Continuous Thermoregulatory Responses to a Mass-Participation 89-km Ultramarathon Road Race

Click name to view affiliation

Christopher Byrne Public Health and Sport Sciences, Faculty of Health and Life Sciences, University of Exeter, Exeter, United Kingdom

Search for other papers by Christopher Byrne in
Current site
Google Scholar
PubMed
Close
*
,
Aurelien Cosnefroy Public Health and Sport Sciences, Faculty of Health and Life Sciences, University of Exeter, Exeter, United Kingdom

Search for other papers by Aurelien Cosnefroy in
Current site
Google Scholar
PubMed
Close
,
Roger Eston Alliance for Research in Exercise, Nutrition and Activity, University of South Australia, Adelaide, SA, Australia

Search for other papers by Roger Eston in
Current site
Google Scholar
PubMed
Close
,
Jason K.W. Lee Heat Resilience and Performance Center, Human Potential Translational Research Program, and Department of Physiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore
N.1 Institute for Health, National University of Singapore, Singapore
Campus for Research Excellence and Technological Enterprise (CREATE), Singapore

Search for other papers by Jason K.W. Lee in
Current site
Google Scholar
PubMed
Close
, and
Tim Noakes Department of Applied Design, Cape Peninsula University of Technology, Cape Town, South Africa

Search for other papers by Tim Noakes in
Current site
Google Scholar
PubMed
Close
Open access

Purpose: To continuously measure body core temperature (Tc) throughout a mass-participation ultramarathon in subelite recreational runners to quantify Tc magnitude and the influence of aerobic fitness and body fat. Methods: Twenty-three participants (19 men and 4 women; age 45 [9] y; body mass 72.0 [9.3] kg; body fat 26% [6%]; peak oxygen uptake 50 [6] mL·kg−1·min−1) had gastrointestinal temperature measured during an 89-km ultramarathon. Prerace-to-postrace changes in body mass, plasma sodium, and fluid and food recall quantified body water balance. Results: In maximal environmental conditions of 26.3 °C and 53% humidity, 21 of the 23 participants finished in 10:28 (01:10) h:min while replacing 49% (27%) of sweat losses, maintaining plasma sodium (140 [3] mmol·L−1), and dehydrating by 4.1% (1.3%). Mean maximum Tc was 39.0 (0.5) (range 38.2–40.1 °C) with 90% of race duration ≤39.0 °C. Mean maximum ΔTc was 1.9 (0.9) (0.9–2.7 °C) with 95% of race duration ≤2.0 °C. Over 0 to 45 km, associations between ΔTc and peak oxygen uptake (positive) and body fat (negative) were observed. Over 58 to 89 km, associations between Tc and peak oxygen uptake (negative) and body fat (positive) were observed. Conclusions: Modest Tc responses were observed in recreational ultramarathon runners. Runners with higher levels of aerobic fitness and lower levels of body fat demonstrated the greatest changes in Tc during the first half of the race. Conversely, runners with lower levels of aerobic fitness and higher levels of body fat demonstrated the greatest absolute Tc in the final third of the race.

Ultramarathons, defined as running events longer than the marathon distance of 42.195 km, are characterized by a diversity of distance, duration, topography, and environmental conditions.1 A key consideration for the ultramarathon runner is the regulation of body core temperature (Tc), as the magnitude of Tc elevation during prolonged exercise has the potential to impact both performance and health.2 Tc and sweat rate during running are primarily determined by metabolic heat production.3,4 Postmarathon Tc typically averages 39 °C with sweat rates over 1 L·h−1 and dehydration of approximately 4% of body mass.4,5 In marathon races, the fastest runners display the highest Tc and greatest dehydration,5 subsequent to greater rates of heat production and required evaporative heat loss. However, few studies and limited data exist for the thermoregulatory responses to ultramarathon running.6

Mass-participation distance running is associated with the greatest risk of exertional heat illness in organized sports, with an estimated prevalence of 6.7%.7 Exertional heat stroke is the rarest and most severe form of heat illness and is characterized by central nervous system dysfunction and Tc exceeding 40.5 °C.7 Improved knowledge of thermoregulatory function during ultramarathon running will better inform athlete preparation and medical risk management. On the one hand, Tc elevation during ultramarathon running may be modest compared with shorter distance events as metabolic heat production (running speed) is lower with increasing distance.8 This may explain the handful of ultramarathon studies reporting modest Tc responses (ie, 36.6 to 39.5 °C).912 On the other hand, numerous factors specific to or exacerbated by ultramarathon running have the potential to elevate Tc via effects on increasing metabolic heat production and/or impairing body heat loss.6 For example, heat production may increase for a given running speed due to the impaired running economy associated with peripheral muscle fatigue and muscle damage, resulting in a higher Tc if speed is maintained.13,14 In addition, greater exposure to maximal daylight environmental conditions may increase skin temperature, reduce the core-to-skin temperature gradient, and necessitate an increase in Tc to reduce whole-body skin blood flow requirements for heat loss, if speed is maintained.15 Furthermore, ultramarathon runners are a population with a wide range of aerobic fitness and body morphology attributes.1 Higher aerobic fitness will be expected to lead to greater ΔTc due to higher rates of metabolic heat production.3 Alternatively, higher levels of body fat produce small but significant increases in Tc for a given level of heat production,16 and this may be exacerbated by the prolonged nature of ultramarathon running.

This study aimed to advance knowledge of the thermoregulatory responses to ultramarathon running that is currently based upon single postrace Tc measures9,10 or few intrarace serial Tc measures.12 The objectives were to continuously measure Tc throughout an ultramarathon in a sample of recreational runners, to quantify the magnitude and pattern of response, to estimate body water balance, and to investigate the influence of aerobic fitness, body fat, and body water balance on the Tc response.

Methods

Participants, Design, and Setting

Twenty-three recreational distance runners (19 men and 4 women), who were registered entrants in an 89-km ultramarathon road race, volunteered with written informed consent to participate in this study. The study procedures were approved by the ethics committees of the School of Sport and Health Sciences at the University of Exeter, United Kingdom, and the Department of Exercise Science and Sports Medicine at the University of Cape Town, South Africa.

The study design represents an observational cross-sectional study of recreational ultramarathon runners employing convenience sampling. The design consisted of participants undertaking a single laboratory visit for physiological assessment followed by field-based physiological measurements during the Comrades Marathon on May 24, 2009. Twenty volunteers were recruited by responding to advertisements to Cape Town-based running clubs, and 3 were recruited by responding to advertisement at the prerace Expo in Durban. Comrades is an annual mass-participation ultramarathon road race of approximately 89 km (56 miles) distance between the cities of Pietermaritzburg and Durban in KwaZulu-Natal Province, South Africa. It is considered the world’s oldest (first race in 1921) and largest (up to 20,000 participants) ultramarathon. Participants are required to complete the distance within 12 hours to receive an official finisher’s medal. The start and finish of the race alternates annually between the 2 cities. The race under study, started at 05:30 hours in Pietermaritzburg (650-m altitude), finished in Durban (sea level), and is considered a “down run.”

