UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of steady rate exercise on the pattern of force production of the lower limbs in cycling Black, Alexander H. 1994

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

Item Metadata

Download

Media
831-ubc_1995-0034.pdf [ 3.67MB ]
Metadata
JSON: 831-1.0077057.json
JSON-LD: 831-1.0077057-ld.json
RDF/XML (Pretty): 831-1.0077057-rdf.xml
RDF/JSON: 831-1.0077057-rdf.json
Turtle: 831-1.0077057-turtle.txt
N-Triples: 831-1.0077057-rdf-ntriples.txt
Original Record: 831-1.0077057-source.json
Full Text
831-1.0077057-fulltext.txt
Citation
831-1.0077057.ris

Full Text

T H E EFFECT OF S T E A D Y R A T E EXERCISE O N T H E P A T T E R N O F F O R C E P R O D U C T I O N OF T H E L O W E R LIMBS I N C Y C L I N G . By Alexander H . Black B.P.E. University of British Columbia. 1989 A THESIS SUBMITTED I N P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF M A S T E R O F S C I E N C E i n T H E F A C U L T Y OF G R A D U A T E STUDIES (School of Human Kinetics) We accept this thesis as conforming to the reauired standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A D E C E M B E R , 1994 © Alexander H . Black, 1994 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. r5e~partnTent of rJ^Wc^ (/M^-g.ln<-i The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A B S T R A C T The role of fatigue in the cessation of activity has been studied extensively and reported i n the physiology literature. Biomechanical compensations that occur as a result of fatigue, however, have had little focus in the scientific literature. Data have been published (Black et al. 1992 and Amoroso et al., 1993) that suggest there are biomechanical compensations that occur just prior to the cessation of activity as a result of the fatigue process in cycling. These compensations include increases in the index of effectiveness, earlier maximum force production in the pedal cycle, and increased dorsiflexion at the ankle. These changes suggest that there are compensations by the muscles in response to the onset of whole body fatigue. The purpose of the present investigation was to quantify the biomechanical changes that result from the progressive onset of fatigue in cycling. The criterion measurements included timings of maximum joint angle excursions, timings of maximum force production and timings of maximum joint moments at the ankle, knee and hip. Male cyclists (n=12) completed a progressive, incremental maximal exercise test at 90 R P M on a bicycle ergometer to determine maximum power outputs. Two steady rate, constant power output rides followed: one at 80% maximum power output (max P.O.) to exhaustion, and one at 30% max P.O., for the same length of time as the 80% max P.O. ride. Force data were collected from the right pedal of an instrumented bicycle for 3 pedal cycles at the end of each minute of the steady rate exercise tests. Kinematic data were recorded ongoing throughout the steady rate exercise test. Kinematic and kinetic data for the initial and final minutes of both steady state rides were then time matched and joint moments for the ankle, knee and hip were calculated using the inverse dynamics approach. There were significant differences between the initial and final minutes of the study. These changes included earlier maximum hip extensor moment, greater ankle plantarflexor moment, increased knee flexor moment and increased hip extensor moment (p<0.05). These increased joint moments summed to produce a significantly larger propulsive moment in the final minute of the exercise (p<0.05). These changes were results of changes in the kinematic data and the kinetic data that were the result of fatigue. / Based on the results of this study, it was concluded that the increase in the propulsive moment was necessary to overcome a decrease in the index of effectiveness (p<0.05), which was a product of whole body fatigue. The increase in the propulsive moment was a result of increases in the maximum joint moments at each of the lower limb joints. In summary, as the results of fatigue affect an athlete, the athlete is forced to change the strategy of muscle recruitment which is used to overcome the given power output in order postpone the cessation of exercise. i i i T A B L E OF C O N T E N T S A B S T R A C T i i T A B L E OF C O N T E N T S i v LIST OF T A B L E S v i LIST OF FIGURES v i i A C K N O W L E D G M E N T S ix I N T R O D U C T I O N 1 Statement of Problem 4 Hypotheses 5 Significance of the Study 6 Definition of terms 6 Delimitat ions 8 R E V I E W OF T H E L I T E R A T U R E 10 Physiology Review 10 Biomechanics and Fatigue 14 Biomechanics of Cycl ing 20 P R O C E D U R E S 25 Subjects 25 Instrumentation and Data Collection 25 Experimental Protocol 27 Data Analysis 28 R E S U L T S 32 Description of Subjects 32 Group Metabolic Data 32 Group Kinematic Data... 35 Group Average Joint Moment Data 48 With in Subject Variability 55 Between Trial Differences 56 i v D I S C U S S I O N 57 Overv iew 57 Comparisons 57 Support for Hypotheses 59 Hypothesis 1 59 Hypothesis 2 60 30% max. P.O. trials 63 Variabi l i ty 64 Big Picture 68 C O N C L U S I O N S A N D R E C O M M E N D A T I O N S 69 Summary 69 Conclusions 70 Recommendations 71 R E F E R E N C E S 73 v LIST OF T A B L E S 1. Subject descriptive data 33 2. Incremental V O 2 max data 33 3. Steady rate, constant power output tests data 32 4. Mean Intra-subject coefficients of variation for joint angles 55 5. Mean Intra-subject coefficients of variation for forces 55 6. Mean Intra-subject coefficients of variation for joint moments 56 v i LIST OF FIGURES 1. Pictorial representation of the forces applied to the pedal and to the crank of a bicycle 7 2. Energy-producing mechanism in muscle 11 3. H i p , knee, and ankle moments reported by Gregor et al. (1985), H u l l and Jorge (1985) and Ericson (1986) 22 4. Pictorial representation of the marker placement and the joint angle definitions 26 5. (a) Average foot angle and (b) average shank angle for the 80% max P.O. trial 36 6. (a) Average thigh angle and (b) average pedal angle for the 80% max P.O. trial 37 7. (a) Average foot angle and (b) average shank angle for the 30% max P.O. trial 39 8. (a) Average thigh angle and (b) average pedal angle for the 30% max P.O. trial 40 9. (a) Average normal pedal force, (b) average tangential pedal force and (c) average resultant pedal force for the 80% max P.O. trial 43 10. (a) Average effective pedal force and (b) average index of effectiveness for 80% max P.O. trial 44 11. (a) Average normal pedal force, (b) average tangential pedal force and (c) average resultant pedal force for the 30% max P.O. trial 46 12. (a) Average effective pedal force and (b) average index of effectiveness for 30% max P.O. trial 48 v i i 13. (a) Average ankle moment and (b) average knee moment for 80% max P.O. trial, initial and final minutes 50 14. (a) Average hip moment and (b) average propulsive moment for 80% max P.O. trial, initial and final minutes 51 15. (a) Average ankle moment and (b) average knee moment for 30% max P.O. trial, initial and final minutes 53 16. (a) Average hip moment and (b) average propulsive moment for 30% max P.O. trial, initial and final minutes 54 17. Comparisons wi th other joint moment data 58 v i i i A C K N O W L E D G M E N T S I wou ld like to acknowledge the following people whose assistance wi th this thesis can not be left unheralded. To Dr . D o n McKenzie for his sense of calm and understanding, even when the going got biomechanical, and for the use of his lab, without which the physiological testing wou ld have been impossible. Both Dr. McKenzie and Diana Jesperson made me comfortable in the Physiology lab at the A l l a n M c G a v i n Sports Medicine Centre. To Dr. Angelo Belcastro who always broke the problem down further and further unt i l the questions became obvious. Angelo's fine-toothed-comb analysis of the drafts of my thesis helped put the entire project into perspective. To my friend and mentor Dr. David Sanderson whose guidance first led me to Biomechanics almost ten years ago and whose persistence and love for the discipline has kept my interest. Through the years we have done many exciting scientific and non-scientific pursuits together. I hope that this relationship w i l l continue. Finally, I wou ld like to acknowledge my family and friends. To my friends an lab mates that helped me through the tough times and helped me enjoy the good times. To Er in Johnston for her love and constant support and to her family who always were interested and supportive of my progress. Most of all , however, to my parents and brother, for without their nurturing and support none of this w o u l d have been possible. A n d for the words my father told me, time and again, which I w i l l never forget "get it done". ix Chapter 1 I N T R O D U C T I O N There are two reasons for studying the biomechanics of cycling. First, because of the increase in popularity of cycling there has been a concomitant increase in the number of overuse injuries in the lower extremity (Francis, 1986, Francis, 1988, Hannaford et al., 1985 ). There is still some uncertainty, however, as to the etiology of these injuries and thus research is warranted to understand the mechanics of cycling. Second, because of the relatively closed loop system that cycling provides, the relative ease in the collection of cycling data, and similarity to other sports, in terms of repetitive motions, cycling is an important model in the exploration of skeletal muscle mechanics. Physiological changes that are a result of athletic performance have been wel l documented. Incremental tests such as the maximum oxygen uptake test, termed V 0 2 m a x test, measure aerobic fitness by inducing fatigue. Fatigue has been defined as the inability to generate a required or expected force (Edwards, 1981). Therefore, just prior to fatigue the athlete is doing his/her best to overcome the load and must therefore process the greatest amount of oxygen ( O 2 ) possible; this point is termed V02max- Incremental tests are performed a number of different ways, some of the most popular being treadmill running and bicycle ergometry. The physiological changes invoked by these tests are relatively wel l described, however the underlying biomechanics are not. Physiologically, for whole body exercise, there is a greater demand for O 2 by the working muscles as the power output increases. This increase in whole body O 2 demand and consumption to meet the requirements of increasing power output does, however, have an upper limit. Whole body fatigue can therefore be thought of as a broad term that refers to a number of physiological and neurological manifestations that leave the individual unable to complete or overcome a given task. This failure to maintain the required or expected force can be manifested in a number of different ways. Whole body fatigue, in general, can be broken down into two main categories: central fatigue and peripheral fatigue. Central fatigue is a failure of the neural drive resulting in 1), a decrease in the number of functioning motor units and 2), a decrease in the motor unit firing frequency. Peripheral fatigue, on the other hand, is a failure of force generation of the whole muscle (Edwards, 1981). Wi th both central fatigue and peripheral fatigue there are a number of mechanisms that individual ly or in combination cause the general failure. Regardless of the mechanisms contributing to whole body fatigue the ultimate outcome, by definition , is a failure to complete the given task. This definition, however, implies that fatigue is a barrier that the athlete all of a sudden reaches. This definition is, in the authors mind a rather narrow view of the concept of fatigue. A s an athlete progresses through an exercise test there is an accumulation of metabolites associated wi th the fuel used by the athlete. In conjunction wi th the increase in the metabolites produced, there is also a decrease in the fuel available to the athlete. Fatigue should therefore be thought of as a continuous slope where at some point the athlete can no longer overcome the effects of the fatigue and cessation of the activity occurs. Al though the physiological characteristics associated wi th whole body fatigue have been studied extensively, the biomechanics underlying the physiological changes have not. Few biomechanical studies have probed the changes that occur as a result of whole body fatigue. Kinematic changes that result from fatigue in running have been investigated by Bates and Osternig, 1977, Chapman, 1982 and 1981, Elliot et al. 1981 and 1980, and Siler and Mart in, 1991 and the kinetic changes accompanying whole body fatigue is even more sparse in the literature, having been studied only by Sprague and Mann, 1983. Bates and Osternig (1977) and Chapman (1982) measured kinematic changes that result from whole body fatigue. In each study these authors found measurable changes in the patterns of motion that the athletes exhibited during separate running protocols leading to fatigue. Al though velocity was measured in each of the studies there was no control or maintenance of constant overground velocity. Therefore, as the athlete exercised the kinematics that changed were a result of not only fatigue but of overground velocity decreases as wel l . Siler and Mart in (1991) removed the confounding variable of decreased overground speed by exercising the athletes on a treadmill at constant velocity. By maintaining the velocity of the run these investigators were able to quantify the kinematic changes in running that accompanied the fatigue process. There is only one study that has described the kinetic changes that accompany the fatigue process in running, Sprague and M a n n (1983). These authors compared joint moments during foot contact wi th the ground at the beginning of a 400 m race to the same joint moments at the end of the race. Al though differences were found, these differences may have been a result of overground velocities which decreased from 9.51 m / s at the start to 7.53 m / s at the end, and not a reflection of the fatigue process during this exercise, as was the case in many of the kinematic studies. There are primarily two major confounding issues in the studies of the relationship between whole body exercise to fatigue and the biomechanical changes that accompany the fatigue process. In the running studies that have investigated the effects of whole body fatigue biomechanically, outside the laboratory (Sprague and Mann , 1983, Bates and Osternig, 1977 and Chapman, 1982) there has been no control for the maintenance of a constant overground velocity. A s the athlete exercised, in general, overground velocity decreased. When the overground velocity decreases there are concomitant changes in the kinematics, such as joint angles, joint angular velocities etc., as wel l as changes in the patterns of the kinetics: the applications of force to the ground. Therefore, the biomechanical changes measured i n these studies (and their subsequent conclusions) are confounded wi th this design issue of decreased overground running velocity. The second confounding issue wi th the majority of the studies to date is that there were no complete analyses of the step cycle. For the most part the studies that relate biomechanical changes of whole body exercise to the accompanying fatigue processes have utilised only a kinematic approach to the analyses. Because of this approach a large part of the "picture" (i.e. analysing