THE EFFECT OF RECOVERY STRATEGIES ON HIGH-INTENSITY EXERCISE PERFORMANCE AND LACTATE CLEARANCE by MON JEF PEETERS B.H.K., The University of British Columbia, 2004 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Human Kinetics) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) March 2008 © Mon JefPeeters, 2008 Abstract PURPOSE: To compare the effects of recovery intensity on performance of a bicycle sprint task and blood L& clearance. METHODS: On three separate days twelve trained male subjects (27.4 ± 3.9 yrs) performed three supramaximal exercise (SE) bouts at 120% of maximum aerobic power (MAP) for 60% of the time to exhaustion (TTE). Bouts were separated by 5 mm of passive recovery (PR), active recovery (AR) or combined active recovery (CAR). The third bout was followed by a 14 mm recovery. Recovery intensities were: PR (rest), AR at 50% of the workload difference between the individual anaerobic threshold (TAT) and the individual ventilatory threshold (IVT) below the IVT (IVT_50%A ), or CAR at the TAT workload for 5 mm and at the IVT_50%AT workload for 9 mm. Five 10 s sprints were performed 2 mm post-recovery. Blood lactate (L&) concentration, power parameters (Peak Power (PP), Mean Power (MP), Fatigue Index (Fl), and Total Work (TW)), Heart Rate (HR), and Oxygen Uptake (V02) were compared using repeated-measures ANOVA. Pairwise comparisons and dependent T-tests were performed to analyze differences. RESULTS: Mean La values for AR and CAR were lower than PR (9.7 ± 3.5, 9.5 + 3.5, 11.7 + 3.6, respectively, pO.O5). La was significantly lower with CAR versus PR at the 3Id, 6th 9th and 14th mm of recovery (pO.OS). AR versus PR La was lower at the 6, 9th and 14th mm of recovery (pO.O5). Mean MP was greater in the AR group compared to the PR group (800.1 ± 114.5 vs 782.2 ± 111.7 W, pO.O5). TW during AR was greater than PR (p0.05) but not CAR (p>O.O5,40003.3 ± 5110.2, 39108.3 ± 4852.9, 39335.8 ± 5022.6 J, respectively). CONCLUSIONS: AR and CAR both demonstrated improved L& clearance when compared to PR, but differences in L& clearance did not determine performance on the sprint task. AR resulted in more TW than PR and greater maintenance of power over the sprints. 11 Table of Contents Abstract.ii Table of Contents iii List of Tables v List of Figures vi CHAPTER I: INTRODUCTION 1 1.1 Introduction 1 1.2 Rationale 2 1.3 Purpose 2 1.4 Significance of the Study 3 1.5 Hypotheses 3 1.6 Definitions 4 1.7 Delimitations 7 1.8 Limitations 8 CHAPTER II: LITERATURE REVIEW 9 2.1 Introduction 9 2.2 The New View on Lactate 10 2.3 Lactate and its Relation to Fatigue During High-Intensity Exercise 10 2.3.1 Inorganic Phosphate as a Cause of Fatigue 11 2.3.2 Other Mechanisms of Fatigue 12 2.4 Active Recovery and Lactate Clearance 13 2.4.1 Exercise Intensity, Mode, and Duration 15 2.4.2 Combined Active Recovery and Lactate 19 2.5 Lactate and Performance 21 2.6 Active Recovery and Performance 22 2.6.1 No Performance Benefit with Active Recovery 23 2.6.2 Performance Benefit with Active Recovery 25 2.6.3 Performance and Active Recovery Duration 28 2.7 Summary 28 CHAPTER III: METHODS 30 3.1 Subjects 30 3.2 Experimental Design 30 3.3 Facilities and Instrumentation 31 3.3.1 Wingate and Sprint Tests 31 3.3.2 Ventilatory and Gas Exchange Variables 32 3.3.3 Blood Lactate 32 3.3.4 Heart Rate 33 3.4 Testing Procedures 33 3.4.1 Day 1: Anthropometric Measures, 7O2max, and WAnT 33 3.4.2 Day 2: Time to Exhaustion Test 34 3.4.3 Day 3: Familiarization Tasks 35 3.5 Data Analysis 38 CHAPTER IV: RESULTS 39 4.1 Subject Characteristics 39 111 4.2 BloodLactate.39 4.3 Sprint Task Performance 44 4.3.1 Peak Power 45 4.3.2 Mean Power 47 4.3.3 Fatigue Index 50 4.3.4 Total Work 52 4.4 Heart Rate 53 4.4.1 Exercise and Recovery Heart Rate 54 4.4.2 Sprint Task Heart Rate 56 4.5 Volume of Oxygen Consumed 58 CHAPTER V: DISCUSSION 61 5.1 Subject Characteristics 61 5.2 Blood Lactate 61 5.2.1 Combined Active Recovery and Lactate Clearance 63 5.3 Sprint Task Performance 64 5.3.1 Peak Power 64 5.3.2 Mean Power 66 5.3.3 Fatigue Index 71 5.3.4 Total Work 73 5.4 Heart Rate and Volume of Oxygen Consumed 74 5.4.1 Exercise and Recovery Heart Rate 74 5.4.2 Sprint Task Heart Rate 75 5.4.3 Volume of Oxygen Consumed 76 5.5 Practical Significance 77 CHAPTER VI: CONCLUSION 79 6.1 Conclusions 79 6.2 Recommendations for Future Research 79 6.2.1 Performance Criteria 80 6.2.2 Active versus Passive Recovery and Duration Dependency 80 6.2.3 Mechanisms 80 6.3.3 Suggested Modifications and Additions 80 References 82 APPENDIX I 93 APPENDIX II 94 APPENDIX III 95 APPENDIX IV 96 APPENDIX V 97 APPENDIX VI 98 APPENDIX VII 99 APPENDIX VIII 100 APPENDIX IX 101 APPENDIXX 10’L iv List of Tables Table 1. Subject Characteristics 40 Table 2. Subject Workload and Recovery Characteristics 41 Table 3. Mean Blood Lactate Values 42 Table 4. Mean Blood Lactate Differences for Sample Time 42 Table 5. Group Means for Peak Power, Mean Power, Fatigue Index and Total Work 45 Table 6. Mean Peak Power Outputs 45 Table 7. Sprint Number Means for Peak Power, Mean Power, and Fatigue Index 46 Table 8. Mean Mean Power Outputs 48 Table 9. Mean Fatigue Indexes 50 Table 10. Mean Exercise and Recovery Heart Rate Values 54 Table 11. Mean Exercise and Recovery Heart Rate Differences for Sample Time 56 Table 12. Mean Maximum Sprint Task Heart Rate 56 Table 13. Mean Sprint Task Heart Rate Differences for Sample Time 57 Table 14. Mean Volume of Oxygen Consumed Values 59 Table 15. Mean Volume of Oxygen Consumed Differences for Sample Time 60 Table 16. Daily and Overall Maximum and Minimum Blood Lactate Values 93 Table 17. Daily and Overall Maximum and Minimum Peak Powers 94 Table 18. Daily and Overall Maximum and Minimum Mean Powers 95 Table 19. Daily and Overall Maximum and Minimum Fatigue Indexes 96 Table 20. Individual Total Work Outputs and Maxima and Minima 97 Table 21. Daily and Overall Maximum and Minimum Exercise and Recovery Heart Rate Values 98 Table 22. Daily and Overall Maximum and Minimum Sprint Task Heart Rate Values 99 Table 23. Daily and Overall Maximum and Minimum Volume of Oxygen Consumed Values 100 Table 24. Sprint Task Fatigue Scores 101 v List of Figures Figure 1. Example of Determination of Individual Anaerobic Threshold in Subject 11 6 Figure 2. Example of Determination of Individual Ventilatory Threshold in Subject 2 7 Figure 3. Schematic Representation of the Testing Day Protocol 37 Figure 4. Mean Blood Lactate Values 43 Figure 5. Mean Peak Power Outputs 46 Figure 6. Mean Mean Power Outputs 49 Figure 7. Mean Fatigue Indexes 51 Figure 8. Total Work 53 Figure 9. Mean Exercise and Recovery Heart Rate Values 55 Figure 10. Mean Sprint Task Heart Rate 58 Figure 11. Mean Volume of Oxygen Consumed Values 60 vi CHAPTER I: INTRODUCTION 1.1 Introduction Historically lactatea (L&) and its accumulation during high-intensity exercise has been identified as a causative factor in fatigue development (Hill & Kupalov, 1929; Karisson, Bonde Petersen, Henriksson, & Knuttgen, 1975; Klausen, Knuttgen, & Forster, 1972; Stamford, Weitman, Moffatt, & Sady, 1981; Yates, Gladden, & Cresanta, 1983). More recently, the role of L& as a causative agent in fatigue development has been questioned, but results remain inconclusive (Fitts, 2003; Gladden, 2004). It appears that the relationship of La accumulation to fatigue is part of a complex interactive process. Nonetheless, the association between L& accumulation and the onset of fatigue during high-intensity exercise remains irrefutable. Though, it must be stressed that the chronological associations of the two events does not necessarily indicate causation. It has been suggested that an increased rate of L& clearance may lead to faster recovery and/or postpone fatigue during repetitive exercise (Gisolfi, Robinson, & Turrell, 1966; Lindinger, McKelvie, & Heigenhauser, 1995). This has large implications for performance athletics, where the ability to recover quickly for subsequent high-intensity exercise bouts is crucial to success. This is especially true for sports where athletes may compete more than once in a day (e.g. track, cycling, swimming, ref. Dodd, Powers, Callender, & Brooks, 1984) or perform repetitive high-intensity bouts within one competition (e.g. hockey, football, soccer, basketball). The unequivocal findings that active recovery (AR, i.e. low-intensity aerobic exercise performed after high-intensity exercise) increases the rate of L& clearance (Ahmaidi et al., 1996; Belcastro & Bonen, 1975; Davies, Knibbs, & Musgrove, 1970; Dupont, Moalla, Guinhouya, Abmaidi, & Berthoin, 2004; Gupta, Goswami, Sadhukhan, & Mathur, 1996; Hermansen & Stensvold, 1972; Jervell, 1928; Newman, Dill, Edwards, & Webster, 1937; Rämmal & Strom, 1949; Siebers & McMurray, 1981; Stamford et al., 1981; Taoutaou et al., 1996; Weltman, Stamford, & Fulco, 1979), has lead to the belief that AR is beneficial to subsequent exercise performance. However, studies investigating the effect of active versus passive recovery (PR, i.e. no activity) on subsequent performance have been inconclusive, with some investigations showing benefits (Ahmaidi et al., 1996; Signorile, Ingalls, & Tremblay, 1993; Spierer, Goldsmith, Baran, Hryniewicz, & Katz, 2004; Thiriet et al., 1993) and others not (Dupont et al., 2004; Franchini, Yuri Takito, Yuzo Nakamura, Ayumi Matsushigue, & Peduti DaUMolin Kiss, 2003; Siebers & McMurray, 1981; Weltman & Regan, 1983). Furthermore, the a The term La will be used rather than lactic acid throughout this paper since at a physiological muscular pH range (6.2O-7.OO) lactic acid is more than 99% dissociated to La and a proton due to its pKa value (pH=3 .87, ref. Gladden, 2004; Robergs, Ghiasvand, & Parker, 2004) 1 optimal method(s) by which to perform AR remains uncertain since previous research has used a wide array of protocols. It has been suggested that recovery (and exercise) intensities should be expressed relative to individual thresholds rather than maximum oxygen consumption (VO2max, ref. Baldari, Videira, Madeira, Sergio, & Guidetti, 2004; McLellan & Skinner, 1982), since the metabolic outcome is dependent on thresholds. There is some evidence to support the use of a combined active recovery (CAR) of varying intensity to improve the clearance rate of La in comparison to AR at one intensity (Gmada et al., 2005). However, further research is needed to congeal this assertion. Additionally, there is currently no research available as to whether or not the increased rate of La clearance with a CAR translates to improved performance or what the effects of CAR are on performance independent of the effect on L& clearance. 1.2 Rationale In 1981, Stamford et al. suggested that there may not be one optimal exercise intensity at which to clear LaZ Rather than use a continuous submaximal constant load for AR (as had previously been used in most research protocols), they suggested using a progressively decreasing load starting above the anaerobic threshold (AT) and finishing below the AT. The author is aware of two studies that have used this suggestion and employed a CAR. Dodd et al. (1984) compared CAR (i.e. moderate-to-high-intensity followed by low-intensity exercise) to both passive and two traditional continuous constant load recoveries at moderate-to-high- and low-intensity, respectively. Their results showed that the CAR and low-intensity AR had the fastest L& clearance rates, but that there were no statistically significant differences between those two recovery modalities. Recently Gmada et al. (2005) showed that CAR resulted in a faster L& clearance rate when compared with constant load moderate-to-high-intensity, low- intensity active, and passive recovery (the effect was more pronounced in trained subjects compared to untrained). Together, these investigations demonstrate that CAR is at least equal to or better than AR with respect to blood La clearance rate. However, neither of these studies examined the effect of the recovery periods on subsequent performance. 1.3 Purpose The purpose of this investigation was two-fold: to examine the effects of two active recoveries, relative to individual thresholds (i.e. CAR and AR), and PR in trained male cyclists on the subsequent performance of five 10 s bicycle sprints after a work bout of three supramaximal exercise (SE) intervals. Secondly, to examine the effects of the three recoveries on blood L& clearance rate. 2 1.4 Significance of the Study The results of investigations examining the effects of different recovery methods on performance have implications for the design and implementation of training programs and competition strategies. Specifically, this investigation provides insight into the use of different recovery intensities and their effects on subsequent performance of high-intensity exercise as well as blood La clearance. The results help to clarify what the effects of various recovery strategies are on performance and whether or not the use of active recoveries provide a benefit to high-intensity athletic performance and training. 1.5 Hypotheses The following hypotheses were suggested: a. Blood Lactate Alternate Hypothesis (Hi): A CAR blood L&> A AR blood L&> A PR blood L& The change in blood La in the CAR trial will be greater than the change in the AR trial which will in turn be greater than in the change PR trial. This hypothesis has been suggested because it has been previously shown that CAR is more effective than (Gmada et al., 2005) or as effective as (Dodd et a!., 1984) AR at one sole intensity with respect to L& clearance. It has been theorized that the faster L& clearance rate with CAR is a result of the increased blood flow from the higher exercise intensity (Gmada et al., 2005). Previously it had been suggested that a CAR may be more effective at removing L& since the decreasing exercise intensity would be commensurate with the decreasing L& concentration (Stamford et al., 1981). The pairing of exercise intensity and L& concentration is predicted to be beneficial because the rate La clearance is proportional to its concentration (Jorfeldt, 1970). Furthermore, it is hypothesized that AR will clear L& faster than PR, since there is an abundance of literature that already supports this hypothesis (Baldari, Videira, Madeira, Sergio, & Guidetti, 2005; Franchini et a!., 2003; McAinch et aL, 2004; Spierer et al., 2004). b. Peak Power (PP) Output H1: CAR = AR = PRpp No differences in PP output on the sprint task are hypothesized between recovery trials. This hypothesis has been suggested because it has previously been shown that AR modalities did not affect the development of PP in subsequent high-intensity exercise tasks with similar recovery durations (Ainsworth et a!., 1993; Spierer et a!., 2004; Weltman, Stamford, Moffatt, & Katch, 1977). Peak power (sometimes called anaerobic power) is mainly dependent on the phosphocreatine (PCr) system and free ATP, and 3 therefore is not expected to change significantly with the recovery durations in this investigation, since the prolonged 14 mm recovery should provide ample time for replenishment of PCr levels. . Mean Power (MP) Output H1: CARMP > ARMP> PR The MP output on the sprint task will be statistically significantly greater in the CAR trial than the MP in the AR trial, which will be greater than the MP output in the PR trial. It is hypothesized that AR will result in a greater MP output than PR recovery since it has been previously shown that AR can improve subsequent performance with respect to MP output (Spierer et a!., 2004; Thiriet et aL, 1993). Combined active recovery is hypothesized to result in the greatest MP output since it has been shown that CAR can clear L& faster than the other two recovery modes being tested (Gmada et al., 2005). High L& and proton transport, as would occur during AR, has been suggested to prevent fatigue because of the beneficial effects of L& anions in providing oxidizable substrate and gluconeogenic precursors (Thomas et al., 2005). The greater clearance of La is therefore not predicted to prevent fatigue by eliminating L&, an agent in fatigue, but rather by utilizing L& as fuel source to prevent fatigue. In the case of this investigation, it is being suggested that the fastest rates of L& transport (i.e. clearance) will occur in the CAR trial and thus in this trial subjects will have the greatest resistance to fatigue. That is, their anaerobic capacity (or MP output) will be maintained. d. Fatigue Index (Fl) H1: PRFI> ARFI> CARFI Fatigue index will be greater in the PR trial than Fl in the AR trial, which will be greater than FT in the CAR trial. Fatigue index is the percent power decrease over the course of a sprint and therefore will have the opposite results of MP. e. Total Work (TW) H1: CARTW> ARTW> PRTW Total work on the sprint task will be statistically significantly greater in the CAR trial than TW in the AR trial, which will be greater than TW in the PR trial. This hypothesis is suggested because TW is the product of MP and time over the five sprints, and thus follows the above hypothesis of MP output. 1.6 Definitions • Maximum Oxygen Consumption (VO2m) — a measure of cardiorespiratory fitness. VO2max is highest amount of 02 the body is able to consume and the product of maximal cardiac output (L.min’) and arterial-venous difference (mL O2L’, ref. ACSM, 2006). 4 • Active Recovery (AR) — moderate- to low-intensity exercise performed after high-intensity exercise to promote a faster return to pre-exercise conditions. • Combined Active Recovery (CAR) — moderate- to high-intensity exercise followed by moderate- to low-intensity exercise performed after high-intensity exercise to promote a faster return to pre-exercise conditions. • Passive Recovery (PR) — resting post high-intensity exercise to promote a faster return to pre-exercise conditions. • Individual Anaerobic Threshold (IAT) — a specific form of anaerobic threshold defined as the highest sustainable workload without an accumulation of La (i.e. maximal L& steady state, ref. Baldari & Guidetti, 2000; Stegmann, Kindermann, & Schnabel, 1981). IAT is operationally defined as the workload corresponding to the second L& increase of at least 0.5 mmolL’ from the previous value (Baldari & Guidetti, 2000, see Figure 1 for visual representation). • Individual Ventilatory Threshold (IVT) — a specific form of ventilatory threshold also known as the point of optimum ventilatory efficiency (Hollmann, 2001). IVT is the lowest value of the ventilatory equivalent (VEIVO2), when VE/V02 is plotted as a function of V02 (Baldari & Guidetti, 2001, see Figure 2 for visual representation). 5 Figure 1. Example of Determination of Individual Anaerobic Threshold in Subject 11 5. -______________ X4.3 4 •3.4 •3.2 E 3 E 4-, ‘I-, 0 -2 12.0 11.6 •1.7 0- I I I 90 120 150 180 210 240 270 300 Workload (W) TAT is the workload corresponding to the second L& increase of at least 0.5 mmol•L1 from the previous value and is represented by the x 6 Figure 2. Example of Determination of Individual Ventilatory Threshold in Subject 2 25 20 •16.5 hiT2 170 16.0 • 16.< 15.6 167 117.6 5 0 I I I I I 1 1.5 2 2.5 3 V02 (LJmin) WT corresponds to the lowest VE/V02 and is represented by the 1.7 Delimitations This study will be delimited by: a. A sample of university-aged (18-35) subjects from University of British Columbia students and others from the Vancouver area. b. Setting the criteria for trained subjects as participating in competitive cycling with a heavy anaerobic component, PP and MP outputs greater than or equal to 11 and 9 W•kg’ on the Wingate Anaerobic Test (WAnT), respectively, and having a VO2m greater than 55 mLkg’•min’. c. A respiratory gas-sampling rate set at 20 s intervals. d. The measurement of performance on the sprint task as a measure of anaerobic capacity. e. The methodology used to determine IVT, IAT, and VO2m. 7 1.8 Limitations This study will be limited by: a. The data collection capabilities of the SensorMedics Vmax 29 series metabolic cart and the interlaced Data Acquisition System. b. The individual’s metabolic responses to the testing protocols. c. The individual’s effort during testing procedures (e.g. the individual’s ability to perform maximally during the exercise tasks). d. The ability to determine TAT and IVT from the data collected. 8 CHAPTER II: LITERATURE REVIEW 2.1 Introduction In performance athletics the ability to maintain power output and ward off fatigue is essential to success. In activities that are intermittent in nature or between competitions that are shortly spaced apart (e.g. tournament setting) the capacity for an athlete to recover quickly is a determining factor in their performance. While the mechanism(s) of fatigue to date remain undetermined, the historical view has identified the accumulation of L& as the causative factor (Hill & Kupalov, 1929). Over time this paradigm has shifted to L& being an agent in the development of fatigue through its influence as an anion on the acid-base status of the muscle and blood (Lindinger et al., 1995; Stewart, 1981). The knowledge that the use of AR results in an increased clearance rate of blood L& and the anecdotal reports of the advantageous use of AR in athletic training regimens has led to a belief that AR is beneficial to repeated performance (Gisolfi et al., 1966). Despite the common belief that AR is beneficial to subsequent performance, the research evidence is inconclusive with some research showing a benefit (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Connolly et aL, 2003; Corder et al., 2000; Signorile et al., 1993; Spierer et al., 2004; Thiriet et al., 1993) and some showing no benefit (Bond, Adams, Teamey, Gresham, & Ruff, 1991; Franchini et al., 2003; McAinch et al., 2004; Siebers & McMurray, 1981; Watson & Hanley, 1986; Weltman et al., 1979; Weltman & Regan, 1983). Furthermore, within the research that supports the use of AR to improve performance, the relationship of L& clearance to performance is inconsistent (Bogdanis et al., 1 996b; Connolly, Brennan, & Lauzon, 2003; Spierer et al., 2004; Thiriet et al., 1993). Interestingly, this lack of correlation between L& clearance and improved performance may be explained by more recent evidence that is shifting the cause of fatigue away from La accumulation and the associated increase in acidity to an accumulation of inorganic phosphate (Westerblad, Allen, & Lannergren, 2002). Nonetheless, the mechanism(s) for the development of fatigue still remain unresolved and are most likely a result of a complex interaction of events related to the availability and accumulation of metabolites. Therefore the role of L& as an agent in the development of muscular fatigue remains valid as the onset of fatigue coincides with its accumulation. From a practical perspective, regardless of the role of L& in fatigue, the use of AR as a means to prevent decrements in performance in successive exercise remains a compelling research area particularly since AR may be beneficial for certain types of physical activity. 9 2.2 The New View on Lactate Historically, the accumulation of L& in the blood associated with high-intensity exercise has been viewed negatively and La has been labeled as a metabolic waste product that resulted from glycolysis in hypoxic conditions (Gladden, 2004). The initial concept of a relationship between L& accumulation and hypoxia stemmed from research in the early part of the last century (Fletcher & Hopkins, 1907; Hill, Long, & Lupton, 1924). Gladden (2004) has identified the time frame from the 1930s to the 1970s as the dead-end waste product era of L&. During this time La was believed to be a dead-end metabolite of glycolysis in hypoxic conditions (Wasserman, 1984). This paradigm has shifted greatly over the past few decades, as it has been repeatedly shown that the production of L& is not necessarily a result of 02 lack (e.g. Richardson, Noyszewski, Leigh, & Wagner, 1998). Currently, the perspective on L& metabolism is very different in light of the advent of the La shuttle hypothesis (Brooks, 1985) and the widely held acceptance of the extracellular L& shuttleb. Lactate is now recognized as a metabolic intermediate rather than an end, and a movable source of energy substrate that can be passed about the body for metabolism. 2.3 Lactate and its Relation to Fatigue During High-Intensity Exercise During high-intensity exercise, anaerobic metabolic processes are heavily utilized to meet the energy demand. As the glycolytic production of ATP increases, the mitochondria’ s ability to aerobically oxidize pyruvate is exceeded (Spriet, Howlett, & Heigenhauser, 2000). This leads to an increased concentration of pyruvate and NADH, which are then converted to L& and NAD by the near-equilibrium enzymatic reaction of L& dehydrogenase (LDH, ref. Spriet et al., 2000). It is the processes that govern the production of pyruvate and NADH that predominantly control the production of L& (Spriet et al., 2000). Blood La concentration is ultimately the result of the balance between production and clearance processes. With sufficiently high-intensities of exercise the balance between the production and clearance of L& is shifted to disequilibrium and La begins to accumulate. The knowledge that L& accumulates in exercised muscle has a long history dating back to the work of Berzelius in 1807, who noted that hunted stags had elevated acid concentrations in their muscles (as cited in Needham, 1971). More recently other researchers noted the b Brooks has also argued for an intracellular L& shuttle (Brooks, 1998) in addition to the original cell-to-cell (or extracellular) L& shuttle, and has provided support for its existence (Brooks, Dubouchaud, Brown, Sicurello, & Butz, 1999). However, other researchers have failed to fmd supporting evidence to some of the central tenets of the hypothesis (Rasmussen, van Hall, & Rasmussen, 2002; Sahlin, Fernstrom, Svensson, & Tonkonogi, 2002; Yoshida et al., 2007). Consequently, the cell-to-cell L& shuttle hypothesis is more or less unanimously accepted while the state of the intracellular L& shuttle remains to be determined with future research 10 accumulation of L& in working muscles and noted the effects of AR on its clearance (Jervell, 1928; Newman et a!., 1937). In the early twentieth century, Hill and Kupalov (1929) proposed that lactic acid accumulation was the cause of muscular fatigue. This formed the initial framework for the conception that La accumulation was the cause of fatigue. This conception was maintained for years to come (Karisson & Saltin, 1970), and is still very prevalent amongst most laypersons. Currently, more researchers attribute the associated development of fatigue that arises with L& accumulation to the associated decrease in pH (i.e. increase hydrogen ion concentration ([Hj)) rather than to the increase in L& anion itself (Fitts, 2003). Remember that the formation of lactic acid at a physiological pH results in its immediate dissociation into the L& anion and a proton. Some researchers have misinterpreted the direct donation of this proton from lactic acid as the cause of the decrease in pH. This may be a result of attempting to simplify the explanation. However, the relationship of body fluid acid-base status is more complex than this. The re-introduction of the concepts of earlier researchers such as Henderson and van Slyke by Peter Stewart (Stewart, 1981) has helped to clarify this understanding (Lindinger, 2003). In this view acid-base status is determined by the independent effects of carbon dioxide (Pcü2), the concentration of weak acid buffers ([Atot], in plasma mainly the amino acids in plasma proteins), and the strong ion difference ([SID], i.e. the sum of the strong cations minus the sum of the strong anions, ref. Kowalchuk, Heigenhauser, Lindinger, Sutton, & Jones, 1988; Stewart, 1981). Lactate being a strong anion decreases pH by causing a decrease in the [SID]. The accumulation of L& in the blood that occurs with high-intensity exercise has been shown to increase the plasma [Hj, PCQ2, and osmolarity (Kowaichuk et al., 1988; Lindinger et al., 1995). The increased [Hj is believed to cause fatigue by: (1) inhibiting the glycolytic enzyme phosphofructokinase, (2) reduction of myosin crossbridge activation via competitive inhibition of Ca2 binding to troponin C, and (3) inhibiting sarcoplasmic reticulum (SR) ATPase reducing Ca2 re-uptake and subsequently Ca2release (Fitts, 2003). 