Journal of Science and Medicine in Sport (2008) 11, 280286

ORIGINAL PAPER

Inuence of exercise mode and maximal

lactate-steady-state concentration on the
validity of OBLA to predict maximal
lactate-steady-state in active individuals
Tiago R. Figueira, Fabrizio Caputo, Jailton G. Pelarigo,
Benedito S. Denadai
Human Performance Laboratory, UNESP, Brazil
Received 20 July 2006 ; received in revised form 7 November 2006; accepted 3 February 2007
KEYWORDS
Constant work-rate;
Aerobic metabolism;
Cycling;
Running

Summary
The aim of this study was to analyze the effects of exercise
mode on the validity of onset of blood lactate accumulation (OBLA3.5-mM
xed blood lactate concentration) to predict the work-rate at maximal lactate steady state (MLSSwork-rate ). Eleven recreationally active males (21.3 2.9
years, 72.8 6.7 kg, 1.78 0.1 m) performed randomly incremental tests to determine OBLA (stage duration of 3 min), and 2 to 4 constants work-rate exercise
tests to directly determine maximal lactate steady state parameters on a cycleergometer and treadmill. For both exercise modes, the OBLA was signicantly
correlated to MLSSwork-rate , (cycling: r = 0.81 p = 0.002; running: r = 0.94, p < 0.001).
OBLA (156.2 41.3 W) was lower than MLSSwork-rate (179.6 26.4 W) during cycling
exercise (p = 0.007). However, for running exercise, there was no difference between
OBLA (3.2 0.6 m s1 ) and MLSSwork-rate (3.1 0.4 m s1 ). The difference between
OBLA and MLSSwork-rate on the cycle-ergometer (r = 0.86; p < 0.001) and treadmill (r = 0.64; p = 0.048) was signicantly related to the specic MLSS. We can
conclude that the validity of OBLA on predicting MLSSwork-rate is dependent on
exercise mode and that its disagreement is related to individual variations in
MLSS.
2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Introduction

Corresponding author.
E-mail address: [email protected] (B.S. Denadai).

The upper boundary of exercise intensities eliciting constant arterial lactate concentration over
time has been termed maximal lactate steady state
(MLSSwork-rate ).1 The determination of MLSSwork-rate

1440-2440/$ see front matter 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.jsams.2007.02.016

Exercise mode and maximal lactate-steady-state

requires 3 to 5 constant work-rate tests performed on different days. The physiological origin
of the mean lactate concentration corresponding to MLSSwork-rate (MLSS) is poorly understood,
but is acknowledged to be dependent on exercise
mode.1 Cycling has often elicited MLSS nearly of
5 mM, while running exercise shows MLSS nearly
of 4 Mm.24 Benets of MLSSwork-rate determination include: aerobic capacity evaluation; training
intensity prescription; prediction of exercise performance; and others of academic interest.5 Since
the MLSSwork-rate determination is time expensive,
many researchers have attempted to use single days
protocols for indirect MLSSwork-rate determination.
Most of these protocols are based on lactate
response to incremental work-rate tests, which provide an exponential curve of blood lactate concentration ([LA]) versus work-rate.2,6,7 The determination of onset of blood lactate accumulation (OBLA,
dened as the work-rate at [LA] of 4 mM) as an
indirect index of MLSSwork-rate has been widely used
and some studies have showed good validity of this
index in different exercise modes.2,810 In addition,
some authors have argued against the utilization of
a xed lactate value of 4 mM as an indirect index of
MLSSwork-rate during incremental exercise test.1113
The earliest criticisms concerned the effects of
aerobic training on lactate response which would
reduce the [LA] corresponding to MLSSwork-rate during an incremental work-rate test.11 However,
experimental results have indicated that OBLA is
a valid indirect index of MLSSwork-rate during cycling
exercise regardless of the aerobic training status.9
More recently, some considerations have been
made based on the great interindividuality of MLSS
(1.97.5 mM), and that mean values of MLSS during cycling exceed 4 mM.14 In accordance with some
authors, these factors (great interindividuality and
MLSS average > 4 mM) will preclude estimates of the
MLSSwork-rate from xed blood [LA] values.3,12,13
In this scenario, it can be hypothesized that
exercise mode and MLSS interindividuality inuence the validity of OBLA to predict MLSSwork-rate .
However, the physiological mechanisms and
experimental proof of this hypothesis are missing. Possible effects of exercise mode and the
interindividuality of MLSS on the validity of OBLA
for estimating MLSSwork-rate can only be evaluated
by using constant work-rate tests and the direct
determination of the MLSS. In addition, to avoid an
eventually random effect of group composition on
the selected variables, the same subject should be
studied during both exercise modes. Thus, the aim
of the present study was to analyze the inuence
of exercise mode on the validity of OBLA to predict
MLSSwork-rate .

