Olympic crosscountry mountain biking (XCO) is a high-intensity intermittent activity of an approximately 2-hour duration. The physiological demands of XCO and the profile of mountain bike (MTB) riders have been previously reviewed by Impellizzeri and Marcora (9). Some physiological variables may play a greater role in determining performance factors, particularly power and aerobic capacity, when normalized to body mass (10,11). However, there is evidence that during various moments in competitions, athletes can maintain exercise intensities either near to or greater than that required to maintain their maximal oxygen consumption (V[Combining Dot Above]O2max). For example, Stapelfeldt et al. (20) reported that athletes increase their power to levels above their maximal power (WMax) on uphill tracks (as demonstrated by a maximal progressive exercise test) at levels surpassing 500 W. Furthermore, higher levels of power are required when overtaking other riders, and during XCO competition, high levels of lactate have been reported (9).
Little attention has been paid to the anaerobic components of cycling performance, especially in MTB. Although it has been suggested that high power and anaerobic capacity are important physiological requirements in XCO competitions (2,8,9,20), no evidence as of yet demonstrates any relationship between anaerobic indicators and performance in this sport. Our objective is to determine the association between anaerobic variables and performance during XCO competition. Taking into account the intermittent and high-intensity characteristic of XCO race, we hypothesized a significant correlation between performance indicators of anaerobic power and XCO race time.
Experimental Approach to the Problem
For testing our hypothesis, we examined the correlation between anaerobic variables and XCO performance during a regional race. Considering the high-intensity intermittent characteristic of XCO, we included 2 Wingate (WIN) tests, the standard test to verify the anaerobic power, and an adaptation of the original test with 5 repetitions of the WIN protocol at a lower workload and intercalated with 30-second rests to verify the ability to repeatedly produce anaerobic efforts. This cross-sectional study required 3 testing sessions to complete the data collection. Two of these took place in the laboratory and the third during a step of the Lagos Mountain Bike Championship (Saquarema, RJ, Brazil). The subjects had from 1 to 7 days between laboratory visits, and there was a 3-week gap before the competition. All the laboratory tests were performed at 21 to 23°C and conducted at the same time (≈2 hours). The subjects were instructed to run as fast as they could and not eat solid foods 3 hours before testing and to take water ad libitum.
The original sample included 13 regional and national level MTB athletes (Table 1) with 2 years of competitive practice and with a maximal aerobic power >60 ml·kg−1·min−1. All the subjects trained 6 d·wk−1 with a total training distance of 300–400 km·wk−1. The subjects were excluded if they had any mechanical problems during the race. From the original 13 subjects recruited, 3 were excluded from the analysis because of mechanical problems during the competition. All the subjects signed a consent form that had been previously approved by the institutional ethical board (Gama Filho University—protocol 0007.0.312.000-09).
The measurement standards established by the International Society for the Advancement of Kinanthropometry (17) were used to determine the following anthropometric measures: body weight and height (Filizola Scale, São Paulo, Brazil) and skinfold (Slim Guide, Rosscraft, Surrey, Canada). Body composition (12,19) was estimated from the collected data.
The subjects were submitted to an incremental maximal test on a mechanical break cycle ergometer (Maxx Pro, Belo Horizonte, Brazil) with a racing saddle and racing pedal system. During the test, the subjects were instructed to maintain a rate of 90 rpm (22). A warm-up was performed for approximately 10 minutes at 88 W. After that, increases of 22.1 W·min−1 (0.25 kp·min−1 at 90 rpm) were implemented until volitional fatigue (10). The test was interrupted when a problem occurred or if the subject was unable to respect the protocol's cadence for 10 seconds. Heart rate (HR) was continuously monitored (Polar® Vantage NV, Polar Electro, Oy, Finland), and the Borg's CR10 scale was recorded 5 seconds before the start of each stage (3). Maximal oxygen consumption (V[Combining Dot Above]O2max) was predicted using the Hawley and Noakes (7) equation (V[Combining Dot Above]O2max [liters per minute] = 0.01141 × WMax [W] + 0.435) using the maximal power achieved during the final 60-second test stage. If the cyclist did not complete the last step of the incremental test, an equation was applied to calculate the peak power (PP) as suggested by Kuipers et al. (14).