2.3.1 Inorganic Phosphate as a Cause of Fatigue Recently the role of increased [Hj in the development of fatigue has come into question as the initial studies that attributed H accumulation to fatigue were not performed at physiological temperatures, and recent investigations at physiological temperatures do not support the role of H in fatigue development (Gladden, 2004; Westerbiad et a!., 2002). Two landmark studies that have been integral in this shift are that of Bangsbo et a!. (1996) and Nielsen et al. (2001). Bangsbo et al. (1996) showed that muscle acidity in humans during intense 11 exercise did not reduce glycogenolysis/glycolysis. While Nielsen et a!. (2001) demonstrated that the reduced muscular force that normally developed with increased extracellular potassium concentration ([Kje) could be reduced with induced acidosis and was accompanied by the regeneration of action potential development. In place, the cause of fatigue is attributed to the accumulation of inorganic phosphate (Pt) within the muscle. It is hypothesized that P1 causes muscular fatigue by: (1) reducing maximum crossbridge force, or (2) altering SR Ca2handling via direct action on SR Ca2 release channels, inhibition of Ca2uptake, or formation of Ca2-P1 precipitate (for review see Westerbiad et al., 2002). However, Gladden (2004) notes that the time course of fatigue development and the accumulation of P1 within the muscle do not coincide, since the majority of PCr is broken down in the initial seconds of high-intensity exercise. Similarly, Fills (2003) states that it may be premature to dismiss the role of H accumulation in fatigue development since studies have not completely elucidated the effects of a combined low pH, elevated P1 and reduced Ca2 release. Recent research by Fitts and colleagues has provided support for the latter statement since they found that a combined reduction in myoplasmic Ca2 and increased concentration of P1 act synergistically to reduce muscular force (Debold, Romatowski, & Fills, 2005). It yet remains to determine the role of changes in acidity. 2.3.2 Other Mechanisms of Fatigue It should be noted that there are previous reports that have linked the L& anion itself to fatigue. In 1995, Hogan et al., using dog preparations showed that L& infusion at a constant arterial pH (7.40) reduced muscle tension development by 15%. However, more recently the effects of the L& anion on muscle contractility have been shown to be minimal (Posterino, Dutka, & Lamb, 2001). Still, other researchers subscribe to the theory of muscular fatigue induced via the accumulation of extracellular IC (Lindinger et a!., 1995; Renaud, 2002), where L& may have an indirect effect on fatigue development. Lindinger (1995) states that in order for proper muscle function to continue L& must be removed, and intracellular IC concentration ([K]) must be maintained, as L& clearance is essential for the recovery of [Hj and restoration of [Kj is necessary for both the recovery of [Hj and sarcolemmal and transverse tubule membrane potentials. In this model La influences acid-base status, which in turn regulates the membrane excitability. Similarly, Renaud (2002) proposes a model of fatigue in which [Kie 5 increased by the activation of ATP-sensitive IC channels (KATP channel) to prevent a decrease in cellular ATP levels or prevent accumulation of metabolic end-products. Lactate, as well as ADP and H have been identified as potential modulators of the KATP channel and thus modulators of fatigue in this model (Renaud, 2002). 12 To date, the exact mechanisms of muscular fatigue during high-intensity exercise remain undetermined and are most likely the result of a multitude of interacting factors. Fitts (2003) states that fatigue recovery from high-intensity exercise has both a rapid and slow component likely caused by separated mechanisms. The rapid component being reversible is related to P1 and changes in the excitation-contraction coupling and Ca2 regulation, while the slower component involves several sites and steps in the contraction process that are at least partially mediated by H and P1 (Fitts, 2003). While the role of L& in the development of fatigue has shifted from causative factor to a potential mediator, the fact that fatigued muscles display increased La concentrations attests to the fact that L& accumulation has some role in the development of fatigue. 2.4 Active Recovery and Lactate Clearance Numerous studies have provided unequivocal evidence that AR expedites blood L& clearance compared to passive recovery (Belcastro & Bonen, 1975; Davies et al., 1970; Hermansen & Stensvold, 1972; Jervell, 1928; Newman et a!., 1937; Rämmal & Strom, 1949; Siebers & McMurray, 1981; Stamford et a!., 1981; Weltman et a!., 1979). Beginning with Jervell (1928) it was noted that blood L& concentration declined more rapidly during exercising recovery. This was the first scientific paper on what would later become known as AR. Later, Margaria et a!. (1933) discovered that L& clearance rate is proportional to its concentration. Newman et a!. (1937) concluded that L& clearance rate increases in approximate proportion to metabolic rate up to a critical intensity, which varies among individuals and is higher in those that are trained. Subsequent studies confirmed these initial findings (Belcastro & Bonen, 1975; Davies et a!., 1970; Hermansen & Stensvo!d, 1972; Rämmal & StrOm, 1949; Siebers & McMurray, 1981; Stamford et al., 1981; Weitman et al., 1979), and the literature is conclusive that AR increases the rate of L& clearance from the blood. At the same time AR has been shown to have the reverse effect on muscle La concentration in the exercised muscle groups (McAinch et al., 2004; Peters-Futre, Noakes, Raine, & Terblanche, 1987). These researchers found that AR increased muscle L& concentration, and attributed the increase to increased local metabolic activity that resulted in L& production. However, these findings are not unanimous as other investigators have found AR to decrease muscle L& concentration (Bangsbo, Graham, Johansen, & Saltin, 1994; Spencer, Bishop, Dawson, Goodman, & Duffield, 2006). While it is agreed that AR clears blood L&, the fate of the cleared L& has been less obvious. There are two main biochemical processes that have been identified as the route for L& elimination: (1) gluconeogenesis and glyconeogenesis, and (2) oxidation in the tricarboxylic 13 cycle to CO2 and H20 with energy production (Rontoyaimis, 1988). Gladden (2003) takes the division further and differentiates between glyconeogenesis in muscle and uptake by the liver and/or kidneys with subsequent formation of glucose and/or liver glycogen. Previously it was believed that L& was predominantly removed via glyconeogenesis in the liver (Rowell et al., 1966). However, evidence shows that post-exercise the majority of L& (55-75%, ref. Brooks & Gaesser, 1980) is not resynthesized to glycogen, but rather is oxidized in the muscles (Bangsbo et al., 1994; Hermansen & Stensvold, 1972; Peters-Futre et al., 1987; Rontoyannis, 1988). Human studies estimating the post-exercise conversion of L& to glycogen have shown varied results: approximately 70% (Hermansen & Vaage, 1977), approximately 50% (Astrand, Huitman, Juhlin-Dannfelt, & Reynolds, 1986), and between 13-27% (Bangsbo, Gollnick, Graham, & Saltin, 1991). More recently it has been shown that in humans post-exercise L& makes minor contributions to glycogen synthesis (Bangsbo, Madsen, Kiens, & Richter, 1997). Furthermore, it has been shown that L& is specifically oxidized in the oxidative muscle fibres (type I), whereas it is predominantly produced in the glycolytic fibres (i.e. type II, ref. Donovan & Pagliassotti, 2000; Gladden, 2000). Taken in consideration with the ‘lactate shuttle’ hypothesis introduced in 1984 by Brooks which states: “the shuttling of lactate through the interstitium and vasculature provides a significant carbon source for oxidation and gluconeogenesis during rest and exercise” (Brooks, 1985) it is apparent that the L& produced by glycolytic muscle fibres is subsequently oxidized by oxidative fibres. When AR is performed after high-intensity exercise the accumulated L& is cleared via oxidation in type I fibres that are active during low-intensity exercise. Therefore, the major clearance pathway of L& post- exercise, especially with AR, appears to be oxidation (Gladden, 2003). It is now known that L& traverses the plasma membrane via stereo-specific, pH-dependant transmembrane proteins called monocarboxylate transporters (MCT5, ref. Bonen, 2001). In human skeletal muscle the transporter is present in two isoforms (MCT1 and MCT4, ref. Bonen, 2001), with MCT1 being important to L& clearance (Thomas et al., 2005). It is proposed that AR maintains blood flow post-exercise thereby allowing the transport and circulation of L& to sites where it can subsequently be oxidized, primarily by skeletal muscle and additionally in smaller quantities by other tissues (i.e. heart, liver, kidneys). In addition to blood flow and membrane transport, L& release is dependent on exercise intensity and duration, training status, and age (Graham, Sinclair, & Chapler, 1976; Juel, 1997). 14 2.4.1 Exercise Intensity, Mode, and Duration In order for AR to successfully clear L&, the intensity must be such that the production of L& does not exceed its clearance rate. If L& production surpasses L& clearance (i.e. utilization) then it will accumulate in the muscle and blood. The general agreement regarding the optimal intensity with which to perform AR is that the intensity should be moderate (approx. 30-45% VO2max, ref. Boileau, Misner, Dykstra, & Spitzer, 1983; Davies et al., 1970). Nevertheless intensities ranging from approximately 16-70% of VO2max have been reported in the literature (Corder, Potteiger, Nau, Figoni, & Hershberger, 2000; Hermansen & Stensvold, 1972). Using a bicycle ergometer Davies et al. (1970) investigated the effects of four different recovery intensities on blood L& clearance in a group of four subjects. Following 6 mm of exercise at 80% VO2max subjects performed 40 mm of recovery exercise at approximately 15, 30, 45 or 60% VO2m. The results showed that the optimal L& clearance rate occurred between 30- 45% VOmax. In contrast, using treadmill exercise consisting of three 60 s maximal efforts separated by approximately 4 mm of rest with the final work bout being followed by 30 mm of AR at one of four intensities (approx. 30, 60, 70 or 80% VO2m) Hermansen and Stensvold (1972) found that L& cleared fastest at approximately 63% (range 5570%) VO2max. In accordance with Davies et al. (1970), Belcastro and Bonen (1975) reported that optimal exercise intensity for La clearance on a bicycle ergometer was predicted at 32% VO2m and additionally those subjects were able to self-select adequate intensities to clear L&. Later, using treadmill exercise Bonen and Belcastro (1976) reported that self-selected running intensity, corresponding to approximately 61.4% (range 45.2-70.6%) VO2m, resulted in statistically significantly faster L& clearance rates compared to self-selected intermittent exercise and resting recovery. They noted that though these findings are similar to that of Hermansen and Stensvold (1972), their study did not allow the assertion of whether this intensity was optimal for L& clearance, since the two AR were self-selected (Bonen & Belcastro, 1976). In an attempt to differentiate between recoveries below and above AT, Stamford et al. (1981) demonstrated that the apparent rate of clearance could be manipulated by selecting different baseline asymptotes. Graphing semilogarithmic plots of La disappearance using a resting baseline L& value of 0.9 mmol•L’ showed statistically significantly faster clearance with AR at 40% VO2m compared to AR at 70% VO2max and PR. Conversely, using the same data with experimentally determined baseline values of L& (1.3 mmolL’ and 3.5 mmolU’, respectively) yielded no differences in disappearance between active recoveries. Therefore, both AR intensities were able to return L& 15 levels to their respective baselines faster than PR, but AR at 40% cleared L& faster overall. It may be that the 70% recovery may not have been above threshold as evidenced by the low baseline La value of 3.5 mmolL1 This may explain the lack of difference between recovery intensities, as L& would not have been accumulating in significant quantity if the workload was in fact below AT. In order to investigate whether the inconsistencies surrounding clearance rate and recovery intensity found above were related to exercise mode (i.e. bicycle vs. treadmill) Boileau et al. (1983) compared bicycle and treadmill recovery exercise at various intensities. They found no statistically significant differences between optimal L& clearance rate intensities across modalities (35.9% and 32.5% V02m for cycling and running, respectively). It should be noted that this experiment was comprised of a small sample of three females. In addition as suggested from another study with a larger sample size of seven males blood L& cleared fastest at moderate intensities (i.e. 28.2-43.1% VO2max, ref. Boileau et al., 1983). This study was conducted using only bicycle ergometry. It appears that the optimal recovery intensity to remove blood L& is moderate. This is more definite for bicycle ergometer work, but it may be the case that the optimal intensity is higher for treadmill recovery. It has been shown that for a given 02 consumption there is a greater rate of lactate production for bicycle exercise compared to running on a treadmill (Peters Futre et al., 1987). From a theoretical perspective this may be a consequence of the greater amount of musculature involved in running versus cycling, which may provide more potential for the oxidation of L&. However, the difference in recovery modes within individuals has not been extensively investigated, as the one study that has addressed this issue used a small sample (Boileau et al., 1983). The inconsistencies of the aforementioned studies (Belcastro & Bonen, 1975; Bonen & Belcastro, 1976; Davies et al., 1970; Hermansen & Stensvold, 1972) with respect to the optimal intensity for clearance may also be a result of the quantification of exercise intensity. It is known that two subjects with similar maximal aerobic power (MAP) may display different thresholds for the accumulation of L& (Dodd et al., 1984). Since previous research has typically quantified recovery loads as a percentage of V02m it follows that within and between studies subjects exercise capacities relative to L& thresholds may have varied. Logically this can be avoided by basing recovery intensities on L& thresholds. McLellan and Skinner (1982) investigated the intersubject variability of L& clearance rates when expressed relative to VO2max or aerobic threshold (AerT). Aerobic threshold was defined by the authors as the first “break” in the plot of 16 VE versus V02 and the initial continuous rise in La. They found that in 15 males (VO2mof 51.5 mL.kg1. in’) AerT values varied between 45-62% (X = 52.9%) VO2m and that recovery intensity expressed relative to AerT explained 13% more variance than VO2max (77% and 64%, respectively). Therefore it is slightly more advantageous to quantify recovery intensity by AerT than solely by VO2max. Moreover, peak L& clearance rate was predicted to be 10% below AerT (i.e. 43% VO2max), which is in agreement with values previously reported (Davies et al., 1970). Subsequent research typically has selected AR intensities from the aforementioned studies (e.g. Thiriet et al., 1993). As a result, the current belief is that AR at intensities between 30-45% VO2max is optimal for La clearance. This perception has been reinforced by research that has shown that L& clears fastest at moderate workloads (Dodd et al., 1984; Gmada et al., 2005; McAinch et al., 2004; Spierer et al., 2004). However, there are still investigations that have used higher intensity recoveries for both cycling (50% VO2max, ref. Monedero & Donne, 2000) and running (approx. 60% VO2m, ref. Bonen & Belcastro, 1976) and reported substantial L& clearance. More recently Baldari et al. (2004; 2005) have quantified recovery intensity in relation to both the IVT and TAT. They compared the effects of four 30 mm recovery intensities on blood L& clearance after 6 mm of treadmill running at 75% of the difference between TAT and VO2max (approx. 90% VO2m) in both soccer players and triathletes (Baldari et al., 2004; Baldari et al., 2005, respectively). The recovery intensities were: IVT, IVT+50%AT, TVT_50%AT and PR, where A is the difference between TAT and TVT. In soccer players it was found that the two lowest intensity recoveries (IVT_50%AT and IVT) were the most efficient for L& clearance and statistically faster than IVT+50%AT and PR (Baldari et al., 2004). Tn the triathietes IVT_50%AT removed L& statistically faster than the other three intensities (Baldari et aL, 2005). The authors note that all recovery intensities used were within the range (3 0-70% VO2m) previously reported for optimal L& clearance. Other investigations have used combined recoveries (CR, i.e. more than one recovery method within one recovery period). Taoutaou et al. (1996) compared PR (20 mm seated rest on bicycle ergometer followed by 40 mm seated rest) to partially AR (20 mm bicycling at 40% VO2max followed by 40 mm seated rest) and found that the partially AR cleared L& 1.5- and 3- fold faster in untrained and trained individuals, respectively. Donne and Monedero (2000) investigated blood L& clearance and subsequent exercise performance after four different 15 mm recovery interventions. The recovery interventions were: PR, AR at 50% VO2max, massage 17 recovery, and CR consisting of AR at 50% VO2max for 3.75 mm then massage for 7.5 mm followed by AR for the remainder. Their performance indicator was the difference in completion time for a 5 km bicycle time trial simulation performed before the recovery intervention and after on each of the testing days. Their results showed that AR and CR were superior to PR recovery with respect to L& clearance, that the fastest clearance rate occurred with AR, and that the clearance rate was fastest during the AR exercise periods of the CR. The increase in 5 km completion time was significantly less in the CR intervention compared to all other recoveries. The authors speculate that the greater performance maintenance in the CR trial was due to the combination of greater L& clearance in the active periods and increased intramuscular glycogen restoration during the passive massage periods. Post-exercise blood L& levels peak after a small lag in time of approximately 1-7 mm (Bret et al., 2003; Dodd et al., 1984; Gmada et al., 2005; Taoutaou et al., 1996; Thiriet et al., 1993), since L& must be transported out of the cell via MCTs to the circulatory system. The time to reach peak blood L& appears to be dependent on metabolic rate, since it peaks faster with higher intensity recovery intensities, and sooner with AR versus PR (Gmada et al., 2005; Stamford et al., 1981; Taoutaou et al., 1996). Afterwards blood L& levels will begin to decrease provided that the energy demand is below the threshold at which blood L& accumulates. Depending on the peak blood L& levels and recovery intensity, values may remain elevated for up to 1.5 hours before reaching resting levels (Bret et al., 2003; Choi, Cole, Goodpaster, Fink, & Costill, 1994; Taoutaou et al., 1996). Studies have utilized a wide variety of recovery durations ranging from 4-90 mm to investigate La clearance (Corder et al., 2000; Taoutaou et al., 1996). When comparing clearance rates between AR and PR statistically significant differences are usually not evident until approximately 10 mm (Gmada et al., 2005; Monedero & Donne, 2000). Interestingly, the two investigation previously discussed by Baldari et al. (2004; 2005) both demonstrated that all active recoveries examined did not show further decreases in blood L& after the twentieth minute of recovery. The investigations examined two different subject pools: soccer players and triathietes (VO2m of 62.3 and 69.7 mL.kg’•min’, ref. Baldari et al., 2004; Baldari et al., 2005, respectively). The recovery intensities were: IVT, IVT+50%iXT, IVT_50%AT and PR, and ranged between the previously described intensities of 3 0-70% VO2max that have been suggested to be optimal for L& clearance (range 39-60% VO2max). 18 2.4.2 Combined Active Recovery and Lactate Dodd et al. (1984) and more recently Gmada et al. (2005) have investigated the effects of a two stage CAR on La clearance. The precedence for their research came from the work of Stamford et al. (1981) who compared the effects of three different 40 mm recoveries on L& clearance: PR, and AR at 40% and 70% VO2max (AR4O% and AR7O%, respectively). They found that L& clearance was greater in the AR4O% from 3 0-40 mm compared to PR and AR7O%. Additionally it was shown that L& levels peaked faster with AR7O%. Since LaT uptake is proportional to its concentration (Jorfeldt, 1970; Margaria et al., 1933) it was theorized that as L& concentration decreases, uptake decreases in proportion, resulting in a decrease in clearance rate since La is still being produced during active recovery, albeit at lower levels. Thus it was hypothesized that optimal L& clearance may occur with an exercise intensity that matches its concentration (Stamford et al., 1981). That is, as L& concentration progressively decreases the exercise intensity should decrease to limit L& production and therefore enhance clearance. The first group to test the latter hypothesis was Dodd et al. (1984). They combined moderate- to high-intensity and moderate- to low-intensity work within one recovery period in an attempt to optimize L& clearance. A sample of seven trained males (VO2max of 48.7 mL•kg ‘.min’) performed 50 s of supramaximal bicycle work at 150% VO2m followed immediately by one of four 40 mm recoveries: PR, AR at 35% VO2m (AR35%), AR at 65% VO2max (AR65%), or 7 mm at 65% followed by 33 mm at 35% VO2max (CAR). Differences in clearance rate at the 6th mm and from mm 20 to 40 of recovery were examined, and results demonstrated that from minute 20 to 40 La clearance was statistically significantly faster in the AR3 5% and the CAR trial. No statistically significant differences were observed between AR3 5% and CAR during this time frame and the authors therefore concluded “that these data do not support the hypothesis that following maximal work, a combination of submaximal exercise intensities is more beneficial in lowering blood L& concentrations than a single intensity” (Dodd et al., 1984). Conversely, in retrospect their results could be interpreted such that CAR was able to clear La to the same extent as AR35% since there were no statistical differences between the two conditions. The question can then be proposed that if the two recovery strategies differ in intensity but still clear La to the same extent, do they differ in any other respects. For example, do they result in differences in performance independent of their respective effects on L&. Gmada et al. (2005) re-investigated this hypothesis with several modifications to the work of Dodd and colleagues. A larger sample of fourteen subjects (7 trained and 7 untrained, VO2of 56.5 and 42.0 n’.L.kg’.min’, respectively) performed three supramaximal intermittent 19 exercise bouts at 120% MAP for 60% of the time to exhaustion (TTE) separated by 5 mm intervals with the third bout being followed by 20 mm of recovery. The recovery bouts were as follows: PR, AR at 20% less than the first ventilatory threshold (VT1), AR at 20% less than the second ventilatory threshold (VT2), and CAR consisting of 7 mm at VT2 followed by 13 mm at VT1. Their findings demonstrated that peak blood L& occurred faster in CAR and VT2 conditions for both trained and untrained subjects (4t11 and 7th mm, respectively). This was in accordance with previous work (Dodd et a!., 1984; Stamford et al., 1981) and confirmed that blood L& peaks sooner post-exercise with higher-intensity recovery. In contrast to Dodd et al. (1984) it was determined that L& disappeared statistically significantly faster in both groups with CAR, the effect being more pronounced in the trained group (Gmada et a!., 2005). The discrepancy between results with respect to the efficacy of CAR of the two studies could be due the differences in subject’s fitness level or the protocols used. Dodd et a!. (1984) used only one group of seven trained subjects with an average VO2m of 48.7 mL.kg’.min’. Whereas Gmada et al. (2005) used two groups of seven subjects, one trained and one untrained with VO2max of 56.5 and 42.0 mL.kg1•min’, respectively. Furthermore, Gmada et a!. (2005) quantified the recovery intensities specific to ventilatory thresholds, while Dodd et al. (1984) only used VO2max. Since two subjects with a similar VO2m can exhibit different anaerobic thresholds, the possibility exist that the recoveries chosen in the study by Gmada et a!. may have been less likely to surpass an individual’s anaerobic threshold and therefore result in less L& production during the recovery period (2005). Furthermore, Gmada et a!. used a repetitive exercise task to induce increased blood La, which resulted in higher blood L& concentrations. Since the rate of L& uptake is proportion to its concentration (Jorfeldt, 1970; Margaria et al., 1933) it may be that the higher blood La levels had some effect on clearance rate. However, with respect to the actually blood L& concentrations this does not appear to be the case, although it may be that the effect was not evident in blood L& concentrations. Additionally, the 5 mm recoveries between the repetitive exercise bouts in the study by Gmada et al. (2005) matched the longer duration recovery period, with respect to recovery intensity, and may have influenced the results. In summary it has been shown that the use of AR facilitates L& clearance compared to PR. It may be that the optimal AR intensity to clear blood L& is dependent on recovery mode or the active muscle mass. Moderate intensity AR is recommended to clear blood La fastest when performing bicycle exercise. However, the optimal intensity with which to clear L& remains 20 undetermined and may be dependent on individual thresholds or occur with the use of more than one recovery intensity. 2.5 Lactate and Performance As shown above AR is an appropriate means to remove L& following high-intensity exercise in which there is an accumulation of L&. However, the real interest surrounding AR for coaches, athletic trainers, strength and conditioning coaches, and athletes is whether or not AR improves performance. The evidence for the use of AR as a means of improving performance following previous exercise is less obvious. Lactate is the one of the most researched metabolites. From an early time it was believed to be the direct cause of fatigue (Hill & Kupalov, 1929). The concept that AR from exercise bouts may be beneficial to subsequent performances like many training and competition techniques in the sport sciences has some of its origin in anecdotal observations. Gisolfi, Robinson and Turrell highlight this in their 1966 paper in which they conclude that their work provides a physiological basis for the practice of AR post-competition; which they note athletes had already learned from experience. The authors formulate this conclusion after finding that moderate aerobic exercise (approx. 