281

Methods
Subjects
Eleven recreationally active male physical education students (21.3 2.9 years, 72.8 6.7 kg,
1.78 0.1 m) volunteered to participate in this
study. The protocol was approved by the universitys ethics committee and all subjects gave
informed consent. The experiments were in accordance with the current laws of the country in which
the experiments were performed.

Experimental designs
Each subject performed one set of tests (an
incremental test and 24 constant load exercise
bouts) in random order for both exercise modes.
All running tests were performed on a motorized treadmill (INBRAMED Super ATL, Brazil) with
the gradient set at 1%. Cycling tests were conducted on a mechanically braked cycle-ergometer
(Monark 828E, Stockholm, Sweden), with pedal
frequency maintained constant at 60 rpm. OBLA,
peak heart rate (HRpeak ) and peak work-rate
(PWR ) were obtained from incremental work-rate
tests, and maximal lactate steady state parameters (MLSSwork-rate, MLSS and MLSSHR ) were directly
determined from constant work-rate tests at
cycling and running exercises. The interval between
each test was at least 48 h. The subjects were
instructed to arrive at the laboratory in a rested
and fully hydrated state, at least 2 h post-prandial,
and to avoid strenuous exercise in the 48 h preceding a test session. Each subject was tested at the
same time of day (9:30 1:00 h) to minimize the
effects of diurnal biological variation.

Procedures
The incremental work-rate test for the cycleergometer started at 60 W and was continuously
increased to exhaustion by 20 W every three minutes. For the treadmill, the incremental work-rate
test started at a velocity of 1.94 m s1 and was
increased to exhaustion by 0.28 m s1 every three
minutes, all stages were followed of 30-s period
of rest. The PWR was determined according to the
equation 15 : PWR (W or m s1 ) = work-rate last stage
completed (W or m s1 ) + [t (s)/step duration (s) x
step increment (W or m s1 )]; t is the time of
the uncompleted stage.
Each constant work-rate test lasted 30 min. The
power output/velocity of the rst constant work-

282

T.R. Figueira et al.

rate test corresponded to a [LA] of 3.5 mM (OBLA)

measured during the incremental work-rate test. A
xed blood [LA] of 3.5 mM, instead of 4.0 mM, has
been used in incremental tests with 3 min stages.2,7
If during the rst constant work-rate test a steady
state or a decrease in lactate was observed, further
subsequent 30 min constant work-rate tests from
4 to 8% (10 W at cycle-ergometer and 0.14 m s1
at treadmill) higher work-rate were performed on
separate days until no [LA] steady state could be
maintained. If the rst constant work-rate test
resulted in a clearly identiable increase of the [LA]
and/or could not be completed due to exhaustion,
further constant work-rate tests were conducted
with subsequently reduced (4 to 8%) work-rate.
The MLSSwork-rate was dened as the highest
power output (cycling exercise) or velocity (running exercise) at which [LA] did not increase by
more than 1 mM between Minutes 10 and 30 of
the constant load test.1,3,4,14 MLSS was calculated as the mean [LA] measured at Minutes 10
and 30 of the MLSSwork-rate . The maximal lactate
steady state intensity (MLSSintensity ) was dened
as the ratio between MLSSwork-rate and PWR . The
[LA] corresponding to the MLSSwork-rate during incre[LA]
mental test (MLSSworkrate ) was calculated from
steps just below and above MLSSwork-rate by linear
interpolation.2 Blood samples (25 l) were collected from the ear lobe into microcentrifuge tubes
containing 50 l NaF (1%) during the nal 15 s of
every 3-m (cycle-ergometer incremental test) or
during the 30-s rest period (treadmill incremental
test) and at 10th and 30th minute during constant work-rate tests. The [LA] was determined by
an electrochemical method (YSL 2300 STAT, Yellow

Table 1

Springs, OH, USA). The heart rate (HR) was continuously monitored by a heart rate monitor (Polar
Vantage NV, Finland) and values recorded every ve
minutes during constant work-rate tests. The HR
corresponding to MLSS (MLSSHR ) was the mean of
the values registered during MLSSwork-rate and is also
reported as percent of the peak HR (%MLLSHR ).

