Altitude is Positively Correlated to Race Time during the Marathon

Abstract

Lara, Beatriz, Juan Jose Salinero, and Juan Del Coso. Altitude is positively correlated to race time during the marathon. High Alt Med Biol. 15:64—69, 2014.—Completing a marathon (42.2 km) is one of the more challenging sports activities. Besides the distance, the ambient conditions of the race (altitude, temperature, etc) can increase the physiological demands of the event. The purpose of this study was to investigate the relation between the altitude of the city in which the marathon is held and the marathon race time. For this purpose, we sought the race times of 16 popular marathons performed at different altitudes above sea level (range from ≈0 to 2800 meters above sea level). In these competitions, we analyzed the race times of the female and male runners who finished from 21^st to 100^th position. We excluded the top 20 male and female finishers from the analysis because elite athletes usually compete in marathons held at low altitudes above sea level. Ambient temperature, the positive cumulative elevation gain, and the number of participants were used as control variables. Finishing time in the marathon was positively correlated with the altitude of the competition for both male (r=0.78; p<0.05) and female participants (r=0.73; p<0.05). On average, each increase of 1000 meters above sea level augmented marathon race time by 10.8±0.6% in men and 12.3±0.7% in women. Compared to race times in the Rotterdam marathon (held at 0 meters above sea level), the time taken to complete the marathon was significantly higher in competitions held at an altitude of over 700 meters. In conclusion, the time taken to complete a marathon strongly depends on the altitude of the city in which the marathon is held. Selecting marathon competitions close to 0 m above sea level is a good strategy to maximize marathon performance.

Introduction

Marathon foot races represent one of the most demanding endurance competitions because they greatly stress both cardiovascular and musculoskeletal systems for a long period of time (Del Coso et al., 2013). Despite the severe physical demands of this competition, the attractiveness of the marathon has increased in the last few decades, mostly in the amateur running population that perceive the distance as a challenge. Due to this increase in popularity, many of the capital cities of the world host a marathon race and thousands of elite and amateur runners compete for the fastest finishing times. All the marathon races entail the same running distance (42.2 km) but the physiological demands can be very different depending on the profile of the course, the ambient temperature the day of the race (Vihma, 2010), environmental pollution (Marr et al., 2010), or the altitude of the city in which the marathon is held (Peronnet et al., 1991). In addition, body fat and speed in running training have been identified as primary factors for race performance in the marathon (Rust et al., 2012).

Elevation in altitude is associated with a progressive reduction in barometric pressure and consequently a reduction in air density and partial pressure of oxygen (Mazzeo, 2008). The lower oxygen pressure in the air at altitude produces a concomitant decrease in arterial partial pressure of oxygen, a reduction of arterial saturation and hence in arterial oxygen content (Mazzeo, 2005). At rest, the human body compensates hypoxemia by increasing cardiac output, mechanical ventilation, and catecholamine and cortisol concentrations (Mazzeo, 2008), but full physiological compensation cannot be reached during intense exercise (Calbet et al., 2003; Saunders et al., 2009). The degree of arterial oxygen desaturation correlated with the decline in performance during a 3000-m race completed at altitude (Chapman et al., 2011) and with the decline of Vo_2max during simulated altitude (Ferretti et al., 1997), suggesting that the lessening of blood oxygen saturation caused by altitude is a key factor for the reduced physical performance at altitude. Nevertheless, the deleterious effects of altitude on maximal endurance performance can be lessened by performing altitude acclimatization for at least 14 days (Schuler et al., 2007).

