Running 8000 m Fast or Slow: Are There Differences in Energy Cost and Fat Metabolism? : Medicine & Science in Sports & Exercise

Journal Logo

Applied Sciences: Physical Fitness and Performance

Running 8000 m Fast or Slow: Are There Differences in Energy Cost and Fat Metabolism?


Author Information
Medicine & Science in Sports & Exercise 37(10):p 1789-1793, October 2005. | DOI: 10.1249/01.mss.0000176401.92974.81
  • Free


Frequently, low intensities are recommended for preventive endurance running to ensure compliance and enhance “fat burning.” In its instructions, the company Polar (Polar Electro, Kempele, Finland) differentiates between training purposes: “weight loss and building endurance” with 60–70% of maximal HR and “weight management and improving cardiovascular fitness” with 70–80% of maximal HR. Recent research showed similar effects on weight loss and improved cardiorespiratory fitness for exercise programs of different intensities and durations combined with diet in sedentary women (11). Furthermore, some publications describe reductions of coronary events and mortality not only due to vigorous exercise but also due to light and moderate physical activity (7,14,18).

Theoretically, to lose weight, the caloric cost of running should be as high as possible. According to the literature, the relationship between oxygen uptake and running speed is supposed to be linear, so that the energy cost of running a given distance should be the same regardless of whether the pace is fast or slow (20). This conclusion is predominantly based on treadmill experiments (3,6,9,19,28). In some of these studies, air resistance was simulated by increasing the incline of the treadmill, but little is reported about the grade and how it was determined, a crucial decision for such investigations. Also, air resistance increases with running speed, whereas treadmill incline usually remains constant. These considerations partly invalidate the transfer of treadmill research to field conditions. In addition, running economy can change when running velocity increases, which further necessitates field studies. Modern equipment for ambulatory gas exchange measurements and other technological developments such as flashing light indicators of speed now enable more appropriate measurements of energy expenditure while track running. Such investigations have not been conducted before.

Due to occupational and other constraints, many people feel that they hardly find spare time to spend on recreational activities. When running for a given duration, it is obvious that higher velocities lead to more energy expenditure than lower ones. However, many individuals prefer to run a given distance rather than for a specified time. To save time, the question arises whether the caloric cost of running a given distance differs meaningfully within realistic training speeds. To investigate this, energy expenditure and substrate use were measured by means of a portable metabolic device while running 8000 m on an indoor track with 70 and 95% of the speed corresponding to the maximal lactate steady state. A priori, a meaningful difference in energy expenditure was arbitrarily defined as at least 10% because smaller differences were considered to be too small to have an impact on an individual's decision about how to conduct endurance running for the purpose of weight loss. It was hypothesized that no meaningful difference in energy expenditure between the runs would occur.



Fourteen healthy endurance-trained male (N = 10) and female (N = 4) subjects (physical education students (N = 4), triathletes (N = 4), and long-distance runners (N = 6)), were recruited for the study. They gave written informed consent after the study was approved by the institutional review board. Before exercise testing, participants were subjected to a medical checkup (medical history and physical examination) to ensure the absence of health risks.

General design.

All subjects performed an initial incremental running protocol to exhaustion and subsequently two 8000-m runs of different intensities in randomized order. From the blood lactate concentrations (La) during the incremental test, the individual anaerobic threshold (IAT) as a measure for the maximum lactate steady state (25,27) was determined to calculate running velocities for the 8000-m runs. Constant velocities for the 8000-m runs were 70 and 95% of the IAT. All tests took place on an indoor track (200-m length, tartan surface). Gas exchange and HR measurements were carried out continuously during all tests. Each subject had all his or her tests within 2 wk including at least 1 d off between the incremental test and the first 8000-m run and at least 2 d off between the two 8000-m runs. They were advised to abstain from strenuous exercise and to keep their nutritional intake similar on the days before all tests. This was controlled by means of a written protocol. Each subject wore the same shoes, similar clothes, and the same face mask for all tests, which were performed at the same time of day. Anthropometric and ergometric data of the participants are given in Table 1.

Anthropometric and ergometric performance data of the subjects from the incremental running test (means and SD).

Incremental running protocol.