Wingate and 5× Wingate Tests
Both tests were performed on the same bicycle and after the same warm-up used in the aerobic test. After a 2-minute rest, the WIN test was started with a workload equivalent of 0.1 kp·kg−1 (1). After 3–5 seconds without a load, weight was applied, synchronized with the chronometer start. The subjects were instructed to accelerate and pedal at the highest cadence they could. Power output was recorded by filming with a digital camera (Sanyo VPC-CA65 EX, Sanyo Electric, Los Angeles, CA, USA) located in a sagittal plane at 200 cm from the subject and perpendicular to 60 cm of soil.
The Virtual Dub software (Version 1.9.6, Free Software Foundation, Cambridge, MA, USA) was used to capture frame by frame. Peak power and mean power (MP) were calculated using Microsoft Office Excel 2007 (Version 12.0 for Windows, Microsoft Corp., Redmond, WA, USA). During the 5× Wingate (WIN5) test, the same procedures described for the WIN test were repeated with 30-second rests between trials and a workload equivalent of 0.05 kp·kg−1 (13). Five minutes after the WIN5 test was conducted, a blood sample (≈25 μL) was taken from the index finger to determine blood lactate concentration ([La]WIN5) using a portable spectrophotometer (Accutrend Lactate, Boehringer-Mannheim, Mannheim, Germany).
The regional level crosscountry circuit competition was conducted insummer between 9:00 AM and 12:00 PM, at a temperature of approximately 26°C and humidity of approximately 45%. Before the competition, all the athletes performed a free warm-up according to their own preferences. During the race, their HR was recorded (Polar® Electro, Oy, Finland). The subjects were instructed to complete the race as fast as possible and to perform a hydration strategy ad libtum. The individual's official race time supplied by the organizers was used for all the analyses.
After checking for normality assumption, Pearson's product moment correlation coefficient was used for examining the association between race performance and the anaerobic variables. Given the low sample size, we presented the estimation of 95% confidence intervals using Fisher's z transformation. Because a different starting position could have influenced competition time and, consequently, its relation to laboratory variables, and given that MTB riders are usually more concerned with the ranking than time, we also calculated Spearman's rank correlation. The correlation coefficients were interpreted using the scale of magnitudes proposed by Hopkins (www.sportsci.org): <0.1, trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large; 0.7–0.9, very large; >0.9, nearly perfect. The analyses were performed using SPSS v. 17.0 (SPSS Inc., Chicago, IL, USA), and GraphPad Prism 5 (GraphPad software Inc., San Diego, CA, USA) at a significance level of p ≤ 0.05.
The crosscountry circuit competition was completed in 142.7 ± 15.4 minutes, at an average HR of 172 ± 13 b·min−1 corresponding to 92% of the maximum HR. The anaerobic parameter derived from the WIN tests showing very large correlation with XCO race time was the average maximal power on the WIN5 test (r = −0.79, p = 0.006). The correlation expressed in absolute values normalized to body mass are presented in Table 2. The average maximal power on the WIN5 test, relative to body mass 1 (WIN5 in watts per kilogram) and 0.79 (WIN5 in W·kg−0,79), showed significant associations with the race time, and the average MP on the WIN5 test (watts per kilogram) showed a significant and large association with the XCO race time. Similarly, the correlations of WIN5 expressed relative to body mass (−0.80, p = 0.005) and body mass−0.79 (−0.77, p = 0.009) with final ranking were significant. Figure 1 shows the scatter plots of the correlations between race time and the average maximal power on the WIN5 test (watts per kilogram) and the WMax test (watts per kilogram).
This is the first study to examine the relationship between anaerobic tests and XCO performance. According to our hypothesis, we found a moderate to large correlation between XCO performance and anaerobic parameters measured during the intermittent test. This may be explained by the physical requirement and technical characteristics of XCO competitions. Indeed, XCO is a highly intense intermittent event with PP outputs ranging between 250 and 500 W and during which there are phases where maximal and supramaximal efforts are necessary such as during steep climbing, at the start of the race and when sprinting to pass slower riders (10). For these reasons, it has been suggested that high anaerobic power, other than high aerobic fitness, may be important for meeting the physiological demands of off-road cycling competitions (9).
