Influence of different rest intervals during active or passive recovery on repeated sprint swimming performance

Abstract

The purpose of this study was to investigate the effect of active or passive recovery after two different rest intervals on performance during repeated bouts of maximal swimming exercise. Sixteen swimmers (eight males and eight females) performed four trials in a counterbalanced order. Eight repetitions of 25-m sprints (8 x 25 m), with a rest interval of 45 or 120 s, followed by a 50-m sprint test 6 min later, were performed in each trial. The 45 or 120-s interval was either active (A45 and A120) or passive (P45 and P120). The intensity of the active recovery corresponded to 60% of the individual best 100-m velocity. Performance time was recorded using an official competition timing system. The first 25-m sprint was comparable across trials (P>0.05), but performance was decreased after the second sprint during active compared to passive recovery, irrespective of the interval duration (P<0.05). The 50-m sprint time was 2.4% better in the P120 and A120 compared to the A45 and P45 trials (P<0.05). After completing the 8x25 m, blood lactate was decreased with active recovery when the interval period was 120 s (P120 vs A120, P<0.05). Blood lactate concentration at the start as well as 5 min after the 50-m sprint was lower in the A120 and A45 compared to the P120 and P45 trials respectively (P<0.05). Plasma glycerol was not different between trials (P>0.05), whereas plasma ammonia was higher in the A45 compared to the P120 trial (P<0.05). The interval period separating short-duration sprints may therefore alter performance when subsequent maximum exertion is applied. For sustained sprinting ability, passive recovery is advised during repeated swimming sprints of short duration.

Full text

ORIGINAL ARTICLE
Argyris G. Toubekis Æ Helen T. Douda
Savvas P. Tokmakidis
Influence of different rest intervals during active or passive
recovery on repeated sprint swimming performance
Accepted: 19 September 2004 / Published online: 20 November 2004
 Springer-Verlag 2004
Abstract The purpose of this study was to investigate the
effect of active or passive recovery after two different rest
intervals on performance during repeated bouts of
maximal swimming exercise. Sixteen swimmers (eight
males and eight females) performed four trials in a
counterbalanced order. Eight repetitions of 25-m sprints
(8·25 m), with a rest interval of 45 or 120 s, followed by
a 50-m sprint test 6 min later, were performed in each
trial. The 45 or 120-s interval was either active (A45 and
A120) or passive (P45 and P120). The intensity of the
active recovery corresponded to 60% of the individual
best 100-m velocity. Performance time was recorded
using an official competition timing system. The first 25-
m sprint was comparable across trials (P>0.05), but
performance was decreased after the second sprint dur-
ing active compared to passive recovery, irrespective of
the interval duration (P<0.05). The 50-m sprint time
was 2.4% better in the P120 and A120 compared to the
A45 and P45 trials (P<0.05). After completing the
8·25 m, blood lactate was decreased with active recov-
ery when the interval period was 120 s (P120 vs A120,
P<0.05). Blood lactate concentration at the start as well
as 5 min after the 50-m sprint was lower in the A120 and
A45 compared to the P120 and P45 trials respectively
(P<0.05). Plasma glycerol was not different between
trials (P>0.05), whereas plasma ammonia was higher in
the A45 compared to the P120 trial (P<0.05). The in-
terval period separating short-duration sprints may
therefore alter performance when subsequent maximum
exertion is applied. For sustained sprinting ability, pas-
sive recovery is advised during repeated swimming
sprints of short duration.
Keywords Active recovery Æ Sprint swimming Æ
Interval duration Æ Performance
Introduction
It has been suggested that when active recovery is ap-
plied between repeated bouts, it may facilitate perfor-
mance more than when applying passive rest (Maglischo
2003). When active recovery is applied during the rest
interval between short-duration sprints in cycling (i.e. 6–
30 s), performance is better maintained compared to
passive recovery (Signorile et al. 1993; Ahma idi et al.
1996; Bogdanis et al. 1996b). This may be attributed to
increased muscle blood flow (Bangsbo et al. 1994), which
in turn may facilitate oxygen supply to the cell for a
faster phosphocreatine (PCr) resynthesis.
During successive bouts of sprint exercise, a short-
duration interval period leads to incomplete resynthesis
of PCr (Gaitanos et al. 1993; Bogdanis et al. 1996a,
1998). At the same time, anaero bic glycolysis results in
high levels of muscle lactate associated with hydrogen
ions. Lactate can no longer be considered the cause of
fatigue (Gladden 2004). However, intramuscular meta-
bolic byproducts, such as inorganic phosphate as well as
hydrogen ions, may act to impair muscle function
(Hermansen 1981). Even though aerobic ATP contri-
bution increases to compensate for the reduced rate of
ATP provided by anaerobic sources (Bogdanis et al.
