as the triathlon cycling session progressed. Possibly this decrease in performance was due to thermal stress and/or the early signs of dehydration. The cumulative effects were even more evident in the run. The post cycling core temperature was 38.4°C versus an initial control run core temperature of 37°C. The subjects perceived the work during the triathlon run to reflect increased physiological stress. A further 1.5% loss of body weight occurred during the run and while core temperature and dehydration responses were not excessive, the thermal stress would only be magnified during actual race environmental conditions. Therefore, triathletes should frequently take in fluids to minimize elevation in core temperature, dehydration, and impaired performance (Kreider et al., 1988).  2.3 Physiological Determinants of Endurance Performance While many variables have been examined relative to endurance performance three key physiological ones have been the focus of research: 1) economy of motion  17 2) maximal aerobic power (t%02mAx ) 3) anaerobic threshold^(Pate and Branch, 1992). Each has been demonstrated to be a determinant of endurance performance, characterizing the ability of an athlete to sustain effort for prolonged periods. Economy refers to the rate of energy expenditure associated with a given rate of power output or speed of movement (Pate and Branch, 1992). This will be desirable to the endurance athlete given that being "economical" will result in a lesser rate of energy expenditure, thus conserving energy to help sustain performance. Maximal aerobic power (1%02mAx ) represents the upper rate limit of one's ability to use oxygen in metabolism. High performance capacity has been strongly linked to this peak of aerobic metabolism (Sjodin and Svedenhag, 1985). Finally, the anaerobic threshold defines a workload intensity at which blood levels of lactic acid begin to rise significantly above normal resting values (MacDougall, 1977). In the simplest sense this is the point of imbalance between lactic acid production and its removal or uptake beyond which progressively more lactate accumulates in the blood. The association of lactate accumulation with fatigue suggests anaerobic threshold represents the peak of the body's ability to perform prolonged exercise. Each of these measures has demonstrated characteristics necessary for endurance success in several sports.  2.3.1 Maximal Aerobic Power  Since the athlete wishes to develop, an ability to perform at the maximal sustainable oxygen consumption, early studies focused on 1%02 MAX as the source of relationships.  18 2.3.1a Swimming The economical use of energy which would be a prerequisite for distance swimming success is of less importance over short distances (Holmer, 1974). As most swim contests are short duration, the swimmer's selection of the optimal combination of local muscle power and endurance will tend to favour power at the expense of endurance (Craig and Pendergast, 1979). Thus, most of the predictive research has focused on generating maximal energy output without much regard to the ability to sustain performance long term. Considering this it is not surprising that the majority of correlations of swim performance have been to 1%02^. Several researchers (Chatard et al., 1985; Costill et al., 1985; Montpetit et al., 1981) have found high correlations between 1%02mAx and performance. Chatard et al. (1990) found 368m swim performance mainly related to 1%0 24,,Ax (r=0.80). Costill et al. (1985) and Nomura (1983) also found high correlations of the order or r=0.80 and r=0.75 respectively between swim performance and tO2 mAx . The observation by LePere and Porter (1975) that 1%02^is greater in more skilled swimmers suggest it could be used to differentiate performance success. Often in endurance settings the measure of 1%02mAx is modified to better represent the actual level of performance and is called the fractional utilization of 1%02mvc (%1%02mAx ). Magel et al. (1975) used a fractional utilization of 70% of maximum determined in a swimming  VO2 MAX test  as the basis of  training and was able to improve swim performance. Holmer (1974) found at a similar workrate of 60-70%  VO2 MAX blood  lactate levels were not significantly  elevated, suggesting an ability for prolonged performance at this level.  19 2.3.1b Cycling  as the criterion measure to Cycling has rarely been examined with VO2 M,ax predict performance. Krebs et al. (1986) found VO2 M,ax to be the most significant physiological predictor related to 25 mile time trial cycling. Malhotra et al. (1984) studying performance time in an 84km cycling event found the highest correlation (r=-0.87) with 1%02mAx . In examining time to exhaustion, Aunola et al. (1990) found that 70% of the variance could be explained by the workrate performed at maximum.  2.3.1c Running 1%02 hfAx is a sensitive indicator of level of marathon running ability (Sjodin  and Svedenhag, 1985) and race pace (r>0.78) (Foster et al., 1977; Farrell et al., 1979; Maughan and Leiper, 1983). It is not only marathons that have produced high correlations with r=0.76 and r=0.97 for 10 km by Williams and Cavanagh (1987) and Morgan and Martin (1986) respectively. In addition, running research has also examined the velocity at 1%0 2 fax as a predictor variable and it was found to have a significant relationship (r=0.87) with 10 km race time (Sjodin and Svedenhag, 1985).  2.3.1d Triathlon  Maximal aerobic power has been found to range from no significant relationship to Ironman performance in cycling (r=0.04) (O'Toole et al., 1987) to a significant relationship (r=-0.78)(Kohrt et al., 1987a). A similar result is seen in the run with no relationship (r=-0.09) (O'Toole et al., 1987) to a moderate relationship (r=-0.68)(Kohrt et al, 1987a). Peak 1%0 2 in tethered swimming correlated (r=-0.50) with swim times to produce no significant relationship (Kohrt  20 et al., 1987a). van Rensburg et al. (1984) found correlations of r=-0.52 for cycling and r=-0.58 for running. As well, relationships between 1%0 2 how and performance times were found for swimming (r=-0.49), cycling (r=-0.32) and running (r=-0.55) in the study of Dengel et al. (1989).  2.3.1e Concerns Although 1%02mAx has demonstrated success as a predictor of performance there are situations where its usefulness is severely reduced. 1%02^has a very low relationship to performance in more homogeneous samples of athlete. Despite heart rate and lactate data reflecting that swimmers were improving endurance, this did not imply equivalent changes in 1%0 2mAx (Costill et al., 1991). Acevedo and Goldfarb (1989) found run times improved without changes in 1%02 M,4x suggesting it may not be the best indicator of endurance performance. When a subgroup of runners with more similar performance capabilities were studied, there was a nonsignificant correlation (r=0.08) between marathon performance and 1%0 2mAx (Costill, 1972). He speculated that it was possible for two runners to have identical 1%0 2 MAX values but differ drastically in their ability to utilize a large fraction of that capacity. Furthermore, relatively low  VO2 nu values x  have been found in some elite  marathoners (Costill et al., 1976; Costill et al., 1973). Thus, while having a high maximal oxygen uptake is of great importance it does not equal success and other factors must be able to compensate for a low capacity.  21 2.3.2 Economy of Motion Svedenhag and Sjodin (1985) found enhancement in running performance, occurring after l%02mAx had reached a plateau, to be associated with slow, steady improvement in economy. There is a paucity of economy of motion studies from a physiological point of view, most focus on more biomechanical aspects.  2.3.2a Swimming Often physiological economy of motion has been successful because of its ability to, in some way, represent technical ability factors. Such is the case with swim economy which has been shown to be a good predictor of technical ability (Pendergast et al., 1977) and a prerequisite for success in performance (van Hardel, 1988). Often, it is these technical ability factors (stroke length, stroke rate, etc.) that are controlling performance (to some degree) and have been shown to have a moderate to high relationship with free swimming (Jensen and Tihanyi, 1978). Economy, expressed as high efficiency for swimming strokes, is a vital prerequisite to the maintenance of a high velocity over long distances (Holmer, 1974). As well, diPrampero et al. (1974) and McArdle (1971) observed that even highly proficient swimmers have considerable variation in the energy cost to swim at a given speed.  2.3.2b Cycling Athletic performance velocity during cycling is determined by the highest steady-state rate of oxygen consumption that can be tolerated and the biomechanical economy of motion, defined as the velocity achieved for a given oxygen consumption (Coyle et al., 1988). While efficiency is clearly important,  22 physiological economy of motion has not been measured in a performance sense.  2.3.2c Running The variation in submaximal oxygen requirements of running at a specific speed is indicative of differences in economy between subjects (Costill et al., 1973; Sjodin and Schele, 1982; Daniels, 1974). It is with this observation in mind that economy of motion has received considerable attention for studying running. Running economy, defined as the steady-state oxygen consumption for a given running speed, has been shown to account for a large and significant proportion of the variation in distance running performance among runners roughly comparable in 1%02mAx (Morgan, 1989; Morgan et al., 1989). In athletes with a similar l%02mAx , Conley and Krahenbuhl (1980) and Morgan and Craib (1992) found running economy to significantly correlate with 10 km time (r=0.79 to 0.83). Daniels (1974) recognized the ability of economy to account for nearly identical 2 mile run times among two champion male runners. Costill and Winrow (1970) state different performance ability in two runners with similar 1,%02mAx values could be attributed to individual differences in economy.  2.3.2d Triathlon Despite its obvious importance very little triathlon research has focused on economy. Roalstad (1989) found economy at a work rate of 160W to significantly relate (r=0.61) to bike finish time. Dengel et al. (1989) reported submaximal V02 measured in each mode to be related to swim (r=0.72), cycle (r=0.60), and run (r=0.64) times. A correlation of r=0.78 between cycle ergometry and bike finish time in a half Ironman triathlon was also demonstrated (Dengel et  23 al., 1986). O'Toole et al. (1989) reported a relationship (r=0.61) between percent peak V02 at 160W and bike finish time.  2.3.2e Concerns Running economy has been found to exhibit the same problem as 1%02  MAX  when a narrower subgroup of runners with similar performances in marathon are studied (Davies and Thompson, 1979). Daniels et al. (1984) found individual stability in run economy varied by as much as 11 % within a particular test period. The relationship between run economy and level of training is equivocal, some studies showing worse economy in untrained and moderately trained (nonelite) runners and other studies demonstrating no difference between trained and untrained (Morgan and Craib, 1992). Relationships of running economy with distance running are not always high, ranging from r=0.36 (Foster et al., 1977) to r=0.83 (Conley and Krahenbuhl, 1980). Most studies involving economy provide little rationale with respect to the chosen level for performance measurements. Considerable concern is raised by Pate et al. (1992) where it was observed that running economy tended to be poorer in subjects with higher maximal aerobic power. This may suggest an artifact in the specific testing protocols used in some studies with respect to the chosen speed for measurement. If a speed is chosen to be submaximal for all athletes, those with the highest l%02mAx will be at a very low relative percentage of  102 x ,  probably  appreciably lower than their typical training intensity (Pate et al., 1992). Thus, runners at lower 1%02^values may have trained more frequently at running velocities similar to those used in the investigation and were more mechanically efficient at those velocities (Bailey and Pate, 1991). The high tO 2mAx runners may have been uncomfortable trying to perform at the prescribed level and a  24 subsequent reduction in economy was found. Consequently, faster runners may be uneconomical at speeds that are slower than their race pace (Williams and Cavanagh, 1987).  