The exhaustive incremental running protocols started with 10 km·h−1 for men and 8 km·h−1 for women. Running speed increased stepwise every 3 min by 2 km·h−1. Capillary blood samples for the determination of blood lactate concentrations (enzymatic-amperometric method; Greiner, Flacht, Germany) were taken from the hyperemized earlobe at rest, during 30-s breaks between the workloads, and at 1, 3, 5, 7, and 10 min after cessation of exercise. IAT was determined from the lactate curve by means of a computer software program following the procedures of Stegmann et al. (25). In brief, a tangent is drawn to the exercise lactate curve departing from a point after cessation of exercise where the (now decreasing) blood lactate concentration is equal to the one at cessation of exercise. IAT has been shown to approximate the maximal lactate steady state well during running and cycling (25,27). Gas exchange measurements (for details, see below) were carried out during the incremental tests to familiarize the subjects with the equipment, and to measure V̇O2max.

8000-m runs.

The 8000-m runs were carried out with velocities corresponding to 70 and 95% IAT. They represent aerobic intensities suitable for recreational endurance training that can safely be maintained over the chosen distance and were chosen to appropriate minimal and maximal speeds of realistic training (4). Precondition for participation was an IAT of at least 11.5 km·h−1. Running velocities were precisely given by a flashing light system next to the track (Gümbel, Ludwigshafen, Germany). The following gas exchange data were measured continuously by means of a portable device (Meta Max II, Cortex, Leipzig, Germany; mixing chamber; sampling frequency: 10 s): minute ventilation (V̇E), oxygen uptake (V̇O2), and carbon dioxide output (V̇CO2). Thus, RER could be calculated. The weight of the device including gear added up to 1000 g. Its validity was recently documented for workloads between 100 and 250 W (15). HR was measured continuously by means of a telemetric system (Polar S 610, Polar Electro; recording frequency: 5 s). Capillary blood samples for lactate determination were taken at rest, after 2600 m, after 5400 m, and immediately after exercise. Therefore, each run was interrupted twice for 30 s. The data were averaged over the entire run. The 30-s breaks as well as the first 3 min after each break (readjustment of metabolic variables) were excluded, and the average value of the remaining time was multiplied by the total exercise duration. Energy consumption and substrate use (fat and carbohydrate metabolism) were calculated by means of indirect calorimetry according to the table of Lusk (17). Endurance runs were preceded by 2-min resting measurements and followed by 10-min resting measurements while subjects were sitting quietly.


Data are given as means and SD throughout. They were checked for normal distribution using the Shapiro–Wilks W-test. As this was present for all dependent variables, parametric tests were chosen for comparisons between the 8000-m runs. Discrete variables were treated with Student's t-test for paired samples. Substrate use and other variables with more than one measurement time (HR, La, energy consumption, and substrate use) were compared using a two-factor ANOVA (factor 1: 70% IAT vs 95% IAT; factor 2: measurement time). The Scheffé test was used for post hoc comparisons. P < 0.05 for the α error was considered significant.

To assess an appropriate sample size a priori (26), effect size (ES = (MX- MY)/SDY) was estimated to be 0.37 using the most suitable (10) equation to predict energy cost of horizontal track running (16), making calculations for an imaginary subject (IAT: 13 km·h−1, running velocity: 9 km·h−1, body weight including equipment: 80 kg), and setting the practically relevant difference in energy expenditure at 10%. Approximately 14 to 16 subjects resulted from estimation with a two-tailed α = 0.05, a power of 0.8, and a calculated ES of 0.37.


The 70% IAT runs corresponded to approximately 56.0 ± 4.0% V̇O2max (min–max: 47.5–61.6% V̇O2max) and were carried out with an average running velocity of 9.6 ± 1.0 km·h−1 and an average time to completion of 50:35 ± 5:25 min. Average HR was 138 ± 12 min−1 (74 ± 5% HRmax). The 95% IAT runs corresponded to approximately 78.5 ± 3.9% V̇O2max (min–max: 72.5–84.8% V̇O2max) with an average speed and exercise duration of 13.0 ± 1.3 km·h−1 and 37:17 ± 3:59 min, respectively. Mean HR was 168 ± 16 min−1 (90 ± 5% HRmax). Ambient air temperature and air pressure averaged 23.2 ± 1.2°C and 986 ± 14 hPa, and varied intraindividually for maximally 2.5°C and 49 hPa, respectively. HR were significantly higher (P < 0.001) during the 95% IAT run than during the 70% IAT run (Fig. 1A). Subdividing the 8000-m runs into three thirds, HR increased significantly from the first to the third part (P < 0.05) of the 95% IAT run, whereas no such trend was observed for the 70% IAT run. Blood lactate concentrations were significantly higher at any time of the 95% IAT run than during the 70% IAT run (P < 0.001, Fig. 1B). They remained at baseline values during the 70% IAT run, but increased significantly from the first to the third measurement (P < 0.05) of the 95% IAT run. The average increase amounted to 0.6 ± 0.4 mmol·L−1 (min–max: 0.3–1.1 mmol·L−1). According to definitions by Beneke (2) and Urhausen et al. (27), lactate steady-state criteria for constant-load tests lasting at least 30 min are met if the increase in blood lactate concentration does not exceed 1.0 mmol·L−1 during the last 10 min of exercise. According to this criterion, lactate steady state was present in 11 of the 14 subjects.