1996a), it does not prevent the significant reduction in
power output and velocity which occurs during repeated
short-duration cycling, swimming or running (Balsom
et al. 1992a; Gaitanos et al. 1993; Peyrebr une et al.1998).
Studies conducted with swimmers examined the
application of active recovery after long-duration (about
2 min) swimming exercise and after long resting periods
(i.e. 14 min, Felix et al. 1997; or 3 min, McMurray
1969). However, it is possible that changes in duration
and intensity of the exercise as well as duration of the
resting interval alter energy contribution. It has been
A. G. Toubekis Æ H. T. Douda Æ S. P. Tokmakidis (&)
Department of Physical Education
and Sport Science, Democritus University
of Thrace, 69100 Komotini , Greece
E-mail: stokmaki@phyed.duth.gr
Tel.: +30-531-39649
Fax: +30-531-39683
Eur J Appl Physiol (2005) 93: 694–700
DOI 10.1007/s00421-004-1244-9

shown that performance af ter active recovery may be
better maintained when the intensity is maximum and
the duration is short (Weltman et al. 1977; Signorile
et al. 1993; Bogdanis et al. 1996b), but not when longer
duration exercise is performed (e.g. 5 min, Weltman
et al. 1979). Despite this fact, the findings on swimming
are inconclusive since only one study found increased
performance on a 200-yard (182.88-m) trial after active
compared to passive recovery (Felix et al. 1997). Data
from our laboratory found negative effects on swimming
velocity after active recovery (Toubekis and Tokmakidis
2003). Thus, there is lim ited information regarding the
effects of active recovery on performance during short
distances (25–50 m), which are frequently used during
sprint training in swimming (Maglischo 2003). The
purpose of this study was to examine the effects of active
recovery on performance during repeated 25-m sprints
and a subsequent 50-m sprint test. Two different rest
periods of 45 and 120 s were selected to examine any
possible influence of active recovery duration on per-
formance. It was hypothesized that active recovery
would lead to better maintenance of sprint swimming
performance.
Methods
Subjects
Sixteen swimmers [eight males, age 21.2 (0.6) years,
height 177.8 (2.1) cm, body mass 73.8 (1.5) kg; and eight
females, age 21.4 (0.6) years, height 169.6 (2.2) cm, body
mass 66.4 (3.0) kg] with past competitive experience
agreed to participate in this study, which was conducted
with the approval of the University’s Ethical Committee.
All swimmers were informed in detail about the experi-
mental procedure and signed an informed consent
statement. Participants refrained from competitive
swimming and formal training for at least 2 years before
the examination period, but they performed recreational
swimming two or three times per week.
Familiarization
Each swimmer participated in two familiarization ses-
sions. Maximum exercise of 25-m bouts with 45 or 120 s
rest interval was performed. The time of self-sel ected
active reco very for at least 200 m after the completion of
four to six sprints was recorded. This self-selected pace
corresponded to 68 (2)% of the 100-m maximum
velocity, and swimmers were instructed to adopt the
pace suggested by Cazorla et al. (1983) corresponding to
60% of the 100-m velocity in a following session. A test–
retest was performed with 4·25 sprints separated with
45-s intervals. No significant difference was observed
between trials in the first sprint or in the mean perfor-
mance of the four sprints (P>0.05). Preliminary trials
started within a week of the familiarization sessions.
Female swimmers were not tested during menstruation
in any of the familiarization, preliminary or main trials
Preliminary trials
All swimmers initially performed a 100-m trial, applying
the rules of an official competition, in order to determine
the best performance time on this distance. A 50-ll
blood sample was taken from a finger 5 min after the
trial. On a second day, each swimmer completed two
repetitions of 200 m, swimming at a pace corresponding
to 60% of the 100-m velocity. A blood sample was taken
at rest and immediately after each 200-m repe tition.
Blood lactate concentration was not changed after the
second [2.26 (0.27) mmol l
1
] compared to the first
200 m [2.49 (0.22) mmol l
1
, P>0.05]. This confirmed
that the adopted velocity for active recovery did not
cause increased accumulation of blood lactate.
