International Journal of Sports Physiology and Performance, 2015, 10, 278-284 http://dx.doi.org/10.1123/ijspp.2014-0013 © 2015 Human Kinetics, Inc.
www.IJSPP-Journal.com ORIGINAL INVESTIGATION
Reliability and Validity of a New Variable-Power Performance Test in Road Cyclists Avish P. Sharma, Adrian D. Elliott, and David J. Bentley Context: Road cycle racing is characterized by significant variability in exercise intensity. Existing protocols attempting to model this aspect display inadequate variation in power output. Furthermore, the reliability of protocols representative of road cycle racing is not well known. There are also minimal data regarding the physiological parameters that best predict performance during variable-power cycling. Purpose: To determine the reliability of mean power output during a new test of variable-power cycling and establish the relationship between physiological attributes typically measured during an incremental exercise test and performance during the variable-power cycling test (VCT). Methods: Fifteen trained male cyclists (mean ± SD age 33 ± 6.5 y, VO2max 57.9 ± 4.8 mL · kg–1 · min–1) performed an incremental exercise test to exhaustion for determination of physiological attributes, 2 VCTs (plus familiarization), and a 30-km time trial. The VCT was modeled on data from elite men’s road racing and included significant variation in power output. Results: Mean power output during the VCT showed good reliability (r = .92, CV% = 1.98). Relative power during the self-paced sections of the VCT was most correlated with maximal aerobic power (r = .79) and power at the second ventilatory threshold (r = .69). Blood lactate concentration showed poor reliability between trials (CV% = 13.93%). Conclusions: This study has demonstrated a new reliable protocol simulating the stochastic nature of road cycling races. Further research is needed to determine which factors predict performance during variable-power cycling and the validity of the test in monitoring longitudinal changes in cycling performance. Keywords: endurance, athletes, reproducibility, testing, physiology Reliability in tests of sports performance is crucial in quantifying the effects of various interventions that may serve to alter training and competitive outcomes.1 For elite athletes, small changes in performance can be the difference between winning and losing,2 and as such, testing needs to have the precision to detect small but meaningful changes in performance over time.3 In a research setting, highly reliable testing protocols allow sports scientists to determine with greater confidence the effects of training or nutrition on performance and ensure that any resulting changes are real and not the result of measurement or biological error.4 A distinguishing characteristic of road racing is variability in the intensity of exercise,5 with periods of lower-intensity exercise frequently interspersed with shorter periods ( peak power output
60
25
~70
Table 1. The presence of repeated high-intensity efforts alleviated the absence of supramaximal efforts in previous studies, which largely focused on varying power output at controlled, submaximal levels. The only information available to participants was actual relative power output (W/kg), the sequence of sections in each lap, and how many laps were remaining in the test. During all tests, heart rate was recorded continuously via radio telemetry (Polar Electro Oy, Kempele, Finland). The participants’ respiratory exchange was measured at regular intervals (Figure 1) throughout the test by indirect calorimetry as previously described for determination of VO2. Economy was established by the ratio of power output (W) to VO2 (W · L–1 · min–1). Blood lactate concentration was measured at 3 points during the test (Figure 1) from a small (50-μL) capillary blood sample using a portable lactate analyzer (Lactate Scout, SensLab GmbH, Germany) whose reliability has been previously established.23 Participants’ perceived exertion was measured each lap (Figure 1) using the Borg rating scale of 6 to 20.24 30-km Time Trial. After a standardized warm-up (Figure 1), participants performed a 30-km time trial that they were instructed to complete as quickly as possible. Cadence and resistance during this test were self-selected, and participants were provided with their speed, power output, average power output, cadence, and heart rate during the test.
power outputs were not significantly different between trials and showed high correlation (both r = .92, P < .001) and low random error (CV% < 2). Calculation of the least significant change revealed that the VCT could detect a 3.9% change in mean self-paced power with 95% confidence based on the CV observed over 2 trials. There was no significant difference for the other performance variables determined test and retest. Mean economy was also not significantly different between trials and shown to have good reliability (r = .83, CV% = 2.05) with the ability to detect a 4% change with 95% confidence. The highest CV was found in blood lactate concentration (CV% = 13.93), while the lowest Pearson correlation was found for rating of perceived exertion between trials (r = .76).
Relationship With Physiological and Performance Variables The correlation coefficients between variables measured during the incremental test and the VCT are shown in Figure 3. Moderate correlations existed between VO2max, VT2, and PPO and mean self-paced power during the VCT (r = .53, .69, and .79, P = .04, .004, and .0004, respectively). PPO was significantly correlated with mean power during the 30-km time trial (r = .88, P = .00002).
