Introduction

Mountain-biking (MTB) is an off-road cycling modality characterized by competition in rugged and uneven terrain as well as a range obstacles such as jumps, drops, and rocks (^{Novak et al., 2018}). The MTB cross-country marathon (XCM-MTB) is defined by the traveled distance of the race, having a minimum of 60 km and a maximum of 160 km, completed in forest roads, gravel paths, single trails, steep climbs, and technically demanding descents (^{Schenk et al., 2010}). Despite to these characteristics, few studies appear to be available on the physiological characteristics of off-road cycling. In this sense, the prescription of physical training to athlete cyclists has been considered is a challenge to technics and trainers. Therefore, the understanding of the physiological variables during the MTB race can contribute to training planning and for improving the physical performance of the athletes.

The control of the external load of the training involves the organization, quality and quantity of physical work prescribed in the training plan (^{Impellizzeri et al., 2019}). In cycling sports, the velocity, power, pacing, and total time of exercise are widely used variables to control the external load and training prescription (^{Halson, 2014}). Moreover, psychophysiological responses, such as heart rate (HR), maximal oxygen uptake (VO2max), rating perceived exertion (RPE), and lactate blood concentration, are used in the monitoring of internal load and can contribute to determining fatigue and consequently the physical performance of athletes (^{Impellizzeri & Marcora, 2007}).

The HR is an important variable to control of internal load and has been used to estimates the exercise intensity during MTB race (^{Atkinson et al., 2003}). Although, the peak oxygen uptake, the 30s all-out fatigue resistance, and the power variables from a repeated sprint test are able to predict the physical performance in cross-country Olympic MTB (^{Bejder et al., 2019}), the variables determinants of performance in XCMMTB are still unknown.

Specifically, in an XCM-MTB race, the variability of the terrain found during the race can promote a high variability in the variables predicted by of the physical performance, hindering the prior planning of the race strategy. Although the knowledge of variables involved in the quantification of the internal and external load of training can contribute to developing techniques, assist trainers in providing training prescriptions and training control for athletes, and assist in the planning of the race strategy, these responses during an official XCM-MTB race have not yet been investigated. Therefore, the objective of this study is comparing the physiological variables of training control responses

esulting from an official XCM-MTB race, and to verify which of these variables predicts physical performance.

Material and Methods

The present study is characterized as quantitative exploratory research and was conducted during an official cross-country marathon MTB race (XCM-MTB) in Brazil (route of 70 km; at the municipality of Lavras to Carrancas-MG). The athletes were informed about the objectives and procedures of the present study and signed the free and informed consent form. The study was approved by the Human Research Ethics Committee (COEP, n°. CAAE: 17544213.4.0000.5148) following the Declaration of Helsinki 1996.

Figure 1 demonstrates the experimental design. Thirteen XCM-MTB athletes (33.3 ± 12.7 years; 71.6 ± 14.5 Kg; 1.7 ± 0,1 m; 24.4 ± 6.1 Kg/m2) of both sexes (11 men and 2 women) with at least two years of experience in MTB racing participated in this study. The inclusion criteria were athletes who followed a training plan, with frequent participation in competitions of MTB and who used a highly reliable device to measure the following variables: energy expenditure (EE), heart rate (HR), altimetry, velocity, temperature, and power. HR was continuously monitored throughout the test using a telemetric HR monitor (Garmin, Switzerland). Athletes in which the device could not provide information on the analyzed variables and those who were unable to complete the race were excluded from the present study. The variables involved in the quantification of internal and external load training, were obtained through STRAVA (San Francisco, CA 94103, USA) and analyzed every 10 km during the official XCM-MTB race. After the description and comparison of the variables during the XCM-MTB race, the results of the winner were compared with the other competitors to verify when and which variables are determinants of physical performance.

BM: Body Mass; HR: Heart Rate; Temperature (°C): Degree Celsius; EE: Energy Expenditure.

Figure 1 Experimental design. 