Methodology

Laboratory Measurements

Twenty of the 23 participants attended the laboratory at the Department of Exercise Science and Sports Medicine at the University of Cape Town (Sports Science Institute of South Africa) on a single occasion 21 (10) (10–52) days before the race for assessment of their anthropometry, running economy, and peak oxygen uptake (VO2peak). Three participants recruited at the prerace Expo did not attend the laboratory due to insufficient time before the race. Percentage of body fat was estimated from the sum of 4 skinfolds using the equations of Peterson et al.17 Skinfold measurements were made in duplicate by one researcher with the mean of the 2 measurements accepted as the criterion for each site. An incremental treadmill run until volitional exhaustion was performed with VO2 and heart rate (HR) measured continuously using a breath-by-breath metabolic cart (Jaeger Oxycon Pro 2, Erich Jaeger GmbH) and short-range telemetry (Suunto T6, Suunto), respectively.18 Individual running speed–VO2 regression equations were established from 30-second average VO2 values (R2 = .98 [.02]) for the prediction of VO2 during the race when speed was known. Running economy was defined as the VO2 at 10 km·h−1 and expressed in milliliters per kilogram per minute and milliliters per kilogram per kilometer. VO2peak and maximal HR (HRmax) were defined as the highest 30-second average values observed during the test. Velocity at VO2peak (vVO2peak) was considered the functional expression of VO2peak.18 Peak treadmill velocity was defined as the highest running speed maintained for 60 seconds.

Field Measurements

Participants reported to a research station near the start line within 90 minutes of the official 05:30 hours race start time for prerace measurements and equipment fitting. Prerace hydration status was assessed by the osmolality of waking urine (Osmocheck, Vitech Scientific Ltd). Venous blood was sampled prerace and postrace for the analysis of plasma sodium using an automated analyzer (EasyLyte, Medica Corp). Prerace and postrace race body mass was measured in duplicate to the nearest 0.100 kg with participants in minimal attire (ie, shorts and vest) and having toweled dry. Participants ingested a telemetric temperature sensing capsule (VitalSense, Philips Respironics) prior to sleeping (approximately 7.5 h before race start) and during the race wore a 250-g data recorder for the continuous measurement of gastrointestinal temperature as an index of Tc.19 Sensor ingestion timing aimed to ensure sensor transit beyond the stomach but not expulsion before data collection.19 A HR telemetry system was fitted to 12 randomly selected participants (Polar Vantage, Polar Electro Oy), due to financial constraints limiting equipment availability. One participant was fitted with a global positioning system running computer (Garmin Forerunner 305, Garmin International Inc) for the measurement of race distance and elevation. Each runner wore the race organizers timing chip system (ChampionChip) on their shoes, which provided split times at 26.77, 44.97, 58.27, 70.97, 82.17, and 89.17 km (finish line). Upon crossing the finish line, participants walked approximately 100 m to the study research station in the race organizers medical tent for immediate postrace measurements. Finally, participants completed a fluid and food recall survey to estimate intake during the race. Environmental conditions for the race were provided by the South African Weather Service.

Data Processing

The Tc and HR data represent 60-second average values of data recorded at 15- and 5-second intervals, respectively. Complete Tc data were recorded in 19 of the 23 participants. Mean Tc during the 60 seconds prior to the race start represented baseline Tc. The loss of 4 Tc data sets was due to complete and unexplained Tc data recording failure in 2 data sets, data loss after 35% of race duration in 1 data set possibly due to excretion of the sensor, and the confounding effect of fluid intake affecting 66% of data in a further data set. Complete HR data were recorded in 9 of the 12 participants fitted. The 3 data sets were lost after 320, 380, and 380 minutes possibly due to battery failure. HR data were expressed relative to laboratory measured HRmax. Mean VO2 per liter per minute was predicted from laboratory-derived speed–VO2 relationships solved for mean 89-km race speed. Mean metabolic heat production was predicted from the product of mean VO2, the energy equivalent of VO2 (assuming a nonprotein respiratory exchange ratio of 0.85, ie, 20.375 kJ·VO2 L·min−1), and conversion factor 16.667; and expressed in absolute terms (in watts), relative to body mass (in watts per kilogram), and relative to body surface area (in watts per meter square). Dehydration was calculated in classic form as the prerace to postrace body mass change expressed as a percentage of the prerace mass. Dehydration was also calculated in contemporary form by correcting for respiratory and gas exchange losses as 0.20 g·kcal−1 of total energy expenditure and expressed as a percentage of the prerace mass.20 Total energy expenditure (in kilocalories) was estimated as the product of mean VO2, race duration, and the energy equivalent of VO2 (assuming a nonprotein respiratory exchange ratio of 0.85, ie, 4.862 kcal·VO2 L·min−1). Whole-body sweat loss (in liters) was calculated as the sum of corrected body mass change and the mass of estimated food and fluid intake. Urine and fecal losses were not recorded or accounted for in the estimation of dehydration and sweat loss. Sweat rate (in liters per hour) was calculated as the sweat loss expressed relative to race duration.

Split times enabled the 89.17-km race and measured data to be split into 6 split sections (S1–S6): S1 = 0 to 26.77 km (26.77 km, 30.0% of total distance), 0.5% mean gradient, 2:52 (0:25) hours:minutes duration; S2 = 26.77 to 44.97 km (18.2 km, 20.4%), −0.5%, 2:04 (0:16) hours:minutes; S3 = 44.97 to 58.27 km (13.3 km, 14.9%), −0.5%, 1:41 (0:11) hours:minutes; S4 = 58.27 to 70.97 km (12.7 km, 14.2%), −2.2%, 1:33 (0:12) hours:minutes; S5 = 70.97 to 82.17 km (11.2 km, 12.6%), −2.0%, 1:28 (0:12); and S6 = 82.17 to 89.17 km (7.0 km, 7.9%), −1.5%, 0:52 (0:08) hours:minutes.

Statistical Analyses

Data were analyzed with SPSS IBM (version 26). Descriptive data are presented as mean (SD) (range). Statistical significance was accepted as P < .05 for all tests. Single-factor repeated-measures analysis of variance with Bonferroni follow-up tests investigated changes in variables (ie, running speed, pacing, %HRmax, mean Tc, maximum Tc, mean ΔTc from baseline, maximum ΔTc from baseline, within-split maximum positive ΔTc, and within-split maximum negative ΔTc) across the 6 split sections. Paired samples t tests analyzed prerace to postrace changes in means for body mass and plasma sodium. Identification of Tc responses classified as meaningful observations was achieved by converting individual maximum Tc data per split and within-split maximum positive and negative ΔTc per split into z scores. An individual z score > 1.96 was considered a meaningful observation. Pearson correlation coefficient and the coefficient of determination (R2) were employed to determine the relationship and strength of association between selected variables. Meaningful relationships were identified as those with a correlation coefficient of r ≥ .50 (R2 ≥ .25), representing a large effect size.21

Results

Participant Characteristics, Course Profile, and Environmental Conditions

Table 1 illustrates the participant’s physical and physiological characteristics. Total race distance was 89.17 km of undulating terrain (3208-m ascent and 3829-m descent) with a net descent of 621 m (Figure 1A). Dry bulb temperature and relative humidity ranged from a minimum of 11.7 °C and 95% at the start (05:30 h) to a maximum of 26.3 °C and 53% after 7.5 hours (13:00 h). Table 2 illustrates the sample demographics and distribution of finishing times in relation to the race population.