2.3.3 Anaerobic Threshold Considering endurance performance, the anaerobic threshold has been described as a key parameter which defines the ability to sustain high-intensity exercise (Caiozzo et al., 1982).  2.3.3a Swimming Anaerobic threshold studies done in swimming have been geared to setting training intensities for athletes in the pool. The observation that the energy expenditure rate increases exponentially with velocity (Holmer, 1979) suggests an optimal point for endurance performance. Olbrecht et al. (1985) found that the speed swum for 30 and 60 minutes was nonsignificantly different from lactate threshold (TLAc) predicted speed. Further, the lactic acid concentration at the end of 30 minutes of swimming was similar (p > 0.05) to lactic acid concentration at Tukc. In swimmers with similar 1%02mAx values anaerobic threshold has demonstrated an ability to differentiate performance capacity (Smith et al., 1984). Consistent with this is the observation that sprinters (100, 200m swimmers), as compared with endurance performers (400, 1500m swimmers), produce high lactate levels at speeds where the heart rate is low relative to its maximum (Treffene, 1979), indicative of a lower anaerobic threshold. Mader et al. (1978), using TLAc, was successfully able to prescribe optimal training pace for developing the speed of a group of East German swimmers.  25 2.3.3b Cycling While anaerobic threshold is often measured in cycling ergometry tests, only recently has it been related to actual cycling performances. In the study of Aunola et al. (1990) it was found that endurance time demonstrated a relationship to oxygen consumption at the anaerobic threshold (r=0.66). Loat and Rhodes (1991), using TvENT as determined by excess CO 2 , predicted a level of work that could be performed for one hour without a significant elevation in blood lactate. The strongest support for anaerobic threshold in cycling are the studies of Coyle et al. (1991, 1988). Coyle et al. (1988) states that it is more accurate to express an individual's metabolic capacity for endurance exercise by reporting 1%02 at Tukc than it is to report 1%0 2 . Further, Coyle et al. (1988) found "time to fatigue" to strongly relate to TLAc (r=0.90). Coyle et al. (1988) also demonstrates the sensitivity of TLAc with respect to "time to fatigue". When subjects performed at 88% of 1%0 2mAx , one group was only 8% above Tukc while the other was 34% above. The average times to fatigue were 60.8 and 29.1 minutes, respectively. Coyle et al. (1991) found the average work rate maintained during a one hour lab cycling performance best correlated to 1%0 2 at Tuk c (r=0.93). They then compared this sustainable Tukc determined work rate  to a 40 km time trial performance with a strong relationship being evident (r=0.88). The oxygen consumption at TLA0 also seemed to be a sensitive indicator of cycling proficiency, as the better endurance cyclists clearly demonstrated higher thresholds.  2.3.3c Running It would be impossible to mention all the anaerobic threshold studies that have produced strong relationships to run performance. Exercise intensities  26 selected quite voluntarily appear to be near the point where other investigators have shown that lactic acid production begins to increase (Karlsson, 1970; Knuttgen and Saltin, 1972). Coen et al. (1991) states that it is possible to provide training recommendations on the basis of anaerobic threshold, especially for endurance training. Variables such as TuNc account for a greater proportion of the variance in running performance than running economy or 1%02 (Farrell et al., 1979; Tanaka et al., 1986; Yoshida et al., 1987). TVENT during incremental exercise seems to provide an indication of lactate maximal steady-state while TVENT + 4.9% showed a marked increase in blood lactate between 15 and 30 minutes (Yamamoto et al., 1991). Farrell et al. (1979) observed runners to set a race pace which closely approximates the run velocity at which lactate begins to accumulate in the plasma. Studies (Farrell et al., 1979; LaFontaine et al., 1981) have used anaerobic threshold predictions for distances from 3.2km to 42.2km with correlations between r=0.91 and r=0.98. Another study, supporting a performance relationship, was reported by Rhodes and McKenzie (1984) where TVENT determined velocity was used to predict (r=0.94) actual marathon performance time. They suggested that run velocity at TVENT may be critical in determining efficient running speed during marathons.  2.3.3d Triathlon Anaerobic threshold determinations have produced similarly conflicting triathlon results. TVENT demonstrated a weak relationship (r=0.58) while TLAc demonstrated no relationship (r=-0.37) with performance times in the cycling portion of the Ironman triathlon (Roalstad, 1989). Van Rensburg et al. (1984) found an r=-0.54 between V02 at the lactate turning point and cycle time. Despite these low correlations for anaerobic threshold Albrecht et al. (1986) was  27 able to set work loads based on TvENT that produced stable heart rate and 1%0 2 from 60 minutes of cycling to the end of 45 minutes of running.  2.3.3e Concerns  Despite the wealth of strong relationships to performances, and the abundance of research, anaerobic threshold remains a debatable and controversial area of research. Arguments question the existence of a threshold (Hughson et al., 1987) and methods of detection based on invasive or noninvasive determination (Davis et al., 1976). The initial definition of anaerobic threshold is constantly examined with respect to whether or not hypoxic conditions are present when lactic acid is being produced in the working tissues (Brooks, 1985). Powers et al. (1984), using gas exchange indices, were not able to reproduce the high correlations found earlier and concluded that TuNc could not be accurately determined by these measures. As well, one of the most often used variables for estimating the point of TLA0 is based on a fixed (4 mmole) lactate concentration in the blood and has been criticized for over and underestimating the true threshold (Stegmann and Kindermann, 1982; McLellan and Jacobs, 1989).  2.4 Conclusions  While the literature provides several lines of evidence to support a prediction concept in single sports there is a lack of research in multisport performances. Even the research that has been done seems to be very conflicting and demonstrates the wide range of variability within triathlon populations. The key determinants of success have all produced strong results in characterizing single sport performance ability although not without some conflicts representing methodological and mechanism concerns. The emergence  28 of other pace determining features with the passage of time and the effect of one sport on another make determination of triathlon success difficult. Despite this difficulty, there is a fundamental necessity for the successful triathlete to be one with the ability to have a highly developed oxygen transport and utilization system as well as the ability to efficiently produce a high energy output for prolonged periods without creating metabolic acidosis (O'Toole et al., 1989). This statement is indicative of an area of study that must have a basis in 1%0 211m , economy of motion, and anaerobic threshold measurements that represent performance success. It is likely a lack of research and inappropriate research that has failed to produce better results in a triathlon setting.  29 CHAPTER 3  3.0 Methods and Procedures  3.1 Subjects Eleven male subjects were selected from various running and triathlon clubs in the Vancouver area. Subjects were tested in the month prior to their participation in the 1991 Canadian Ironman Triathlon; all were in a highly trained state. The subjects were asked to refrain from heavy exercise 24 hrs prior to testing and to be 3 hrs postabsorptive.  3.2 Testing Procedures All testing was performed at the University of British Columbia with the cycling and running sessions in the J.M. Buchanan Exercise Science Lab and the swimming sessions in the Aquatic Centre. During the first session, after the appropriate consent forms were signed, baseline measures of height, weight and body composition (hydrostatic weighing) were determined. The formula for hydrostatic weighing was as follows:  body density -  where  W (W -UWW) C  RV  ^  W = weight on land (kg) UWW = under water weighing (kg) RV = residual volume C = temperature correction  Body density was then converted to percent body fat using the Sid equation  30 (%fat = ((4.95/density) - 4.5) *100). In addition, the first of three maximal oxygen consumption tests (1%02mAx ) was performed either swimming, cycling or running. The remaining lab sessions consisted of the administration of the other two 1%02^tests not administered on the first day. A minimum of 48 hrs separated the testing sessions to try and prevent any carry-over from the previous test. After the lab sessions were completed the last phase of testing consisted of time measurements at the actual triathlon in Penticton, B.C. All the timing was done electronically by Racemate, the official timers for the Ironman Canada Triathlon. This timing resulted in both transition times being included in the cycling component time. To eliminate this from the cycling time each subject was instructed to start their bike cyclometer (timer) as they left the transition area at the start of the bike segment and stop it when they re-entered the transition at the end of the bike segment. The difference between official time and the cyclometer time was a measure of the transition time and could then be eliminated from calculations.  3.3 Testing Protocols  During all lab testing sessions heart rate was monitored with a Sporttester PE3000 heart rate meter. Expired gases were collected and analyzed using a Beckman Metabolic Measurement Cart interfaced with a Hewlett Packard Data Acquisition System. All maximal oxygen consumption tests for determination of TVENT were preceded with a warm-up and terminated upon volitional fatigue. The warm-up consisted of an explanation of the testing procedures followed by a physiological warm-up with all the equipment in place against a light resistance load. This provided the subjects with an opportunity to become familar with the equipment and the actual process for gathering data during the test. Volitional  31 fatigue was defined by the criteria for the particular test or by the reaching of 1%02^as defined by the following criteria: 1) The oxygen consumption ceases to increase linearly with rising workload and approaches a plateau or drops slightly, the last two values agreeing within ±2 ml•kg•min-1 . 2) Heart rate should be close to the age predicted maximum. 3) Respiratory exchange ratio is greater than 1.10.  The three sport protocols were as follows:  1) Swimming All swimming sessions were confined to a roped-off section of the indoor pool to minimize interference from other swimmers and any wave action they may cause. The water temperature in the pool at this time of year (July, August) tends to vary between 26 and 29 °C. The subjects were instructed to wear their wetsuits (they would be worn in the race) and a rope tether was attached via two belt loops on either side of a waist harness. The rope pulled at a slight angle from the bottom pulley of the frame to the waist harness. The protocol started at an initial load of 3.5 kg with 250 g increases each minute while the subject was required to maintain swimming over a marked section on the bottom of the pool. The additional measure of stroke rate was determined for each minute of the test by counting the number of strokes over a 30 second interval. This procedure was done from 15 seconds to 45 seconds of each minute to minimize the effect of workload changes on the stroke rate determination. Volitional fatigue was defined by either the subject's inability to maintain swimming over the marked position in the pool against the load or their own termination. Subjects were  32 given a warning when they started to drop back to reassume the position over the marker, if they were unable the test was terminated.  