Medicine & Science in Sports & Exercise
Issue: Volume 28(10), October 1996, pp 1327-1330
Copyright: © Williams & Wilkins 1996. All Rights Reserved.
Publication Type: [Applied Sciences: Physical Fitness and Performance]
Keywords: ANAEROBIC TRAINING, AEROBIC TRAINING
[Applied Sciences: Physical Fitness and Performance]
Effects of moderate-intensity endurance and high-intensity
intermittent training on anaerobic capacity and ·VO2max
TABATA, IZUMI; NISHIMURA, KOUJI; KOUZAKI, MOTOKI; HIRAI, YUUSUKE; OGITA, FUTOSHI;
MIYACHI, MOTOHIKO; YAMAMOTO, KAORU
Department of Physiology and Biomechanics, National Institute of Fitness and Sports,
Shiromizu-cho 1, Kanoya City, Kagoshima Prefecture, 891-23 JAPAN
Submitted for publication November 1994.
Accepted for publication December 1995.
The training protocol used in experiment 2 was first introduced by Kouichi Irisawa, who
was a head coach of the Japanese National Speed Skating Team. The training has been used
by the major members of the Japanese Speed Skating Team for several years.
Present addresses: I. Tabata, Laboratory of Exercise Physiology, Division of Health
Promotion, National Institute of Health and Nutrition, 1-23-1 Toyama, Shinjuku City, Tokyo
162 Japan; Y. Hirai, Number 1 Fitness Club, 5-14-6 Shimo-Takaido, Suginami City, Tokyo 168
Japan; K. Nishimura, General Research and Development Section, Product Development
Department, Moon-Star Chemical Corporation, Kurume City, Fukuoka Prefecture, 830-91
Japan; F. Ogita, Swimming Performance Laboratory, National Institute of Fitness and Sport,
Shiromizu-cho 1, Kanoya City, Kagoshima Prefecture, 891-23 Japan; M. Miyachi, Department
of Health and Sports Sciences, Kawasaki University of Medical Welfare, 288 Matsushima,
Kurashiki City, Okayama Prefecture, 701-01 Japan; and K. Yamamoto, Nagoya YMCA, 2-5-29
Kamimaezu, Naka-ku, Nagoya City, Aichi Prefecture, 460 Japan.
Address for correspondence: I. Tabata, Ph.D., Laboratory of Exercise Physiology, Division
of Health Promotion, National Institute of Health and Nutrition, 1-23-1 Toyama, Shijuku City,
Tokyo 162, Japan.
This study consists of two training experiments using a mechanically braked cycle
ergometer. First, the effect of 6 wk of moderate-intensity endurance training
(intensity: 70% of maximal oxygen uptake (·VO2max), 60 min·d-1, 5 d·wk-1) on the
anaerobic capacity (the maximal accumulated oxygen deficit) and ·VO2max was
evaluated. After the training, the anaerobic capacity did not increase significantly(P >
0.10), while ·VO2max increased from 53 ± 5 ml·kg-1·min-1 to 58 ± 3 ml·kg-1·min-1 (P <
0.01) (mean± SD). Second, to quantify the effect of high-intensity intermittent
training on energy release, seven subjects performed an intermittent training exercise
5 d·wk-1 for 6 wk. The exhaustive intermittent training consisted of seven to eight
sets of 20-s exercise at an intensity of about 170% of ·VO2max with a 10-s rest between
each bout. After the training period, ·VO2max increased by 7 ml·kg-1·min-1, while the
anaerobic capacity increased by 28%. In conclusion, this study showed that moderate-
intensity aerobic training that improves the maximal aerobic power does not change
anaerobic capacity and that adequate high-intensity intermittent training may
improve both anaerobic and aerobic energy supplying systems significantly, probably
through imposing intensive stimuli on both systems.
During high-intensity exercise lasting more than a few seconds, adenosine
triphosphate (ATP) is resynthesized by both aerobic and anaerobic processes (7). The
ability to resynthesize ATP may limit performance in many sports. Thus, if possible,
the training of athletes for sports involving high-intensity exercise should improve the
athletes' ability to release energy both aerobically and anaerobically. The success of
different training regimens can and should be evaluated by the athletes' performance.
However, performance is influenced by other factors such as psychology. In addition,
an adequate training regimen may have several different components, all of which
may not improve the athletes' ability to resynthesize ATP. Training programs should
therefore be evaluated by other means, e.g., by laboratory experiments.