Figure 1— A. HR during the 8000-m runs (:
N = 13; means and SD). *** P < 0.001 for 70 vs 95% IAT. B. Blood lactate concentrations during the 8000-m runs ( N = 14; means and SD). *** P < 0.001 for 70 vs 95% IAT.

Significantly more energy was expended during the 95% IAT run than during the 70% IAT run (2650 ± 276 and 2554 ± 348 kJ, respectively; P < 0.05; Fig. 2). However, the difference of 3.8 ± 4.8% was not meaningful according to the a priori defined relevant difference of at least 10%. Including measurements up to 10th minute postexercise, the total difference between 95 and 70% IAT rose to 5.1 ± 4.7% (2830 ± 301 kJ vs 2692 ± 368 kJ; P < 0.01). Energy expenditure during this 10-min period was significantly higher after the 95% IAT run than after 70% IAT (184 ± 29 vs 138 ± 29 kJ; P < 0.01). Considering the three parts of the running distance, the difference in energy expenditure between the two running velocities increased during the duration of the test with no significant difference during the first third but significant differences during the second (P < 0.01) and third (P < 0.001) third (Fig. 3).

Figure 2— Energy expenditure during the 8000-m runs (in the foreground) and energy expenditure until the 10th min postexercise (in the background;:
N = 14; means and SD). * P < 0.05; ** P < 0.01.
Figure 3— Energy expenditure during the 8000-m runs separately calculated for three parts of the distance (:
N = 14; means and SD). ** P < 0.01; *** P < 0.001 for 70 vs 95% IAT.

There was no significant difference between the absolute amount of fat being metabolized during the runs (70% IAT: 26 ± 5 g, 95% IAT: 20 ± 5 g; P = 0.14). However, during the 95% IAT run, significantly more carbohydrates were used than during the 70% IAT run (108 ± 14 g vs 90 ± 15 g, respectively; P < 0.001; Table 2).

Absolute amounts of fat and carbohydrate metabolized during the 8000-m runs (means, SD, and min–max).


The energy cost of running 8000 m differed slightly between a fast and a slow velocity chosen from the range of realistic intensities of recreational endurance exercise. Although significant, this difference was not meaningful as related to the a priori defined “relevant difference” of at least 10%. Also, the absolute amount of metabolized fat did not differ significantly between intensities. The results do not indicate that low-intensity exercise is advantageous for running a given distance when weight loss is the primary concern. Even though lower intensity preventive exercise was considered to be efficient in recent publications (7,11,14,18), it certainly does not represent the optimal choice for those with limited time sources being interested in weight loss.

Several former studies described a linear relationship of running velocity and energy expenditure. But these investigations were carried out on a treadmill (3,6,9,19,28) and might therefore not be transferable to field conditions. While conducting identical incremental running protocols on the treadmill and on a track, Meyer et al. (22) in a recent investigation observed significantly higher oxygen uptakes and minute ventilations on the treadmill. The treadmill inclination remained constantly at 0.5% to simulate air resistance, which, however, grows with increasing running velocity under real conditions. Those examinations carried out in the field, such as the one of Pugh (24), point to the assumption of speed-independent energy expenditure, but running conditions were poorly controlled and measurement devices have been much improved since then.

The difference in energy expenditure between the two running velocities increased over the course of the run because energy expenditure increased slightly during the 95% IAT run, and no such trend was observed for 70% IAT. This is probably primarily due to rising core temperature, which was not measured in this study but is described in literature (5,23). When investigating longer running distances, it might thus be possible to demonstrate larger differences in energy expenditure between fast and slow running. However, the choice of distances for such analyses is limited because recreational athletes must be able to cope with the fast pace over the entire distance.