Main trials
All swimmers completed four main trials under different
conditions. Eight repetitions of 25-m sprints (8·25 m),
with an interval period of 45 or 120 s followed by a 50-m
sprint test 6 min later, were performed in each trial. The
45 or 120-s interval was either active (A45 and A120) or
passive (P45 and P120). When passive recovery was
applied, swimmers rested standing still in the water,
while in active recovery, they swam at a pace corre-
sponding to 60% of the best 100-m individual velocity.
A period of about 6–8 s preceding and following each
sprint was allowed before the swimmers could get ready
and take the starting position for active recovery in the
A45 and A120 trials. As a consequence, about 30 s and
100 s were available for active recovery in the A45 and
A120 trials respectively. When the interval period in the
A45 trial was not enough to cover the 25-m distance of
the swimming pool with the required velocity for active
recovery, the swimmers were asked to swim at a prede-
termined point and return to the finishing side of the
pool. During the 6-min interval period separating the
8·25-m and the 50-m sprint, a time of 150 s was allowed
for blood sampling while the swimmers rested passively
next to the swimming pool. As a consequence, the time
available for active recovery after the 8·25 m in the
A120 and A45 trials was about 210 s.
All trials were performed at the same time of the day
after a controlled warm-up (600 m swim, 200 kicks, 200
pulls, 4·10-m spri nts) in counterbalanced order. The
front crawl swimming style was used in all trials starting
with a push-off from within the water in a 25-m indoor
swimming pool. The water temperature was kept at 25–
26C.
The time required to complete each 25 or 50-m bout
was recorded with an accuracy of 1/1,000 s by an official
timing system used during swimming competition
(TancCo timing system). Briefly, two touchpads were
695

affixed to the walls at both ends of the swimming pool
and both were connected to the recording system where
swimming time was printed after touching any side of
the pool. All swimmers were familiarized with the
starting and finishing procedure, and easily adopted
the pace of active recovery. At the same time, one of the
investigators was walking alongside the pool deck giving
instructions about pacing when necessary during active
recovery.
Diet and exercise restrictions
Participants were asked to follow similar activities, and
apply the same diet with controlled carbohydrate con-
tent 2 days before each trial. They were also advised to
avoid heavy exercise, alcohol or caffeine consumption
for 24 h preceding each trial.
Blood sampling and analysis
On arrival at the pool for each main trial, swimmers
rested for 20 min and then a venous blood sample (5 ml)
was collected from an antecubital vein. One more ve-
nous sample (5 ml) was collected during the second
minute of recovery after the last 25-m sprint. Capillary
blood samples (50 ll) were also taken from a fingertip as
follows: at rest, after warm-up, at the second minute of
recovery after the 8·25 m, as well as before and 5 min
after the 50-m sprint. A time of about 100 s was used to
get the swimmer dried after the 8·25 m, and an addi-
tional 50 s to get the finger blood sample when active
recovery was applied.
The venous sample was placed into tubes containing
lithium heparin, centrifuged immediately at 4,000 rpm
(2,361 g) for 10 min, and the supernatant was then
placed into an ice bath. Ammonia was determined
within 3 h after sampling, using the procedures
described by Sigma (kit no. 171-UV, coefficient of var-
iation 4.2%). The remaining plasma was stored at
80C and later analyzed for glycerol concentration
(Sigma kit no. 337, coefficient of variation 2.2%).
Capillary samples were deproteinized in tubes con-
taining 100 ll trichloroac etic acid. The tubes were
centrifuged at 3,200 rpm (1,522 g) for 10 min, stored
at  80C, and later analyzed enzymically for lactate
concentration using a spectrophotometer (Hitachi
U-2001) and the procedure described by Sigma (kit
no. 826-UV).
Triplicate samples of 75 ll were collected at rest
and after the 8·25 m into heparinized capillary tubes.
After centrifuging for 15 min at 13,000 rpm (15,682 g),
the haematocrit was determined. Haemoglobin was
measured using the Cyanmethemoglobin Method
(Sigma, kit no. 525). Changes in plasma volume were
estimated from the changes in haemoglobin and
haematocrit values as described by Dill and Costill
(1974).
Statistical analysis
A three-way analysis of variance for repeated measures
was used to examine differences between means
(sex · trials · sprints). Since no significant interaction
for sex was observed in most variables, a two-way
analysis of variance was performed on the whole sample
(trials · sprints, independent of sex). The Tukey post-
hoc test was used to locate the observed differences. The
Pearson correlation coefficient was used to examine
correlation between variables. The accepted level of
significance was set at P<0.05. The results are presented
as means (SE).