The main physiological determinants of marathon performance are Vo_2max and running velocity at the lactate threshold (Joyner et al., 2011). During exercise, maximal oxygen uptake (Vo_2max) falls progressively with increasing altitude (Mollard et al., 2007) due to the reduced O₂ pulmonary gas exchange in the lungs, but also because cardiac output and leg blood flow are blunted (Calbet et al., 2009). Consequently, performing the same absolute workload at altitude represents a higher relative effort (Wolfel et al., 1998). In a normobaric hypoxic environment (e.g., simulated altitude), the running velocity at the lactate threshold is markedly reduced (Friedmann et al., 2004). Hypoxia similarly reduces both Vo_2max and the lactate threshold and thus, the relative intensity of running expressed as a percentage of Vo_2max can be maintained when running in altitude (Friedmann et al., 2004). However, acclimatization to altitude (5000 m above sea level) for more than one week can produce a reduction in the lactate concentration for given workload, which suggests that acclimatization to altitude represents a meaningful advantage for endurance performance in altitude (Grassi et al., 1996). In addition, altitude increases the reliance on carbohydrate as fuel for the exercising muscle (Roberts et al., 1996) and may produce an earlier depletion of muscle and liver glycogen stores. Finally, acute exposure to altitude frequently produces a pathological effect on humans known as acute mountain sickness that causes an increased prevalence of headaches, nausea, loss of appetite, and sleeplessness (Fulco et al., 2013). All these factors suggest that the altitude of the city in which the marathon race is held may represent a major negative factor for distance running performance.

Peronnet et al. (1991) presented a mathematical model to calculate the effect of altitude on running performance, based on the influence of aerobic and anaerobic metabolism on different athletic disciplines (from 60 m to the marathon). According to this theoretical analysis, altitude may benefit sprint performance since the reduction of aerodynamic resistance present at altitude is more beneficial than the reduction of maximal aerobic power. On the contrary, the reduction in maximal aerobic power with altitude is associated with a reduction in performance over middle and long distances (from 800 m to the marathon). The reduction in air density with increasing altitude would not represent an advantage for long distance runners, while altitude markedly lessens Vo_2max and running velocity at the lactate threshold. Under this mathematical model, race time during a marathon could increase by 7% at 2240 m in both men and women. Roi et al. (1999) determined the effects of running at sea level versus 4300 m and found that the faster runners at sea level were the faster ones at 4300 m, but there was a 35±9% reduction in race time at this altitude.

Nevertheless, there are no data regarding the effects of altitude on race time during a marathon competition. The aim of this investigation was to analyze the finishing times of marathon competitions held in cities situated at different altitudes. A second purpose was to examine any sex difference in marathon performance impairment related to altitude. We hypothesized that marathon competitions held in cities at an altitude of over 1000 m would produce significantly higher race times in both male and female runners.

Material and Methods

Marathon racing performances were collected from 16 competitive races held in cities located at different altitudes above sea level (Table 1). These data are in the public domain and thus, written and informed consent was not required from individual runners. We obtained data from the 100 best performances for men and women to obtain a representative sample of high performance runners in each marathon. We excluded the top 20 finishers in each marathon to eliminate the effect of elite runners who only compete in low altitude marathons. Thus, data for each marathon represent mean±SD for runners between the 21^st and 100^th positions. Finishing race times were obtained for all the races in their 2012 edition, except the New York marathon race (2011 edition). Female data from the Quito marathon corresponded to 13 participants. Mean ambient temperature corresponding to the race day, profile of the course of the race, and the number of participants in each marathon were collected from marathon websites or by contacting marathon organizers. These data were collected to be used as control variables during the statistical analysis. The study was revised by a Research Ethics Committee in accordance with the latest version of the Declaration of Helsinki. The Research Ethics Committee indicated that this investigation did not require approval.

Table 1.

Cities, Dates, Mean Altitude Above Sea Level, Mean Ambient Temperature the Day of the Race, Total Positive Elevation Changes during the Course of the Marathon and Overall Number of Participants in the Analyzed Marathons

City	Country	Date	Altitude (m)	Temperature (°C)	Elevation change (m)	Participants (×1000)
Rotterdam	Netherlands	14 Apr	0	10	60	10
New York	United States	6 Nov	10	12	253	35
London	UK	25 Apr	30	27	220	30
Berlin	Germany	30 Sep	34	17	147	34
Boston	United States	18 Apr	60	17	245	20
Rome	Italy	18 Mar	100	13	308	10
Chicago	United States	7 Oct	182	24	273	33
Prague	Czech Rep.	13 May	200	9	189	7
Madrid	Spain	22 Apr	700	27	151	12
Caracas	Venezuela	26 Feb	800	29	225	6
Salt Lake	United States	21 Apr	1320	27	130	7
Denver	United States	22 Sep	1600	27	291	14
Colorado	United States	6 May	1700	34	69	14
Provo	United States	8 Jun	1833	29	91	2
Mexico city	Mexico	2 Sep	2250	16	90	7
Quito	Ecuador	8 Jul	2800	22	163	2

Mean altitude of the races is an approximation because some marathon organizers did not have the exact data. All this information is related to the 2012 editions, except the New York marathon that corresponds to 2011 (the 2012 edition was cancelled).

