Abstract
There is an optimal load and corresponding velocity at which peak power output occurs. It is reasonable to expect that these conditions will change as a result of fatigue during 30 s of all-out cycling. This study evaluated optimal velocity after 30 s of maximal isokinetic cycle ergometer exercise and tested the hypothesis that progressive adjustment of velocity (optimized) during 30 s of all-out cycling would permit greater short-term work capacity (STWC). Non-fatigued optimal cadence [NFOC, 109.6 (2.5) rpm] was determined for ten males on an SRM ergometer using regression analysis of the torque–angular velocity relation during a 7-s maximal acceleration. Fatigued optimal cadence [73.4 (2.4) rpm] was determined in the same way, immediately after a 30-s isokinetic test at NFOC. A subsequent trial with cadence decreasing in steps from NFOC to a conservative estimate of fatigued optimal cadence [83.9 (2.8) rpm] was completed to see if more work could be done with a more optimal cadence during the test. STWC was not different (P=0.50) between the constant [23,681 (764) J] and optimized [23,679 (708) J] conditions. Another more radical progressive change in cadence with four subjects yielded the same result (no increase in STWC). Extraneous factors apparently contribute more to variability in STWC than differences between constant and adjusted optimization of conditions.
Similar content being viewed by others
Introduction
For short duration athletic events, the ability to generate high peak power output (PPO) is critical, as is the ability to sustain a high average power output and resist fatigue. In cycle ergometry it has been shown that the resistive forces the athlete is working against influence these variables (MacIntosh et al. 2003; Vandewalle et al. 1987). Scores obtained in tests of PPO are influenced by less than optimal conditions (load and cadence), making direct comparisons between athletes inappropriate. Accordingly, a test protocol designed to allow athletes to achieve the best possible scores for both PPO and average power output would be extremely useful. The ideal protocol should maintain optimal conditions for power output throughout the test, but it is not known how this optimal condition changes during such a test.
The torque–angular velocity relationship is linear for both friction-loaded (Vandewalle et al. 1987) and isokinetic cycle ergometry (Baron et al. 1999; Sargeant et al. 1984). Since power is the product of torque and angular velocity, establishing the torque–angular velocity relation allows identification of the optimal condition for PPO (MacIntosh and MacEachern 1997; Vandewalle et al. 1987). The apex of the dome-shaped curve relating power and angular velocity represents this unique condition. It has been shown that the linear torque–angular velocity relationship can be obtained with maximal-effort acceleration (Arsac et al. 1996). In practical terms, this means that individualized optimal testing conditions for PPO can be readily determined and immediately used, without the need for additional laboratory visits or multiple fatiguing efforts. However, the impact of fatigue on these relationships has received limited attention.
The effect of fatigue on maximal torque and angular velocity over four successive sprint efforts was studied by Buttelli et al. (1996). All four brief sprints were completed within 30 s, and evidence of fatigue (decreased torque and angular velocity) was observed following the first effort. The slopes of the torque–angular velocity relationships were reported to be similar over the entire test period, resulting in a downwards and leftwards parallel shift. Evaluation of fatigue has not been done after 30 s of continuous effort. If a similar shift occurred during a 30-s maximal effort trial, the torque–angular velocity relationship would change in the manner presented in Fig. 1. This theoretical parallel shift represents a change in the crank angular velocity that is optimal for PPO from 11.35 rad·s−1 at the start of the test to 8.4 rad·s−1 at the end of the 30 s. The corresponding power output would change from 1,227 W to 552 W if the cadence was kept constant (vertical down-arrow, Fig. 1). However, if optimal conditions could be maintained during the test, the power output at the end of the test should be 630 W (angled down-arrow, Fig. 1).
Clearly, fatigue alters the optimal condition for PPO, and according to the evidence presented by Butelli et al. (1996) these changes occur rapidly upon initiation of intense effort. This information supports the idea that changing conditions during a test should permit greater short-term work capacity (STWC), which is defined here as the total work in 30 s of all-out cycling. This has previously been difficult to investigate due to the inability to instantly measure and precisely control resistance settings using friction-loaded ergometers. However, new isokinetic ergometers supported by computer software should permit rapid adjustment of parameters, allowing changes to be made during the course of a test.