the forces involved i n the exercise) is not available. To date most of the biomechanical research into the relationship between whole body exercise and the fatigue process have used a running model. The present study however, proposes to utilise a cycling model in order to minimise the confounding issues introduced above, and analyse consecutive cycles of kinematic and kinetic data. In cycling the foot is in contact wi th the pedal throughout the cycle, therefore forces can be collected for the entire cycle as wel l as the pattern of l imb motion. Wi th data for the whole cycle more complete analyses are possible and thus questions regarding the influence biomechanical factors during whole body exercise and the fatigue process can be investigated. Such studies using a cycling model are non-existent in the literature. It is therefore the aim of this study to measure and describe the changes in biomechanical measures, specifically changes in joint kinematics, pedal force kinetics and joint moments, during a steady rate exercise test to voluntary cessation of exercise. Statement of Problem In the literature there is minimal evidence of changes in the biomechanics of cycling, kinetically or kinematically, that accompany the fatigue process. There are data to support the hypothesis that fatigue affects the kinematics of running (Bates and Osternig, 1977; Chapman, 1982; Siler and Mart in, 1991) and the kinetics of running (Sprague and Mann, 1983), although kinetic data are sparse and i n most cases the protocols are confounded. There are minimal data in the literature that support the hypothesis that there are relative load changes at the three joints (ankle, knee, and hip) as determined by changes in patterns of joint moments. If changes in the patterns of joint moments throughout the pedal cycle occurs during prolonged exercise which accompanies the fatiguing process, this w i l l allow for a better -4-understanding of biomechanical mechanisms utilised by an athlete for prolonged activity. Hypotheses There are two hypotheses: 1. in order to maintain the test cadence during a steady rate constant power output exercise test the athlete w i l l change the pattern of force application such that the maximum joint moments w i l l be greater and occur earlier in the pedal cycle when compared to similar power outputs in an unfatigued state. 2. the changes described in hypothesis 1 were a direct result of the athlete changing his pattern of force application such that the index of effectiveness for the complete pedal cycle w i l l increase. This increase w i l l be accomplished by an increase in the positive impulse, earlier maximum force because of an increase in the negative impulse in the second half of the pedal cycle, and thus an increased positive impulse in the first half of the pedal cycle. The increase in the positive impulse w i l l be accomplished by an increase in the forces applied to the pedal i n the propulsive phase of the pedal cycle. Furthermore, the maximum joint angles w i l l occur earlier in the pedal cycle, which in conjunction wi th the earlier maximum forces, w i l l result in earlier maximum joint moments . Therefore the objectives of the proposed study were: A) a steady rate, constant power output test at 80% of the athlete's maximum power output (max. P.O.) was used to induce biomechanical changes resulting from the fatigue process. Comparisons of the joint moments, t iming of maximum force production and kinematic changes between the "fatigued" state, at the end of the test, and the "unfatigued" state, the beginning of the - 5 -test, w i l l give insight into accommodations the athletes makes through the fatigue process. B) to ensure that the results of the 80% constant power output test were truly reflective of the fatigue process accompanying prolonged exercise and are not confounded by exercise time a second steady rate constant power output test was used. It was theorized that there should be minimal changes between the first minute of the test and the final minute of the test in the variables that were to be measured. Significance of the Study This study addressed the importance of both the kinematic changes and the kinetic changes that occur as a result of prior exercise, ultimately leading to fatigue, at constant power output and cadence. To date none of the running studies that have investigated the fatigue process have been able to consistently maintain cadence and speed as wel l as power output, in conjunction wi th the collection of kinetic data. This study contributed ideas and data to the area of muscle mechanics as related to prior exercise, or to fatigue. Definition of terms Norma l Force (FM ) - The component of the force that is applied vertically and perpendicular to the pedal surface. Tangential Force (¥j) - The component of the force that is applied horizontally and parallel to the pedal surface. Resultant Force (FT?) - the vector sum of the normal and tangential pedal forces. Unused Force (Fn) - The portion of the resultant force that is parallel to the crank. This force is ineffective because it produces either tension or compression of the crank. Effective Force (FF) - The portion of the resultant force that is perpendicular to the crank. This force is the sole contributor to crank torque. 57 -( — "V \ \ \ \ \ \ F E /I #FR F T Figure 1. Pictorial representation of the forces applied to the pedal and to the crank of a bicycle. Index of Effectiveness (IEFF) - The integral of the effective force, calculated over the entire pedal cycle, divided by the integral of the total force applied to the pedal, calculated over the whole pedal cycle (Black et al., 1992, Lafortune and Cavanagh, 1983, Sanderson and Cavanagh, 1985). t" \FEdt IEFF(%) = x 100 \FRdt Jto Power Output - The amount of work done by the cyclist per unit time, measured in watts (W). Cadence - The rate at which the pedals turn, measured in revolutions per minute (rpm). Triaxial Piezo-Electric Force Transducer - A device that measures component forces in three plans; X , Y , and Z . The output is in volts which are the changed to Newtons by the analysis program in the computer that collects the data. Joint Moment - (moment of force, torque) A rotatory force, produced by the muscles, that produces angular acceleration (of a limb segment). Propulsive Moment - Defined by Winter (1980) as the net support moment i n the stance phase of walking and was the summation of the lower l imb extensor -7-moments (ankle, knee and hip). Because the body is not supported by the limbs in cycling a more appropriate name for the summation of the lower l imb moments is the propulsive moment. Thus the propulsive moment is the sum of the lower limb joint moments that aid in propulsion of the bicycle. A negative propulsive moment, therefore, would be one that retards the creation of positive torque, and thus wou ld hinder performance. Torque - A rotary force that produces angular acceleration. More specifically it is the force that causes the crank on the bicycle to rotate. The calculation of torque in the present context is: where Fe is the effective force applied to the crank, and CL is the length of the crank (0.170 m). Coefficient of Variation (CV) - A measure of the total variability of any ensemble averaged patterns of motion, i.e. ensemble averaged force, joint moments, angles. The C V is the root mean square of the standard deviation of the moment (or force, or angle) divided by the absolute moment (or force, or angle) over the pedal cycle (Winter, 1984). where: N is the number of intervals over the pedal cycle, Mj is the average amplitude of the normalized moment of force (or force, or angle) at the ith interval, and a/ is the standard deviation of Mi at the ith interval. "Thus C V represents the r.m.s. width of the standard deviation 'band' expressed as a percent of the magnitude of the signal pattern itself" (Winter, 1984, p.55). Del imitat ions Because data were collected from only one side of the body, the right side, bilateral symmetry was assumed. This is to say that for one complete pedal cycle the same motion and forces occurred in both legs 180° out of phase. This may be a 1'crank ~ F e * CL - 8 -source of error, a source which was not measurable wi th the present data collection system at the U .B .C . Biomechanics laboratory; There are reports in the literature that it is relatively common to find riders who have asymmetrical patterns of force application between the two legs (Sanderson, 1990, Daly and Cavanagh, 1976). It was suggested that this asymmetry had little effect on the present study as it d id not affect the measurement of joint moments or the relative changes in the joint moments, as the present study investigated relative changes of the same limb at different times. A second assumption is that the marker placement for the kinematic data collection remained consistent for the duration of the exercise and that marker motion that occurred wi th the flexion and extension of the limb segments incurred negligible errors. Lafortune et al. (1992) and Black (1993a) studied the effects of skin motion on the errors incurred in the collection of kinematic data. Both these authors found that there was substantial error in the calculation of joint centres as a result of marker motion. Black (1993), however, found that in studies that used the marker placement to determine differences between an initial and a final trial in the same event, and where the markers were not moved wi th in the trial, the error incurred was irrelevant to the outcome. For example if there was an error incurred as a result of skin movement, this error would be consistent throughout the testing procedure. Thus the changes that were seen in this investigation may have incurred skin motion error, however, because this error was consistent throughout the test the changes that were seen when the unfatigued state was compared to the fatigued state were a result of fatigue and not a result of a change in the skin motion error. Finally, during long periods of time there is a certain amount of "drift" that occurs in the force transducers. This "drift" is a result of electrostatic forces being built up i n the transducers. The duration of the present tests wi th in this study were short enough, however, that "drift" was not a problem. Chapter 2 R E V I E W OF T H E L I T E R A T U R E Most of the cycling research that has dealt wi th whole body fatigue has been oriented towards the metabolic aspects of acute and chronic exhaustive exercise. There is no research that has investigated the biomechanical outcomes and the compensation to prolonged exercise and the accompanying fatigue process. The following literature review w i l l discuss the research that has been done in both the fields of the biomechanics of cycling and of fatigue itself. The literature review w i l l be broken down into three main categories: 1) a brief review of physiological characteristics of fatigue, 2) the biomechanics of fatigue, and 3) the biomechanics of cycling. In the review of the biomechanics of fatigue kinematics, kinetics and joint moment literature w i l l be discussed in terms of not only cycling, but also other sports such as running. Finally the review of the pertinent physiological data w i l l briefly cover the mechanisms of fatigue as wel l as discuss incremental tests such as the ones that were used in the present study. Physiology Review Al though this study is not a metabolic one, rather biomechanical, it is necessary to review the literature that has examined the phenomenon of fatigue. In general, fatigue is defined as a failure to maintain the required or expected force (Edwards, 1981). From a more metabolic perspective Sahlin (1985) states that "when the rate of A T P production is insufficient to meet the demands the muscle becomes fatigued." Furthermore "the metabolic change in muscle is dependent upon the work intensity at which the fatigue occurs and w i l l be quite different at low intensity (glycogen depletion) than at high intensity (LA accumulation and P C r depletion)" (Sahlin, 1985). -10-In order for exercise to continue, A T P , the most immediate energy source at the muscle cell's disposal, must be resynthesised at the same rate at which it is being utilised. This resynthesis of A T P can be achieved by three mechanisms (Figure 2): " (a) by the creatine kinase reaction whereby the storage of the high-energy phosphate compound phosphocreatine (PCr) is utilised, (b) through glycolysis or glycogenolysis resulting i n lactic acid (LA) formation, and (c) through complete oxidation of carbohydrates (CHO) or free fatty acids (FFA) within the mitochondria" (Sahlin, 1985). The first two mechanisms are anaerobic processes and the third mechanism is aerobic. ATP < ADP + Pi + Energy Cytoplasm: ADP + PCreatine < Creatine + ATP ADP + Glucose (Glycogen) < Lactate + ATP Mitochondria: ADP + Substrate + O2 < CO2 + ATP Figure 2: Energy producing mechanism in muscle (from Sahlin, 1985). Fatigue cannot, however, be a result simply of a depletion of energy stores to any particular level and therefore must be further defined or broken down into smaller more definable subsets. There are two main subsets in the realm of fatigue: they are central and peripheral fatigue. Central and Peripheral Fatigue. Failure anywhere along the command chain between the brain and the formation of the actin and myosin cross bridges wi th in the muscle may result in fatigue. The first classification of fatigue is central fatigue. Central fatigue is caused by a failure in the neural drive. The second classification of fatigue is peripheral fatigue which is a result of impairment of force generation by the muscle (MacLaren et al., 1989 ). -11-Central fatigue may occur because of malfunction of nerve cells or inhibition of voluntary effort; the section of sensory pathways in the reticular formation has been suggested as a critical mechanism (MacLaren et al., 1989). Other causes of central fatigue that have been mentioned in the literature are: inhibitions in the chemoreceptors in the fatigued muscle; increased levels of plasma tryptothan: branched chain amino acid ratio which in turn increases the level of neurotransmitters in the brain; and finally an increase in the ammonia content of the blood, as a byproduct of muscle during exercise, which is directed mainly towards the central nervous system. In the brain the N H 3 accumulation may alter the concentration of neurotransmitters and reduce the level of A T P (MacLaren et al., 1989). Peripheral fatigue is a failure of force generation of the whole muscle and can occur at three possible sites: the neuromuscular junction and the muscle cell membrane (excitation); the calcium release mechanism (activation); and the sl iding filaments (contractile processes) (McLaren et al. 1989). Al though I have defined the relative sites of fatigue and the working definitions of fatigue, these have a relatively insignificant role in the present investigation. Of greater importance is the notion of volitional fatigue or volit ional exhaustion. U p to this point in the paper, fatigue has been defined as the inability of the athlete to accomplish the given task, in this case the given power output at which point there is a cessation of exercise, and the term fatigue is used. However, the data collection and measurements of forces and kinematics were done just prior to "fatigue". Therefore, true fatigue is a misnomer in this case. Furthermore, the types of measurements that