One purpose of this study was to describe finishing times in marathons held in cities located at different altitudes. Simple descriptive statistics such as mean and SD were used to illustrate the effect of altitude on marathon performance. In addition, we tested the normality of each variable with the Kolgomorov-Smirnov test. A second purpose was to quantify any decrease in marathon performance associated with altitude; we performed a correlation analysis (r Pearson) between race times in the marathon and the altitude of the city in which the marathon was held. Afterwards, linear regression analysis (adjustment by least squares) was performed to model the relationship between race time and altitude. We used partial correlations to eliminate the effects of ambient temperature, the number of participants, and the total positive elevation changes during the race in the correlation between running performance and altitude. The differences in the finishing times among marathons at different altitudes were examined using two-way ANOVA (altitude×sex). All analyses were performed using the statistical package SPSS 20.0 (SPSS Inc., Chicago, IL). The significance level was set at p<0.05.

Results

The best race time in the marathon was obtained at an altitude of 0 m for men (146±5 min) and 30 m for women (174±7 min). The worst race time in the marathon was obtained at an altitude of 2800 m for both men (202±13 min) and women (295±10 min). In comparison to the race times obtained at 0 m (e.g., the Rotterdam marathon), finishing times were higher in marathon competitions conducted at 700 m or above in both male and female participants (p<0.05). Figure 1 depicts the relationship between the mean race time of male and female runners and the altitude of the city in which the marathon was held. Finishing times and altitude were positively correlated in both male (r=0.78; p<0.05) and female runners (r=0.73; p<0.05). The equation for the estimated race time with the linear regression analysis was:=0.0176×altitude+155.72 for men and 0.0247×altitude+189.17 for women. As a mean, each increase of 1000 m above sea level increased the race time in the marathon by 10.8±0.6% in men and by 12.3±0.7% in women. Race time was negatively correlated with the number of participants in male (r=−0.58; p<0.05) and female runners (r=−0.76; p<0.05). Mean ambient temperature the day of the race was also correlated to race time in male participants (r=0.64; p<0.05) but the correlation was not significant in female runners (r=0.34; p=0.19). Race time in the marathons did not correlate to total positive elevation changes (p>0.05). Race time and altitude were still correlated in men and women after eliminating the effect of ambient temperature and the number of participants by using partial correlations.

FIG. 1.

Race time for marathon races held in cities at different altitudes. Each data point represents the mean±SD for the 21^st to 100^th best race times for men and woman in the 2012 edition. Horizontal error bars represent maximal and minimal altitude of the course. Note: Elite runners (e.g., the top 20 finishers in each marathon) were excluded from the statistical analysis because this type of runner usually avoids races at cities located at moderate or high altitudes.

Discussion

The main finding of this investigation was that the altitude of the city in which the marathon is held was positively correlated to the marathon finishing times in both male and female endurance runners. Elevation in altitude reduces barometric and partial pressure of oxygen in the air and thus oxygen diffusion in the lungs. The main effect of altitude on body homeostasis is the reduced arterial oxygen saturation (Mazzeo, 2005), while a number of essential physiological (e.g., increased heart rate and ventilation) and metabolic adjustments are required to maintain proper tissue oxygenation at rest (Mazzeo, 2008). Nevertheless, when exercising at altitude, there are two different independent stresses to which the body must respond and adapt (Mazzeo, 2008). Major physiological adaptations to altitude (e.g., increased heart rate and ventilation) may be effective to maintain body functioning during submaximal exercise (Lhuissier et al., 2012), but they are insufficient to maintain maximal aerobic capacity (Chapman et al., 2011; Ferretti et al., 1997). Although marathon races are not completed at maximal aerobic capacity, the present investigation demonstrates that altitude greatly influences performance during endurance events.