The purpose of this study was twofold: (1) to evaluate fatigue after a maximal isokinetic cycle ergometer test lasting 30 s, and (2) to determine if changing the conditions during a maximal isokinetic cycle ergometer test (to account for fatigue) affects total work performed in 30 s. We hypothesized that the torque–angular velocity relation generated in a fatigued condition would exhibit a downwards and leftwards parallel shift, similar to that reported by Butelli et al. (1996). It was also hypothesized that changing the cadence towards the fatigued optimal cadence during the last 20 s of the test would result in greater STWC.
Methods
Subjects
Ten males volunteered to be tested on four occasions over a 4-week period. Additional tests were done with four subjects. The subjects were regular participants in competitive sports, including road and track cycling, and were familiar with intense exercise. Subjects did not have any previous experience with maximal effort testing on an isokinetic cycle ergometer. All subjects gave informed consent prior to their participation. The university Research Ethics Board approved the procedures. All subjects maintained their habitual level of activity during the study, but were required to avoid vigorous exercise in the 24 h preceding any test. The subjects were asked to report to the laboratory well rested, well hydrated, and free from injury or illness. They were also asked not to consume food for 2 h prior to testing, and to refrain from consuming caffeine for 12 h and alcohol for 24 h prior to testing.
Measurements
All tests were performed on an SRM ergometer (Schoberer Rad Messtechnik, Jülich, Germany). The ergometer was operated in either ‘open-ended’ (non-isokinetic) or isokinetic mode. In isokinetic mode, an upper limit for cadence was set and controlled by the system software via an eddy current brake in the ergometer, and subjects could freely accelerate to this cadence. Torque was measured directly at the cranks by a system of eight strain gauges housed within the crankset. Power output was calculated by the system software, based on the equation: power (watts) = torque (newton meters) × angular velocity (radians per second). Power and cadence data were recorded every 0.5 s, and represented an average of the previous complete crank revolution. From the power and cadence data, torque and angular velocity at the cranks were calculated using Excel (Microsoft, Redmond, Wash.), where angular velocity ( rad·s−1) = cadence (revolutions per minute) × 0.1047 ( rad·s−1·rpm−1); and torque (N·m) = power (watts) × angular velocity−1. These values were used for subsequent regression analysis.
Procedures
The following procedures were applied in all laboratory visits. The handlebar and saddle positions of the ergometer were adjusted for each individual, and identical positioning was used for all tests for a given individual. Feet were fixed to the pedals, either with standard cycling shoes and clipless pedals, or running shoes with toe-strap pedals. The crank length was 170 mm. Subjects completed a 10-min warm-up before each test, at a power output not exceeding 150 W. The warm-up included three short sprint efforts at 7, 8, and 9 min. Three minutes separated the warm-up and initiation of the test. During each test, subjects remained seated to standardize body position. Subjects were instructed to give maximal effort and avoid pacing, and were given vigorous verbal encouragement during all tests. A minimum of 48 h separated laboratory visits.
Testing session 1
The purpose of the first testing session was to establish an estimate of the non-fatigued (NFOC) and fatigued (FOC) optimal cadence, and to allow the subjects to practice the 30-s isokinetic constant cadence test. The results generated during this 30-s trial were not used in subsequent data analysis.
The test to determine NFOC began with a controlled rolling start (1 min at 50 rpm with low effect), followed by an all-out effort lasting approximately 7 s with the ergometer set in open-ended mode. Subjects were instructed to accelerate to maximum velocity as quickly as possible. NFOC was calculated by linear regression of the torque and angular velocity data obtained from this all-out effort (MacIntosh et al. 2003). Optimal angular velocity was converted to revolutions of the crank per minute (rpm) to represent optimal cadence. A 10-min active recovery period followed the acceleration test.