The aerobic energy releasing system is conventionally evaluated by maximal
oxygen uptake (·VO2max) (10), and there are many studies on the effect of training on
·VO2max(9). However, until recently methods for quantifying anaerobic energy release
have been inadequate and thus information on the effect of training on anaerobic
capacity, i.e., the maximum amount of energy available from anaerobic sources, is
incomplete. We have proposed that the accumulated oxygen deficit, first introduced
by Krogh and Lindhard in 1920 (4), is an accurate measure of the anaerobic energy
release during treadmill running (6) and bicycling (7). This principle may allow
examination of the anaerobic capacity (3), taken as the maximal accumulated oxygen
deficit during 2-3 min of exhaustive exercise (6,7). Therefore, the effect of specific
training on the anaerobic capacity may be evaluated by measuring the maximal
accumulated oxygen deficit before and after training. Generally, the more demanding
the training, the greater the fitness benefits. Therefore, we were interested in
learning whether the effects of training on anaerobic capacity are dependent on the
magnitude of anaerobic energy release developed by the specific training. To study
this issue, we compared two different training protocols: a moderate-intensity
endurance training that is not supposed to depend on anaerobic metabolism and a
high-intensity intermittent training that is supposed to recruit the anaerobic energy
releasing system almost maximally.
MATERIALS AND METHODS
Subjects. Young male students majoring in physical education volunteered for
the study (Table 1). Most were physically active and were members of varsity table
tennis, baseball, basketball, football (soccer), and swimming teams. After receiving a
detailed explanation of the purposes, potential benefits, and risks associated with
participating in the study, each student gave his written consent.
TABLE 1. Characteristics of the subjects.
Protocol. All experiments, as well as pretests, were done on a mechanically
braked cycle ergometer (Monark, Stockholm, Sweden) at 90 rpm. Each test or high-
intensity intermittent training session was introduced by a 10-min warm-up at about
50% of ·VO2max.
Experiment 1. The subjects started training after their·VO2max and maximal
accumulated oxygen deficit were measured. They exercised 5 d·wk-1 for 6 wk at an
intensity that elicited 70% of each subject's ·VO2max. The pedaling rate was 70 rpm,
and the duration of the training was 60 min. As each subject's ·VO2max increased
during the training period, exercise intensity was increased from week to week as
required to elicit 70% of the actual ·VO2max. During the training, the maximal
accumulated oxygen deficit was measured before, at 4 wk, and after the training.
·VO2max was determined before and after the training and every week during the
Experiment 2. Subjects exercised for 5 d·wk-1 for 6 wk. For 4 d·wk-1, they
exercised using exhaustive intermittent training. They were encouraged by the
supervisor to complete seven to eight sets of the exercise. Exercise was terminated
when the pedaling frequency dropped below 85 rpm. When they could complete more
than nine sets of the exercise, exercise intensity was increased by 11 W. One day per
week the subjects exercised for 30 min at an intensity of 70% ·VO2max before carrying
out four sets of the intermittent exercise at 170%·VO2max. This latter session was not
exhaustive. The anaerobic capacity was determined before, at 2 wk, and 4 wk into
the training, and after the training. ·VO2max was determined before, at 3 wk, 5 wk,
and after the training.
Pretest. Each subject's oxygen uptake was measured during the last 2 min of six
to nine different 10-min exercise sets at constant power. The power used during each
set ranged between 39% and 87% of the·VO2max. In addition, the power that would
exhaust each subject in 2-3 min was established. These pretests were carried out on
3-5 separate days.
·VO2max. After a linear relationship between exercise intensity and the steady-
state oxygen uptake had been determined in the pretests, the oxygen uptake was
measured for the last two or three 30-s intervals during several bouts of supramaximal
intensity exercise that lasted 2-4 min. The highest ·VO2 was determined to be the
Anaerobic capacity. Anaerobic capacity, the maximal accumulated oxygen
deficit during a 2-3-min exhaustive bicycle exercise, was determined according to the
method of Medbø et al.(6,7). The exercise intensity used to cause exhaustion within
the desired duration (2-3 min) was established on pretests. On the day that anaerobic
capacity was measured, the subjects exercised at the preset power to exhaustin
(defined as when they were unable to keep the pedaling rate above 85 rpm).
Methods of analysis. Fractions of oxygen and carbon dioxide in the expired air
were measured by a mass spectrometer (MGA-1100, Perkin-Elmer Cetus, Norwalk CT).
The gas volume was measured by a gasometer (Shinagawa Seisakusho, Tokyo, Japan).
Values are shown as means ± SD. The data were compared using a paired t-test. The
significance level for all comparisons was set at P < 0.05.