were taken in this experiment w i l l not be sensitive enough to differentiate between central or peripheral fatigue. It is more sensible to think of fatigue, for the purposes of this investigation, as volit ional exhaustion on the athlete's part, and that volitional exhaustion was a result of one or more of the mechanisms defined as fatigue in the above literature review. Thus, the point of -12-the exercise which is of interest in the present investigation is that point just prior to volit ional exhaustion. Furthermore, although fatigue has been defined as a "barrier" to the continuation of the exercise it is felt by the author that fatigue is more of a progressive process whereby as the fuel depletes and the metabolites increase there is a point at which the athlete can no longer carry on, at this point the athlete is fatigued. -13-Biomechanics and Fatigue Biomechanical data on the effects of prior exercise or fatigue are relatively sparse in the literature. The kinematics of fatigue in running have been analysed by a few authors (Bates and Osternig, 1977, Chapman, 1982. Siler and Mart in , 1991, Wil l iams et al., 1991) and even more difficult to find are data on the kinetic changes resulting from fatigue (Sprague and Mann, 1983). Even though cycling is an excellent medium for the study of muscle mechanics there are no published data relating the effect of prior exercise to biomechanical measures. Bates and Osternig (1977) compared selected temporal and kinematic parameters of 12 female subjects in a maximal running test at two stages, unfatigued and fatigued. Temporal changes included significant increases in the percentage of time the athlete was in foot strike and foot descent, along wi th significant decreases in the percentage of time the foot is in forward swing, while take off and follow through remained unchanged (Bates and Osternig, 1977). These temporal changes indicated to these investigators that fatigue causes a breakdown i n the internal t iming mechanisms of the athlete. Kinematically there were changes in the step velocity wi th fatigue, this change was a result of a decreased step length rather than a change in the stepping frequency. Furthermore, there were changes in a number of joint angles when the unfatigued condition was compared to the fatigued condition, however these are too numerous to mention in this context. Finally these authors concluded that "the concept of fatigue does not simply produce a consistent uniform reduction in the components of the movement pattern but rather changes in their relationship completely" (Bates and Osternig, 1977, p. 207). Al though there were no kinetic data collected in this study the authors concluded as cited in the previous sentence, thus indicating that it is possible that the recruitment strategies utilised by the athlete in order to extend the length of the exercise changed as the athlete became fatigued. -14-Chapman (1982) researched the hierarchies in sprinting fatigue. Us ing five female sprinters, Chapman measured a number of kinematic variables at the beginning of a 400m run and at the end of the same run. The variables considered in this study were velocity, stride length, cycle time, as wel l as times of stance, recovery, absorption and driving phases of the 400m runs of each subject. Chapman found that velocity and stride length were significantly reduced (p<0.05) while cycle time, stance time and the time in driving phase were significantly increased. In terms of the hierarchical changes Chapman found that: "Velocity decreased due to a decrease in step length which (was) sufficiently large to counteract the decrease in cycle time. The decrease in cycle time (was) due to a decrease in recovery time which again (was) sufficiently large to counteract an increase in support time" (Chapman, 1982, p. 120). In conclusion Chapman stated that the response to fatigue manifests differently depending on the individual athlete, and that although there are common trends in the changes that result from fatigue individual variability hinders the researcher i n finding significance in many of the variables measured. Chapman's results were supported by Sprague and M a n n (1983) in a study that not only compared fatigued and unfatigued states kinematically but also kinetically. Sprague and M a n n found that in their group of 15 elite middle distance runners there were significant kinematic differences between the beginning of a 400m race (40m into the race) and the end of a 400 m race (380m into the race). In fact they found that there were differences in the stride frequency, ground phase time, body centre of gravity (C of G) horizontal velocity as wel l as foot C of G velocity prior to heel strike. They compared not only kinematic differences but also kinetic changes that resulted from fatigue. The kinetic variable that was selected for comparison was joint-muscle moment. The joint moments for elbow, shoulder, hip, knee, and ankle were calculated at each phase of the movement. The joint moments were -15-then integrated, and the difference between the unfatigued integrated moments and the fatigued state integrated moments were then compared. Sprague and M a n n found that "poorer" sprinters tended to generate a significant increase in both the hip moment and the knee moment during foot contact wi th the ground i n the fatigued state, when compared to the unfatigued state. This increase in hip and knee moments was attributed to "significant kinematic alterations i n mechanics beginning prior to foot strike" (Sprague and Mann, 1983). In conclusion they stated that although fatigue occurred in both the better and poorer sprinters, the better sprinters were able to maintain both mechanics and level of effort as fatigue set in. "Due to fatigue altered mechanics beginning prior to foot strike, the poorer sprinter was forced to generate greater productive muscular activity during the initial portion of the ground-phase... The inefficient mechanics of the fatigued poorer sprinter continued to affect the activity during the latter portion of the ground phase, since productive muscular activity was l imited due to altered body position" (Sprague and Mann, 1983, p. 66). The results of these predominantly kinematic investigations are, in general, that as the athlete becomes fatigued velocity of the centre of mass decreases, stride length shortens, and the range of motion exhibited at the joints of the lower extremity is reduced. However, Siler and Mart in (1991) complained that the absence of speed control presented a confounding factor in all the previously cited studies in that al l of the changes in running pattern have also been seen wi th decreased running speeds in the absence of fatigue. "Consequently, by not forcing the runners to maintain a constant running velocity, these studies d id not attempt to differentiate between behavioural changes occurring wi th the development of fatigue and those associated wi th a reduced running speed" (Siler and Mart in , 1991). Siler and Mar t in (1991) presented a technique to avoid the confounding presence of decreasing speed in the attempt at studying fatigue in running. The purpose, that is of interest in this discussion of the effects of fatigue on performance, -16-was to describe the compensation that distance runners make in order to maintain a given treadmill speed during an exhausting run. The second purpose of their study, and of less interest in the present discussion, was the attempt to determine whether the response to fatigue differs between two groups of runners, an elite group and a recreational group. There were a total of 19 subjects, wi th 9 being in the fast group and 10 in the slow group. Each of the subjects completed a fatigue run that was conducted at VC>2's between 80% and 88% of the runners' V O 2 max • In general these authors found that the runners demonstrated subtle compensations i n running pattern as they approached volitional exhaustion. Specifically, Siler and Mar t in (1991) found that the stride length increased; this finding differs from other fatigue studies that d id not control for speed. In the non-speed-controlled studies previously cited, the authors invariably found that the stride length decreased wi th fatigue. However this finding is confounded by the fact that concomitant wi th stride length decreases came velocity decreases. Furthermore, Siler and Mar t in (1991) found that as the stride length increased there was an accompanying modification in the joint angular kinematics of the thigh and knee. "The range of motion of the thigh was significantly expanded due to a significant increase in the maximum thigh flexion and a trend towards increases in the maximum thigh extension" (Siler and Mart in, 1991, p. 19). In terms of the knee angle the range of motion was unaltered wi th fatigue. This maintenance of knee R O M was a result of parallel changes in knee flexion and knee extension, increased knee flexion and decreased knee extension. Siler and Mart in (1991) concluded that although there were some changes that were measurably different in the fatigued state, many of the kinematic changes seemed to be individual differences rather than group differences. In fact "some individuals appeared to be more sensitive to the effects of fatigue, as indicated by pronounced adaptations in running style in the run to volit ional exhaustion" (Siler and Mart in, 1991, p. 27). -17-The final study that w i l l be discussed in this review of biomechanical fatigue in running articles is one performed by Will iams et al. (1991). In a similar vein as Siler and Mar t in (1991), Wil l iams et al. investigated "the changes in running kinematics as fatigue progressed that were not confounded by accompanying changes in running speed (p. 140)." Interestingly, these authors found that changes in kinematics wi th fatigue during distance running are minimal when data are analysed across a group of subjects. This being said, however, there were significant changes in some of the individual variables, most notably increased step length, maximal angle of the thigh during hip flexion, and the maximal angle of the knee during swing phase. There are two studies that serve as pilot studies to the present investigation. The first of theses studies is one by Black et al. (1993c), in which the authors studied the kinematic and kinetic effects of a V O 2 max test on the lower limb during cycling. In this experiment the athletes were asked to complete an incremental V O 2 max style test while force and video data were being collected. These investigators found that the final minute and the initial minute of the test were very different both kinetically and kinematically. Kinetically, the forces were greater at the end of the incremental test; this of course is to be expected as the power output increases each minute, and therefore the force needed to overcome the increased power output also increased. Interestingly, there were shifts in the locations of the maximum resultant force as wel l as the resultant force component Fz and Fy wi th in the pedal cycle. In each case as the power output increased and the athlete became fatigued the maximum force applied to the pedal occurred earlier in the pedal cycle. Furthermore, the component of the resultant force that is effective in creating torque at the crank axle, the effective force, increased wi th respect to the resultant force. In other words, as the power output increased and the athlete fatigued, the Index of Effectiveness (IEFF) increased from 0.30 in the initial trial to 0.60 i n the final trial. This shift indicates that not only is the athlete increasing the force -18-applied to the pedal but he is also making the application of the force more efficient in terms of the torque that it produces. Other interesting findings in this incremental study were that kinematically there were changes in the joint angles of the lower limb as wel l as pedal angle changes. In fact as the power output increased, and the athlete fatigued, the ankle became significantly more dorsiflexed. There were concurrent changes at both the knee and the hip joint as wel l . These kinematic changes give further evidence for a change in the pattern of force generation as an athlete either fatigues or as a result of increased power output. Therein lies the deficiency of the Black et al. (1993c) investigation. The changes in both the kinematics and kinetics of the incremental exercise could be a result of increased power output, a result of fatigue or some combination of the two. Thus, although there is evidence for an interaction, no real conclusions as to the etiology of the changes can be made. Finally, an investigation by Amoroso et al. (1993) investigated the changes in both kinematics and kinetics of the lower limb during cycling at steady rate. The protocol for this particular study was to induce fatigue in highly trained cyclists during a constant power output, constant cadence ride. Al though the power output used in this study was not sufficient to produce cessation of the exercise in some of the athletes, none the less there were some significant changes in pattern of force generation as wel l as kinematic changes. Amoroso et al. (1993) found that fatigue resulted in greater peak hip extension by 1.5° in the fatigued condition when compared to the non-fatigued condition. Furthermore, although the changes at the knee were not significant, there were changes in the ankle angle. There was a significant shift towards dorsiflexion, as was found in the Black et al. (1993c) study, in the ankle angle and these changes were similar to changes in the pedal angle. These kinematic changes were felt to be an adaptation by the body to fatigue. -19-A comparison of the force profile by Amoroso et al. (1993) showed that, concomitant wi th the changes in the kinematics, there were changes i n the kinetic patterns. These investigators found that wi th fatigue came an increase i n the maximum normal force (from 339 N in the non-fatigued trial to 369 N in the fatigued trial) as wel l as shifts in the tangential force. The shear force showed not only a shift in the maximum and minimum forces, but also a t iming shift. Dur ing the fatigued trial the maximum anterior/posterior force decreased and occurred earlier in the pedal cycle and the min imum anterior/posterior force was more negative in the fatigued trial than in the non-fatigued trial and also occurred earlier in the pedal cycle. Amoroso et al. (1993) felt that these changes were linked to the changes in the pedal angle and ankle angle/and were a result of "adaptations in the body's mechanical response to the increasing demands placed on it" (Amoroso et al., 1993). It is clear that wi th fatigue that the body makes adaptations. However, as a result of the inadequacies of the reviewed literature, especially the cycling literature, it is also clear that this area of biomechanics has not been fully explored. Biomechanics of Cycl ing The biomechanical literature as it pertains to cycling has grown quickly in the last 20 years. The use of the bicycle ergometer in both metabolic and biomechanical studies has increased as a result of both the ease of data collection wi th the ergometer and the cyclical pattern of the motion, which is similar to other activities, such as walking and running. The mechanics of force application in cycling dates back to the late 1800's, in research by Sharp. Sharp was able, through the use of a spring loaded pedal and smoked drum, to determine the vertical forces applied to the pedal in cycling. It was not, however, unti l the late 1960's that pedal mechanics were dealt wi th seriously again. Hoes et al. (1968) designed a device to measure the normal force applied to the pedal. They found that during the recovery phase of the pedal cycle there were -20-retarding forces. A s the technology of pedal force measurement advanced it was found by Soden and Adeyefa (1979) that during low load conditions, such as the ones used by Hoes et al. (1968), there was little or no pul l ing up on the pedal in the recovery phase of the cycle, and hence a retarding force. However during high load conditions, such as sprint starting and h i l l climbing, Soden and Adeyefa (1979) found that there was in fact pul l ing up in the recovery phase of the pedal cycle. Previously D a l Monte et al.