The constant cadence isokinetic test and fatigued acceleration were then performed. The test began with a controlled start (1 min at 50 rpm with low effect). At the end of this minute, the subject accelerated to the prescribed cadence (NFOC) and continued pedaling as hard as possible for 30 s. Immediately after the 30 s, the programmed cadence slowed to 50 rpm for 5 s, during which time effort was decreased. A second 7-s acceleration was then performed. This portion of the protocol was programmed with a high isokinetic velocity, to allow the subject to freely accelerate from 50 rpm to maximum velocity, duplicating the conditions of the non-fatigued acceleration trial.
From the second 7-s acceleration, the FOC was determined using linear regression. For several subjects, this trial did not intersect the final point of the 30-s trial, suggesting that subjects may not have given maximal effort during this acceleration. Because inappropriate effort was suspected, a conservative estimate for fatigued optimal cadence (e-FOC) was used for the optimized isokinetic protocol. To determine the e-FOC, optimal angular velocity was re-calculated using the same slope as the non-fatigued effort for a line passing through the point of the fatigued acceleration data that gave the highest intercept on the angular velocity axis (y axis). This decision was based on the assumption that for at least one crank revolution, the subject had been capable of true maximal effort. The resulting regression line passed very close to the final point of the 30-s isokinetic test (see Fig. 2) for all subjects.
Testing session 2
The purpose of the second testing session was to allow subjects a practice trial of the optimized isokinetic protocol with cadence changing throughout the test. These results were not used for subsequent data analysis. Subjects were not aware that results of test 1 and test 2 were not to be used in the final analysis.
The optimized test began with a controlled rolling start (1 min at 50 rpm with low effort), followed by 30 s of all-out effort. The cadence was programmed at the NFOC for the first 10 s of the test. During the last 20 s of the test, the cadence was decreased every 5 s towards the e-FOC. The cadence decreased in steps corresponding to 20% of the difference between the NFOC and e-FOC values, assuming a gradual change in the true optimal condition.
Testing session 3
The third test was a 30-s isokinetic protocol at NFOC. The results of this test were used to quantify STWC for a test at constant cadence.
Testing session 4
The 30-s optimized cadence protocol was repeated during the fourth testing session. The results of this test were compared with the constant cadence test to determine if adjusting the cadence during the test permitted greater STWC.
Additional testing
It became clear that the optimized test did not permit greater total work in 30 s of all-out cycling. To evaluate the possibility that this result was due to our conservative estimate of the FOC, we did additional tests with a greater change in cadence.
Subjects participating in the additional tests did two 30-s trials, separated by at least 48 h. The first (isokinetic) test was immediately followed by an acceleration to obtain the FOC. Cadence was decreased every second from 10 s to the end of the test during the subsequent 30-s optimized trial. The calculated FOC was used as an end-point for this more radical optimized protocol. The cadence was decreased every second rather than every 5 s, to eliminate sudden changes in cadence. This was expected to improve subject comfort. Finally, to account for limitations with the SRM software, adjustments were made in the optimized protocol program to ensure that the cadence at the end of the 30 s was as close as possible to the FOC value. This meant overestimating the decrease in cadence to overcome the delay that is apparently built into the SRM software. All other aspects of the methods remained the same.
Statistics
All calculations and statistics were done with Excel. All mean values are presented with standard error of the mean (SEM). Linear regression was performed on all torque–angular velocity plots in order to determine optimal cadence. Paired t-tests were used to determine if a difference existed between the non-fatigued and fatigued optimal cadence, PPO, and slopes of the regression lines. Mean power output was calculated for the first 10 s and the last 20 s of the 30-s tests. Paired t-tests were used to determine if there was a significant difference in the mean power and STWC attained using the constant cadence and optimized protocols. Differences were considered significant for P<0.05.
Results
The subjects who volunteered for this study were active healthy males, 179 (2) cm tall, 81.3 (3.0) kg in mass, and 26.5 (1.2) years of age.
For all isokinetic tests, cadence was maintained to within 2% of the programmed value, with a tendency for the resultant cadence to be slightly higher than the set cadence. There was a 2-s delay in reaching the prescribed cadence. These results appear to be inherent errors in the SRM software.
