Exercise intensity domains (i.e., moderate, heavy, and severe) have been defined based on the oxygen uptake (V[Combining Dot Above]O2) and blood lactate responses to constant work rate cycling exercise (11). During the transition from rest to the constant work rate of the severe-intensity domain, blood lactate and V[Combining Dot Above]O2 increase progressively in a biexponential fashion (by the development of the V[Combining Dot Above]O2 slow component, at intensities below V[Combining Dot Above]O2max) or are projected exponentially (at intensities at or above V[Combining Dot Above]O2max), reaching their maximal values at the end of exercise (6). The severe-intensity domain may be defined as the range of work rates over which V[Combining Dot Above]O2max can be elicited during constant load exercise (6,16). The asymptote of the power-duration relationship (i.e., critical power [CP]) has been considered the lower boundary of the severe domain (23). The homeostatic profiles in V[Combining Dot Above]O2 and lactate responses, blood acid-base balance, and concentrations of intramuscular metabolites (phosphocreatine, inorganic phosphate, and hydrogen ions) were reported when exercising below, but not above, the CP (15,23).
The CP model can be used to prescribe exercise training intensity, to assess aerobic capacity, and to predict aerobic performance and exercise tolerance (i.e., the time limit [Tlim]) within the severe domain (16). The exercise tolerance above the CP is defined by the curvature constant of the power-duration relationship (W′), which indicates a fixed amount of work that can be performed above the CP. To estimate the CP and W′, it is necessary to complete several (ideally 4 or more) independent high-intensity constant power exercise bouts for which the tolerable duration lies between 2 and 15 minutes (i.e., within the severe-intensity domain). Hill et al. (12) have demonstrated that the time to achieve the V[Combining Dot Above]O2max (TAV[Combining Dot Above]O2max) during the predictive tests can be expressed as a hyperbolic function of power. The power asymptote was not different from the CP, and they were highly correlated (r = 0.95–0.97). Using the linear relationship between the TAV[Combining Dot Above]O2max and its respective Tlim, it was possible to estimate the unique Tlim at which the V[Combining Dot Above]O2max is achieved at the point of fatigue (6). In addition to the CP and W′, the CP concept provides an estimate of the lowest and the highest exercise intensity at which the V[Combining Dot Above]O2max can be reached.
Despite the potential importance of the CP model to understand exercise performance, limited data related to the reliability of the cardiorespiratory parameters during exercise performed at the severe domain have been published. Ozyener et al. (22) found that the time constant and amplitude of the slow component (V[Combining Dot Above]O2SC) during cycling exercise was highly variable from day to day. Ozyener et al. (22) have estimated the V[Combining Dot Above]O2SC using only 1 transition. The utilization of only 1 exercise transition to analyze the response of the V[Combining Dot Above]O2 kinetics parameters during exercise performed at the heavy and severe-intensity domains might be associated with low confidence levels (17). In addition, the magnitude and distribution of breath-by-breath fluctuation (noise) might reduce the precision of the parameters estimation during nonsteady state exercise conditions. The averaging of repeated bouts has been traditionally performed to improve the signal-to-noise ratio of the data (17). Carter et al. (8) used multiple transitions to examine the day-to-day reproducibility of the V[Combining Dot Above]O2 kinetic response during treadmill running. During moderate-intensity and high-intensity exercise, Carter et al. (8) found that the amplitudes of the response were more robust than the temporal parameters (time constants and time delay) (coefficient of variation = ∼ 6% vs. 16–23%, respectively). Many intervention studies (e.g., training, pacing, and prior exercise) used to alter the parameters of the power-time relationship have been conducted in active individuals during cycling exercise (16). It is important to note that the time constants and V[Combining Dot Above]O2SC are higher for cycling than running during the heavy and severe exercise domains (7,19). It is important to use multiple transitions to examine the reproducibility of the V[Combining Dot Above]O2 kinetic response during cycling exercise.
The purpose of this study was to determine the test-retest reliability of the cardiorespiratory parameters during cycling exercise performed at the severe domain in active individuals.