Results
Preliminary tests
The mean time for the 100 m was 63.12 (2.40) s for
males and 75.90 (1.80) s for females; blood lactate values
5 min after this trial were 13.0 (0.6) mmol l
1
and 10.5
(0.6) mmol l
1
respectively (time and lactate, males vs
females, P<0.05).
Main trials
Swimming velocity decreased more during active com-
pared to passive recovery between sprints (main effect
trials and interaction, P<0.05, Fig. 1). This was ob-
served irrespectively of the time interval between sprints.
The average times for 8·25 m were 15.12 (0.39), 15.72
(0.44) 15.93 (0.43) and 16.49 (0.49) s for the P120, A120,
P45 and A45 trials respec tively (P120 vs A45, P<0.05).
The time to complete the first 25-m sprint was not sta-
tistically different between trials [P120, 14.93 (0.39) s;
A120, 15.00 (0.38) s; P45, 15.25 (0.39) s; A45 15.17
Fig. 1 Time to complete each 25-m sprint in each trial.
a
P<0.05
between the corresponding sprint in trials at the same interval,
b
P<0.05 compared to first sprint in the A120 and A45 trials,
c
P<0.05 compared to first sprint in the P120 and P45 trials, [n=16,
mean (SE)]
696

(0.42) s; P>0.05; Fig. 1]. However, during the A120 and
A45 trials, velocity decreased by 5% and 6% respec-
tively in the second sprint compared to the first, whereas
no difference was observed in the corresponding sprint
for the P120 and P45 trials (Fig. 1). Swimming velocity,
compared to the first sprint was well maintained, up to
the sixth sprint in the P120 and up to the second sprint in
the P45 trials (Fig. 1).
The ti me to complete the 50-m sprint, performed
6 min after the 8·25 m, was not affected by the type of
recovery, but was faster during the longer interval [P120,
33.59 (1 .05) s; A120, 33.58 (1.02) s] compared to the
shorter interval trials [P45, 34.33 (1.06) s; A45, 34.48
(1.17) s; P<0.05].
Blood metabolites
Blood lactate concentration increased in all four trials
after the 8·25 m compared to resting and post-warm-up
values, and was higher in the P120 compared to the
A120 trial (main effect trials and interaction, P<0.05).
The corresponding lactate values in the P45 and A45
trials were not statistically different ( P>0.05). Six min-
utes later, at the start of the 50-m sprint, blood lactate
concentration decreased, co mpared to the post 8·25 m
value, only when recovery was active (P<0.05, Fig. 2).
After the 50-m sprint, lactate concentration was in-
creased in all four trials compared to the pre-start val-
ues. However, it was lowe r during both the active
recovery trials compared to the corresponding passive
recovery (P<0.05, Fig. 2). No difference was observed
between passive P120 and P45 or active A120 and A45
trials at any point. The blood lactate concentration at
the end of the 6-min recovery period was not correlated
with performance on the 50-m sprint (r=0.22–0.37,
P>0.05). The blood lactate measured 5 min after the
50-m sprint was correlated to performance time in the
P120 and A120 (r=0.54 and 0.56 respectively,
P<0.05), but not in the P45 and A45 trials (r= 0.41
and 0.51 respectively, P>0.05).
Plasma ammonia was lower in women compared to
men (main effect sex, P<0.05). However, no interaction
was observed between sexes (P>0.05). When the results
were analysed for the whole group, ammonia increased
compared to resting values in all trials, but no difference
was observed between trials with the same interval
duration. Ammonia concentration was higher in the A45
compared to P120 trial (Table 1). Plasma ammonia was
strongly correlated to blood lactate concentration after
the 8·25 m (r=0.71–0.86, P<0.01).
Plasma glycerol increased compared to resting values,
but no differences were observed between trials
(Table 1). No alterations in the results were observed in
trials for ammonia or glycerol concentration when the
values were corrected for plasma volume changes.
Haematocrit and haemoglobin increased after the last
25-m sprint; however, no differences were observed
between trials (P>0.05, Table 1). Plasma volume
changes were not statistically different between trials
(P>0.05, Table 1).