Results

The athletes completed the XCM-MTB race in 227.20 ± 16.26 min and presented an energy expenditure of 2546.15 ± 681.32 kcal, without a significant difference in body mass compared to baseline (71.6 ± 14.5 vs. 69.05 ± 14.25 kg; p = 0.769). The heart rate, altimetry, velocity, temperature, pacing, and power during the MTB race are demonstrated in Figure 2. When compared to baseline, the heart rate increased significantly in the 10 km, 20 km, 30 km and 60 km (F = 4.65; p < 0.001; d = 0.11; Figure 2a). The altimetry varied significantly during the XCM-MTB race presenting a peak elevation at 30 km (F = 18.10; p < 0.001; d = 0.61; Figure 2b) and the velocity increased significantly compared to that at 30 km (F = 16.66; p < 0.001; d = 0.56; Figure 2c). The temperature increased during the XCM-MTB race (F = 49.65; p < 0.001; d = 0.63: Figure 2d), reaching 21.62 ± 1.56°C at the end of the race. The athletes also presented a variation in pacing (F = 35.82; p < 0.001; d = 0.66; Figure 2e) and power (F = 7.20; p < 0.001; d = 0.18; Figure 2f).

Data are expressed at mean and standard deviation. * Significant difference from baseline (0 km): * p < 0.05; ** p < 0.01; *** p < 0.001; ^a Significant difference from baseline (10 km): ^a p < 0.05; ^aa p < 0.01; ^aaa p < 0.001; ^b Significant difference from baseline (20 km): ^bb p < 0.01; ^bbb p < 0.001; ^c Significant difference from baseline (30 km): ^c p < 0.05; ^ccc p < 0.001; ^d Significant difference from baseline (40 km): ^dd p < 0.01; ^ddd p < 0.001; ^e Significant difference from baseline (50 km): ^e p < 0.05; ^ee p < 0.01; ^eee p < 0.001; ^g Significant difference from baseline (70 km): ^ggg p < 0.001.

Figure 2 Heart rate, altimetry, velocity, temperature and, pacing during mountain biking performance. 

The correlations between heart rate, velocity, pacing, and power and the total time to complete the XCM-MTB race are presented in Figure 3. Only pacing demonstrated a significant correlation with the total time to complete the XCM-MTB race (Figure 3c). The analysis resulted in a significant model (F = 16.14; p = 0.001; R2 = 0.891). Only pacing could be considered a predictor of the total time (β = 0.958; t = 7.30; p <0.001). Therefore, we performed an analysis of the pacing obtained every 10 km in predicting physical performance. The analysis resulted in a significant model (F = 10.23; p = 0.004; R2 = 0.82), and pacing at 20 km could be considered a predictor of the total time.

Figure 3 Correlation between heart rate, velocity, pacing, and power with the total time to complete the competition. 

The comparison of the heart rate, velocity, pacing, and power between the winner and other competitors are presented in Figure 4. The HR of the winner was higher during the XCM-MTB race (Figure 4a). The winner also presented higher velocities at 10 km, 20 km, and 50 km (Figure 4b) and lower pacings at 20 km, 30 km, 60 km and 70 km (Figure 4c). Moreover, the winner presented higher power during the XCM-MTB (Figure 4d).

Data are expressed as the mean and standard deviation.

Figure 4 Comparison of heart rate, velocity, pacing and power between the winner and other competitors. 

Figure 3a-d: Only the main effects by ANOVA that were significant are displayed on the graphs; Post hoc analyses for a group effect: § p < 0.05; §§§ p < 0.001 for the winner vs. other competitors.

Discussion

The present study described and compared the training control variables and the physiological responses during an official XCM-MTB race. The results demonstrated that athletes are able to maintain an elevated HR for a long period of time, maintaining a constant HRmax of 80% in the corresponding zone throughout the race. The high variability found in the variables of training control demonstrated that velocity, power, and pacing are associated with terrain characteristics and altimetry variation during an XCM-MTB race. Moreover, the present study demonstrated that pacing was able to predict the physical performance of athletes; specifically the pacing at 20 km was able to predict 82% of the total time to complete the XCM-MTB race.