Experimental Approach to the Problem
To analyze the reliability of the cardiorespiratory parameters during cycling exercise performed at the severe domain, 13 active males were recruited. All the subjects performed 4 repetitions of square-wave transitions from rest to a power corresponding to 95% IV[Combining Dot Above]O2max (severe-intensity exercise) to determine the parameters of the V[Combining Dot Above]O2 kinetics and Tlim (dependent variables) in 2 sessions separated by 48–72 hours. The procedures employed in these visits were identical. The V[Combining Dot Above]O2 responses to the 2 severe exercise bouts performed in each session were averaged before the analysis to reduce the breath-to-breath noise and enhance confidence in the parameters derived from the modeling process (17). The intraclass correlation coefficient and typical error as the coefficient of variation were used to assess reliability (14).
Thirteen active males (24.5 ± 4.5 years; 77.8 ± 11.3 kg; 177 ± 5.5 cm) volunteered for the study. The subjects were physical education students (undergraduate and postgraduate) involved in recreational sports (soccer, basketball, and volleyball) who had not participated in regular aerobic training for at least 6 months before the start of the study. All of the participants were healthy and free of cardiovascular, respiratory, and neuromuscular disease. All of the risks associated with the experimental procedures were explained before involvement in the study, and each participant completed a written informed consent. The study was performed according to the Declaration of Helsinki, and the protocol was approved by the São Paulo State University's Ethics Committee.
Each participant was required to visit the laboratory 3 times, and the following tests were performed: (a) an incremental test to exhaustion to determine the V[Combining Dot Above]O2max and the intensity associated with the V[Combining Dot Above]O2max (IV[Combining Dot Above]O2max); and (b) a total of 4 repetitions of square-wave transitions from rest to a power corresponding to 95% IV[Combining Dot Above]O2max to determine the parameters of the V[Combining Dot Above]O2 kinetics. The participants performed only 2 transitions on any given day (i.e., session 1: 2 transitions; session 2: 2 transitions). The interval between the 2 sessions was 48–72 hours. The entire testing was completed within a period of 2 weeks. The participants were instructed to arrive at the laboratory in a rested and fully hydrated state, at least 3 hours postprandial, and they were asked not to perform any strenuous activity during the day before each test. All of the tests were performed in a climate-controlled (21–22° C) laboratory at the same time of day (±2 hours) to minimize the effects of diurnal biological variation on the results (1).
Each participant performed an incremental exercise test to volitional fatigue on an electronically braked cycle ergometer (Excalibur sport, Groningen, Netherlands) to determine the V[Combining Dot Above]O2max and lactate threshold (LT). On this visit, the seat height and handlebar positions were individually adjusted for comfort, and the adjustments were recorded and replicated in subsequent tests. The incremental protocol started at a power output of 35 W, with increments of 35 W every 3 minutes. The test was performed until voluntary exhaustion with strong encouragements provided throughout the test. Pedal cadence was kept constant (70 rpm). Throughout the tests, the respiratory and pulmonary gas exchange variables were measured using a breath-by-breath gas analyzer (Quark PFTergo, Cosmed, Italy). Before each test, the oxygen and carbon dioxide analysis systems and pneumotachograph were calibrated using precision reference gases and a syringe of known volume (3 L), respectively, according to the manufacturer's instructions. The heart rate (HR) was monitored (Polar, Kempele, Finland) throughout the test. The V[Combining Dot Above]O2max was defined as the highest average 15-second V[Combining Dot Above]O2 value recorded during the incremental test. All of the participants fulfilled at least 3 of the following 4 criteria to ascertain that the V[Combining Dot Above]O2max was attained: (1) a respiratory exchange ratio (R) greater than 1.1; (2) a blood lactate concentration greater than 8 mmol·L−1; (3) a peak HR at least equal to 90% of the age-predicted maximal; and (4) the identification of a plateau of less than 150 ml·min−1 between 2 subsequent stages (13). The intensity associated with the V[Combining Dot Above]O2max (IV[Combining Dot Above]O2max) was defined as the power output at which V[Combining Dot Above]O2max occurred. At the end of each stage, earlobe capillary blood samples (25 µL) were collected into an eppendorf tube and analyzed for lactate concentration ([La]) using an automated analyzer (YSI 2300 STAT; Yellow Spring, OH, USA). Plots of blood [La] against the power output and V[Combining Dot Above]O2 were provided to 2 independent reviewers, who determined the LT as the first sudden and sustained increase in blood lactate above the resting concentrations (7).