2) Cycling A mechanically braked Monarch bicycle ergometer was used with the protocol starting at an initial workload of 88.2W and having progressive increases of 22.1 W each minute. Volitional fatigue was defined by either the inability to maintain the required test cadence of 90rpm or the subject's own termination of the test. The cadence was clearly visible to the athlete at all times on the Monarch's readout screen. If the subject cycled above or below the test cadence he was instructed to resume 90 rpm and maintain it for the duration of the test.  3) Running Procedure involved a continuous (zero grade) treadmill run protocol started at an initial velocity of 5 mph and increased 0.5 mph each minute. Volitional fatigue was defined by the athlete no longer being able to maintain the treadmill velocity.  Ventilatory thresholds were determined on the basis of a disproportionate (nonlinear) increase in the excess CO 2 (EXCO2 ) elimination curve over time. This determination was made by visual inspection of three observers not directly involved in the study, who had experience in measuring threshold. All three were volunteers with considerable background in testing endurance athletes and using excess CO 2 for determining  TVENT  in a research setting. Observations  were made separately and an agreement was reached on the location of the  33 threshold. In the event of a disagreement the VE /VOZ curve was used to clarify the actual threshold point. The velocity that TvENT corresponded to was made for each component sport as follows:  1) Swimming The stroke rate (strokes.minute -1 ) measured at TvENT was converted to velocity by using a pool swim to measure the stroke length (metres•stroke -1 ). While swimming at the desired stroke rate the stroke length was measured over a marked (middle 15 m) section of the 25 m pool. This minimized the effect of the turn and the subjects were directed not to push-off hard from the ends of the pool. These two values were then multiplied together to give the velocity (see Appendix A for a sample calculation).  2) Cycling TvENT was converted to velocity using a developed regression equation  relating workload at TvENT to velocity during a 40 km cycling performance (see Appendix B for a precise explanation).  3) Running Simply the measured velocity of the treadmill at the determined threshold point (see Appendix C for sample calculation).  Each velocity was then converted to an estimated performance time based on the distance of that component of the triathlon.  34 3.4 Experimental Design and Data Analysis An analysis of variance was used to compare the dependent variables I%02mAx, TVENT  absolute, TVENT relative, and heart rate at TVENT (absolute and  relative) for each component sport. A statistical significance level of 0.01 was used in the analyses. Simple linear regression correlation coefficients were calculated to compare estimated and actual performance times (minus the transition times) for each component sport. A multiple linear regression was then performed to predict overall triathlon time from the estimated component sport times.  35 CHAPTER 4  4.0 Results and Discussion  4.1 Results  Eleven trained triathletes participated in this study prior to competition in the 1991 Ironman Canada Triathlon. Note that one subject dropped out during the cycling portion of the triathlon, consequently his data were not used in the study. Descriptive data (age, height, weight, and body fat) for the 10 subjects are presented in Table 1. Triathlon experience and training mileage is summarized in Table 2.  Table 1: Descriptive data for all subjects.  AGE  HEIGHT  WEIGHT  BODY FAT  (years)  (cm)  (kg)  (%)  5?  29.7  179.8  76.8  11.4  SD  ±3.1  ±5.6  ±7.9  ±2.2  36 Table 2: Weekly training mileage and previous triathlon experience for all participants.  WEEKLY MILEAGE (km)  NUMBER OF  SWIMMING  CYCLING  RUNNING  PREVIOUS TRIATHLONS  X  16  10  296  48  SD  ±12  ±3.5  ±73  ±8  The variables of heart rate and VO2 (ml.kg-1 .min-1 ) were measured as an absolute value at TvENT and as a relative percentage of maximum (TvENT/MAX) for the three events of swimming, cycling, and running (Table 3). An ANOVA comparing swimming, cycling, and running revealed significant differences (p<0.01) between cell means for the absolute and relative scores on heart rate and 1%02 . Tukey's HSD post hoc analysis demonstrated that the significant difference in heart rate at TvENT (HSD = 17.27, p = 0.01) was between swimming and running (difference between pair of means (diff) = 20.5) and cycling and running (diff = 17.5) while differences between cycling and  37 swimming were nonsignificant (diff = 3). This pattern was also observed in the relative heart rate (HSD = 4.75, p = 0.01) where significant results were swimming and running (diff = 5.2) and cycling and running (diff = 6.6) with nonsignificance between cycling and swimming (diff = 0.6). Similarly, VO2 at TvENT (HSD = 8.35, p = 0.01) revealed significant results for swimming and  running (diff = 16.6) and cycling and running (diff = 11.6) with a nonsignificant result between cycling and swimming (diff = 5). The final variable showing this pattern was the relative TvENT (HSD = 7.56, p = 0.01) where swimming and running (diff = 8.9) and cycling and running (diff = 14) were significant while cycling and swimming (diff = 5.1) was nonsignificant. Investigation of the variable  VO2 mAx (HSD = 8.83, p = 0.01) revealed swimming and running significant (diff = 14.7), swimming and cycling significant (diff = 10.7) and cycling and running nonsignificant (diff = 4).  38 Table 3: Heart rate and VO2 measured at  TVENT  and expressed as a percentage  of their maximal value.  SWIMMING  CYCLING  RUNNING  X  SD  )7  SD  5e  SD  147.6  ±13.2  150.6  ±13.5  168.1  ±10.3  HEART RATE AT TVENT  *It  (bpm)  *  HEART RATE RELATIVE  86.9  ±3.5  85.5  ±3.9  92.1  ±2.7  **  (%) *  V02 AT TVENT  36.2  ±5.2  41.2  ±6.7  52.8  ±5.9  Irk  (ml.kg-1 .min-1 )  *  V02 RELATIVE  74.5  ±5.9  69.4  ±5.3  83.4  ±4.9  48.6  ±5.8  59.3  ±7.5  63.3  ±5.5  irk  (%) 1102 mAx  *  (ml.kg-1 -min-1) * **  = significant difference (p < 0.01) between swimming and running = mg ' nificant difference (p < 0.01) between cycling and running = significant difference (p < 0.01) between cycling and swimming  39 Estimated times and race paces for each component event of the triathlon were determined utilizing physiological testing related to that sport. The variables stroke rate, in strokes•min -1^= 32.5, SD = ±2.5), and stroke length, in metres•stroke -1^= 1.88, SD = ±0.20), measured during the swim testing sessions were used to estimate time (min) and pace (min•mile -1 ) for this component of the triathlon. The resulting estimated and actual times (min) and paces (min•mile -1 ) for swim testing are shown in Table 4 (for more detail regarding the actual determination of the swim estimation refer to Appendix A). Since mph could not be determined directly from testing for the cycling component, a linear regression was developed using excess CO 2 at TvENT• (Table 5). This equation related mph and this variable measured during the cycling component of the 1991 English Bay triathlon. Since the resulting prediction equation (r = 0.68) was based on a short course triathlon (40 km) cycling component it was necessary to extrapolate this to mph in Ironman distance (112 mile) cycling. This was accomplished using 26 subjects who participated in both the Ironman and the English Bay triathlon. The mph in both cycling performances were correlated (r = 0.91) to estimate the necessary distance correction factor (Table 5).  40 Table 4: Estimated and actual swimming times and paces for the Ironman Triathlon.  ESTIMATED  ACTUAL  TIME  PACE  TIME  PACE  (min)  (min•mile-1)  (min)  (min•mile-1)  )7  64.2  26.7  67.1  27.97  SD  ±5.6  ±2.3  ±7.8  ±3.2  Table 5: Equations used in the determination of the cycling portion of the Ironman Triathlon.  EQUATION  r  MPH = (EX CO2 * 0.845) + 12.07  0.68  MPH (112 mile) = (40 km MPH * 0.9988) - 3.35  0.91  PREDICTED MPH (based on 40 km performance) CONVERSION FACTOR (40 km to 112 mile)  41 These equations resulted in the estimated time (min) and pace (mph) shown with the actual race values in Table 6. A more detailed explanation of this calculation for the cycling component is presented in Appendix B.  Table 6: Estimated and actual cycling times and paces for the lronman Triathlon.  ESTIMATED  5?  TIME  PACE  TIME  PACE  (min)  (mph)  (min)  (mph)  380.0  17.9  344.4  19.5  (6:20)  SD  ACTUAL  ±45.2  (5:44)  ±2.1  ±13.0  ±0.75  The run component of the triathlon utilized the velocity at TvENT on the treadmill to produce the estimated time (min) and pace (mph) shown with the actual time and pace for the triathlon in Table 7. An explanation of the development of the run estimation is presented in Appendix C.  42 Table 7: Estimated and actual run times and paces for the Ironman Triathlon.  ESTIMATED  5?  ACTUAL  TIME  PACE  TIME  PACE  (min)  (mph)  (min)  (mph)  174.5  9.1  258.6  6.4  (2:55)  SD  ±15.7  (4:19)  ±0.9  ±64.4  ±1.4  From these results linear regression equations were developed to estimate actual time for each respective component of the triathlon, and to use that to develop the overall predictor equation (Table 8).  43 Table 8: Conversion factors from estimated to actual for each component and overall time in the Ironman Triathlon.  EQUATION  r  SWIMMING  ACTUAL SWIM = 1.15' ESTIMATED SWIM - 6.75  0.83  CYCLING  ACTUAL CYCLE = 0.22 * ESTIMATED CYCLE + 262.6  0.70  RUNNING  ACTUAL RUN = 3.03 * ESTIMATED RUN - 267.1  0.76  OVERALL  ACTUAL OVERALL = (-3.58 * EST. SWIM) + (-0.10 * EST. CYCLE) + (3.76 * EST. RUN)+ 291.35  0.89  44 4.2 Discussion Despite the controversy and debate with respect to nomenclature, mechanisms, detection, and interpretation (Brooks, 1985; Davis, 1985), there is little dispute over the strong relationships that have been demonstrated between anaerobic threshold and performance. While much of this research has focused on running studies (Farrell et al., 1979; Morgan et al., 1989; Maughan and Leiper, 1983), other investigations involving cycling, swimming, cross-country skiing, etc. have produced similarly strong results (Mader et al., 1978; Coyle et al., 1991). Most of this performance research has been confined to single sport endurance events. Now, with the emergence of multiple sport endurance contests (triathlon, duathlon), the ability of ventilatory threshold (T vEN T) to characterize performance in this new setting is questionable. Not only must measurements characterize the component sports but they must also take into account the prior exercises which increase the physiological demands of performing subsequent events (Kreider et al., 1988). The purpose of this study was to investigate if TvENT measured in swimming, cycling and running was capable of predicting triathlon performance (time) success. The subject pool used in this study appears to be typical of other studies in terms of the descriptive characteristics of the triathletes (Burke and Read, 1987; Roalstad et al., 1987; Ireland and Micheli, 1987). The triathletes tend to be older than comparable single sport endurance performers (Kreider et al., 1988; Kohrt et al., 1987a; O'Toole et al., 1987). This is probably a factor of the "newness" of the sport and as it grows the population of triathletes will be younger. As well, most athletes come to triathloning after competing in one of the single sports, so it would be expected that they are older. In terms of the other descriptive characteristics of height, weight, and body fat, they are  45 undistinguished from other single sport athletes (O'Toole et al., 1987; Roalstad, 1989). The weekly mileage for these subjects is similar to other triathletes with respect to swimming (typical = 10.5 km) and cycling (typical = 300 km) while the running is low (typical = 72 km) (O'Toole et al., 1989). This low weekly run mileage will be discussed in the context of how it may relate to the results found in this study. As with other studies, there is a wide interindividual variability with respect to training mileage in the three component sports (O'Toole, 1989). The values