(1973) mentioned this effect, however, their methods and presentation are relatively suspect as they used illogical units of measure and only one subject. Further evidence of this pul l ing up in the recovery phase was provided by Davis and H u l l (1981) . These researchers found that wi th the aid of toe clips and straps there was pul l ing up in the recovery phase which, they stated, increased the section over which effective positive torque producing force could be applied by allowing greater plantar flexion and shear loads in the beginning of the cycle. It was these initial kinetic studies that led to the use of inverse dynamics for the calculation of joint moments in the cycling literature. In the mid 1980's the study of joint moment patterns in cycling became of interest to a few researchers, for a couple of reasons. The first of these reasons is that in the early 1900's W.P. Lombard studied and defined the actions of two joint muscles and found that the simultaneous activity of these muscles proved paradoxical. The paradox was exemplified in motions such as rising from a chair, simple flexion and extension of the leg, and cycling (Andrews, 1987). Lombard's Paradox, as it has been aptly named, is essentially the contradictory activities of the quadriceps and the hamstring muscles -21-Hip Moment 6 Z C 01 s I g o 90 135 180 225 270 315 360 Crank Angle (deg) Knee Moments 90 135 180 225 270 315 360 Crank Angle (deg) Ankle Moments 0 45 90 135 180 225 270 315 360 Crank Angle (deg) Figure 3. H i p , knee, and ankle moments reported by Gregor et al. (1985) , H u l l and Jorge (1985) and Ericson (1986) . (Adapted from Ericson, 1986) groups in flexion and extension activities. For example the quadriceps produce both flexion of the hip and extension of the knee, whereas the -22-hamstrings produce hip extension and knee flexion, both examples prove paradoxical in the actions of these two muscle groups. Lombard's paradox was looked at in the cycling model by Andrews (1987) in a kinematic analysis. Kinetically, Lombard's paradox was examined first, however, by Gregor et al. (1985) in an attempt to define the muscle function across the hip and the knee joints and relate their activity to the joint moments produced across these joints. Gregor et al. (1985) found that during the first 180° of the pedalling cycle the hip moment was always extensor, whereas the knee moment was first extensor but became flexor at approximately 100° of the cycle (Figure 3). The net torque, or moment, about the ankle was plantar flexor for the first 200° of the pedal cycle and close to zero for the remainder of the cycle. These data are the results of five subjects cycling at a cadence of 60 R P M for four minutes at a power output of 160W. The second major reason for the study of joint moments is one of understanding the relative loads that contribute to injury and for the sake of modeling the lower extremity. Figure 3 represents graphically the data presented in the three main works in the fields of joint moments in cycling. The first of these sets of data has been discussed, Gregor et al. (1985). The second and third sets of data represent the work of H u l l and Jorge (1985) and of Ericson (1986). In each case the general pattern of joint torque remains consistent even though the cadences and power outputs differ between studies. H u l l and Jorge (1985) were interested in determining the functional role of the muscles that participate in pedalling. Using one subject these authors attempted to compare variations of power output and cadence to a reference case in which the cadence was 80 R P M wi th a gear ratio of 52 X 19 ( there was no measure of power output). The data from Case 1 of the H u l l and Jorge study are presented in Figure 3 and correspond closely to the other data presented. H u l l and Jorge found that when the power output was increased and the cadence was decreased there was a tendency for all the joint moments of the lower l imb to increase (increased maxima and more negative minima). When the cadence increased and the power output decreased the maximum joint moments decreased and the min imum joint moments moved closer to zero; in other words the range of the joint moments decreased. When cadence remained constant and the power output was increased there was a trend to increased maxima and minima joint moments at each of the joints and this led to a larger range in the joint moments through the pedal cycle. Finally, when the cadence remained constant, 80 R P M , and the power output decreased there was a decrease i n the ranges of the moments at each joint (Hul l and Jorge, 1985). The final group of data presented in Figure 3 are those of Ericson (1986). Using eleven male subjects, Ericson perturbed the system in a number of ways. The Ericson data in Figure 3 represents a power output of 120 W and 60 R P M . Aga in the data presented for all three studies are relatively similar in shape. Ericson (1986) - 2 3 -looked at the effects of increased workload, from 0 watts to 240 watts, increased pedalling rate, from 40 R P M to 100 R P M , and finally increased saddle height, from 102% of the distance between the ischial tuberosity to the medial malleolus, to 120% of the same distance. These authors found that increased workload increased the hip flexor and extensor moments, the knee flexor and extensor moments and the ankle dorsiflexor moment. Increased cadence had significant increases in the hip flexor and extensor moments, and the knee extensor moments, but had no significant effect on the knee flexor or the ankle dorsiflexor moments. Increasing the saddle height decreased the hip extensor and the knee flexor moments and increased the knee extensor moment, but had no effect on the hip flexor or the ankle dorsiflexor moments. The results of a number of cycling-related joint moment investigations have been reviewed in this section. It is clear that none of the literature has looked at the effects of fatigue on the magnitude of the joint moments of the lower l imb during cycling or on the relative timing changes that may result from fatigue. These changes are the focus of the present study and the procedure for determining these joint moment changes w i l l be discussed in the following section. - 2 4 -Chapter 3 P R O C E D U R E S Subjects. The study group consisted of 12 male cyclists wi th racing experience. Selection was based on competitive level and training background wi th the criteria of current Canadian Cycl ing Association (CCA) road racing license of category three or higher (1 or 2) or C C A mountain bike racing license of Sportsman, Expert or Elite. It was felt that athletes wi th these credentials would be able to complete the test protocol, were familiar wi th r iding at or above their anaerobic threshold and were sufficiently wel l trained. A l l persons were made fully aware of the experimental details prior to assuming their involvement in the program. They were informed that they were able to withdraw from the testing procedure at any time without prejudice and signed informed consent which conformed wi th the ethical guidelines of the Universi ty of British Columbia. Instrumentation and Data Collection Bicycle. A standard bicycle instrumented wi th two piezo-electric triaxial force transducers (Kistler 9251A) in the right pedal was used in this study. A continuous output potentiometer monitored pedal angle and an optical encoder monitored data collection and top dead centre (TDC). The T D C encoder triggered the data collection, which lasted for 3.5 seconds. Thus at 90 R P M approximately five revolutions of force data were collected. The bicycle was mounted on a Schwinn Velodyne® electronically braked cycle ergometer. The Schwinn Velodyne® simulates inertial characteristics of road r id ing and modulate given power outputs based on cadence. The bicycle was set up so that it would match, as closely as possible, the athletes own bicycle wi th one exception. This exception was that the bicycle seat height was set to 100% trochanteric height. This height was found to be optimal in terms of oxygen consumption by Nordeen-Snyder (1977). - 2 5 -2-D Video. The kinematics of the limb segments were recorded at 60 H z using a Panasonic video camera (WDV 5100), wi th the lens axis oriented perpendicularly to the sagittal plane of the rider. The pattern of limb motion was defined by highly reflective markers placed at key landmarks. These landmarks were on the right side of the body and included: front and centre of the toe clip, lateral aspect of the fifth metatarsal head, posterior lateral aspect of the calcaneus, lateral malleolus, and the greater trochanter of the femur. Three cycles from each minute of the test were digitized using the Peak Performance Technologies software package. Two-dimensional coordinates for the limb segment markers were calculated as wel l as joint angles, angular velocities and angular accelerations. These coordinate data were then time matched to the corresponding force data file for subsequent processing. The angles within this frame of reference were as follows: the hip angle was defined as the angle between the thigh segment and the vertical; the knee angle was the angle between the shank segment and the vertical; and the ankle angle was the angle between the foot and the vertical. The marker placements and the joint angles are pictorially represented in Figure 4. Figure 4. Pictorial representation of the marker placement and the joint angle definitions (adapted from P015 Manual by D . Derosa). -26-Force Data Collection. Force data were recorded continuously at a rate of 240 H z . Data collection was triggered by the optical encoder indicating T D C . Force data collection occurred at the end of each minute of the test protocol and lasted for 5 revolutions of the crank (approximately 3.5 seconds) per collection period. The outputs from the two force transducers were summed prior to amplification, translated from analogue to digital (12-bit), and then stored for subsequent processing. Metabolic Data Collection. Metabolic data were measured wi th the Rayfield open circuit system. This system consists of the Rayfield computational software, a Rayfield gasmeter, which measures ventilation and volume, an Amtek O2 analyser and a Beckman C O 2 analyser. The total system measures V O 2 , V C O 2 , ventilation ( V E ), respiratory rate (RR), tidal volume (Vy), as wel l as heart rate (HR) and plots these data in 15 second intervals. This equipment was used for only the initial test protocol. Experimental Protocol Initial Visit . Each subject completed a five minute warm-up period at the cadence and power output of his choice. The test began at a power output setting of zero on the Velodyne and a cadence of 90 R P M , giving an effective power output of approximately 60 watts. After the first minute the power output of the Velodyne was increased to 100 watts. A t the end of each subsequent minute interval the power output was increased by 25 watts per minute until such time as the athlete was unable to maintain the test cadence ( 90 ± 5 RPM) . Metabolic data collection took place at 15 second intervals. This initial visit was used as a guide for the second visit. In the initial test the maximum power output of each subject was determined. This determination of the -27-maximal power output was then used to calculate the appropriate power output of the bicycle ergometer for the second visit. Second Visit: In the second visit the athlete completed a steady rate ride at 80% of his maximum power output (80% max P.O.) in the initial V02max test. This ride began wi th the rider choosing a power output and cadence of his choice for a five minute warm-up period. A t the end of the warm-up interval the power output setting on the Velodyne was set corresponding to the 80% of the athlete's maximum power output determined in the initial visit. The test continued unti l the athlete was no longer able to maintain the test cadence of 90 ± 5 R P M . Video data and force data were collected at the end of each minute. Third Visit: In the third visit the athlete completed a steady rate ride at 30% of his maximum power output (30% max P.O.) in the initial V02max test. This ride began wi th the rider choosing a power output and cadence of his choice for a five minute warm-up period. A t the end of the warm-up interval the power output setting on the Velodyne was set corresponding to the 30% of the athlete's maximum power output determined in the initial visit. The test continued for the same length of time as the second visit, for example if the athlete was able to maintain 80% max P.O. for 12 minutes then the third test would continue for 12 minutes. Video data and force data were collected at the end of each minute as was the protocol i n the 80% max P.O. test. Data Analysis Kinematic Data. Kinematic data were digitized using the Peak Performance Technologies software package. This software computed joint angles, angular velocities, and angular accelerations as wel l as linear displacements, velocities and accelerations of limb segments. Kinematic data were filtred at 6 H z . Because of small variations in cadence, joint angular displacements and joint angular velocities were normalized to 38 points for one cycle from T D C to T D C using in-house software. Three cycles in each condition were averaged wi th in - 28 -subjects. These Intra-subject averages were then averaged across all subjects for both the initial and the final minutes of both steady rate exercise tests and were then plotted. Maximums, minimums, and ranges of angular displacements of each joint were then calculated for each subject using the average of three pedal cycles. The relative positions within the pedal cycle (percent cycle) of the maximums and minimums were also calculated and analysed. Two-tailed paired Student's t-Tests were used to compare the maximums, the minimums and the ranges of the init ial test trial to the final test trial. One-tailed paired Student's t-Tests were used to compare the timings of the maximums and the minimums wi th in the pedal cycle of the initial test trial to the final test trial. Significance was set at p< 0.05. Kinetic Data. Normal force (Fz) and tangential force (Fy) data were collected during the two steady rate exercise tests at the end of each minute, time matched to the kinematic data and normalized to match the per cycle number of kinematic data points. From the Fz and Fy data resultant forces (Fr) were calculated, the effective components (Fe) of the Fr were calculated as was the instantaneous index of effectiveness and the overall index of effectiveness (IEFF) for the entire pedal