The participants performed a series of constant work rate transitions at 95% IV[Combining Dot Above]O2max on separate days. On a given day, the participant completed 2 transitions that were separated by 60 minutes of passive recovery. It has been demonstrated that the Tlim (4) and the V[Combining Dot Above]O2 kinetic parameters (60 minutes of recovery between the transitions) (3) are not modified when 2 severe exercise transitions are performed in the same day. The first transition lasted 6 minutes and was conducted to determine the V[Combining Dot Above]O2 kinetics. The second transition was conducted until voluntary exhaustion to determine the V[Combining Dot Above]O2 kinetics (first 6 minutes) and the Tlim (time to exhaustion). The V[Combining Dot Above]O2 data of the 2 exercise transitions of the day (i.e., 6 minutes) were time aligned for the V[Combining Dot Above]O2 kinetics analysis (see the section, “Modeling of V[Combining Dot Above]O2”). The exercise protocol began with a 5-minute warm-up at 50% IV[Combining Dot Above]O2max followed by a 7-minute rest. The participants performed 3 minutes of unloaded cycling at 20 W, followed by a step change in the power output to 95% IV[Combining Dot Above]O2max, which was maintained for 6 minutes or until exhaustion (Figure 1). The second transition was terminated when the participant could not maintain a cadence of >65 rpm despite verbal encouragement. The Tlim was measured to the nearest second. During the constant work rate tests, the peak of the V[Combining Dot Above]O2 (V[Combining Dot Above]O2peak) was defined as the highest 15-second average value. The capillary blood samples were collected before and after 1, 3, and 5 minutes of the second transition of the day for peak blood lactate concentration ([La]peak) determination.
Modeling of V[Combining Dot Above]O2
The breath-by-breath data from each exercise test were filtered manually to remove outlying breaths, defined as breaths ± 3 SD from the adjacent 5 breaths. For each exercise transition, the breath-by-breath data were interpolated to give second-by-second values. The 2 transitions performed in each day were then time aligned to the start of exercise and averaged to enhance the underlying response characteristics. The first 20 seconds of data after the onset of exercise were deleted to eliminate the phase I component (i.e., the cardiodynamic phase) from the analysis (24). The first 6 minutes of data (20–360 seconds) were then modeled with a biexponential function to analyze the V[Combining Dot Above]O2 responses to severe exercise, as described by the following equation:
where V[Combining Dot Above]O2(t) is the absolute V[Combining Dot Above]O2 at a given time t; V[Combining Dot Above]O2baseline is the mean V[Combining Dot Above]O2 in the baseline period; Ap, TDp, and τp are the amplitude, time delay, and time constant, respectively, describing the phase II increase in V[Combining Dot Above]O2 above baseline; and As, TDs, and τs are the amplitude of, time delay before the onset of, and time constant describing the development of the V[Combining Dot Above]O2 slow component, respectively. An iterative process was used to minimize the sum of the squared errors between the fitted function and the observed values. The V[Combining Dot Above]O2baseline was defined as the mean V[Combining Dot Above]O2 measured over the final 60 seconds of exercise preceding the step transition to severe exercise. Because the asymptotic value (As) of the exponential term describing the V[Combining Dot Above]O2 slow component may represent a higher value than is actually reached at the end of the exercise, the actual amplitude of the V[Combining Dot Above]O2 slow component at the end of exercise was defined as As'. The end-exercise V[Combining Dot Above]O2 was defined as the mean V[Combining Dot Above]O2 measured over the final 30 seconds of severe exercise.
The data are presented as the mean ± SD. The normality of the distribution was verified by the Shapiro-Wilk’s test. Student's paired t-test was used to compare the data between sessions (S1 and S2). The V[Combining Dot Above]O2max and V[Combining Dot Above]O2peak obtained during S1 and S2 were compared using 1-way analysis of variance with the Tukey's post hoc tests where appropriate. The intraclass correlation coefficients (ICC), typical error of measurement (TE), and coefficient of variation (CV) were calculated according to Hopkins (14) to determine the test-retest reliability. The CV represents the variation in a subject's test score from measurement to measurement (14). The confidence intervals were calculated to determine the limits of agreement between the trials. The reliability was classified in accordance with the ICC values, as poor (<0.4), moderate (0.4–0.75), or excellent (>0.75) (10).