Discussion
The present study examined the effects of tw o different
types of recovery on repeated sprint swimming perfor-
mance. Active or passive recovery at 45 and 120-s
intervals was applied during the period separating 25-m
sprints, and the findings clearly indicate that active
recovery increases fatigue following both recovery
intervals. It is evident that when swimmers rested pas-
sively between sprints, velocity was better maintained
irrespective of the interval duration. Furthermore,
velocity during the subsequent 50-m sprint was not
affected by the type of recovery (active or passive), but
was impaired to a greater extent when the interval
during the 8·25 m sprints was shorter (45 s vs 120 s).
Performance during the 25-m sprints
The duration of the exercise as well as the rest inte rval
separating successive sprints are crucial factors for per-
formance maintenance (Holmyard et al. 1987; Balsom
et al. 1992a, b). The duration of each 25-m sprint in the
present study ranged from 13 to 17 s. During a sprint
that lasts 10–20 s, muscle lactate and hydrogen ion (H
+
)
concentrations reach high levels (Bogdanis et al. 1998).
The increased concentration of H
+
may be respon sible
for the decreased performance during subsequent
sprints, since it may impair the function of glycolytic
enzymes (Spriet et al. 1989 ). Additionally, the increased
accumulation of inorganic phosphate (P
i
) may depress
the muscular function by several mechanisms (Wester-
blad and Allen 2003). This may explain the decrement in
swimming velocity observed in our second sprint during
the active recovery trials. However, the fact that a drop
Fig. 2 Blood lactate concentration during each trial.
a
P<0.05
between A120 versus P120,
b
P<0.05 between P45 versus A45,
c
P<0.05 with the previous sampling point in the A45 and A120
trials,
d
P<0.05 from the previous sampling point in all trials
[n=16, mean (SE)]
697

in velocity from sprint 1 to sprint 2 was not seen when
passive recovery was applied shows that muscle acidosis
and increased accumulation of P
i
are not the only factors
affecting performance during the A120 and A45 trials.
During repeated short-duration sprints, the rate of
PCr resynthesis is crucial for performance maintenance,
(Gaitanos et al. 1993), although the availability of this
substrate rapidly decreases within 10–20 s of dynamic
exercise (Bogdanis et al. 1996a, 1998). Several metabolic
alterations within the muscle cell, such as pH and oxy-
gen availability, may affect the resynthesis of PCr
(Sahlin et al. 1979; Haseler et al. 1999). Any changes in
these factors caused by the type of recovery after
sprint 1 may be reflected in the performance of sprint 2.
The intensity of active recovery in the present study
corresponded to 60% of the maximum 100-m velocity.
This intensity was selected because it facilitates blood
lactate removal during swimm ing (Cazorla et al. 1983)
and prevents lactate accumulation, as shown in a pre-
liminary trial in the present study (see Methods).
Unpublished observations from our laboratory have
shown that this velocity corresponds to an intensity of
about 40–45% of the maximum oxygen uptake
(
_
V O
2max
). When swimming slowly, swimmers rely
mainly on arm work, while the limited leg action is used
to keep body balance. Additionally, slow arm move-
ments increase the time between breaths, since swimmers
normally breathe at every stroke cycle. This allows
swimmers to breathe every 2–3 s, while inspiration time
must be short to keep up with stroke coordination.
Under these conditions, the volume of inspired oxygen
may be reduced, causing a reduction in arterial oxygen
pressure and arterial hypoxaemia (Yamamoto et al.
1987). The fact that the participants in the present study
were untrained may have further challenged the
breathing pattern.
Even further, Dupont et al. (2004) used near-infrared
spectroscopy in cycling, and demonstrated a slower
reoxygenation of oxyhaemoglobin during active recov-
ery, corresponding to an intensity of 40% of maximum
oxygen uptake. Oxyhaemoglobin saturation may reflect
the balance of oxygen supply and utilization. It is also
possible that the energy demand increases during active
recovery even at a slow swimming pace. Increased oxy-
gen uptake during active recovery has been reported for
cycling (Bogdanis et al. 1996b), and for high-intensity
intermittent running (Dupont et al. 2003). It is likely
then that there is an imbalance between the oxygen
required and the oxygen available. In this case the
reduced oxygen availabili ty may increase the time
constant for PCr resynthesis (Haseler et al. 1999), and
decrease performance during repeated short-duration
sprints (Balso m et al. 1994). Applying a different
swimming velocity during the active recovery interval
period could have led to a different metabolic demand
and could possibly have affected performance.