Taking into account the characteristics of XCO performance and the aforementioned suggestions of previous authors (13), we developed the WIN5 test to measure the power output during repeated anaerobic efforts, thus reproducing the intermittent nature of XCO performance. The high lactate blood concentration levels measured at the end of the test (16.0 ± 2.3 mmol·L−1) confirmed that the anaerobic energy system was heavily involved. A significant correlation was observed between race time and the average peak and MP measured during the WIN5 test (expressed as watts per kilogram and W·kg−0.79). These relationships were confirmed by the rank correlations. The findings indicate that this test is promising and interesting for evaluating XCO athletes' performance. The significant correlation with WIN5 test and XCO race time may be because of the intermittent characteristic of the test reflecting the intermittent nature of XCO competitions as hypothesized.
Among the tests that are traditionally used to evaluate cycling, the WIN test has been consistently used to determine power and anaerobic capacity (1). Studies by Machado et al. (16) and Costa and De-Oliveira (4) found no significant correlations between PP and MP on the WIN test and simulated MTB uphill time and XCO race time. Similarly, no significant associations between PPWIN and MPWIN and XCO race time were observed in this study even if we cannot exclude that the lack of significance was because of the low statistical power. Conversely, Davison et al. (5) found a significant correlation between average power measured using the WIN test in relation to body mass and performance time while investigating road cyclists cycling 1 km uphill with a 12% inclination (r = −0.92) and 6 km uphill with a 6% inclination (r = −0.90). However, a short performance time was used (∼4–16 minutes), which is considerably less than the duration of an XCO competition.
The reasons for this discrepancy are not clear but it seems that the ability to repeatedly produce anaerobic efforts is a more important determinant of performance than is the maximal anaerobic power. However, Prins et al. (18) found that intermittent tests did not show greater correlation with performance compared with the PP and lactate threshold in a traditional incremental test. Further studies are needed to clarify this issue.
Compared with the findings of previous studies (4,11), we observed greater relations between XCO performance and WMax normalized by body mass. In a previous study, Costa and De-Oliveira (4) reported an association between XCO performance of Brazilian riders and WMax when normalized to body mass (r > 0.88). Impellizzeri et al. (10,11) reported greater correlations between XCO performance and aerobic parameters such as PP output, maximal oxygen uptake, ventilatory and lactate thresholds when normalized to body mass. Similar findings have been reported by Gregory et al. (6) and Prins et al. (18). Overall, the association between performance and PP output suggests that the aerobic metabolism is largely involved in XCO performance. However, according to the results of this study, anaerobic power seems to be another important physiological capacity related to XCO performance as previously suggested by Baron (2). Other authors such as Costa and De-Oliveira (4) and Impellizzeri et al. (9,10) have underlined that XCO competitions, although predominantly aerobic, probably also require high power and anaerobic capacity. The results of this study support their conclusions. Furthermore, similarly to aerobic power, the relationship between the parameters derived from the anaerobic tests and XCO performance was higher when normalized to body mass compared with the absolute values. The importance of the power-to-weight ratio is because of the several climbs characterizing XCO races (6,8–11,15). Indeed, body mass can explain 10–20% of uphill performance (21). Therefore, it can be speculated that the lower body mass of XCO athletes may be because of a natural selection process determined by the specific physiological requirements of XCO competitions (15).
The use of a real competition instead of a simulation may be seen as a limitation of the study. However, this approach (similar to most of the previous studies) has increased the external validity of the study. Although the effect of confounders cannot be excluded, the associations found are similar to those previously reported. Another limitation can be the completion of the tests 3 weeks before the competition. However, this time frame was because of the postponement of the race decided by the organizers for technical reasons. Therefore, we cannot exclude changes in aerobic and anaerobic fitness, but if they occurred, their influence was not enough to obscure the associations. Finally, given the low sample size, a multiple regression (probably more appropriate for examining combined contribution of aerobic and anaerobic systems) was not possible. Because for each predictor at least 10 cases are needed, this would require at least 20 riders. Considering our results, future studies should be conducted to investigate the combined metabolism (aerobic and anaerobic) collaboration to determine race performance.
In conclusion, the results of this study support the use of anaerobic tests for assessing mountain bikers participating in XCO races and suggest that anaerobic power is another important determinant of performance even if we were not able to clarify the unique contribution of the various energy systems to XCO performance. The present findings further confirm the importance of analyzing the physiological test results of MTB riders normalizing the values to body mass.
The major finding of this study was that, given the significant and high relationship between a modified version of the WIN test (five 30-second consecutive stimuli with 30-second rest) and XCO performance during a mountain bike race, this test appears to be appropriate for measuring the anaerobic-intermittent performance in MTB riders. Because also power and aerobic capacity are paramount in the training of XCO athletes, training programs should be designed to specifically improve both the aerobic and anaerobic energy production systems.