The V[Combining Dot Above]O2max measured in the incremental test was 3306.7 ± 441.9 ml·min−1, with the LT occurring at 1859.7 ± 503.1 ml·min−1, which corresponded to a work rate of 262.3 ± 40.7 and 120.0 ± 31.5 W, respectively (Table 1). There was no significant difference between the V[Combining Dot Above]O2max and V[Combining Dot Above]O2peak obtained during S1 and S2 (p > 0.05).
The results regarding the test-retest reliability of the V[Combining Dot Above]O2peak, peak HR (HRpeak), [La]peak, and Tlim are presented in Table 2. The Tlim (p = 0.02) and [La]peak (p = 0.04) were significantly higher in S2 than in S1. Following the benchmarks of Fleiss et al. (10), the reliability was excellent for the Tlim, V[Combining Dot Above]O2peak, HRpeak, and [La]peak. The CV for these variables ranged between 1.6% and 9.6%.
The reliability data of the V[Combining Dot Above]O2 kinetics parameters are presented in Table 3. No significant difference was found between S1 and S2 for all the variables (p > 0.05). Following the benchmarks of Fleiss et al. (10), the reliability was poor for τs, TDs, and TDp and moderate for τp. The CV for the time-based parameters ranged between 21% and 68%. For the amplitudes of the V[Combining Dot Above]O2 response, the reliability of the measurements was excellent (10). The CV for the amplitudes of the V[Combining Dot Above]O2 response ranged between 4.3% and 13.6%.
The severe-intensity domain is characterized by the attainment of V[Combining Dot Above]O2max. In our study, the V[Combining Dot Above]O2peak values obtained during S1 and S2 were not significantly different than those of the V[Combining Dot Above]O2max. The aim of this study, which was to determine the test-retest reliability of the cardiorespiratory parameters during cycling exercise performed at the severe domain in active individuals, was attained successfully. Similar to the results found by Carter et al. (8) during treadmill exercise, the amplitudes of the V[Combining Dot Above]O2 kinetics responses were more robust than the temporal parameters during severe cycling exercise. The second of 2 Tlim measurements in a group of active individuals was related to (ICC = 0.78; p ≤ 0.01), but significantly greater than, the first (p = 0.02).
The reliability of the performance tests seems to be influenced by the exercise intensity and the protocol utilized in these tests and by the aerobic fitness level of the volunteers. It is generally reported that submaximal exercise intensity (70–80% V[Combining Dot Above]O2max) produces higher variability (CV = 15–25%) on time to exhaustion than maximal and supramaximal (115–125% V[Combining Dot Above]O2max) exercise intensities (CV = 1–10%) (9). For similar exercise intensity (100% IV[Combining Dot Above]O2max), Laursen et al. (18) found in highly trained cyclists a better day-to-day reproducibility of the time to exhaustion (CV = 6%) than found in our study (CV = 9.6%). Open-loop tests, in which the task is open ended (i.e., time-to-exhaustion tests), may result in a greater degree of variation compared with closed-loop tests (i.e., time trials) (9). In our study, the higher [La]peak during S2 suggests that in nonfamiliarized individuals, the anaerobic store (i.e., W′) was not fully depleted at the tolerable limit during S1.
To our knowledge, this is the first study to use multiple transitions to examine the reproducibility of the V[Combining Dot Above]O2 kinetic response during severe cycling exercise. We have found that the within-subject variability for the amplitudes of the V[Combining Dot Above]O2 kinetic responses to severe cycling exercise was lower than for the time-based parameters. Others studies performed in cycling (22) and treadmill exercise (8) have found similar results. Considering that the influence of the interbreath “noise” on the V[Combining Dot Above]O2 response is reduced by averaging the results of several identical tests, the biological variability can explain in part this different within-subject variability for the amplitudes of the V[Combining Dot Above]O2 kinetic responses and the time-based parameters.