Table 1

Physical and Physiological Characteristics of Study Participants (Mean [SD])

Males

(n = 19)
Females

(n = 4)
Sample

(N = 23)
Age, y47 (8)36 (11)45 (9)
Stature, m1.76 (0.07)1.58 (0.04)1.73 (0.1)
Body mass, kg73.8 (7.2)58.6 (9.2)72.0 (9.3)
BMI, kg·m−224.2 (3.2)24.0 (3.2)24.2 (3.1)
Body fat, %24 (5)32 (4)26 (6)
VO2peak, mL·kg−1·min−151 (6)46 (4)50 (6)
Running economy at 10 km·h−1
 mL·kg−1·min−133 (3)35 (2)34 (3)
 mL·kg−1·km−1200 (17)209 (12)202 (16)
vVO2peak, km·h−115.3 (1.7)13.8 (1.1)15.0 (1.7)
Peak treadmill velocity, km·h−115.8 (1.5)13.8 (1.2)15.4 (1.7)

Abbreviations: BMI, body mass index; vVO2peak, velocity at VO2peak; VO2peak, peak oxygen uptake.

Figure 1
Figure 1

—Course profile illustrating distance versus elevation and the 6 (S1–S6) split sections (A); mean (SD) running speed for each split section (B, n = 21); mean (SD) pacing as percentage of mean 89-km speed for each split section (C, n = 21); and mean (SD) %HRmax for each split section (D, n = 8). HRmax indicates maximal heart rate; S, split section value. S is significantly greater than designated number split section(s), P < .05.

Citation: International Journal of Sports Physiology and Performance 17, 11; 10.1123/ijspp.2022-0043

Table 2

Demographics of Study Participants and Total Comrades Marathon Race Entrants

Study participantsRace entrants
Starters/finishers, n23/2011,345/10,005
Finishers, %87.088.2
Finishers (male/female, n)17/38256/1749
Finishers (male/female, %)85.0/15.082.5/17.5
Finish time 05:23 < 07:30 h, %05.8
Finish time 07:30 < 09:00 h, %20.019.3
Finish time 09:00 < 11:00 h, %50.047.0
Finish time 11:00 < 12:00 h, %30.027.9

Performance, Pacing, and Exercise Intensity

Twenty of the 23 participants completed the distance within 12 hours, 1 participant completed the distance 3 minutes after the 12-hour cutoff, and 2 participants voluntarily withdrew after 08:01 and 11:21 hours:minutes, respectively. Table 3 illustrates mean finishing time, running speed, estimated intensity, and HR for the finishers. Figure 1B and 1C illustrates the running speed and pacing profile, respectively. Figure 1D illustrates %HRmax across the race.

Table 3

Environmental Conditions, Finishing Time, Running Speed, and Exercise Intensity During the 89-km Ultramarathon

Mean (SD)Range
Dry bulb temperature, °C19.9 (5.5)11.7 to 26.3
Relative humidity, %70 (19)53 to 95
Finish time, h:min10:28 (01:10)08:25 to 12:03
Running speed, km·h−18.6 (1.0)7.4 to 10.6
VO2, mL·kg−1·min−130 (4)23 to 37
%VO2peak60 (9)46 to 77
%Peak treadmill velocity56 (5)46 to 65
Metabolic heat production
 W717 (129)472 to 961
 W·kg−110.2 (1.2)7.9 to 12.5
 W·m−2392 (51)301 to 484
HR
 Beats·min−1148 (9)133 to 158
 %HRmax79 (3)76 to 84
Race time in %HRmax zone, %
 40% < 55%0.4 (0.8)0 to 2.5
 55% < 70%6.7 (8.8)0.2 to 26.0
 70% < 80%39.2 (13.8)22.0 to 56.0
 80% < 90%48.2 (14.1)32.5 to 72.1
 ≥90%5.2 (7.2)0 to 19.4

Abbreviations: HR, heart rate; HRmax, maximal HR; VO2, oxygen uptake; VO2peak, peak VO2. Note: Values represent n = 21 for finish time and running speed; n = 18 for VO2, %VO2peak, %peak treadmill velocity, and heat production; and n = 8 for HR and %HRmax.

Mean race speed demonstrated positive relationships with peak treadmill velocity (R2 = .51, P = .001), vVO2peak (R2 = .48, P = .001), and VO2peak (R2 = .43, P = .003); and a negative relationship with body fat percentage (R2 = .35, P = .007).

Thermoregulatory Responses

Body Tc

Figure 2A2D illustrates the individual mean Tc and ΔTc per hour. Table 4 provides a summary of Tc and ΔTc variables and the proportion of race time in Tc and ΔTc zones. All participants exhibited maximum Tc > 38.0 °C, 95% ≥38.5 °C, 47% ≥39.0 °C, 10.5% ≥39.5 °C, and a single participant exceeded 40.0 °C (ie, 40.1 °C). The 2 nonfinishers exhibited unremarkable Tc of 38.5 °C (ΔTc = 1.0 °C) and 38.4 °C (ΔTc = 0.9 °C) at withdrawal, having spent the majority of race time (≈75%) with Tc < 38.5 °C. Their peak Tc and ΔTc during the race were 39.0 °C (ΔTc = 1.6 °C) and 38.8 °C (ΔTc = 1.6 °C).

Figure 2
Figure 2

—Individual baseline and mean Tc (A), individual baseline and peak Tc (B), mean ΔTc from baseline (C), and peak ΔTc from baseline (D) per hour of running for 17 finishers (n = 13 males ○ and n = 4 females ▴) and 2 nonfinishers (n = 2 males •). Tc indicates core temperature.