Statistical analyses
Data are reported as means SD. The paired ttest was applied to compare MLSS, MLSSintensity
and MLSSHR between exercise modes and to compare OBLA and MLSSwork-rate within the same
exercise mode. Pearson product moment correlation coefcient was used to assess magnitude
of relationship between the selected variables. In
addition, the typical error of estimate was calculated as described by Hopkins16 . Signicance was
set at p 0.05.

Results
Table 1 contains the variables determined from
the incremental work-rate tests (PWR, OBLA and
HRPeak ) and at the constant work-rate tests
[LA]
(MLSSwork-rate , MLSSintensity , MLSS, MLSSworkrate ,
MLSSHR and %MLSSHR ) for both exercise modes.
OBLA for the cycle-ergometer was lower than
MLSSwork-rate (p = 0.007), however for the treadmill there was no difference between OBLA
and MLSSwork-rate . The correlation between the

Selected variables obtained from constant and incremental work-rate tests for both ergometers
Cycle-ergometer

PWR
MLSSwork-rate
OBLA
MLSSintensity (%)
MLSS (mM)
MLSSwork-rate [LA] (mM)
HRPeak (bpm)
MLSSHR (bpm)
%MLSSHR (%)

230.3
179.6
156.2
78.2
5.8
4.5
183.1
162.9
89.0

34.1a
26.4a,b
41.3a
4.9
2.1d
1.2d
10.3d
9.4d
3.2

Treadmill
4.1
3.1
3.2
75.9
3.8
3.2
194.3
172.6
89.0

0.3c
0.4c
0.6c
5.1
0.9
0.7
7.1
7.9
5.8

C95%
n.a.
n.a.
n.a.
6.91.8
0.33.2
0.32.3
4.917.5
3.516.9
0.040.05

The 95% condence intervals of the difference between ergometers are shown. PWR , peak work-rate; MLSSwork-rate , maximal
lactate steady state work-rate; OBLA, onset of blood lactate accumulation; MLSSintensity , MLSSwork-rate as percent of PWR ; MLSS,
[LA]

lactate concentration at MLSSwork-rate ; (MLSSworkrate ), calculated blood lactate concentration corresponding to the MLSSwork-rate
during incremental test; HRPeak , peak heart rate; MLSSHR , heart rate average at MLSSwork-rate ; % MLSSHR , MLSSHR as a percent of
peak heart rate.
a Units are Watts.
b Signicantly different from OBLA (p < 0.05)
c Units are metres per minute; n.a., not applicable.
d Signicantly different from treadmill (p < 0.05).

Exercise mode and maximal lactate-steady-state

Figure 1 Relationships between MLSSwork-rate and OBLA

at cycle-ergometer (upper panel) and at treadmill (lower
panel). (, line of identity; , regression line).

OBLA and MLSSwork-rate were signicant for both

exercise modes (cycling: r = 0.81, p = 0.002; running: r = 0.94, p < 0.001) (Fig. 1). The extent
of disagreement between MLSSwork-rate and OBLA
(MLSSwork-rate OBLA) was signicantly related to
MLSS for the cycle-ergometer (r = 0.86, p < 0.001)
and for the treadmill (r = 0.64, p = 0.048) (Fig. 2).
The typical error of estimate suggests a good
validity of OBLA to estimate MLSSwork-rate during
running exercise (5.5%) but not in cycling (20.7%).
There were signicantly correlations between
[LA]
MLSSworkrate and MLSS for both cycling (r = 0.70,
p = 0.017) and running (r = 0.70, p = 0.025).

Discussion
The main nding of this investigation indicates
that the validity of OBLA to predict MLSSwork-rate is
dependent on exercise mode (cycling vs. running)
and that the extent of disagreement is related to
individual variations in MLSS (Fig. 2).
Some studies had veried that exercise mode has
an important inuence on the MLSS.3,12,14 The highest values of MLSS (5.46.6 mM) have been found
in exercises with a lower amount of active muscle
(e.g. speed skating and cycling), while the lowest
values (3.14.0 mM) have been veried in rowing and running.2,3 Thus, Beneke1 suggested that

283

Figure 2 Relationship between maximal lactate steady

state concentration (MLSS) and the difference between
MLSSwork-rate and OBLA for the cycle-ergometer (upper
panel) and treadmill (lower panel).