Non-fatigued and fatigued optimal cadence
All torque–angular velocity relations were linear with a negative slope. The NFOC was 109.6 (2.5) rpm (range 96.3–123.4 rpm), and the FOC was 73.4 (2.4) rpm (range 60.7–83.1 rpm). There was a 33% decrease (P<0.001) between the NFOC and FOC. An example of the non-fatigued and fatigued torque-angular velocity relationships for one subject is presented in Fig. 2. Individual values for optimal cadence are presented in Table 1.
The average slopes of the regression lines for the two 7-s trials were −0.1129 (0.01) (non-fatigued) and −0.0922 (0.01) (fatigued); these values were significantly different (P<0.05). Although a downward and leftward shift did occur, it was not parallel. Therefore, the first hypothesis was rejected. The change in slope may have been caused by either a real change in the muscle properties or by unwillingness to produce the required effort. In order to reduce the effect of inconsistent subject effort, a conservative estimate of the fatigued optimal cadence (e-FOC) was used for the initial optimized 30-s protocols.
The PPO attained for a single crank revolution during the non-fatigued 7-s effort was 1,227.5 (48.6) W [15.7 (0.51) W·kg−1]. This was significantly different (P<0.001) and 45% greater than the PPO attained for a single crank revolution during the fatigued 7-s trial [676.5 (26.0) W; 8.65 (0.32) W·kg−1].
STWC was not significantly different (P=0.50) between the constant and optimized protocols [23,681 (764) J and 23,679 (708) J, respectively]. Furthermore, no significant difference (P=0.21) was found between the mean power output during the first 10 s of the tests [905 (32) W constant, 917 (34) W optimized]. This result was as expected, since the controlled cadence was identical during the first 10 s of all trials. However, no significant difference (P=0.27) was detected between the mean power output during the final 20-s portion of the two trials [734 (23) W constant, 728 (19) W optimized]. It was expected that the values for the optimized protocol would be greater, since the cadence was changing towards the e-FOC during that portion of the protocol. Figure 3 shows STWC and mean power output (first 10 s and last 20 s) for each protocol for each subject. Variability is clearly evident, even during the first 10 s of the test where it was expected that results would be consistent because conditions were identical. The relationship between the STWC values from the constant cadence and optimized protocols is presented in Fig. 4.
Due to our conservative estimates for protocol optimization, the cadence at the end of the 30-s optimized test was always greater than the FOC. The potential impact of this limitation was subsequently tested using four additional subjects.
The results from the supplementary testing were consistent with the previously reported results. No significant differences (P>0.15) were seen in STWC and average power during the first 10 s and final 20 s of the tests. STWC values were 26,516 (316) J (constant cadence) and 27,146 (1,111) J (optimized). Mean power over the first 10 s of the tests was slightly higher in the optimized condition [1,090 (11) W] as compared to the constant cadence condition [1,048 (35) W]. The same trend was observed over the last 20 s of the test [815 (25) W optimized, 804 (11) W constant]. Although the STWC tended to be greater in the optimized condition, a substantial part of the increase could be attributed to differences within the first 10 s of the test when conditions were the same between the two protocols. These data strengthen the decision to reject the second hypothesis.
Discussion
The purposes of this study were to evaluate fatigue after a maximal isokinetic cycle ergometer test lasting 30 s, and to determine whether changing the conditions during a maximal effort cycle ergometer test (to account for fatigue) affects total work performed in 30 s. Torque–angular velocity relationships were used to estimate optimal cadence in both non-fatigued and fatigued states. This information was used to develop two 30-s cycling protocols (constant cadence and optimized cadence), and STWC was compared between the two trials. To our knowledge, this is the first study that has examined the effects of changing conditions during an isokinetic cycle ergometer test protocol.