Citation: International Journal of Sports Physiology and Performance 17, 11; 10.1123/ijspp.2022-0043

Table 4

Overview of Tc Responses During the 89-km Ultramarathon Race

Mean (SD)Range
Baseline Tc, °C37.1 (0.3)36.6 to 37.9
Tc rate of rise in 30 min, °C·min−10.05 (0.01)0.01 to 0.07
Mean Tc, °C38.4 (0.3)37.8 to 39.1
Maximum Tc, °C39.0 (0.5)38.2 to 40.1
Final Tc, °C38.6 (0.6)37.7 to 40.1
Mean ΔTc, °C1.3 (0.3)0.6 to 1.7
Maximum ΔTc, °C1.9 (0.9)0.9 to 2.7
Final ΔTc, °C1.5 (0.5)0.8 to 2.7
Race time in Tc zone, %
 <38.0 °C18.5 (21.2)1.5 to 82.0
 >38.0 ≤ 38.5 °C47.4 (20.3)7.3 to 73.1
 >38.5 ≤ 39.0 °C24.5 (19.6)0 to 66.6
 >39.0 ≤ 39.5 °C8.1 (15.4)0 to 48.2
 >39.5 < 40.0 °C1.3 (5.4)0 to 22.3
 ≥40.0 °C0.1 (0.3)0 to 1.3
Race time in ΔTc zone, %
 ≤0.5 °C4.1 (6.0)1.4 to 26.7
 >0.5 ≤ 1.0 °C20.8 (28.1)1.0 to 82.3
 >1.0 ≤ 1.5 °C39.5 (20.9)0 to 67.9
 >1.5 ≤ 2.0 °C31.0 (26.0)0 to 82.7
 >2.0 ≤ 2.5 °C4.2 (8.0)0 to 28.0
 ≥2.5 °C0.2 (1.0)0 to 4.2

Abbreviation: Tc, body core temperature. Note: Values represent n = 17 for race finishers Tc.

Figure 3 illustrates box plots of peak Tc and ΔTc data sets across the 6 split sections. Maximum Tc (Figure 3A) and maximum ΔTc from prerace baseline (Figure 3B) were similar across split sections (P = .429). Within-split maximum positive ΔTc (Figure 3C) was greater in S1 versus S2 to S6 (P ≤ .001), and within-split maximum negative ΔTc was greater in S1 and S3 versus S6 (P ≤ .014; Figure 3D). Within S1, 81 (17) (50%–100%) of the maximum ΔTc for the entire race was achieved in the first 60 minutes of running. The net change in Tc over S2 to S6 (27–89 km) was 0.04 (0.51) (−0.87 to 0.89 °C).

Figure 3
Figure 3

—Box plots of data sets for peak Tc (A), peak ΔTc from baseline Tc (B), peak positive ΔTc within each split (C), and peak negative ΔTc within each split (D) across the 6 split sections for 17 finishers. Box plots represent minimum and maximum values (whiskers); 25th percentile, median (line), mean (×), 75th percentile, and interquartile range (box). Data points that exceed 1.5 times the interquartile range are considered outliers and are illustrated outside the whiskers.

Citation: International Journal of Sports Physiology and Performance 17, 11; 10.1123/ijspp.2022-0043

Maximum Tc responses identified as meaningful observations were observed in 3 runners. These were in 4 consecutive split sections for 1 runner (ie, S3 = 39.6 °C; S4 = 39.7 °C; S5 = 39.7 °C; and S6 = 40.1 °C); 2 split sections for a further runner (ie, S1 = 39.4 °C and S5 = 39.6 °C); and a single split section in another runner (ie, S2 = 39.3 °C). Within-split peak positive ΔTc responses identified as meaningful were observed in 4 runners, including the 2 latter runners above. One runner experienced a 1.0 °C change in S2 in 17 minutes (0.058 °C·min−1) to 39.3 and 0.9 °C in S3 in 19 minutes (0.047 °C·min−1) to 39.2 °C. A further runner produced a change of 1.1 °C in S6 to 38.8 °C, with 0.8 °C of the 1.1 °C change occurring in the final 13 minutes of the race (0.062 °C·min−1). In 2 further runners, similar magnitudes of change to the above (S4 = 0.96 °C and S5 = 0.81 °C), albeit at slower rates (S4 = 0.013 °C·min−1 and S5 = 0.015 °C·min−1), produced peak Tc of 39.3 and 39.0 °C, respectively.

Correlates of Tc

The ΔTc in the first half of the race (ie, S1–S2) was positively related to measures of aerobic fitness and negatively related to body fat and baseline Tc. For example, S1 mean ΔTc was positively related to VO2peak (R2 = .29, P = .027) and negatively related to % body fat (R2 = .37, P = .008) and baseline Tc (R2 = .45, P = .002). The proportion of race time with ΔTc 1.0 to 2.0 °C was also positively related to measures of aerobic fitness (eg, peak treadmill velocity R2 = .31, P = .026) and negatively related to baseline Tc (R2 = .60, P ≤ .001).

The Tc in the last third of the race (ie, S4–S6) demonstrated positive relationships with measures of body fat and baseline Tc and negative relationships with measures of aerobic fitness. For example, S6 peak Tc was positively related to body mass index (R2 = .43, P = .004) and baseline Tc (R2 = .25, P = .041) and negatively related to vVO2peak (R2 = .52, P = .002).

Baseline Tc was a consistent correlate of both ΔTc (negative) and Tc (positive) responses. In turn, baseline Tc was positively related to measures of body fat (eg, body mass index R2 = .56, P ≤ .001; %fat R2 = .54, P ≤ .001) and negatively related to measures of aerobic fitness (eg, VO2peak R2 = .47, P = .001; vVO2peak R2 = .39, P = .005).

Body Water Balance

Table 5 illustrates body water balance variables in response to the race. Body mass was reduced postrace (P ≤ .001). Plasma sodium was unchanged (P = .988), with individual changes ranging from −8 to 7 mmol·L−1. The proportion of the sample with dehydration < 2%, 2% < 3%, 3% < 4%, and >4% was 5%, 15%, 30%, and 50%, respectively. The proportion of the sample with corrected dehydration <2%, 2% < 3%, 3% < 4%, and >4% was 45%, 15%, 25%, and 15%, respectively. One participant who withdrew at 11:21 hours:minutes demonstrated unremarkable values for corrected body mass loss (1.8 kg), corrected dehydration (1.4%), postplasma sodium (143 mmol·L−1), plasma sodium change (−0.6 mmol·L−1), sweat rate (0.7 L·h−1), and sweat replaced (73%).

Table 5

Body Water Variables in Response to the 89-km Ultramarathon Race

Mean (SD)Range
Prerace urine osmolality, mOsmol·kg−1588 (223)250 to 1090
Prerace plasma sodium, mmol·L−1140 (3)133 to 145
Postrace plasma sodium, mmol·L−1140 (3)135 to 145
Change in plasma sodium, mmol·L−10 (4)−8 to + 7
Body mass loss, kg2.9 (1.0)1.5 to 5.4
Dehydration, %4.1 (1.3)1.9 to 7.1
Corrected body mass loss, kg1.7 (1.0)0.2 to 4.0
Corrected dehydration, %2.3 (1.3)0.3 to 5.3
Food intake, kg0.6 (1.0)0.1 to 4.0
Fluid intake, L3.3 (2.9)0.1 to 10.6
Fluid intake rate, L·h−10.3 (0.3)0 to 1.0
Sweat loss, L5.8 (3.0)2.0 to 11.4
Sweat rate, L·h−10.6 (0.3)0.2 to 1.1
Sweat loss replaced, %49 (27)3 to 93

Note: Values represent n = 21 for urine osmolality, body mass loss, and dehydration; n = 18 for serum sodium; and n = 15 for food, fluid, and sweat loss variables.