One potential limitation of this study is that the ambient temperature the day of the race and the number of participants also correlated to the marathon finishing time, as has been found in previous publications (Ely et al., 2008; Hunter et al., 2013). For this reason, we performed a partial correlation analysis that measured the degree of association between race time and altitude without the effect of these variables. With this control, marathon finishing times and the altitude of the city were still significantly correlated (r=0.74 for men and 0.72 for women; p<0.05). This suggests that race time and altitude were associated even without the influence of ambient temperature or the number of participants. Another limitation was related to the experimental design, since different marathoners participated in each competition. To avoid this effect, we selected the top-100 finishers in each marathon, but there was still the possibility that the results of this study could be affected by the different levels of the competitors in each race. To minimize this effect, elite runners (e.g., the top 20 finishers in each marathon) were excluded from the statistical analysis because this type of runner usually avoids races at cities located at moderate or high altitudes. Finally, it is possible that some race participants were high-altitude residents and they could be naturally acclimatized to altitude, improving their exercise capacity for races held at altitude (Brutsaert, 2008; Muza et al., 2010).

Several investigations have demonstrated that altitude or simulated altitude (by setting hypobaric or hypoxic environments) reduces maximal cardiac output (Fukuda et al., 2010), Vo_2max (Mollard et al., 2007), running velocity at the lactate threshold (Wolfel et al., 1998), and might even affect gross efficiency, at least during cycling (Noordhof et al., 2012). Because the main physiological determinants of endurance running are these same elements (Joyner et al., 2011), it could be hypothesized that altitude might reduce running performance in distance athletes. However, there are no data about the magnitude of the effects of altitude on marathon performance, apart from the mathematical calculation reported by Peronnet and co-workers (1991). These authors suggested that race time during a marathon could increase by 7% in a city located at 2240 m above sea level as compared to sea level. The present results confirm the calculations of Peronnet and co-workers, in this case by using real data, because there was a progressive reduction in the marathon performance with altitude (Fig. 1).

The majority of the most popular marathons are held at very low altitudes, which in turn adds to the attractiveness of the races, because finishing times are typically lower. For example, among the 100 best all-time records in the marathon (International Associantion of Athletic Federations 2013), the city with the highest altitude is Prague (≈200 m above sea level) for men and Chicago for women (≈200 m). In addition, most of the best all-time records have been obtained in cities at altitudes lower than 100 m above sea level. Thus, the altitude of the city seems an important factor for obtaining personal best marathon times and this is carefully considered by elite distance runners.

In the present investigation, finishing times were considerable higher in all the marathons conducted above 700 m in both male and female participants (p<0.05), in comparison to the mean race times obtained at 0 m. Although the deleterious effects of altitude on performance are typically related with locations at over 1500 m above sea level (Wilmore et al., 2008) this investigation indicates that competing in a marathon held at over 700 m above sea level can significantly reduce endurance running performance. This information agrees with investigations that found that even a moderate altitude (700–800 m above sea level) can produce a reduction in Vo_2max, endurance running performance (Wehrlin et al., 2006), and blood oxygen saturation (Goldberg et al., 2012). As previously indicated, ambient temperature and the number of participants were significantly associated with marathon finishing times. For this reason, the reduced endurance performance in marathons held above 700 m might be also related to participants' level and the environmental conditions the day of the race.