for heart rate and oxygen consumption are consistent with respect to other research (Dengel et al., 1989). The hierarchy of 1%02mAx values is typical (run>cycle>swim) but the absolute scores are less than comparable single sport athletes (Astrand and Rodahl, 1977). This may be reflective of the reduced training volume in any one area and/or specific adaptations that triathletes undergo in training related to their sport (Kohrt et al., 1987b). The absolute scores demonstrate the expected pattern between running and swimming, and running and cycling with the exception of run-cycle I% 02mAx • While the differences between run-cycle were nonsignificant on this variable the drop from running to cycling is consistent with previous triathlon research. The difference of 6.7% compares well to the work of Kohrt et al. (1987a) and O'Toole et al. (1987) where the maximums are 3 to 6% less in cycling. This drop of 3 to 6% for I%02 „,„u( from running to cycling is somewhat less than highly trained runners would show (9 to 11%) (Roalstad, 1989). The hierarchy tends to demonstrate the highest^values for treadmill running, with cycling approximately 90% (Faulkner et al., 1971; Miyamura et al., 1978) and swimming approximately 80% (Holmer et al., 1974; Magel et al., 1975) of the running value. This relationship is altered in athletes who are trained cyclists or  46 swimmers. Highly trained cyclists may equal or exceed their treadmill running VO2MAx on cycle ergometry testing (Kohrt et al., 1989) likely by virtue of a more sport specific testing procedure. Several studies (Roalstad, 1989; Hagberg et al., 1978; Withers et al., 1981) state that triathletes most resemble cyclists. While the triathletes generally do not equal or exceed their treadmill run VO2 , like elite cyclists there is a training specificity, resulting in less of a gap between the run and cycle VO2 mAx . This relationship is not unique to this study (O'Toole et al., 1987; Kohrt et al., 1987b) and may reflect a specific adaptation to triathlon type training versus single sport training. For the variable V02 „„Ax only swimming and cycling illustrated the hypothesized result, with the cycling 1102mAx significantly greater than the swimming V02MAX. This study demonstrated a swim V02 AfAx value at 77% of the running VO2 hfAx which is typical of trained runners tested in both modes. Since highly trained swimmers may reduce the difference between swimming and running, it was thought triathletes might demonstrate this result, although to a lesser degree. Studies by Dengel et al. (1986) and Kohrt et al. (1987b) found a reduction in the difference, while Kohrt et al. (1989) did not. Thus, the literature tends to be equivocal on this point. The similar finding to Kohrt et al. (1989) may indicate that the relatively low weekly volume of training (elite swimmers' daily training distance is about equal to triathletes' weekly swim training distance) was insufficient to develop training specificity as seen in cycling. The variables HR and VO2 at TvENT were similar (p > 0.05) for swimming and cycling. This might be accounted for by a slightly low estimated threshold in cycling (Kohrt et al., 1989). The mean VO2mAx value is typical of other studies which have ranged from 54.4 ml.kg -1 .min -1 (Kohrt et al., 1987b) to 66.7 ml-kg -1 .min -1 (O'Toole et al., 1987) for men on cycle ergometry testing. The low threshold value might be  47 accounted for by the type of training most subjects performed. Most participants followed a high volume (about 296 km•week -1 ) format with very little "quality" training. To improve TvENT most studies indicate a high intensity (near or slightly above threshold) type protocol (Pate and Branch, 1992). O'Toole (1989) documented a low amount of interval training ( 20%) on the bike by a large survey of triathletes. Thus, since most subjects have trained based on volume for cycling as opposed to higher intensity, the ability of TvENT to separate differences in levels of performance is lessened. Subjects are working at a lower percentage of maximum in cycling than the other two component sports as demonstrated in these results. On the relative measures of HR and VO2 it was hypothesized that there would be nonsignificant differences between swimming, cycling and running. This was only true for swimming and cycling with running being significantly greater than either of the other component sports. The high percentage may reflect the previous training of these athletes. Approximately 75% of triathletes have a running background (Ireland and Micheli, 1987) and the higher TvENT may reflect a relationship between experience and performance. Certainly, studies have shown years of experience to be highly correlated with performance (Magel et al., 1975; O'Toole, 1989). Kohrt et al. (1989) suggest that this lower percentage found in cycling may indicate that triathletes have a greater potential to improve in cycling than in running. This lower TvENT in cycling may reflect the need to devote a greater proportion of training or at least an increase in the intensity of the training in order to achieve the similar training status as found in running. In this study, the one subject from a cycling background had the highest cycling TvENT estimated percentage. This may also simply reflect large interindividual variability, coupled with the small sample size  48 of the present study. This condition may result in extreme scores influencing results more than if a large sample study were employed. The results of swim testing produced an estimate 2.9 minutes faster than actual swim time for that component of the Ironman triathlon. This result (r=0.83) indicates that 69% of the actual time variability was accounted for by the estimated time value. This result seems consistent with the TvENT concept. TvENT assumes participants will set a maximum pace that could be sustained  indefinitely, without inducing metabolic acidosis (Coyle et al., 1988; Williams and Cavanagh, 1987; Kumagai, 1982). Since this is the first event, sport intrinsic fatiguing factors (fluid, fuel, etc.) are not likely to severely limit pace (O'Toole et al., 1989). As well, the low number of confounding factors would suggest that the correlation between actual and estimated swim performance should be the highest of the three component sports, and indeed this was the case. Thus, this physiological variable, explaining 69% of the variability, stresses the importance of other factors, particularly mechanical efficiency which may also affect performance (Kohrt et al., 1989). The study of Toussaint (1990), comparing elite swimmers to triathletes, found that they both had similar ability to do work but the differences in technique separated the triathletes from the more skilled swimmers. The inefficient swimmer may expend more than twice the calories during submaximal swimming compared to the efficient swimmer. This would lead to a generalized and localized fatigue (Holly et al., 1986). This is paramount in a triathlon when considering the majority of the competitive distance is still to come and it is necessary to conserve energy. From the perspective of competitive experience, the subjects were well prepared (average of 16 triathlons completed) and less likely to push pace this early in the event. Only two subjects were faster (both 4 minutes) than their estimated time, one being the strongest swimmer in the study.  49 Cycling results were much more confusing, with the estimated time being 36 minutes slower than actual time. The correlation between estimated and actual time resulted in 49% (r=0.70) of the variance being explained by the estimator. It was expected that all estimated results would be faster than actual times. Given the increasing influences of intrinsic factors (fuel, fluid, thermoregulatory mechanisms, etc.) and extrinsic factors (heat, humidity, terrain, etc.) which would reduce the ability to maintain speed and the inability of TvENT to account for these, it seemed a reasonable assumption. There are several possible explanations for this observation. Firstly, it may reflect an inadequacy in the chosen methodology. The inability to measure speed directly in the testing environment necessitated the indirect determination using a short course triathlon performance. Since many of the factors that adversely. effect performance will also be present in a short course triathlon (however to a lesser degree) the cycling equation developed will in some way account for this variability and reduce the ideal speed determined by a solely lab developed estimation. As well, the extrinsic factors of terrain, weather, etc. will also be present, confounding the determination of cycling pace. Another possible explanation is that the methodology employed underestimates the true threshold. The few studies that have looked at triathletes TvENT relative to performance have been unsuccessful with correlations between TvENT and bike finish times in an Ironman triathlon (r =-0.26)(0 Toole et al., 1989). While some 1  of the reduced relationship from swim to cycle is methodological, considerable variability is due to factors not present while swimming (terrain, heat, etc.). Also the cumulative effect of time magnifies the factors that were evident in the swim. Certainly, one of the conclusions of the low correlation studies has been that the upper limit of pace for cycling is being set by other factors as opposed to physiological considerations represented by TvENT (O'Toole et al., 1989). The  50 emergence of other factors was evident in the bike portion of the triathlon, as one subject dropped out due to dehydration. By this time in the triathlon, the cumulative effects of time, distance, and heat are beginning to take their toll in terms of dehydration, electrolytic abnormalities, and exhaustion (Laird, 1989). Although the swim is a relatively small component of the total event, it may have pronounced effects on subsequent performance. Kreider et al. (1988) found that prior swimming would increase core temperature at the onset of cycling. Thusly, thermoregulatory and cardiovascular adjustments occur sooner than they would with no prior swimming. In this study the lower correlation may also be reflective of a very homogeneous sample (344.4 ±13 minutes), with respect to actual cycling time. In addition, the idea of maintenance of efficiency as the ultimate determinant of triathlon success (Kreider et al., 1988) may be less physiological (TVENT) and more biomechanical (technique), accounting for a considerable portion of the variability. The run results produced the greatest disparity between estimated and actual times with an average of one hour and 24 minutes difference. As would be expected the estimated time was faster than the actual time. There was considerable variability (258.6 ±64.4 minutes) in run performances with the estimated time able to explain 58% of the variability in actual time. The effects of prior activity in swimming and cycling (subjects have been active between about 6 and 7 hours) are likely to be even more pronounced. The large difference between estimated and actual times provides support for other factors setting pace limits. While these intrinsic and extrinsic factors are present, another observation may explain some of the decrease in run race pace. The difference in combined swim and cycle time from fastest to slowest subject is one hour and one minute while the combined swim, cycle, and run difference from fastest to slowest is 3 hours and 48 minutes. This is a considerable amount  51 of variability to be added during the run segment. This difference is largely accounted for by the inability of slower subjects to maintain pace. The welltrained runners are better able to maintain race paces that approximate their training pace, while the less well-trained athletes demonstrate an inability to maintain training pace (O'Toole, 1989). This may be the result of these subjects having difficulty maintaining a constant TvENT estimated pace for some portion of the triathlon prior to running. The variable terrain found at the Ironman Canada Triathlon would also be an interfering factor. The considerable elevation changes during the bike portion produce extreme speed variability making it difficult to maintain a steady work output. Roalstad