38-53% VO2max) for 30-35 minutes following exhaustive treadmill running reduced the “oxygen debt” and cleared L& faster than a resting recovery in the four subjects they examined (Gisolfi et al., 1966). No attempt was made to examine performance in this investigation and their conclusion was likely established based on current beliefs of that time pertaining to L& and fatigue. Later, more empirical evidence would imply that high “lactic acid” concentration may be the reason for exhaustion in high-intensity exercise (Karlsson & Saltin, 1970). To test the hypothesis that high blood La concentration may limit maximal exercise performance Klausen et al. (1972) had subjects perform maximal bicycle work that was either preceded by rest or high-intensity arm ergometry. Therefore, high blood L& concentration was induced and its effect on a previously non-exercised muscle group was examined. While their results were not statistically significant they did observe an average 10% reduction in TTE in the condition in which leg work was preceded by arm work. The authors concluded that increased L& concentration in the working muscles inhibited further La production, but that the hypothesis that L& is a limiting factor in exercise was not confirmed since only a trend to reduced endurance time was observed. In a similar experiment in which arm exercise preceded leg exercise (series A-L) and vice versa (series L-A, i.e. leg exercise preceding arm exercise) on a separated occasion, it was observed that TTE occurred earlier in both conditions with respect to 21 TTE without preceding exercise (Karlsson et al., 1975). The authors concluded that because L& was elevated prior to the second exercise bout in both conditions and therefore reached peak values sooner, these peak La values (20-30 mmolkg’ wet muscle) or related factors serve as limiting factors to muscular performance. However, the latter two studies (Karlsson et a!., 1975; Klausen et a!., 1972) had small sample sizes (n = 4 and 3, respectively) limiting the conclusiveness of their results. It is also worthwhile to note that the two investigations did not use SE bouts. Nonetheless, the investigations are foundational to the belief that increased blood L& concentration can limit performance, or perhaps more appropriately stated, is related to performance decrements. Furthermore, it is this foundational belief that allows the conclusion that AR may be beneficial to performance since AR can clear blood L& faster than a resting recovery. 2.6 Active Recovery and Performance The first empirical evidence that AR was beneficial to subsequent high-intensity performance came from Weitman et a!. (1977). After performing an initial all-out SE task (bicycling for 1 mm at 5.5 kg resistance) subjects underwent one of eight recovery interventions followed by the same supramaximal criterion exercise task. Recovery interventions consisted of AR at 1 kg resistance or PR for either 10 or 20 mm breathing either room air or 100% 02. All combinations of the aforementioned variables were examined. Their findings showed that L& clearance and subsequent performance were statistically significantly improved with active and 20 mm recovery compared to the other interventions. However, they also concluded that other factors beside L& clearance are critical to subsequent performance since these variables were not correlated. In the time to come research findings would be divided on the effects of AR on subsequent performance. While all studies showed that AR could decrease L& concentration better than PR, many showed no additional benefits to subsequent exercise performance (Bond, Adams, Tearney, Gresham, & Ruff, 1991; Franchini et a!., 2003; McAinch et al., 2004; Siebers & McMurray, 1981; Watson & Hanley, 1986; Weltman et al., 1979; Weltman & Regan, 1983), while others would show benefits (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Connolly et al., 2003; Corder et a!., 2000; Signorile et al., 1993; Spierer et al., 2004; Thiriet et al., 1993). Typically studies aimed at investigating subsequent performance have used recovery durations of 10-20 mm. This is likely due to the fact that this time frame coincides with requirements and restraints put on sporting events. Previously it has been suggested that 20-40 mm of AR should be used to prevent a decrease in power output (Ainsworth et al., 1993). 22 Interestingly, with respect to performance, recent evidence suggests that shorter durations of AR may be beneficial to performance (i.e. approx. 6 mm) rather than longer durations (i.e. 15 mm or longer, ref. Ahmaidi et al., 1996; Bogdanis et a!., 1996b; Spierer et al., 2004). 2.6.1 No Performance Benefit with Active Recovery A substantial body of research exists that shows that AR does not significantly improve performance. Weltman et al. (1979) studied the effects of four recoveries (PR, ARAT, and AR>AT + 100% 02) on La clearance and subsequent performance for an endurance task consisting of 5 mm of cycling in nine males. Subjects cycled for 5 mm at MAP, recovered for 20 mm at one of the respective recovery intensities, and then performed a second 5 mm cycle at MAP. They reported that while AR>AT + 100% 02 cleared La significantly faster than PR, and AR>AT there were no significant differences in performance among recovery conditions. Performance was determined by work done over the 5 mm and assessed by pedal revolutions completed. Siebers and McMurray (1981) investigated blood L& clearance and subsequent performance of a 200 yd swim 15 mm after a 2 mm swim at 90% VO2m on a swimming ergometer in six females. The 15 mm recoveries consisted of either walking at a moderate pace (2.5-3 mph) for 10 mm followed by 5 mm of PR or swimming continuous front crawl lengths at a moderate pace for 10 mm followed by 5 mm of PR. Despite reporting that swimming recovery cleared 22% more La than walking, there were no statistically significant differences for 200 yd swim time. Subjects took slightly longer than 2 mm to complete the 200 yd swims for the swim and walk recoveries (125.9 ± 5.9 s and 127.2 ± 5.8 s, respectively). In an attempt to amalgamate the literature, Weltman and Regan (1983) investigated the effects of 20 mm of active and passive recovery on subsequent maximal constant load (i.e. no work drop-off allowed) exercise performance. At this point in time research supporting the contention that elevated blood L& concentration has a detrimental effect on subsequent performance had used a constant load protocol (Karlsson et al., 1975; Klausen et al., 1972) and research not in support had used maximal effort work (i.e. allowing a work drop-off, ref. Weltman et al., 1979). In contrast to the previous work utilizing constant load tasks, they found no statistical differences in work output between the recovery conditions. An applied investigation had eight hockey players perform two simulated hockey tasks separated by one of three 15 mm recovery modes: skating, bench-stepping, or PR (Watson & Hanley, 1986). The hockey task was comprised of six 45 s sprints each separated by 90 s PR, with the distance skated determining each player’s performance. Results demonstrated that only bench-stepping reduced L& values with statistical significance compared to rest, but that neither 23 of the AR altered performance with statistical significance. However, the practical nature of this investigation resulted in difficulty standardizing recovery duration and intensity, since the AR condition had substantial passive periods to take measurements and the PR condition had substantial active periods to prepare for the subsequent exercise task. The authors note that the skating recovery may have resulted in slowed La clearance because of the ability of the players to glide, limiting the amount of muscular work they performed. The effect of this on performance cannot be determined. Bond et al. (1991) investigated the effects active and passive recovery on subsequent isokinetic muscle function. They used a 60 s bicycle ergometer task at 150% VO2max to elevate blood L& levels followed by either 20 mm of AR at 30% VO2max or PR. Subjects then performed 60 repeated isokinetic knee extensions at an angular velocity of 1 500.s4.Results again confirmed the superiority of AR to PR with respect to La clearance but no differences in peak torque, total work or fatigue index were noted between recovery conditions or control values. Another applied investigation used two simulated hockey tasks consisting of seven 40 s ‘shifts’ separated by 90 s rest with 15 mm of AR or PR between skating tasks (Lau, Berg, Latin, & Noble, 2001). The AR, which consisted of self-selected resistance “low-intensity” cycling, did not clear L& faster than the PR or effect subsequent performance in a beneficial way with respect to statistical significance. However, the authors did note a trend towards a greater distance skated in the second bout of simulated hockey shifts with AR. Additionally, the authors note that their findings may have been limited by the fact that the recovery intensity was self-selected and that the skating bouts only induced moderate L& concentrations. Furthermore, similar to Watson & Hanley’s (1986) investigation the recoveries were not strictly passive or active due to constraints of the experimental design. Recently two investigations have examined the effects of active versus passive recovery on La clearance and subsequent anaerobic and aerobic performance, respectively (Franchini et a!., 2003; McAinch et a!., 2004). In one of the investigations a group of seventeen subjects first participated in a 5 mm judo combat followed by 15 mm of AR (running at 70% of the anaerobic threshold velocity) or PR (Franchini et al., 2003). They then completed an intermittent anaerobic task comprised of four upper body WAnT each separated by 3 mm PR. Performance on the WAnT was not altered by recovery mode. The authors note that this finding is in accordance their observations that AR seems to be beneficial to performance when the recovery period is 6 mm or less, and that with recoveries of 15 mm or longer AR does not appear to be beneficial. This is further supported by the work of McAinch et a!. (2004). They investigated the effects of 24 AR (40% VO2max) and PR on muscle biopsies and plasma L& clearance, as well as performance of intense aerobic exercise. Seven male subjects performed two 20 mm bicycle trials separated by 15 mm of recovery. No differences in work performed or muscle glycogen and La concentrations were observed, but plasma L& concentration was significantly lower in the AR protocol. 2.6.2 Performance Benefit with Active Recovery Despite the wealth of literature that has shown that AR does not appear to be beneficial to performance there is a similar amount of literature to the contrary. In addition to the initial research that exhibited that AR may be advantageous to recovery and/or maintenance of performance (Karisson et al., 1975; Klausen et al., 1972; Weitman et a!., 1977) others have produced supporting literature. Pendergast et al. (1983) confirmed the earlier results of Karisson et al. (1975) that preceding high-intensity exercise considerably reduces the potential for further supramaximal performances. They found that endurance for both aerobic and anaerobic work was reduced in the presence of high blood L&. Other investigators have documented a similar relationship of increased blood L& concentration and reduced muscular endurance (Yates et al., 1983). These investigators looked at the muscle contractile properties (maximum voluntary contraction (MVC), peak rate of tension development, peak rate of relaxation, one-half contraction time, and one-half relaxation time) of the elbow flexors 6 mm after 1 mm of intense cycling at a fixed load of 5 kg versus a control (i.e. no prior cycling). They found no statistically significant differences in muscle contractile properties after the cycle ergometer bout, but endurance time at 40% MVC was reduced by 25% with prior exercise. After the endurance task there was a statistically significant reduction in MVC, peak rate of tension development and peak rate of relaxation. It was concluded that the elevation of blood L& by intense exercise of one muscle group reduced the endurance of a second non-exercised muscle group. However, the aforementioned investigations (Pendergast et al., 1983; Yates et al., 1983) did not actually investigate the effects of AR, but rather showed that increased blood L& and associated changes result in decrements of performance. The investigators both suggest that the reduction of L& concentration should therefore be beneficial to performance, as increases in L& are detrimental to performance. This suggestion is more of an anecdotal assertion than scientific fact, but nonetheless provided a conceptual framework for the conviction that AR and hence blood L& clearance is beneficial to performance. Thiriet et al. (1993) investigated the effects of AR and PR on repeated SE. They had 16 male subjects perform four cycling bouts to exhaustion at 130% MAP. Each bout was separated 25 by a 20 mm recovery period of either leg or arm ergometric exercise at 30% MAP or rest. Active recovery cleared L& faster and maintained work performance more than PR. However, they also noted a non-significant correlation between power output and L& levels, which has been observed by others (Siebers & McMurray, 1981; Weitman et aL, 1977; Weitman et al., 1979; Weitman & Regan, 1983). Therefore, they concluded that the relationship between power output and L& is not one of cause and effect. Other researchers have corroborated the finding that power output may be maintained to a greater extent with AR compared to PR (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Connolly et al., 2003; Signorile et al., 1993; Spierer et al., 2004). Signorile et al. (1993) examined the effect of AR versus PR on power output during eight 6 s supramaximal bicycle sprints separated by 30 s. Active recovery consisted of pedaling against 1 kg of resistance at 60 rpm while PR consisted of sitting on the bicycle motionless. Mean PP and mean TW performed were statistically significantly greater in the AR protocol. This investigation used a fixed load for the recovery intensity based on previous work (Weitman et al., 1977), which limits the control of interindividual differences in fitness and work capacity. Biochemical variables were not measured and therefore the interpretation of the data is limited solely to performance parameters. Bogdanis et al. (1 996b) compared the effects of recovery type on performance of two maximal 30 s bicycle sprints separated by 4 mm. Active recovery resulted in statistically significantly higher MP output compared to PR. The difference in power output could be attributed to the differences observed in the initial 10 s of the sprint. The authors suggest that the increased blood flow during AR may have increased resynthesis of PCr or allowed an initially faster glycolytic rate as an explanation for the performance improvement based on results of their previous work (Bogdanis, Nevill, Lakomy, & Boobis, 1994b). Blood La concentration did not differ significantly between recovery conditions (Bogdanis et al., 1 996b). Another investigation examined the effects of recovery type on repetitive 6 s bicycle sprints using incremental resistive forces separated by 5 mm of active (32% MAP) or passive recovery in ten male subjects (Abmaidi et al., 1996). The results showed that at the higher resistive forces the AR protocol enabled greater maintenance of power and also cleared L& faster than the PR protocol. Dorado et al. (2004) examined the effects of recovery mode on aerobic and anaerobic energy yield as well as performance during high-intensity intermittent exercise. Ten trained subjects (VO2m of 58 mL.kg’.min’) performed four supramaximal constant intensity cycling bouts to exhaustion at 110% maximal power output each separated by one of three 5 mm recoveries. The recoveries were: AR at 20% VO2max (HITA), stretching recovery of the lower 26 limbs (HITS), or PR (HITP). Performance was 3-4% better and aerobic energy yield was 6-8% greater in the HITA condition. The greater aerobic yield was due to faster V02 kinetics and the authors concluded that this was the source of improved performance in the AR trial. It was proposed that the faster V02kinetics were a result of either increased blood flow or maintenance of aerobic regulatory enzyme activation (Bangsbo et al., 1994). It has previously been shown that aerobic metabolism makes a significant contribution to metabolism during high-intensity exercise (Bogdanis, Nevill, Boobis, & Lakomy, 1 996a) and the authors argued that this was the case in their investigation (Dorado et al. 2004). More recently, a study investigated the use of a short 3 mm recovery period on La clearance and power output (Connolly et al., 2003). In congruence with previous work utilizing similar work-to-rest intervals (Bogdanis et al., 1996b), it was found that power output on six 15 s bicycle ergometer sprints was greater with AR compared to PR, but L& values did not differ with respect to the recovery used (Connolly et al., 2003). Spierer et al. (2004) examined both moderately trained ice hockey players (VO2max of 45.6 mL.kg’.min’) and sedentary (VO2max of 36.9 mLkg’ .min’) subjects on their ability to perform serial WAnT interspersed with either 4 mm AR at 28% VO2max or PR. Their results showed that PP output did not differ significantly between recovery types in both groups. However, sedentary subjects displayed statistically significantly greater MP output and TW values with AR, whereas only TW performed was significantly improved in moderately trained individuals with AR. Capillary blood L& differed with statistical significance in the moderately trained group only when AR was employed. It should be noted that there were statistical differences between groups with respect to age, gender, height, and mass in addition to VO2max, which complicates the inter-group comparison since the group differences outside of fitness level are a source of uncontrolled variability. Ainsworth et al. (1993) examined the effect of AR duration on blood L& and power in 16 male competitive cyclists (VO2max of 67.6 mL.kg’.min’). Following a 45 s bicycle bout subjects performed 6, 9, or 12 mm of AR at a fixed resistance of 5.5 kg, after which they immediately performed another 45 s bicycle bout. Results showed that power output was decreased with statistical significance between bouts during the 6 mm recovery, but was maintained in the 9 and 12 mill recoveries. However, no statistically significant differences were observed in the ability to produce PP in all of the recovery durations. Recovery blood L& was only statistically decreased in the 12 mm recovery protocol. It was therefore concluded that in this population 9 mm of AR at approximately 30% VO2max was sufficient to restore power output to resting levels 27 following 45 s of supramaximal cycling. In support of this finding, another investigation in which 8 males performed two 30 s bouts of bicycle ergometer work separated by 6 mm of PR, PP and MP output were only 92% and 85% of initial control values (Bogdanis, Nevill, & Lakomy, 1994a). However, this investigation used passive recovery and therefore a direct comparison of results is contentious. Nonetheless, it appears that following maximal work durations of 30-45 s greater than 6 mm of recovery is necessary for power recuperation. 2.6.3 Performance and Active Recovery Duration One commonality of the aforementioned research on AR and performance that has not found an improvement in subsequent performance is that the recovery durations were all 15 minutes or longer. Therefore, it may be that recoveries of this duration mask any additional benefits of an AR that may only be evident in the initial part of the recovery period. This contention is supported by the fact that much of the literature that has shown improved performance with AR has used recovery durations of 6 mm or less (Ahmaidi et al., 1996; Bogdanis et a!., 1996b; Signorile et al., 1993; Spierer et a!., 2004). This trend has been previously noted (Franchini et al., 2003). Of the studies reporting improved performance with AR in this literature review only one used a long recovery duration (i.e. 20 mm, ref. Thiriet et a!., 1993). Interestingly, a study that investigated performance on a resistance training (i.e. parallel squat) task in which exercise sets were separated by 4 mm of cycling or PR it was shown that AR cleared L& faster and resulted in better performance (Corder et al., 2000). This is in agreement with the aforementioned trend. Also of interest, is the fact that several of the studies reporting improved performance with AR did not correlated or associated this improvement with L& clearance (Bogdanis et al., 1996b; Connolly et al., 2003; Signorile et al., 1993; Spierer et al., 2004; Thiriet et al., 1993). Therefore, while it appears that a short duration AR may be beneficial to performance it may not be related to the clearance of La and could be a result of increased blood flow and its effects on subsequent metabolism (Bogdanis et al., 1994b; Bogdanis et al., 1 996b). 2.7 Summary It has been unequivocally shown that the use of AR can remove L& at a faster rate than PR. It is proposed that the accelerated clearance rate with AR is a result of heightened blood flow that serves to circulate L& and provide a fuel source for the working muscles while also being oxidized by resting muscles and other tissues to a lesser extent. While it has been postulated that increased L& clearance should be beneficial to subsequent exercise performance the literature is indeterminate. Active recovery may be only beneficial for short duration 28 recovery because it can maintain blood flow increasing the aerobic energy contribution to exercise, whereas with longer durations the return to homeostasis may be more similar between recovery types and result in similar performance. 29 CHAPTER III: METHODS 3.1 Subjects The subjects were comprised of a sample of 12 volunteer, trained, male, university-aged (18-35 years) cyclists. An a priori power calculation performed using the G*Power software package (G*Power Version 2.0, Germany) was used to determine sample size (Faul & Erdfelder, 1992). The analysis was performed with the intent of detecting a 10% decrement in MP output assuming an average final power output of 800 W with a standard deviation of 50 W. A 10% power decrement was selected because a decrease of this magnitude or less in muscle performance may be sufficient to limit whole body exercise performance in a competitive setting (Sprague & Mann, 1983). The power output stated above is based on previous research examining the performance of competitive road cyclist on a single WAnT (Tanaka, Bassett, Swensen, & Sampedro, 1993) and preliminary pilot data from this laboratory. Subjects were recruited from the University of British Columbia campus and Vancouver area. All subjects were non-smokers and not under any pharmacological or special dietary treatment during the investigation. Subjects were defined as trained for this investigation if their training included both a heavy aerobic and anaerobic component. Additionally, subjects were required to meet at least three of the following four criteria: currently participating in regular competitive level cycling, have PP and MP outputs on the WAnT greater than or equal to 11 and 9 W.kg’, respectively, and have a VO2mgreater than 55 mL.kg’.min’. Prior to participation all subjects were informed of potential risks and benefits associated with participation and completed a written Informed Consent, approved by the UBC Clinical Research Ethics Board (see Appendix X), and a Physical Activity Readiness Questionnaire (PAR-Q). Subjects were required to weigh less than 95 kg, as the resistance load on the Monark (Ergomedic 874E, Monark Exercise AB, Sweden) ergometer becomes less accurate with athletes over this weight (Inbar, Bar-Or, & Skinner, 1996). However, no subjects outside of this requirement attempted to participate in the study. 3.2 Experimental Design A randomized counterbalanced within-subjects design was used to evaluate the treatment effects. Subjects were randomly assigned to one of three recovery modes (AR, PR, and CAR) and the order in which they performed each trial was counterbalanced in order to control for treatment order effects. 30 3.3 Facilities and Instrumentation All testing was completed at the John M. Buchanan Exercise Science Laboratory within the University of British Columbia Aquatic Centre. Equipment was calibrated, as per manufacturers’ instructions, prior to testing. Anthropometric measurements of height and weight were taken for each subject on the first testing day prior to exercise. Additionally, the sum of five skinfolds (biceps, triceps, subscapular, iliac crest, and medial calf) were taken as a measurement of body composition according to the procedures set out in The Canadian Physical Activity, Fitness & Lifestyle Appraisal Manual (CSEP, 1998), using Standard Harpenden calipers (Baty International, UK). 3.3.1 Wingate and Sprint Tests Performance was measured using a repetitive sprint task comprised of five 10 s maximal sprints separated by 30 s of PR. Both the sprints and WAnT were performed on a pan load Monark Ergomedic 874E (Monark Exercise AB, Sweden) bicycle ergometer. Subjects cycled against a set load of 0.09 kg.(kg body mass)’, for the respective test durations, as recommended for athletes on the WAnT (Inbar et a!., 1996). In order to calculate power outputs the velocity of the bicycle flywheel is determined by way of an optical sensor (SMI OptoSensor, USA) that records pulses from reflective markers fitted to the flywheel. The sensor was interfaced with a PC equipped with SMI POWER Version 5.2.8 software (SMI, USA). The software then calculates power parameters based on the measured flywheel velocity and belt friction (i.e. applied resistive force). Peak Power and MP were calculated for the WAnT for use as subject inclusion criteria. Peak Power, MP, and Fl were calculated for each of the five sprints, and TW was calculated for the entire sprint trial using Microsoft Excel Version 5.1.2600 (Microsoft Corporation, USA). The SMI POWER (SMI, USA) standard power outputs were used rather than the corrected power outputs. The variables calculated were defined as: • Peak power — the highest mechanical power output over 1 s achieved during the sprint. • Mean power — the average power output maintained over the entire 10 s sprint. • Fatigue index — the difference between PP and the lowest 1 s power output divided by PP and expressed as a percentage. • Total Work — the sum of the product of mean power and time across all five sprints. The ability to objectively quantify the capacity for intense activity is one of the most difficult components of measuring athletic performance (Inbar et a!., 1996). Since anaerobic ATP production is an intracellular process there are no precise methods to quantify the energy release, and therefore no direct “gold standard” method of validation (Gastin, 2001; Inbar et al., 31 1996). Therefore like other measures of anaerobic capacity the validity of this type of sprint task is contestable. However, it has been stated that “the choice of an anaerobic test depends on the aims and subjects of a study and its practicability” (Vandewalle, Peres, & Monod, 1987). Therefore, due to practicability and the desire to examine repeat high-intensity exercise performance, the model of five 10 s sprints was chosen as the performance indicator for this investigation. 3.3.2 Ventilatory and Gas Exchange Variables Maximum oxygen consumption was evaluated by an incremental stage bicycle ergometer test, performed on an electronically braked SensorMedics Egrometrics 800 bicycle (SensorMedics, USA) utilizing the SensorMedics Vmax 29 series metabolic measurement cart (SensorMedics, USA). Breath-by-breath values of V02, VCO2 (volume of CO2 produced), VE (minute ventilation) were averaged over 20 s intervals and recorded for analysis. The exact protocol used was adapted from previous incremental bicycle and treadmill tests (Baldari & Guidetti, 2000; Stegmann et al., 1981). Subjects started at an initial workload of 120 W and performed 3 mm stages with step increments of 30 W until the attainment of their TAT. After which, the stage durations were decreased to 1 mm and the step increments were maintained at 30 W until exhaustion. The criteria used to determine the attainment of VO2max was operationally defined as the achievement of any two of the following: volitional fatigue, plateau in V02 (i.e. change < 2.1 mL.kg’.min), 90% age predicted maximum HR, respiratory exchange ratio (VCO2/ 0)exceeding 1.15, or blood L& concentration greater than 8 mmolL’ (Duncan, Howley, & Johnson, 1997). Subjects were asked to maintain a pedal cadence above 80 rpm when possible to facilitate the ease of cycling on the mechanically braked bike, but were allowed to pedal at any cadence that was above 60 rpm. The same apparatus was used on subsequent testing days to record ventilatory data during the exercise and recovery bouts. Data was again sampled breath-by-breath and averaged over 20 s intervals for analysis. 