Free Access
Issue
Mov Sport Sci/Sci Mot
Number 104, 2019
Masters athletes: Age is just a number / Les athlètes masters : l’âge n’est qu’un nombre
Page(s) 13 - 19
DOI https://doi.org/10.1051/sm/2019015
Published online 12 June 2019

© ACAPS, 2019

1 Introduction

Aging is an inherent and progressive process accompanied by alterations of physiological function including cardiovascular and neuromuscular systems (Hunter, Pereira, & Keenan, 2016; Wilson & Tanaka, 2000), resulting in the decline of functional capacities, reduced autonomy and ultimately death. It is well recognized that physical performance decreases with age starting approximately at 30 years old (Baker & Tang, 2010). For example, maximal oxygen consumption declines by 10% per decade (Hawkins & Wiswell, 2003; Trappe, Costill, Vukovich, Jones, & Melham, 1996), with progressive changes in the endocrine and neuromuscular systems leading to a deterioration in maximal force generation capacity and endurance (Faulkner, Davis, Mendias, & Brooks, 2014; Tanaka & Seals, 2008). The decline in physiological function is accelerated by sedentary behavior and lack of physical activity that are both increasing with age (Bouchard, Shephard, & Stephens, 1994). However, in most studies on human aging physical activity level is not controlled, inducing confusion between aging and inactivity effects (Hunter et al., 2016). For example, as elderly population relies more on the oxydative than glycolytic metabolism (Lanza, Befroy, & Kent-Braun, 2005), the loss in endurance performance with aging could be less than expected, especially in physically active individuals. In this context, trained individuals like master athletes (i.e. > 40 years old) are an interesting model to study human aging, as they remain free from the deleterious effects of inactivity (Lazarus & Harridge, 2017; Tanaka & Seals, 2003).

Performance of master athletes during endurance and ultra-endurance (> 6 h) competitions have shown a growing interest, especially since the number of finishers older than 40 years considerably increased in the last decade (Lepers & Cattagni, 2012) inducing a rapid improvement in performance in this particular category of athletes (Lepers & Stapley, 2016). Analyzing performances during ultra-endurance competitions allows investigating the evolution of the physiological limits of the body across ages. The results from ∼10 000 athletes performing 161-km North-American ultramarathons between 1977 and 2008 showed a ∼4% performance loss per decade starting from 40 years old (Hoffman, 2010). The loss was even greater (∼15%) when only the Top 5 finishers from ten specific races were considered, consistent with the performance observed from the Top 5 of the Western States 100-mile Endurance Run (Hoffman & Wegelin, 2009). A similar trend was observed during a 100-km ultramarathon with limited elevation gain (600 m) (Knechtle, Rüst, Rosemann, & Lepers, 2012a). However, when there is an important elevation gain during the race, a different trend is noticed regarding the age-related decline in performance. For example, at the Swiss Alpine Marathon (78 km, 2260 m of elevation gain), winner’s performance per category started decreasing at 55 years old (Eichelberger & Bilodeau, 2007). Considering that the distance of this race was shorter than the races previously mentioned, it is not clear how age-related performance is influenced by elevation gain during ultra-endurance races. Thus, age-related changes in running performance for long distance races with higher elevation gains need to be investigated.

The Ultra-Trail du Mont-Blanc (UTMB®) is a famous international ultra-endurance event. The main event is a 170-km ultra-trail with 10 000 m of elevation gain. This race brings together well-trained athletes of all age groups as the participation is subject to a qualification system and limited to 2300 participants. Shorter races are also organized at the UTMB® on segments of the main race such as the Courmayeur-Champex-Chamonix (CCC®, 101 km and 6100 m of elevation gain, 1900 participants allowed) and the Orsières-Champex-Chamonix (OCC, 55 km and 3500 m of elevation gain, 1200 participants allowed). The large number of participants and the various races on the same circuit give the opportunity to investigate the effects of distance and difficulties on age-related effects on performance.

The aim of the present study was thus to compare the number of finishers and the performance in three ultra-trail races with different distances and elevation gains (UTMB®, CCC®, OCC) across age categories. We hypothesized that (i) the age group with the largest number of finishers would be the 40–49 yr as previously observed (Hoffman & Wegelin, 2009; Knechtle et al., 2012a, b) and (ii) performance would be similar between 23 and 49 yr but the decline would start from 50, as previously observed in the literature.

2 Methods

Considering that the study involved the analysis of publicly available data, written consent was not required. The total race time and age category of all the male finishers of the UTMB®, CCC® and OCC from 2014 (i.e. the date of the first edition of the OCC) and 2018 were analyzed. The data set from this study was obtained from the race’s website (http://www.utmbmontblanc.com).