Discussion
Statistical Analyses Data were analyzed using an Excel spreadsheet for reliability developed by Hopkins25 and used to determine Pearson product–moment correlation coefficients with 95% confidence intervals, as well as coefficient of variation (CV, %) between test and retest data. The least significant change with 95% confidence was determined per variable using the equation
(
)
1.96 CV / n × 2 where n = number of trial replicates. Regression analysis/Pearson product–moment correlations were performed to determine relationships between variables from the incremental cycling test (PPO, VT2W, and VO2max) and those from the first VCT (mean self-paced power W/kg) and 30-km time trial (mean power W/kg).
Results Reliability of the VCT The mean ± SD for each variable measured during the VCT are presented in Table 2. Total mean self-paced power and overall
The aim of the study was to determine the reliability of mean power output during a new test of variable-power cycling. A secondary aim was to establish whether a relationship existed between power during the VCT and physiological attributes measured during an incremental exercise test. The results suggest that the VCT provides a reliable measure of mean power output. Mean self-paced power output was strongly correlated with PPO. The first main finding of the study was the reliability of both self-paced mean power output (r = .92, CV% = 1.98) and total mean power output (r = .92, CV% = 1.24) during the VCT, thus confirming the primary hypothesis. Notably, this study highlights the ability of this test to detect modest (>3.9%) changes in performance with a high level of confidence. While the reliability of variable-power protocols has not been extensively studied, the results of the current study are consistent with those of Abbiss et al,12 who reported a mean CV% of 2.4 of a 30-km dynamic time trial for 2 trials after a single familiarization. While a familiarization identical to the subsequent trials of the current study would have been ideal, in this instance a single familiarization was sufficient to ensure high test–retest reliability of variable-power cycling tests conducted within a week— however, we did not investigate the effect multiple familiarizations may have had on the reliability of the VCT. Despite this limitation,
282
14.32 (1.55)
Rating of perceived exertion (Borg)
14.41 (1.78)
6.71 (2.55)
79.99 (5.63)
6.66 (0.78)
1.74 (0.23)
3.22 (0.16)
3.03 (0.24)
Trial 2
0.09 (1.18)
–0.73 (1.70)
–0.72 (3.19)
0.16 (0.36)
–0.06 (0.13)
0.01 (0.07)
0.01 (0.10)
Mean difference
.76* (.4–.91)
.83** (.56–.94)
.85** (.59–.95)
.89** (.69–.96)
.88** (.66–.96)
.92** (.77–.97)
.92** (.76–.97)
r
.78 (.45–.92)
.85 (.6–.95)
.87 (.65–.96)
.9 (.74–.97)
.89 (.69–.96)
.93 (.81–.98)
.93 (.8–.97)
ICC
1.2 (0.88–1.9)
0.84 (0.61–1.32)
2.25 (1.63–3.63)
0.26 (0.19–0.4)
0.05 (0.04–0.08)
0.09 (0.07–0.14)
0.07 (0.05–0.11)
TEM
4.75 (3.37)
13.93 (9.81)
2.05 (1.79)
3.38 (2.37)
4.49 (3.09)
1.24 (0.75)
1.98 (1.15)
CV (%)
9.3%
27.3%
4%
6.6%
8.8%
2.4%
3.9%
LSC (95%)
Abbreviations: ICC, intraclass correlation; TEM, typical error of measurement; CV, coefficient of variation; LSC, least significant change. Note: The trial data and mean difference are expressed as mean (SD), Pearson r values are expressed as mean (95% confidence intervals), and CV is expressed as mean (SD). LSC calculated as 1.96(CV / n ) × 2 , where n = number of trial replicates. *P < .001. **P < .0001.