MLSS is inversely related to the specic muscle

mass primarily engaged in the exercise, although,
the lactate production is likely higher when larger
amount of muscle mass is added to exercise.17
On the other hand, variables that can inuence
lactate exercise metabolism, such as monocarboxylate transporters isoforms, myosin heavy chain
isoforms and aerobic enzymes do not seem to be
related to MLSS during cycling.18 Even using similar muscle groups, comparing cycling to running,
specic biomechanical characteristics impose large
differences on physiological demand, mainly the
amount of muscle mass and the intramuscular pressure may elicit particular systemic cardiovascular
response.19,20 Likely, an integrative effect, related
to intrinsic hemodynamic and neuro-endocrine
response to different exercise modes remains to
be claried and could play a role in MLLS dependence on exercise mode. Recently Beneke et al. 14
and Schuylenbergh et al. 12 , analyzing the same
subjects (rowers and triathletes, respectively) in
different exercise modes, conrmed the effects of
exercise mode on MLSS. This experimental design
is particularly important, since there is a great
interindividuality on MLSS.21 Therefore, the results
of the present study (Table 1), seem to conrm
that the MLSS is dependent on exercise mode, being
higher in the exercise with lower amount of primarily muscle mass engaged (e.g. cycling exercise).

284
In agreement with Beneke et al. 3 , we also
found that the MLSSintensity was similar between
exercise modes. However, it is important to note
that the characteristics of the incremental test
(stage duration and increment rate) inuence PWR ,
and consequently, the MLSSintensity .22 In the present
study, MLSSintensity during cycling was higher (78.2
vs. 70.5% PWR ) than previously reported data in
subjects with similar MLSSwork-rate 4 and close to the
values found in well trained cyclists (79.5%).9 Thus,
we argue that the high MLSSintensity found here for
untrained subjects during cycling exercise is due
to the low increment rate (6.7 W min1 ), which
resulted in a low PWR . Conrming this possibility,
there is some evidence that the MLSSwork-rate
occurs at higher % VO2max during running compared
to cycling exercise.23,24
For running, the good validity of OBLA to estimate MLSSwork-rate found in the present study agrees
with previous studies conducted in this exercise
mode.2,8,10 Although these studies had analyzed
subjects with different aerobic tness levels
(sedentary, active, well trained runners and soccer
players), the values of OBLA and MLSSwork-rate were
statistically similar and highly correlated (r > 0.80).
The highest correlation level (r = 0.97) was found by
Heck et al.,2 likely due to a greater heterogeneity
of the aerobic training status of their subjects (e.g.
25% = trained runners and 75% = active subjects). To
our knowledge, this is the rst study that analyzed
the validity of OBLA in a more homogeneous sample
of healthy subjects with no previous training history
in running and cycling. Therefore, the present study
conrms that OBLA, when identied in an incremental test with a 3-min stage and by considering a
xed 3.5-mM of [LA] has a good validity to predict
MLSSwork-rate during running in active person.
The OBLA signicantly underestimated the
actual MLSSwork-rate during cycling exercise. Moreover, the typical error of estimate was higher in
this exercise mode (20.7%) than in running (5.5%).
This low validity of OBLA to estimate MLSSwork-rate
is apparently contradictory to the data obtained by
Denadai et al.,9 which has shown a good validity of
OBLA regardless the training status. However, an
aspect that should be taken into account is the different increment rate utilized in the incremental
protocol test between these studies. In the present
study, the increment rate (6.7 W min1 ) was lower
than used by Denadai et al.9 (11.7 W min1 ). Heck
25 demonstrated an increase of approximately
1.4 W in OBLA when the increment rate was
increased by 1.0 W min1 . Therefore, differences
in the increment rate could explain, at least
in part, the contradictory results of these two

T.R. Figueira et al.