cycle. Furthermore, torque, total work, positive work, negative work, power, angular impulse, positive angular impulse, negative angular impulse and positions of where the negative angular impulse began and ended within the pedal cycle were also calculated for each of the initial and the final minutes of both steady rate test protocols. A l l the kinetic data were ensemble averaged in order to compare the initial and final minutes of the tests. Statistical tests of the force data used one-tailed paired Student's t-Tests to compare the maximums and the positions of the maximum forces (Fz, Fy, Fr, Fe and torque) within the pedal cycle of the initial test trial to the final test trial in each of the two steady rate test protocols. The initial and final minute values of the IEFF, angular impulse, positive impulse, negative impulse, as wel l as the positions of the -29-beginning and end of the negative impulse and range of the negative impulse wi th in the pedal cycle were compared using one-tailed paired Student's t-Tests. The init ial and final minute values of the total work, positive work, negative work, and power output two-tailed paired Student's t-Tests. Significance was set at p< 0.05. Toint Moments. Calculations of joint moments were done using in house software. These moments were calculated using the inverse dynamics approach which is used by others in the literature (Gregor et al., 1985, Redfield and H u l l , 1984 a & b, H u l l and Jorge, 1985, and Black, 1993b). The propulsive moment (Winter, 1980) was calculated by summing the extensor moments of the ankle, knee and hip. Joint moments for three cycles of data were averaged for each subject. Ensemble averages were then used in the calculations of the total average ankle, knee, hip and propulsive moments. One-tailed paired Student's t-Tests were used to compare the maximum flexor and extensor moments, and the timings of these maxima wi th in the pedal cycle between the initial and final minutes in each of the two steady rate testing protocols. Significance was set at p< 0.05. Within-subject comparisons. Winter's (1983, 1984) Coefficient of Variation (CV) was used to compare within-subject variations. The C V was used to compare patterns of force application, joint angle patterns and joint moment patterns. The C V ' s from the initial trials were then compared to the CV ' s of the final trials to determine whether there was an increase in the variability as the athlete began to fatigue, or in the case of the 30% max P.O. test, whether the athletes remained consistent in the patterns of motion in terms of variability changes. One-tailed Student's t-Tests were again used for these comparisons. Significance was set at p<0.05. Between power output comparisons. In order to determine whether the changes seen in the 80% max P.O. trials are different from the changes seen in the 30% max P.O. trial repeated measures A N O V A ' s were used. The following comparisons were made for the initial and final minutes of both the 30% max P.O. - 3 0 -trial and the 80% max P.O. trial: maximum Fy, Fz, Fr, Fe as wel l as the timings of these variables within the pedal cycle, IEFF, angular impulse, positive and negative angular impulses as wel l as the maximum joint moments and timings of the m a x i m u m joint moments. -31-Chapter 4 R E S U L T S Description of Subjects. Twelve trained male cyclists were subjects for this study. The subject's descriptive data are presented in Table 1. The mean age, height, and weight were 27.75 ± 6.77 years, 179.28 ± 7.52 cm, and 75.05 ± 11.15 kg respectively. The subject's average training distance per week and years cycling competitively were 315.45 ± 159.52 k m and 5.83 + 3.38 years respectively. Group Metabolic Data Individual metabolic data are presented in Table 2. Incremental Exercise test. The average maximum power output, max V 0 2 (ml /kg /min ) , absolute V 0 2 were 412.5 ± 54.88 Watts, 66.7 ± 6.66 m l / k g and 4.95 + 0.63 1/min respectively. Heart rates for the incremental V02max test reached 186.25 ± 4.45 beats per minute (bpm) which on average was 97% of the athletes projected maximum heart rate of 192.25 ± 6.47 bpm. This maximum heart rate was calculated based on the rule that maximum heart rate is equal to 220 minus the athlete's age. Steady rate test at 80% max power output: Table 3 presents the power outputs and percentages of max P.O. for each subject. The average power output setting, time to fatigue and maximum heart rate was 331.31 ± 41.46 watts, 13.04 ± 5.17 minutes and 181.6 ± 6.60 bpm respectively. The maximum heart rate was 94.6% of the maximum attainable heart rate of 192.6 ± 6.99 bpm. Steady rate test at 30% max power output: The average power output for the 30% max P.O. test was 125.00 ± 18.46 watts. This power output setting elicited an average maximum heart rate of 120.9 ± 5.86 bpm which was 63.0 ± 2.9% of the athlete's maximum heart rate. - 3 2 -Table 1. Subject descriptive data Subject # Cycling Cat. Age Mass Height Years Dist/week (Yrs) (Kg) (m) Cycling (km) 1 C C A M t n Exp 26 81.7 183.7 3 120 2 C C A 2 / M t n E l . 22 74.65 178 4 350 3 C C A M t n Sport 25 75.55 187.1 5 140 , 4 C C A Vet B 46 88.52 181 15 260 5 Elite Duathlete 24 54.4 163.6 4 200 6 C C A 2 23 73.5 177.8 3 500 7 C C A 1 27 88.5 189.1 4 600 8 C C A 4 / M t n Exp 23 74.3 181.3 8 250 9 C C A 2 30 87.6 187.3 7 500 10 C C A 3 28 78.5 179.6 4 11 C C A 1 32 63.4 171.2 8 350 12 C C A 1 27 60 171.6 5 200 Average 27.75 75.05 179.28 5.83 315.45 S T D E V 6.47 11.15 7.52 3.38 159.52 Table 2. Incremental V O 2 max data Subject # Max P.O Max Vo2 Max Vo2 (W) (ml/kg/min) (1/min) STPD 1 425 67.56 5.52 2 475 65.51 4.89 3 425 63.27 4.78 4 375 50.27 4.45 5 325 69.49 3.79 6 450 68.71 5.05 7 525 68.14 6.03 8 375 67.43 5.01 9 400 66.1 5.79 10 350 63.69 5.00 11 425 79.5 4.87 12 400 70.67 4.24 Average 412.50 66.70 4.95 S T D E V 54.88 6.66 0.63 - 3 3 -Table 3. Steady rate, constant power output tests data Subject # "80% Max" P.O. % of max P.O. "30% Max" % of max P.O. setting (W) P.O. setting (W) 1 375 88.2 125 29.4 2 375 78.9 150 31.6 3 375 88.2 125 29.4 4 325 86.6 100 26.7 5 250 . 76.9 100 30.8 6 350 77.7 150 33.3 7 425 80.9 150 28.6 8 300 80.0 125 33.3 9 300 75.0 125 31.3 10 300 85.7 100 28.6 11 325 76.5 125 29.4 12 300 81.3 125 31.3 Average 331.25 81.33 125.00 30.30 S T D E V 41.46 4.72 18.46 1.99 - 3 4 -Group Kinematic Data Mean angle profiles for the ankle, knee and the hip for the 80% max P.O. trial are presented in Figures 5 a-b and 6 a-b. Each curve represents the ensemble average of 3 trials collected from each of the 12 subjects. The abscissa of all plots are expressed wi th respect to the position of the crank angle within the pedal cycle. The initial minute of the test was compared to the final minute of the test, or more simply the unfatigued state compared to the fatigued state. The ankle angle (Figure 5a) shows that the foot initially dorsiflexed as the crank angle approached 90° and then became plantarflexed as the crank angle approached the 270° position. The knee angle (Figure 5b), or the shank angle was calculated wi th respect to the vertical, wi th the angle becoming more negative as the shank moved clockwise wi th respect to the vertical. Using this reference frame the shank approached the vertical as the crank angle approached bottom dead centre (BDC) of the pedal stroke, which intuitively is correct. During the second half of the pedal stroke the knee angle became more flexed to a maximum flexion angle near 300° of crank angle. Visual inspection of the graphs revealed relatively small differences between the initial and the final minutes. Statistically, comparisons were made between the maximum extensions and flexions and the t iming of these maxima wi th in the pedal cycle. Significant differences were seen in the maximum hip extension and the maximum knee flexion, wi th the hip angle becoming more extended in the final trial when compared to the initial trial, 24.51 ± 1.8° initial to 21.82 ± 2.55° final (p<0.05), and the knee angle becoming more flexed, -46.74 ±3 .05° initial to -47.86 ± 4.14° final (p<0.05). There were differences in the timings of the maximum joint angles wi th the maximum knee flexion angle, and the maximum ankle dorsiflexion angle occurring later in the pedal cycle in the final trial than in the initial trial, (p<0.05). There was an increase in the range of motion at the hip joint in the final trial, 41.38 ± 2.25° in the initial trial to 43.87 ± 2.49° in the final trial (p<0.05). - 3 5 -Average Foot Angle 0 90 180 270 360 Crank Angle (deg) Average Shank Angle O n 0 90 180 270 360 Crank Angle (deg) Figure 5. (a) Average foot angle, ± 1 stdev, and (b) average shank angle, ± 1 stdev, for the 80% max P.O. trial -36-Average Thigh Angle 70-1 CD CD a> O) c < 20-Initial Minute Final Minute 90 180 270 Crank Angle (deg) c o X CD O Q 360 Average Pedal Angle b CD c < Figure 6. -1 0 90 180 Crank Angle (deg) 270 360 (a) Average thigh angle, ± 1 stdev, and (b) average pedal angle, ± 1 stdev, for the 80% max P.O. trial - 3 7 -Figures 7 a-b and 8 a-b present the 30% max P.O. data graphically. In the 30% max P.O. trial there were some interesting results. It was hypothesized that there would be no differences between the initial and the final trials for this particular set of tests because the athlete would not be stressed. There were however significant differences in the angle data. Max imum hip extension, knee flexion and ankle dorsiflexion angles all changed in the final minute of the test when compared to the initial minute of the test. The maximum hip extension angle changed from 68.51 ± 2.46° in the initial trial to 69.40 ± 3.00° in the final trial (p<0.05). The maximum knee flexion angle increased from -5.72 + 2.27° in the initial trial to -6.44 ±2.45° in the final trial (p<0.05). Finally the maximum ankle dorsiflexion angle decreased from 60.13 ± 5.92° in the initial trial to 57.66 ± 5.15° in the final trial (p<0.05). There were also changes in the timings of the maximum angles wi th in the pedal cycle wi th the maximum knee extension angle occurred earlier in the pedal cycle in the final trial than in the initial trial (p<0.05). There were no other significant changes in the angular data collected in the 30% max P.O. test. - 3 8 -Average Foot Angle 0 90 180 270 360 Crank Angle (deg) Average Shank Angle b TO CD T3 TO C < 0 90 180 270 Crank Angle (deg) g 'w c CD X LU 360 Figure 7. (a) Average foot angle, ± 1 stdev, and (b) average shank angle, ± 1 stdev, for the 30% max P.O. trial -39-Average Thigh Angle 0 90 180 270 360 Crank Angle (deg) Average Pedal Angle 60-I 0 90 180 270 360 Crank Angle (deg) Figure 8. (a) Average thigh angle, ± 1 stdev, and (b) average pedal angle, ± 1 stdev, for the 30% max P.O. trial -40-Group Pedal Force Data 80% max P.O. The group mean profiles for the vertical (Fz) and the anterior-posterior (Fy) components and the resultant (Fr) pedal forces are presented in Figures 9 a-c. In addition the effective component of the resultant pedal force (Fe), as wel l as instantaneous index of effectiveness (IEFF) and torque are plotted in Figures 8 a-c. In each of the graphs the initial minute is compared to the final minute of the trial . In general there were significant differences between the initial and the final trials of the 80% max P.O. trials in most of the maximum forces that were compared. It was hypothesized that as the athlete fatigued he would increase the maximum force applied to the pedal and change the timing such that the maximum force occurred earlier in the pedal cycle. The results of the statistical tests do not necessarily support these hypotheses. There were no differences i n the maximum anterior-posterior forces or the timing of these forces when the first and last trials were compared. The maximum Fz force was greater in the final trial (393.56 ± 80.23N) than in the initial trial (354.38 ± 66.08N), (p<0.05); the timings of these maximums were, however, not different. There was a similar increase in the maximum Fr between the initial and the final minutes; i n the init ial minute of the ride the max Fr was 356.64 ± 67.3 N and increased to 393.57 ± 74.16 N in the final minute (p<0.05). These changes are reflected in the some of the calculated variables from the pedal forces, such as work and angular impulse. Al though there were no changes in the total power output of the rider or changes in the total work or angular impulse there were changes in the components of these measures. Positive work, negative work, positive angular impulse, negative angular impulse al l increased in the final minute when compared to the initial minute. Positive work increased from 118.62 ± 19.02 J in the initial minute to 127.751 ± 20.3 J in the final minute (p<0.05). Negative work in the initial minute of the test averaged -5.99 ± 6.88 J to -9.88 ± 6.37 J (p<0.05) in the final minute. Finally, the positive angular -41-impulse increased from 28.319 ± 4.54 Ns to 30.498 ± 4.85 Ns (p<0.05) and negative impulse increased from-1.43 ± 1.63 Ns to -2.34 ± 1.5 Ns (p<0.05) in the final minute. -42-b Average Fz 3 200-J Initial Minute — - Final Minute 90 180 Crank Angle (deg) Average Fy 270 CD O i r 90 180 Crank Angle (deg) Average Fr 270 180 Crank Angle (deg) 270 360 360 360 Figure 9. (a) Average normal pedal force, ± 1 stdev, (b) average tangential pedal force, ± 1 stdev, and (c) average resultant pedal force, ± 1 stdev, for the 80% max P.O. trial - 4 3 -Average Fe 400-, 0 90 180 270 360 Crank Angle (deg) Figure 10. (a) Average effective pedal force, ± 1 stdev, and (b) average index of effectiveness, + 1 stdev, for 80% max P.O. trial Figures 10 a-b represent the ensemble averages of the effective force component of the resultant pedal force (Fe), and instantaneous index of effectiveness (IEFF) throughout the pedal stroke. In the Fe graph the final minute of the test yielded a greater maximum than the initial minute of the test, Fe increased from 339.57 ±65.28 N to 375.5 ± 74.16 N (p<0.05) and, although not presented graphically, the torque increased from 57.73 ± 11.1 N m to 63.835 ± 12.7 N m (p<0.05). Finally the instantaneous index of effectiveness is presented to give a sense of where the riders are effective in the application of force within the pedal cycle and where they are ineffective. In both the initial and the final trials the athletes are effective for the - 4 4 -first 250° of the pedal cycle. In the final trial, however, the athletes have a greater ineffectiveness and they are ineffective for a greater portion of the pedal cycle. 