3.3.3 Blood Lactate Lactate measurements were made using a portable L& analyser (ARKRAY Inc., Japan). Briefly, the finger to be lanced was disinfected with an alcohol pad before a fingertip puncture was made. The first drop of blood was cleared away (along with any non-evaporated alcohol) with sterile gauze and then a second drop was pressured out to fill a reagent strip via capillary action with approximately 5 iL of blood. Lactate in the sample reacts with potassium ferricyanide and La oxidase to form potassium ferrocyanide and pyruvate. A given voltage is 32 then applied oxidizing ferrocyanide, releasing electrons and creating a current. The current is measure amperometrically and is directly proportional to the blood L& concentration of the sample (Pyne, Boston, Martin, & Logan, 2000). Capillary blood L& values have been shown to accurately reflect arterial blood L& values (Williams, Armstrong, & Kirby, 1992). The Lactate Pro has been shown to be accurate, reliable and demonstrate a high degree of agreement with other L& analysers (Pyne et al., 2000). 3.3.4 Heart Rate Heart rate was measured and recorded using a Polar s120 HR Monitor and uploaded to a PC using Polar Infrared Connection (Polar Electro, Finland). Heart rate data was provided every 5 s. 3.4 Testing Procedures Subjects attended the lab on six days separated by at least 48 hours. One exception to this guideline was made for Subject 3 who completed testing days 2 and 3 within 28 hours due to a conflict of schedule. On all testing days subjects were asked to report to the laboratory two hours after a suggested snack of a whole wheat bagel and banana. In addition, to control for the effects of nutritional and exercise status on performance subjects were asked to follow similar dietary and exercise habits throughout the experiment on both testing and off-days. Furthermore, in the 24 hours preceding a testing day subjects were asked to avoid intense physical activity, alcohol intake, and refrain from caffeine intake in the 3 hours preceding a testing session. Prior to all testing days, subjects were asked to report whether or not they complied with the aforementioned guidelines. 3.4.1 Day 1: Anthropometric Measures, VO2and WAnT On subjects’ first testing day anthropometric measurements and skinfolds were taken before performing a VO2max test (as described above), followed by the WAnT. Bicycle seat and handle bar height were adjusted to the individuals comfort and recorded for subsequent use throughout the experiment. Blood L& measures were taken during the VO2max test at the end of each 3 mm stage, until the IAT was determined. A final blood L& measure was taken 3 to 4 minutes after the incremental test. During this time subjects cycled at 40 W in order to alleviate any discomfort associated with the cessation of intense physical activity. The IAT was operationally defined as the workload corresponding to the second L& increase of at least 0.5 mmolL1 from the previous value (Baldari & Guidetti, 2000). That is, the TAT is the workload corresponding to the stage that elicits a second L& increase greater than or equal to 0.5 mmolL1.However, in the aforementioned investigation the investigators demonstrated that the 33 maximal L& steady state (MLSS) was more accurately predicted when the workload from the antecedent test stage was attributed to the blood L& value, rather than the workload of the same stage. Therefore in this investigation the workload used to quantify the IAT was the workload corresponding to the stage antecedent to that in which the IAT blood L& value was observed. Additionally the respiratory data collected during the final minute of each test stage was averaged and used to calculate the IVT. Specifically, VE/V02was plotted as a function of V02 and the level of V02 at which VE/V02 was the lowest corresponds to the IVT (Baldari & Guidetti, 2001, see Figure 2 for visual representation). The workload corresponding to the stage in which the IVT was found was used to quantify the recovery wattage. Twenty minutes P05t02max subjects performed a WAnT. During the 20 mm recovery period subjects cycled at 40 W and were permitted to consume fluids at will. In the last 5 mm of the recovery period subjects were allowed to dismount the bicycle ergometer and stretch. Subjects then transferred to the Monark (Ergomedic 874E, Monark Exercise AB, Sweden) bicycle and the seat and handle bar height were adjusted and recorded for subsequent use. Subjects were instructed to start from a “rolling” start, cycling at a light cadence (approximately 60-80rpm) against no resistance. They were then given a 10 s and 5 s warning for the commencement of the test. The SMI POWER software (SMI, USA) was set to have a 3 s count down which was audibly counted for subjects. Subjects were instructed to achieve maximum pedal cadence by the end of the 3 s count and at this time the pan loaded resistance (0.09 kg.(kg body mass)’) was applied. Subjects were instructed to remain seated and to attempt to maintain maximum pedal cadence throughout the entire 30 s period. At the completion of the WAnT subjects were allowed to cycle at a self selected resistance on either bicycle ergometer to allow for venous blood return and prevent blood pooling as well as alleviate any discomfort associated with the cessation of high-intensity exercise. 3.4.2 Day 2: Time to Exhaustion Test On Day 2, after a warm-up (10 mm at a workload corresponding to 50% of VO2max), subjects performed a TTE ride at an intensity corresponding to 120% MAP. The same bike used for the VO2m test was used for this ride and later exercise trials (Egrometrics 800, SensorMedics, USA). Subjects were given verbal encouragement throughout the task and were instructed to maintain a cadence over 60 rpm. A verbal warning was given when their cadence fell below 60 rpm and they were allowed a few seconds to bring their cadence back up. Exhaustion was operationally defined as an inability to maintain a cadence above 60 rpm, and subjects were told they could receive up to three warnings before the test would be terminated. 34 However, in practice all but one subject were unable to bring their cadence back above 60 rpm after the initial decline and warning. Sixty percent of the TTE was then used for the duration of the subjects’ work intervals on the following familiarization and trial days. 3.4.3 Day 3: Familiarization Tasks For Day 3 subjects performed familiarizations of the sprint and work tasks in order to ensure that they could complete the required intensities and to improve their ability to reproduce the test protocol (Le Panse et al., 2005). After an equivalent warm-up to Day 2, followed by a brief transition to the Monark (Ergomedic 874E, Monark Exercise AB, Sweden) bicycle, subjects performed the sprint familiarization. Identical instructions to the WAnT were given with respect to the start and countdown. Subjects performed five 10 s sprints against a resistance of 0.09 kg.(kg body mass)1,with 30 s of recovery between sprints. During the 30 s recoveries subjects were instructed not to pedal for the initial 20 s, but were allowed to pedal for the final 10 s of recovery leading into the ensuing sprint. Post-sprint task subjects performed 20 mm of light cycling at a self-selected resistance, between 40-100 W, before performing a familiarization trial of the exercise task. The familiarization trial consisted of three square wave bouts at 120% MAP for 60% of the TTE, at that same workload, with 5 mm of cycling at 50 W between bouts at any cadence, provided it was greater than 60 rpm. 3.4.4 Days 4 to 6: Testing Testing days (Day 4, 5, and 6) were comprised of a similar protocol to the familiarization trial, the only difference being the randomized recovery intensity (i.e. AR, PR, or CAR) over the three days. Subjects reported to the laboratory rested and hydrated, two- to three- hours post partum. After an identical warm-up to the previous data collection days, subjects were given a short period of time (3-5 mm) to allow them to recuperate, mentally prepare, as well as fit the mouth piece for respiratory data collection, before commencing the exercise trial. The exercise trial was similar to the familiarization trial with respect to duration and work intensity, with the recovery intensity varying in addition to the addition of a third recovery period of 14 mm that separated the exercise trial from the sprint task (discussed below). The 5 mm recoveries were as follows on the respective test days: • PR — subjects remained seated and stationary on the bicycle ergometer, but were permitted to cycle against negligible resistance for five to ten revolutions to prevent blood pooling and discomfort every minute. • AR — subjects cycled at an intensity equivalent to the 50% of the difference between the IAT and the IVT (AT) below the IVT (IVT_50%A ). That is, the corresponding workloads 35 for the IAT and IVT (WIAT and WIVT, respectively) were used to calculate the workload at IVT_5o%zT (W IvT-5o%AT) from the following equation: W IVT-5O%T = WIVT - Y2(WIAT - WJvT) The use of IVT_50%AT to determine recovery intensity has been previously shown to be the optimal intensity for L& clearance in soccer players and triathletes during treadmill running exercise when compared against three other single intensity recoveries (Baldari et al., 2004; Baldari et al., 2005). Thus, it was selected as the intensity quantification for this investigation. Evidence exists that recovery intensities for L& clearance between cycling and running exercise may be different (Belcastro & Bonen, 1975; Bonen & Belcastro, 1976; Davies et al., 1970; Hermansen & Stensvold, 1972). However, the IVT_ 50%AT intensity relative to VO2mwas 39.3 ± 6.8% and 51.1 ± 4.9% for the soccer players and triathletes, respectively, which fall within the range of 30-70% VO2m previously reported to be optimal for L& clearance. Additionally, an investigation by Boileau et al. (1983), showed that recovery intensity for L& clearance did not significantly differ between cycling and running exercise, and therefore the IVT_50%AT intensity was deemed appropriate for this investigation. CAR — subjects cycled at an intensity equivalent to the TAT for 2 mm and then the IVT_ so%AT for the remaining 3 mm. Blood L& measures were taken immediately after the first two SE bouts and just before the second and third exercise bouts (see Figure 3). Following the third exercise bout the subjects recovered for 14 mm before performing the sprint task. This recovery was performed at the same intensity(s) as that performed between the SE bouts. During the PR trial subjects were allowed to dismount the bicycle, but were required to remain seated for the initial 8 mm of the recovery period. During the eighth minute subjects remounted the bicycle so that for the final 5 mm of recovery they could ride at their IVT_50%AT. This was done so that subjects did not begin the sprints cold. Additionally this effectively started subjects with similar cardiac outputs for all recoveries as the ended as the same exercise intensity. For the AR trial subjects cycled for the full recovery duration at their IVT_50%AT. In the case of the CAR trial the subjects cycled at their IAT intensity for 5 mm and at their IVT_50%AT for the remaining 9 mm. In all trials post-recovery 2 mm were allotted to allow subjects to transition to the Monark (Ergomedic 874E, Monark Exercise AB, Sweden) bicycle. Therefore, in actuality, subjects underwent 14 mm of the respect AR intensity, with an additional 2 mm of 36 recovery to allow the bike transition. Approximately fifteen minutes of recovery was selected because it is the typical intermission period for most intermittent sports (e.g. hockey, basketball, soccer). During the recovery after the third exercise bout blood L& measures were made at the 3rd 6th, 9th and 14th mm. The final procedure on the testing days was the sprint task. Subjects performed five 10 s sprints against a resistance of 0.09 kg.(kg body mass)’, with an identical protocol and instructions to the familiarization sprint task. Three minutes post-sprint task a final blood L& measure was taken. After the sprints subjects were recommended to perform and additional recovery to help return the body to homeostasis and relieve any discomfort associated with the intense sprint tasks. However, this was not mandatory. Heart rate was recorded throughout the testing procedure, during both the exercise bouts and recovery intervals, as well as during the sprint task. Samples were taken every 5 s during exercise, and sample points that temporally corresponded to the closest L& sample time were used for analysis (i.e. HR1-9). In addition, a HR measure was made at the start of first SE bout and used as a baseline value (HRB). During the sprint task the maximum HR achieved after each individual sprint was selected for analysis. Ventilatory parameters were recorded throughout the exercise bout and recovery periods, but not for the sprint task. These measurements were sampled breath-by-breath and averaged over 20 s. Again, the 20 s average that corresponded temporally closest to the La sample time was selected for analysis (i.e. V021-8, since ventilatory parameters were not measured after the 14 mm recovery period). Similar to the HR measures an additional baseline measure was taken (VO2B). Figure 3 visual depicts the temporal layout of the HR and V02 sample times. Figure 3. Schematic Representation of the Testing Day Protocol B Li L2 L3 L4 L5 L6 L7 L8 L9 Warm- SE1 5 mm SE2 5 mm SE3 31 6th 9th 14th Sprint 3 mm up Recovery Recovery 14 mm Recovery Task Post Sprint B, baseline HR and V02 sample time; L1-9, L& (HR and V02) sample times from post-SE 1 to 3 mm post-sprint task (no V02 measure was made for the L9 time point) 37 3.5 Data Analysis The dependent variables that were measured in this investigation were: Blood L& Concentration (mmolL1),PP Output (W), MP Output (W), Fl (%), and TW (J). Additionally HR (bpm) and V02 (mLkg’min’) were recorded. Blood La comparisons were made using a group (3) by time (9) repeated-measures analysis of variance (ANOVA). Similarly, a group (3) by time (5) repeated-measures ANOVA was used to compare parameters from the sprint tests (i.e. PP and MP outputs and Fl). A one-way repeated-measures ANOVA was used to compare TW in the three treatment groups. Heart rate and V02 during the exercise and recovery bouts were assessed using group (3) by time (10 and 9, respectively) ANOVA’s. While HR during the sprint task was assessed using a group (3) by time (5) repeated-measures ANOVA. Statistical significance was set a priori at a level of p 0.05. When the omnibus F-test showed a significant interaction effect dependent T-tests were used post hoc to determine where differences were, since the repeated-measures within-subjects design for ANOVA in SPSS version 15.0 (SPSS Inc., USA) does not produce post hoc tests. Significant omnibus F-tests for the main effects were followed up with pairwise comparisons to determine where differences occurred. The Bonferroni adjustment was made to account for multiple comparisons. All values are reported as means ± standard deviations (5 ± SD). Statistical analyses of ANOVA were performed using SPSS version 15.0 for Windows (SPSS Inc., USA) and T-tests were performed with Microsoft Excel Version 5.1.2600 (Microsoft Corporation, USA). 38 CHAPTER IV: RESULTS 4.1 Subject Characteristics Twelve trained male cyclists volunteered to participate in this investigation. All subjects completed all of the testing protocols. However, four L& values were not obtained due to complications measuring the samples in two subjects. In Subject 2 the PR L3, AR Li, and CAR L2 values are missing. In Subject lithe PR L4 value is missing. Therefore, in the initial analysis these subjects were excluded from the blood L& ANOVA, since the default setting for SPSS (SPSS Inc., USA) is to do a listwise deletion for missing data. Since statistical significance was achieved, despite the smaller sample size, no measures were taken to replace the missing data. Additionally, HR data from Subject 10 on the AR day and Subject 12 on all days was not able to be uploaded due to technological difficulties. Furthermore, HR could not be electronically monitored in Subject 11 due to a morphological anonmaly. Therefore, subjects 10-12 were excluded from the HR analysis. Subject’s characteristics and workload and recovery characteristics are presented in Table 1 and 2, respectively. 4.2 Blood Lactate On each of the Testing Days (i.e. Day 4, 5 and 6) nine blood L& samples were taken. Sample times were: post-exercise bout 1 (Li), pre-exercise bout 2 (L2), post-exercise bout 2 (L3), pre-exercise bout 3 (L4), 3 mm post-exercise bout 3 (L5), 6 mm post-exercise bout 3 (L6), 9 mm post-exercise bout 3 (L7), 14 mm post-exercise bout 3 (L8), and 3 mm post-sprint (L9, see Figure 3 for visual description). Subject 2 and ii both had missing blood L& values and are therefore excluded from this analysis. The ANOVA revealed a significant interaction (group x time) effect for the blood L& data (p0.001). Thus, the effects of the recovery group on blood L& differed depending on the sampling time. Larger decreases in blood L& were observed in the AR and CAR protocols compared to the PR protocol. This effect was more evident in the recovery portion of the protocol (i.e. L5-8, see Figure 4). Mean differences from L5 to L8 were 4.2, 6.7, and 6.5 mmolU’ for PR, AR and CAR, respectively. Targeted dependent T-tests were performed for the recovery and post-sprint blood L& values (i.e. L5-9). T-test showed that during PR blood La values were significantly greater from the L6-8 sample times when compared to AR (pO.O5). Furthermore, PR blood La values were significantly greater than CAR values for all recovery samples (i.e. L5-8, pO.O5). However, the final blood L& values post-sprint (L9) in both AR and CAR were not statistically different from PR (p>0.05, see Table 3). Differences between the two active recoveries were not significantly different, despite the lower blood L& values throughout 39 the CAR from L5-8 (p>O.05). Post-sprint (L9) blood L& was highest in the PR, followed by AR and then CAR (see Table 3). However, none of these differences were statistically significant (p>O.05). Table 1. Subject Characteristics Subject Age Height Weight So5S VO2m PP MP (yrs) (cm) (kg) (mm) (mLkg’min1 (Wkg1 (Wkg1) 1 29.5 176.0 72.7 29.5 71.2 13.7 8.5 2 21.8 181.5 68.1 31.5 76.5 15.5 11.2 3 28.8 185.6 77.9 54.4 58.9 13.6 10.2 4 29.6 182.2 69.8 25.0 75.2 15.2 9.1 5 27.0 172.2 75.1 41.2 58.1 14.5 9.6 6 27.1 188.0 87.9 40.1 57.2 16.5 10.6 7 34.0 181.0 65.3 34.1 65.4 16.8 9.7 8 30.1 169.6 68.5 45.8 59.5 14.3 9.9 9 23.8 186.4 72.0 28.5 61.1 13.7 8.5 10 21.2 181.0 71.6 39.9 61.5 14.1 9.3 11 25.0 180.3 69.8 43.5 65.7 15.1 9.4 12 30.9 193.5 89.4 34.3 63.8 15.6 9.0 X 27.4 181.4 74.0 37.3 64.5 14.9 9.6 SD 3.9 6.7 7.6 8.4 6.6 1.1 0.8 So5S, sum of five skinfolds; PP, peak power (on WAnT, Wingate Anaerobic Test); MP, mean power (on WAnT); X, mean value; SD, standard deviation The main effect for (treatment) group was also significant (pO.OO1). Pairwise comparisons demonstrated that both AR and CAR groups resulted in significantly lower blood L& values compared to the PR group (p0.05 and p0.00i, respectively for the former and latter). 40 Table 2. Subject Workload and Recovery Characteristics 50%AT, 50% of the difference between the TAT and IVT (individual ventilatory threshold) below TVT The sphericity assumption was violated for the main effect of (sample) time and therefore the Greenhouse-Geisser correction was applied. The main effect of time on blood L& values was significant (p0.001). This demonstrated that the work and recovery protocol intensities were sufficient to induce statistically significant increases in blood L& as well as being appropriate for clearance. Pairwise comparisons revealed where the difference existed between sample times and Table 4 presents the differences. Blood La levels increased throughout the exercise protocol and peaked at L5 before decreasing during the recovery portion of the exercise protocol. Lactate levels where again elevated post-sprint task (L9) to an overall maximum across all nine sample times. Subject 120% MAP 60% TTE IAT IVL50%ATVO2IAT VO2I_5o%T (W) (s) (W) (% max) (W) (% max) 1 540 76.5 270 69.6 180 47.7 2 576 72.4 330 82.3 240 61.7 3 396 111.2 270 76.3 180 54.4 4 504 52.4 240 73.9 195 52.0 5 468 91.4 210 65.1 120 44.7 6 540 89.4 270 66.1 180 55.5 7 432 76.8 240 75.9 105 43.1 8 396 75.4 240 75.5 150 51.3 9 468 81.9 210 66.7 120 46.9 10 432 74.5 240 67.6 150 51.3 11 504 72.9 270 75.9 135 45.0 12 612 127.6 390 83.1 165 49.0 X 489.0 83.5 265.0 73.2 160.0 50.2 SD 69.4 19.7 50.9 6.1 38.0 5.3 MAP, maximum aerobic power; TTE, time to exhaustion; TAT, individual anaerobic threshold; IVT_ 41 Table 3. Mean Blood Lactate Values Blood Lactate Sample (mmolL’) PR AR CAR Li 5.6+ 1.6 5.6± 1.8 5.4±2.2 L2 8.9±2.0 7.8± 1.3 8.2± 1.3 L3 12.1 ± 2.1 9.2 ± 2.7 10.6 ± 2.4 L4 12.0±3.0 9.9±2.2 10.0±2.1 L5 14.4± 1.8 13.6±2.6 12.4±2.2* L6 13.8 ± 1.8 10.7 ± 2.7* 10.3 ± 3.0* L7 13.4±3.6 9.6±3.2* 9.2±3.1* L8 10.3 ± 3.1 6.9 ± 2.5* 5.9 ± 2.6* L9 14.6± 1.6 13.8±2.3 13.6±2.4 X 11.7±3.6 9.’7±3.5 9.5±3.5 n = 10, Subjects 2 and 11 are excluded; T-tests were only performed for recovery L& (i.e. L5-9); L1-9, blood L& sample times (see Figure 3) * T-tests significantly different from PR value (pO.O5); Pairwise comparison significantly different from PR (pO.O5); Pairwise comparison significantly different from PR (pO.OO1) Table 4. Mean Blood Lactate Differences for Sample Time Sample Time Mean Blood Lactate (mmoFL’) Significantly Different From (pO.O5) Li 5.6± 1.8 L3-7,L9 L2 8.3 ± 1.6 L3-6, L9 L3 10.6±2.6 L1-2 L4 10.6 ± 2.6 L1-2, L5, L8-9 L5 13.5 ± 2.3 Li-2, L4-8 L6 11.6±2.9 Li-2,L5,L8-9 L7 10.7 ± 3.7 Li, L5, L8-9 L8 7.7 ± 3.3 L4-9 L9 14.0±2.1 Li-2,L4,L6-8 n = 10, Subjects 2 and 11 are excluded 42 Figure 4. Mean Blood Lactate Values ---PR AR —&-CAR n = 10, Subjects 2 and 11 are excluded * Significantly different from PR value (pO.O5) Comparing the ten subjects included in the blood L& statistical analysis, eight reached their overall (i.e. among all three testing days) maximum blood L& value on the PR day. The other three overall maximum values occurred on the AR day, as one subject shared the same maximum value on both the AR and PR days. Of the two subjects excluded from the analysis, one was the only subject to have their overall maximum blood L& on the CAR day. The other excluded subject followed the group trend, having his highest overall value on the PR day. Daily maximum blood L& values typically occurred at the third minute of recovery (L5), with 5, 8, and 6 of the subjects having their daily maximum L& at L5 for PR, AR, and CAR days respectively. In general, the overall highest post-sprint blood L& (L9) predominantly occurred on the PR test day, with 7 of 10 subjects peaking on this day. One subject peaked post-sprint on the AR day and three peaked on the CAR day (with one subject having the same L9 La value on both the PR and CAR days). Of the excluded subjects, Subject 2 had the same maximum L9 value on the PR and CAR day, while Subject 11 had their overall L9 maximum on the PR day. 0 E E C.) -J 0 0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 LI L2 L3 L4 L5 L6 L7 L8 L9 Sample Time 43 The Li -L4 scores were excluded from the determination of minimum L& values as these samples were taken during the fatiguing exercise bout and the interest for lowest L& values was within the 14 mm recovery period. The majority of subjects had their lowest overall La values on the CAR day (7 of 10). The other three subjects had their lows on the AR day. Subject 2 and 11 had their overall minimum L& values on the PR and CAR day, respectively. Overall minimum L& values were observed at L8 in all subjects (n = 10). Subject 2 (excluded) was the only participant to have an overall minimum L& value at a sample time other than L8 (i.e. L7). On the CAR day all subjects (n = 10) had their daily minimum L& value at the L8 sample. Nine subjects had their daily low at L8 and one at L7 on the PR day. Eight subject’s daily lows occurred at L8 and two at L7 on the AR day. Subject 11 had his daily L& minimum at L8 on all days, while Subject 2 had his daily lows at L7 on the PR and AR days and L8 for the CAR day. Of the subjects included in the L& ANOVA, none experienced their overall L& minimum on the PR day, while the majority of maxima (i.e. 8) were achieved on this day. However, Subject 2 (excluded from ANOVA) was an exception to this, as he experienced his overall La minimum of the PR day. Conversely, no subjects experienced an overall La maximum on the CAR day, while the majority of L& minima occurred on this day. Again, Subject 2 was the exception to this, as he was the sole participant to achieve his overall L& maximum on the CAR day. Subject 9 was the only subject to experience their overall maximum and minimum La values on the same day. This occurred on the AR day. Of the eight subjects that achieved their overall maximum L& value on the PR day, six also had their overall final (i.e. 3 mm post-sprint) maximum La value too. Subject 7 was the only participant to achieve his overall final maximum on the AR day, and interestingly, also had his overall minimum on this day just prior to the sprint task. On the CAR day, three subjects achieved their overall final maximum L& value after previously having their overall minimum L& value at L8. Additionally, Subject 2 achieved his overall final maximum L& score on the CAR day (which was the same on the AR day), but did not experience an overall minimum at L8. Table 5 summarizes the daily and overall maximum and minimum L& values for all subjects. 4.3 Sprint Task Performance Subjects performed five 10 s sprints, each separated by 30 s of recovery. Each individual sprint was assessed for PP, MP and Fl. In addition, the performance across the sprint task was assessed by analyzing TW across all five sprints. All subjects were able to complete all five sprints on each test day. Group means for all sprint task performance variables are presented in Table 6. 44 Table 5. Group Means for Peak Power, Mean Power, Fatigue Index and Total Work PP (W) MP (W) FT (%) TW (J) PR 1017.7± 134.4 782.2± 111.7 37.0±9.2 39108.3 ±4852.9 AR 1013.6± 145.6 800.1 ± 114.5* 34.2± 11.3 40003.3±5110.2* CAR 1021.8± 134.4 786.7± 118.0 37.4± 10.6 39335.8± 5022.6 * Significantly different from PR value (pO.O5) 4.3.1 Peak Power The interaction (group x time) effect for PP was not found to be significant (p>0.05). Therefore, the effect of time on significantly decreasing PP did not change as a result of the group level (i.e. recovery mode). Peak power decreased in all three groups as the number of sprints performed increased. Table 6 shows the mean PP outputs and Figure 5 graphically depicts them. Table 6. Mean Peak Power Outputs Peak Power (W) PR AR CAR 1 1067.2±137.1 1054.8±160.1 1072.5±158.1 2 1046.1± 137.6 1040.7± 130.0 1060.2± 141.5 3 998.8± 121.1 1000.1± 152.1 1011.0± 126.2 4 992.1± 145.2 989.2± 151.1 1001.7± 143.9 5 984.2± 132.4 983.3±143.9 963.6± 113.5 The group main effect for PP was not significant (p>0.05). Subjects achieved similar PP outputs regardless of the recovery intensity. Nonetheless, the group mean PP output was greatest in the CAR group, followed by PR, then AR (see Table 5). The PP output was 4.1 W greater than the next closest group mean PP output for each group difference. 