The authors thank all the volunteers who participated in this study. Allan Inoue and Alberto S. SáFilho were supported by scholarship from CNPq, and Tony M Santos was supported by research grant # E26/190.127/2010 from FAPERJ. All the authors declare that there is no potential conflict of interests regarding this article. The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association. Finally, the authors would like to thank the reviewers for their important comments.
1. Bar-Or O. The Wingate anaerobic test. An update on methodology, reliability and validity. Sports Med 4: 381–394, 1987.
2. Baron R. Aerobic and anaerobic power characteristics of off-road cyclists. Med Sci Sports Exerc 33: 1387–1393, 2001.
3. Borg G. Borg's Perceived Exertion and Pain Scales (2nd ed.). Champaign, IL: Human Kinetics, 1998.
4. Costa VP, De-Oliveira FR. Physiological variables to predict performance in cross-country mountain bike
races. J Exerc Physiol Online 11: 14–24, 2008.
5. Davison RC, Swan D, Coleman D, Bird S. Correlates of simulated hill climb cycling performance. J Sports Sci 18: 105–110, 2000.
6. Gregory J, Johns DP, Walls JT. Relative vs. absolute physiological measures as predictors of mountain bike
cross-country race performance. J Strength Cond Res 21: 17–22, 2007.
7. Hawley JA, Noakes TD. Peak power output predicts maximal oxygen uptake and performance time in trained cyclists. Eur J Appl Physiol Occup Physiol 65: 79–83, 1992.
8. Impellizzeri F, Sassi A, Rodriguez-Alonso M, Mognoni P, Marcora S. Exercise intensity during off-road cycling competitions. Med Sci Sports Exerc 34: 1808–1813, 2002.
9. Impellizzeri FM, Marcora SM. The physiology of mountain biking. Sports Med 37: 59–71, 2007.
10. Impellizzeri FM, Marcora SM, Rampinini E, Mognoni P, Sassi A. Correlations between physiological variables and performance in high level cross country off road cyclists. Br J Sports Med 39: 747–751, 2005.
11. Impellizzeri FM, Rampinini E, Sassi A, Mognoni P, Marcora S. Physiological correlates to off-road cycling performance. J Sports Sci 23: 41–47, 2005.
12. Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr 40: 497–504, 1978.
13. Kearney JT, Rundell KW, Wilber RL. Measurement of work and power in sport. In: Exercise and Sport Science. William E.G.J., Donald T.K., eds. Philadelphia, PA: Lippincott Williams & Wilkins, 2000. pp. 31–52.
14. Kuipers H, Verstappen FT, Keizer HA, Geurten P, van Kranenburg G. Variability of aerobic performance in the laboratory and its physiologic correlates. Int J Sports Med 6: 197–201, 1985.
15. Lee H, Martin DT, Anson JM, Grundy D, Hahn AG. Physiological characteristics of successful mountain bikers and professional road cyclists. J Sports Sci 20: 1001–1008, 2002.
16. Machado CEP, Caputo F, Lucas RD, Denadai BS. Physiological and anthropometrical factors associated with uphill off-road cycling performance. Braz J Sci Mov 10: 35–40, 2002.
17. Norton K, Olds T, eds. Anthropometrica. Sydney, Australia: University of New South Wales Press, 1996.
18. Prins L, Terblanche E, Myburgh KH. Field and laboratory correlates of performance in competitive cross-country mountain bikers. J Sports Sci 25: 927–935, 2007.
19. Siri WE. Body composition from fluid spaces and density. In: Techniques of Measuring Body Composition. Brozek J., Henschel A., eds. Washington, DC: National Academy of Science, 1961. pp. 233–244.
20. Stapelfeldt B, Schwirtz A, Schumacher YO, Hillebrecht M. Workload demands in mountain bike
racing. Int J Sports Med 25: 294–300, 2004.
21. Swain DP. The influence of body mass in endurance bicycling. Med Sci Sports Exerc 26: 58–63, 1994.
22. Takaishi T, Yasuda Y, Moritani T. Neuromuscular fatigue during prolonged pedalling exercise at different pedalling rates. Eur J Appl Physiol Occup Physiol 69: 154–158, 1994.