studies. In our study, the increment rates (cycletreadmill = 0.28 m s1 )
ergometer = 6.7 W min1 ;
were established to equal the relative increment
of the expected MLSSwork-rate in both exercises
exercise modes. In fact, they were very similar
(cycle-ergometer = 11%; treadmill = 9%) to the
specic MLSSwork-rate . In our opinion, this aspect
must be considered in the analysis of the possible
inuences of exercise mode on the validity of OBLA
to predict MLSSwork-rate .
The MLSS (5.8 mM) found in the present study
during cycling is slightly higher than observed by
Beneke et al. 21 and in others studies conducted
in our laboratory (5.0 mM) using the same [LA]
analysis.4,9,26 In our sample, three subjects presented MLSS higher than 8.0 mM during cycling.
Interestingly, these subjects had presented the
highest difference between MLSSwork-rate and OBLA
(>50 W), which can be better observed by the significant correlation (r = 0.86, p = 0.0000) between this
difference (MLSSwork-rate OBLA) and MLSS during
cycling (Fig. 2). Although there was no signicant
difference between OBLA and MLSSwork-rate , a
signicant correlation (r = 0.64, p = 0.0482) was
also found during running. These results suggested
that MLSS interferes on the xed [LA] value, which
must be utilized during the incremental protocol
in order to predict MLSSwork-rate , irrespective
of exercise mode. Furthermore, the calculated
[LA] corresponding to the MLSSwork-rate during the
[LA]
incremental test (MLSSworkrate ) was signicantly
related to MLSS during cycling (r = 0.70, p = 0.017)
and running (r = 0.70, p = 0.025). However, in the
study conducted by Heck et al.,2 there was no signicant correlation (calculated from the individual
values presented in the table II) between MLSS and
[LA]
MLSSworkrate . It is possible that the differences
on the increment rate utilized during running in
our study (1.5 103 m s2 ) and by Heck et al. 2
(2.2 103 m s2 ), could explain these different
results. The total time duration of the incremental
test is dependent on the increment rate. When the
increment rate is rather small, a given work-rate
will be attained later. As the [LA] is work-rate and
time dependent,2 protocols with small increment
rates can proportionate sufcient time to a more
individual and realistic lactate kinetics during
incremental tests, when compared to constant
work-rate tests.
The results of the present study together with
previous published data 2,9 initiate an important
issue in regard to the interaction between the
characteristics of incremental test (increment
rate) and the interindividual variations of MLSS on
the validity of OBLA to estimate MLSSwork-rate . It

Exercise mode and maximal lactate-steady-state

is possible that the determination of OBLA from
some protocols, (e.g. with large increment rates),
may be less inuenced by interindividual variations
of the MLSS. Thus, the major limitation of this
study was the employment of only one incremental
protocol per exercise mode. Future research is
required to investigate the possible interaction
between the characteristics of different incremental protocols and MLSS on the validity of OBLA to
estimate MLSSwork-rate for different exercise modes.
We can conclude that the validity of OBLA to
predict MLSSwork-rate , at least in the experimental
conditions of this study, is dependent on exercise mode (cycling vs. running) with OBLA being
valid to predict MLSSwork-rate on running but not
on cycling. The disagreement between MLLSwork-rate
and OBLA is related to individual variations on MLSS.
We therefore, recommend that the MLLSwork-rate be
determined directly when precision is required.

Practical implications
OBLA determination may be an accurate index
for maximal lactate steady state prediction
depending on exercise mode and likely test
protocol.
Progressive exercise test protocol might
be important, which implies the need of
previously validated protocols of OBLA determination.
For the same subject the highest constant
blood lactate concentration over time is higher
in cycling than running.

Acknowledgements
We thank the subjects for participation in this study,
FAPESP and CNPQ for nancial support. We are
grateful to Ms. Anna Wittekind for reviewing the
paper.

References
1. Beneke R. Maximal lactate steady state concentrations
(MLSS): experimental and modelling approaches. Eur J Appl
Physiol 2003;88:3619.
2. Heck H, Mader A, Hess G, Mucke S, Muller R, Hollmann W.
Justication of the 4 mmol/l lactate threshold. Int J Sports
Med 1985;6:11730.
3. Beneke R, von Duvillard S. Determination of maximal lactate steady state response in selected sports events. Med
Sci Sports Exerc 1996;28:2416.
4. Denadai BS, Ruas VDA, Figueira TR. Maximal lactate steady
state concentration independent of pedal cadence in active
individuals. Eur J Appl Physiol 2006;96:47780.