45 Figure 5. Mean Peak Power Outputs 0 I Table 7. Sprint Number Means for Peak Power, Mean Power, and Fatigue Index Sprint Number PP (W) MP (W) Fl (%) 1 1064.8± 147.9*35 864.5± 108.7*25 29.8± ii.i*45 2 1049.0 ± 132.7*35 825.7 ± 99 7*’ 34.8 ± 10.2*45 3 1003.3 ± 130.1*12 778.6 ± 102.1*1.2,45 36.4 ± 93*45 4 994.3 ± 135.4*1.2 748.4 ± 995*1.3 39.6 ± 94*3 5 977.0± 127.1*1.2 730.9 ± 101.8*13 40.5 ± 8.8*13 * Significantly different from listed sprints (pO.O5) 1080 1060 1040 1020 1000 •PR I AR 980 DCAR 960 940 920 900 A significant main effect of (sprint number) time on PP output was found (p0.001). Therefore, subjects were unable to maintain PP across the sprint task. Pairwise comparisons showed where the differences were (see Table 7). Peak power output was significantly greater on the first two sprints compared to the last three (p0.05). No difference in PP was observed between Sprint 1 and 2 or between the last three sprints (p>0.05). 1 2 3 4 5 Sprint Number 46 In general the overall maximum PP outputs were achieved in either the first or second sprint. The exception was Subject 4, who shared the same overall maximum score for both his second and third sprint. The majority of overall maximum PP outputs were attained on the CAR day (8 of 12). Additionally, three overall maxima were attained on the AR day and one on the PR day. As a general rule, most subjects achieved their highest daily PP output in either the first or second sprint. Three subjects achieved their daily maximum PP in the third sprint with one on the AR day and two on the CAR day (one subject shared the same score in the second and third sprint on the CAR day) and one in the fifth sprint on the PR day. The other daily maximum values were distributed between the first and second sprint (23 and 10, respectively). Seven daily maxima occurred in the first sprint on the PR and CAR days respectively, and nine in the first sprint on the AR day. Four daily maxima occurred in the second sprint on the PR and CAR days respectively, and two happened in the second sprint on the AR day. Overall minimum PP outputs tended to occur on the PR and CAR day, with five on each respectively. The other two occurred on the AR day. In which sprint the overall minima occurred was more varied than the maxima and displayed no distinct trend. Seven arose in the fifth sprint, with one of these being shared with the second sprint score in Subject 3. Of the seven in the fifth sprint, three occurred on each of the PR and CAR day, respectively, and one on the AR day. The rest of the overall minimum PP outputs were spread over all the other sprints, with one in the first sprint, two in the second (one being a shared score), two in the third, and one in the fourth. Daily minimum PP outputs tended to take place in the later sprints, but were evident throughout the sprint task. On the PR day six daily minima (two were shared) occurred in the fifth sprint, four in the fourth (one shared), three in the third, and one (shared) in the second sprint. Five occurred in the fifth sprint on the AR day, five in the fourth, and one in both the first and third sprints. On the CAR day, seven occurred in the fifth, one in the first, second and fourth sprint, and two in the third. Table 17 summarizes the daily and overall maximum and minimum PP outputs. 4.3.2 Mean Power The interaction (group x time) effect for MP was not statistically significant (p>O.O5). Thus, similar to the effect of time on PP, MP decreased in all three groups as the number of sprints performed increased. Table 8 shows the mean MP outputs, which are also illustrated in Figure 6. 47 Table 8. Mean Mean Power Outputs Mean Power (W) PR AR CAR 1 861.3± 102.9 872.3±114.0 859.7± 118.0 2 820.3 ± 97.2 834.3 ± 106.1 822.7 ± 103.9 3 767.9±100.6 791.1±109.5 776.9±103.8 4 738.3 ± 106.7 757.5 ± 99.2 749.5 ± 100.2 5 723.1±104.1 744.9±109.1 724.8±99.6 The main effect for group MP was statistically significant (pO.O5). The group mean MP was highest on the AR day, followed by CAR, then PR (see Table 5). Pairwise comparisons showed that the significant differences were between the PR and AR MP outputs (pO.O5). Neither of the other two pairwise comparisons (AR vs. CAR and PR vs. CAR) were significant (p>O.OS). The sphericity assumption for the main effect of (sprint number) time was violated and therefore the Greenhouse-Geisser correction was applied. The main effect of time on MP outputs was significant (pO.OO 1). Therefore, in addition to subjects not being able to maintain PP across the sprint task, they were also unable to maintain MP. Pairwise comparisons showed where the differences were (see Table 7). All MP outputs, except on the fourth and fifth sprint, were significantly different from all other sprints. The fourth and fifth sprint did not differ significantly from each other (p>0.05), but were significantly less than the initial three sprints (p0.05). Mean power outputs decreased as the number of sprints completed increased. Each subsequent sprint was significantly less than the previous sprint (p0.05), except for the final two sprints. The majority of overall maximum MP outputs were accomplished on the AR day. Seven overall maximum MP outputs were achieved on the AR day, three on the PR day, and two on the CAR day. All but one of these occurred in the first sprint, with the exception occurring in the second sprint. All daily maximum MP outputs, except for one, occurred in either the first or second sprint, with the vast majority (32 of 35) occurring in the first sprint. In the one exception the maximum transpired in the third sprint. On the PR day twelve of the daily maxima happened in the first sprint and one occurred in the second sprint. Subject 11 reproduced the same daily maximum value for the first and second sprint on the PR day. Ten daily maxima were achieved 48 in the first sprint on the AR day, one in the second, and one in the third. On the CAR day ten daily maxima were achieved in the first sprint and two in the second. •PR E: AR DCAR On the PR day five of the overall minimum MP outputs occurred. Of these, two occurred in the fourth sprint and the other three in the fifth sprint. Only one overall minimum occurred on the AR day. It was obtained in the fourth sprint. Six overall minimum MP outputs occurred on the CAR day and they were all in the fifth sprint. On the PR day, three daily minimum MP outputs occurred in the fourth, and nine in the fifth sprint. Subject 1 shared the same daily minimum MP score on the first and fifth sprint for the AR day. Three subjects had their daily minimum in the fourth sprint and nine had it in the fifth. All but one subject had their daily minimum in the fifth sprint on the CAR day. The one anomaly experienced his daily minimum in the fourth sprint. Subject 1 and 4 were the only participants to experience both their overall maximum and minimum MP output on the same day (i.e. on the CAR and PR days respectively). On the AR day there were five subjects that experienced their overall maximum MP in addition to also 49 Figure 6. Mean Mean Power Outputs 900 850 800 4- 750 700 650 600 1 2 3 4 5 Sprint Number achieving their overall final (i.e. fifth sprint) maximum MP. In total there were eight subjects that achieved their overall final maximum MP output on the AR day. Table 18 summarizes the daily and overall maximum and minimum MP outputs. 4.3.3 Fatigue Index The interaction (group x time) effect for FT was not statistically significant (p>0.05). Fatigue index increased in all three groups as the number of sprints increased. The Fl for all three treatment groups across the sprint times are shown in Table 9 and graphically depicted in Figure 7. Table 9. Mean Fatigue Indexes Fatigue Index (%) PR AR CAR 1 30.3 ± 9.6 28.4 ± 12.5 30.7 ± 12.0 2 35.0± 10.4 32.7± 11.0 36.6± 12.0 3 37.0 ± 7.5 33.7 ± 10.3 38.5 ± 10.0 4 40.9± 7.4 37.3±11.1 40.7± 9.7 5 41.8±7.1 38.9±9.9 40.7±9.6 The main effect for group for FT was not found to be significant (p>0.05). The group mean for Fl was greatest in the CAR protocol, followed by the PR then AR protocols. However, the differences were very minute (see Table 5). The sphericity assumption for the main effect for (sprint number) time was violated and consequently the Greenhouse-Geisser correction was applied. The main effect for time was found to be significant (p0.00 1). Fatigue index increased across the sprint task. Pairwise comparisons demonstrated where the differences occurred (see Table 7). The Fl for the first three sprints were significantly less than the last two sprints (p0.05). Conversely, there were no significant differences amongst the first three sprints or amongst the final two sprints. Thus, over the course of the sprint task, subjects developed larger differences between their maximum and minimum power outputs within a sprint. 50 Figure 7. Mean Fatigue Indexes 45.0 40.0 35.0 30.0 25.0 •PR AR 20.0 DCAR 15.0 10.0 5.0 0.0 Sprint Number The trends for maximum and minimum data values for Fl were less discernible than the other performance variables. The overall maximum Fl were relatively evenly distributed among the three protocols. Four maxima occurred on the PR day, three on the AR day, and five on the CAR day. Of the four subjects who experienced an overall maximum Fl on the PR day one was in the fourth sprint and the other three were in the fifth sprint. On the AR day all three occurred in the fifth sprint. Lastly, on the CAR day one occurred in the second sprint, three in the fourth, and one in the fifth. Daily maximum FT tended to occur in the later sprints, but several subjects experienced their maxima early on in the sprint task. During the PR trial, two subjects experienced their daily maximum Fl in the second sprint, six in the fourth, and four in the fifth. On the AR day, two experienced their maxima in the first sprint, three in the fourth, and seven in the fifth. During the CAR protocol, one subject had their daily maximum in the first sprint, two in the second, one in the third, and four in each of the fourth and fifth sprint. Four subjects experienced overall minimum FT on the PR day. Five subjects had their overall minimum FT on the AR day and four had theirs on the CAR day. Subject 3 had the same overall minimum FT on both the AR and CAR days. On the PR day, two of the overall minima 1 2 3 4 5 51 that occurred were in the first sprint and one occurred in each of the second and third sprints respectively. Three overall minimum FT occurred in the first sprint on the AR day, and one occurred in each of the second and third sprints respectively. On the CAR day the overall minima were divided between the first and second sprint, with two occurring in each. The majority of daily minimum Fl occurred in the first sprint, followed next by the second sprint. On the PR day, seven daily minima happened in the first sprint. In the second and third sprint there were three daily minima that occurred in each (Subject 2 had the same Fl in both the second and third sprint). During the AR protocol, eight subjects had their daily minimum FT score in the first sprint. There were two subjects who had their daily minimum FT in the second sprint, one in the third, and one in the fifth. On the CAR day, nine subjects had their daily minimum FT in the first sprint and three had it in the second sprint. Overall initial and final maximum FT values were relatively evenly dispersed among the three testing days. Table 19 summarizes the daily and overall Fl scores and in which sprint they occurred. 4.3.4 Total Work The one-way repeated-measures ANOVA performed to compare the treatment effects on TW revealed the same results as the ANOVA MP group effect. This is because the TW score is simply an aggregate score of the MP across the sprint trial. Therefore, there were significant differences between the treatments with respect to TW (pO.O5). Again, the pairwise comparison between AR and PR was found to be significant (pO.05). The mean values for TW for each group are shown in Table 5. Figure 8 graphically presents the data. The AR group mean TW was greatest, followed by the CAR, then the PR. No significant differences existed among AR and CAR, and PR and CAR pairwise comparisons (p>O.05). On an individual basis, seven subjects had their highest TW output on the AR day. Four subjects had their overall maximum TW output on the CAR day, and only one subject had their highest TW output on the PR day. Conversely, six subjects had their overall minimum TW output on the PR day, five did on the CAR day, and only one did on the AR day. Table 14 summarizes the maximum and minimum TW outputs for each subject. 52 Figure 8. Total Work 40200 40000 39800 39600 IPR 39400 EJAR 0 DCAR 39200 39000 38800 38600 * Significantly different from PR value (pO.O5) 4.4 Heart Rate Heart rate was recorded during both the exercise and recovery bouts, as well as during the sprint task. During the exercise and recovery bouts HR measures were matched to the corresponding L& sample points for analysis, with the addition of a baseline measure (HRB) prior to SE1. However, due to technological difficulties several HR files were not able to be uploaded for analysis, while in another subject a HR reading was not obtainable due to the subject’s (Subject 11) chest morphology. Therefore the exercise and recovery HR ANOVA was comprised of nine subjects. The maximum HR achieved after each individual sprint was selected for analysis of HR during the sprint task. In addition to the aforementioned technical difficulties in uploading HR data, two other HR data sets were excluded from analysis due to erratic values during the sprint task. Thus the sprint task HR (SHR) data set was comprised of seven subjects. 53 4.4.1 Exercise and Recovery Heart Rate There was a significant interaction (group x time) effect for the HR results (pO.OOl). Heart rate responses varied differently over the three recovery intensities for the different sample times. Dependent T-tests revealed that HR was greater in both the AR and CAR compared to the PR from the HR2 sample time to HR7 (p0.05). Heart rate was greater in the AR trial compared to the CAR trial at HR4, but HR was then greater at HR5-6 in the CAR trial compared to the AR trial (pO.05). Mean HR values are shown in Table 10. Table 10. Mean Exercise and Recovery Heart Rate Values Heart Rate (bpm) PR AR CAR HRB 115.0± 11.0 117.2± 12.9 115.0± 8.5 HR1 173.0 ± 8.8 172.2 ± 10.0 171.9 ± 9.9 HR2 119.2± 10.1 139.1 ± 10.5* 141.6± 8.3* HR3 175.7±9.8 181.1±10.1* 180.6±11.4* HR4 120.9 ± 10.1 148.7 ± 8.4* 146.2 ± HR5 109.2 ± 9.7 142.4 ± 93* 167.3 ± HR6 102.2 ± 8.1 139.4 ± 8.4* 156.3 ± l2.l’ HR7 108.2±8.9 142.1±9.4* 144.2±11.1* HR8 139.2±8.9 143.7± 11.3 144.4± 12.9 HR9 110.3 ± 9.0 112.3 ± 11.4 112.1 ± 6.3 5 127.3 ± 27.0 143.9 ± 22.3 148.0 ± 23.5 n 9, Subjects 10-12 were excluded; T-test were performed for all HR (i.e. HRB-9); HRB, HR-Baseline * Significantly different from PR value (pO.O5); significantly different from AR value (pO.05) The sphericity assumption for the group main effect was violated and therefore the Greenhouse-Geisser correction was applied. The main effect for group was significant (p0.001). Pairwise comparisons showed that group mean HR was significantly greater in the AR and CAR versus PR (p0.001), and in CAR versus AR (p0.05). Figure 9 graphically presents the mean HR values. 54 Figure 9. Mean Exercise and Recovery Heart Rate Values 60 -4-PR AR —&-CAR n = 9, Subjects 10-12 are excluded * Significantly different from PR value (pO.O5); significantly different from AR value (pO.O5) The main effect for time was found to be significant (p0.001). Pairwise comparisons revealed where the differences between sample times existed. Table 11 shows the sample time means and describes the differences and Table 21 presents individual daily and overall maximum and minimum HR values for the nine subjects evaluated. 200 180 160 140 80 40 20 0 HRB HRI HR2 HR3 HR4 HR5 HR6 HR7 HR8 HR9 Sample Time 55 HRB 116.0± 10.6 HR1 172.4 ± 9.2 HR2 133.1 ± 13.8 HR3 179.1±10.3 HR4 138.6 ± 15.6 HR5 139.9±26.4 HR6 132.8 ±24.9 HR7 131.6 ± 19.3 HR8 142.6± 10.8 HR9 112.0±8.8 n 9, Subjects 10-12 were excluded 4.4.2 Sprint Task Heart Rate The interaction (group x time) effect for SHR reached statistical significance (pO.O5). The increase in SHR across the sprint task differed depending on the treatment group. Dependent T-tests revealed that SHR was greater in both the CAR compared to the PR after the third and fourth sprint (pO.O5). Mean SHR values are shown in Table 12. Table 12. Mean Maximum Sprint Task Heart Rate Heart Rate (bpm) PR AR CAR SHR1 167.3 ± 6.6 168.7 ± 5.8 168.0 ± 6.8 SHR2 173.1 ± 6.5 175.9 ± 6.5 176.4 ± 6.8 SHR3 175.0 ± 6.4 177.3 ± 7.7 178.1 ± 59* SHR4 175.6 ± 5.1 178.6 ± 7.4 180.1 ± 74* SHR5 176.1 ± 5.2 179.3 ± 7.3 179.4 ± 7.1 173.4 ± 6.5 175.9 ± 7.6 176.4 ± 7.8 Table 11. Mean Exercise and Recovery Heart Rate Differences for Sample Time Sample Time Heart Rate (bpm) Significantly Different From (p0.05) HR1-8 HRB, HR2-9 HRB, HR1, HR3, HR9 HRB-2, HR4-9 HRB-1, HR3, HR7, HR9 HRB-1, HR3, HR6-7, HR9 HRB-1, HR3, HR5, HR8-9 HRB-1, HR3-5, HR8-9 HRB-1, HR3, HR6-7, HR9 HR1-8 n = 7, Subjects 7 and 9-12 are excluded * Significantly different from PR value (pO.O5) 56 The group main effect for SHR was not significant (p>O.05). The group mean for SHR was greatest in the CAR protocol, followed by the AR then PR protocols. However, the differences were very small. The main effect for time was found to be significant (pO.OOl). Maximum HR values post-sprint increased across the sprint task. Pairwise comparisons demonstrated where the differences occurred (see Table 13). Mean SHR steadily climbed over the course of the sprint task. The SHR for the first sprint was significantly less than all other values (p0.05). After the second sprint maximum HR was significantly greater than the first sprint, but less than the last two sprints (p0.05). Heart rate after the third sprint was only significantly different form the SHR1 value (p0.05). Heart rate after the last two sprints was significantly greater than after the first three sprints (pO.O5). The SHR for all three groups across the sprint times are graphically depicted in Figure 10. Table 22 presents individual daily and overall maximum and minimum SHR values for the seven subjects evaluated Table 13. Mean Sprint Task Heart Rate Differences for Sample Time Sprint Number Heart Rate (bpm) Significantly Different From (pO.05) SHR1 168.0±6.1 SHR2-5 SHR2 175.1 ± 6.4 SHR1, SHR4-5 SHR3 176.8 ± 6.5 SHR1 SHR4 178.1 ± 6.7 SHR1-2 SHR5 178.3 ± 6.4 SHR1-2 n = 7, Subjects 7 and 9-12 are excluded 57 Figure 10. Mean Sprint Task Heart Rate 185 180 0. . 175 t 170 165 160 n 7, Subjects 7 and 9-12 are excluded 4.5 Volume of Oxygen Consumed The interaction (group x time) effect for the V02 scores was significant (pO.OO 1). Responses of V02 to the exercise and recovery protocol varied differently over the three treatments across the sample times. Dependent T-tests revealed that V02 in both the AR and CAR protocols was significantly greater than the PR protocol at the V02 sample, and from V024-7 (pO.0O1). Volume of 02 consumed during CAR was greater than AR at the V025 and V026 sample times (p0.O0l). Mean V02 was lower during the PR protocol from VO2B to V027, except at the V021 and V023 samples. At the V028 sample the mean V02 during PR was actually significantly higher than during AR (pO.O5), and greater than during the CAR (p>O.O5) Mean V02 was higher in the CAR protocol than in the AR protocol at all measures except at baseline (VO2B) and pre-exercise bout 3 (V024). Mean V02 values are shown in Table 14. 1 2 3 4 5 Sprint Number 58 Table 14. Mean Volume of Oxygen Consumed Values Volume of Oxygen Consumed (mL.kg’.min’) PR AR CAR VO2B 13.6 ± 3.0 15.1 ± 4.1 14.8 ± 3.9 V021 56.5 ± 6.8 57.0 ± 7.1 57.8 ± 6.8 V02 14.3 ± 4.9 36.5 ± 6.1* 377 ± 8.0* V023 58.9±6.5 60.5±7.4 61.6±8.8 V024 13.6±4.0 38.6±6.1* 38.2±6.7* V025 10.6± 1.7 36.5 ±6.6* 5149*t V026 8.0± 1.6 34.3 ±6.4* 40082*t V027 11.1 ± 3.2 32.9 ± 6.6* 34.6 ± 6.8* V028 33.0±6.4 31.7±6.0* 31.9±7.3 5 24.4± 19.7 38.1 ± 14.2 40.9± 15.4 T-test were performed for all V02 (i.e. VO2B-8) * Significantly different from PR value (pO.OO 1); significantly different from AR value (pO.O5) The sphericity assumption for the main effect of group was violated and therefore the Greenhouse-Geisser correction was applied. The effect was then found to be significant (p0.00 1). Pairwise comparisons showed that all three groups were significantly different with respect to group mean V02 scores (p0.001). Volume of 02 consumed was greatest in the CAR protocol, followed by AR then PR. The sphericity assumption for the main effect for time was violated and consequently the Greenhouse-Geisser correction was applied. The V02 main effect for time was found to be significant (p0.001). Pairwise comparisons revealed where the differences between sample times existed. Table 15 shows the sample time means and describes the differences. Figure 11 graphically presents the mean V02 values. Table 23 presents individual daily and overall maximum and minimum V02 values for all subjects. 59 VO2B v021 V02 V023 V024 V025 V026 V027 V028 0) . -j E •0a) E 0 0 C.) a) 0) 0 II 0 a) E 0 > 14.5 ± 3.6 57.1 ± 6.7 29.5 ± 12.6 60.4 ± 7.5 30.1 ± 13.1 32.8 ± 18.2 27.5 ± 15.3 26.2 ± 12.2 32.2 ± 6.4 V021-8 VO2B,V02,V024-8 VO2B-1, V023,V025 VO2B-1, V024-8 VO2B-1, V023,V026-7 VO2B-4, V026-7 VO2B-1, V023-5, V028 VO2B-1, V023-5, V028 VO2B-1, V023,V026-7 Table 15. Mean Volume of Oxygen Consumed Differences for Sample Time Sample Time Mean Volume of Oxygen Consumed Significantly Different From (pO.O5) (mL.kg1.min’) Figure 11. Mean Volume of Oxygen Consumed Values 70.0 60.0 50.0 40.0 -4—PR AR 30.0 —&-CAR 20.0 10.0 0.0 VO2B V021 V022 V023 V024 V025 Sample Time * Significantly different from PR value (p0.O5); t significantly different from AR value (pO.O5) V026 V027 V028 60 CHAPTER V: DISCUSSION 5.1 Subject Characteristics The mean age (27.4 yrs) of the subjects tested in this investigation was similar to that of Stamford et a!. (1981) and Dodd et a!. (1984), but older than that of more recent investigations (Baldari et al., 2004; Baldari et al., 2005; Gmada et al., 2005). Furthermore, the subjects investigated displayed greater VO2max scores than those reported by Stamford et al. (1981), Dodd et al. (1984), and Gmada et al. (2005), but slightly lower than the populations investigated by Baldari et al. (2004; 2005). This was expected as the two early investigations did not use trained populations. Interestingly, the subjects had VO2max scores that were reasonably higher than the trained group examined by Gmada et al. (2005). This is attributed to the criteria utilized to determine training status in the respective investigations. However, the group did exhibit similar VO2max, threshold and MAP data to those reported in other groups of trained cyclists (Faria, Parker, & Faria, 2005; Tanaka et al., 1993). Compared to Gmada et al. (2005) workloads at 120% MAP were substantially higher (489.0 ± 69.4 vs. 310.0 ± 14.0 W (trained) and 280.0 ± 15.0 W (untrained) respectively). The 60% TTE duration was shorter than the trained group (83.5 + 19.7 vs. 102.0 ± 20.0 s), but longer than the untrained group (73.0 ± 21.0 s). These differences may be explained by differences in the parameters used to define training status, as well as differences in the protocol used to achieve VO2m. During the VO2m protocol in the study by Gmada et a!. (2005) subjects were required to pedal at a set cadence (60 rpm) which is suggested to be more economical, but lower than preferred cadences of trained cyclists (Marsh, Martin, & Foley, 2000). The only cadence restriction in the current investigation was that subjects could not drop below 60 rpm. This may have had some effect on the VO2max and, more particularly, MAP scores. Though the major contributor to the differences is more likely directly related to fitness levels. 5.2 Blood Lactate The changes observed in mean blood La during the recovery portion of the exercise bout were as anticipated. That is, mean blood L& was lowest in the CAR trial at the end of the recovery period, albeit only slightly less than the AR trial. Mean blood L& was always higher in the PR protocol. However, the differences observed between clearance rates in the AR and CAR protocols were non-significant, and thus it cannot be said with any certainty that chance alone could not account for the differences. Nonetheless, blood L& concentration was lower at all sample points during the 14 mm recovery period for CAR. Furthermore, the fact that blood La was lowest in the CAR trial at the L5 sample shows that L& likely peaked fastest during CAR. 61 This is in agreement with the suggestion put forth by Stamford et al. (1981) and also agrees with the findings of Gmada et al. (2005). Blood L& almost always reached a minimum at the L8 sample time, with the exception of three cases in the ten subjects analyzed over all the testing days. This suggests that L& values were still decreasing at the 14t1i mm of recovery. Analysis of the final L& scores at the end of the recovery period renders this more plausible, as the values are still sufficiently above resting levels. However, due to the experimental protocol design, this was not to be investigated. Recently, Baldari et al. (2004; 2005) reported that during a 30 mm recovery period, significant decreases in L& did not occur after the 20th mm of recovery. However, decreases in L& in their investigation were expressed as percentages, whereas the current investigation used absolute values. When expressed as percentages of maximum mean values, L& values at the end of recovery are substantially higher (72, 51, and 48% for PR, AR and CAR respectively) in this study than those reported by Baldari et al. (2004; 2005) at 15 mm of similar intensity recovery (approx. 20%, (2004); and 12-25% (2005)). It therefore follows that blood L& was likely still significantly decreasing after 14 mm of recovery and would require a longer recovery time to reach baseline levels after similar intensity work. Blood La values did however decrease to similar absolute values as those reported by Gmada et al. (2005). This difference is probably due to the type of exercise performed, as Gmada et a!. (2005) used an intermittent exercise protocol which resulted in higher overall L& values, similar to this investigation, compared to the single high-intensity bout employed by Baldari et al. (2004; 2005). As such, the finding that blood La clearance was greater with an AR protocol is in agreement with numerous other investigations (Baldari et al. 2005, Dodd et al., 1984; Gmada et al., 2005; McAinch et al., 2004; Spierer et al., 2004; Siebers & McMurray, 1981; Stamford et al., 1981). Among all the subjects analyzed, blood La was always lowest across the three testing days in either the AR or CAR trial. Therefore, the findings are in complete agreement with the vast majority of findings that show blood La clearing faster with AR. Investigations that do not show improved L& clearance with AR have used shorter recovery periods, much less than that of the present study (Dorado et al., 2004; Spencer et a!., 2006). Such short durations appear to not allow enough time for L& levels to decrease substantially. It is now currently believed that the major fate of La is oxidation (Gladden, 2003). This is especially true during exercise, when as much as 75% of L& is oxidized, with the remainder being disposed through gluconeogenesis (Brooks, 2007). During high-intensity exercise La is predominantly produced in the glycolytic muscle fibres and then is shuttled to adjacent and 62 remote oxidative muscle fibres, as well as other tissue sites, such as the heart and liver. During AR the maintenance of a lower metabolic power output, compared to the previously higher power output during exercise, provides an energy demand that is met by a large contribution of L& metabolism (i.e. oxidation, Bangsbo et al., 1994; Rontoyannis, 1988). Additionally, the metabolic power demand results in higher cardiac output and hence, higher blood flow, which is believed to contribute to largescale L& shuttling to more remote bodily locations. The collective outcome is a greater clearance of blood L&. 