2.1 Data analysis

Data were available from 18 903 male athletes. Athletes where divided in age groups based on 5 official categories: seniors (23–39 yr), masters 1 (40–49 yr), masters 2 (50–59 yr), masters 3 (60–69 yr) and masters 4 (70–79 yr). For the 3 races over the 5-year period, the average speed and finishing time of the Top 10 finishers and all finishers were assessed separately for each category. Values for the Top 10 finishers were pooled for the 5 editions resulting in 50 data in each category. For the 70–79 yr group, as only 2 participants finished the UTMB® and CCC® and 30 finished the OCC, this category was not included in the present analysis. Performance density was expressed as the time difference between the winner and the 10th finisher for each age category and expressed as a percentage of the winner performance. This parameter was calculated for each year (from 2014 to 2018) and averaged for each age category and race. Because of the large differences in the number and performance level of finishers in each age category, which can lead to a misinterpretation of the results from the Top 10, finishing time of the Top 10% of the finishers of each category for the three competitions was also considered.

2.2 Statistical analysis

Descriptive statistics were used to report finishing time and average speed of the finishers, Top 10 and Top 10% as mean ± standard deviation (SD) in the text and tables. All variables were tested for normal distribution by means of the Shapiro–Wilk test if n < 2000 and Kolmogorov–Smirnov test if n ≥ 2000. The difference between age categories for each competition and the difference between competitions for each age category were analyzed using Kruskal–Wallis test for non-parametric parameters and one-way ANOVA for parametric parameters. If the Kruskal–Wallis test provided difference between mean ranks, Dunn’s post hoc test adjusted using the Bonferroni correction was performed to make comparisons among groups. For one-way ANOVA, Bonferroni’s post hoc was chosen if pairwise comparison was required. Significance was set as P < 0.05 for all tests. Statistical analysis was conducted using IBM SPSS software (Version 23.0, SPSS, Inc, Chicago, IL).

3 Results

The total numbers of finishers from 2014 to 2018 for each age category during the three races (UTMB®, CCC® and OCC) are reported in Table 1. The number of finishers decreased with age from the youngest group in CCC® and OCC, whereas the age group with the largest number of finishers was 40–49 yr at UTMB®. More than 80% of the finishers were aged 23–49 yr when considering the three races (Fig. 1).

Table 1

Number of finishers of each age category and at the UTMB®, CCC® and OCC during the 2014–2018 period.

thumbnail Fig. 1

Percent of finishers per age category in Ultra-Trail du Mont-Blanc® (UTMB®, 171 km, 10 000 m of elevation gain; white), Courmayeur-Champex-Chamonix (CCC®, 101 km, 6100 m of elevation gain; grey) and Orsières-Champex-Chamonix (OCC, 55 km, 3500 m of elevation gain; black).

3.1 Top 10 results

For each competition, Top 10 average speed significantly decreased with age (P < 0.001; Fig. 2). For UTMB® the best average speed was performed by the 23–39 yr category (7.69 ± 0.49 km.h−1), followed by 40–49 yr (7 ± 0.46 km.h−1), 50–59 yr (5.51 ± 0.55 km.h−1) and 60–69 yr (4.32 ± 0.44 km.h−1). The density of the Top 10 finishers for UTMB® was 14.3 ± 2.5% (23–39 yr), 16.8 ± 3.9% (40–49 yr), 30.7 ± 8.2 (50–59 yr) and 32.4 ± 6.9% (60–69 yr). Density for 23–39 yr and 40–49 yr significantly differed from 50–59 yr and 60–69 yr (P < 0.05), while no differences were found between 23–39 yr and 40–49 yr and between 50–59 yr and 60–69 yr. Regarding CCC®, the best average speed for the Top 10 finishers was performed by the 23–39 yr category (8.34 ± 0.62 km.h−1), followed by 40–49 yr (7.53 ± 0.69 km.h−1), 50–59 yr (5.95 ± 0.46 km.h−1) and 60–69 yr (4.48 ± 0.45 km.h−1). The density of the Top 10 finishers (calculated based on finishing time) showed significant difference only between 23–39 yr and 60–69 yr (P < 0.05), and was 14.4 ± 3.5% (23–39 yr), 23.3 ± 3.4% (40–49 yr), 18.7 ± 4.5 (50–59 yr) and 30.1 ± 4.2% (60–69 yr). For OCC, the Top 10 best average speed was performed by the 23–39 yr category (9.53 ± 0.55 km.h−1), followed by 40–49 yr (8.55 ± 0.71 km.h−1), 50–59 yr (6.88 ± 0.48 km.h−1) and 60–69 yr (5.64 ± 0.52 km.h−1). The density of the Top 10 finishers (calculated based on finishing time) was significantly different only between the 23–39 yr and 60–69 yr groups (P < 0.05) and was 13.8 ± 6.9% (23–39 yr), 22.7 ± 3.0% (40–49 yr), 21.5 ± 3.1 (50–59 yr) and 28.5 ± 9.1% (60–69 yr). Relative data, expressed as percentage of the 23–39 yr group, are presented in Figure 2A. Finishing times for the Top 10 of UTMB®, CCC® and OCC are reported in Table 2.