285
5. Billat V, Sirvent P, Py G, Koralsztein JP, Mercier J. The concept of maximal lactate steady state: a bridge between
biochemistry, physiology and sport science. Sports Med
2003;33:40726.
6. Beneke R. Anaerobic threshold, individual anaerobic
threshold, and maximal lactate steady state in rowing. Med
Sci Sports Exerc 1995;27:8637.
7. Smith CGM, Jones AM. The relationship between critical
velocity, maximal lactate steady-state velocity and lactate turnpoint velocity in runners. Eur J Appl Physiol
2001;85:1926.
8. Jones AM, Doust JH. The validity of the lactate minimum
test for determination of the maximal lactate steady state.
Med Sci Sports Exerc 1998;30:130413.
9. Denadai BS, Figueira TR, Favaro OP, Gonc
alves M. Effect
of the aerobic capacity on the validity of the anaerobic
threshold for determination of the maximal lactate steady
state in cycling. Braz J Med Biol Res 2004;37:15516.
10. Denadai BS, Gomide EBG, Greco CC. The relationship
between onset of blood lactate accumulation, critical
velocity, and maximal lactate steady state in soccer players. J Strength Cond Res 2005;19:3648.
11. Keul J, Simon G, Berg A, Dickhuth H, Goertler I, Kubel R.
Determination of individual anaerobic threshold for performance test and exercise prescription. Deutsche Zeitschrift
f
ur Sportmedizin 1979;30:2128.
12. Schuylenbergh R, Eynde B, Hespel P. Prediction of sprint
triathlon performance from laboratory tests. Eur J Appl
Physiol 2004;91:949.
13. Kilding AE, Jones AM. Validity of a single-visit protocol to
estimate the maximum lactate steady-state. Med Sci Sports
Exerc 2005;37:173440.
14. Beneke R, Leithauser R, Hutler M. Dependence of the maximal lactate steady state on the motor pattern of exercise.
Br J Sports Med 2001;35:1926.
15. Kuipers H, Verstappen FT, Keizer HA, Geurten P, van
Kranenburg G. Variability of aerobic performance in the
laboratory and its physiologic correlates. Int J Sports Med
1985;6:197201.
16. W.G. Hopkins (2000). Analysis of validity by linear
regression (Excel spreadsheet). In: A new view of
statistics. Available at: http://www.sportsci.org/resource/
stats/xvalid.xls. Accessed 19 October 2006.
17. Jersen-Urstad M, Svedenhag J, Sahlin K. Effect of muscle
mass on lactate formation during exercise in humans. Eur J
Appl Physiol Occup Physiol 1994;69:18995.
18. Loekkegaard J, Pedersen PK, Juel C, Sjoegarrd G. Individuals variations in maximal lactate steady state and their
relationship with muscle buffering capacity and lactate
transporters. Med Sci Sports Exerc 2001;33:S330.
19. Clarys JP, Cabri J, Gregor RJ. The muscle activity paradox during circular rhythmic leg movements. J Sports Sci
1988;6:22937.
20. Van Ingen Schenau GJ, Bobbert MF, De Haan A. Does
elastic energy enhance work and efciency in the stretchshortening cycle? J Appl Biomech 1997;13:389415.
21. Beneke R, Hutler M, Leithauser RM. Maximal lactate-steadystate independent of performance. Med Sci Sports Exerc
2000;32:11359.
22. Coen B, Urhausen A, Kindermann W. Individual anaerobic
threshold: methodological aspects of its assessment in running. Int J Sports Med 2001;22:816.
23. Billat V, Mille-Hamard L, Petit B, Koralsztein JP. The role of
cadence on the VO2 slow component in cycling and running
in triathletes. Int J Sports Med 1999;20:42937.
24. Caputo F, Denadai BS. Effects of aerobic endurance
training status and specicity on oxygen uptake kinet-

286

T.R. Figueira et al.

ics during maximal exercise. Eur J Appl Physiol 2004;93:

8795.
25. Heck H, von-Rosen I, Rosskopf P. Dynamik des Blutlaktats
bei konstanter Fahrrad- und Drehkurbelarbeit. In: Liesen
H, Wei M, Baum M, editors. Regulations und Repairmechanismen.. Cologne: Dt Arzte-Verlag, Ko; 1994. p. 18790.

26. Ruas VDA, Figueira TR, Caputo F, Barbeitos DF, Denadai BS.
Effects of pedaling cadence on the relationship between
anaerobic threshold and maximal lactate steady state
in active male individuals. Port J Sport Sci 2006;6:
1520.

Available online at www.sciencedirect.com