5.2.1 Combined Active Recovery and Lactate Clearance Stamford et al. (1981) initially suggested that a recovery intensity that decreased in relation to a decreasing blood L& concentration may clear blood L& faster than a single intensity recovery. Dodd et al. (1984) and Gmada et al. (2005) have previously examined this, with conflicting results. The current study is in agreement with the latter two, in that CAR was able to clear L& at a rate at least equal to AR performed at a single intensity. However, results did not confirm Gmada et al.’s (2005) finding that CAR was able to clear La faster than AR in a trained population and to a lesser extent in an untrained population. Differences in findings between the current investigation and those of Gmada et al. (2005) could be related to subject fitness level and/or recovery intensity. The trained group in the study by Gmada et al. (2005) had a mean VO2max of 56.5 ± 3.5 mL.kg’•min’, compared to 64.5 ± 6.6 mL•kg1•min’ in this study. Additionally, while recovery intensities in Gmada et al.’s (2005) work were based on individual thresholds, the workloads were determined relative to ventilatory thresholds minus 20% VO2max (specifically VT1 and VT2). The authors selected these workloads because they would be roughly equivalent to 35% and 65% of VO2max; both which had been previously used for La clearance analysis in other investigation (Dodd et al., 1984; Stamford et al., 1981). Recovery intensities in this investigation were quantified relative to IAT and IVT, determined via blood L& analysis and ventilatory data (respectively), and recently examined in a similar population for their effectiveness for L& clearance (Baldari et al., 2005). Furthermore, we felt that recovery intensities based on the AT determined with blood L& would be more accurate than solely from ventilatory data since the measure is more direct. It is important to have a recovery intensity that does not surpass the AT and result in a significant increase in L& production in order to facilitate L& clearance with AR (Dodd et al., 1984). Relative to VO2max, recovery intensity in this study was higher than that of Gmada et al. (2005). Volume of 02 consumed at VT2 was 61.0 ± 4.5 and 54.5 ± 6.0% VO2m, and at VT1 was 37.5 ± 5.0 and 33.5 ± 4.5% VO2m for trained and untrained subjects, respectively. 63 Conversely, V02 at IAT and IVL5O%AT in the present investigation were 73.2 ± 6.1 and 50.2 ± 5.3% VOmax, respectively. Accordingly, it may be that CAR did not result in significantly greater La clearance than AR because the IAT workload may have resulted in substantial production of L& within the muscle as the workload was higher than Gmada et al. (2005), which may have limited clearance from the blood. Interestingly however, it was observed that L& was lowest at the L5 sample point in CAR, so it seems that the TAT workload was at a sufficient intensity to clear La quickly enough to have an earlier peak. Despite the early peak, it seems that CAR did not incur a significant advantage to overall La clearance, above that obtain with AR. It appears that the protocol used to determine the TAT in this investigation was adequate, as there was no evidence of an increase in blood L& in the CAR trial. However, it may be that the workload was below the AT. Nonetheless, the workload could not have been much lower than the AT since subjects V02 at IAT was sufficiently high. Therefore the treadmill protocol adopted from Baldari et al. (2000) and modified for use on a bicycle ergometer seems to be valid for determination of the IAT. 5.3 Sprint Task Performance Performance results on the sprint task followed the hypothesized trends with respect to PP, but did not fulfill the other predictions. Across the three testing days, no significant differences were observed in PP output or FT. Mean power outputs was significantly greater in the AR group compared to the PR group. Additionally, TW was significantly greater during AR compared to PR. 5.3.1 Peak Power Peak Power did not significantly differ amongst the recovery interventions. However, over the course of the five sprints, subjects fatigued and PP significantly dropped. As a trend among individuals, subjects tended to achieve their highest PP on the CAR day. These occurred in either the first or second sprint. Conversely, the highest PP in the final sprint tended to occur on the AR and PR days. The finding that PP was not significantly affected by recovery in this investigation is in agreement with several other studies (Ainsworth et al., 1993; Spierer et al., 2004; Weitman, Stamford, Moffatt, & Katch, 1977). Ainsworth et al. (1993) demonstrated that as little as 6 mm of AR at 80 W was sufficient to restore 5 s PP output on a 45 s bicycle sprint task. While Spierer et al. (2004) showed that PP on repeat 30 s WAnT in both trained and untrained subjects was not significantly different when interspersed with 4 mm of either PR or AR at 28% VO2max. Based on these findings, it was hypothesized that 14 mm of recovery (with an additional 2 mm 64 transition period, totaling 16 mm) would be more than adequate to allow complete restoration of PP output in this study. Then from this point the sprint task would be identical, regardless of the recovery trial, and thus no differences in PP were expected on the latter sprints. Recent investigations have shown that AR can actually be detrimental to performance of sprint tasks when compared to PR (Dupont et a!., 2004; Dupont, Moalla, Matran, & Berthoin, 2007; Spencer et al., 2006). However, these investigations used very short recovery durations (15 s and 21 s, respectively), and therefore cannot be directly compared to the present study. There are likely different mechanisms responsible for changes in performance over such short durations compared to longer durations. It was suggested that AR impaired performance in the aforementioned investigations by limiting PCr resynthesis via competition for limited 02 supplies (Spencer et al., 2006). However, any potential limitations to PCr resythesis in this investigation did not cause any observable detrimental effect to PP performance in either of the active recoveries. Peak power output, when assessed over a short duration, is primarily controlled by energy release from free ATP and the cycling of the PCr system. Indeed, PCr resynthesis has been strongly correlated with subsequent sprint performance (Bogdanis, Nevill, Boobis, Lakomy, & Nevill, 1995). Both free ATP and PCr energy sources are quickly depleted during high-intensity exercise, yet the half-time for PCr resynthesis is short, approximately 21-60 s (Harris et al., 1976; Yoshida & Watari, 1993; Bogdanis et a!., 1995). Accordingly, maintenance of PP output requires adequate replenishment of PCr. As maintenance of PP output was observed in the current study, it follows that PCr levels were adequately replenished in all three recoveries. Thus, during 14 mm recovery, there seems to have been ample time to replenish PCr levels, despite a significant 02 demand to the working muscles in both the AR and CAR. Previously, it has been reported that AR can result in decreased oxyhemoglobinloxymyoglobin recovery as well as prolong PCr resynthesis when a short (15 s to 2 mm) recovery period is used (Dupont et al., 2007; Spencer et a!., 2006; Yoshida, Watari, & Tagawa, 1996). Oxygen competition between working muscle and the aerobic process of PCr repletion did not appear to be a factor over the longer recovery period in this study. It may be that 02 competition effects were dispersed over the extended recovery period and were thus nullified. That is to say, despite recovery intensities ranging from the IAT (73.2 + 6.1% V02m) to the IVT_50%AT (50.2 ± 5.3% VO2max), which imposed substantial V02 demands, it seems the overall recovery duration was sufficient to allow potentially limited PCr repletion over a long enough time period to result in adequate replenishment. However, as PCr levels were not actually measured in the current 65 investigation, their successful repletion can only be inferred from subject’s ability to achieve PP outputs equivalent to their baseline values. During the 30 s recovery periods between the five sprints PCr levels likely depleted, and this is evident in the significant main effect for time. Inter- sprint recoveries were equivalent across testing days and therefore this effect was similar on all days and is assumed to have had no discernable effect on the treatment groups. Peak power was operationally defined as the highest 1 s mechanical power output during each individual sprint. Therefore, it may be that due to the operational definition of PP the sprint task was not sensitive enough to find statistical differences in the assessment of fatigue. It is possible that any potential effects of energy metabolism on PP output were masked by the criteria used to determine PP. That is, subjects may have been able to reproduce PP for 1 s despite physiological changes that would not allow them to maintain those power outputs over a longer duration. Though it is worth mentioning that the cyclist tested were significantly trained, as well as motivated, and thus it is more likely that they would have been able to reproduce their performances with greater reliability. A 1 s duration was selected for the determination of PP as the proportion of the total sprint (10%) is most similar to the proportion of typical time, 5 s (16.7%), used to determine PP in a standard 30 s WAnT. Though, it should be noted that the measurement precision is lower (approx. ± 1.7%) with a shorter sample time for PP (SMI POWER, 2000). Furthermore, flywheel inertia was not considered in the calculations of power output. Lakomy (1986) showed that during a WAnT maximal power output was achieved before peak velocity and was 3 0-40% greater than power output at peak velocity. Therefore, it may be that power outputs reported in this study are lower than the power outputs actually achieved. However, as subjects were allowed to start sprints from a rolling start, the effects of not considering flywheel inertia in power calculations were reduced. Additionally, as all sprints were started in a similar matter, any error incurred became systematic and would not have compromised results with random error. 5.3.2 Mean Power Mean MP output during AR was significantly greater than during PR. This appeared to be a result of a greater ability to maintain power output in the latter sprints in the AR trial compared to the PR trial. As the difference between MP scores became more prominent in the last three sprints. Additionally, MP decreased significantly over time within a given sprint task. At the individual level, maximum MP over the three days tended to happen in the AR trial. Furthermore, the majority of the initial and final maximum MP outputs occurred during the AR trial. Concurrent with these trends, subjects tended to experience their minimum MP outputs on 66 either the CAR or PR days. Thus, in this investigation AR showed a strong tendency toward maintenance of power output during the sprint task, compared to the other two recovery modes. Several other researchers have noted a greater ability to maintain performance with the use of an AR (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Connolly et al., 2003; Corder et al., 2000; Signorile et al., 1993; Spierer et al., 2004; Thiriet et al., 1993). Of the aforementioned investigations, only several have noted benefits to MP specifically with an AR (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Spierer et al., 2004). Ahmaidi et al. (1996) found that AR resulted in higher MP outputs at high braking forces compared to PR during a repeat force-velocity test with increasing loads. The observed increased power performance with AR was associated with a decreased plasma L& concentration. The authors suggest that the greater L& clearance may have resulted in improved power outputs by reducing the amount of H accumulation, which has been previously shown to inhibit glycolysis. While the finding that AR recovery resulted in improved MP output is in agreement with the current investigation, the interpretation of why is not. In this study AR and CAR both resulted in enhanced L& clearance compared to PR. However, despite a L& clearance rate that was equivalent to AR with CAR, no performance benefit was observed in the CAR trial. Therefore, the performance benefit observed in the AR trial cannot be solely attributed to enhanced La clearance. Bogdanis et al. (1996b) noted greater MP output on a repeat 30 s sprint task separated by 4 mm, when AR was used compared to PR. The improvement was not associated with a lower blood L& concentration or higher blood pH. These findings are very similar to this investigation, with respect to the improved performance being dissociated from L& clearance. The authors suggest four possible mechanisms for the improved performance with AR: (1) greater PCr resynthesis, (2) lower muscle L& and [Hj, (3) increased aerobic contribution to energy supply, and (4) changes in mechanical efficiency from increased muscle water content; all of which have the potential to be affected by blood flow (Bogdanis et al., 1996b). Previously it has been shown that during the initial 10 s of a 30 s sprint (approx. 45% of the TW of the sprint) a large portion of the ATP demand is supplied through anaerobic metabolism (Bogdanis et al., 1996a). Specifically, in the initial 10 s PCr accounted for 34% and glycolysis 42% of the energy supply. Since the noted performance benefits in Bogdanis et al. ‘s (1 996a) investigation were attributed to a greater power output in the initial 10 s of the sprint, it was hypothesized that AR may have either enhanced PCr resynthesis, or increased the initial glycolytic contribution to the energy supply via H removal. Nevertheless, despite this supposition, more recent works, with recovery durations varying from 15 s to 15 mm, have shown a decreased muscle oxygenation (Dupont et 67 al., 2004; Dupont et a!., 2007) and PCr resynthesis (McAinch et a!., 2004; Spencer et a!., 2006) with AR compared to PR. These findings imply that PCr resynthesis is impaired with AR, rather than enhanced. Thus it may be more plausible to infer that the improvement in performance noted by Bogdanis et a!. (1 996a) was likely due to an increased initial glycolytic contribution to the energy demand. It is possible that a similar mechanism would be responsible for the improved performance noted in this investigation. Muscle L& concentration was not determined in this investigation, but it has been previously shown that changes in plasma L& and muscle La are independent of one another (McAinch et al., 2004). Thus, the finding that the two AR intensities in this investigation resulted in lower blood L& concentrations has no direct bearing on the muscle L& concentrations. Furthermore, investigations have reported both higher (McAinch et a!., 2004; Peters-Futre et a!., 1987) and lower (Bangsbo et al., 1994; Spencer et al., 2006) muscle L& concentrations with AR. Therefore, a postulation as to what the effects of the recovery intensities used in this study were on muscle L& cannot be made. However, it can be hypothesized that muscle pH would have been higher in the active recoveries as AR has been shown to increase muscle pH (Sairyo, Ikata, Takai, & Iwanaga, 1993; Sairyo et al., 2003) as well as blood pH (Siegler, Bell-Wilson, Mermier, Faria, & Robergs, 2006). A potentially lower [Hj in the muscle may have contributed to improved performance via a reduced inhibition of glycolytic enzymes and consequently greater glycolysis in the AR trial (Ahmaidi et aL, 1996; Karisson, Hulten & Sjodin, 1974). Slightly contrary to this statement is the fact that Siegler et a!. (2006) did not find any performance benefits with AR despite an increased blood pH. However, blood pH is not necessarily indicative of muscle pH, and it may be that a decreased muscle pH results in improved performance. Numerous investigations have shown that with repeat sprint tasks the aerobic contribution to energy supply increases with the number of sprints performed (Bogdanis et al., 1996a; Gaitanos, Williams, Boobis, & Brooks, 1993; Dorado et al., 2004; Trump, Heigenhauser, Putman, & Spriet, 1996). When AR is performed between the sprints the subsequent aerobic contribution is greater than with PR (Dorado et al., 2004). The increase in aerobic contribution has been attributed to faster V02 kinetics (Dorado et a!., 2004; Spriet, Lindinger, McKelvie, Heigenhauser, & Jones, 1989), similar to those observed when a warm-up is performed prior to high-intensity exercise (Bangsbo et al., 1994). Dorado et al. (2004) concluded that AR enhanced work capacity during high-intensity intermittent exercise by increasing the aerobic energy yield via faster V02 kinetics and a longer working time. Though the precise mechanism by which AR 68 enhances work capacity was unclear. It may be that the improved performance in the AR trial in this investigation was a result of expedited V02 kinetics too. However, ventilatory data was not measured during the sprint task, and therefore it cannot be determined for certain if this was in fact the case. At the V028 sample (i.e. 14th mm recovery) the V02 was highest in the PR protocol, followed by the CAR then AR protocols. This seems contrary to expected outcome since V02 should be lower during PR as the metabolic demand is minimal. However, the exercise and recovery protocol design had subjects complete the last 5 mm of the 14 mill recovery at the IVT_50%ATh in order to give them a ‘warm-up’ prior to the sprint task, as well as to minimize the differences in pre-sprint starting conditions with respect to HR, blood flow, and V02, by having all three recoveries finish at the same workload. In actuality, the ‘warm-up’ in the PR trial resulted in a significantly greater V02 (33.0 ± 6.4 mL.kg’.min’) at the end of recovery (V028) compared to AR (31.7 ± 6.0 mL.kg’.min1) It may be that the addition of exercise and increased V02 may have resulted in 02 competition between energy supply for exercise and the restorative processes of recovery. This contention is further support by the differences in the HR7 and HR8 values for PR and AR (108.2 ± 8.9 vs. 142.1 ± 9.4 bpm, pO.OS; and 139.2 ± 8.9 vs. 143.7 ± 11.3 bpm, p>0.05, respectively). From the 9th to 14th mm of recovery during PR HR and V02 increased drastically (11.1 ± 3.2 to 33.0 ± 6.4 mL.kg1. mn’) whereas in AR HR was relatively steady while V02 was gradually decreasing (32.9 ± 6.6 to 31.7 ± 6.0 mL.kg’.min’). Thus it would seem 02 extraction would be greater in the PR trial at this time since overall V02 was on the rise but close to the values from the AR and CAR trials, while HR was lower. The further uptake would be a result of the added exercise stimulus since any restorative processes would have already been in operation prior to the additional workload. Thus, it would appear that the V02 kinetics may very well have been slower in the PR trial, since it has been previously shown that V02 kinetics are faster when the 02 content of arterial blood is elevated (Balsom, Ekblom, & Sjodin, 1994). Two possible mechanisms suggested for faster V02 kinetics are: increased blood flow to exercised muscle group and/or greater maintenance of aerobic regulatory enzyme activation levels (Bangsbo et al., 1994; Dorado et al., 2004). As to why only AR recovery resulted in improved MP output in this investigation remains unclear. To our knowledge this is the only investigation to have examined the effects of a two-tiered intensity recovery on subsequent performance. It seems plausible that both the CAR and AR would both maintain leg blood flow. Furthermore, leg blood flow during the PR recovery would have been close to the two active 69 recoveries at the start of the sprint task since all subjects finished at least the last 5 mm of the last recovery period cycling at their respective IVT_SOO,’OIXT workloads. Though, throughout the greater portion of the last recovery (i.e. 9 mm of the 14 mm) leg blood flow would have been substantially lower in the PR trial compared to the two AR trials. Thus, while reduced leg blood flow could account for the absence of a performance benefit in the PR trial, it does not provide an explanation for the lack of a benefit in the CAR trial. The higher intensity portions of the CAR (at IAT) may have been too intense to allow sufficient recuperation, despite a greater blood flow. This would have then had to have had a carry-over effect into the following recovery period, as the remainder of the recovery period was performed at the same workload as the AR trial. It is unlikely that any carry over effects would have had to do with PCr levels, as the duration at the IVT_50%AT recovery intensity would have allowed sufficient PCr resynthesis. The elevated intensity may have had an effect on blood or muscle pH levels, or the hydroelectric balance of the muscle by changing NatK pump activity (Bangsbo et al., 1992). Similarly there may have been some form of carry-over effect in the CAR trial with respect to oxidative enzyme activation levels. Again, the higher intensity recovery may have caused physiological changes, such as a decreased pH, that would have potentially limited oxidative enzyme activation. However, the aforementioned arguments are speculative and further research is needed to substantiate the claims. Lastly, it has been suggested that water shifts from the blood to the muscle, as seen during intense sprint exercise, may increase intramuscular pressure and alter mechanical efficiency (Bogdanis et al., 1 996b). Large increases in total muscle water from intense exercise have previously been reported (Sjogaard & Saltin, 1982). The shifts in water are likely driven by changes in osmolarity due to the production of metabolites and increases in blood pressure (Bogdanis et al., 1996b). The elevated intramuscular pressure over that of blood pressure may offset vasodilation and restrict local blood flow to certain areas of the muscle (Bogdanis et a!., 1 996b). A faster removal of such metabolites (e.g. L&) may help to reduce the osmolarity imbalance and restore homeostatic conditions more rapidly and reduce muscular inefficiency. Notwithstanding to these findings, there is some evidence to the contrary. In a very applied study Franchini et al. (2003) used a similar duration recovery (15 mm) to examine the effects of a prior judo combat match on repeat upper body WAnT performance. No differences in performance were observed in this investigation between the different recoveries. Franchini et al. (2003) note that previous investigations finding performance benefits with AR versus PR, in general, used a shorter recovery duration (i.e. 6 mm or less). Interestingly, this study is then one 70 of few investigations finding improved performance with an AR recovery over a duration greater than 6 mm. One major difference between this investigation and that by Franchini et a!. (2003) is that the current investigation exercise modalities were similar between the fatiguing work bouts, recovery, and performance task (i.e. all performed on bicycle ergometers). In the investigation by Franchini et al. (2003), subjects were fatigued during a simulated judo combat, recovered running, and had their performance assessed by repeat upper body WAnT. Thus, direct comparisons between the studies are limited due to changes in the exercised muscle groups as well as the intensities used. There is similar trend throughout the literature that limits between study comparisons since there are a wide variety of experimental designs that have been used to assess the effects of recovery intensity on performance (e.g. exercise intensity/durationltype, recovery intensity/durationltype, etc.). 5.3.3 Fatigue Index There were no significant differences in Fl amongst the recovery groups, contrary to the prediction that differences in Fl would be inverse to MP. As expected FT increased over the number of sprints performed, demonstrating a larger difference between the PP and minimum power outputs. Interestingly, the mean FT was lower for all sprints in the AR trial, indicating a trend towards greater maintenance of power output. Despite being evident as a significantly greater MP output in the AR trial, the trend of a lower Fl in the AR trial was not statistical significant. Individual scores for FT showed less distinct trends than the other performance task variables. The amount of subjects who had their daily and overall maximum Fl scores was relatively evenly distributed amongst the three testing days. This was also the case for the minimum Fl scores as well as the initial and final maximum scores. Two possible explanations can account for this amount of variance among the individual scores. Firstly, it has been shown that performance decrement and fatigue indices scores inherently have large variations, even in trained populations, and thus should be interpreted cautiously (Glaister et al., 2007; McGawley & Bishop, 2006). Secondly, it is possible that the FT data did not reach significance for similar reasons described for PP. That is, the operational definitions for PP and minimum power used 1 s averages, which may not have been sensitive enough to detect statistical differences. Again, the 1 s duration for PP, and also minimum power, was selected to minimize the proportion of the sprint that each variable comprised. However, the shorter time comes at a cost of reduced precision for the measurement (SMI POWER, 2000). Decrement scores have been suggested to be more reliable for repeat-sprint exercise compared to Fl scores (Glaister, Stone, Stewart, 71 Hughes, & Moir, 2004). As such, it is worth noting that assessment of fatigue across the sprint task (TaF) was analyzed using the formula described by Fitzsimons et al. (1993) and recommended by Glaister et al. (2004) and did not show statistical significance. This formula uses MP power to assess the amount of decrement from an ideal power output to an actual power output and therefore avoids the problem of reduced precision that the 1 s average may have introduced to the FT score. The main limitation to this formula is the assumption that maximum power output occurs on the first sprint (Glaister et al., 2004). As this criterion was not met, FT were presented in the results. Interestingly, despite the use of MP (i.e. a 10 s versus a 1 s average) the TaF score was still not significantly different between groups. This lends more support to the first interpretation, that a lack of statistical significance was due to the inherent variability in measures of fatigue, rather than the second interpretation. The TaF formula and scores (see Table 24) are presented in Appendix TX. Many studies examining performance after active or PR have not included a measure of fatigue. This is partly due to the fact that several investigations have used TTE tests as their performance criterion (Dorado et al., 2004; McAinch et al., 2004; Siegler et al., 2006). Other investigations have simply not reported a measure of fatigue (Gaitanos et al., 1993) or used a design that did not allow assessment of fatigue by way of Fl or decrement scores (e.g. Ahmaidi et al., 1996). Bogdanis et al. (1996b), Signorile et al. (1993), and Spencer et al. (2006) all reported fatigue scores, and did not find any significant changes to fatigue after AR. Bogdanis and colleagues (1996b) used FT as their measure of fatigue, whereas Signorile et al. (1993) used fatigue rate (i.e. power decrement over time, Ws’), and Spencer et al. (2006) used a work decrement score (Fitzsimons et al., 1993). Furthermore, more similar to this study, Bogdanis et a!. (1996b) and Signorile et a!. (1993) did not find statistical differences in their measures of fatigue despite improved power outputs with AR. Conversely, Spencer et al., (2006) did not find statistical differences in work decrement, but found a reduced power output with AR. The similar lack of change to the fatigue measures is likely due to the greater variability associated with the measure of fatigue. In contrast, Spierer et al. (2004) found that AR resulted in a reduced fatigue index per bout (i.e. FT/# of bouts) during AR compared to PR in sedentary subjects but not in moderately trained hockey players. Congruent with this, sedentary subjects achieved higher MP outputs with AR whereas there was no significant difference in MP output between recoveries in the hockey players. Differences in the findings from Spierer et al. (2004) compared to the other investigations are likely due to methodological differences. Specifically, the other investigations performed a fixed number of sprints, whereas subjects in the latter investigation 72 performed serial WAnT until exhaustion (i.e. inability to complete the test) or until power output was reduced to less than or equal to 70% of the first sprint. 