The effects of altitude on physical performance during large-muscle exercise (e.g., bicycle ergometer or treadmill) are of a similar magnitude in both men and women (Fulco et al., 1998). In fact, sex is the least related factor to the changes in maximal and submaximal exercise performance that occur at different altitudes, when compared to the athlete's fitness level, pre-exposure to altitude, and duration of altitude exposure (Fulco et al., 1998). In the present study, male and female race times were similarly affected by altitude (see slopes of Fig. 1). As a mean, each increase of 1000 m above sea level increased the race time in the marathon by 10.8±0.6% in men and by 12.3±0.7% in women. The slightly higher increase in running time/1000 m of altitude in the female population may be related to lower participation rates and a lower level of talent among women competitors (Hunter et al., 2013). In both male and female runners, the increase in the finishing time per 1000 m of altitude is higher than previously calculated by a mathematical model (e.g., 3.1%; (Peronnet et al., 1991) but similar to the one found during a real competition held at 4300 m above sea level (8.1% per 1000 m increase (Ely et al., 2008)). Figure 2 represents race time for the marathon finishers (from 21^st to 100^th position) in the marathons held at 0, 700, 1600, and 2800 m above sea level. Interestingly, women marathoners at 0 m above sea level obtained similar race times to their male counterparts when competing at 1600 above sea level. Thus, the sex differences in the marathon performance could be offset when male runners participate in competitions at over 1600 m above sea level.

FIG. 2.

Performance of the 21^st to 100^th best race times in marathons held at 0 m (Rotterdam), 700 m (Madrid), 1600 m (Denver), and 2800 m above sea level (Quito). Black-filled shapes represent race times for male runners and white-filled shapes represent race time for female runners. Note: Elite runners (e.g., the top 20 finishers in each marathon) were excluded from the statistical analysis because this type of runner usually avoids races at cities located at moderate or high altitudes.

Conclusion

In summary, marathon race performance strongly depended on the altitude of the city in which the marathon was held. On average, each increase of 1000 m above sea level extended the race time by 10.8±0.6% in men and 12.3±0.7% in women. In comparison to race times of the marathon held at 0 m (Rotterdam), the finishing times in competitions run above 700 m were significantly higher. The magnitude of the deleterious effect of altitude on performance was similar in male and female participants. Selecting marathon competitions close to 0 m is a good strategy for maximizing marathon performance.

Footnotes

Author Disclosure Statement

The authors declare that they have no conflict of interest derived from the outcomes of this study. This study did not receive any funding.

Beatriz Lara was supported by a grant from the Camilo Jose Cela University.

References

Brutsaert

. (2008). Do high-altitude natives have enhanced exercise performance at altitude?. Appl Physiol Nutr Metab, 33:582–592.

Calbet

, Boushel

, Radegran

, Sondergaard

, Wagner

, and Saltin

. (2003). Determinants of maximal oxygen uptake in severe acute hypoxia. Am J Physiol Regul Integr Comp Physiol, 284:R291–303.

Calbet

, and Lundby

. (2009). Air to muscle O2 delivery during exercise at altitude. High Alt Med Biol, 10:123–134.

Chapman

, Stager

, Tanner

, Stray-Gundersen

, and Levine

. (2011). Impairment of 3000-m run time at altitude is influenced by arterial oxyhemoglobin saturation. Med Sci Sports Exerc, 43:1649–1656.

Del Coso

, Salinero

, Abian-Vicen

, et al. (2013). Influence of body mass loss and myoglobinuria on the development of muscle fatigue after a marathon in a warm environment. Appl Physiol Nutr Metab, 38:286–291.

Ely

, Martin

, Cheuvront

, and Montain

. (2008). Effect of ambient temperature on marathon pacing is dependent on runner ability. Med Sci Sports Exerc, 40:1675–1680.

Ferretti

, Moia

, Thomet

, and Kayser

. (1997). The decrease of maximal oxygen consumption during hypoxia in man: A mirror image of the oxygen equilibrium curve. J Physiol, 498:231–237.

Friedmann

, Bauer

, Menold

, and Bartsch

. (2004). Exercise with the intensity of the individual anaerobic threshold in acute hypoxia. Med Sci Sports Exerc, 36:1737–1742.

Fukuda

, Maegawa

, Matsumoto

, Komatsu

, Nakajima

, Nagai

, and Kawahara

. (2010). Effects of acute hypoxia at moderate altitude on stroke volume and cardiac output during exercise. Int Heart J, 51:170–175.

10.

Fulco

, Beidleman

, and Muza

. (2013). Effectiveness of preacclimatization strategies for high-altitude exposure. Exerc Sport Sci Rev, 41:55–63.

11.