The key findings of this study were: (1) there was a non-parallel shift in the torque–angular velocity relationship, resulting in a substantial decrease in optimal velocity during the 30 s test, and (2) there was no difference in STWC between the test conditions. The decrease in optimal velocity was not surprising, since it has been demonstrated that the torque–angular velocity properties for cycling change in this direction with fatigue from repeated brief sprints (Buttelli et al. 1996). However, the absence of a significant difference in STWC values between the two test conditions was not expected. Possible factors contributing to this lack of difference in STWC include: counteracting differences in fatigue, counteracting differences in energetics, inappropriate optimization, and day-to-day variation.
Although a substantial decrease in optimal angular velocity was apparent between the NFOC and FOC, it is possible that a portion of this difference may be attributed to subject effort (central fatigue) rather than muscle fatigue induced by the 30-s trial. The slopes of the regression lines of the non-fatigued and fatigued torque–angular velocity relationships were significantly different. Assuming the non-fatigued 7-s acceleration was representative of a ‘best effort’, this finding could be attributed to either a real change in the slope, or to less than maximal effort in performing the acceleration after the 30-s test. Buttelli et al. (1996) reported a parallel shift of the torque-angular velocity relationship during a series of four sprint efforts lasting 30 s in total. However, they did not report values for the slope, and did not apparently evaluate the slope change with statistics. It is possible that the protocol differences in this study (a longer duration effort on an isokinetic ergometer) may have caused a real change in the slope of the torque–angular velocity relationship, making it impossible for subjects to achieve the expected level during the fatigued acceleration. Alternatively, it is possible that Buttelli et al. (1996) actually did have a small change in slope, similar to us. The impact of the change in slope is relatively minor in comparison with the leftward shift in this torque–angular velocity relationship.
Further insight into effort and fatigue was gained by examination of the point at which the actual and estimated fatigued torque–angular velocity line intersected the torque–angular velocity values from the constant cadence isokinetic test. In only one case was the fatigued torque–angular velocity line to the right of the final value from the 30-s test. In this single case, either recovery had occurred during the 5-s wait for deceleration of the ergometer or less than maximal effort was given during the 30-s test. In four cases, subjects did not accelerate to the expected level coinciding with the final value from the 30-s test and the fatigued torque–angular velocity line was to the left of the final value from the 30-s isokinetic test, as shown in the example in Fig. 5. This may indicate insufficient effort during the 7-s acceleration, or additional fatigue resulting from the effort of the acceleration. Finally, five subjects produced fatigued torque–angular velocity lines that passed very close to the final point of the 30-s test. An example of this is shown in Fig. 2. However, despite this consistent effort, the optimized STWC of these five subjects was not consistently greater than the STWC during the constant cadence protocol.
The theory developed in the introduction of this paper gives compelling reason to expect greater STWC in the optimized test than in the isokinetic test. The theory relies on the assumption that load does not affect the energy cost of the task, but this may not be a valid assumption. If additional energy is used under the optimal conditions, then greater fatigue would be expected. This issue would tend to counteract any benefit of optimizing the condition for power output. The fact that the final value of the optimized trial falls closer to the fatigued acceleration values than the final value of the isokinetic trial in Fig. 5 is consistent with this notion. Another factor that may have prevented us from seeing an effect of optimizing cadence in these trials was that we changed the cadence in a linear fashion after 10 s in both the conservative and the radical approaches. There may be a more appropriate pattern of change that would result in improved STWC.
A comparison of 30-s isokinetic cycle ergometer protocols was previously done by McCartney et al. (1983). No significant difference in average power output over 30 s was found between isokinetic crank velocities of 60, 100, and 140 rpm. Although initial power output was greater at higher crank velocities, the rate and extent of decline was also greater. This is consistent with a higher energy cost of the optimized condition. Based on these findings, slowing the programmed cadence during a 30-s isokinetic cycle ergometer test would seem like a reasonable method of optimizing the amount of work performed during the test, and a difference in STWC should have been evident between the constant cadence and optimized conditions. However, the energetics of this condition are unknown.