5.3.4 Total Work Total work demonstrated the same results as MP, which is expected, as TW is the aggregate of MP scores. That is, contrary to the hypothesized results, AR in fact resulted in the greatest TW followed by CAR then PR. Differences were only significant between AR and PR TW scores. However, CAR may result in greater TW than PR, but these findings cannot say with certainty whether the improvement was a result of chance alone. Among individual subjects the majority (seven) had their overall maximum TW output on the AR day. That was followed by the CAR day (four) then the PR day (one). Consistent with this trend is the fact that most (six) of subjects had their overall minimum TW output on the PR day, followed by the CAR day (five), then the AR day (one). Interestingly, Subjects 3 and 4, the exceptions to the trend, experienced their overall minimum and maximum on the AR and PR days respectively, and had the other extreme score default to the CAR day. The effect of the CAR on TW appears to be quite varied as an almost equal number of subjects experience TW maxima and minima on this day (four and five, respectively). The exact mechanism by which AR resulted in improved TW remains unclear, although it is obviously due to the same factors proposed for the improved MP output. Thus, it seems that there is an interaction between maintenance of blood flow to the working muscle and an optimal intensity with which to allow restorative processes to take place. An alternative interpretation would be that there may be an optimal intensity that allows improved metabolism on subsequent exercise rather than effecting restorative processes. In either case, the variance in individual TW outcomes may be an artifact of the procedures used to determine threshold intensities. Since, workloads during the VO2max test were increased by 30 W per stage, the thresholds could only be determined accurately within a 60 W range. Therefore, it could be that some of the subjects performed better in the CAR trial because their IAT workload was on the lower end on that range, whereas others may have performed worse because the workload may have been too high. Differences in relative workload intensities may have subsequently altered the aerobic contribution to metabolism by changing the chemical environment within the muscle and hence oxidative enzyme activation levels (Balaban, 1990). Or conversely, it may be that the glycolytic contribution to metabolism was hindered via an increased acidity (Spriet et al., 1989), since mild-intensity recovery has been shown to reduce intracellular pH levels (Sairyo et al., 2003). However, neither of these suppositions can be supported of refuted from the present study 73 results, and further research is necessary to clarify the mechanism for improved work capacity during AR. Two other investigations have noted improved TW with AR compared to PR (Signorile et al., 1993; Spierer et al., 2004). Interestingly, in the investigation conducted by Spierer and colleagues (2004), both sedentary and moderately trained subjects performed more TW with AR compared with PR despite no significant differences in MP output in the trained group. This is likely the result of the criteria used to determine the cut-off point for exercise bouts performed. Subjects performed bouts until they were unable to achieve 70% of the PP output from the first sprint bout. This resulted in three moderately trained subjects performing an extra sprint bout on the AR day. In contrast, a recent investigation found no significant difference in TW performed between active and PR, despite a greater power decrement and lower final PP during AR (Spencer et al., 2006). Similarly, Franchini et a!. (2003) measured TW performed but did not find any significant differences between active and PR. Other investigations have used TTE trials as their performance indicator (Dorado et a!., 2004; Dupont et al., 2004; Dupont et al., 2007; McAinch et al., 2004) and therefore are not relevant for a direct comparison. However, it could be argued that a longer TTE is indicative of more TW, but it is not known whether or not this would hold for a repeat sprint task. 5.4 Heart Rate and Volume of Oxygen Consumed Heart rate and V02 data were collected as measures of exercise intensity, to allow comparisons of the recovery workloads throughout the exercise and recovery periods as well as prior to the initiation of the sprint task. Additionally, HR was measured to provide an indirect and crude measure of blood flow, albeit total body blood flow. Despite the varying recovery intensities resulting in different metabolic demands, the final 5 mm of the 14 mm recovery period was performed at the same intensity in all three recovery protocols. This was an attempt to standardize the pre-sprint physiological conditions. Assessment of HR and VO2 at this sample point allowed the determination as to whether or not this goal was achieved. 5.4.1 Exercise and Recovery Heart Rate As expected mean HR values were different among the three recovery intensities during the SE bouts and recovery periods. Mean HR was greatest during the CAR trial, followed by AR, then PR. This was to be expected since the recovery intervals performed during the CAR were at a higher intensity for 2 mm of the 5 mm recovery periods, and 5 mm of the 14 mm recovery period compared to AR, as well as for the entire recovery intervals compared to PR. Similarly, the intensity during AR recovery intervals was higher throughout the entire recovery 74 periods compared to PR. This demonstrates that the amount of work performed prior to each sprint task was different between recovery conditions and had a significant physiological effect. Mean HR after the first SE bout (HR1) was highest in the PR trial, but there were no significant differences between the three recovery trials. This was to be expected since the workloads were the same across all three recovery trials and subjects performed equivalent warm-ups on all testing days. However, after the second SE bout (HR3) the effect of PR on HR became evident. The lack of activity during the PR interval resulted in a significantly lower HR after the SE2 bout (HR3) compared to the other two recoveries. The lower HR value at the start of the PR SE2 bout (HR2) prevented the HR from reaching the same peak values as in the AR and CAR protocols. Heart rate prior to SE3 (HR4) was slightly greater in the AR protocol compared to the CAR protocol, which in turn were both significantly greater than the PR protocol. It is likely that this small difference is a result of a slower pedal cadence in the CAR protocol during the higher intensity portions. Subjects were only required to keep their cadence above 60 rpm, but were allowed to cycle at whatever speed they found comfortable. Consequently, while not systematically observed, anecdotal observations suggest that subjects had lower cadences during the higher workload periods of the CAR, as several subjects were warned to maintain their cadence above 60 rpm. The slower cadences apparently allowed for a lower HR during the first 2 mm of the recovery interval that carried over into a slightly lower final HR at the mm of recovery. During the final recovery period mean HR followed the expected trends, with the highest values occurring in the CAR trial. At the end of recovery period (HR7) HR values were not significantly different from each other. This provides some evidence that the equivalent workload during the final 5 mm of the 14 mm recovery period was successful in standardizing the pre-sprint conditions. Therefore it is quite likely that subjects would have began each sprint task with a comparable cardiac output and blood flow to the legs on each testing day. Thus, it may be that differences observed in performance are related to changes that occurred earlier in the recovery period. The sprint task performance was not related to HR pre-sprint. Thus, the sprint task performance did not appear to be heavily controlled by pre-exercise cardiac output or blood flow factors. 5.4.2 Sprint Task Heart Rate During the sprint task, HR did not differ significantly between groups. Mean HR across all five sprints did however mirror the trend observed during the SE and recovery period. The CAR SHR was highest, followed by AR, and then PR. This suggests that there was some 75 residual effect on HR from the SE bouts and recovery period to the sprint task. Interestingly, the greater amount of TW performed in the AR trial did not result in a significant change in HR compared to the other recoveries. The differences in performance on the sprint task appeared to be independent of HR. This is likely due to the fact that energy metabolism during high-intensity short duration exercise is predicated by local muscle factors, rather than systemic attributes. 5.4.3 Volume of Oxygen Consumed Mean V02 throughout the SE and recovery period was greatest in the CAR trial, followed by AR, then PR. This was a result of the higher recovery intensities and hence greater amount of work performed in the two active recoveries prior to the sprint task. As expected, this again demonstrates that the amount of work performed prior to each sprint task was different between recovery conditions and had a significant physiological effect. The first and second SE bouts caused a similar V02 peak in all three recovery trials. After the first 5 mm recovery interval (V02), V02 was similar between the two active recoveries, which were both greater than the PR V02. Despite a substantially greater workload during the initial 2 mm of the recovery interval in the CAR, equivalent workloads for the final 3 mm of the interval resulted in a similar V02 value at the V02 sample during to the AR trial. This finding was similar to the HR data. However, contrary to the HR scores at the same sample point, V02 scores post-SE2 (V03) were not significantly different between recoveries. Leading into the third SE bout (V024), V02 showed the same trend as the previous bout, with the exception that V02 was slightly greater in the AR trial compared to the CAR trial. The difference is negligible and likely due to the summation of breaths over the 20 s sample period for V02 data. At the V025 sample, V02 was significantly higher in the CAR trial compared to the other two recoveries, since subjects were still riding at their IAT workload at this time. Active recovery V02 was in turn significantly greater than PR V02, as subjects were working at their IVT_50%AT workload compared to seated rest. Although the difference was less than the previous sample, V02 remained significantly greater in the CAR at the V026 sample time compared to the other recoveries, in spite of the reduced workload. Obviously there was a lag in V02 to accommodate the new workload. At the 9th mm of recovery (VO27) the differences in V02 between the two active recoveries were no longer significant. Fascinatingly, the addition of IVT_50%AT workload to the PR trial at the 9th mm of recovery resulted in a substantial increase in V02, to the point that the PR V02 was significantly greater than the AR V02 at the V028 sample. Though the difference is hardly substantial, it perhaps reflects a lack of metabolic 76 efficiency with the onset of exercise in the PR, compared to greater efficiency in the AR from a sustained effort. Thus, while the HR and V02 data demonstrate that the SE bouts provoked similar physiological responses among the three testing days, it is apparent that the overall physiological effects throughout the exercise and recovery period are different. This was to be expected, as obviously the workloads between the three recoveries were not equivalent. Interestingly, the least amount of work prior to the sprint task (i.e. PR trial) did not result in the best performance, as one might logically reason. In fact, it was the middle amount of work performed pre-sprint (i.e. AR trial) that resulted in the best performance, as measured by TW performed. 5.5 Practical Significance Active recovery has been suggested to result in superior recovery from exercise compared to PR (Ahmaidi et al., 1996; Bogdanis et al., 1996b; Spierer et al., 2004). However, the empirical evidence of this is more equivocal than the common perception. The belief that AR is beneficial to subsequent performance appears to result from historic belief that L& is a cause of fatigue in addition to the finding that AR does in fact clear La faster than PR. From a practical standpoint in performance athletics, AR would only be beneficial if it in fact did transmit some sort of improvement to performance, regardless of the effect on L& concentration. Furthermore, the effect on performance would need to impart a significant competitive advantage. The results from this investigation showed that the AR recovery employed over the two 5 mm and one 14 mm recovery intervals resulted in statistically significantly more TW during five 10 s sprints interspersed by 30 s, when compared to PR. However, conditions in a laboratory setting are different from those in an athletic competition. Specifically, when dealing with sports with multiple repetitive high-intensity bouts within one competition (e.g. hockey, football, soccer, basketball), a favourable and successful outcome is rarely determined purely by total work capacity. The unpredictable nature of the activities is determined by a combination of work capacity, physical fitness, skill, and motivation, to name a few. Similarly, in more fitness driven competitions (e.g. track, cycling, swimming), work capacity is not the sole factor determining success. Though, a greater work capacity would likely increase the chances for a successful outcome and thus is a desirable attribute. Interestingly, the data obtained in this study appear to show practical significance when the differences in mean MP between groups are converted from power outputs to velocities. Using an on-line bicycle speed and power calculator (Zom, 2005), differences in mean MP wattage demonstrated that the average subject tested would have gone 10 kmhr’ faster in the 77 AR trial compared to the PR trial and 8.1 kmbr’faster than the CAR trial. The CAR trial would have had a velocity 3.1 kmhf1 faster than the PR trial. Over 50 s of total sprint time, these velocities would correspond to distances of 138.9, 112.5 and 43.1 m, respectively, assuming that the mean MP (i.e. differences in velocity) was maintained. However, these calculations are somewhat contrived as they do not consider changing conditions, but nonetheless demonstrate the potential for application to competition. For example it cannot be determined from this experiment whether these effects would hold over an event like an individual pursuit (4000m), or be relevant in the final sprint to the finish line of a longer distance event. 78 CHAPTER VI: CONCLUSION 6.1 Conclusions This study adds further knowledge to the wealth of literature surrounding the use of AR and its effects on subsequent performance. The following was concluded based on the results of the study: a. Active and CAR both cleared blood L& significantly faster than PR. b. Active recovery resulted in greater TW performed on the five 10 s sprints via a greater maintenance of MP in the latter sprints compared to PR. c. Despite similar decreases in blood L& concentration between AR and CAR, no significant performance benefits were observed in the CAR trial. Therefore the improved performance in the AR trial appears to be independent of blood L& clearance. d. Active recovery at a moderate intensity (i.e. approx. 50% VO2m) may be an effective recovery intervention for recovery intervals of approximately 15 mm when subsequent high-intensity intermittent exercise requires the maintenance of power output. Therefore, this investigation confirms that a CAR, utilizing a moderate-to-high-intensity followed by a moderate-intensity of exercise, can successfully clear blood L&. However, the intensities used in this investigation for the CAR did not impart any significant clearance benefit above that of a single intensity moderate AR. And furthermore, did not statistically improve performance above that achieved from PR. Thus, for similar exercise requirements and recovery durations it appears that a moderate intensity AR is most beneficial. This study provides more evidence that L& is not a causative factor in the development of fatigue and that benefit of AR to performance is not directly related to the faster clearance of La. Additionally, it provides support for the use of an AR to promote improved performance over moderate recovery durations (i.e. approx. 15 mm). However, further research is needed to confirm the effect of AR on performance over this recovery duration as the literature is somewhat divided, as well as to determine the exact mechanism by which AR is beneficial. 6.2 Recommendations for Future Research Future research areas for the continued acquisition of knowledge regarding AR, lactate kinetics, performance and fatigue are diverse. While it is likely that future research endeavours will remain subdivided and specific, it is important for the resulting interpretations to remain unified and generalized to facilitate further discussion and dissemination. 79 6.2.1 Performance Criteria It is imperative that future research into the effects of AR incorporates some form of performance assessment. Anecdotal evidence of improved performance or logical reasoning that increased blood L& clearance should prevent fatigue (e.g. Gisolfi et al., 1966) can longer be the basis for the notion that AR is beneficial to performance. There must be empirical evidence of a performance benefit that can only come through objective assessment of performance variables. 6.2.2 Active versus Passive Recovery and Duration Dependency Given that recent investigations have demonstrated improved performance with PR over short duration sprint and recovery intervals (Dupont et al, 2004; Dupont et al., 2007; Spencer et al, 2006), further research is still required to elucidate the time frames as well as intensities at which active or PR is most beneficial. Specifically, active recoveries of a moderate duration (i.e. ranging from 3-20 mm) have shown equivocal findings as to whether they are beneficial or detrimental to performance. Therefore, future research should focus on illuminating the circumstances and intervals over which active or PR is most appropriate. Particularly the reproducibility of some of the previous works should be tested using larger sample sizes to confirm previous results. 6.2.3 Mechanisms The mechanisms by which AR is beneficial are still undetermined and further studies delving into the mechanisms are needed. Similarly, as the mechanisms of muscular fatigue still remain elusive, it follows that any insight into methods to maintain or improve performance will be limited. As such, it follows that further research into the mechanisms of fatigue are needed and will lead to new ideas and interpretations of the processes of recovery, specifically, how or why AR is beneficial. The development, utilization and refinement of new technologies will play a pivotal role in new research. Particular, Magnetic Resonance Spectroscopy (MRS) looks to be a promising tool as it allows for non-invasive and real-time measurement of the working muscle intracellular metabolism (Sairyo et al., 2003). Specifically, further research focused on the role of P1 and {H+] via MRS looks to be intriguing. Additionally, MRS technology may also prove to be useful for the assessment of muscular blood flow. 6.3.3 Suggested Modifications and Additions Further research using similar investigative models to the one employed in this study look to benefit from the addition of more blood measures as well as the inclusion of muscle 80 measures. Metabolites of key interest would be L&, H, Pi, and Ca2, in addition to examine the effects of temperature as well as the mediation of acid-base status via Po2. 81 References ACSM. (2006). ACSM’s Guidelinesfor Exercise Testing and Prescription (7th Ed.). Baltimore: Lippincott Williams & Wilkins. Ahmaidi, S., Granier, P., Taoutaou, Z., Mercier, J., Dubouchaud, H., & Prefaut, C. (1996). Effects of active recovery on plasma Lactate and anaerobic power following repeated intensive exercise. Medicine and Science in Sports and Exercise, 28(4), 450-456. Ainsworth, B. E., Serfass, R. C., & Leon, A. S. (1993). Effects of recovery duration and blood lactate level on power output during cycling. Canadian Journal ofApplied Physiology, 18(1), 19-30. Astrand, P. 0., Hultman, E., Juhlin-Dannfelt, A., & Reynolds, G. (1986). Disposal of Lactate during and after strenuous exercise in humans. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 61(1), 338-343. Balaban, R. S. (1990). Regulation of oxidative phosphorylation in the mammalian cell. The American Journal ofPhysiology, 258(3 Pt 1), C377-89. Baldari, C., & Guidetti, L. (2001). VO2max, ventilatory and anaerobic thresholds in rhythmic gymnasts and young female dancers. The Journal ofSports Medicine and Physical Fitness, 41(2), 177-182. Baldari, C., & Guidetti, L. (2000). A simple method for individual anaerobic threshold as predictor of max lactate steady state. Medicine and Science in Sports and Exercise, 32(10), 1798-1802. Baldari, C., Videira, M., Madeira, F., Sergio, J., & Guidetti, L. (2005). Blood lactate clearance during recovery at various intensities below the individual anaerobic threshold in triathletes. The Journal ofSports Medicine and Physical Fitness, 45(4), 460-466. Baldari, C., Videira, M., Madeira, F., Sergio, J., & Guidetti, L. (2004). Lactate clearance during active recovery related to the individual anaerobic and ventilatory thresholds in soccer players. European Journal ofApplied Physiology, 93(1-2), 224-230. Balsom, P. D., Ekblom, B., & Sjodin, B. (1994). Enhanced oxygen availability during high intensity intermittent exercise decreases anaerobic metabolite concentrations in blood. Acta Physiologica Scandinavica, 150(4), 455-456. Bangsbo, J., Golinick, P. D., Graham, T. E., & Saltin, B. (1991). Substrates for muscle glycogen synthesis in recovery from intense exercise in man. The Journal ofPhysiology, 434, 423- 440. Bangsbo, J., Graham, T., Johansen, L., & Saltin, B. (1994). Muscle lactate metabolism in recovery from intense exhaustive exercise: Impact of light exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 77(4), 1890-1895. Bangsbo, J., Graham, T., Johansen, L., Strange, S., Christensen, C., & Saltin, B. (1992). Elevated muscle acidity and energy production during exhaustive exercise in humans. The American Journal ofPhysiology, 263(4 Pt 2), R891-9. 82 Bangsbo, J., Madsen, K., Kiens, B., & Richter, E. A. (1997). Muscle glycogen synthesis in recovery from intense exercise in humans. The American Journal ofPhysiology, 2 73(2 Pt 1), E416-24. Bangsbo, J., Madsen, K., Kiens, B., & Richter, E. A. (1996). Effect of muscle acidity on muscle metabolism and fatigue during intense exercise in man. The Journal ofPhysiology, 495 (Pt 2)(Pt 2), 5 87-596. Belcastro, A. N., & Bonen, A. (1975). Lactic acid clearance rates during controlled and uncontrolled recovery exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 39(6), 932-93 6. Bogdanis, G. C., Nevill, M. B., Boobis, L. H., & Lakomy, H. K. (1996a). Contribution of phosphocreatine and aerobic metabolism to energy supply during repeated sprint exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 80(3), 876-884. Bogdanis, G. C., Nevill, M. E., Boobis, L. H., Lakomy, H. K., & Nevill, A. M. (1995). Recovery of power output and muscle metabolites following 30 s of maximal sprint cycling in man. The Journal ofphysiology, 482 (Pt 2)(Pt 2), 467-480. Bogdanis, G. C., Nevill, M. E., & Lakomy, H. K. (1994a). Effects of previous dynamic arm exercise on power output during repeated maximal sprint cycling. Journal ofSports Sciences, 12(4), 363-370. Bogdanis, G. C., Nevill, M. E., Lakomy, H. K., & Boobis, L. H. (1994b). Muscle metabolism during repeated sprint exercise in man. Journal ofPhysiology (London), 475, 25. Bogdanis, G. C., Nevill, M. E., Lakomy, H. K., Graham, C. M., & Louis, G. (1996b). Effects of active recovery on power output during repeated maximal sprint cycling. European Journal ofApplied Physiology and Occupational Physiology, 74(5), 461-469. Boileau, R. A., Misner, J. E., Dykstra, G. L., & Spitzer, T. A. (1983). Blood lactic acid clearance during treadmill and bicycle exercise at various intensities. The Journal ofSports Medicine and Physical Fitness, 23(2), 159-167. Bond, V., Adams, R. G., Tearney, R. J., Gresham, K., & Ruff, W. (1991). Effects of active and passive recovery on lactate clearance and subsequent isokinetic muscle function. The Journal ofSports Medicine and Physical Fitness, 31(3), 357-361. Bonen, A. (2001). The expression of lactate transporters (MCT1 and MCT4) in heart and muscle. European Journal ofApplied Physiology, 86(1), 6-11. Bonen, A., & Belcastro, A. N. (1976). Comparison of self-selected recovery methods on lactic acid clearance rates. Medicine and Science in Sports, 8(3), 176-178. Bret, C., Messonnier, L., Nouck Nouck, J. M., Freund, H., Dufour, A. B., & Lacour, J. R. (2003). Differences in lactate exchange and clearance abilities in athletes specialised in different track running events (100 to 1500 m). International Journal ofSports Medicine, 24(2), 108- 113. 83 Brooks, G. A. (1985). Lactate: Glycolytic product and oxidative substrate during sustained exercise in mammals—the “lactate shuttle.”. First international congress ofcomparative physiology and biochemistry, Liege, Belgium, A 208-218. Brooks, G. A. (2007). Lactate - link between glycolytic and oxidative metabolism. Sports Medicine, 37(4-5), 341-343. Brooks, G. A. (1998). Mammalian fuel utilization during sustained exercise. Comparative Biochemistry and Physiology. Part B, Biochemistry & Molecular Biology, 120(1), 89-107. Brooks, G. A., Dubouchaud, H., Brown, M., Sicurello, J. P., & Butz, C. E. (1999). Role of mitochondrial lactate dehydrogenase and lactate oxidation in the intracellular lactate shuttle. Proceedings ofthe National Academy ofSciences ofthe United States ofAmerica, 96(3), 1129-1134. Brooks, G. A., & Gaesser, G. A. (1980). End points of lactate and glucose metabolism after exhausting exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 49(6), 1057-1069. Choi, D., Cole, K. J., Goodpaster, B. H., Fink, W. J., & Costill, D. L. (1994). Effect of passive and active recovery on the resynthesis of muscle glycogen. Medicine and Science in Sports and Exercise, 26(8), 992-996. Connolly, D. A. J., Brennan, K. M., & Lauzon, C. D. (2003). Effects of active versus passive recovery on power output during repeated bouts of short term, high intensity exercise. Journal ofSports Science and Medicine, 2(2), 47-51. Cooper, S. M., Baker, J. S., Eaton, Z. E., & Matthews, N. (2004). A simple multistage field test for the prediction of anaerobic capacity in female games players. British Journal ofSports Medicine, 38(6), 784-789. Corder, K. P., Potteiger, J. A., Nau, K. L., Figoni, S. F., & Hershberger, S. L. (2000). Effects of Active and Passive Recovery Conditions on Blood Lactate, Rating ofPerceived Exertion, and Performance during Resistance Exercise. UNITED STATES: CSEP. (1998). The Canadian physical activity, fitness & flfestyle appraisal (2nd ed.). Ottawa, ON: Canadian Society for Exercise Physiology. Davies, C. T., Knibbs, A. V., & Musgrove, J. (1970). The rate of lactic acid clearance in relation to different baselines of recovery exercise. Internationale Zeitschrfl Fur Angewandte Physiologie, Einschliesslich Arbeitsphysiologie, 28(3), 155-161. Debold, E. P., Romatowski, J. G., & Fitts, R. H. (2005). The depressive effect of pi on the force calcium relationship in skinned single muscle fibers is temperature dependent. American Journal ofPhysiology. Cell Physiology, Dodd, S., Powers, S. K., Callender, T., & Brooks, E. (1984). Blood lactate disappearance at various intensities of recovery exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 57(5), 1462-1465. 