Corrected dehydration was positively related to plasma sodium change (R2 = .47, P = .002), VO2peak (R2 = .39, P = .006), and mean race speed (R2 = .25, P = .023). Corrected dehydration was negatively related to baseline Tc (R2 = .27, P = .019), S4 peak Tc (R2 = .28, P = .024), and S5 mean and peak Tc (R2 = .30, P = .023; R2 = .35, P = .013).

Discussion

Our main finding is that recreational runners in a mass participation 89-km road-based ultramarathon lasting up to 12 hours in mild environmental heat, display thermoregulatory responses within normal limits. The incidence of marked hyperthermia (ie, Tc ≥ 40 °C)22 in this subelite population appears low as evidenced by a single observation (ie, Tc = 40.1 °C) in the current study.

The novel feature of the current study was the continuous measurement of Tc throughout an 89-km ultramarathon. Our data are in general agreement with Dancaster and Whereat9 who provide the only comparative Tc data for the Comrades marathon and reported postrace (5–10 min) rectal temperatures in the range 38.3 to 38.9 °C in 9 runners following an “up run.” We observed mean final Tc of 38.6 (0.6 °C) and 59% of final Tc values and 72% of race time was spent within the Tc range 38.0 to 38.9 °C. However, we did observe a greater upper range of final Tc (21% ≥39.0 °C) and maximum Tc (47% ≥39.0 °C). Taken together, both studies suggest the majority of Comrades runners experience a modest level of hyperthermia. Our observations are also consistent with the modest levels of hyperthermia reported in the existent ultramarathon evidence base. For example, postrace Tc ranged from 38.7 to 39.2 °C following 56 km (3 h 48 min to 5 h 24 min) of road running,10 37.2 to 39.4 °C during 161 km (26 h 48 min) of trail running,12 and peaked at 39.0 to 39.5 °C during 217 km (36 h) of desert running.11

A novel observation was that baseline Tc was inversely related to ΔTc and directly related to measures of Tc. This apparent paradox (ie, higher baseline Tc not only associated with lower ΔTc but also higher Tc) is explained by the interaction of baseline Tc, aerobic fitness, body fat, and the parts of the race when these relationships were significant (ie, 0–45 km for baseline Tc: ΔTc relationship; and 58–89 km for baseline Tc: absolute Tc relationship). Baseline Tc, measured at 05:30 hour when air temperature was a cool 11.7 °C, was directly related to measures of body fat and inversely related to measures of aerobic fitness. Two independent or interacting factors could potentially explain these relationships: (1) the role of subcutaneous fat and muscle tissue providing thermal insulation and limiting heat loss and Tc decline during cold air exposure before the race23 and (2) lower resting Tc representing a biomarker of enhanced heat adaptation and endurance training status.24,25 The lack of standardization in baseline Tc measurement means that prerace physical activity and clothing are potential confounding influences. However, the relationships were robust, linear, uninfluenced by outliers, and warrant confirmation from future well-controlled field or laboratory studies.

Our data indicate that individuals with lower baseline Tc had lower levels of body fat, higher levels of aerobic fitness, and displayed higher changes in Tc from baseline in the first 45 km of the race. However, due to their lower baseline Tc, the leaner fitter runner’s absolute Tc responses were not elevated above the fatter,less fit runner’s absolute Tc responses. Conversely, individuals with a higher baseline Tc had higher levels of body fat, lower levels of aerobic fitness, and displayed higher absolute levels of Tc in the 58 to 89 km part of the race. However, due to their higher baseline Tc, the fatter-less fit runner’s ΔTc responses were not elevated above the leaner-fitter runner’s ΔTc responses. These latter relationships were evident during a part of the race characterized by downhill running (Figure 1A, S4–S6). It is possible that the consequences of downhill running on heat production and subsequent Tc were greater for the fatter-less fit runners than the leaner-fitter runners.

Marked hyperthermia (ie, Tc ≥ 40 °C) was observed in a single participant for the final 8 minutes (1.3%) of their 10-hour 14-minute race. This runner exhibited Tc ≥ 39 °C for the final 7 hours 19 minutes (71.5%) of their race resulting in the highest final and peak Tc (40.1 °C) and ΔTc (2.7 °C) of the sample. Notable features from this male runner’s physical and physiological profile in comparison with the sample (Tables 1 and 3) include a high baseline Tc (37.4 °C), high level of adiposity (body mass index 28.0; %fat 29.1), low aerobic fitness parameters (VO2peak 45 mL·kg−1·min−1; running economy 236 mL·kg−1·km−1; and vVO2peak 11.6 km·h−1), and high estimated race VO2 (35 mL·kg−1·min−1), relative intensity (77 %VO2peak), and metabolic heat production (12 W·kg−1), despite a modest race speed (8.7 km·h−1). The role of poor running economy in elevating heat production and ΔTc for a given running speed3 and higher levels of body fat producing small but significant increases in ΔTc for a given heat production16 have been demonstrated. The complex interaction between metabolic heat production, body morphology, and the physical properties of the skin and environment determining heat loss are considered the principal components determining Tc responses to exercise26 and are likely determinants of the individual variability in Tc responses observed in the current study (Figures 2 and 3).

Our estimated body water balance responses (ie, body mass loss = 4.1% [1.3%] [1.9%–7.1%]) were in general agreement with previous Comrades studies.9,27 We did not record or account for urine and fecal losses, and therefore, a slight overestimation of dehydration and sweat loss will be inherent in our results. Dancaster and Whereat9 observed body mass losses of 5.2% (1.7%) (0.7%–8.2%) and  Kelly and Godlonton27 observed body mass losses of 3.6% (maximum 7.1%). The sweat losses and fluid replacement data of Dancaster and Whereat9 (ie, sweat rate = 0.88 [0.24] [0.51–1.49] L·h−1; fluid intake rate = 0.48 [0.30] [0.03–1.37] L·h−1; and sweat loss replaced = 53% [20%] [5%–93%]) were also similar to our observations (ie, sweat rate = 0.6 [0.3] [0.2–1.1] L·h−1; fluid intake rate = 0.3 [0.3] [0–1.0] L·h−1; and sweat loss replaced = 49% [27%] [3%–93%]). In addition, plasma sodium was well regulated within the 135 to 145 mmol·L−1 normal range in all participants (ie, postrace plasma sodium = 140 [3] [135–145] mmol·L−1). This indicator of euhydration was in the presence of 95% of participants exhibiting body mass loss ≥ 2.0% with 50% ≥4.0%. This supports the view that maintaining body mass loss < 2% for body water homeostasis during ultra-endurance exercise is not warranted.28 Moreover, dehydration did not appear to negatively impact Tc or performance. Indeed, dehydration demonstrated inverse relationships with S4 and S5 Tc and was positively related to mean race speed (R2 = .25, P = .023). The latter finding supports previous observations on 100-km ultramarathon29 and 42.2-km marathon30 races, that faster runners exhibit the greatest levels of dehydration, suggesting their performance is not impaired by their greater dehydration.

