Tapering for Marathon and Cardiac Autonomic Function

Abstract

The purpose of this study was to investigate changes in post-exercise heart rate recovery (HRR) and heart rate variability (HRV) during an overload-tapering paradigm in marathon runners and examine their relationship with running performance. 9 male runners followed a training program composed of 3 weeks of overload followed by 3 weeks of tapering (-33±7%). Before and after overload and during tapering they performed an exhaustive running test (Tlim). At the end of this test, HRR variables (e.g. HRR during the first 60 s; HRR60 s) and vagal-related HRV indices (e.g. RMSSD5-10 min) were examined. Tlim did not change during the overload training phase (603±105 vs. 614±132 s; P=0.992), but increased (727±185 s; P=0.035) during the second week of tapering. Compared with overload, RMSSD5-10 min (7.6±3.3 vs. 8.6±2.9 ms; P=0.045) was reduced after the 2nd week of tapering. During tapering, the improvements in Tlim were negatively correlated with the change in HRR60 s (r=-0.84; P=0.005) but not RMSSD5-10 min (r=-0.21; P=0.59). A slower HRR during marathon tapering may be indicative of improved performance. In contrast, the monitoring of changes in HRV as measured in the present study (i.e. after exercise on a single day), may have little or no additive value.

Full text

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
accepted after revision
October 21 , 2013
Bibliography
DOI http://dx.doi.org/
10.1055/s-0033-1361184
Published online: 2014
Int J Sports Med
© Georg Thieme
Verlag KG Stuttgart · New York
ISSN 0172-4622
Correspondence
Bernhard Hug
Swiss Federal Institute
of Sport
Section for Elite Sport
Magglingen
Heiligenschwendi 3625
Switzerland
Tel.: + 41/76/377 68 98
Fax: + 41/21/692 32 93
bernhardhug@bluewin.ch
Key word
●
▶
endurance performance
●
▶
cardiac autonomic activity
●
▶
heart rate recovery
●
▶
heart rate variability
●
▶
parasympathetic
reactivation
Tapering for Marathon and Cardiac Autonomic
Function
Searching for a minimally invasive and minimally
disturbing method to optimize preparation for
peak performance has always been a matter of
interest in exercise physiology and sports medi-
cine [ 7 ] . The autonomic nervous system (ANS)
function plays an important role in the training
responses and in the functional adaptations
occurring from a given training stimulus [ 7 , 37 ] .
The mono-exponential decrease in heart rate
after maximal exercise is primarily modulated by
the ANS, and short-term post-exercise heart rate
recovery (HRR) can therefore be used as a marker
of cardiac parasympathetic outfl ow [ 29 ] . Heart
rate variability (HRV) measurements are also
well-accepted procedures for the assessment of
the cardiac vagal function [ 12 , 16 ] . In this con-
text, HRV monitoring has been proposed as a
valuable tool for detecting the complex changes
in ANS activity in athletes [ 10 , 11 , 37 , 48 ] .
Consequently, either post-exercise HRR, resting-
HRV, or post-exercise HRV have been suggested
as indirect markers of cardiac autonomic control
and may off er practical and simple ways of quan-
tifying the physiological eff ects of training
[ 8 , 16 , 17 , 24 , 32 ] . The ability of these indices to
predict endurance performance in the field and
Introduction
▼
Reduced training load (tapering) during the
preparation for an important competition aims
to minimize fatigue without restricting the posi-
tive training eff ects. The overload-tapering para-
digm is characterized by an initial increase in
training load for a time period of several weeks,
followed by a reduction of training load during a
time period of 1–4 weeks [ 39 , 40 , 54 ] . Taper-
induced performance increase is generally
greater when the taper phase is preceded by an
overload period with increased training load (up
to 50 %). Maintaining high training intensities as
well as a high frequency ( > 80 % of normal) dur-
ing the taper phase seems to be important
[ 9 , 39 , 40 , 54 ] . Training intensity seems to be the
key factor for optimized performance prior to a
main competition [ 54 ] . Race preparation accord-
ing to current tapering recommendations can
lead to performance gains from 2 to 9 % [ 9 , 54 ] . To
date, except for competition results, there is no
practical and reliable method for measuring the
eff ect of tapering on the athletes’ training status,
fatigue and performance.