7.44 (2.98)
80.71(5.87)
Blood lactate (mmol/L)
·
6.4 (0.74)
Total hard + sprints mean power (W/kg)
Economy at 3.5 W/kg (W ·
1.8 (0.26)
Total recovery mean power (W/kg)
min–1)
3.21 (0.17)
Total mean power (W/kg)
L–1
3.03 (0.25)
Total self-paced mean power (W/kg)
Trial 1
Table 2 Physiological and Performance Variables Measured During Test and Retest of the Variable Cycle Test, N = 15
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Reliability of a New Variable-Power Cycle Test 283
the high reliability can also be attributed to subjects’ participation in local cycling competitions that were in criterium/circuit style—as such, participants were already familiar with the nature and demands of the VCT. Power output during “recovery” and “hard + sprints” sections showed increasing variability compared with the entire trial (Table 2), with mean CVs of 4.49 and 3.38, respectively. This is in agreement with previous findings that report that the CV of power output tends to decrease as intensity increases.13,26 Rating of perceived exertion displayed the weakest correlation between trials (Table 2). Blood lactate concentration showed the highest variability of all measures between trials (Table 2), a finding in accordance with those of Swart and Jennings,27 who showed large fluctuation in finger-prick lactate measurements. Mechanistically, this reflects the variability in lactate production and subsequent diffusion from the skeletal muscle to blood.28 This finding shows that finger-prick lactate measurements are unreliable, and extreme caution should be taken when using this method. The second main finding of the study was the moderate to strong correlations between self-paced mean power during the VCT and performance parameters determined during the incremental exercise test (Figure 3). The results are similar to those presented by Levin et al,13 who reported correlations of .83 and .74 for PPO and VT2, respectively, with power during a 30-km time trial containing 250-m and 1-km sections of intense effort, respectively completed at 127% and 105% of PPO. While the aforementioned high-intensity efforts are applicable to road racing, they do not mirror the shortduration, supramaximal efforts frequently seen during mass-start road stages and criteriums in cycling.6 Thus, the inclusion of these repeated high-intensity efforts in the VCT addresses this limitation of previous protocols. Despite the moderate strength of correlations with these performance variables, future studies are required to establish the longitudinal changes in the results of the VCT and how they relate to both field and test performance in well-trained or elite athletes. PPO showed a stronger correlation with mean power during the 30-km time trial (r = .88), a finding consistent with data from previous studies. Bentley et al14 and Lamberts et al15 reported very strong correlation (r = .91 and .90, respectively) between PPO and time trials of 90 minutes and 40-km duration. This suggests that PPO is not as valid a predictor of variable-power cycling performance as it is for time-trialing and, further, indicates that factors other than those also relevant to time-trial influence performance during variable-power cycling. A significant proportion of the VCT was conducted at intensity greater than lactate threshold, above which point the body increases the amount of energy provided by anaerobic metabolism.16,22 Thus, physiological variables associated with fatigue resistance may be highly relevant to performance during variable-power cycling, including muscle-buffering capacity19,29 and clearance of lactate and other metabolites.16,28 Future studies should be directed toward identifying the physiological characteristics most associated with variable-power cycling performance and the development of a prediction model for performance during such events.
Practical Applications The results of this study suggest that tests simulating the stochastic nature of road cycle racing can be conducted in a reliable way. As such, future studies examining the effect of interventions on performance during road cycling may make use of variable-power cycling tests.
Figure 3 — The correlation between mean self-paced power during the Variable Cycle Test (VCT) and performance parameters from an incremental exercise test. Abbreviations: PPO, peak power output; VT2, second ventilatory threshold; VO2max, maximal oxygen uptake.
284 Sharma, Elliott, and Bentley
In physiological performance diagnostics, the VCT is a race-specific protocol that allows for the assessment of cyclingperformance aspects that cannot be quantified by incremental or time-trial protocols, such as repeat-sprint ability and recovery capacity of cyclists. The VCT can be used as training tool to simulate the demands of criterium-style road racing.
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Conclusion The VCT provides a reliable measure of power output during variable-intensity cycling and shows moderate correlation with a number of physiological variables relevant to performance during road cycling. The presence of repeat supramaximal efforts in the VCT addresses the limitations of previous variable-power protocols and allows for the assessment of the repeat-sprint ability and recovery capacities of athletes during a race-specific protocol. Future investigations are required to establish the validity of the VCT in monitoring longitudinal changes in performance during the test with different interventions and determine the physiological factors most relevant to stochastic cycling.