There might have resulted an even larger difference in energy expenditure between fast and slow running velocities if more distant exercise intensities had been employed. However, 70 and 95% IAT were chosen because they represent reasonable exercise intensities for preventive training and building endurance and can be maintained over 8000 m. Running speeds were given as percentages of IAT because this concept, compared to concepts based on V̇O2max or HRmax, seems to enable the attainment of more homogeneous metabolic responses to exercise and, thus, individual degree of demand (21). However, both represent intensities within the range of American College of Sports Medicine recommendations (1).

Calculations of energy expenditure and substrate use by means of indirect calorimetry are based on the assumption that expiratory gas composition precisely reflects oxygen consumption and carbon dioxide production from fuel metabolism on the tissue level. Carbon dioxide output can, however, be affected by nonmetabolic processes such as the HCO3 buffering of hydrogen ions that appear when blood lactate concentration rises above baseline. As a result of buffering hydrogen ions, additional CO2 (“excess CO2”) is released so that RER is influenced nonmetabolically. This might lead to a slight overestimation of energy consumption and to an underestimation of fat metabolism. In this investigation, blood lactate concentrations stayed on baseline values during the 70% IAT run but increased significantly above baseline during 95% IAT. Thus, theoretically, our findings might have been affected by a systematic error overestimating the difference in energy expenditure (in absolute terms: 96 ± 121 kcal) and in fat metabolism (in absolute terms: 6 ± 4 g) between the two running velocities. Reports from other groups, however, indicate a very small size of this error except for very rapidly increasing blood lactate concentrations and exercise intensities above 75% of maximal oxygen uptake (8,13). Ninety-five percent IAT corresponded to approximately 78.5% V̇O2max. When calculating energy expenditure with a mean caloric equivalent of 20.31 kJ·L−1 O2 (which is independent from V̇CO2 and RER), the difference between 70 and 95% IAT appears to be smaller by 0.8%. Altogether, excess CO2 does not seem to have affected the present results relevantly.

According to the MetaMax II validation study of Larsson et al. (15), the device slightly overestimates V̇O2 during high workloads on the cycle ergometer (0.11 L·min−1 at 325 W). In the present study, V̇O2 measurements while running 95% IAT reached similar values. The real difference in energy expenditure might have been magnified if such a measurement error had really occurred in our study, too. Therefore, a meaningful difference between the two chosen intensities is even more unlikely.

Four female subjects were recruited for this study. Because all of a subject's tests had to be carried out within 2 wk, the subject's menstrual cycle phases could not be taken into account. Two of these subjects were on a single-phase oral contraceptive during the examination, which generally leads to smaller hormonal fluctuations during the menstrual cycle. According to a recent review (12), most research is in favor of no changes over the menstrual cycle with regard to V̇O2 and substrate use during submaximal aerobic exercise. Only few studies observed increased V̇O2 and enhanced lipid metabolism during the mid-luteal phase (12). Furthermore, there was no overt difference between male and female subjects' data. Therefore, the influence of the menstrual cycle on the results of the examination seems to be minor.

In conclusion, this study demonstrated that the energy cost of running 8000 m at both fast and slow speeds differs significantly but not meaningfully within reasonable training intensities. These results are in good agreement with the assumption of a speed-independent energy cost for running a given distance. Therefore, earlier findings could be validated by means of an improved simulation of the real training situation using more appropriate equipment. Endurance training for weight loss of high intensity, which is below the IAT, can be considered at least as suitable as low intensities and reduces the time needed to consume energy. Also, fat metabolism does not differ relevantly.