84) Donovan, C. M., & Pagliassotti, M. J. (2000). Quantitative assessment of pathways for lactate disposal in skeletal muscle fiber types. Medicine and Science in Sports and Exercise, 32(4), 772-777. Dorado, C., Sanchis-Moysi, J., & Calbet, J. A. (2004). Effects of recovery mode on performance, 02 uptake, and 02 deficit during high-intensity intermittent exercise. Canadian Journal of Applied Physiology, 29(3), 227-244. Duncan, G. E., Howley, E. T., & Johnson, B. N. (1997). Applicability ofVO2max criteria: Discontinuous versus continuous protocols. Medicine and Science in Sports and Exercise, 29(2), 273-278. Dupont, G., Moalla, W., Guinhouya, C., Ahmaidi, S., & Berthoin, S. (2004). Passive versus active recovery during high-intensity intermittent exercises. Medicine and Science in Sports and Exercise, 36(2), 302-308. Dupont, G., Moalla, W., Matran, R., & Berthoin, S. (2007). Effect of short recovery intensities on the performance during two wingate tests. Medicine and Science in Sports and Exercise, 39(7), 1170-1 176. Faria, E. W., Parker, D. L., & Faria, I. E. (2005). The science of cycling: Physiology and training - part 1. Sports Medicine (Auckland, NZ), 35(4), 285-312. Faul, F., & Erdfelder, E. (1992). GPO WER: A Priori, Post-Hoc, and Compromise Power Analysesfor Windows. Bonn, FRG: Bonn University, Dept. of Pyschology. Fitts, R. H. (2003). Mechanism of muscular fatigue. In J. R. Poortmans (Ed.), Principles of exercise biochemistry (3rd rev. ed.) (pp. 279-300). Basel; New York: Karger. Fitzsimons, M., Dawson, B., Ward, D., & Wilkinson, A. (1993). Cycling and running tests of repeated sprint ability. Australian Journal ofScience and Medicine in Sport, 25(4), 82-87. Fletcher, W. M., & Hopkins, F. G. (1907). Lactic acid in amphibian muscle. Journal of Physiology, 35, 247-309. Franchini, E., Yuri Takito, M., Yuzo Nakamura, F., Ayumi Matsushigue, K., & Peduti Dal’Molin Kiss, M. A. (2003). Effects of recovery type after a judo combat on blood lactate clearance and on performance in an intermittent anaerobic task. The Journal ofSports Medicine and Physical Fitness, 43(4), 424-431. Gaitanos, G. C., Williams, C., Boobis, L. H., & Brooks, 5. (1993). Human muscle metabolism during intermittent maximal exercise. Journal ofApplied Physiology (Bethesda, Md.: 1985), 75(2), 712-719. Gastin, P. B. (2001). Energy system interaction and relative contribution during maximal exercise. Sports Medicine (Auckland, NZ), 31(10), 725-741. Gisolfi, C., Robinson, S., & Turrell, E. S. (1966). Effects of aerobic work performed during recovery from exhausting work. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 21(6), 1767-1772. 85 Gladden, L. B. (2003). Lactate metabolism during exercise. In J. R. Poortmans (Ed.), Principles ofexercise biochemistry (3rd rev. ed.) (jp. 152-196). Basel; New York: Karger. Gladden, L. B. (2004). Lactate metabolism: A new paradigm for the third millennium. The Journal ofPhysiology, 558(Pt 1), 5-30. Gladden, L. B. (2000). Muscle as a consumer of lactate. Medicine and Science in Sports and Exercise, 32(4), 764-771. Glaister, M., Howatson, G., Lockey, R. A., Abraham, C. S., Goodwin, J. E., & Mclnnes, G. (2007). Familiarization and reliability of multiple sprint running performance indices. Journal ofStrength and Conditioning Research /National Strength & Conditioning Association, 21(3), 857-859. Glaister, M., Stone, M. H., Stewart, A. M., Hughes, M., & Moir, G. L. (2004). The reliability and validity of fatigue measures during short-duration maximal-intensity intermittent cycling. Journal ofstrength and conditioning research /National Strength & Conditioning Association, 18(3), 459-462. Gmada, N., Bouhiel, E., Mrizak, I., Debabi, H., Ben Jabrallah, M., & Tabka, Z. et al. (2005). Effect of combined active recovery from supramaximal exercise on blood lactate disappearance in trained and untrained man. International Journal ofSports Medicine, 26(10), 874-879. Graham, T. E., Sinclair, D. G., & Chapler, C. K. (1976). Metabolic intermediates and lactate diffusion in active dog skeletal muscle. The American Journal ofPhysiology, 231(3), 766- 771. Gupta, S., Goswami, A., Sadhukhan, A. K., & Mathur, D. N. (1996). Comparative study of lactate clearance in short term massage of extremities, active recovery and a passive recovery period after supramaximal exercise sessions. International Journal ofSports Medicine, 17(2), 106-110. Harris, R. C., Edwards, R. H., Hultman, E., Nordesjo, L. 0., Nylind, B., & Sahlin, K. (1976). The time course of phosphoryicreatine resynthesis during recovery of the quadriceps muscle in man. Pflugers Archiv : Europeanjournal ofphysiology, 367(2), 137-142. Hermansen, L., & Stensvold, I. (1972). Production and clearance of lactate during exercise in man. Acta Physiologica Scandinavica, 86(2), 19 1-201. Hermansen, L., & Vaage, 0. (1977). Lactate disappearance and glycogen synthesis in human muscle after maximal exercise. The American Journal ofPhysiology, 233(5), E422-9. Hill, A. V., & Kupalov, F. R. 5. (1929). Anaerobic andAerobic Activity in Isolated Muscle. London: Hill, A. V., Long, C. N. H., & Lupton, H. (1924). Muscular exercise, lactic acid, and the supply and utilisation of oxygen. part VI. oxygen debt at the end of exercise. Proceedings ofthe Royal Society ofLondon. Series B, Containing Papers ofa Biological Character, 97(68 1), 127-137. 86 Hollmann, W. (2001). 42 years ago--development of the concepts of ventilatory and lactate threshold. Sports Medicine (Auckland, NZ.), 31(5), 315-320. Inbar, 0., Bar-Or, 0., & Skinner, J. S. (1996). The Wingate anaerobic test. Champaign, IL: Human Kinetics. Jervell, 0. (1928). Investigation of the concentration of lactic acid in blood and urine under physiologic and pathologic conditions. Acta Medica Scandinavica, Suppi. 24, 1-135. Jorfeldt, L. (1970). Metabolism of L(plus)- lactate in human skeletal muscle during exercise. Acta Physiologica Scandinavica. Supplementum, 338, 1-67. Juel, C. (1997). Lactate-proton cotransport in skeletal muscle. Physiological Reviews, 77(2), 321-35 8. Karisson, J., Bonde-Petersen, F., Henriksson, J., & Knuttgen, H. G. (1975). Effects of previous exercise with arms or legs on metabolism and performance in exhaustive exercise. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 38(5), 763- 767 Karisson, J., Hulten, B., & Sjodin, B. (1974). Substrate activation and product inhibition of LDH activity in human skeletal muscle. Acta Physiologica Scandinavica, 92(1), 2 1-26.. Karisson, J., & Saltin, B. (1970). Lactate, ATP, and CP in working muscles during exhaustive exercise in man. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 29(5), 596-602. Klausen, K., Knuttgen, H. G., & Forster, H. V. (1972). Effect of pre-existing high blood lactate concentration on maximal exercise performance. Scandinavian Journal ofClinical and Laboratory Investigation, 30(4), 415-419. Kowalehuk, J. M., Heigenhauser, G. J., Lindinger, M. I., Sutton, J. R., & Jones, N. L. (1988). Factors influencing hydrogen ion concentration in muscle after intense exercise. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology, 65(5), 2080- 2089. Lakomy, H. K. (1986). Measurement of work and power output using friction-loaded cycle ergometers. Ergonomics, 29(4), 509-517. Lau, S., Berg, K., Latin, R. W., & Noble, J. (2001). Comparison of active and passive recovery of blood lactate and subsequent performance of repeated work bouts in ice hockey players. Journal ofStrength Conditioning Research, 15(3), 367-37 1. Le Panse, B., Collomp, K., Portier, H., Lecoq, A. M., Jaffre, C., & Beaupied, H. et al. (2005). Effects of short-term salbutamol ingestion during a wingate test. International Journal of Sports Medicine, 26(7), 518-523. Lindinger, M. I. (2003). Exercise: A paradigm for multi-system control of acid-base state. The Journal ofPhysiology, 550(Pt 2), 334. 87 Lindinger, M. I., McKelvie, R. S., & Heigenhauser, G. J. (1995). K+ and lac- distribution in humans during and after high-intensity exercise: Role in muscle fatigue attenuation? Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 78(3), 765- 777. Margaria, R., Edwards, H. T., & Dill, D. B. (1933). The possible mechanisms of contracting and paying the oxygen debt and the role of lactic acid in muscular contractions. American Journal ofPhysiology, 106, 689-715. Marsh, A. P., Martin, P. E., & Foley, K. 0. (2000). Effect of cadence, cycling experience, and aerobic power on delta efficiency during cycling. Medicine and science in sports and exercise, 32(9), 1630-1634. McAinch, A. J., Febbraio, M. A., Parkin, J. M., Zhao, S., Tangalakis, K., & Stojanovska, L. et al. (2004). Effect of active versus passive recovery on metabolism and performance during subsequent exercise. International Journal ofSport Nutrition and Exercise Metabolism, 14(2), 185-196. McGawley, K., & Bishop, D. (2006). Reliability of a 5 x 6-s maximal cycling repeated-sprint test in trained female team-sport athletes. European Journal ofApplied Physiology, 98(4), 383-393. McLellan, T. M., & Skinner, J. S. (1982). Blood lactate clearance during active recovery related to the aerobic threshold. International Journal ofSports Medicine, 3(4), 224-229. Monedero, J., & Donne, B. (2000). Effect of recovery interventions on lactate clearance and subsequent performance. International Journal ofSports Medicine, 21(8), 593-597. Needham, D. M. (1971). Machina carnis . The biochemistry ofmuscular contraction in its historical development. Cambridge Eng.: University Press. Newman, E. V., Dill, D. B., Edwards, H. T., & Webster, F. A. (1937). The rate of lactic acid clearance in exercise. Journal ofApplied Physiology, 118, 457-462. Nielsen, 0. B., de Paoli, F., & Overgaard, K. (2001). Protective effects of lactic acid on force production in rat skeletal muscle. The Journal ofPhysiology, 536(Pt 1), 16 1-166. Pendergast, D., Leibowitz, R., Wilson, D., & Cerretelli, P. (1983). The effect of preceding anaerobic exercise on aerobic and anaerobic work. European Journal ofApplied Physiology and Occupational Physiology, 52(1), 29-35. Peters-Futre, E. M., Noakes, T. D., Raine, R. I., & Terbianche, S. E. (1987). Muscle glycogen repletion during active postexercise recovery. The American Journal ofPhysiology, 253(3 Pt 1), E305-1l. Posterino, G. S., Dutka, T. L., & Lamb, G. D. (2001). L(+)-lactate does not affect twitch and tetanic responses in mechanically skinned mammalian muscle fibres. Pflugers Archiv: European Journal ofPhysiology, 442(2), 197-203. Pyne, D. B., Boston, T., Martin, D. T., & Logan, A. (2000). Evaluation of the lactate pro blood lactate analyser. European Journal ofApplied Physiology, 82(1-2), 112-116. 88 Rämmal, K., & Strom, G. (1949). The rate of lactate utilization in man during work and at rest. Acta Physiologica Scandinavica, 17, 452-456. Rasmussen, H. N., van Hall, G., & Rasmussen, U. F. (2002). Lactate dehydrogenase is not a mitochondrial enzyme in human and mouse vastus lateralis muscle. The Journal of Physiology, 541(Pt 2), 575-580. Renaud, J. M. (2002). Modulation of force development by Na+, K+, Na+ K+ pump and KATP channel during muscular activity. Canadian Journal ofApplied Physiology, 27(3), 296-315. Richardson, R. S., Noyszewski, E. A., Leigh, J. S., & Wagner, P. D. (1998). Lactate efflux from exercising human skeletal muscle: Role of intracellular P02. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 85(2), 627-634. Robergs, R. A., Ghiasvand, F., & Parker, D. (2004). Biochemistry of exercise-induced metabolic acidosis. American Journal ofPhysiology. Regulatory, Integrative and Comparative Physiology, 287(3), R502-16. Rontoyannis, G. P. (1988). Lactate elimination from the blood during active recovery. The Journal ofSports Medicine and Physical Fitness, 28(2), 115-123. Rowell, L. B., Kraning, K. K.,2nd, Evans, T. 0., Kennedy, J. W., Blackmon, J. R., & Kusumi, F. (1966). Splanchnic clearance of lactate and pyruvate during prolonged exercise in man. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 21(6), 1773-1783. Sahlin, K., Fernstrom, M., Svensson, M., & Tonkonogi, M. (2002). No evidence of an intracellular lactate shuttle in rat skeletal muscle. The Journal ofPhysiology, 541(Pt 2), 569- 574. Sairyo, K., Ikata, T., Takai, H., & Iwanaga, K. (1993). Effect of active recovery on intracellular pH following muscle contraction, a 31 P-MRS study. The Annals ofPhysiological Anthropology = Seiri Jinruigaku Kenkyukai Kaishi, 12(3), 173-179. Sairyo, K., Iwanaga, K., Yoshida, N., Mishiro, T., Terai, T., Sasa, T., et al. (2003). Effects of active recovery under a decreasing work load following intense muscular exercise on intramuscular energy metabolism. International Journal ofSports Medicine, 24(3), 179-182. Siebers, L. S., & McMurray, R. G. (1981). Effects of swimming and walking on exercise recovery and subsequent swim performance. Research Quarterlyfor Exercise and Sport, 52(1), 68-75. Siegler, J. C., Bell-Wilson, J., Mermier, C., Faria, E., & Robergs, R. A. (2006). Active and passive recovery and acid-base kinetics following multiple bouts of intense exercise to exhaustion. International Journal ofSport Nutrition and Exercise Metabolism, 16(1), 92- 107. Signorile, J. F., Ingalls, C., & Tremblay, L. M. (1993). The effects of active and passive recovery on short-term, high intensity power output. Canadian Journal ofApplied Physiology, 18(1), 3 1-42. 89 Sjogaard, G., & Saltin, B. (1982). Extra- and intracellular water spaces in muscles of man at rest and with dynamic exercise. The American Journal ofPhysiology, 243(3), R271-80. SMI POWER [computer program] (2000). Version 5.2.8. Delray Beach, FL. Help on SMI Power: Measurements and Calculations (Velocity Measurement). Spencer, M., Bishop, D., Dawson, B., Goodman, C., & Duffield, R. (2006). Metabolism and performance in repeated cycle sprints: Active versus passive recovery. Medicine and Science in Sports and Exercise, 38(8), 1492-1499. Spierer, D. K., Goldsmith, R., Baran, D. A., Hryniewicz, K., & Katz, S. D. (2004). Effects of active vs. passive recovery on work performed during serial supramaximal exercise tests. International Journal ofSports Medicine, 25(2), 109-1 14. Sprague, P., & Mann, R. V. (1983). The effects of muscular fatigue on the kinetics of sprint running. Research Quarterlyfor Exercise and Sport, 54(1), 60-66. Spriet, L. L., Howlett, R. A., & Heigenhauser, G. J. (2000). An enzymatic approach to lactate production in human skeletal muscle during exercise. Medicine and Science in Sports and Exercise, 32(4), 756-763. Spriet, L. L., Lindinger, M. I., McKelvie, R. S., Heigenhauser, G. J., & Jones, N. L. (1989). Muscle glycogenolysis and H+ concentration during maximal intermittent cycling. Journal ofApplied Physiology (Bethesda, Md.: 1985), 66(1), 8-13. Stamford, B. A., Weltman, A., Moffatt, R., & Sady, S. (1981). Exercise recovery above and below anaerobic threshold following maximal work. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 51(4), 840-844. Stegmann, H., Kindermann, W., & Schnabel, A. (1981). Lactate kinetics and individual anaerobic threshold. International Journal ofSports Medicine, 2(3), 160-165. Stewart, P. A. (1981). How to understand acid-base . A quantitative acid-base primerfor biology and medicine. London: Edward Arnold. Tanaka, H., Bassett, D. R.,Jr, Swensen, T. C., & Sampedro, R. M. (1993). Aerobic and anaerobic power characteristics of competitive cyclists in the United States Cycling Federation. International Journal ofSports Medicine, 14(6), 334-338. Taoutaou, Z., Granier, P., Mercier, B., Mercier, J., Ahmaidi, S., & Prefaut, C. (1996). Lactate kinetics during passive and partially active recovery in endurance and sprint athletes. European Journal ofApplied Physiology and Occupational Physiology, 73(5), 465-470. Thiriet, P., Gozal, D., Wouassi, D., Oumarou, T., Gelas, H., & Lacour, J. R. (1993). The effect of various recovery modalities on subsequent performance, in consecutive supramaximal exercise. The Journal ofSports Medicine and Physical Fitness, 33(2), 118-129. Thomas, C., Perrey, S., Lambert, K., Hugon, G., Mornet, D., & Mercier, J. (2005). Monocarboxylate transporters, blood lactate clearance after supramaximal exercise, and fatigue indexes in humans. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 98(3), 804-809. 90 Trump, M. E., Heigenhauser, G. J., Putman, C. T., & Spriet, L. L. (1996). Importance of muscle phosphocreatine during intermittent maximal cycling. Journal ofApplied Physiology (Bethesda, Md.: 1985), 80(5), 1574-1580. Vandewalle, H., Peres, G., & Monod, H. (1987). Standard anaerobic exercise tests. Sports Medicine (Auckland, NZ), 4(4), 268-289. Wasserman, K. (1984). The anaerobic threshold measurement to evaluate exercise performance. The American Review ofRespiratory Disease, 129(2 Pt 2), S35-40. Watson, R. C., & Hanley, R. D. (1986). Application of active recovery techniques for a simulated ice hockey task. Canadian Journal ofApplied Sport Sciences, 11(2), 82-87. Weitman, A., & Regan, J. D. (1983). Prior exhaustive exercise and subsequent, maximal constant load exercise performance. International Journal ofSports Medicine, 4(3), 184- 189. Weitman, A., Stamford, B. A., & Fulco, C. (1979). Recovery from maximal effort exercise: Lactate disappearance and subsequent performance. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 47(4), 677-682. Weltman, A., Stamford, B. A., Moffatt, R. J., & Katch, V. L. (1977). Exercise recovery, lactate clearance, and subsequent high intensity exercise performance. Research Quarterly, 48(4), 786-796. Westerblad, H., Allen, D. G., & Lannergren, J. (2002). Muscle fatigue: Lactic acid or inorganic phosphate the major cause? News in Physiological Sciences . An International Journal of Physiology Produced Jointly by the International Union ofPhysiological Sciences and the American Physiological Society, 17, 17-21. Williams, J. R., Armstrong, N., & Kirby, B. J. (1992). The influence of the site of sampling and assay medium upon the measurement and interpretation of blood lactate responses to exercise. Journal ofSports Sciences, 10(2), 95-107. Yates, J. W., Gladden, L. B., & Cresanta, M. K. (1983). Effects of prior dynamic leg exercise on static effort of the elbow flexors. Journal ofApplied Physiology: Respiratory, Environmental and Exercise Physiology, 55(3), 891-896. Yoshida, T., & Watari, H. (1993). 3 1P-nuclear magnetic resonance spectroscopy study of the time course of energy metabolism during exercise and recovery. European Journal of Applied Physiology and Occupational Physiology, 66(6), 494-499. Yoshida, T., Watari, H., & Tagawa, K. (1996). Effects of active and passive recoveries on splitting of the inorganic phosphate peak determined by 31 P-nuclear magnetic resonance spectroscopy. NMR in Biomedicine, 9(1), 13-19. Yoshida, Y., Holloway, G. P., Ljubicic, V., Hatta, H., Spriet, L. L., Hood, D. A., et al. (2007). Negligible direct lactate oxidation in subsarcolenirnal and intermyofibrillar mitochondria obtained from red and white rat skeletal muscle. The Journal ofPhysiology, 582(Pt 3), 13 17-133 5. 91 Zorn, W. (2005). Speed and Power. Homepage. June 4, 2005. January 11, 2008. < http ://www.kreuzotter.de/englisblespeed.htm> 92 APPENDIX I Table 16. Daily and Overall Maximum and Minimum Blood Lactate Values Subject 1 2 3 4 5 6 7 8 9 10 11 12 PR Max. l4.8 16.3 15.1 l3.2 17 15.6 l5.O l5.6 l’7.2 l4.3 l5.O ll.8 Time3 2 7 5 4 5,6,77 5 5 5 6 3 Mm. 7.8 7.6 13.2 4.7 12.9 12.8 12.1 9.3 14.3 14.7 8.8 5.3 Time8 7 8 7 8 8 8 8 8 8 8 8 Final 15.6* 16.9* 15.1* 15.3* 15.9* 15.7 14.8 14.4* 15.8* 13.4 16.2* 10.7* AR Max. 11.1 15.6 l7.& 11.4 13.4 l6.O 14.6 12.0 l7.2 13.2 17.2 11.2 Time5 6 5 5 5 7 5 5 5 5 5 3 Mi 2.8 9.0 10.0 7.1 6.2 9.2 8.3 5.1 7.6 6.0 5.2 4.2 Time8 7 7 7 8 8 8 8 8 8 8 8 Final 11.6 15.2 13.3 14.4 13.1 16.8 15.8* 13.4 17.0 11.9 13.9 10.2 CAR Max. 13.4 l9.6 13.8 10.0 14.4 15.2 14.9 11. 15.2 12.8 12.4 11.1 Time5 7 3 5 3 5 5 3 5 5 5 3 Mm. 5.6 14.0 6.6 3.2 3.8 6.1 10.3 4.0 10.6 5.1 4.8 4.1 Time8 8 8 8 8 8 8 8 8 8 8 8 Final 14.6 16.9* 15.1* 12.9 10.6 16.0 14.3 11.6 15.0 16.6* 12.8 9.2 Subjects 2 and 11 had two and one missing data point, respectively; Overall maximum blood La; Overall minimum blood L&; * Overall final maximum blood L& 93 APPENDIX II Table 17. Daily and Overall Maximum and Minimum Peak Powers Subject 1 2 3 4 5 6 7 8 9 10 11 12 PR Max. 1113 1105 1012 1179 1050 l359 975 982 1055 894 942 1290 Sprint 2 5 1 1 1 2 1 2 1 1 2 1 Mm. 991 1045 907 1060 936 1189 847 859: 920 759 904 1142 Sprint 3 4 2, 5 5 4, 5 5 5 5 4 4 3 3 Initial 1068 1081 1012 1179w 1050 1328 975 951 1055 894w 923 1290 Final 1023* 1105 907* 1060 936 1189 847 859 926* 817 929* 1212 AR Max. 1010 1189 977 1135 1057 1320 lO34 1013 1023 894 985 l342 Sprint3 1 1 1 1 1 2 1 2 1 1 1 Mm. 856 1057 794 1048 969 1143 905 890 862 791 904 1228 Sprintl 4 3 5 4 5 5 5 5 4 4 4 Initial 856 1189 977 1135 1057 1320 911 1013 978 894 985 1342w Final 952 1147* 822 1048 1023* 1143 905* 890* 862 811 929* 1267* CAR Max. ll52 1141 lO68 ll8S IO’77 1312 901 lOY7 ll39 9S2 1010 1337 Sprint I 1 2 2, 3 1 1 3 1 1 2 1 2 Mm. 1010 1032: 9071 1091 9021 1073 850 884 8881 785 8731 1149 Sprint3 2 3 5 5 5 5 5 5 1 5 4 Initial 1152 1141 914 1135 1077 1312 878 1037w 1139 785 1010w 1290 Final 1017 1093 892 1091* 902 1073 850 884 888 843* 873 1157 Overall maximum PP; Overall minimum PP; Overall initial maximum PP; * Overall final maximum PP 94 APPENDIX III Table 18. Daily and Overall Maximum and Minimum Mean Powers Subject 1 2 3 4 5 6 7 8 9 10 11 12 PR Max. 818 878 9l9 814 865 1039 696 809 932 821 730 1014 Sprinti 1 1 1 1 1 1 1 1 1 1 1,2 Mm. 792 853 741 668 734 802 586 615 689 569 675 913 Sprint5 4 5 5 4 5 5 5 5 4 5 5 Initial 818 878 919 814 865 1039 696 809 932 821 730 1014 Final 792 860 741 668 739 802 586 615 689 597 675 913 AR Max. 844 933 888 791 862 1111r 822 923 830 7’75 1033 Sprint3 1 1 1 1 1 2 1 1 1 1 1 Mm. 796 875 696 672 759 827 610 630 709 650 669 976 Sprintl,5 5 4 5 4 5 5 5 5 4 5 5 Initial 796 933 888 791 862 1111 707w 822w 923 830 775 1033w Final 796* 875 751 672* 761* 827* 610 630* 709* 652* 669* 976* CAR Max. 894 921 916 757 839 1079 686 817 939 806 743 1018 Sprinti 1 2 1 1 1 1 1 1 2 1 1 Mi 786 885 777 676 712 799 611 592 708 638 635 872 Sprint5 5 4 5 5 5 5 5 5 5 5 5 Initial 894 921 873 757 839 1079 686 817 939 750 743 1018 Final 786 885* 784* 676 712 799 611* 592 708 638 635 872 Overall maximum MP; Overall minimum MP; Overall initial maximum MP; * Overall final maximum MP 95 APPENDIX IV Table 19. Daily and Overall Maximum and Minimum Fatigue Indexes Subject 1 2 3 4 5 6 7 8 9 10 11 12 PR Max. 41.6 36.6 31.3 56.2 38.1 47.2 46.4 46.9 4O.3 457 47.O 42.7 Sprint2 5 4 4 4 4 2 4 5 5 5 4 Mm. 35.7 27.6 13.2 47.6 25.3 32.0 40.7 24.5 21.3 18.0 35.8 34.9 Sprint3 2,3 2 1 2 1 1 1 1 1 1 3 thitial 38.0 28.5 16.0 47.6 25.6 32.0 40.7 24.5 21.3 18.0w 35.8 35.2 Final 38.4 36.6 27.9* 55.9 36.0 46.1* 44.8 44.3 40.3* 457* 47.0* 38.7 AR Max. 28.4 38.9 19.4 57.1 38.8 46.0 4’7.l 48.4 32.7 36.5 43.0 40.4 Sprint5 5 1 5 5 4 5 4 4 5 5 1 Mlii. 11.3 25.6 11.5 48.4 28.0 28.1 34.0 32.7 9.9 14.4 37.5 38.3 Sprintl 2 3 1 1 1 1 1 1 1 2 5 Initial 11.3 32.0 19.4 48.4 28.0 28.1 34.0 32.7 9.9 14.4 42.4 40.4 Final 28.4 38.9* 18.8 57.1 38.8* 42.6 47.1* 46.9 30.6 36.5 43.0 38.3 CAR Max. 47.l 33.1 32.3 64.6 34.9 41.7 41.7 S4.2 39.9 42.0 44.4 so3r Sprint4 5 4 4 2 5 4 5 3 4 1 2 Min. 35.2 25.7 11.5 50.0 31.9 27.3 34.6 34.3 26.6 7.4 32.4 35.8 Sprintl 2 1 1 1 2 1 1 1 1 2 1 Initial 35.2 28.6 22.8 50.0 31.9 28.2 34.6 34.3 26.6 7.4 44.4 35.8 Final 40.5* 33.1 11.5 58.9* 32.1 41.7 41.2 54.2* 34.8 42.0 44.3 42.6* Overall maximum Fl; Overall minimum Fl; Overall initial maximum Fl; * Overall final maximum Fl 96 APPENDIX V Table 20. Individual Total Work Outputs and Maxima and Minima Subject PR AR CAR 1 39890 40730 4l42O 2 43320 4492O 44790 3 40870 39500 4.l55O 4 36’74O 36330 36110 5 39400 4OOlO 38440 6 46180 4’72AO 46750 7 32190 33’72O 31990 8 34730 3532O 33500 9 39880 41150 4ll’7O 10 33570 35600 3562O 11 35170 3572O 34640 12 47360 4.98OO 46050 Overall maximum TW; Overall minimum TW 97 APPENDIX VI Table 21. Daily and Overall Maximum and Minimum Exercise and Recovery Heart Rate Values Subject 1 2 3 4 5 6 7 8 9 PR Max. 178 180 170 170 185 160 176 172 192 Time 1,3 1,3 3 3 3 1 3 3 3 Mi 97 99 96 94 108 102 118t 96 110 Time 6 6 6 6 6 6 6 6 6 Initial 178w 180 167 166 177 160 175 166 188 Final 107 102 107 104 116 112 119* 99 127 AR Max. 179 l9l 1’73 174 186 l6& 179 179 201 Time 3 3 3 3 3 3 3 3 3 Mm. 119 150 140 141 129 130 138 138 144 Time 2 2 2 5, 6 5, 6 2 6, 7 7 6 Initial 170 180w 164 169 179 161w 167 167 193w Final 96 111 110* 101 112 113* 116 115* 137* CAR Max. l8& 188 169 l’78 l9O l6& 183 175 2Ol Time 3 3 3 3 3 3 5 3 3 Mm. 133 154 141 149 129 136 136 131 149 Time 2,7 2 2 4 2 2 2 8 2 Initial 177 176 160 166 181 159 172 167c0 189 Final 113* 119* 110* 105* 117* 110 115 106 124 n 9, Subjects 10-12 were excluded; Initial, FIR at HR1; Final, HR at HR9; Overall maximum HR; Overall minimum HR; Overall initial maximum HR; * Overall final maximum HR 98 APPENDIX VII Mm. 166 167 162 Sprint 1 1 1 Initial 166 167 162 Final 176 176 171 AR Max. 178 190 171 Sprint 5 3, 5 3, 5 Mi 165 178 163 Sprint 1 1 1 Initial 165 178 163w Final 178 190* 171 CAR Max. l8l 193 1’76 Sprint 4, 5 4 4 Mm. 168 179 162 Sprint 1 1 1 Initial 168 179w 162 Final 181* 190* 173* 174 176 157 169 1 1 1 1 174 176 157 169 178 185* 169 178 178 184 l’7O l84 4,5 5 5 4,5 171 169: 162 173 1 1 1 1 171 169 162 173 178 184 170* 184* l83 l8S l7O 175 4 5 3,4,5 4,5 173 170 159 165 1 1 1 1 173 170 159 165 182* 185* 170* 175 Table 22. Daily and Overall Maximum and Minimum Sprint Task Heart Rate Values Subject 1 2 3 4 5 6 8 PR Max. 176 177 171 180 l8S 169 178 Sprint 4, 5 3 3, 4, 5 3 5 5 5 n = 7, Subjects 7 and 9-12 are excluded; Initial, HR at SHR1; Final, HR at SHR5; Overall maximum SHR; Overall minimum SUR; Overall initial maximum SHR; * Overall final maximum SHR 99 APPENDIX VIII 4 5 6 7 8 9 10 11 12 59.4 56.5 54.7 61.0 54.9 53.6 59.7 58.5 64.0 3 3 3 3 1 3 1 1 1,3 6.6 9.7 10.2 11.1 12.3 13.9 15.8 6.6 16.0 6 6 6 6 6 6 6 7 6 51.7 54.7 50.6 59.3 54.9w 46.5 59.7 58.8 64.0 38.5 25.1* 30.2* 29.1 30.2 26.8 35.8* 377* 31.5* 60.6 S6.S 601 59.2 56.5 56.5 6O.O 55.7 65.3 3 3 3 1 3 1,3 3 1 1 38.3 24.0 29.3 26.8 28.6 23.3 31.1 29.9 29.5 8 8 8 6 7 7 8 8 6 57.6 50.9 52.7 59.2 54.3 56.5 50.6 55.7 65.3 38.3 24.0 29.3 30.2* 31.1* 27.3 31.1 29.9 29.9 68.2 54.1 57.8 63.7 51.8 53t 59.2 6l.O 6S.’7 3 3 3 3 3 3 3 3 1 42.4 24.6 29.6 26.5 28.4 31.0 31.3 27.1 23.8 7 8 8 2 8 8 8 8 8 Table 23. Daily and Overall Maximum and Minimum Volume of Oxygen Consumed Values Subject 1 2 3 PR Max. 67.6 73.0 S2.2 Time 1 3 3 Mm. 13.1 9.1: 8.7 Time 6 7 6 Initial 67.6 63.4 46.8 Final 33.7 48.4 28.5 AR Max. ‘73.S 75.8 50.1 Time 3 3 3 Miii. 31.4 47.1 30.2 Time 8 6 8 Initial 70.0 65.9 45.7 Final 31.4 47.6 30.2 CAR Max. 70.1 82.3 51.0 Time 3 3 3 Miii. 36.4 48.6 32.3 Time 8 8 8 Initial 69.2 69.5 48.8w 57.0 52.8 53.6 55.7 51.3 54.7 57.9 58.0 65.7 Final 36.4* 48.6 32.3* 42.5* 24.6 29.6 27.4 28.4 31.0* 31.3 27.1 23.8 Initial, V02 at V021; Final, V02 at V028; Overall maximum V02; Overall minimum V02; Overall initial maximum V02; * Overall final maximum VO2 100 APPENDIX IX Sprint Task Fatigue Formula Fatigue = the percentage decrement score Fatigue = 100 — [(Total power output ± Ideal power output) x 100] Where Total power output = sum of MP outputs from all sprints Ideal power output = the number of sprints x MPmax Table 24. Sprint Task Fatigue Scores Subject PR AR CAR 1 2.5 3.5 7.3 2 1.3 3.7 2.7 3 11.1 11.0 9.3 4 9.7 8.1 4.6 5 8.9 7.2 8.4 6 11.1 15.0 13.3 7 7.5 8.2 6.7 8 14.1 14.1 18.0 9 14.4 10.8 12.3 10 18.2 14.2 11.6 11 3.6 7.8 6.8 12 6.6 3.6 9.5 5 9.1±5.1 8.9±4.2 9.2±4.1 101