References 1. Hopkins WG, Schabort EJ, Hawley JA. Reliability of power in physical performance tests. Sports Med. 2001;31:211–234. PubMed doi:10.2165/00007256-200131030-00005 2. Jeukendrup AE, Craig NP, Hawley JA. The bioenergetics of world class cycling. J Sci Med Sport. 2000;3:414–433. PubMed doi:10.1016/ S1440-2440(00)80008-0 3. Driller MW. The reliability of a 30-minute performance test on a Lode cycle ergometer. J Sci Cycling. 2012;1:21–27. 4. O’Hara JP, Thomas A, Seims A, Cooke C, King R. Reliability of a highintensity endurance cycling test. Int J Sports Med. 2012;33:18–25. PubMed doi:10.1055/s-0031-1284340 5. Vogt S, Schumacher YO, Blum A, et al. Cycling power output produced during flat and mountain stages in the Giro d’Italia: A case study. J Sports Sci. 2007;25:1299–1305. PubMed doi:10.1080/02640410601001632 6. Ebert TR, Martin D, Stephens B, Withers R. Power output during a professional men’s road-cycling tour. Int J Sports Physiol Perform. 2006;1:324–335. PubMed 7. Currell K, Jeukendrup AE. Validity, reliability and sensitivity of measures of sporting performance. Sports Med. 2008;38:297–316. PubMed doi:10.2165/00007256-200838040-00003 8. Abbiss CR, Menaspà P, Villerius V, Martin DT. Distribution of power output when establishing a breakaway in cycling. Int J Sports Physiol Perform. 2013;8:452–455. PubMed 9. Theurel J, Lepers R. Neuromuscular fatigue is greater following highly variable versus constant intensity endurance cycling. Eur J Appl Physiol. 2008;103:461–468. PubMed doi:10.1007/s00421-008-0738-2 10. Palmer GS, Borghouts L, Noakes T, Hawley J. Metabolic and performance responses to constant-load vs. variable-intensity exercise in trained cyclists. J Appl Physiol. 1999;87:1186–1196. PubMed 11. Bernard T, Vercruyssen F, Mazure C, Gorce P, Hausswirth C, Brisswalter J. Constant versus variable-intensity during cycling: effects on subsequent running performance. Eur J Appl Physiol. 2007;99:103–111. PubMed doi:10.1007/s00421-006-0321-7 12. Abbiss CR, Levin G, McGuigan M, Laursen P. Reliability of power output during dynamic cycling. Int J Sports Med. 2008;29:574–578. PubMed doi:10.1055/s-2007-989263
13. Levin GT, Laursen PB, Abbiss CR. Reliability of physiological attributes and their association with stochastic cycling performance. Int J Sports Physiol Perform. 2014;9(2):309–315. 14. Lamberts RP, Lambert MI, Swart J, Noakes TD. Allometric scaling of peak power output accurately predicts time trial performance and maximal oxygen consumption in training cyclists. Br J Sports Med. 2012;46:36–41. PubMed doi:10.1136/bjsm.2010.083071 15. Bentley DJ, McNaughton L, Thompson D, Vleck V, Batterham A. Peak power output, the lactate threshold, and time trial performance in cyclists. Med Sci Sports Exerc. 2001;33:2077–2081. PubMed doi:10.1097/00005768-200112000-00016 16. Coleman D. The physiological demands of road cycling. In: Hopker J, Jobson S, eds. Performance Cycling: The Science of Success. 1st ed. London: Bloomsbury; 2013:13–32. 17. Lucia A, Hoyos J, Pérez M, Santalla A, Earnest CP, Chicharro JL. Which laboratory variable is related with time trial performance time in the Tour de France? Br J Sports Med. 2004;38:636–640. PubMed doi:10.1136/bjsm.2003.008490 18. Bentley DJ, Newell J, Bishop D. Incremental exercise test design and analysis: implications for performance diagnostics in endurance athletes. Sports Med. 2007;37:575–586. PubMed doi:10.2165/00007256200737070-00002 19. Chicharro JL, Hoyos J, Lucía A. Effects of endurance training on the isocapnic buffering and hypocapnic hyperventilation phases in professional cyclists. Br J Sports Med. 2000;34:450–455. PubMed doi:10.1136/bjsm.34.6.450 20. Hopker J, Myers S, Jobson S, Bruce W, Passfield L. Validity and reliability of the Wattbike cycle ergometer. Int J Sports Med. 2010;31:731–736. PubMed doi:10.1055/s-0030-1261968 21. McLaughlin JE, King GA, Howley ET, Basset DR, Ainsworth BE, Jr. Validation of Cosmed K4b2 portable metabolic system. Int J Sports Med. 2001;22:280–284. PubMed doi:10.1055/s-2001-13816 22. Lucía A, Hoyos J, Pérez M, Chicharro J. Heart rate and performance parameters in elite cyclists: a longitudinal study. Med Sci Sports Exerc. 2000;32:1777–1782. PubMed doi:10.1097/00005768-20001000000018 23. Tanner RK, Fuller K, Ross M. Evaluation of three portable blood lactate analysers: Lactate Pro, Lactate Scout and Lactate Plus. Eur J Appl Physiol. 2010;109:551–559. PubMed doi:10.1007/s00421-0101379-9 24. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2:92–98. PubMed 25. Hopkins WG. Reliability from consecutive pairs of trials [Excel spreadsheet]. In: Internet Society for Sport Science, A New View of Statistics. 2000. sportsci.org/resource/stats/xrely.xls. 26. Lamberts RJ, Swart J, Noakes T, Lambert M. A novel submaximal cycle test to monitor fatigue and predict cycling performance. Br J Sports Med. 2011;45:797–804. PubMed doi:10.1136/bjsm.2009.061325 27. Swart J, Jennings CL. Use of blood lactate concentration as a marker of training status. S Afr J Sports Med. 2004;16:3–7. 28. Brooks GA. The lactate shuttle during exercise and recovery. Med Sci Sports Exerc. 1986;18:360–368. PubMed doi:10.1249/00005768198606000-00019 29. Weston AR, Myburgh KH, Lindsay FH, Dennis SC, Noakes TD, Hawley JA. Skeletal muscle buffering capacity and endurance performance after high-intensity interval training by well-trained cyclists. Eur J Appl Physiol Occup Physiol. 1997;75:7–13. PubMed doi:10.1007/s004210050119