30% max P.O. In the 30% of Max power output trials there were changes as wel l . Figures 9 a-c are ensemble averaged Fz, Fy and Fr. Al though there were differences in the magnitude of the maximums in the 80% max P.O. trials there were no such changes in the 30% max P.O. trials when comparing the init ial minute to the final minute of the test. There were, however, changes in the timings of the maximum forces within the pedal cycle when the initial trial was compared to the final trial. The maximum Fz and the maximum Fr occurred later in the pedal cycle of the final trial than in the initial trial (p<0.05). Al though there were few differences in the force data , there were some significant changes in most of the variables calculated from the forces. There were changes in the IEFF, which decreased in the final trial to 0.433 ± 0.005 from an initial value of 0.500 ± 0.011 (p<0.05), and the single leg power output of the riders was also changed in the final minute. The single leg power output produced by the athletes decreased from 81.49 ± 14.18 W to 74.94 ± 11.83 W in the final minute of the test (p<0.05). - 4 5 -Average Fz 2 5 0 n 0 90 180 270 360 Crank Angle (deg) Average Fy 8 0 n T i M T 0 90 180 270 360 Crank Angle (deg) Average Fr 2 5 0 n 0 90 180 270 360 Crank Angle (deg) Figure 11. (a) Average normal pedal force, ± 1 stdev, (b) average tangential pedal force, ± 1 stdev, and (c) average resultant pedal force, ± 1 stdev, for the 30% max P.O. trial -46-There was an overall decrease in the total angular impulse that was applied to the pedal in the final minute of the test. The total angular impulse dropped from 12.97 ± 2.25 N s in the initial minute to 11.93 ± 1.89 Ns in the final minute (p<0.05). This change was induced by an increase in the negative impulse as there were no changes in the average positive impulse. The negative impulse increased negatively from -2.95 ± 1.37 Ns in the initial minute of the test to -3.68 ± 1.5 N s in the final minute of the test (p<0.05). Furthermore, the t iming of the negative impulse wi th in the pedal cycle changed such that the beginning of the negative impulse occurred earlier in the pedal cycle in the final trial, 210.78 ± 7.07°, than in the initial trial, 216.3 + 11.3° (p<0.05). Although there were no t iming changes in the end of the negative impulse within the pedal cycle, the changes in the onset of negative impulse were great enough to increase the range of the negative impulse wi th in the pedal cycle (114.47 ± 29.8° initial to 127.1 ±21.7° final, p<0.05). Figures 12 a-b present the ensemble average effective force component of the resultant force and the instantaneous index of effectiveness for the 30% of max power output test. The effective force showed no changes between the initial and the final minutes in terms of maximums or the timings of the maximums wi th in the pedal cycle. Furthermore, there were less pronounced changes in the shape of the instantaneous index of effectiveness when compared to the differences seen in the 80% max P.O. trial as previously discussed. -47-Average Fe 0 90 180 270 360 Crank Angle (deg) Figure 12. (a) Average effective pedal force, ± 1 stdev, and (b) average index of effectiveness, ± 1 stdev, for 30% max P.O. trial Group Average Joint Moment Data The group mean profiles for the ankle, knee, hip and propulsive moments are presented in Figure 13 a-b and 14 a-b. The average ankle moment is plantarflexor for the entire pedal cycle. In the initial stages of the pedal cycle the ankle moment is small and then just after 90° reaches a maximum plantarflexor moment. The knee moment initially begins the cycle in an extensor moment and then at approximately 90° of crank angle becomes flexor. The knee moment remains flexor unt i l 300° where it again becomes extensor. The maximum extensor moment - 4 8 -occurs at approximately 20° and the maximum flexor moment at approximately 150° of crank angle. The hip moment is extensor for the first 215° of the pedal cycle and then becomes flexor until approximately 300° where it again becomes extensor, the maximum hip extensor moment occurred at approximately 100° of crank angle, and the maximum hip flexor moment occurred at approximately 275° of crank angle. Statistically there were changes in some of the maximum and m i n i m u m joint moments. The maximum ankle plantar flexor moment, knee flexor moment and hip extensor moments were all significantly greater in the final minute of the 80% max P.O. trial than in the initial minute of the test. The maximum ankle plantarflexor moment, maximum knee flexor moment, and the maximum hip extensor moment increased from -41.58 ± 8.17 N m , -52.82 ± 11.78 N m , and -79.96 ± 20.70 N m to -48,68 ± 10.31 N m , -59.17 ± 11.48 N m and -94.28 ±17.29 N m respectively. Contrasting these increases in the maximum moment in the final minute of the test, there was a decrease in the maximum hip flexor moment in the final minute of the 80% max P.O. test from 18.66 ± 3.24 N m to 17.54 ± 3.46 N m in the final minute (p<0.05). There were only two changes in the timings of the maximum joint moments wi th in the pedal cycle. The maximum plantarflexor moment occurred later in the pedal cycle in the final minute than in the initial minute, 326.05 ± 30° to 333.95 ± 33.95° (p<0.05). The maximum hip flexor moment occurred earlier in the final minute of the test than it d id in the initial minute of the test, 109.74 ± 15.33° to 95.53 ± 19.13° (p<0.05). -49-b c CD E o c o c <u E o Average Ankle Moment 1 0 01 -10 -20 -30 -40 -50 -60 0 Initial Minute Final Minute o x cu o Q 90 180 Crank Angle (deg) 270 360 Average Knee Moment 50 70 o CO c 3 >< HI 0 ~1 1— 90 180 270 Crank Angle (deg) 360 Figure 13. (a) Average ankle moment, ± 1 stdev, and (b) average knee moment, ± 1 stdev, for 80% max P.O. trial, initial and final minutes. - 5 0 -Average Hip Moment 40i -1 2 0 i 1 1 1 1 0 90 180 270 360 Crank Angle (deg) Average Propulsive Moment 150n -5 On 1 1 1 1 0 90 180 270 360 Crank Angle (deg) Figure 14. (a) Average hip moment, ± 1 stdev, and (b) average propulsive moment, ± 1 stdev, for 80% max P.O. trial, initial and final minutes. -51-30 % max P.O. The 30% max P.O. trials showed patterns similar to those from the 80% max P.O. trials, and different maxima and minima as expected. Statistically there are changes in the maximum dorsiflexion angle, maximum knee extension moment, and the maximum hip flexion moment. Figures 15 a-b and 16 a-b show ankle dorsiflexor moment decreased from 1.1 ±2.37 N m to 0.16 ± 1.98 N m , the knee moment decreased from 18.29 ± 4.65 N m to 14.11 ± 4.48 N m and the hip flexor moment decreased from 16.10 ± 2.19 N m to 14.45 ± 3.46 N m in the final minute of the test ride. There were two changes in the timings of the maximum moments. These changes were an earlier maximum hip flexor moment in the final trial, 280.26 ± 6.33° in the initial minute to 274.73 ± 9.03° in the final minute, and a later onset of the maximum plantarflexor moment in the final trial, from 126.31 + 19.92° in the initial trial to 135.79 ± 19.92° in the final trial. There were no other significant changes in the joint moments of the 30% max P.O. trials -52-c CD E o c 'o b Average Knee Moment Initial Minute Final Minute o 03 c cu >< LU 180 Crank Angle (deg) 270 360 Figure 15. (a) Average ankle moment, ± 1 stdev, and (b) average knee moment, ± 1 stdev, for 30% max P.O. trial, initial and final minutes. -53-Average Hip Moment 2 0 n 0 90 180 270 360 Crank Angle (deg) Average Propulsive Moment LU 0 90 180 270 360 Crank Angle (deg) Figure 16. (a) Average hip moment, ± 1 stdev, and (b) average propulsive moment, ± 1 stdev, for 30% max P.O. trial, initial and final minutes. - 5 4 -With in Subject Variabil i ty Winter's Coefficient of Variation (CV) was used to describe the consistency of the patterns of motion for intra-subjects trials. This part of the results section w i l l be broken into three categories: forces, angles and joint moments. Average intra-subject variabilities in the limb patterns of motion as defined by joint angles are presented in Table 4. A comparison of the initial and final minutes of the 80% max P.O. trial showed a significant increase in the variability between trials at all three joints (p<0.05). The 30% max P.O. trial however d id not show the same increases in variability between the initial trial and the final trial. Table 4. Mean Intra-subject C V ' s for joint angles 80% Max P.O Hip Angle Knee Angle Ankle Angle Initial Minute 1.84 ± 0.7% 3.37 ± 1.4% 3.44 ±1 .4% Final Minute 3.22 ± 7.5% * 4.75 ± 1.8% * 4.83 ± 1.8% * 30% Max P.O. Initial Minute 2.09 ± 0.7% 3.86 ± 1.1% 4.34 ± 2.2% Final Minute 1.68 ± 0.5% 2.83 ± 0.9% * 3.73 ± 1.0% * indicates a significant difference between the initial and final minutes of the test (p<0.05). Table 5 presents the C V ' s for intra-subject pedal forces. Al though there is a trend towards increased variability, none of the changes reported are significantly different. This indicates that the pattern of force generation is consistent between pedal strokes in the 80% max P.O. trials. A n interesting finding in the 30% max P.O. trial was that as the test continued there was a decrease in the variability between trials for the Fz, Fr and Fe (p<0.05). Table 5. Mean Intra-subject C V ' s for forces. 80% Max P.O F V Fz Fr Fe Initial Minute 21.2 ± 6.0% 10.4 ± 6.3% 8.6 ± 3.7% 11.5 ± 3.9% Final Minute 27.5 ± 1.9% 11.3 ± 9 . 7 % 9.5 ± 8.0 % 12.6 ± 7 . 0 % 30% Max P.O. Initial Minute 26.4 ± 16.5% 13.9 ± 10.0% 12.1 ± 8.2 % 13.9 ± 8.0% Final Minute 25.1 ± 13.9% 9.7 ± 5.5% * 8.7 ± 4.2 % * 11.6 ± 5.7% * * indicates a significant difference between the initial and final minutes of the test (p<0.05). -55-Table 6 summarizes the changes in the C V ' s of the joint moment patterns. Again , as wi th the forces, there is a trend toward greater variability in the 80% max P.O. trial i n the final minute as compared to the initial minute, however these changes are not significant. A s wi th the forces in the 30% max P.O. trial there are decreases in the variability in the patterns of joint moments in the hip (p<0.05) and trends towards decreased variability in the other joints as wel l as in the propulsive moment . Table 6. Mean intra-subject C V ' s for joint moments 80% Max P.O Hip Mom Knee Mom Ankle Mom Prop Mom Initial Minute 12.7 + 5.4% 13.7 ± 3.1% 9.8 ± 4.3 % 13.5 ± 5.7% Final Minute 12.9 ± 10.9% 16.1 ± 7.0 % 11.0 ± 7.5 % 14.8 + 12.2% 30% Max P.O. Initial Minute 22.3 + 11.4% 19.3 + 9.7% 13.83 ± 8.3 % 25.5 + 15.2% Final Minute 16.9 + 7.7% * 18.8 + 5.9% 10.92 + 3.9 % 20.6 ± 11.7% * indicates a significant difference between the initial and final minutes of the test (p<0.05). Between Trial Differences. In order to determine whether the differences seen in the 80% max P.O. trial were different from the changes seen in the 30% max P.O. trial repeated measures A N O V A ' s were used. Of interest to the present discussion was the interaction effect as determined by the A N O V A . Significant interactions were seen for the following variables: maximum Fe, angular impulse, positive angular impulse, maximum knee flexor moment, maximum propulsive extensor moment, and finally the t iming of the maximum ankle dorsiflexion moment wi th in the pedal cycle. A l l differences were significant at p<0.05. -56-Chapter 5 D I S C U S S I O N Overv iew. The main aim of this thesis was to determine the effects of fatigue on the ability of an athlete to maintain a given level of performance. It was hypothesized that as an athlete approaches the cessation of exercise, because of fatigue there wou ld be a number of compensations that would be made in order to maintain a given cadence at a given power output. The main criterion measurements were timings and magnitudes of forces, joint angles and joint moments of the initial minute compared to the final minute of an 80% max P.O. exercise test. In order to determine whether the changes found in the 80% of maximum power output trial were indeed a result of fatigue a comparison test was used. This comparison test was of the same duration as the high power output trial, however the power output that the riders rode at was considerably lower - 30% max P.O. It was found that there were changes in some of the criterion measures in both the 80% and the 30% max P.O. trials. Some of these changes support the hypotheses. A discussion of the results follows. Comparisons Earlier in this thesis there was a presentation of joint moments as shown by other investigators. A s each of these investigators used different techniques to calculate joint moments, and al l found similar patterns in the joint moments, it was felt that a comparison between the results of the present investigation and the previous studies was necessary. Figure 17 presents joint moments as calculated by Gregor et al. (1985), H u l l and Jorge (1985) and Ericson (1986). In addition data from the present investigation have been overlaid in order for the reader to get an impression of the similarity in the present data to those already published. Interestingly the ankle moments from all three studies are remarkably similar. A s the calculations ascend the limb segment there are large variations in the -57-magnitudes of the maximas and minimas although the patterns remain similar. These changes in magnitudes are a result of differences in the power outputs used in each of the studies. Hip Moment Gregor 0 45 90 135 180 225 270 315 360 Crank Angle (deg) Knee Moments 1 - 1 -601 0 45 90 135 180 225 270 315 360 Crank Angle (deg) Ankle Moments 0 45 90 135 180 225 270 315 360 Crank Angle (deg) Figure 17. Comparisons wi th other joint moment data. - 5 8 -Support for Hypotheses Hypothesis 1. The first hypothesis stated that as the riders approached fatigue during a steady rate constant power output test the timings of the maximum joint moments wou ld occur earlier in the pedal cycle. The data presented in the results section of this paper indicate that, for the most part, this in fact was not the case. Only two relative t iming changes occurred in the joint moments. The first was a later maximum dorsiflexor moment of the ankle joint in the final minute of the exercise test. This later maximum moment does not support the first hypothesis. The second timing change which is of importance to the support of the earlier t iming of the maximum joint moments is the t iming change that occurs at the hip joint. The maximum hip extensor moment occurs earlier in the pedal cycle in the final minute when compared to the initial minute. This hip moment t iming change is of greater importance than the t iming change at the ankle because it occurs within the propulsive phase of the pedal cycle, whereas the maximum dorsiflexor moment at the ankle occurs in the recovery phase of the pedal cycle and therefore has little consequence for the motion of the bicycle. A n earlier maximum hip extensor moment indicates that as the athlete tired it was necessary to increase the output of the large muscle groups that produce extension at the hip and to shift the timing earlier in order to overcome maintain the test cadence. Al though not specifically defined in hypothesis 1, there were changes in the maximum joint moments i n the final minute of the exercise test. These changes included increases in the maximum plantarflexor moment at the ankle, maximum knee extensor moment and the maximum hip extensor moment. In each case these maximum moments occur in the propulsive phase of the pedal cycle and are therefore of interest in the present discussion. Because of these increases in the maximum joint moments at each of the lower limb joints there was a concomitant - 5 9 -increase in the propulsive moment in the final minute when compared to the ini t ia l minute. Hypothesis 2. There was only one significant change that supported hypothesis number 1. This change was the earlier t iming of the maximum hip extensor moment. Hypothesis 2 stated that the changes described in hypothesis 1, earlier maximum moments, were a direct result of the athlete changing his pattern of application of force to the bicycle. These changes would be reflected in the components of the joint moments, such as joint angles and pedal forces. The next section in this paper w i l l look at the relationship of these variables to the support of hypothesis 2. Index of effectiveness. It was hypothesized that as the athlete fatigued the force applied to the pedal would become more effective in order to maintain the test cadence. Black et al. (1993) found that during an incremental exercise test to exhaustion