Lack of familiarity with all-out pedaling on an isokinetic ergometer may have had an impact on performance during the remainder of the 30-s tests and the resulting STWC values. The sensation of not being able to accelerate the cranks during the tests was somewhat unfamiliar to the subjects, and very different than the sensation experienced while performing a similar test on a friction-loaded ergometer. The subjects were given two practice trials (first and second laboratory visits), and were also instructed that power output was calculated from torque measured at the cranks and they were to ‘push as hard as possible’ despite not being able to accelerate the cranks beyond the programmed level. In spite of these precautions, neither the conservative estimate of optimal conditions nor the more radical adjustment resulted in enhanced work.
It may be argued that we underestimated the change in optimal conditions for our optimized trial. We recognized this possibility and did additional trials with a more radical change in cadence. This still did not result in increased STWC, supporting the notion that optimization does not result in greater STWC.
Insufficient or inconsistent effort during the 30-s trials is a compelling explanation for the variation in STWC results. The torque and angular velocity values generated from the first 7-s effort were consistently greater than those generated during the acceleration phase of the 30-s tests. In eight out of ten subjects this line did not coincide with any data points from the 30-s efforts (see Figs. 2 and 5). Clearly, more effort was given during the non-fatigued 7-s acceleration than during any other test.
The maximum effort that subjects were capable of achieving may have been substantially different on a random basis from day to day. Evidence of this was seen in the first 10 s of the 30-s tests (Fig. 3b). The programmed cadence was identical for the first 10 s of each 30-s protocol, so the mean power output should have also been identical. For all but one subject, substantial variation was shown. Additional variation was present in the final 20 s of the test. For example, subject no. 8 had a higher STWC with the optimized protocol, but did less work during the last 20 s of the optimized protocol compared to the constant protocol. In this case, the higher optimized STWC resulted from extra work done in the first 10 s of the optimized test.
In this study, imposing optimal conditions for PPO for an entire 30-s test did not enhance STWC. No difference in STWC was apparent between the constant cadence and optimized protocols. From a theoretical point of view, it appears that STWC should be increased by a properly optimized protocol. However, in practical terms, the variability between tests due to variations in subject effort and normal day-to-day differences is greater than the variability (improvement) that could be attributed to optimizing the protocol. Furthermore, there is a possibility that increased energy cost and therefore greater fatigue results during such an optimized protocol.
References
Arsac LM, Belli A, Lacour JR (1996) Muscle function during brief maximal exercise: Accurate measurements on a friction-loaded cycle ergometer. Eur J Appl Physiol 74:100–106
Baron R, Bachl N, Petschnig R, Tschan H, Smekal G, Pokan R (1999) Measurement of maximal power output in isokinetic and non-isokinetic cycling. A comparison of two methods. Int J Sports Med 20:532–537
Buttelli O, Seck D, Vandewalle H, Jouanin JC, Monod H (1996) Effect of fatigue on maximal velocity and maximal torque during short exhausting cycling. Eur J Appl Physiol 73:175–179
MacIntosh BR, MacEachern P (1997) Paced effort and all-out 30 s power tests. Int J Sports Med 18:594–599
MacIntosh BR, Rishaug P, Svedahl K (2003) Assessment of peak power and short-term work capacity. Eur J Appl Physiol 88:572–579
McCartney N, Heigenhauser GJF, Jones NL (1983) Power output and fatigue of human muscle in maximal cycling exercise. J Appl Physiol 55:218–224
Sargeant AJ, Dolan P, Young A (1984) Optimal velocity for maximal short-term (anaerobic) power output in cycling. Int J Sports Med 5:124–125
Vandewalle H, Peres G, Heller J, Panel J, Monod H (1987) Force–velocity relationship and maximal power on a cycle ergometer: Correlation with the height of a vertical jump. Eur J Appl Physiol 56:650–656
Acknowledgement
This work was supported by the Natural Science and Engineering Research Council (Canada).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
MacIntosh, B.R., Svedahl, K. & Kim, M. Fatigue and optimal conditions for short-term work capacity. Eur J Appl Physiol 92, 369–375 (2004). https://doi.org/10.1007/s00421-004-1177-3
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00421-004-1177-3