A Kruskal–Wallis test provided very strong evidence of a difference (P < 0.001) for running speed between the three competitions. Dunn’s pairwise tests showed significant differences between the three races in each age category except for the 60–69 yr age group between UTMB® and CCC®. Independently of the race distances, compared to the 23–39 yr category, running speed decreased by 10% for 40–49 yr, by 28% for 50–59 yr, and by 44% for 60–69%, respectively.

thumbnail Fig. 2

Average speed for the Top 10 (A), Top 10% (B) and Overall finishers (C) of each age category in UTMB®, CCC®, and OCC, expressed as percentage of the 23–39 yr group. * = significant difference from 23–39 yr (P < 0.05); # = significant difference from 40–49 yr (P < 0.05); $ = significant difference from 50–59 yr (P < 0.05).

Table 2

Average time for all the finishers of each age category at the UTMB®, CCC® and OCC during the 2014–2018 period.

3.2 Top 10% results

Top 10% performance decreased with age (P < 0.001), yet without any significant difference between 50–59 yr and 60–69 yr groups. Finishing times for the Top 10% are reported in Table 2. For UTMB®, compared to the 23–39 yr Top 10% (6.6 ± 0.7 km.h−1), average speed was 12% slower for 40–49 yr (5.8 ± 0.6 km.h−1), 22% slower for 50–59 yr (5.2 ± 0.5 km.h−1) and 24% slower for 60–69 yr (5.0 ± 0.3 km.h−1; Fig. 2B). Regarding CCC®, the 23–39 yr Top 10% ran at an average of 7.4 ± 0.7 km.h−1, the 40–49 yr category was 13% slower (6.4 ± 0.7 km.h−1), the 50–59 yr category was 23% slower (5.7 ± 0.4 km.h−1) and the 60–69 yr category was 30% slower (5.2 ± 0.3 km.h−1; Fig. 2B). For OCC, the 23–39 yr performed the fastest average speed at 8.6 ± 0.7 km.h−1. The 40–49 yr category was 12% slower (7.6 ± 0.8 km.h−1), the 50–59 yr category was 22% slower (6.7 ± 0.5 km.h−1) and the 60–69 yr category was 28% slower (6.2 ± 0.3 km.h−1; Fig. 2B).

3.3 Overall results

For each competition, overall average speed significantly decreased with age (P < 0.05). No significant difference was found for the groups 50–59 yr and 60–69 yr in the OCC. For UTMB® the best average speed was performed by the 23–39 yr category (4.57 ± 0.87 km.h−1), followed by 40–49 yr (4.33 ± 0.63 km.h−1), 50–59 yr (4.14 ± 0.45 km.h−1) and 60–69 yr (4.03 ± 0.41 km.h−1). Regarding CCC®, the best average speed for the Top 10 finishers was performed by the 23–39 yr category (5.03 ± 1.06 km.h−1), followed by 40–49 yr (4.65 ± 0.8 km.h−1), 50–59 yr (4.36 ± 0.57 km.h−1) and 60–69 yr (4.2 ± 0.42 km.h−1). For OCC, the Top 10 best average speed was performed by the 23–39 yr category (5.87 ± 1.26 km.h−1), followed by 40–49 yr (5.42 ± 1.04 km.h−1), 50–59 yr (5 ± 0.79 km.h−1) and 60–69 yr (4.74 ± 0.68 km.h−1). Average speed of all the UTMB® and CCC® finishers decreased linearly with a ∼4% rate per decade, while average speed of all the OCC finishers showed a linear decrease of ∼6% per decade (Fig. 2C). Average speed of all finishers significantly decreased with increase in distance and difficulty of the race (OCC faster than CCC® faster than UTMB®) for each age group.