athletes became more effective. Amoroso et al. (1993) however, found that the IEFF d id not change during a steady rate exercise test to exhaustion, although the effective component of the resultant force was significantly increased in the final minute of the exercise test. The findings of the present study support the Amoroso et al. (1993) study that found no differences in the IEFF. These findings indicate that the results of the Black et al. (1992) study were an outcome of increased power output, not of fatigue, a conclusion which also supports the findings of Kautz et al. (1991), but do not support those of Patterson and Moreno (1990). Patterson and Moreno found that IEFF decreased wi th increased power output but lower power outputs and untrained cyclists were used as subjects in their study. Therefore as athletes become fatigued during a steady rate exercise test to exhaustion they do not become more effective in the application of force to the pedals. Angular Impulse. It was hypothesized that as the athlete fatigued the IEFF wou ld increase. These increases in the IEFF would be a result of increased positive -60-angular impulse applied to the pedal. Although the index of effectiveness d id not increase there were changes in the angular impulse applied to the pedal. There were no changes in the total impulse applied to the pedal, however there were changes i n the components of the total impulse, positive and negative impulse. The positive angular impulse applied to the pedal increased, however the negative impulse applied to the pedal became more negative, which negated any gains made in the initial propulsion phase of the pedal stroke. Furthermore, in order for the total angular impulse to remain constant as the negative impulse increased during the recovery portion of the pedal stroke there had to be the concomitant increase in the positive angular impulse to compensate. Forces. It was hypothesized that as the rider became fatigued there wou ld be an increase i n the positive angular impulse and an increase in the negative impulse. These changes would cause the applications of force to the pedal to change. A s the negative impulse increased there was an increase in the net positive angular impulse as wel l . It was hypothesized that this increase in the net positive angular impulse wou ld be driven by an earlier application of force to the pedal wi th a subsequent earlier maximum of forces applied to the pedal. However, this earlier maximum of pedal forces d id not occur. There were no changes in the timings of the maximum normal or tangential forces. This leaves us wi th the question of " i f the maximum forces d id not occur earlier what made up for the increase in the net positive angular impulse?" The increase in the net positive angular impulse was a result of an increase in the in the maximum forces applied to the pedal. These data in part support the finding by Amoroso et al. (1993), who found that as athletes fatigued there was an increase in the maximum normal and tangential forces. However, Amoroso et al. (1993) also found earlier maximum anterior and posterior tangential forces. These last-mentioned findings were not supported i n the present study. i -61-The tangential force curve for the final minute of the exercise is different in shape than it was in the first minute of the exercise. Al though this was not tested, nor was it hypothesized, there was a change in the pattern of force application in the final minute. This change was a shift to earlier negative tangential force, or a pul l ing across the pedal platform. It is believed that this force pattern change could be l inked to changes in the joint angle at the ankle. These joint angle, or kinematic changes w i l l be covered in the next section. Joint Angles. A s the athlete fatigued it was shown in the results section that there were some changes in the joint kinematics. These changes included greater thigh extension, increased knee flexion as wel l as some timing changes of these maxima wi th in the pedal cycle: later maximum, knee extension and foot dorsiflexion. Not only were there statistical changes, there were also changes in the pattern of the joint movement. These pattern changes were not analysed statistically, however. Figure l a shows that in the final minute of the exercise, as the athlete approaches fatigue, there is a longer portion of the pedal cycle where the athlete remains in dorsiflexion. This increased time in dorsiflexion results in changes at both the knee and the hip. The timing of the maximum knee extension occurred later and there was greater hip extension. Pattern changes in the foot angle have been documented by a number of authors. Black et al. (1993) found that as the power output increased that there was a change in the ankle angle towards dorsiflexion throughout the pedal stroke. Similar findings were presented by Kautz et al. (1991). These authors found that athletes could be grouped into two separate categories based on the changes in pedal angle as power output was increased. One of the two groups showed no change in the pedal angle as the power output was increased and the other group, called the ankling group, showed a more dorsiflexed pedal style throughout one pedal cycle. The present study d id not try to differentiate between pedal techniques, which may be a weakness. -62-These kinematic changes are linked to the change in the pattern of force application utilised by the athlete to maintain the given level of performance. A s the ankle dorsiflexes there is a mechanical barrier to motion at the end of the ankle's range of motion. This mechanical barrier could enhance the plantarflexor muscle groups of the lower l imb by changing the role of the plantarflexors from a support of the ankle joint and ankle plantarflexion to simply ankle plantarflexion. This is a relatively plausible explanation because at the same point in the pedal cycle where maximum dorsiflexion occurs, the maximum ankle plantarflexor moment occurs. The pattern of the ankle plantarflexor moment is also changed in the final minute of the exercise. The change is to a larger area through which the plantarflexor moment is near its maximum. This longer maximum plantarflexor moment also adds support to the idea that there is a mechanical barrier at maximum dorsiflexion that aids i n the production of a maximum plantarflexor moment. These data support earlier findings of both Black et al. (1993) and Amoroso et al. (1993) and Amoroso (1994), who all felt that as the athlete fatigues the plantarflexors fatigue first, as they are a smaller muscle group. In order to maintain a high plantarflexor moment, the ankle dorsiflexes to its maximum dorsiflexed position, at which point there is a mechanical barrier to motion, thus stabilizing the ankle joint and thus allowing the transmission of force from the larger muscle further up the leg through the lower limb and into the pedal. 30% max. P.O. trials. It was hypothesized that there would be no differences between the initial and the final minute of the 30% max. P.O. test. This however was not the case. There were changes in a number of the variables that were measured. loint moments. The changes that occurred in the 30% max. P.O. trials were different from the changes that occurred in the 80% max. P.O. trial. Compensations that the riders make as a result of time are increased ankle dorsiflexor moment, increased knee flexor moment, increased hip flexor moment and later occurrences -63-of maximum hip extensor and flexor moments. Thus the only changes in the 80% max. P.O. trial that are the same in the 30% max. P.O. trial, and therefore may simply be accommodations by the rider to the length of time on the bicycle, are the maximum hip flexor moment and the t iming of the maximum hip extensor m o m e n t Forces and Joint Angles. The changes that were experienced by the riders in the 30% max. P.O. trial that were significant and the same as the changes in the 80% max. P.O. trial were negative work and negative angular impulse, both decreased. Al though there were other significant changes they were either not the same as occurred in the 80% max. P.O. trial or the changes were in the opposite direction to the changes in the 80% max. P.O. trial and w i l l therefore not be discussed here. Even though there were significant changes in the same direction in the two testing protocols and in some of the same variables, these changes are felt to have little effect on the result of the 80% max. P.O. results. The reasons for this w i l l be discussed in the following section, which looks at intra-subject variability. Variabi l i ty. The question of variability was not addressed in the hypotheses, however it seems necessary to look at the within-subject variability. One reason to look at the variability is that it gives insight into what happen to athletes as fatigue becomes a factor in performance. The measure of variability is the coefficient of variation (C.V.), which is described in detail by Winter (1984). Winter used the C V . initially to determine how consistent are patterns of motion during walking. Because cycling is a cyclical motion, as is walking, the primary comparison for the following discussion w i l l be to Winter's work. Joint Moments. Dur ing normal walking average within-subject C V . was found to be 71.6%, 67.0%, 22.1% and 25.1% for the hip, knee ankle and support moments respectively (Winter, 1984). In contrast the present investigation of cycling found average C V . ' s of 12.7%, 13.7%, 9.8% and 13.5% for the hip, knee, ankle -64-and propulsive moments respectively during the initial minute of the 80% max. P.O. trial. The C.V. 's for walking are much higher than those for the cycling activity. This is not surprising since the number of degrees of freedom in walking is much higher than the degrees of freedom in cycling. For instance the hip and foot are fixed in cycling because of the mechanical constraints introduced by the athlete sitting on the bicycle seat, and the foot being attached to the pedal which is attached to the crank and can travel only in a circular path. Thus cycling is a defined system in which the degrees of freedom are limited by mechanical attachments. Perhaps the greatest degree of freedom in cycling is at the knee joint, because it is not directly attached to the bicycle. This lack of attachment allows for larger variation in the motion or the production of force, thus the C V . for the knee moment is the highest of all the joints of the lower limb. In contrast, walking is mechanically constrained only at the floor, when the foot is in contact wi th the floor. Therefore, as the joint moments are calculated up the lower limb the C.V. 's increase as there are an increase in the degrees of freedom at each level up the lower limb. In general, fatigue also has an effect on the C.V. 's of the lower limb joint moments. A s the athlete becomes fatigued there is a trend towards increased C.V. ' s , although not significant. This investigation also found that as time progresses, and the athlete is at a low power output, the variability of the exercise decreases, this could indicate that in the initial minute of the test at the 30% max. P.O. the athletes were not sufficiently prepared, i.e. warmed-up, for the test to start. These larger C.V. ' s in the 30% max. P.O. test may be an indication, therefore, that some of the changes seen in the 30% max. P.O. test were simply an accommodation of the subject to the experimental set-up rather than to the cadence and power output setting. Forces. In walking trials Winter (1984) found that the C.V. 's for the ground reaction forces were 20% and 7% for the horizontal and vertical forces respectively. The present study found that the C.V. 's for the same forces, horizontal and vertical -65-forces, were 21.2% and 10.4% respectively for the initial minute of the 80% max. P.O. trial. From these results it can be seen that there is very little difference between the variability in the application of force between walking and cycling. There is a trend towards increased variability as the athlete becomes fatigued in the 80% max. P.O. trial, however as wi th the joint moments there are no significant differences. These consistent C.V. 's indicate that even wi th fatigue that pattern of force production remains consistent between pedal cycles. The C.V. 's in the 30% max. P.O. trials showed a significant decrease in the variability of the normal, vertical, force component. A s wi th the joint moment changes it would seem that as the athlete becomes comfortable wi th the cadence/power output setting the intra-subject variability decreases. Toint Angles. Because of the small number of degrees of freedom in the lower limb in cycling there are large differences between C.V. 's of joint angles in walking as compared to the C.V. 's of the joint angles in cycling. Winter (1984) found C.V. ' s for joint angles of 19%, 10%, and 9% for the hip, knee and ankle joints respectively. These C.V. ' s are compared to hip, knee and ankle C.V. 's in cycling of 1.8%, 3.4% and 3.4% respectively for the 80% max. P.O. trials. There were significant changes in the C.V. 's of the lower limb angles when the initial minute of the 80% max. P.O. test was compared to the final minute of the same test. This indicates that as the athletes fatigue their pattern of l imb motion becomes more variable. The variability of the joint angles however is very low, and although there are changes in the C.V. 's between fatigued and non-fatigued trials these differences are small when compared to the C.V. 