et al. (1987) found triathletes with the least fluctuations in heart rates tended to have better finishing times than those with wide variability. The terrain creates a condition that must ultimately place some constraints on the ability of TvENT to estimate during the bike portion of the triathlon. This result is certainly consistent with the concept of TvENT and the expected result if a subject were to perform for any length of time  above the defined limit of aerobic ability (Farrell et al., 1979). The homogeneous cycling performances may also support this result, if the slower cyclists' actual performances were faster than expected. The greatest difference between actual and estimated time was in fact between the slowest cyclists with all performances much faster than estimated (by an average of 74 minutes). Therefore, there is some evidence for cycling above TvENT with a subsequent adverse effect on running. This would explain some of the wide variability seen in the slower subjects. Another factor that may account for variability in running is the low mileage done for this component in training. O'Toole (1989) found that of the subjects who did not finish the two main reasons were; (1). inappropriate race strategy and (2). too little training (particularly long distance bike rides or long runs). Some coaches suggest that training in swimming and cycling  52 provides a "crosstraining" benefit that would offset the low mileage training in running. Previous research suggests that this is not true. Swim training (Gergley et al., 1984; Magel et al., 1975) and cycle training (Town and Sinning, 1982; Pechar et al., 1974) have both failed to demonstrate improvement in run training. Thus, a lack of sufficient run mileage may also represent some of the variability. When all three component sports are taken into account in a multiple linear regression a strong correlation to the actual overall time (r=0.89) resulted. Thus, the combined variability represented by each of the component sports is able to predict overall time better, accounting for 78% of variability, than its individual component sport. This also suggests that each sport's variability represents an important part of the variability seen in a combined sport like triathlon. These results support the importance of an understanding of these other factors influencing performance. Successful triathlon performance appears to be impacted upon by the ability to maintain thermal and mechanical efficiency throughout the event (Kreider et al., 1988). Most of the metabolic abnormalities that develop are time dependent and cumulative as the triathlon progresses. An example of this time effect is found in the research of Laird (1989) where in going from a short course triathlon to Ironman distance, the average percent dehydration increased from 1.7 to 3.7%. Certainly, absolute fluid loss increases over longer distances. While the most marked differences are seen in the run, they are more the effect of the total duration as opposed to the marathon itself (Farber et al., 1991). Relationships reflect this emergence of pace determining factors if we consider that in the study of Rhodes and McKenzie (1984) an r=0.94 was found between estimated and actual time. While this study, using the exact same equipment and methodology found r=0.76 for the marathon run component of the triathlon. In comparison to other endurance activities  53 indications of dehydration were much greater in Ironman triathletes (van Rensburg et al., 1986). The most common reason for requiring medical attention at the Hawaii Ironman is dehydration (Hiller et al., 1987; Laird, 1987). Sawka et al. (1985) recognized that dehydration accompanied with thermal stress may significantly decrease work output. In terms of electrolytes, incidence of hyponatraemia are directly related to race distance (O'Toole et al., 1989). Saltdepletion heat exhaustion is characterized by fatigue, nausea, muscle cramps, etc. (Laird, 1989) This combined with dehydration and increased sweat rate will have profound results on fatigue. Certainly, nutritional replacement has been well documented with respect to its role in endurance performance (Costill, 1988). Emerging research suggests there are likely extra considerations that need to be made in events of the duration of ultraendurance triathlons. Typical carbohydrate replacement protocol, like that done in marathoning, may not be a sufficient rate of replacement for these increased duration events (Applegate, 1989). As well, the use of solid food is well accepted in ultraendurance triathlons (Applegate, 1989). Further research is needed with respect to the appropriate protocols for nutritional replacement during Ironman distance triathlons to minimize the adverse effects of running out of energy. Despite the other factors interfering with the actual estimation of performance, the underlying mechanisms support the concept of TvENT• Ultimately, the triathlon is a contest requiring a series of endurance events to be done sequentially at optimal pace without creating fatigue that will hinder performance in the next event (O'Toole et al., 1989). TvENT is designed to identify the point where work may be performed for indefinite durations as energy requirements are being supplied predominantly through unlimited aerobic energy sources, while waste products are being adequately removed (Anderson and Rhodes, 1989). While there are many different methods for determining  54 TVENT, excess carbon dioxide (CO 2 ) combines a strong mechanism base with a strong relationship to performance research. Carbon dioxide production will increase with exercise due to the metabolism of fats and carbohydrates until the TVENT breakaway point at which time there will be a secondary (nonmetabolic) increase in CO 2 due to the buffering of lactic acid by bicarbonate (Anderson and Rhodes, 1989). This provides for a direct tie between the generation of excess CO2 and the increased production of lactic acid which is strongly related to fatigue (Sjodin and Svedenhag, 1985). Studies suggest that 90-94% of the lactic acid produced is immediately buffered by the bicarbonate buffering system (Wasserman et al., 1986). Thus, excess CO 2 is representing the magnitude of lactic acid production through the glycolytic pathways and the organism's buffering capacity (Anderson and Rhodes, 1991) and will continue to increase as long as the rate of lactic acid production is increasing (Loat and Rhodes, 1991). While excess CO 2 changes do not represent the actual occurrence of blood lactate accumulation they do seem to track the changes taking place (Anderson and Rhodes, 1989). This "tracking effect" demonstrates the strong relationship (r=0.92) between excess CO 2 and blood lactate changes (Langill and Rhodes, 1992). It is suspected that this difference is due to the excess CO 2 being freely diffusable across the cell membrane while the lactate molecule experiences a "translocation hindrance" delaying increases in the blood (Stainsby, 1986). It is suggested that this hindrance represents a delayed lactate transport caused by low muscle membrane lactate permeability and a change in capacitive effects in muscle concentration of lactic acid (Stainsby, 1986). This resulting delay in blood lactate accumulation may in fact suggest that excess CO 2 more directly approximates changes in intramuscular lactate production and accumulation (Volkov et al., 1975; Issekutz and Rodahl, 1961). Combined with this support for excess CO 2 's ability to define a critical intensity of lactate accumulation, the  55 performance research supports similarly strong ties. Hearst and Rhodes (1982) found exercising at TVENT to be an appropriate intensity to maintain a significantly low blood lactate concentration while exercising at an intensity 1 km•hr-1 above TVENT elevated the lactate. The study of Loat and Rhodes (1991) found excess CO 2 defined a TVENT that could be sustained for one hour of cycling without significant elevations in blood lactate. Endurance, as it relates to competition, can be thought of as the capacity to achieve and maintain a high average speed over the distance of the race. It is with this in mind, that the concept of TVENT has emerged to try to identify this point of maximum sustainable pace. While other factors such as substrate depletion, temperature regulation, fluid and electrolyte balance will contribute to fatigue and make it difficult to represent a sustainable work rate (Carnevale and Gaesser, 1991) there is still underlying physiological mechanisms that TVENT appears to represent. Between 49 and 69% of the individual component sport variability and 78% of the total triathlon variability is accounted for by these measurements. Despite adverse conditions having reduced the ability of TVENT to determine pace there still remains a connection between performance and threshold measurements that extends from single to multiple endurance sport contests. Triathlon is a comparatively new sport and considerably more research is needed into the determinants of success. Particular emphasis should focus on understanding the unique physiological adaptations that take place in triathletes relative to single sport athletes and examining the role the various factors play in decreasing performance. However, research should not dismiss the use of TVENT in this sport setting.  56 CHAPTER 5  5.0 Summary and Conclusions  5.1 Summary  In theory, there is considerable evidence to suggest that it is possible to estimate athletic performance in endurance sports. These sports are contested by athletes who will, regardless of their competitive ability, try to sustain a race pace that maximizes their physiological capacities. The use of lab data to estimate such performance is an extremely attractive concept since it could offer a more scientific basis for training and selection of athletes. Several physiological factors (^Aux , economy of motion, and anaerobic threshold) have been investigated with varying degrees of success. Among these, the anaerobic threshold seems to have the most potential for characterizing endurance sport performance. The term "anaerobic threshold" has undergone considerable change in the years since its initial development by Wasserman and Mcllroy (1964). Its measurement based on the ventilatory measure excess CO 2 has produced strong relationships to endurance sport performance in running and cycling (Rhodes and McKenzie, 1984; Loat and Rhodes, 1991; Hearst and Rhodes, 1982). It has also been shown to have a strong relationship to the lactate threshold (Anderson and Rhodes, 1991; Langill and Rhodes, 1992) and possibly the intracellular production and accumulation of lactate (Issekutz and Rodahl, 1961). Anaerobic thresholds have been used in sports like swimming, cycling, and particularly running to develop performance based relationships with some success. The comparatively new sport of triathloning has presented a new set of  57 questions with regard to estimating performance. Due to the multiple modes of competition within the same contest it is not known whether the same features governing single sports also apply to triathlon. Problems of performing sequential events and extreme duration pose restrictions that must be considered. It was the purpose of this investigation to determine if ventilatory threshold measurements in swimming, cycling, and running could be extended to triathlon in an effort to characterize performance. This study found the estimated single component sport times to correlate to actual triathlon times in swimming, cycling, and running with r=0.83, 0.70, 0.76 respectively. Overall triathlon performance found r=0.89 between the combined three sport estimated times and the actual total time. Linear regression equations were produced such that actual time could be calculated from estimated times (in minutes) with the following results: actual swim = 1.15 * estimated swim - 6.75 actual cycle = 0.22 * estimated cycle + 262.6 actual run = 3.03 * estimated run - 267.1 actual overall = (-3.58 * est. swim) + (-0.10 * est. cycle) + (3.76 * est. run)+ 291.35  These findings demonstrated an underlying physiological base for triathlon performance that was represented by ventilatory threshold. Unfortunately, other factors resulting from the cumulative effect of time, heat, and terrain diminish the ability to use ventilatory threshold as a predictor.  