The physiological correlates of performance in the current study (ie, peak treadmill velocity, vVO2peak, VO2peak, and %fat) are in good agreement with previous studies.31 The negative impact of body fat on marathon and ultramarathon32 performance was confirmed by our observation of an inverse relationship between %fat and mean speed. The observed progressive reduction in speed, representing a positive pacing pattern, is consistent with previous pacing observations on marathon and ultramarathon races ranging from 100 to 161 km.33,34 While the mechanisms underlying this pacing pattern were not the focus of this study, it appears that marked hyperthermia is not a contributing factor.

Practical Applications

Our study revealed that 89-km ultramarathon running in temperatures up to 26 °C is not associated with high body temperatures, suggesting that specific heat mitigation strategies (eg, precooling, per-cooling, aggressive hydration) may have limited effectiveness compared to warmer environmental conditions. Nevertheless, we revealed that modifiable internal factors such as high body fat and lower aerobic fitness were associated with higher body temperatures later in the race. Indeed, these same internal factors were associated with poorer ultramarathon running performance, just as they are with distance running over shorter distances. This suggests that dual benefits on performance and body temperature regulation are likely to result from preparation strategies that produce positive adaptations in body fat and markers of aerobic fitness. Runners who withdrew from the race appeared to run at an intensity lower than runners who completed the race and exhibited unremarkable body temperature and hydration responses suggesting their performance limitation was not of a thermoregulatory origin. Due to ultramarathons being characterized by a diversity of distance, duration, topography, and environmental conditions, our study findings are generalizable to the specific demands of the Comrades Marathon event. Our results should be viewed in the context of the modest distance (89.17 km), duration (≤12 h), topography (tarmacadam running surface, 3208-m ascent, and 621-m net descent), and environmental heat (≤26.3 °C) of the Comrades event versus the more extreme ultramarathon running races.1,6 Our small sample size was representative of the subelite recreational male and female Comrades runner. Future research focusing on the thermoregulatory responses of elite and subelite male and female runners across the full spectrum of ultramarathon events is warranted.

Conclusions

By measuring Tc continuously throughout an 89-km mass-participation ultramarathon, we revealed thermoregulatory responses within normal limits in a representative sample of subelite runners. The greatest and most consistent changes in Tc occurred during the first hour of running. While evidence of heat storage and meaningful changes in Tc were revealed thereafter, this typically did not manifest in marked hyperthermia except in a single case of 40.1 °C. Runners with lower body fat and higher aerobic fitness demonstrated the greatest changes in Tc during the first half of the race. These faster runners demonstrated the greatest degree of dehydration. Conversely, runners with higher body fat and lower aerobic fitness demonstrated the greatest Tc in the final third of the race and demonstrated lower levels of dehydration. In this study, faster runners did exhibit greater dehydration, but did not display greater absolute Tc responses due in part to a lower starting Tc, which appeared to reflect lower body fat and higher aerobic fitness and/or heat adaptation status.

Acknowledgments

The authors would like to thank Theresa Mann, Elske Schabort, and Nicholas Tam for their help with data collection, and Dr Jeremy Boulter and the Comrades Marathon medical team for accommodating our research stations. Jason Lee was supported by the National Research Foundation, Prime Minister’s Office, Singapore, under its Campus for Research Excellence and Technological Enterprise (CREATE) program.

References

  • 1.

    Knechtle B, Nikolaidis PT. Physiology and pathophysiology in ultra-marathon running. Front Physiol. 2018;9:634. PubMed ID: 29910741 doi:10.3389/fphys.2018.00634.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2.

    Sawka MN, Leon LR, Montain SJ, Sonna LA. Integrated physiological mechanisms of exercise performance, adaptation, and maladaptation to heat stress. Compr Physiol. 2011;1(4):18831928. PubMed ID: 23733692

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3.

    Smoljanić J, Morris NB, Dervis S, Jay O. Running economy, not aerobic fitness, independently alters thermoregulatory responses during treadmill running. J Appl Physiol. 2014;117(12):14511459. PubMed ID: 25301893 doi:10.1152/japplphysiol.00665.2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Noakes TD, Myburgh KH, du Plessis J, et al. Metabolic rate, not percent dehydration, predicts rectal temperature in marathon runners. Med Sci Sports Exerc. 1991;23(4):443449. PubMed ID: 2056902 doi:10.1249/00005768-199104000-00009

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Cheuvront SN, Haymes EM. Thermoregulation and marathon running: biological and environmental influences. Sports Med. 2001;31(10):743762. PubMed ID: 11547895 doi:10.2165/00007256-200131100-00004

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Bouscaren N, Millet GY, Racinais S. Heat stress challenges in marathon vs. ultra-endurance running. Front Sports Act Living. 2019;1:59. doi:10.3389/fspor.2019.00059

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.

    Gamage PJ, Fortington LV, Finch CF. Epidemiology of exertional heat illnesses in organised sports: a systematic review. J Sports Sci Med. 2020;23(8):701709. PubMed ID: 32144023 doi:10.1016/j.jsams.2020.02.008

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Davies CT, Thompson MW. Aerobic performance of female marathon and male ultramarathon athletes. Eur J Appl Physiol. 1979;41(4):233245. PubMed ID: 499187 doi:10.1007/BF00429740

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Dancaster CP, Whereat SJ. Fluid and electrolyte balance during the comrades marathon. S Afr Med J. 1971;45(6):147150. PubMed ID: 5553596

  • 10.

    Noakes TD, Adams BA, Myburgh KH, Greeff C, Lotz T, Nathan M. The danger of an inadequate water intake during prolonged exercise. A novel concept re-visited. Eur J Appl Physiol Occup Physiol. 1988;57(2):210219. PubMed ID: 3349989 doi:10.1007/BF00640665

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Brown JS, Connolly DA. Selected human physiological responses during extreme heat: the Badwater Ultramarathon. J Strength Cond Res. 2015;29(6):17291736. PubMed ID: 25463692 doi:10.1519/JSC.0000000000000787

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Valentino TR, Stuempfle KJ, Kern M, Hoffman MD. The influence of hydration state on thermoregulation during a 161-km ultramarathon. Res Sports Med. 2016;24(3):197206. doi:10.1080/15438627.2016.1191491

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13.

    Davies CT, Thompson MW. Physiological responses to prolonged exercise in ultramarathon athletes. J Appl Physiol. 1986;61(2):611617. doi:10.1152/jappl.1986.61.2.611

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Fortes MB, Di Felice U, Dolci A, et al. Muscle-damaging exercise increases heat strain during subsequent exercise heat stress. Med Sci Sports Exerc. 2013;45(10):19151924. PubMed ID: 23559121 doi:10.1249/MSS.0b013e318294b0f8

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15.