Decreased performance on the following sprints
(3–8) in the P45 trial may be attributed to combined
effects of reduced PCr resynthesis rate and decreased
rate of anaerobic glycolysi s caused by H
+
accumula-
tion (Hermansen 1981; Spriet et al. 1989; Hargreaves
et al. 1998). It is likely that the increased interval
period of the P120 trial enabled adequate PCr resyn-
thesis and availability during subsequent sprinting. A
decreased rate of PCr resynthesis (with potential
mechanisms described earlier) may have further
reduced sprinting ability during the active recovery
periods of the A120 and A45 trials.
Improved performance after active recovery was
reported during sprint cycling (Signorile et al. 1993;
Bogdanis et al. 1996b; Ahmaidi et al. 1996). Female
swimmers swam faster on a 200-yard (182.88-m) trial
after a long duration active recovery study (Felix et al.
1997). In contrast, the time to exhaustion was decreased
when short-duration running or cycling bou ts at an
intensity of 120%
_
V O
2max
were separated with active
recovery (Dupont et al. 2003, 2004). In addition, data
from our laborat ory confirmed a decrease in perfor-
mance after active recovery (Toubekis and Tokmakidis
2003). The exercise intensity and duration of various
experimental protocols may explain the different results
found in the present study and previous studies in
cycling or swimming (Signorile et al. 1993; Bogdanis
et al. 1996b; Ahma idi et al. 1996; Felix et al. 1997;
Toubekis and Tokmakidis 2003).
Plasma ammon ia levels were not different during
trials of the same interval duration, despite a tendency
for increased values after active recovery. Higher
ammonia values after active recovery have been reported
by Bogdanis et al. (1996b). Indeed, the increased plasma
ammonia after active recovery in the present study may
be attributed to increased efflux from the muscle because
Table 1 Haematocrit, haemoglobin, plasma volume changes (PVC), plasma ammonia and glycerol, at rest and after 8·25-m sprints in the
four trials. Values are means (SE) for 16 subjects. Ammonia values, n=13
Trial Haematocrit
(%)
Haemoglobin
(g dl
1
)
PVC
(%)
Plasma ammonia
(lmol l
1
)
Plasma glycerol
(mmol l
1
)
Rest Post 8·25 m Rest Post 8·25 m Rest Post 8·25 m Rest Post 8·25 m
P120 45.0 (0.8) 48.5 (0.9) 14.0 (0.5) 14.7 (0.5) 11.1 (0.9) 24.4 (2.9) 81.9 (7.5) 0.068 (0.013) 0.157 (0.028)
A120 45.0 (1.0) 48.3 (1.1) 13.7 (0.5) 14.6 (0.5) 12.1 (1.2) 20.8 (2.0) 98.5 (11.9) 0.074 (0.011) 0.133 (0.016)
P45 45.1 (0.8) 49.0 (0.9) 13.9 (0.5) 15.0 (0.5) 13.9 (1.0) 21.9 (2.4) 103.7 (15.8) 0.080 (0.013) 0.148 (0.023)
A45 45.5 (0.7) 49.1 (0.8) 13.9 (0.5) 14.8 (0.5) 11.7 (1.1) 23.8 (2.4) 115.7 (17.9)* 0.067 (0.010) 0.121 (0.015)
*P<0.05 compared to P120 trial; all post values compared to rest P<0.05
698

of increased muscle blood flow. In addition, a decreased
ammonia concentration would support an assumption
for less energy deficit in the passive recovery trials rather
than in the active.
Performance of the 50-m sprint
Improvement of performance was observed for the
50-m sprint after the long- (P120, A120) compared to
short-interval (P45, A45) trials. To our knowledge, no
study has reported similar findings. For the same rest
interval after the last 25-m sprint (6 min in all trials),
velocity decreased when the previously performed
sprints were separated by short-duration intervals.
This is of practical importance for coaches when
aiming to design repeated sets of short-duration bouts.
Blood lactate concentration after all trials, irre-
spective of active or passive recovery, was similar at
the end of 8·25 m, and decreased after 6 min only
during the A120 and A45 trials. Despite these de-
creased values in the active recovery trials, blood
lactate concentration at the start of the 50-m sprint
test does not correlate with performance. Blood lactate
values may not represent the intramuscular lactate
content, and the interval time available for the re-
moval of this metabolite is crucial. If we co nsider the
total duration of the resting interval periods between
sprints, the time available for lactate removal was
about 9 min longer in the A120 and P120 (recovery
time 14 min) compared to the A45 and P45 trials
(recovery time 5.15 min). This would have affected the
amount of lactate removed from the muscle. It is
likely that higher muscle lac tate content was present in
the P45 and A45 compared to P120 and A120 trials.