5.2 Conclusions 1) Ventilatory threshold accounted for 78% of the variability in pace in an ultraendurance sport setting. 2) This ability is compromised by other factors that confound the ability to estimate pace.  58  5.3 Recommendations 1) It would be beneficial to make some measurements of pace at the beginning, middle, and end of each component sport during the race to see how well subjects actually do maintain constant pace. 2) The development of a correction factor that could be carried-over from one component sport to the next to account for cumulative effects is needed. 3) A more direct methodology for estimating the cycling component time would likely improve this weakest of the correlations. 4) Instead of comparing the TvENT determined speeds to an actual triathlon, compare then to a simulated laboratory triathlon to eliminate some of the extraneous variables. 5) At this stage most of the triathlon research is equivocal, so continued research in this area is necessary to gain a better idea of those factors contributing to triathlon success. 6) Still more research is • needed into anaerobic threshold, particularly the underlying intracellular mechanisms.  59 BIBLIOGRAPHY Acevedo EO, Goldfarb AH. Increased training intensity effects on plasma lactate, ventilatory threshold, and endurance. Medicine and Science in Sports and Exercise 21(5): 563-568, 1989. Albrecht TJ, Foster VL, Dickinson AL, DeBever JM. Triathletes: exercise parameters measured during bicycle, swim bench, and treadmill testing. Medicine and Science in Sports and Exercise 18 (suppl): S86, 1986. Anderson GS, Rhodes EC. A review of blood lactate and ventilatory methods of detecting transition thresholds. Sports Medicine 8(1): 43-55, 1989. Anderson GS, Rhodes EC. The relationship between blood lactate and excess CO2 in elite cyclists. Journal of Sports Sciences 9: 173-181, 1991. Applegate E. Nutritional concerns of the ultraendurance triathlete. Medicine and Science in Sports and Exercise 21(5 suppl): S205-S208, 1989. Astrand PO, Rodahl K. Textbook of Work Physiology, McGraw-Hill Book Company, New York, 1977. Aunola S, Alanen E, Marniemi J, Rusko H. The relation between cycling time to exhaustion and anaerobic threshold. Ergonomics 33(8): 1027-1042, 1990. Aunola S, Rusko H. Reproducibility of aerobic and anaerobic thresholds in 20-50 year old men. European Journal of Applied Physiology 53: 260-266, 1984. Bailey S, Pate RR. Feasibility of improving running economy. Sports Medicine 12: 228-236, 1991. Brooks GA. Anaerobic threshold: review of the concept and directions for the future. Medicine and Science in Sports and Exercise 17(1): 22-31, 1985. Brooks GA. The lactate shuttle during exercise and recovery. Medicine and Science in Sports and Exercise 18(3): 360-368, 1986. Burke LM, Read RSD. Diet patterns of elite Australian male triathletes. The Physician and Sportsmedicine 15(2): 140-155, 1987. Caiozzo VJ, Davis JA, Ellis JF, Azus JL, et al. A comparison of gas exchange indices used to detect the anaerobic threshold. Journal of Applied Physiology 53(5): 1184-1189, 1982.  60 Carnevale TJ, Gaesser GA. Effects of pedalling speed on the power-duration relationship for high-intensity exercise. Medicine and Science in Sports and Exercise 23(2): 242-246, 1991. Chatard JC, Padilla S, Cazorla G, Lacour JR. Influence of body height, weight, hydrostatic lift and training on the energy cost of the front crawl. The New Zealand Journal of Sports Medicine 13(3): 82-84, 1985. Chatard JC, Lavoie JM, Lacour JR. Analysis of determinants of swimming economy in front crawl. European Journal of Applied Physiology 61: 88-92, 1990. Clausen JP, Klausen K, Rasmussen B, Trap-Jensen J. Central and peripheral circulatory changes after training of the arms or legs. American Journal of Physiology 225(3): 675-682, 1973. Clode M, Clark T, Campbell E. The immediate CO 2 storage capacity of the body during exercise. Clinical Science 32: 161-165, 1967. Coen B, Schwartz L, Urhausen A, Kindermann W. Control of training in middleand long-distance running by means of the individual anaerobic threshold. International Journal of Sports Medicine 12(6): 519-524, 1991. Conconi F, Ferrari M, Ziglio PG, Droghetti P, Codeca L. Determination of the anaerobic threshold by a noninvasive field test in runners. Journal of Applied Physiology 52: 869-873, 1982. Conley D, Krahenbuhl G. Running economy and distance running performance of highly trained athletes. Medicine and Science in Sports 12: 357-360, 1980. Costill DL, Thomason H, Roberts E. Fractional utilization of the aerobic capacity during distance running. Medicine and Science in Sports 5(4): 248-252, 1973. Costill DL, Miller JM. Nutrition for endurance sport: carbohydrate and fluid balance. International Journal of Sports Medicine 1: 2-14, 1980. Costill DL, Fink WJ, Pollock ML. Muscle fiber composition and enzyme activities of elite distance runners. Medicine and Science in Sports 8: 96-100, 1976. Costill DL, Winrow E. Maximal oxygen intake among marathon runners. Archives of Physical Medicine and Rehabilitation 51: 317-320, 1970.  61 Costill DL, Kovaleski J, Porter D, Kirwan J, et al. Energy expenditure during front crawl swimming; predicting success in middle-distance events. International Journal of Sports Medicine 6: 266-270, 1985. Costill DL, Thomas R, Robergs RA, Pascoe D, et al. Adaptations to swim training: influence of training volume. Medicine and Science in Sports and Exercise 23(3): 371-377, 1991. Costill DL, Hinrichs D, Fink WJ, Hoopes D. Muscle glycogen depletion during swimming interval training. Journal of Swimming Research 4: 15-18, 1988. Costill DL. Physiology of marathon running. Journal of the American Medical Association 221: 1024-1029, 1972. Coyle EF, Coggan AR, Hopper MK, Walters TJ. Determinants of endurance in well-trained cyclists. Journal of Applied Physiology 64(6): 2622-2630, 1988. Coyle EF, Feltner ME, Kautz SA, Hamilton MT, et al. Physiological and biomechanical factors associated with elite endurance cycling performance. Medicine and Science in Sports and Exercise 23(1): 93-107, 1991. Craig AB, Pendergast DR. Relationships of stroke rate, distance per stroke, and velocity in competitive swimming. Medicine and Science in Sports 11: 278283, 1979. Daniels J, Scardina N, Hayes J, Foley P. Variations in 1%0 2 submax during treadmill running. Medicine and Science in Sports and Exercise 16: 108, 1984. Daniels J. Physiological characteristics of champion male athletes. Research Quarterly 45: 342-348, 1974. Davies CTM, Thompson MW. Aerobic performance of female marathon and male ultramarathon athletes. European journal of Applied Physiology 41: 233-245, 1979. Davis HA, Basset J, Hughes P, Gass G. Anaerobic threshold and lactate turnpoint. European Journal of Applied Physiology 50: 383-392, 1983. Davis JA, Vodak P, Wilmore JH, Vodak J, Kurtz P. Anaerobic threshold and maximal aerobic power for three modes of exercise. Journal of Applied Physiology 41: 544-550, 1976.  62 Davis JA, Frank MH, Whipp BJ, Wasserman K. Anaerobic threshold alterations caused by endurance training in middle-aged men. Journal of Applied Physiology 46: 1039-1046, 1979. Davis JA. Anaerobic threshold: review of the concept and directions for the future. Medicine and Science in Sports and Exercise 17(1): 6-18, 1985. Dengel DR, Flynn MG, Costill DL, Kirwin JP, et al. Metabolic determinants of success during triathlon competition. Medicine and Science in Sports and Exercise 18 (suppl): S87, 1986. Dengel DR, Flynn MG, Costill DL, Kirwan JP. Determinants of success during triathlon competition. Research Quarterly for Exercise and Sport 60(3): 234238, 1989. diPrampero PE, Pendergast DR, Wilson DW, Rennie DW. Energetics of swimming in man. Journal of Applied Physiology 37: 1-5, 1974. Farber HW, Schaefer EJ, Franey R, Grimaldi R, Hill NS. The endurance triathlon: metabolic changes after each event. and during recovery. Medicine and Science in Sports and Exercise 23(8): 559-565, 1991. Farrell PA, Wilmore JH, Coyle EF, Billing JE, Costill DL. Plasma lactate accumulation and distance running performance. Medicine and Science in Sports 11(4): 338-344, 1979. Faulkner JA, Roberts DE, Elk RL, Conway J. Cardiovascular responses to submaximum and maximum cycling and running. Journal of Applied Physiology 30: 457-461, 1971. Foster C, Daniels JT, Yarbrough RA. Physiological and training correlates of marathon running performance. Australian Journal of Sports Medicine 9: 5861, 1977. Gergley TJ, McArdle WD, DeJesus P, Toner MM, et al. Specificity of arm training on aerobic power during swimming and running. Medicine and Science in Sports and Exercise 16(4): 349-354, 1984. Gollnick P, Bayly W, Hodgson D. Exercise intensity, training, diet, and lactate concentration in muscle and blood. Medicine and Science in Sports and Exercise 18(3): 334-340, 1986. Hagberg JM, Giese MD, Schneider RB. Comparison of the three procedures for measuring 11021Ax in competitive cyclists. European Journal of Applied Physiology 39: 47-52, 1978.  63 Harrison TR, Pitcher C. Studies in congestive heart failure II: the respiratory exchange during and after exercise. The Journal of Clinical Investigation 8: 291, 1930. Hearst WE, Rhodes EC. Relationship between anaerobic threshold, excess CO 2 and blood lactate in elite marathon runners. Canadian Journal of Applied Sport Science 7: 230, 1982. Hiller WDB, O'Toole ML, Fortess EE, Laird RH, et al. Medical and physiological considerations in triathlons. American Journal of Sports Medicine 15(2): 164167, 1987. Hiller WDB. Dehydration and hyponatremia during triathlons. Medicine and Science in Sports and Exercise 2(5 suppl): S219-S221, 1989. Holly RG, Barnard RJ, Rosenthal M, Applegate E, Pritikin N. Triathlete characterization and response to prolonged strenuous competition. Medicine and Science in Sports and Exercise 18(1): 123-127, 1986. Holmer I, Lundin A, Eriksson BO. Maximum oxygen uptake during swimming and running by elite swimmers. Journal of Applied Physiology 36: 711-714, 1974. Holmer I. Energy cost of arm stroke, leg kick and the whole stroke in competitive swimming styles. European Journal of Applied Physiology 33: 105-118, 1974. Holmer I. Physiology of swimming man. Exercise and Sport Sciences Reviews 7: 87-123, 1979. Hughson RL, Green HJ. Blood acid - base and lactate relationships studied by ramp work tests. Medicine and Science in Sports and Exercise 14(4): 297302, 1982. Hughson RL, Weisiger KH, Swanson GD. Blood lactate concentration increases as a continuous function in progressive exercise. Journal of Applied Physiology 62(5): 1975-1981, 1987. Hultman E, Sahlin K. Acid-base balance during exercise. Exercise and Sport Sciences Review 8: 41-128, 1980. Ireland ML, Micheli U. Triathletes: biographic data, training, and injury patterns. Annals of Sports Medicine 3: 117-120, 1987. lssekutz B, Rodahl K. Respiratory quotient during exercise. Journal of Applied Physiology 16: 606-610, 1961.  64 Jensen RK, Tihanyi J. Fundamental studies of tethered swimming. In Landry, Orban (eds.) Biomechanics of Sports and Kinanthropometry, pp. 143-148, Symposia Specialists, Miami, 1978. Jones N, Ehrsam R. The anaerobic threshold. Exercise and Sports Sciences Review 10: 49-83, 1982. Jones N. Hydrogen ion balance during exercise. Clinical Science 59: 85-91, 1980. Karlsson J. Lactate and phosphagen concentrations in working muscles in man. Acta Physiologica Scandinavia 82(suppl): 258, 1971. Knuttgen H. Oxygen debt, lactate, pyruvate and excess lactate after muscular work. Journal of Applied Physiology 17(4): 639-644, 1962. Knuttgen HG, Saltin B. Muscle metabolites and oxygen uptake in short-term submaximal exercise in man. Journal of Applied Physiology 32: 690-694, 1972. Kohrt WM, O'Connor JS, Skinner JS. Effects of reduced training on the physiological profile of triathletes. Medicine and Science in Sports and Exercise 19 (suppl): S48, 1987b. Kohrt WM, Morgan DW, Bates B, Skinner JS. Physiological responses of triathletes to maximal swimming, cycling, and running. Medicine and Science in Sports and Exercise 19(1): 51-55, 1987a. Kohrt WM, O'Connor JS, Skinner JS. Longitudinal assessment of responses by triathletes to swimming, cycling, and running. Medicine and Science in Sports and Exercise 21(5): 569-575, 1989. Krebs PS, Zinkgraf S, Virgilio SJ. Predicting competitive bicycling performance with training and physiological values. Journal of Sports Medicine 26: 323330, 1986. Kreider RB, Boone T, Thompson WR, Burkes S, Cortes CW. Cardiovascular and thermal responses of triathlon performance. Medicine and Science in Sports and Exercise 20(4): 385-390, 1988. Kumagai S, Tanaka K, Matsuura Y, Matsuzaka A, et al. Relationships of the anaerobic threshold with the 5km, 10km, and 10mile races. European Journal of Applied Physiology 49: 13-23, 1982.  