Due to the high aerobic demands of the competition, mountain biker athletes have high cardiovascular capacities and are able to maintain elevated velocity and power throughout the event (^{Granier et al., 2018}). Studies (^{Gregory et al., 2007}; ^{Impellizzeri & Marcora, 2007}) show that mountain biking athletes showed significant correlations between running time and maximal and submaximal aerobic fitness, (^{Gregory et al., 2007}) cross-country running time correlates with physiological responses such as power in relation to body mass (r = 0.93) and relative Vo2peak (r = 0,80). Moreover, athletes of cyclocross showed a sustained elevated HR and high lactate concentration during the cyclocross race (^{Carmichael et al., 2017}). In the present study, we demonstrated that XCM-MTB athletes also showed a significant increase in HR in the first kilometers and maintained a constant elevation in the corresponding HRmax zone of 80% throughout the race. Similar results for HR were observed in the study by ^{Bertuzzi et al. (2006)}, when analyzing the behavior of the HR during an endurance test (performance of a 10 km race). The authors observed a significant increase after 12 minutes and it remained high until the end of the performance. Cross-country mountain biking races present variations in intensity, with a high demand on cardiovascular conditioning and high power (^{Arriel et al., 2022}). In addition to the metabolic demand required by exercise, this high HR value may also be associated with extrinsic factors such as environmental temperature and the hydration status of individuals (^{Achten & Jeukendrup, 2003}; ^{Bertuzzi et al., 2006}). HR is an important variable to control internal training load and intensity in endurance sports (^{Mujika, 2017}). Therefore, understanding the values and HR intensity zones during an official MTB race can be useful for developing techiniques and for trainers to provied better training prescriptions and training control for athletes.

Velocity and power have been extensively used to determine external load training; however, in cycling these variables can be influenced by the characteristics of the terrain. In a cross-country race, athletes are required to ascend hills more often and at high speeds, thus explaining why power output plays an important role (^{Engelbrecht & Terblanche, 2018}). Specifically, in an MTB race, the power increases on steep terrain sections, whereas in the subsequent terrain, the power output is reduced, and at these moments, the power is inversely associated with velocity (^{Gregory et al., 2007}). In the present study, the mean power peak ranged between 136 and 348 watts, which could be explained by phases where maximal and supra-maximal efforts were necessary, including steep climbing, and/or when sprinting to pass slower riders (^{Impellizzeri et al., 2005}).

In downhill moments, the required effort and power were reduced, whereas the HR remained elevated, presumably because of the effect of isometric muscle contractions, which could impair recovery after periods of high-power output (^{Gregory et al., 2007}). In the present study, we also identified that the HR remained elevated independently of the variation in altimetry. The total elevation gain is a major contributor to increasing overall loads of exercise (^{Sanders & Heijboer, 2019}) and therefore is another important variable for training control.

Although the VO2max, 60 s maximal mean power output, peak grip strength, and decision-making ability are likely to be beneficial in the performance of mountain bikers during 4 h event (^{Novak et al., 2018}), to the best of our knowledge, this is the first study to demonstrate the determinant variables of physical performance in an XCM-MTB race. The multivariate linear analysis demonstrated that only pacing can be considered a predictor of physical performance in an XCM-MTB race; specifically, pacing at 20 km can predict 82% of the total time to complete an MTB race. In the present study, the winner of the XCMMTB showed lower pacing throughout the race when compared to the other competitors. Economy of movement during the run, presents a moderate inverse correlation (r = -0.53) with the final time race in Olympic Cross-Country mountain biking athletes, showing a low developed power (^{Inoue et al., 2022}). Pacing is a variable associated with the strategy of the race with the objective of avoiding premature fatigue (^{De Koning et al., 2011}). Specifically, in XCM-MTB, a pacing strategy with a higher speed while climbing and a higher speed while descending is typical of faster performances (^{Moss et al., 2019}). Therefore, pacing control is of great importance for training control and can be considered a variable determinant of physical performance in an XCM-MTB race.

In addition to the presented results, other variables may also be associated with fatigue and physical performance in MTB athletes, including psychophysiological responses, race strategy, anaerobic power, age, sex, and race experience (^{Inoue et al., 2012}; ^{Moss et al., 2019}). One of the limitations of the study was not knowing the data from a progressive test, and thus determining which intensity domains were more determinant in the test and thus relating these domains to the variables studied. Therefore, future research should emphasize that analysis of the other variables associated with physical performance and training programs should be designed to specifically improve both aerobic and anaerobic systems, considering the internal and external load control through the variables studied.

Conclusion

The present study concluded that the analyzed variables are reliable for training control in an official XCM - MTB race. The HR and power are variables that can be used to prescribe training, as they change according to the requirements of the race. Power can also be used as a performance predictor as it is directly influenced by terrain. For the prescription and control of mountain biking training, it is important to identify how heart rate and power respond to the demands of the race, promoting adaptations and changes in a training periodization.