www.IJSPP-Journal.com ORIGINAL INVESTIGATION
Reliability and Validity of a New Variable-Power Performance Test in Road Cyclists Avish P. Sharma, Adrian D. Elliott, and David J. Bentley Context: Road cycle racing is characterized by significant variability in exercise intensity. Existing protocols attempting to model this aspect display inadequate variation in power output. Furthermore, the reliability of protocols representative of road cycle racing is not well known. There are also minimal data regarding the physiological parameters that best predict performance during variable-power cycling. Purpose: To determine the reliability of mean power output during a new test of variable-power cycling and establish the relationship between physiological attributes typically measured during an incremental exercise test and performance during the variable-power cycling test (VCT). Methods: Fifteen trained male cyclists (mean ± SD age 33 ± 6.5 y, VO2max 57.9 ± 4.8 mL · kg–1 · min–1) performed an incremental exercise test to exhaustion for determination of physiological attributes, 2 VCTs (plus familiarization), and a 30-km time trial. The VCT was modeled on data from elite men’s road racing and included significant variation in power output. Results: Mean power output during the VCT showed good reliability (r = .92, CV% = 1.98). Relative power during the self-paced sections of the VCT was most correlated with maximal aerobic power (r = .79) and power at the second ventilatory threshold (r = .69). Blood lactate concentration showed poor reliability between trials (CV% = 13.93%). Conclusions: This study has demonstrated a new reliable protocol simulating the stochastic nature of road cycling races. Further research is needed to determine which factors predict performance during variable-power cycling and the validity of the test in monitoring longitudinal changes in cycling performance. Keywords: endurance, athletes, reproducibility, testing, physiology Reliability in tests of sports performance is crucial in quantifying the effects of various interventions that may serve to alter training and competitive outcomes.1 For elite athletes, small changes in performance can be the difference between winning and losing,2 and as such, testing needs to have the precision to detect small but meaningful changes in performance over time.3 In a research setting, highly reliable testing protocols allow sports scientists to determine with greater confidence the effects of training or nutrition on performance and ensure that any resulting changes are real and not the result of measurement or biological error.4 A distinguishing characteristic of road racing is variability in the intensity of exercise,5 with periods of lower-intensity exercise frequently interspersed with shorter periods ( peak power output
60
25
~70
Table 1. The presence of repeated high-intensity efforts alleviated the absence of supramaximal efforts in previous studies, which largely focused on varying power output at controlled, submaximal levels. The only information available to participants was actual relative power output (W/kg), the sequence of sections in each lap, and how many laps were remaining in the test. During all tests, heart rate was recorded continuously via radio telemetry (Polar Electro Oy, Kempele, Finland). The participants’ respiratory exchange was measured at regular intervals (Figure 1) throughout the test by indirect calorimetry as previously described for determination of VO2. Economy was established by the ratio of power output (W) to VO2 (W · L–1 · min–1). Blood lactate concentration was measured at 3 points during the test (Figure 1) from a small (50-μL) capillary blood sample using a portable lactate analyzer (Lactate Scout, SensLab GmbH, Germany) whose reliability has been previously established.23 Participants’ perceived exertion was measured each lap (Figure 1) using the Borg rating scale of 6 to 20.24 30-km Time Trial. After a standardized warm-up (Figure 1), participants performed a 30-km time trial that they were instructed to complete as quickly as possible. Cadence and resistance during this test were self-selected, and participants were provided with their speed, power output, average power output, cadence, and heart rate during the test.
power outputs were not significantly different between trials and showed high correlation (both r = .92, P < .001) and low random error (CV% < 2). Calculation of the least significant change revealed that the VCT could detect a 3.9% change in mean self-paced power with 95% confidence based on the CV observed over 2 trials. There was no significant difference for the other performance variables determined test and retest. Mean economy was also not significantly different between trials and shown to have good reliability (r = .83, CV% = 2.05) with the ability to detect a 4% change with 95% confidence. The highest CV was found in blood lactate concentration (CV% = 13.93), while the lowest Pearson correlation was found for rating of perceived exertion between trials (r = .76).
Relationship With Physiological and Performance Variables The correlation coefficients between variables measured during the incremental test and the VCT are shown in Figure 3. Moderate correlations existed between VO2max, VT2, and PPO and mean self-paced power during the VCT (r = .53, .69, and .79, P = .04, .004, and .0004, respectively). PPO was significantly correlated with mean power during the 30-km time trial (r = .88, P = .00002).