65 Kyle CR, Mastropaolo J. Predicting racing bicyclist performance using the unbraked flywheel method of bicycle ergometry. Biomechanics of Sport and Kinanthropometry 6: 211-219, 1978. LaFontaine TP, Londeree BR, Spath WK. The maximal steady state versus selected running events. Medicine and Science in Sports and Exercise 13(3): 190-192, 1981. Laird RH. Medical care at ultraendurance triathlons. Medicine and Science in Sports and Exercise 21(5 suppl): S222-S225, 1989. Langill RH, Rhodes EC. Comparison of the lactate and ventilatory response during a progressive intensity test. Australian Journal of Science and Medicine in Sport. (in press). LePere CB, Porter GH. Cardiovascular and metabolic response of skilled and recreational swimmers during running and swimming. In Taylor (ed.) Application of Science and Medicine to Sport, pp. 234-247, Thomas, Springfield, 1975. Loat CER, Rhodes EC. Comparison of the lactate and ventilatory thresholds during prolonged work. University of British Columbia.(Unpublished Thesis). MacDougall JD. The anaerobic threshold: its significance to the endurance athlete. Canadian Journal of Applied Sports Sciences 2: 137-140, 1977. Mader A, Heck H, Hollmann W. Evaluation of lactic acid anaerobic energy contribution by determination of postexercise lactic acid concentration of ear capillary blood in middle-distance runners and swimmers. In Landry, Orban (eds.) Exercise Physiology, pp. 187-200, Symposia Specialists, Miami, 1978. Magel JR, Foglia GF, McArdle WD, Gutin B, et al. Specificity of swim training on maximum oxygen uptake. Journal of Applied Physiology 38(1): 151-155, 1975. Magel JR, McArdle WD, Toner M, Delio DJ. Metabolic and cardiovascular adjustment to arm training. Journal of Applied Physiology 45: 75-79, 1978. Malhotra MS, Verma SK, Gupta RK, Khanna GL. Physiological basis for selection of competitive road cyclists. Journal of Sports Medicine 24: 49-57, 1984. Maughan R, Leiper J. Aerobic capacity and fractional utilization of aerobic capacity in elite and non-elite male and female marathon runners. European Journal of Applied Physiology 52: 80-87, 1983  66 McArdle WD, Glaser RM, Magel JR. Metabolic and cardiorespiratory response during free swimming and treadmill walking. Journal of Applied Physiology 30: 733-738, 1971. McArdle WD, Magel JR, Delio DJ, Toner M, Chase JM. Specificity of run training on ki2mAx and heart rate changes during running and swimming. Medicine and Science in Sports 10(1): 16-20, 1978. McLellan TM, Jacobs I. Active recovery, endurance training, and the calculation of the individual anaerobic threshold. Medicine and Science in Sports and Exercise 21(5): 586-592, 1989. Miller FR, Lindholm S, Manredi TG. Anaerobic threshold and 15km cycling performance. Medicine and Science in Sports and Exercise 17(2): 217, 1985. Montpetit RR, Leger LA, Lavoie J-M, Cazorla G. l0 2 peak during free swimming using the backward extrapolation of the 02 recovery curve. European Journal of Applied Physiology 47: 385-391, 1981. Morgan DW, Baldini FD, Martin PE, Kohrt WM. Ten kilometer performance and predicted velocity at VO2 MAX among well-trained male runners. Medicine and Science in Sports and Exercise 21(1): 78-83, 1989. Morgan DW, Craib M. Physiological aspects of running economy. Medicine and Science in Sports and Exercise 24(2): 456-461, 1992. Morgan DW, Martin P. Effects of stride length alteration on race-walking economy. Canadian Journal of Applied Sports Sciences 11: 211-217, 1986. Morgan DW, Martin PE, Krahenbuhl GS, Baldini FD. Variability in running economy and mechanics among trained male runners. Medicine and Science in Sports and Exercise 23(3): 378-383, 1991. Morgan DW. Factors affecting running economy. Sports Medicine 7: 310-330, 1989. Nomura T. The influence of training and age on kr12 MAX during swimming in Japanese elite age group and Olympic swimmers. In Hollander et al. (eds.) Biomechanics and Medicine in Swimming, pp. 251-257, Human Kinetics Publishers, Champaign, Illinois, 1983.  67 O'Toole ML, Hiller WDB, Crosby LO, Douglas PS. The ultraendurance triathlete: a physiological profile. Medicine and Science in Sports and Exercise 19(1): 45-50, 1987. O'Toole ML, Douglas PS, Hiller WDB. Applied physiology of a triathlon. Sports Medicine 8(4): 201-225, 1989. O'Toole ML, Douglas PS. Introduction: the ultraendurance triathlete: physiologic and medical considerations. Medicine and Science in Sports and Exercise 21(5 suppl): S198-S199, 1989. O'Toole ML. Training for ultraendurance triathlons. Medicine and Science in Sports and Exercise 21(5 suppl): S209-S213, 1989. Olbrecht J, Madsen 0, Mader A, Liesen H, Hollmann W. Relationship between swimming velocity and lactic concentration during continuous and intermittent training exercises. International Journal of Sports Medicine 6(2): 74-77, 1985. Owles WH. Alterations in the lactic acid content of the blood as a result of light exercise, and associated changes in CO2 combining power of the blood and into alveolar CO2 pressure. Journal of Physiology (London) 69: 214-237, 1930. Pate RR, Branch JD. Training for endurance sport. Medicine and Science in Sports and Exercise 24(9 suppl): S340-S343, 1992. Pate RR, Macera CA, Bailey SP, Bartoli WP, Powell KE. Physiological, anthropometric, and training correlates of running economy. Medicine and Science in Sports and Exercise 24(10): 1128-1133, 1992. Pechar GS, McArdle WD, Katch FI, Magel JR, DeLuca J. Specificity of cardiorespiratory adaptation to bicycle and treadmill training. Journal of Applied Physiology 36(6): 753-756, 1974. Pendergast DR, diPrampero PE, Craig AB, Wilson DR, Rennie DW. Quantitative analysis of the front crawl in men and women. Journal of Applied Physiology 43: 475-479, 1977. Powers SK, Dobb S, Deason R, Byrd R, McKnight T. Ventilatory threshold, running economy and distance running performance of trained athletes. Research Quarterly for Exercise and Sport 54: 179-182, 1983.  68 Powers SK, Dobb S, Garner R. Precision of ventilatory and gas exchange alterations as a predictor of the anaerobic threshold. European Journal of Applied Physiology 52: 173-177, 1984. Rhodes EC, McKenzie DC. Predicting marathon time from anaerobic threshold measurements. The Physician and Sportsmedicine 12(1): 95-98, 1984. Roalstad M, Crosby L, O'Toole ML, Zigler A, et al. Heart rate monitoring during an ultraendurance event. Medicine and Science in Sports and Exercise 19(suppl 2): S89, 1987. Roalstad MS. Physiologic testing of the ultraendurance triathlete. Medicine and Science in Sports and Exercise 21(5 suppl): S200-S204, 1989. Sawka MN, Young AJ, Francesconi RP, Muza SR, Pandolf KB. Thermoregulatory and blood responses during exercise at graded hypohydration levels. Journal of Applied Physiology 59(5): 1394-1401, 1985. Schneider DA, Lacroix KA, Atkinson GR, Troped PJ, Pollack J. Ventilatory threshold and maximal oxygen uptake during cycling and running in triathletes. Medicine and Science in Sports and Exercise 22(2): 257-264, 1990. Sjodin B, Jacobs I. Onset of blood lactate accumulation and marathon running performance. International Journal of Sports Medicine 2: 23-26, 1981. Sjodin B, Schele R. Oxygen cost of treadmill running in long distance runners: in Komi (ed.) Exercise and Sport Biology, pp. 61-67 (Human Kinetics Publishers, Champaign, Illinois 1982). Sjodin B, Svedenhag J. Applied physiology of marathon running. Sports Medicine 2: 82-99, 1985. Smith BW, McMurray RG, Symanski JD. A comparison of the anaerobic threshold of sprint and endurance trained swimmers. Journal of Sports Medicine 24: 94-99, 1984. Stainsby W. Biochemical and physiological bases for lactate production. Medicine and Science in Sports and Exercise 18(3): 341-343, 1986. Stegmann H, Kindermann W. Comparison of prolonged exercise tests at the individual anaerobic threshold and the fixed anaerobic threshold of 4mmol/L. International Journal of Sports Medicine 3: 105-110, 1982.  69 Sutton JR, Jones NL. Control of pulmonary ventilation during exercise and mediators in the blood: CO 2 , and hydrogen ion. Medicine and Science in Sports 11(2): 198-203, 1979. Swanson G. Overview of ventilatory control during exercise. Medicine and Science in Sports 11(2): 221-228, 1979. Tanaka K, Yoshimura T, Okuda T, Konishi Y, Sumida S, et al. Physiological alterations in obese women: the effects of caloric reduction plus physical training at an intensity corresponding to or above anaerobic threshold. Bulletin of the Physical Fitness Research Institute 62(suppl): 26-40, 1986. Tanaka K, Matsuura Y, Moritani T. A correlational analysis of maximal oxygen uptake and anaerobic threshold as compared with middle and long distance performance. Japanese Journal of Physical Fitness and Sports Medicine 30: 94-102, 1981. Toussaint HM, Beek PJ. Biomechanics of competitive front crawl swimming. Sports Medicine 13(1): 8-24, 1992. Toussaint HM. Differences in propelling efficiency between competitive and triathlon swimmers. Medicine and Science in Sports and Exercise 22(3): 409415, 1990. Town G, Sinning W. Specificity of training effects as measured by serum enzymes. Medicine and Science in sports 14(2): 172, 1982. Treffene RJ. Swimming performance control using plasma lactate and heart rate measurements. International Swimmer 15(12): 19-20, 1979. van Handel PA, Katz A, Morrow JR, Troup JP, et al. Aerobic economy and competitive performance of U.S. elite swimmers. In Ungerechts, Wilke, Reischle (eds.) Swimming Science V, Human Kinetics Publishers, Champaign, Illinois, pp. 219-227, 1988. van Rensburg JP, Kielblock AJ, van der Walt WH. Maximal aerobic capacity and endurance fitness as determinants of rowing, cycling and running performance during a triathlon competition. The Australian Journal of Science and Medicine in Sport 16(3): 12-16, 1984. van Rensburg JP, Kielblock AJ, van der Linde A. Physiologic and biochemical changes during triathlon competition. International Journal of Sports Medicine 7(1): 30-35, 1986.  70 Volkov NI, Shirkovets EA, Borilkevich VE. Assessment of aerobic and anaerobic capacity of athletes in treadmill running tests. European Journal of Applied Physiology 34: 121-130, 1975. Wasserman K, Beaver W, Whipp BJ. Mechanisms and patterns of blood lactate increase during exercise in man. Medicine and Science in Sports and Exercise 18(3): 344-352, 1986. Wasserman K, Whipp BJ, Koyal SN, Beaver WL. Anaerobic threshold and respiratory gas exchange during exercise. Journal of Applied Physiology 35(2): 236-243, 1973 Wasserman K, Whipp BJ. Exercise physiology in health and disease. American Review of Respiratory Disease 112: 219-249, 1975. Wasserman K, Mcfiroy MB. Detecting the threshold metabolism in cardiac patients during exercise. American Journal of Cardiology 14: 844-859, 1964. Wenger HA, Reed AT. Metabolic factors associated with muscular fatigue during aerobic and anaerobic work. Canadian Journal of Applied Sports Sciences 1: 43-48, 1976. Whipp BJ, Davis JA, Torres F, Wasserman K. A test to determine parameters of aerobic function during exercise. Journal of Applied Physiology 50(1): 217221, 1981. Williams C, Nute MLG. Some physiological demands of a half-marathon race on recreational runners. British Journal of Sports Medicine 17: 152-161, 1983. Williams KR, Cavanagh PR. Relationship between distance running mechanics, running economy, and performance. Journal of Applied Physiology 63: 12361245, 1987. Withers RT, Sherman WM, Miller JM, Costill DL. Specificity of the anaerobic threshold in endurance trained cyclists and runners. European Journal of Applied Physiology 47: 93-104, 1981. Yamamoto Y, Miyashita M, Hughson RL, Tamura S, et al. The ventilatory threshold gives maximal lactate steady state. European Journal of Applied Physiology 63: 55-59, 1991. Yoshida T, Chida M, Ichioka N, Suda Y. Blood lactate parameters related to aerobic capacity and endurance performance. European Journal of Applied Physiology 56: 7-11, 1987.  