Authors B. Hug
1 , L. Heyer
1 , N. Naef
1 , M. Buchheit
2 , J. P. Wehrlin
1 , G. P. Millet
3
Affi liations
1 Swiss Federal Institute of Sport, Section for Elite Sport, Magglingen, Switzerland
2 Myorobie Association, Sport Science Unit, Montvalezan, France
3 ISSUL Institute of Sport Sciences, Department of Physiology, University of Lausanne , Switzerland
Abstract
▼
The purpose of this study was to investigate
changes in post-exercise heart rate recovery
(HRR) and heart rate variability (HRV) during an
overload-tapering paradigm in marathon run-
ners and examine their relationship with run-
ning performance. 9 male runners followed a
training program composed of 3 weeks of over-
load followed by 3 weeks of tapering ( − 33 ± 7 %).
Before and after overload and during tapering
they performed an exhaustive running test (T
lim ).
At the end of this test, HRR variables (e. g. HRR
during the fi rst 60 s; HRR
60 s ) and vagal-related
HRV indices (e. g. RMSSD
5–10 min ) were examined.
T
lim did not change during the overload training
phase (603 ± 105 vs. 614 ± 132 s; P = 0.992), but
increased (727 ± 185 s; P = 0.035) during the sec-
ond week of tapering. Compared with overload,
RMSSD
5–10 min (7.6 ± 3.3 vs. 8.6 ± 2.9 ms; P = 0.045)
was reduced after the 2
nd week of tapering. Dur-
ing tapering, the improvements in T
lim were
negatively correlated with the change in HRR
60 s
(r = − 0.84; P = 0.005) but not RMSSD
5–10 min
(r = − 0.21; P = 0.59). A slower HRR during mara-
thon tapering may be indicative of improved
performance. In contrast, the monitoring of
changes in HRV as measured in the present study
(i. e. after exercise on a single day), may have lit-
tle or no additive value.
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
to monitor positive or negative adaptations could help in the
design of individualized HRR- or HRV-guided training programs
for athletes [ 10 , 31 , 44 ] .
However, the relationship between training load, fatigue, per-
formance and changes in HR-derived indices [ 25 , 26 , 35 , 38 , 42 , 55 –
57 ] has led to confl icting results. Hedelin et al. [ 26 ] reported, in
an overtrained cross-country skier, reduced competition per-
formance and decreased profi le of mood states, along with an
increased cardiac parasympathetic modulation. In contrast, total
power spectral density of HRV was decreased in 5 overtrained
female endurance athletes undergoing heavy training over a 6 to
9-week period [ 57 ] . In another study, Hedelin et al. [ 25 ] reported
no signifi cant change in HRV in 9 overexerted canoeists after
increasing training load by 50 % over a 6-day training camp.
These discrepancies between results are probably due to the dif-
ferences in methodology [ 45 ] or protocols, such as the nature of
the overload period, and/or the number and performance level
of the athletes involved in the diff erent studies.
At rest or while exercising at moderate intensity, aerobically-
trained athletes have a greater cardiac parasympathetic activity
compared to untrained subjects [ 18 , 50 , 51 ] . In moderately-
trained athletes, Buchheit et al. [ 15 ] reported a positive relation-
ship between changes in performance and parasympathetic
reactivation following an 9-week training program. In contrast,
well-trained elite athletes have been reported to exhibit
decreased vagal-related HRV indices following a large-volume
training program [ 28 ] . Iwasaki et al. [ 30 ] and Manzi et al. [ 37 ]
have reported an inverted U-shaped relationship between train-
ing load and vagal-related HRV indices in marathon runners.
Pichot et al. [ 42 ] showed that HRV may follow changes in train-
ing load in untrained athletes, with reduced training load being
associated with increased vagal-related HRV indices.
Few authors have investigated the relationship between reduced
training load and changes in cardiac parasympathetic activity.