    Cheuvront SN, Kenefick RW, Montain SJ, Sawka MN. Mechanisms of aerobic performance impairment with heat stress and dehydration. J Appl Physiol. 2010;109(6):19891995. doi:10.1152/japplphysiol.00367.2010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Dervis S, Coombs GB, Chaseling GK, Filingeri D, Smoljanic J, Jay O. A comparison of thermoregulatory responses to exercise between mass-matched groups with large differences in body fat. J Appl Physiol. 2016;120(6):615623. doi:10.1152/japplphysiol.00906.2015

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Peterson MJ, Czerwinski SA, Siervogel RM. Development and validation of skinfold-thickness prediction equations with a 4-compartment model. Am J Clin Nutr. 2003;77(5):11861191. PubMed ID: 12716670 doi:10.1093/ajcn/77.5.1186

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Jones AM, Burnley M, Vanhatalo A. Aerobic exercise performance. In: Norton K, Eston R, eds. Kinanthropometry and Exercise Physiology. Routledge; 2019:318352.

    • Search Google Scholar
    • Export Citation
  • 19.

    Byrne C, Lim CL. The ingestible telemetric body core temperature sensor: a review of validity and exercise applications. Br J Sports Med. 2007;41(3):126133. PubMed ID: 17178778 doi:10.1136/bjsm.2006.026344

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Cheuvront SN, Kenefick RW. CORP: improving the status quo for measuring whole body sweat losses. J Appl Physiol. 2017;123(3):632636. PubMed ID: 28684591 doi:10.1152/japplphysiol.00433.2017

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.

    Cohen J. A power primer. Psychol Bull. 1992;112(1):155159. PubMed ID: 19565683 doi:10.1037/0033-2909.112.1.155

  • 22.

    Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, Roberts WO. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc. 2007;39(3):556572. PubMed ID: 17473783 doi:10.1249/MSS.0b013e31802fa199

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Castellani JW, Young AJ. Human physiological responses to cold exposure: acute responses and acclimatization to prolonged exposure. Auton Neurosci. 2016;196:6374. doi:10.1016/j.autneu.2016.02.009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24.

    Taylor NA. Human heat adaptation. Compr Physiol. 2014;4(1):325365. PubMed ID: 24692142

  • 25.

    Buono MJ, Heaney JH, Canine KM. Acclimation to humid heat lowers resting core temperature. Am J Physiol. 1998;274(5):R1295R1299. PubMed ID: 9644042

    • Search Google Scholar
    • Export Citation
  • 26.

    Cramer MN, Jay O. Biophysical aspects of human thermoregulation during heat stress. Auton Neurosci. 2016;196:313. doi:10.1016/j.autneu.2016.03.001

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Kelly JC, Godlonton JD. The 1980 Comrades Marathon. S Afr Med J. 1980;58(13):509510. PubMed ID: 7423281

  • 28.

    Hoffman MD, Stellingwerff T, Costa RJS. Considerations for ultra-endurance activities: part 2—hydration. Res Sports Med. 2019;27(2):182194. doi:10.1080/15438627.2018.1502189

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Rust CA, Knechtle B, Knechtle P, Wirth A, Rosemann T. Body mass change and ultraendurance performance: a decrease in body mass is associated with an increased running speed in male 100-km ultramarathoners. J Strength Cond Res. 2012;26(6):15051516. PubMed ID: 22614141

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Zouhal H, Groussard C, Minter G, et al. Inverse relationship between percentage body weight change and finishing time in 643 forty-two-kilometre marathon runners. Br J Sports Med. 2011;45(14):11011105. PubMed ID: 21160081 doi:10.1136/bjsm.2010.074641

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Noakes TD, Myburgh KH, Schall R. Peak treadmill running velocity during the VO2 max test predicts running performance. J Sports Sci. 1990;8(1):3545. PubMed ID: 2359150 doi:10.1080/02640419008732129

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Hoffman MD, Lebus DK, Ganong AC, Casazza GA, Van Loan M. Body composition of 161-km ultramarathoners. Int J Sports Med. 2010;31(2):106109. PubMed ID: 20222002 doi:10.1055/s-0029-1241863

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Lambert MI, Dugas JP, Kirkman MC, Mokone GG, Waldeck MR. Changes in running speeds in a 100 km ultra-marathon race. J Sports Sci Med. 2004;3(3):167173. PubMed ID: 24482594

    • Search Google Scholar
    • Export Citation
  • 34.

    Tan PL, Tan FH, Bosch AN. Similarities and differences in pacing patterns in a 161-km and 101-km ultra-distance road race. J Strength Cond Res. 2016;30(8):21452155. PubMed ID: 26808845 doi:10.1519/JSC.0000000000001326

    • Crossref
    • Search Google Scholar
    • Export Citation

Byrne (c.byrne3@exeter.ac.uk) is corresponding author.

  • Collapse
  • Expand
  • Figure 1

    —Course profile illustrating distance versus elevation and the 6 (S1–S6) split sections (A); mean (SD) running speed for each split section (B, n = 21); mean (SD) pacing as percentage of mean 89-km speed for each split section (C, n = 21); and mean (SD) %HRmax for each split section (D, n = 8). HRmax indicates maximal heart rate; S, split section value. S is significantly greater than designated number split section(s), P < .05.

  • Figure 2

    —Individual baseline and mean Tc (A), individual baseline and peak Tc (B), mean ΔTc from baseline (C), and peak ΔTc from baseline (D) per hour of running for 17 finishers (n = 13 males ○ and n = 4 females ▴) and 2 nonfinishers (n = 2 males •). Tc indicates core temperature.

  • Figure 3

    —Box plots of data sets for peak Tc (A), peak ΔTc from baseline Tc (B), peak positive ΔTc within each split (C), and peak negative ΔTc within each split (D) across the 6 split sections for 17 finishers. Box plots represent minimum and maximum values (whiskers); 25th percentile, median (line), mean (×), 75th percentile, and interquartile range (box). Data points that exceed 1.5 times the interquartile range are considered outliers and are illustrated outside the whiskers.

  • 1.

    Knechtle B, Nikolaidis PT. Physiology and pathophysiology in ultra-marathon running. Front Physiol. 2018;9:634. PubMed ID: 29910741 doi:10.3389/fphys.2018.00634.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2.

    Sawka MN, Leon LR, Montain SJ, Sonna LA. Integrated physiological mechanisms of exercise performance, adaptation, and maladaptation to heat stress. Compr Physiol. 2011;1(4):18831928. PubMed ID: 23733692

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3.

    Smoljanić J, Morris NB, Dervis S, Jay O. Running economy, not aerobic fitness, independently alters thermoregulatory responses during treadmill running. J Appl Physiol. 2014;117(12):14511459. PubMed ID: 25301893 doi:10.1152/japplphysiol.00665.2014

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4.

    Noakes TD, Myburgh KH, du Plessis J, et al. Metabolic rate, not percent dehydration, predicts rectal temperature in marathon runners. Med Sci Sports Exerc. 1991;23(4):443449. PubMed ID: 2056902 doi:10.1249/00005768-199104000-00009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5.