71 APPENDIX A  Sample calculation for the estimation of time in the swimming component of the Ironman Triathlon.  1) The ventilatory threshold (TvENT) point was determined by a graph of excess CO2 vs. time during the tethered swim performance.  2) Since stroke rate was measured each minute, the TvENT stroke rate was simply the value at its time of occurrence. eg: TVENT = 4 minutes^stroke rate at TvENT= 30 strokes•min-1 3) To determine the stroke length that corresponded to this stroke rate several swim repeats were performed. The subject swam the measured stroke rate over a marked section of the pool while the number of strokes to cover the distance was counted.  72  eg: Swimming at TvENT stroke rate = 30 strokes•rnin -1 , determined stroke length = 1.96 metres•stroke -1  4) These 2 values were then multiplied together to give the velocity at TvENT eg: 30 strokes•min -1 * 1.96 metres•stroke -1 = 58.8 metres•min -1  5) This was then converted to minutes using the known distance of the swim component (2.4 miles) eg: 1 minute•58.8 metres -1 * 1600 metres•mile -1 *2.4 miles = 65.3 minutes  73 APPENDIX B Sample calculation for the estimation of time in the cycling component of the Ironman triathlon.  The above graph indicates the time of occurrence of TvENT. Unfortunately, velocity cannot be readily determined therefore a series of stages were incorporated to estimate this value.  Stage 1: A relationship was developed between Excess CO 2 at TvENT and mph in a short course triathlon (1991 English Bay Triathlon). This variable was chosen based on several studies that have shown its strong relationship to performance in endurance sports (Langill and Rhodes, 1992; Loat and Rhodes, 1991; Rhodes and McKenzie, 1984). A separate subject pool was used and the following linear regression was developed:  mph = ( EXCO 2 * 0.845) + 12.07^r = 0.68  74 Stage 2: Since this equation was being developed in one subject pool and used in another it was desired that the measured variable show similar values for both populations. An ANOVA comparing Ironman and English Bay triathletes found a nonsignificant difference between excess CO 2 used in the prediction equation but a significant difference in speed (in mph).  t  TVENT  p value  correction for the differences in speed  VARIABLE EXCO 2  This necessitated some form of  1.93  > 0.05  between cycling 40 km and cycling 112 miles in a triathlon.  Stage 3: A sample of 26 subjects made up of males in the same age category as this study who took part in both events was determined. A linear regression was performed to relate mph in the Ironman cycling to mph in the English Bay cycling. This equation was then used as a speed correction factor:  mph (112 mile) = (mph (40 km) * 0.9988) - 3.35r = 0.91  Stage 4: These equations were then used to determine mph at subjects as follows:  eg: EXCO 2 at TVENT = 11.54  TVENT  for the Ironman  75 mph at TvENT = (11.54 * 0.845) + 12.07 = 21.82 mph (uncorrected for speed difference)  mph (112 mile) at TvENT = 21.82 * 0.9988 - 3.35 = 18.4 (corrected for speed difference)  Stage 5: Mph were then converted to time using the distance for this component of the triathlon:  112 miles•18.4 mph-1 *60 min•hr-1 = 364.4 min (6:04)  76 APPENDIX C  Sample calculation for the estimation of time in the running component of the Ironman triathlon.  1) In the graph of excess CO 2 vs. time each minute corresponded to a different treadmill velocity eg: minute 8 = 8.5 mph  2) Mph at the threshold was then converted to minutes since the distance of the run component was known (26.2 miles) 26.2 miles•8.5 mph-1 *60 min•hrl = 185 min (3:05)  77  APPENDIX D  Individual data comparing estimated and actual times for swimming, cycling, and running. 1) Swim times (minutes) for individual subjects.  Ell estimated I=1 actual  SUBJECT  1  2  3  4  5  6  7  8  9  10  ESTIMATED  64  66  76  64  62  58  58  65  61  72  ACTUAL  71  70  78  65  58  54  60  76  67  75  78 2) Cycle times (minutes and hours:minutes) for individual subjects.  SUBJECT  1  2  3  4  5  6  7  8  9  10  ESTIMATED  6:04  6:24  5:14  7:47  6:33  5:33  6:54  5:59  6:58  5:54  ACTUAL  5:39  5:40  5:38  6:08  5:44  5:18  5:48  5:56  6:02  5:55  79 3) Run times (minutes and hours:minutes) for individual subjects.  SUBJECT  1  2  3  4  5  6  7  8  9  10  ESTIMATED  2:30  2:55  2:37  2:55  2:55  2:30  3:05  3:05  3:17  3:05  ACTUAL  3:40  3:36  3:14  3:41  4:35  3:26  5:56  4:49  6:18  3:51  80 APPENDIX E  Mechanisms of Lactate and Ventilatory Threshold  The recognition by Owles (1930) that there was a critical exercise intensity above which there was accumulation of blood lactate, combined with increased CO2 excretion and ventilation is the beginning of what would become the anaerobic threshold concept. It is hypothesized that above this "critical intensity" there is a failing of the cardiovascular and/or the respiratory response to supply the energy demanded through the aerobic pathways resulting in work capacity being limited (Knuttgen, 1962). Theoretically, at the anaerobic threshold point work may be performed for indefinite durations since aerobic energy sources are supplying the required energy and waste products are being adequately removed. Thus, the mechanisms responsible for threshold are those which limit aerobic performance and, in some way, control the relationship between endurance capacity and induced fatigue. The determination of this point by invasive and noninvasive means and their underlying mechanisms are the source of considerable debate. One of the predominant categories of threshold determination involves the invasive measurement of lactate. Lactate is produced to supplement the aerobic energy supply and its presence in the bloodstream reflects increased reliance on the glycolytic pathways of energy production (Jones and Ehrsam, 1982). If there is an imbalance between pyruvate formation and its oxidation in the Kreb's Cycle a subsequent conversion to lactic acid will occur to allow for the continuation of glycolysis (Stainsby, 1986; Jones, 1980). The lactate formed in the dissociation of lactic acid provides a source of fuel (Anderson and Rhodes, 1989) to the working muscle and other tissues of the body.  81 In progressive intensity exercise there is an initial rise in blood lactate (Brooks, 1986) which is maintained (or slightly increased) until a point where there is a disproportional or abrupt increase in concentration. Whether tissue hypoxia is accompanying this increase in lactic acid production is still questioned (Brook, 1985; Gollnick et al., 1986; Wasserman et al., 1986). It is speculated that this point represents an imbalance in lactate production and removal resulting in an increased reliance on anaerobic metabolism and the release of lactate into the blood (Davis et al., 1983). If the biproducts of lactic acid are allowed to accumulate within the working tissues there will be a rapid onset of fatigue (Anderson and Rhodes, 1989). Increased lactate production results in an increased hydrogen ion release, decreasing muscle and blood pH (Wenger and Reed, 1976) which would limit energy production by anaerobic glycolysis (Hultman and Sahlin, 1980). The fatigue accompanying decreased cellular pH may be the result of an alteration in the membrane permeability or interference with calcium ion binding at the actomyosin binding sites (Wenger and Reed, 1976). The actual lactate threshold (Tukc), determination based on these mechanisms, has involved several methodologies. TLAc has been characterized by an absolute blood lactate concentration of 2mmole (Hughson et al., 1982) or 4mmole (Sjodin and Jacobs, 1981), an increase above resting levels (Wasserman et al., 1973) and that point where there is an abrupt increase in lactate accumulation (Aunola and Rusko, 1984). Most of the methods are not without controversy, particularly the absolute blood lactate concentrations which have been questioned by several researchers (McLellan and Jacobs, 1989; Stegmann and Kindermann, 1982) on the basis of individual lactate kinetics. The other predominant category for threshold determination is based on noninvasive ventilatory parameters. The prime stimulators of ventilation  82 accompanying increasing levels of exercise are increased CO 2 production, an associated fall in blood bicarbonate due to buffering of metabolic acids, and a rise in arterial pH (Anderson and Rhodes, 1989). Changes in breathing rate and depth are made to provide a matching of alveolar ventilation to blood perfusion (Sutton and Jones, 1979). Ventilation is strongly linked to CO 2 output (Wasserman and Whipp, 1975) with the relationship demonstrating ventilation increases in proportion to CO 2 production (Swanson, 1979). Prior to the threshold, CO 2 production will rise in response to the aerobic metabolism of fats and carbohydrates. Above the threshold, in this case ventilatory threshold (TVENT), there will be an increased CO 2 load resulting from the production and buffering of lactic acid by bicarbonate as follows: HLa + NaHCO 3 = NaLa + H 2 CO3 = CO2 + H2 0 Thus, above TVENT increases in ventilation above those responsible for arterial CO 2 compensation do not allow for the complete compensation for the pursuing lactic acidosis (Wasserman and Whipp, 1975). An independent ventilatory stimulus is created to increase ventilation and lower the hydrogen ion content of the blood. This will result in a constraining of the buffering capacity of the blood and pH will drop (Whipp et al., 1980) with the accompanying 'fatigue" results already discussed. The ability to use ventilatory measures to represent invasive changes has received considerable attention. While Powers et al. (1984) were not able to produce high correlations and concluded that TLAc could not be accurately determined by gas exchange variables several other researchers have employed TVENT with considerable success (Wasserman et al., 1973; Clode et al., 1969; Anderson and Rhodes, 1991). Again, as with TLAc, considerable variability exists with respect to the different methods of detection. Caiozzo et al. (1982) found R to be the least sensitive, with  VE /VO2 1/CO2 to all provide ,  .  83 excellent predictions of T ugs . One other measure used in detection has produced very strong links to performance (Rhodes and McKenzie, 1984; Hearst and Rhodes, 1982; Loat and Rhodes, 1991). It has been found that the nonmetabolic (excess) CO 2 , resulting from the buffering of lactate, will be generated as long as the rate of lactic acid production is increasing as this will mean additional hydrogen ions to buffer (Wasserman et al., 1986). While the hydrogen ion and CO 2 within the muscle readily diffuse into the bloodstream the lactate molecule has a translocation hindrance to its movement out of the muscle (Stainsby, 1986). This may result in increases in excess CO 2 being detected before a significant rise in blood lactate, allowing excess CO 2 to more accurately reflect cellular lactate production and accumulation (Issekutz and Rodahl, 1961). Volkov et al. (1975) found the excess CO 2 "index" to directly relate to the magnitude of lactate production through the glycolytic pathway and the organism's buffering processes. Further investigation by Langill and Rhodes (1992) found that a strong relationship exists between excess CO 2 and blood lactate (r=0.92) and that their respective curves over time with increasing exercise parallel one another. Despite the controversy surrounding Tukc and T vENT there is considerable support for the concept (Wasserman et al., 1986; Davis, 1985). While many of the exact mechanisms are still being questioned, the inconclusive results are more likely a result of the need for further research, particularly the intracellular events that these thresholds are based on.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
United States 15 2
Canada 8 0
China 5 31
Russia 4 0
Germany 4 7
Unknown 1 0
Poland 1 0
Spain 1 0
Japan 1 0
City Views Downloads
Ashburn 9 0
Unknown 6 33
Shenzhen 5 31
Saint Petersburg 4 0
Vancouver 3 0
Menlo Park 3 0
Ottawa 3 0
Coquitlam 2 0
Cheyenne 2 0
Madrid 1 0
Bellingham 1 0
Tokyo 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items