Discussion
Statistical Analyses Data were analyzed using an Excel spreadsheet for reliability developed by Hopkins25 and used to determine Pearson product–moment correlation coefficients with 95% confidence intervals, as well as coefficient of variation (CV, %) between test and retest data. The least significant change with 95% confidence was determined per variable using the equation
(
)
1.96 CV / n × 2 where n = number of trial replicates. Regression analysis/Pearson product–moment correlations were performed to determine relationships between variables from the incremental cycling test (PPO, VT2W, and VO2max) and those from the first VCT (mean self-paced power W/kg) and 30-km time trial (mean power W/kg).
Results Reliability of the VCT The mean ± SD for each variable measured during the VCT are presented in Table 2. Total mean self-paced power and overall
The aim of the study was to determine the reliability of mean power output during a new test of variable-power cycling. A secondary aim was to establish whether a relationship existed between power during the VCT and physiological attributes measured during an incremental exercise test. The results suggest that the VCT provides a reliable measure of mean power output. Mean self-paced power output was strongly correlated with PPO. The first main finding of the study was the reliability of both self-paced mean power output (r = .92, CV% = 1.98) and total mean power output (r = .92, CV% = 1.24) during the VCT, thus confirming the primary hypothesis. Notably, this study highlights the ability of this test to detect modest (>3.9%) changes in performance with a high level of confidence. While the reliability of variable-power protocols has not been extensively studied, the results of the current study are consistent with those of Abbiss et al,12 who reported a mean CV% of 2.4 of a 30-km dynamic time trial for 2 trials after a single familiarization. While a familiarization identical to the subsequent trials of the current study would have been ideal, in this instance a single familiarization was sufficient to ensure high test–retest reliability of variable-power cycling tests conducted within a week— however, we did not investigate the effect multiple familiarizations may have had on the reliability of the VCT. Despite this limitation,
282
14.32 (1.55)
Rating of perceived exertion (Borg)
14.41 (1.78)
6.71 (2.55)
79.99 (5.63)
6.66 (0.78)
1.74 (0.23)
3.22 (0.16)
3.03 (0.24)
Trial 2
0.09 (1.18)
–0.73 (1.70)
–0.72 (3.19)
0.16 (0.36)
–0.06 (0.13)
0.01 (0.07)
0.01 (0.10)
Mean difference
.76* (.4–.91)
.83** (.56–.94)
.85** (.59–.95)
.89** (.69–.96)
.88** (.66–.96)
.92** (.77–.97)
.92** (.76–.97)
r
.78 (.45–.92)
.85 (.6–.95)
.87 (.65–.96)
.9 (.74–.97)
.89 (.69–.96)
.93 (.81–.98)
.93 (.8–.97)
ICC
1.2 (0.88–1.9)
0.84 (0.61–1.32)
2.25 (1.63–3.63)
0.26 (0.19–0.4)
0.05 (0.04–0.08)
0.09 (0.07–0.14)
0.07 (0.05–0.11)
TEM
4.75 (3.37)
13.93 (9.81)
2.05 (1.79)
3.38 (2.37)
4.49 (3.09)
1.24 (0.75)
1.98 (1.15)
CV (%)
9.3%
27.3%
4%
6.6%
8.8%
2.4%
3.9%
LSC (95%)
Abbreviations: ICC, intraclass correlation; TEM, typical error of measurement; CV, coefficient of variation; LSC, least significant change. Note: The trial data and mean difference are expressed as mean (SD), Pearson r values are expressed as mean (95% confidence intervals), and CV is expressed as mean (SD). LSC calculated as 1.96(CV / n ) × 2 , where n = number of trial replicates. *P < .001. **P < .0001.