4 Discussion

This study examined the 2014 to 2018 participation and performance for three different ultra-endurance races in the context of the Ultra-trail du Mont-Blanc (UTMB®, CCC®, OCC) across age. It was hypothesized that performance would decrease similarly with age for the three races starting from 50–59 yr. Average speed of the Top 10 and all finishers decreased linearly with age in the three races. However, performance of the Top 10 decreased by 10% between 23–39 yr and 40–49 yr and by 20% per decade from 50 years old, evidencing a drop in performance from this age category. Average speed decreased as the distance increased for all age groups but 60–69 yr, where no difference was found between the CCC and UTMB in average speed.

4.1 Participation

As previously observed in ultra-endurance competitions (Hoffman, 2010; Knechtle et al., 2012a, b, c), the age group with the largest number of finishers in the UTMB® race was 40–49 yr. In contrast, the two shorter races (CCC®, OCC) showed larger participation for the youngest group (23–39 yr). This could be partially explained by the fact that 23–39 yr is the largest age-group and by the higher difficulty of the UTMB® race in terms of elevation gain and duration (20 to 48 h) compared to the CCC® and OCC. Young participants may start with the shorter races like CCC® (11–28 h) and OCC (5–15 h) and would rather run the UTMB® as they get more experienced. Furthermore, the goal of the majority of the UTMB® participants may be to complete the race rather than doing the best performance, whereas young athletes may be more attracted by ultra-trail like the CCC® and OCC where it is possible to run at higher speed than UTMB® (Baker & Tang, 2010; Bernard, Sultana, Lepers, Hausswirth, & Brisswalter, 2009). The number of finishers drastically decreased after 50 years in the three races: finishers aged 23–49 yr represent more than 80% of the total, whereas it falls to 15% for the 50–59 yr, between 1 and 4% for the 60–69 yr and a negligible number (< 1%) over 70 years old. The actual number of finishers aged over 50 contrasts with other ultra-endurance competitions of 161 and 100 km, where more than 30% of the finishers are part of the older age groups (Hoffman, 2010; Knechtle et al., 2012a, b, c). The important elevation gain associated with the duration of the race may discourage the older athletes to participate. Finally, cultural or socio-economic factors may influence the participation rate in this kind of competition.

4.2 Age-related changes in performance

The Top 10 finishers results showed a progressive decrease in performance with age for the three races. Previous studies on ultra-endurance competitions observed that the best athletes were aged between 35 and 45 (Eichenberger, Knechtle, Rüst, Rosemann, & Lepers, 2012; Hoffman & Wegelin, 2009; Knechtle et al., 2012a, b, c). In UTMB® and associated races (CCC® and OCC) where elevation gain is high, the best performance was achieved by the youngest age group (23–39 yr). Despite different age categories of the competitions analyzed that could be a confounding factor (23–39 yr vs. 35–45 yr), our results could suggest that age of peak performance does not increase with the race distance (Schulz & Curnow, 1988). Performance of the Top 10 decreased by approximately 10% at the age of 40–49 for the three races. However, the rate of decline accelerated at 20% per decade starting from 50 years old, and might reflect a limitation in training adaptations and tolerance of high volumes of training (Kallinen & Markku, 1995). This breaking point set at 50 years old corresponds to the age group at which the number of finishers drastically declined for the three races. The 50–59 yr group at UTMB® as well as the 60–69 yr group for the three races showed a lower density (increased difference between the winner and the 10th finisher) than the youngest group. This lower density may be explained by the lower number of participants in the older age categories, but could also be due to the larger age-related variability in neuromuscular (Hunter et al., 2016) and aerobic performance (Whipple et al., 2018). In addition to physiological variability, differences in training history and strategy may also explain the larger heterogeneity in the older groups. Whereas the Top 10 finishers of the 23–39 and 40–49 yr categories are mostly full-time sponsored athletes, this is not the case for older athletes, mainly amateurs with less time and/or external support for training. In a socio-economical perspective, older age groups (60–69 yr) represent an elite sample of the overall elderly population, providing further evidences about the capacity to maintain an excellent functionality in the advanced stage of life (Karlsen et al., 2015). Further studies should examine the physical capacities of this population in the context of ultra-endurance competition to get more insight in the physiological profile of those athletes. Top 10% analysis was used to obtain a more representative sample for each age category than Top 10, avoiding misleading interpretations due to confounding factors like number of full-time sponsored athletes or number of overall competitors per category, in addition to the heterogeneous density and variability in the Top 10 data. Top 10% results confirmed the trend shown by the Top 10 analysis: performance consistently declined with age, but the rate of decline was ∼10–15% per decade across all age categories in the three competitions, except between the 50–59 yr and 60–69 yr groups where the level of significance was not reached. In the context of ultra-trail competitions with high elevation gains, muscle damage can increase due to downhill phases where external negative work increases, which has a structural impact on the joints, especially ankle and knee (Vernillo et al., 2016, 2017). Uphill phases require more propulsive force and power generation compared with flat running, inducing greater activation of the muscles and greater energy cost (Giandolini et al., 2016; Vernillo et al., 2017). Runners need either to decrease drastically their running speed or increase their power generation. Thus, muscle mass, strength and fatigability may play a greater role in performance during hilly than flat ultra-endurance races. The progressive loss in muscle mass (Doherty, 2003; Frontera, Hughes, Lutz, & Evans, 1991) and strength with aging (Lindle et al., 1997), and the greater fatigability in older adults during dynamic contractions (McNeil & Rice, 2007; Sundberg, Kuplic, Hassanlouei, & Hunter, 2018), may explain why the performance decline was exacerbated in the older age categories. Furthermore, the three races with different distances (from 55 to 170 km) and elevation gains (from 3500 to 10 000 m) showed similar trends in performance across age. The shortest race, OCC, remains a mountain ultra-trail of ∼6 h, lasting as long as a flat 100-km ultra-marathon (Knechtle, Rüst, Rosemann et al., 2012a). It should be noted that the Top 10 60–69 yr ran UTMB® and CCC® at the same speed, but they ran the OCC faster. This result suggests that beyond a certain duration, older athletes may adopt an optimal speed that they can sustain even if the difficulty and/or the duration of the race increase.