The idea of lactate as a fatigue agent is outdated
(Gladden 2004), and several other hypotheses, mainly
based in vitro studies, have emerged (Westerblad and
Allen 2003). However, the contribution of H
+
ions to
muscular fatigue during whole-body sprint exercise
cannot be dismissed (Hermansen 1981; Spriet et al 1989).
There is evidence that muscle pH is further reduced even
after a dramatic reduction in the rate of anaerobic gly-
colysis during repeated sprints (Spriet et al. 1989;
Bogdanis et al. 1996a). Given the different recovery
duration between the 120 s and 45 s trials, it is likely
that the intramuscular acidity varied across trials after
the 8·25 m in our study. Thus, this effect on perfor-
mance in the 50-m sprint test should be considered, since
it may impair the glycolytic rate in a subsequent sprint
(Spriet et al. 1989 ).
In order to explain any differences in the 50-m sprint
performance, we should also take into accou nt the PCr
levels at the start of this bout, 6 min after the completion
of the 8·25 m. This becomes an important parameter
since PCr resynthesis cannot be completed within 6 min
after maximum exercise (Bogdanis et al. 1995 ), and its
availability is crucial for performance maintenance
during a sprint (Bogdanis et al. 1996a). Under these
conditions, PCr may have been recovered at higher
levels when a longer resting interval was applied. This
would have allowed a higher intramuscular content at
the end of 8·25 m in the P120 and A120 trials. Dawson
et al. (1997 ) have shown that PCr restorat ion may take
longer if this process starts at lower levels. Additionally,
any differences in the intramuscular pH may be con-
nected to the PCr resynthesis rate during the later stages
of recovery (McMahon and Jenkins 2002). Therefore, it
appears that any changes in the PCr content will affect
velocity in a 30–34 s sprint as in the present study.
In conclusion, active recovery should not be applied
during the interval period (lasting from 45 to 120 s)
separating short-duration sprints in swimming. How-
ever, when the interval period is longer (6 min), the type
of recovery does not affect performance. The interval
period between sprints may determine the length of time
before a new set of maximum exercise intensity should
be performed. The causes of reduced velocity after active
recovery could be better explored with specific histo-
chemical and biochemical parameters.
References
Ahmaidi S, Granier P, Taoutaou Z, Mercier J, Dubouchaud H,
Prefaut C (1996) Effects of active recovery on plasma lactate
and anaerobic power following repeated intensive exercise. Med
Sci Sports Exerc 28(4):450–456
Balsom P, Seger J, Sjodin B, Ekblom B (1992a) Physiological
responses to maximal intensity intermittent exercise. Eur J Appl
Physiol 65:144–149
Balsom P, Seger J, Sjodin B, Ekblom B(1992b) Maximal-intensity
intermittent exercise: effect of recovery duration. Int J Sports
Med 13(7):528–533
Balsom P, Gaitanos G, Ekblom B, Sjodin B(1994) Reduced oxygen
availability during high intensity intermittent exercise impairs
performance. Acta Physiol Scand 152:279–285
Bangsbo J, Graham T, Johansen L, Saltin B(1994) Muscle lactate
metabolism in recovery from intense exhaustive exercise: impact
of light exercise. J Appl Physiol 77(4):1890–1895
Bogdanis GC, Nevill M, Boobis L, Lakomy H, Nevill A (1995)
Recovery of power output and muscle metabolites following
30 s of maximal sprint cycling in man. J Physiol (Lond)
482(2):467–480
Bogdanis GC, Nevill ME, Boobis LH, Lakomy HKA (1996a)
Contribution of phosphocreatine and aerobic metabolism to
energy supply during repeated sprint exercise. J Appl Physiol
80(3):876–884
Bogdanis GC, Nevill ME, Lakomy H, Graham C, Louis G(1996b)
Effects of active recovery on power output during repeated
maximal sprint cycling. Eur J Appl Physiol 74:461–469
Bogdanis GC, Nevill ME, Lakomy HKA, Boobis LH (1998) Power
output and muscle metabolism during and following recovery
from 10 and 20 s of maximal sprint exercise in humans. Acta
Physiol Scand 163:261–272
Cazorla G, Dufort C, Cervetti J (1983) The influence of active