It has been suggested that some factors, such as the magnitude and distribution of breath-by-breath fluctuations (noise), can affect the precision of the estimation of the V[Combining Dot Above]O2 kinetics parameters (i.e., τp), particularly during non–steady-state conditions (i.e., severe domain) (17). As the breath-to-breath variability in the V[Combining Dot Above]O2 signal is largely independent of the work rate (approximately 200 ml·min−1), greater amplitude of the response (as analyzed in present study) reduces the degree to which this inherent noise negatively affects confidence in the parameter estimates (17). During severe-intensity exercise, 2 identical transitions have been found to be suitable for achieving confidence in characterizing the dynamics of V[Combining Dot Above]O2 (4,7). The CV for the time-based parameters (e.g., τp ∼ 21%) found in this study suggests that 2 identical transitions have a moderate ability or likelihood of detecting an individual change after an intervention (e.g., training).
In practical terms, Hopkins (14) demonstrated that approximately 1.5–2.0 times the TE could be used to detect the smallest amount of individual change required to designate a change as real and beyond the bounds of measurement error. In our study, the TE expressed as the CV was 13.6% for V[Combining Dot Above]O2SC, suggesting that the amplitudes of the V[Combining Dot Above]O2 kinetics responses can identify the effects of the interventions aimed at improving the V[Combining Dot Above]O2SC in active individuals. Ocel et al. (21) analyzed untrained individuals and obtained training-related changes in the V[Combining Dot Above]O2SC of approximately 300–500 ml·min−1 (57–71%), substantially exceeding the 1.5–2.0 times the TE obtained in our study. The TE expressed as CV for τp (∼21%) suggests that the time-based parameters have a moderate chance of reliably detecting a change after the interventions aimed at improving the τp in active individuals. Berger et al. (2) verified that 6 weeks of continuous and interval training have promoted improvements in the V[Combining Dot Above]O2 kinetics (τp ∼8–11 seconds) during severe-intensity cycle exercise. This improvement on the V[Combining Dot Above]O2 kinetics (∼22–31%) is less than the limits (1.5–2.0 x TE) proposed by Hopkins (14) (i.e., 31.5–42%), suggesting that multiple transitions are recommended to monitor the changes in an individual over any time frame.
In conclusion, this study has shown that the Tlim scores measured in different sessions during severe cycling exercise in active individuals were significantly different. We demonstrated that the amplitudes of the V[Combining Dot Above]O2 response have significant moderate to high reliabilities. The time-based parameters (i.e., τ and time delay) presented an important day-to-day intraindividual variation, and several identical transitions at the severe-intensity domain are recommended to improve the signal-to-noise ratio of the data.
Knowledge of an individual's CP and W' can be used to refine and monitor the training protocols and to predict exercise tolerance (i.e., the Tlim values between approximately 2 and 15 minutes). To estimate the CP and W′, it is necessary to complete several (ideally 4 or more) independent high-intensity constant power exercise sessions for which the tolerable duration lies between 2–15 minutes. Analyzing the V[Combining Dot Above]O2 kinetics response during these sessions allows for estimation of both the lowest and the highest exercise intensity at which the V[Combining Dot Above]O2max can be reached. Active individuals should be familiarized with the performance protocol (i.e., time to exhaustion tests) by at least 1 trial before measurement commences. The amplitudes of the V[Combining Dot Above]O2 response are more reliable than the time-based parameters obtained during severe cycling exercise. The amplitudes of the V[Combining Dot Above]O2 kinetics responses will be helpful in monitoring individual performance changes over time and assessing the effectiveness of physical training and exercise interventions in active individuals. Caution is required when interpreting the time constants and time delays of the various phases. Several (e.g., 4–6) identical transitions are recommended to bring the 95% confidence limits of the V[Combining Dot Above]O2 time constant (τp) estimation to within 5 seconds (20).
We thank the subjects for participation in this study, FAPESP CNPq, and FUNDUNESP for financial support.
1. Atkinson G, Reilly T. Circadian variation in sports performance. Sports Med 21: 292–312, 1996.
2. Berger NJA, Tolfrey K, Williams AG, Jones AM. Influence of continuous and interval training on oxygen uptake on-kinetics. Med Sci Sports Exerc 38: 504–512, 2006.
3. Burnley M, Doust JH, Jones AM. Time required for the restoration of normal heavy exercise VO2 kinetics following prior heavy exercise. J Appl Physiol 101: 1320–1327, 2006.