The significant drop in ultra-trail performance is greater and starts earlier than observed during ultra-cycling races (Pozzi et al., 2014), but similar to mountain bike races (Haupt et al., 2013). The age-related decline in performance seems specific to the discipline. During ultra-triathlon, performance decreases earlier for running than swimming and cycling (Knechtle, Rüst, Knechtle, Rosemann, & Lepers, 2012c). It can be suggested that mountain biking, running and even more trail running are more traumatic for the muscle than swimming or cycling, and performance would be more altered because of the reduction in muscle mass and strength with aging.

5 Conclusion

The number of finishers drastically fell at the age of 50 and this was true not only for the UTMB® but also for the CCC® and OCC. However, whereas the finishers of the largest age group (23–39 yr) are the most numerous at CCC® and OCC, the largest number of finishers at UTMB® were aged 40–49 yr. Average speed progressively decreased with age in the three races and this decline in performance was accelerated at 50 years old. Comparing the three race courses, participants ran at a lower speed as distance increased, while the oldest age group analyzed (60–69 yr) ran the CCC® and the UTMB® at the same average speed. Future interventional studies need to investigate physiological factors (e.g. neuromuscular fatigue) during ultra-running races such as UTMB®, CCC® and OCC to better understand the causes of the age-related decline in ultra-running performance.

Acknowledgements

The authors gratefully acknowledge the contribution of Pr. Romuald Lepers, who helped to design the data analysis.

Author contributions statement

GV, FSP, GYM, and RV contributed to the conception and design of the work. GV, FSP and RV contributed to the acquisition and analysis. GV, FSP, GYM, and RV interpreted the data. GV and RV drafted the manuscript. GV made the figures and tables. FSP and GYM critically revised the manuscript. All gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy.

References

Cite this article as: Varesco G, Sabater-Pastor F, Millet GY, & Rozand V (2019) Age-related performance at the Ultra-Trail du Mont-Blanc®. Mov Sport Sci/Sci Mot, 104, 13–19

All Tables

Table 1

Number of finishers of each age category and at the UTMB®, CCC® and OCC during the 2014–2018 period.

Table 2

Average time for all the finishers of each age category at the UTMB®, CCC® and OCC during the 2014–2018 period.

All Figures

thumbnail Fig. 1

Percent of finishers per age category in Ultra-Trail du Mont-Blanc® (UTMB®, 171 km, 10 000 m of elevation gain; white), Courmayeur-Champex-Chamonix (CCC®, 101 km, 6100 m of elevation gain; grey) and Orsières-Champex-Chamonix (OCC, 55 km, 3500 m of elevation gain; black).

In the text
thumbnail Fig. 2

Average speed for the Top 10 (A), Top 10% (B) and Overall finishers (C) of each age category in UTMB®, CCC®, and OCC, expressed as percentage of the 23–39 yr group. * = significant difference from 23–39 yr (P < 0.05); # = significant difference from 40–49 yr (P < 0.05); $ = significant difference from 50–59 yr (P < 0.05).

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.