    Cheuvront SN, Haymes EM. Thermoregulation and marathon running: biological and environmental influences. Sports Med. 2001;31(10):743762. PubMed ID: 11547895 doi:10.2165/00007256-200131100-00004

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Bouscaren N, Millet GY, Racinais S. Heat stress challenges in marathon vs. ultra-endurance running. Front Sports Act Living. 2019;1:59. doi:10.3389/fspor.2019.00059

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7.

    Gamage PJ, Fortington LV, Finch CF. Epidemiology of exertional heat illnesses in organised sports: a systematic review. J Sports Sci Med. 2020;23(8):701709. PubMed ID: 32144023 doi:10.1016/j.jsams.2020.02.008

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Davies CT, Thompson MW. Aerobic performance of female marathon and male ultramarathon athletes. Eur J Appl Physiol. 1979;41(4):233245. PubMed ID: 499187 doi:10.1007/BF00429740

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Dancaster CP, Whereat SJ. Fluid and electrolyte balance during the comrades marathon. S Afr Med J. 1971;45(6):147150. PubMed ID: 5553596

    • Search Google Scholar
    • Export Citation
  • 10.

    Noakes TD, Adams BA, Myburgh KH, Greeff C, Lotz T, Nathan M. The danger of an inadequate water intake during prolonged exercise. A novel concept re-visited. Eur J Appl Physiol Occup Physiol. 1988;57(2):210219. PubMed ID: 3349989 doi:10.1007/BF00640665

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Brown JS, Connolly DA. Selected human physiological responses during extreme heat: the Badwater Ultramarathon. J Strength Cond Res. 2015;29(6):17291736. PubMed ID: 25463692 doi:10.1519/JSC.0000000000000787

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Valentino TR, Stuempfle KJ, Kern M, Hoffman MD. The influence of hydration state on thermoregulation during a 161-km ultramarathon. Res Sports Med. 2016;24(3):197206. doi:10.1080/15438627.2016.1191491

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13.

    Davies CT, Thompson MW. Physiological responses to prolonged exercise in ultramarathon athletes. J Appl Physiol. 1986;61(2):611617. doi:10.1152/jappl.1986.61.2.611

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Fortes MB, Di Felice U, Dolci A, et al. Muscle-damaging exercise increases heat strain during subsequent exercise heat stress. Med Sci Sports Exerc. 2013;45(10):19151924. PubMed ID: 23559121 doi:10.1249/MSS.0b013e318294b0f8

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15.

    Cheuvront SN, Kenefick RW, Montain SJ, Sawka MN. Mechanisms of aerobic performance impairment with heat stress and dehydration. J Appl Physiol. 2010;109(6):19891995. doi:10.1152/japplphysiol.00367.2010

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Dervis S, Coombs GB, Chaseling GK, Filingeri D, Smoljanic J, Jay O. A comparison of thermoregulatory responses to exercise between mass-matched groups with large differences in body fat. J Appl Physiol. 2016;120(6):615623. doi:10.1152/japplphysiol.00906.2015

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Peterson MJ, Czerwinski SA, Siervogel RM. Development and validation of skinfold-thickness prediction equations with a 4-compartment model. Am J Clin Nutr. 2003;77(5):11861191. PubMed ID: 12716670 doi:10.1093/ajcn/77.5.1186

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Jones AM, Burnley M, Vanhatalo A. Aerobic exercise performance. In: Norton K, Eston R, eds. Kinanthropometry and Exercise Physiology. Routledge; 2019:318352.

    • Search Google Scholar
    • Export Citation
  • 19.

    Byrne C, Lim CL. The ingestible telemetric body core temperature sensor: a review of validity and exercise applications. Br J Sports Med. 2007;41(3):126133. PubMed ID: 17178778 doi:10.1136/bjsm.2006.026344

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Cheuvront SN, Kenefick RW. CORP: improving the status quo for measuring whole body sweat losses. J Appl Physiol. 2017;123(3):632636. PubMed ID: 28684591 doi:10.1152/japplphysiol.00433.2017

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.

    Cohen J. A power primer. Psychol Bull. 1992;112(1):155159. PubMed ID: 19565683 doi:10.1037/0033-2909.112.1.155

  • 22.

    Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, Roberts WO. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc. 2007;39(3):556572. PubMed ID: 17473783 doi:10.1249/MSS.0b013e31802fa199

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Castellani JW, Young AJ. Human physiological responses to cold exposure: acute responses and acclimatization to prolonged exposure. Auton Neurosci. 2016;196:6374. doi:10.1016/j.autneu.2016.02.009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24.

    Taylor NA. Human heat adaptation. Compr Physiol. 2014;4(1):325365. PubMed ID: 24692142

  • 25.

    Buono MJ, Heaney JH, Canine KM. Acclimation to humid heat lowers resting core temperature. Am J Physiol. 1998;274(5):R1295R1299. PubMed ID: 9644042

    • Search Google Scholar
    • Export Citation
  • 26.

    Cramer MN, Jay O. Biophysical aspects of human thermoregulation during heat stress. Auton Neurosci. 2016;196:313. doi:10.1016/j.autneu.2016.03.001

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Kelly JC, Godlonton JD. The 1980 Comrades Marathon. S Afr Med J. 1980;58(13):509510. PubMed ID: 7423281

  • 28.

    Hoffman MD, Stellingwerff T, Costa RJS. Considerations for ultra-endurance activities: part 2—hydration. Res Sports Med. 2019;27(2):182194. doi:10.1080/15438627.2018.1502189

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Rust CA, Knechtle B, Knechtle P, Wirth A, Rosemann T. Body mass change and ultraendurance performance: a decrease in body mass is associated with an increased running speed in male 100-km ultramarathoners. J Strength Cond Res. 2012;26(6):15051516. PubMed ID: 22614141

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Zouhal H, Groussard C, Minter G, et al. Inverse relationship between percentage body weight change and finishing time in 643 forty-two-kilometre marathon runners. Br J Sports Med. 2011;45(14):11011105. PubMed ID: 21160081 doi:10.1136/bjsm.2010.074641

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Noakes TD, Myburgh KH, Schall R. Peak treadmill running velocity during the VO2 max test predicts running performance. J Sports Sci. 1990;8(1):3545. PubMed ID: 2359150 doi:10.1080/02640419008732129

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Hoffman MD, Lebus DK, Ganong AC, Casazza GA, Van Loan M. Body composition of 161-km ultramarathoners. Int J Sports Med. 2010;31(2):106109. PubMed ID: 20222002 doi:10.1055/s-0029-1241863

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Lambert MI, Dugas JP, Kirkman MC, Mokone GG, Waldeck MR. Changes in running speeds in a 100 km ultra-marathon race. J Sports Sci Med. 2004;3(3):167173. PubMed ID: 24482594

    • Search Google Scholar
    • Export Citation
  • 34.

    Tan PL, Tan FH, Bosch AN. Similarities and differences in pacing patterns in a 161-km and 101-km ultra-distance road race. J Strength Cond Res. 2016;30(8):21452155. PubMed ID: 26808845 doi:10.1519/JSC.0000000000001326

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3437 1209 159
PDF Downloads 1488 440 33