's in walking. The changes in the 30% max. P.O. test were again towards less variability as the test continued. There was significantly less variability in the knee angle when the final minute of the test was compared to the initial minute of the test. These changes again indicate that as the test progressed the rider became more comfortable wi th the testing setting. Bruggermann and Arndt (1994) found, as a byproduct of -66-their study investigating the role of muscular and aerobic fatigue in the control of lower extremity function in running, that there was a period of adaptation required before the athlete produced consistent kinematic and kinetic patterns. This period of adaptation was in the order of 10 minutes. Thus the changes that were recorded in the 30% max P.O. test were a result of this adaptation process. Interactions To determine whether the changes seen in the 80% max P.O. trials was different from the changes in the 30% max P.O. trial repeated measures A N O V A ' s were used. The results of the A N O V A ' s tended not to strengthen the argument that the changes seen in the 80% max P.O. trials were different than the changes in the 30% max P.O. trials when the initial and the final trials were compared. Some of the significant changes in the interactions between the two power outputs are none-the-less interesting. For instance, as the athlete becomes fatigued in the 80% max P.O. trial the angular impulse applied to the pedals increases. However, the changes in the 30% max P.O. trial indicate that as the athlete progresses through the trial the angular impulse decreases. This change could be related to the accommodations to the exercise test as Bruggermann and Arndt (1994) hypothesized. In most of the cases of the significant interactions, as defined in the Results section, indicate that the changes in the final trial of the 30% max P.O. trial are in most a decrease i n the test variable, whereas the changes in the 80% max P.O. trial are generally an increase in the test variable. Examples of this phenomenon are as follows; angular impulse, positive angular impulse, maximum knee flexor moment and maximum propulsive extensor moment all decrease when the initial minute is compared to the final minute whereas in the 80% max trials these variables all increase between the initial minute and the final minute. These changes indicate that the compensations made in the 80% trial are different than the compensations made in the 30% max P.O. trial, thus the changes that result from whole body fatigue are different than those made in compensation to the duration of the test. -67-Big Picture Fatigue, as defined earlier in this paper, is the inability to generate a required or expected force. This definition indicates that when the athlete becomes "fatigued" he can no longer carry on the exercise. This definition of fatigue, therefore, is not good one in explaining the changes that were documented by this thesis. Thus, fatigue in the case of this paper must be thought of as a progression towards, ultimately, the inability to generate force. "Fatigue" is therefore thought of as a progression towards a cessation of activity. In this investigation it has been shown that there are differences at the end of a whole body fatiguing exercise test in terms of the applications of forces, the angles of the limb segments and the joint moments. These changes that result from whole body fatigue are different from the changes that result simply from exercising at a low power output for a similar length of time. These changes indicate that athletes change the strategy as they progress towards "fatigue." It is hypothesized that as the athlete becomes "fatigued" there is a subconscious change made by the athlete to compensate for the local muscle fatigue. These changes include altering the joint angles such that different muscle groups become responsible for the propulsion of the bicycle. Thus the cessation of the activity is delayed and the athlete is able to continue the exercise for an extended period of time. -68-Chapter 6 C O N C L U S I O N S A N D R E C O M M E N D A T I O N S Summary This study was designed to address the issue of compensation by athletes to whole body fatigue. The initial focus was to determine whether specific biomechanical parameters changed as a result of whole body fatigue. A steady rate, constant power output test at 80% of the athlete's max. P.O. was used to induce changes resulting from whole body fatigue. Comparisons of the joint moments, t iming of maximum force production and kinematic changes between the final minute of the test and the unfatigued state, the beginning of the test, gave insight into accommodations the athletes make to the onset of whole body fatigue. A final test was conducted to ensure the results of the constant power output test were truly compensations to whole body fatigue. This third test was the same as the second test, a steady rate constant power output test, except the power output was 30% of max. P.O. rather than 80%. It was hypothesized that there wou ld be no changes between the first minute of the test and the final minute of the test in the variables that were to be measured. This ensured that the changes in the second test were truly a result of whole body fatigue, and not simply an accommodation as a result of time. It was hypothesized that there would be specific changes in these biomechanical variables as an athlete approached the cessation of the exercise, or whole body fatigue. The following section w i l l summarize the results of this study by addressing each of the hypotheses individually. 1. Wi th fatigue it was hypothesized that there would be changes in the pattern of force application at the pedal during a steady rate exercise test to exhaustion such that the maximum joint moments would occur earlier in the pedal cycle when compared to similar power outputs in an unfatigued state. The only joint that supported this hypothesis was the hip joint. Both the ankle and the knee joints showed no differences in the t iming of the maximum joint moments. -69-2. The second hypothesis was that the changes described in hypothesis 1 were a direct result of the athlete changing his pattern of force application such that the index of effectiveness for the complete pedal cycle would increase. This increase in the IEFF wou ld be accomplished by an increase in the positive angular impulse, which wou ld be a result of an earlier maximum force and an increase in the negative angular impulse in the second half of the pedal cycle. The IEFF in the final minute of the 80% max. P.O. exercise test showed an opposite result to the hypothesized result. IEFF in the final minute of the test was significantly lower than in the initial minute of the test. Positive angular impulse d id indeed increase as was hypothesized and so too d id the negative angular impulse. These off-setting changes in the positive and negative angular impulses had no effect on the total angular impulse and therefore no effect on the IEFF. Finally there were no changes in the timings of the forces. However, there were significant results that were not hypothesized. These results w i l l be summarized in the following section. 1. Increased maximum ankle plantarflexor moment, increased maximum knee extensor moments and increased hip extensor moments which added up to produce a significantly larger propulsive moment. 2. There was a longer portion of the pedal cycle where that ankle was in dorsiflexion and an earlier shift from positive tangential force to negative tangential force. These changes are believed to be linked. Conclusions. Based on the results of this study, the following conclusions were made. 1. There were changes in the pattern of force application as defined by joint moments. The athlete's increased joint extensor moments produced an increased propulsive moment as they became fatigued. These increases were a result of the larger muscle groups of the upper leg taking a more active role in propulsion and the smaller plantarflexor groups of the lower leg taking a smaller role as a result of fatigue. -70-2. This increase in the propulsive moment was necessary because the effectiveness of force application at the pedal decreased. 3. The ankle plantarflexor moment increased wi th the aid of a mechanical stop at the end of the ankle's range of motion in the dorsiflexed position. This was seen by the changes in foot angle, towards dorsiflexion, and the changes in the application of tangential force. Al though not specifically addressed in the present study it is felt that the results found here could be applied to not only cyclist but other endurance type athletes as wel l . Finally, during studies in which "fatigue" could be a factor the role of changing muscle recruitment strategies must be accounted for in the experimental design. Recommendations. 1. In order to understand the individual muscles' response to fatigue E M G should be used i n conjunction wi th joint moment calculations. This addition of data wou ld give the researcher a greater understanding of the underlying changes that result from fatigue. Furthermore, the hypothesis of a mechanical stop at the ankle joint in dorsiflexion as the athlete becomes fatigued could be investigated based on E M G bursts from the Gastrocnemius and Soleus muscle groups. 2. Analysis of the each minute of the exercise test would give researchers an idea where these changes occur within the fatiguing process. Furthermore, the adaptation period hypothesized by Briiggermann and Arndt (1994) should be determined as it relates to cycling. This could be done simply by looking at each minute of the 30% max P.O. protocol both kinetically and kinematically. 3. A stricter cadence protocol may reduce variability based on cadence. The present study had an effective cadence range of 10 R P M . Cadence changes may have had an effect on the results of the present study 4. Finally, and perhaps most important, rather than a straight 80% of max P.O., which elicit a different metabolic response between subjects, a more precise -71-measurement of anaerobic threshold should be made. Using the measure of anaerobic threshold the researcher could regulate 5 or 10% above this point and thus elicit metabolically a more precise, between subject, fatigue. -72-R E F E R E N C E S Amoroso, A .T . , Sanderson, D.J. and Hennig, E.W. (1994). Consequences of fatigue on a rider's kinematics and kinetics during endurance cycling. In Proceedings of the Xth Annua l Congress of the American Society of Biomechanics. Amoroso, A .T . , Sanderson, D.J. and Hennig, E.W. (1993). Consequences of fatigue on a rider's kinematics and kinetics during endurance cycling. Unpubl i shed course project. Andrews, J.G. (1987). The functional roles of the hamstrings and quadriceps during cycling: Lombard's paradox revisited. T. Biomech. 20,565 - 575 Bates, B.T. and Osternig, L.R. (1977). Fatigue effects in running. J. Motor Behav. 9(3):203 - 207. Black, A . H . , Sanderson, D.J. and Hennig, E.W. (1992). The effectiveness of force application in cycling. Proceedings of The Second Nor th American Conference on Biomechanics. Chicago, Illinois. 245 - 246. Black, A . H . (1993). Errors incurred in the collection of kinematic data. Unpubl i shed course project. Black, A . H . (1993). Joint moment calculations using calculated kinematic data: project P008. Unpublished project. Black, A . H . , Sanderson, D.J. and Hennig, E.W. (1993). Kinematic and kinetic changes during an incremental exercise test on a bicycle ergometer. Proceedings of B I O M E C H A N I C S XlVth ISB Congress. Paris, France. 186-187. Briiggerman, G.-P., and Arndt , A . (1994). Fatigue and lower extremity function. Proceedings of The Eighth Biennial Conference and Symposium of the Canadian Society for Biomechanics. Calgary, Alberta. 316-317. Chapman, A . E . (1981). Cyclographic identification of fatigue in sprinting. J. H u m a n Mvnt . Studies 7. 255-272. Chapman, A . E . (1982). Hierarchy of changes induced by fatigue in sprinting. Cdn . T. A p p l . Sports Sci. 7(2):116-122. D a l Monte, A . , Moreno, A . , & Fucci, S. (1973). Biomechanical study of competitive cycling. S. Cerquiglini , A . Venerando, J., Wartenweiler (eds.). Biomechanics III (pp 434-439). Basel: Karger. Daly, D.J., and Cavanagh, P.R. (1976). Asymmetry in bicycle ergometer pedaling. M e d . Sci. Sports Exerc. 8(3). 204. Davis, R.R., & H u l l , M . L . (1981). Measurements of pedal loading in bicycling: II. Analysis and results. T. Biomech.. 14. 857-872. -73-Derosa, D . (1990). Joint moments, cadence and power outputs. Unpublished course project. Edwards, R.H.T. (1981). Human muscle function and fatigue. In H u m a n muscle fatigue: physiological mechanisms Porter. R.; WhelanJ. , (eds). London: Pitman Medical: 1-18. Elliot, B .C. and Roberts, A . D . (1980). A biomechanical investigation of the role of fatigue in middle-distance running. Can. I. A p p l . Spt. Sci. 5(41 203-207. Elliot, B.C. and Ackland, T. (1981). Biomechanical effects of fatigue on 10000 meter running technique. Res. Quart. Exerc. Sport 52(2). 160-166. Ericson, M . O . (1986). O n the biomechanics of cycling: a study of joint and muscle load during exercise on the bicycle ergometer. Scand. J. Rehab. Med. , Supp. 16. 1-43. Francis, P.R. (1986). Injury prevention: a biomechanical approach. In, Science of Cycl ing . Burke, E. R. (ed), Champaign, 111., Human Kinetics Pub. Francis, P.R. (1988). Pathomechanics of the lower extremity in cycling. In, Med ica l and Scientific Aspects of Cycling. Burke, E. R. (ed), Champaign, 111., H u m a n Kinetics Pub. Gregor, R.J., Cavanagh, P.R., & LaFortune, M . , (1985). Knee flexor moments during propulsion in cycling: a creative solution to Lombard's paradox. J. Biomech. 18. 307 - 316. Hannaford, D.R., Moran, G.T. and Hlavec, H.F. (1985). Video analysis and treatment of overuse knee injury in cycling: a limited clinical study. Biomechanics i n Sport II. San Diego: Academic Press. H u l l , M . L . , & Jorge, M . , (1985). A method for biomechanical analysis of bicycle pedalling. T. Biomech. 18. 631 - 644. Hoes, J.J., Binkhorst, R .A. , Smeekes-Kuyl, A . E . , & Vissers, A . C . (1968). Measurement of forces exerted on pedal and crank during work on a bicycle ergometer at different loads. Int. Zeitschrift fur A n g . Physiol.. 26. 33-42 Kautz, S.A., Feltner, M . E . , Coyle, E.F., and Baylor, A . M . (1991). The pedaling technique of elite endurance cyclists: Changes wi th increased workload at constant cadence. Int. J. Sport Biomech., 7, 29-53. Lafortune, M . A . and Cavanagh, P.R. (1983). Effectiveness and efficiency during bicycle riding. In Biomechanics VII-B (eds.)Matsui and Kobayashi. H u m a n Kinetics Pub., Champaign, 111. 928-936. Lafortune, M . A . , Lambert, C. and Lake, M . (1992). Skin marker displacement at the knee joint. Proceedings of the Second Nor th American Congress on Biomechanics. Chicago, 111. eds. L . Draganich, R. Wells and J. Bechtold. 101-103. -74-MacLaren, D . P . M . , Gibson, H . , Parry-Billings, P., & Edwards, R.H.T., (1989). A review of metabolic and physiological factors in fatigue. In Exercise and Sports Science Reviews. V o l . 17. Pandolf, K . B . (ed), Will iams & Wilkins , Baltimore. 29 - 66. Nordeen-Snyder, K . (1977) The effect of bicycle seat height variation upon oxygen consumption and lower limb kinematics Med, and Sci. in Sport, 9(2). 113-117. Patterson, R.P. and Moreno, M.J . (1990). Bicycle pedalling forces as a function of pedalling rate and power output. Med . Sci. Sports Exerc. 22(4) 512-516. Redfield, R. and H u l l , M . L . (1984). Joint moments and pedalling rates in bicycling. Proceedings of International Symposium on Biomechanics of Sport, (eds) J. Terauds, K . Barthels, E. Kreighbaum,R. Mann, and J. Crates. 247-258. Redfield, R. and H u l l , M . L . (1984). O n the relation between joint moments and pedalling rates at constant power in bicycling. I. Biomech. 19. 317-329. Sahlin, K . (1985). Metabolic changes l imiting muscle performance. In Biochemistry of exercise VI. Saltin, B. (ed), Champaign, 111., Human Kinetics Pub. Sanderson, D.J. and Cavanagh, P.R. (1985)An investigation of the effectiveness of force application in cycling. Med . Sci. Sports Exerc. 17. 222. Sanderson, D.J. (1990). Effects of cadence and power output on asymmetry of force application during steady-rate cycling. T. Human Mvmnt . Studies. 19, 1-9. Siler, W . L . and Mart in , P.E. (1991). Changes in running pattern during a treadmill run to volit ional exhaustion: fast versus slower runners. Int. J. Sports Biomech 7,12 - 28. Soden , P.D., & Adeyefa, B .A . (1979). Forces applied to a bicycle during normal cycling. T. Biomech.. 12. 527-541. Sprague, P. and Mann, R.V. (1983). Effects of muscular fatigue on the kinetics of sprint running. Res. Quart. Sports and Ex. 54(l),60-66. Wil l iams, K.R. , Snow, R. and Arguss, C. (1991). Changes in distance running kinematics wi th fatigue. Int. T. Sports Biomech. 7,138 - 162. Winter, D . A . (1980). Overall principle of lower limb support during stance phase of gait. T. Biomech. 13. 923-927. Winter, D . A . (1983). Biomechanical motor patterns in normal walking. T. Motor Behaviour. 15(4). 302-330. Winter, D . A . (1984). Kinematic and kinetic patterns in human gait: variability and compensating effects. Human Movement Sci. 3. 51-76. - 7 5 -

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0077057/manifest

Comment

Related Items