7.44 (2.98)
80.71(5.87)
Blood lactate (mmol/L)
·
6.4 (0.74)
Total hard + sprints mean power (W/kg)
Economy at 3.5 W/kg (W ·
1.8 (0.26)
Total recovery mean power (W/kg)
min–1)
3.21 (0.17)
Total mean power (W/kg)
L–1
3.03 (0.25)
Total self-paced mean power (W/kg)
Trial 1
Table 2 Physiological and Performance Variables Measured During Test and Retest of the Variable Cycle Test, N = 15
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Reliability of a New Variable-Power Cycle Test 283
the high reliability can also be attributed to subjects’ participation in local cycling competitions that were in criterium/circuit style—as such, participants were already familiar with the nature and demands of the VCT. Power output during “recovery” and “hard + sprints” sections showed increasing variability compared with the entire trial (Table 2), with mean CVs of 4.49 and 3.38, respectively. This is in agreement with previous findings that report that the CV of power output tends to decrease as intensity increases.13,26 Rating of perceived exertion displayed the weakest correlation between trials (Table 2). Blood lactate concentration showed the highest variability of all measures between trials (Table 2), a finding in accordance with those of Swart and Jennings,27 who showed large fluctuation in finger-prick lactate measurements. Mechanistically, this reflects the variability in lactate production and subsequent diffusion from the skeletal muscle to blood.28 This finding shows that finger-prick lactate measurements are unreliable, and extreme caution should be taken when using this method. The second main finding of the study was the moderate to strong correlations between self-paced mean power during the VCT and performance parameters determined during the incremental exercise test (Figure 3). The results are similar to those presented by Levin et al,13 who reported correlations of .83 and .74 for PPO and VT2, respectively, with power during a 30-km time trial containing 250-m and 1-km sections of intense effort, respectively completed at 127% and 105% of PPO. While the aforementioned high-intensity efforts are applicable to road racing, they do not mirror the shortduration, supramaximal efforts frequently seen during mass-start road stages and criteriums in cycling.6 Thus, the inclusion of these repeated high-intensity efforts in the VCT addresses this limitation of previous protocols. Despite the moderate strength of correlations with these performance variables, future studies are required to establish the longitudinal changes in the results of the VCT and how they relate to both field and test performance in well-trained or elite athletes. PPO showed a stronger correlation with mean power during the 30-km time trial (r = .88), a finding consistent with data from previous studies. Bentley et al14 and Lamberts et al15 reported very strong correlation (r = .91 and .90, respectively) between PPO and time trials of 90 minutes and 40-km duration. This suggests that PPO is not as valid a predictor of variable-power cycling performance as it is for time-trialing and, further, indicates that factors other than those also relevant to time-trial influence performance during variable-power cycling. A significant proportion of the VCT was conducted at intensity greater than lactate threshold, above which point the body increases the amount of energy provided by anaerobic metabolism.16,22 Thus, physiological variables associated with fatigue resistance may be highly relevant to performance during variable-power cycling, including muscle-buffering capacity19,29 and clearance of lactate and other metabolites.16,28 Future studies should be directed toward identifying the physiological characteristics most associated with variable-power cycling performance and the development of a prediction model for performance during such events.
Practical Applications The results of this study suggest that tests simulating the stochastic nature of road cycle racing can be conducted in a reliable way. As such, future studies examining the effect of interventions on performance during road cycling may make use of variable-power cycling tests.
Figure 3 — The correlation between mean self-paced power during the Variable Cycle Test (VCT) and performance parameters from an incremental exercise test. Abbreviations: PPO, peak power output; VT2, second ventilatory threshold; VO2max, maximal oxygen uptake.
284 Sharma, Elliott, and Bentley
In physiological performance diagnostics, the VCT is a race-specific protocol that allows for the assessment of cyclingperformance aspects that cannot be quantified by incremental or time-trial protocols, such as repeat-sprint ability and recovery capacity of cyclists. The VCT can be used as training tool to simulate the demands of criterium-style road racing.
Downloaded by University of Exeter on 09/24/16, Volume 10, Article Number 3
Conclusion The VCT provides a reliable measure of power output during variable-intensity cycling and shows moderate correlation with a number of physiological variables relevant to performance during road cycling. The presence of repeat supramaximal efforts in the VCT addresses the limitations of previous variable-power protocols and allows for the assessment of the repeat-sprint ability and recovery capacities of athletes during a race-specific protocol. Future investigations are required to establish the validity of the VCT in monitoring longitudinal changes in performance during the test with different interventions and determine the physiological factors most relevant to stochastic cycling.