Fulco

, Rock

, and Cymerman

. (1998). Maximal and submaximal exercise performance at altitude. Aviat Space Environ Med, 69:793–801.

12.

Goldberg

, Buhbut

, Mimouni

, Joseph

, and Picard

. (2012). Effect of moderate elevation above sea level on blood oxygen saturation in healthy young adults. Respiration, 84:207–211.

13.

Grassi

, Marzorati

, Kayser

, Bordini

, Colombini

, Conti

, Marconi

, and Cerretelli

. (1996). Peak blood lactate and blood lactate vs. workload during acclimatization to 5,050 m and in deacclimatization. J Appl Physiol, 80:685–692.

14.

Hunter

, and Stevens

. (2013). Sex differences in marathon running with advanced age: Physiology or participation?. Med Sci Sports Exerc, 45:148–156.

15.

International Associantion of Athletic Federations. (2013). www.iaaf.org. Accesed on 1/2/2013.

16.

Joyner

, Ruiz

, and Lucia

. (2011). The two-hour marathon: Who and when?. J Appl Physiol, 110:275–277.

17.

Lhuissier

, Brumm

, Ramier

, and Richalet

. (2012). Ventilatory and cardiac responses to hypoxia at submaximal exercise are independent of altitude and exercise intensity. J Appl Physiol, 112:566–570.

18.

Marr

, and Ely

. (2010). Effect of air pollution on marathon running performance. Med Sci Sports Exerc, 42:585–591.

19.

Mazzeo

. (2005). Altitude, exercise and immune function. Exerc Immunol Rev, 11:6–16.

20.

Mazzeo

. (2008). Physiological responses to exercise at altitude: An update. Sports Med, 38:1–8.

21.

Mollard

, Woorons

, Letournel

, et al. (2007). Determinants of maximal oxygen uptake in moderate acute hypoxia in endurance athletes. Eur J Appl Physiol, 100:663–673.

22.

Muza

, Beidleman

, and Fulco

. (2010). Altitude preexposure recommendations for inducing acclimatization. High Alt Med Biol, 11:87–92.

23.

Noordhof

, Schoots

, Hoekert

, and de Koning

. (2013). Is gross efficiency lower at acute simulated altitude as compared to sea level?. Int J Sports Physiol Perform, 8:319–322.

24.

Peronnet

, Thibault

, and Cousineau

. (1991). A theoretical analysis of the effect of altitude on running performance. J Appl Physiol, 70:399–404.

25.

Roberts

, Reeves

, Butterfield

, Mazzeo

, Sutton

, Wolfel

, and Brooks

. (1996). Altitude and beta-blockade augment glucose utilization during submaximal exercise. J Appl Physiol, 80:605–615.

26.

Roi

, Giacometti

, and von Duvillard

. (1999). Marathons in altitude. Med Sci Sports Exerc, 31:723–738.

27.

Rust

, Knechtle

, and Rosemann

. (2012). Similarities and differences in anthropometry and training between recreational male 100-km ultra-marathoners and marathoners. J Sports Sci, 30:1249–1257.

28.

Saunders

, Telford

, Pyne

, Hahn

, and Gore

. (2009). Improved running economy and increased hemoglobin mass in elite runners after extended moderate altitude exposure. J Sci Med Sport, 12:67–72.

29.

Schuler

, Thomsen

, Gassmann

, and Lundby

. (2007). Timing the arrival at 2340 m altitude for aerobic performance. Scand J Med Sci Sports, 17:588–594.

30.

Vihma

. (2010). Effects of weather on the performance of marathon runners. Int J Biometeorol, 54:297–306.

31.

Wehrlin

, and Hallen

. (2006). Linear decrease in VO2max and performance with increasing altitude in endurance athletes. Eur J Appl Physiol, 96:404–412.

32.

Wilmore

, Costill

, and Kenney

. (2008). Physiology of Sports and Exercise. Human Kinetics, Champaign, IL.

33.

Wolfel

, Selland

, Cymerman

, et al. (1998). O2 extraction maintains O2 uptake during submaximal exercise with beta-adrenergic blockade at 4,300 m. J Appl Physiol, 85:1092–1102.