1. American College of Sports Medicine Position Stand: The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscularfitness, and flexibility in healthy adults. Med.Sci. Sports Exerc. 30:975–991, 1998.
2. Beneke, R. Methodological aspects of maximal lactate steady state-implications for performance testing. Eur. J. Appl. Physiol. 89:95–99, 2003.
3. Bransford, D., and E. Howley. Oxygen cost of running in trained and untrained men and women. Med. Sci. Sports. 9:41–44, 1977.
4. Coen, B., L. Schwarz, A. Urhausen, and W. Kindermann. Control of training in middle- and long-distance running by means of the individual anaerobic threshold. Int. J. Sports Med. 12:519–524, 1991.
5. Daniels, J. A physiologist's view on running economy. Med. Sci. Sports Exerc. 17:332–338, 1985.
6. Dill, D. Oxygen used in horizontal and grade walking and running on the treadmill. J. Appl. Physiol. 20:19–22, 1965.
7. Fletcher, G., G. Balady, and S. Blair. Statement on exercise: benefits and recommendations for physical activity programs for all Americans. Circulation 94:857–862, 1996.
8. Frayn, K. Calculation of substrate oxidation rates in vivo from gaseous exchange. J. Appl. Physiol. 55:628–634, 1983.
9. Givoni, B., and R. Goldman. Predicting metabolic energy cost. J. Appl. Physiol. 30:429–433, 1971.
10. Hall, C., A. Figueroa, B. Fernhall, and J. Kanaley. Energy expenditure of walking and running: comparison with prediction equations. Med. Sci. Sports Exerc. 36:2128–2134, 2004.
11. Jakicic, J., B. Marcus, K. Gallagher, M. Napolitano, and W. Lang. Effect of exercise duration and intensity on weight loss in overweight, sedentary women: a randomized trail. JAMA. 290:1323–1330, 2003.
12. Janse de Jonge, X. Effects of the menstrual cycle on exercise performance. Sports Med. 33:833–851, 2003.
13. Jeukendrup, A., and G. Wallis. Measurement of substrate oxidation during exercise by means of gas exchange measurements. Int. J. Sports Med. 26 Suppl 1:S28–37, 2005.
14. Kujala, U., J. Kaprio, S. Sarna, and M. Koskenvuo. Relationship of leisure-time physical activity and mortality. JAMA. 279:440–444, 1998.
15. Larsson, P. U., K. M. E. Wadell, E. I. Jakobsson, L. U. Burlin, and K. B. Henriksson-Larsen. Validation of the MetaMax II portable metabolic measurement system. Int. J. Sports Med. 25:115–123, 2004.
16. Leger, L., and D. Mercier. Gross energy cost of horizontal treadmill and track running. Sports Med. 1:270–277, 1984.
17. Lusk, G. Animal calorimetry. Analysis of the oxidation of mixtures of carbohydrate and fat. A correction. J. Biol. Chem. 59:41–42, 1924.
18. Manson, J., F. Hu, and J. Rich-Edwards. A prospective study of walking as compared with vigorous exercise in the prevention of coronary heart disease in women. N. Engl. J. Med. 341:650–658, 1999.
19. Margaria, R., P. Cerretelli, P. Aghemo, and G. Sassi. Energy cost of running. J. Appl. Physiol. 18:367–370, 1963.
20. McArdle, W., F. Katch, and V. Katch. Exercise Physiology. Energy, Nutrition, and Human Performance. Baltimore: Williams & Wilkins, 1996, p. 174.
21. Meyer, T., H. Gabriel, and W. Kindermann. Is determination of exercise intensities as percentages of V̇O2max or HRmax adequate? Med. Sci. Sports Exerc. 1342–1345, 1999.
22. Meyer, T., J. P. Welter, J. Scharhag, and W. Kindermann. Maximal oxygen uptake during field running does not exceed that measured during treadmill exercise. Eur. J. Appl. Physiol. 88: 387–389 2003.
23. Morgan D., P. Martin, and G. Krahenbuhl. Factors affecting running economy. Sports Med. 7:310–330, 1989.
24. Pugh, L. Oxygen intake in track and treadmill running with observations on the effect of air resistance. J. Physiol. 207:823–835, 1970.
25. Stegmann, H., W. Kindermann, and A. Schnabel. Lactate kinetics and individual anaerobic threshold. Int. J. Sports Med. 2:160–165, 1981.
26. Thomas, J., M. Lochbaum, D. Landers, and C. He. Planning significant and meaningful research in exercise science: estimating sample size. Res. Q. Exerc. Sport. 68:33–43, 1997.
27. Urhausen, A., B. Coen, B. Weiler, and W. Kindermann. Individual anaerobic threshold and maximum lactate steady state. Int. J. Sports Med. 14:134–139, 1993.
28. van der Walt, W., and C. Wyndham. An equation for prediction of energy expenditure of walking and running. J. Appl. Physiol. 34:559–563, 1973.


©2005The American College of Sports Medicine