References 1. Hopkins WG, Schabort EJ, Hawley JA. Reliability of power in physical performance tests. Sports Med. 2001;31:211–234. PubMed doi:10.2165/00007256-200131030-00005 2. Jeukendrup AE, Craig NP, Hawley JA. The bioenergetics of world class cycling. J Sci Med Sport. 2000;3:414–433. PubMed doi:10.1016/ S1440-2440(00)80008-0 3. Driller MW. The reliability of a 30-minute performance test on a Lode cycle ergometer. J Sci Cycling. 2012;1:21–27. 4. O’Hara JP, Thomas A, Seims A, Cooke C, King R. Reliability of a highintensity endurance cycling test. Int J Sports Med. 2012;33:18–25. PubMed doi:10.1055/s-0031-1284340 5. Vogt S, Schumacher YO, Blum A, et al. Cycling power output produced during flat and mountain stages in the Giro d’Italia: A case study. J Sports Sci. 2007;25:1299–1305. PubMed doi:10.1080/02640410601001632 6. Ebert TR, Martin D, Stephens B, Withers R. Power output during a professional men’s road-cycling tour. Int J Sports Physiol Perform. 2006;1:324–335. PubMed 7. Currell K, Jeukendrup AE. Validity, reliability and sensitivity of measures of sporting performance. Sports Med. 2008;38:297–316. PubMed doi:10.2165/00007256-200838040-00003 8. Abbiss CR, Menaspà P, Villerius V, Martin DT. Distribution of power output when establishing a breakaway in cycling. Int J Sports Physiol Perform. 2013;8:452–455. PubMed 9. Theurel J, Lepers R. Neuromuscular fatigue is greater following highly variable versus constant intensity endurance cycling. Eur J Appl Physiol. 2008;103:461–468. PubMed doi:10.1007/s00421-008-0738-2 10. Palmer GS, Borghouts L, Noakes T, Hawley J. Metabolic and performance responses to constant-load vs. variable-intensity exercise in trained cyclists. J Appl Physiol. 1999;87:1186–1196. PubMed 11. Bernard T, Vercruyssen F, Mazure C, Gorce P, Hausswirth C, Brisswalter J. Constant versus variable-intensity during cycling: effects on subsequent running performance. Eur J Appl Physiol. 2007;99:103–111. PubMed doi:10.1007/s00421-006-0321-7 12. Abbiss CR, Levin G, McGuigan M, Laursen P. Reliability of power output during dynamic cycling. Int J Sports Med. 2008;29:574–578. PubMed doi:10.1055/s-2007-989263
13. Levin GT, Laursen PB, Abbiss CR. Reliability of physiological attributes and their association with stochastic cycling performance. Int J Sports Physiol Perform. 2014;9(2):309–315. 14. Lamberts RP, Lambert MI, Swart J, Noakes TD. Allometric scaling of peak power output accurately predicts time trial performance and maximal oxygen consumption in training cyclists. Br J Sports Med. 2012;46:36–41. PubMed doi:10.1136/bjsm.2010.083071 15. Bentley DJ, McNaughton L, Thompson D, Vleck V, Batterham A. Peak power output, the lactate threshold, and time trial performance in cyclists. Med Sci Sports Exerc. 2001;33:2077–2081. PubMed doi:10.1097/00005768-200112000-00016 16. Coleman D. The physiological demands of road cycling. In: Hopker J, Jobson S, eds. Performance Cycling: The Science of Success. 1st ed. London: Bloomsbury; 2013:13–32. 17. Lucia A, Hoyos J, Pérez M, Santalla A, Earnest CP, Chicharro JL. Which laboratory variable is related with time trial performance time in the Tour de France? Br J Sports Med. 2004;38:636–640. PubMed doi:10.1136/bjsm.2003.008490 18. Bentley DJ, Newell J, Bishop D. Incremental exercise test design and analysis: implications for performance diagnostics in endurance athletes. Sports Med. 2007;37:575–586. PubMed doi:10.2165/00007256200737070-00002 19. Chicharro JL, Hoyos J, Lucía A. Effects of endurance training on the isocapnic buffering and hypocapnic hyperventilation phases in professional cyclists. Br J Sports Med. 2000;34:450–455. PubMed doi:10.1136/bjsm.34.6.450 20. Hopker J, Myers S, Jobson S, Bruce W, Passfield L. Validity and reliability of the Wattbike cycle ergometer. Int J Sports Med. 2010;31:731–736. PubMed doi:10.1055/s-0030-1261968 21. McLaughlin JE, King GA, Howley ET, Basset DR, Ainsworth BE, Jr. Validation of Cosmed K4b2 portable metabolic system. Int J Sports Med. 2001;22:280–284. PubMed doi:10.1055/s-2001-13816 22. Lucía A, Hoyos J, Pérez M, Chicharro J. Heart rate and performance parameters in elite cyclists: a longitudinal study. Med Sci Sports Exerc. 2000;32:1777–1782. PubMed doi:10.1097/00005768-20001000000018 23. Tanner RK, Fuller K, Ross M. Evaluation of three portable blood lactate analysers: Lactate Pro, Lactate Scout and Lactate Plus. Eur J Appl Physiol. 2010;109:551–559. PubMed doi:10.1007/s00421-0101379-9 24. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2:92–98. PubMed 25. Hopkins WG. Reliability from consecutive pairs of trials [Excel spreadsheet]. In: Internet Society for Sport Science, A New View of Statistics. 2000. sportsci.org/resource/stats/xrely.xls. 26. Lamberts RJ, Swart J, Noakes T, Lambert M. A novel submaximal cycle test to monitor fatigue and predict cycling performance. Br J Sports Med. 2011;45:797–804. PubMed doi:10.1136/bjsm.2009.061325 27. Swart J, Jennings CL. Use of blood lactate concentration as a marker of training status. S Afr J Sports Med. 2004;16:3–7. 28. Brooks GA. The lactate shuttle during exercise and recovery. Med Sci Sports Exerc. 1986;18:360–368. PubMed doi:10.1249/00005768198606000-00019 29. Weston AR, Myburgh KH, Lindsay FH, Dennis SC, Noakes TD, Hawley JA. Skeletal muscle buffering capacity and endurance performance after high-intensity interval training by well-trained cyclists. Eur J Appl Physiol Occup Physiol. 1997;75:7–13. PubMed doi:10.1007/s004210050119