recovery on blood lactate disappearance after supramaximal
swimming. In: Hollander P, Huijing P, de Groot G (eds) Bio-
mechanics and medicine in swimming. International series on
sport sciences, vol 14. Human Kinetics, Champaign, Ill., pp
244–250
Dawson B, Goodman C, Lawrence S, Preen D, Polglaze T, Fitzi-
mons M, Fournier P (1997) Muscle phosphocreatine repletion
following single and repeated short sprint efforts. Scand J Med
Sci Sports 7(4):206–213
699

Dill D, Costill D (1974) Calculation of percentage changes in
volumes of blood, plasma, and red cells in dehydration. J Appl
Physiol 37(2):247–248
Dupont G, Blondel N, Berthoin S (2003) Performance for short
intermittent runs: active vs passive recovery. Eur J Appl Physiol
89:548–554
Dupont G, Moalla W, Guinhouya C, Ahmaidi A, Berthoin S
(2004) Passive versus active recovery during high-intensity
intermittent exercises. Med Sci Sports Exerc 36:302–308
Felix S, Manow T, Jarviw A, Jensen B, Headley S (1997) Swim-
ming performance following different recovery protocols in fe-
male collegiate swimmers. J Swim Res 12:1–6
Gaitanos G, Williams C, Boobis L, Brooks S (1993) Human muscle
metabolism during intermittent maximal exercise. J Appl
Physiol 75(2):712–719
Gladden LB (2004) Lactate metabolism: a new paradigm for the
third millenium. J Physiol (Lond) 558:5–30
Hargreaves M, McKenna M, Jenkins D, Warmington S, Li J, Snow
R, Febbraio M (1998) Muscle metabolites and performance
during high-intensity intermittent exercise. J Appl Physiol
84:1687–1691
Haseler L, Hogan M, Richardson R (1999) Skeletal muscle phos-
phocreatine recovery in exercise-trained humans is dependent
on O
2
availability. J Appl Physiol 86(6):2013–2018
Hermansen L (1981) Effect of metabolic changes on force genera-
tion in skeletal muscle during maximal exercise. Ciba Foun-
dation Symposium, Human muscle fatigue: physiological
mechanisms. Pitman Medical, London, pp 75–82
Holmyard D, Cheetham M, Lakomy H, Williams C (1987) Effect
of recovery duration on performance during multiple treadmill
sprints. In: Reilly T, Lees A, Davids K, Murphy W (eds) Sci-
ence and football. Spon, New York, pp 134–142
Maglischo E (2003) Swimming fastest. Human Kinetics, Cham-
paign, Ill.
McMahon S, Jenkins D (2002) Factors affecting the rate of phos-
phocreatine resynthesis following intense exercise. Sports Med
32(12):761–784
McMurray R (1969) Effects of body position and immersion on
recovery after swimming exercise. Res Q 40(4):738–742
Peyrebrune M, Nevill M, Donaldson F, Cosford D (1998) Effects
of oral creatine supplementation on performance in single and
repeated sprint swimming. J Sports Sci 16:271–279
Sahlin K, Harris R, Hultman E (1979) Resynthesis of creatine
phosphate in human muscle after exercise in relation to intra-
muscular pH and availability of oxygen. Scand J Clin Lab
Invest 39:551–558
Signorile J F, Ingalls C, Tremblay L (1993) The effects of active and
passive recovery on short-term high intensity power output.
Can J Appl Physiol 18(1):31–42
Spriet L, Lindiger M, McKelvie R, Heigenhauser G, Jones N
(1989) Muscle glycogenolysis and H
+
concentration during
maximal intermittent cycling. J Appl Physiol 66(1):8–13
Toubekis A, Tokmakidis S (2003) Active recovery decreases per-
formance during repeated bouts of sprint swimming irrespective
of resting interval duration. In: Chatard JC (ed) Biomechanics
and medicine in swimming IX. Publications de l’ Universite
´
de
Saint-E
´
tienne, Saint-E
´
tienne, France, pp 469–474
Weltman A, Stamford BA, Moffatt RJ, Katch LV (1977) Exercise
recovery, lactate removal, and subsequent high intensity exer-
cise performance. Res Q 48(4):787–796
Weltman A, Stamford BA, Fulco C (1979) Recovery from maximal
exercise: lactate disappearance and subsequent performance.
J Appl Physiol 47(4):677–682
Westerblad H, Allen D (2003) Cellular mechanisms of skeletal
muscle fatigue. Adv Exp Med Biol 538:563–570
Yamamoto Y, Mutoh Y, Kobayashi H, Miyashita M (1987) Effects
of reduced frequency breathing on arterial hypoxemia during
exercise. EurJ Appl Physiol 56:522–527
700