4. Burnley M, Davison G, Baker JR. Effects of priming exercise on VO2
kinetics and the power–duration relationship. Med Sci Sports Exerc 43: 2171–2179, 2011.
5. Caputo F, Denadai BS. Exercise mode affects the time to achieve VO2
max without influencing maximal exercise time at the intensity associated with VO2
max in triathletes. Int J Sports Med 27: 798–803, 2006.
6. Caputo F, Denadai BS. The highest intensity and the shortest duration permitting attainment of maximal oxygen uptake during cycling: Effects of different methods and aerobic fitness level. Eur J Appl Physiol 103: 47–57, 2008.
7. Carter H, Jones AM, Barstow TJ, Burnley M, Williams CA, Doust JH. Oxygen uptake kinetics
in treadmill running and cycle ergometry: A comparison. J Appl Physiol 89: 899–907, 2000.
8. Carter H, Jones AM, Maxwell NS, Doust JH. The effect of interdian and diurnal variation on oxygen uptake kinetics
during treadmill running. J Sports Sci 20: 901–909, 2002.
9. Currell K, Jeukendrup AE. Validity, reliability and sensitivity of measures of sporting performance. Sports Med 38: 297–316, 2008.
10. Fleiss JL, Levin B, Paik M. Statistical Methods for Rates and Proportions. New York, NY: John Wiley and Sons, 2003.
11. Gaesser GA, Poole DC. The slow component
of oxygen uptake kinetics
in humans. Exerc Sports Sci Rev 24: 35–71, 1996.
12. Hill DW, Poole DC, Smith JC. The relationship between power and time to achieve VO2
max. Med Sci Sports Exerc 34: 709–714, 2002.
13. Howley ET, Bassett DR Jr, Welch HG. Criteria for maximal oxygen uptake: Review and commentary. Med Sci Sports Exerc 27: 1292–1301, 1995.
14. Hopkins W. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
15. Jones AM, Wilkerson DP, DiMenna F, Fulford J, Poole DC. Muscle metabolic responses to exercise above and below the “critical power” assessed using 31P-MRS. Am J Physiol Regul Integr Comp Physiol 294: R585–R593, 2008.
16. Jones M, Vanhatalo A, Burnley M, Morton RH, Poole DC. Critical power: Implications for determination of VO2
max and exercise tolerance. Med Sci Sports Exerc 42: 1876–1890, 2010.
17. Lamarra N, Whipp BJ, Ward SA, Wasserman K. Effect of interbreath fluctuations on characterizing exercise gas exchange kinetics. J Appl Physiol 62: 2003–2012, 1987.
18. Laursen PB, Shing CM, Jenkins DG. Reproducibility of the cycling time to exhaustion
peak in highly trained cyclists. Can J Appl Physiol 28: 605–615, 2003.
19. Machado FA, Guglielmo LG, Greco CC, Denadai BS. Effects of exercise mode on the oxygen uptake kinetic response to severe-intensity exercise in prepubertal children. Pediatr Exerc Sci 21: 159–170, 2009.
20. Murgatroyd SR, Ferguson C, Ward SA, Whipp BJ, Rossiter HB. Pulmonary O2
uptake kinetics as a determinant of high-intensity exercise tolerance in humans. J Appl Physiol 110: 1598–1606, 2011.
21. Ocel JV, Miller LE, Pierson LM, Wootten DF, Hawkins BJ, Myers J, Herbert WG. Adaptation of pulmonary oxygen consumption slow component
following 6 weeks of exercise training above and below the lactate threshold in untrained men. Chest 124: 2377–2383, 2003.
22. Ozyener F, Rossiter HB, Ward SA, Whipp BJ. Influence of exercise intensity on the on- and off-transient kinetics of pulmonary oxygen uptake in humans. J Physiol 533: 891–902, 2001.
23. Poole DC, Ward SA, Gardner GW, Whipp BJ. Metabolic and respiratory profile of the upper limit for prolonged exercise in man. Ergonomics 31: 1265–1279, 1988.
24. Whipp BJ, Rossiter H. The kinetics of oxygen uptake: Physiological inferences from the parameters. Chapter 3. In: Oxygen Uptake Kinetics
in Sport, Exercise and Medicine (1st ed.). Jones A.M., Poole D.C., eds. London, United Kingdom: Routledge, 2005.