Calculations. For each subject linear relationships between the oxygen demand
and power (experiment 1: r = 0.997 ± 0.001, experiment 2: r = 0.998 ± 0.001) were
established from the measured steady state oxygen uptake at different power during
the pretests. The regression parameters are shown in Table 2. The regression
parameters did not change during training periods in either experiment 1 or 2.
TABLE 2. Regression characteristics of the subjects.
The oxygen demands of the 2-3 min of exhausting exercise were estimated by
extrapolating these relationships to the power used during the experiment. The
accumulated oxygen demand was taken as the product of the estimated oxygen
demand and the duration of the exercise, while the accumulated oxygen uptake was
taken as the measured oxygen uptake integrated over the exercise duration. The
accumulated oxygen deficit was taken as the difference between these two entities.
Experiment 1. After the 6 wk of training, the anaerobic capacity did not
change (Fig. 1) (P > 0.10). The·VO2max increased significantly during the training (Fig.
2) (P < 0.01).
Figure 1-Effect of the endurance training (ET, experiment 1) and the intermittent
training (IT, experiment 2) on the anaerobic capacity. Significant increase from
the pretraining value at*P < 0.05 and **P< 0.01; significant increase from the 2-wk
value at#P < 0.05.
Figure 2-Effect of the endurance training (ET, experiment 1) and the intermittent
training (IT, experiment 2) on the maximal oxygen uptake; significant increase
from the pretraining value at*P < 0.05 and **P< 0.01, respectively.
Experiment 2. The anaerobic capacity increased by 23% after 4 wk of training
(P < 0.01, Fig. 1). It increased further toward the end of the training period. After the
training period, the anaerobic capacity reached 77 ± 9 ml·kg-1, 28% higher than the
After 3 wk of training, the ·VO2max had increased significantly by 5 ± 3 ml·kg-
1·min-1 (P < 0.01, Fig. 2). It tended to increase in the last part of the training period,
but no significant changes were observed. The final·VO2max after 6 wk of training was
55 ± 6 ml·kg-1·min-1, a value of 7 ± 1 ml·kg-1·min-1 above the pretraining value.
The main finding of this study was that 6 wk of aerobic training at
70%·VO2max improved the ·VO2max by 5 ml·kg-1·min-1 in moderately trained young men
but that the anaerobic capacity, as judged by the maximal accumulated oxygen
deficit, did not change. The second finding is that 6 wk of training using high-intensity
intermittent exhaustive exercise improved ·VO2max by 7 ml·kg-1·min-1 and the
anaerobic capacity by 28%.
The observation in experiment 1 that anaerobic capacity did not change after 6
weeks of moderate-intensity endurance training but that·VO2max did increase has
several implications. First, it shows the specificity of training; aerobic training does
not change anaerobic capacity. Since lactate production accounts for about 75% of
maximal anaerobic energy release (11), significant improvements in anaerobic
capacity will probably require that the subjects can produce more lactate after
training. Consequently, lactate production should be stressed to increase the
anaerobic capacity during “anaerobic” training. However, since the blood lactate
concentration during the exercise was low (about 2 mmol·l-1), the major part of
anaerobic energy released during the exercise probably comes from the breakdown of
phosphocreatine (PCr). Therefore, the training sessions in experiment 1 probably did
not tax the lactate producing system much and therefore did not tax the whole
anaerobic energy releasing system to any significant extent. Actually, the
accumulated oxygen deficit during the first minutes of the exercise at 70%·VO2max was
only 37 ± 6% (N = 7) of the maximal accumulated oxygen deficit (data not shown).
Second, the results of experiment 1 support the idea that the accumulated
oxygen deficit is a specific measure of the maximal anaerobic energy release. Due to
the increased ·VO2max after the training period, the subjects could exercise for more
than 6 min at the power used for the pretraining 2- to 3-min anaerobic capacity test.
Therefore, the exercise power for the posttraining anaerobic capacity test was
increased by 6 ± 3% to exhaust each subject in 2-3 min. However, the accumulated
oxygen deficit appeared unaffected by the higher power used at the posttraining test,
suggesting that this entity is able to distinguish between aerobic and anaerobic energy
release at different powers. The alternative interpretation, that there was a change
in the anaerobic capacity but that this change was obscured by a bias in the
accumulated oxygen deficit, cannot be ruled out, but the findings here suggest that
the latter interpretation is less likely.
The high-intensity intermittent training in experiment 2 improved anaerobic
capacity by 28%. Medbø and Burgers (5) reported that 6 wk of the intermittent
training (their group B) increased the anaerobic capacity of untrained men by 16%.
Since there are no clear differences in exercise intensity, exercise duration, and
number of exercise bouts between the two studies, this quantitative difference in
improving anaerobic capacity is probably explained by the difference between the
two studies in magnitude of the anaerobic energy release during each training session.
The peak blood lactate concentration after each training session in the previous
study (5) was 69% of the peak blood lactate concentration after the 2-min exhaustive
running. Therefore, anaerobic metabolism, and especially the lactate producing
system, was probably not taxed maximally. In contrast, the peak blood lactate
concentration after the intermittent training in this investigation was not significantly
different from the value observed after the anaerobic capacity test that recruited
anaerobic energy releasing systems maximally. In addition, our subjects exercised to
exhaustion, but in the previous study, the subjects' rating of perceived
exertion (1) was only 15 (“hard”). This difference may also reflect the recruited level
of anaerobic energy release. Therefore, these results support our hypothesis that the
higher the anaerobic energy release during each training session the higher the
increase in anaerobic capacity after a training period.
In addition to anaerobic capacity, the intermittent training
increased·VO2max significantly in experiment 2. This is to our knowledge the first study
to demonstrate an increase in both anaerobic capacity and maximal aerobic power. It
should be emphasized that during the last part of each training session the oxygen
uptake almost equaled each subject's maximal oxygen uptake (data not shown). High-
intensity intermittent training is a very potent means of increasing maximal oxygen
uptake (2). It is interesting to note that the increase in the maximal oxygen uptake
that we found is almost identical to that expected for intermittent training by
Fox (2). Consequently, the protocol used in training in experiment 2 may be optimal
with respect to improving both the aerobic and the anaerobic energy releasing
The intensive bicycle training may have affected the efficiency of cycling,
meaning that the relationship between power and ·VO2 may have changed. This
change may affect the measurement of anaerobic capacity because the accumulated
oxygen deficit is a calculated entity assuming a constant mechanical efficiency.
However, our subjects were sufficiently familiar with bicycle exercise through
repeated testing and experiments so that the relationship between the steady state
oxygen uptake and power did not change during the training periods. Therefore, the
pre- and posttraining data of the accumulated oxygen deficit should be comparable.
In summary, this investigation demonstrated that 6 wk of moderate-intensity
endurance training did not affect anaerobic capacity but that 6 wk of high-intensity
intermittent training (20 s exercise, 10 s rest; intensity 170%·VO2max) may improve
both anaerobic capacity and ·VO2max simultaneously.
1. Borg, G. Perceived exertion as an indicator of somatic stress. Scand. J. Rehabil.
Med. 2-3:92-98, 1970. [Context Link]
2. Fox, E. Sport Physiology. Philadelphia: W.B. Saunders, 1979, pp. 226. [Context
3. Hermansen, L., J. I. Medbo, A.-C. Mohn, I. Tabata, and R. Bahr. Oxygen deficit
during maximal exercise of short duration. (Abstract).Acta Physiol. Scand. 121:39A,
1984. [Context Link]
4. Krogh, A. and J. Lindhard. The changes in respiration at the transition from work to
rest. J. Physiol. 53:431-437, 1920. Full Text Bibliographic Links [Context Link]
5. Medbø, J. I. and S. Burgers. Effect of training on the anaerobic capacity. Med. Sci.
Sports Exerc. 22:501-507, 1990. Ovid Full Text Request Permissions Bibliographic
Links [Context Link]
6. Medbø, J. I., A.-C. Mohn, I. Tabata, R. Bahr, O. Vaage, and O. M. Sejersted.
Anaerobic capacity determined by maximal accumulated O2 deficit. J. Appl.
Physiol. 64:50-60, 1988. [Context Link]
7. Medbø, J. I. and I. Tabata. Relative importance of aerobic and anaerobic energy
release during short-lasting exhaustive bicycle exercise. J. Appl. Physiol. 67:1881-
1886, 1989. [Context Link]
8. Medbø, J. I. and I. Tabata. Anaerobic energy release in working muscle during 30 s
to 3 min of exhausting bicycling.J. Appl. Physiol. 75:1654-1660, 1993.
9. Saltin, B., G. Blomqvist, J. H. Mitchell, R. L. Johnson, Jr., and C. B. Chapman.
Responses to exercise after bed rest and after training. A longituinal study of adaptive
changes in oxygen transport and body composition.Circulation 38 (Suppl. 7):1-78,
1968. [Context Link]
10. Taylor, H. L., E. Buskirk, and A. Henshel. Maximal oxygen intake as an objective
measure of cardiorespiratory performance.J. Appl. Physiol. 8:73-80, 1955. [Context
ANAEROBIC TRAINING; AEROBIC TRAINING