Atlaoui et al. [ 2 ] found during a tapering regime in highly-
trained swimmers a positive correlation between resting cardiac
parasympathetic activity and performance. Le Meur et al. [ 34 ]
divided a group of trained male triathletes into a normal train-
ing and an intensifi ed training group who performed 3 weeks of
overload followed by 1 week of taper. The overreached triath-
letes showed an increase in cardiac parasympathetic activity,
whereas performance in an incremental performance test had
decreased. However, these responses were reversed during the
taper. Despite its potentially high practical interest, the rele-
vance of HR-derived indices for monitoring changes in fatigue
and/or performance in athletes during an eff ective training pro-
gram is still unclear. To improve our understanding of the useful-
ness of these non-invasive markers for monitoring training
adaptations, we investigated changes in running performance
and parasympathetic reactivation following maximal exercise in
response to an overload-tapering paradigm in well-trained mar-
athon runners. Therefore, the aims of the present study were to
(1) examine the respective changes in HRR and HRV indices dur-
ing a 3-week overload followed by a 3-week tapering period in
well-trained runners and (2) assess the possible relationships
between these indices and running performance (time to
exhaustion). Based on previous research, it was hypothesized
that, during the preparation for a marathon, post-exercise para-
sympathetic reactivation would be slower after a 3-week over-
load period but increased during a subsequent 3-week taper
period [ 2 , 42 ] .
Methods
▼
Subjects
11 well-trained marathon runners satisfi ed the inclusion criteria
(male, personal best marathon performance under 3 h, no inju-
ries in the last 3 months, absence of clinical signs or symptoms
of infection, absence of cardiovascular diseases or injuries and a
minimum weekly training dose of 5 running sessions) and gave
written informed consent to the study, which was approved by
the internal review board of the Swiss Federal Institute of Sport
and was performed in accordance with the ethical standards of
the IJSM [ 23 ] . Throughout the normal training phase, the ath-
letes had 6–7 running sessions per week (training volume of
8.2 ± 1.4 h). Each runner had a history of at least 5 years of train-
ing for running competitions. During the study period, one run-
ner dropped out due to injury and another one due to illness.
Therefore, 9 subjects (34.6 ± 5.7 years; 180 ± 9 cm; 69.0 ± 6.3 kg)
completed all measurements.
Experimental design
An outline of the study design is shown in
●
▶
Fig. 1 . Athletes
completed a 10-week study period before participating at the
Lucerne Marathon in Switzerland. They started with a 4-week
training phase with record of their “usual” training regimen
(normal training). In the subsequent overload phase, training
load was increased for 3 weeks by 23 ± 10 %. During the overload
phase, each runner had one additional 1-h high intensity run-
ning session per week and the long run was prolonged by 30 min.
During the tapering phase, the number of high intensity training
sessions was kept similar to that of the previous (overload)
phase, while the training volume was reduced gradually (1 ses-
sion less per week and reduction in the average duration of the
other sessions). The mean reduction in volume between the
overload and tapering phase was 33 ± 7 %. During the 10-week
study period, athletes kept a training log and a history of health
status and nutrient intake. The athletes had to record session-
RPE within 30-min of fi nishing their workout [ 20 ] . A series of
diff erent tests was completed at diff erent time points
(T1-T5,
●
▶
Fig. 1 ). At T1, athletes performed a submaximal run-
ning test and a maximal oxygen uptake (V
˙ O
2peak ) test. These tests
were used to calculate the running speeds for the subsequent
performance tests. At T2, T3, T4 and T5, the athletes performed a
time to exhaustion test (T
lim ) to examine performance changes
during the study period and post-exercise HRR and parasympa-
thetic reactivation was analysed for 10 min.
Normal training
T1
12
Time to exhaustion test
Time to exhaustion test
Time to exhaustion test
Time to exhaustion test
Marathon
345678910
[weeks]
T2 T3 T4 T5
Overload Tapering
Fig. 1 Study design. Testing at time point T1 was composed of a
“submaximal running test” and a V
˙O
2peak test. At T2, T3, T4 and T5, the
athletes performed an identical “Time to exhaustion test” as well as heart
rate recovery (HRR) and heart rate variability (HRV) assessment.
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
The runners were advised to maintain the same preparation
procedure (e. g. habitual nutrition plan; warm-up; hydration)
before each testing day. Preceding all tests, the runners abstained
from alcohol and caff einated beverages and refrained from
medium intensity (48 h) and heavy training (24 h) prior to all
testing days. The athletes completed their test at the same hour
( ± 1 h) between 09:00 and 16:00 h to avoid possible circadian
infl uences on the parameters. Room temperature (18–19 °C) and
humidity (38–40 %) were kept constant for all tests.
Training intervention
As recommended [ 9 , 39 , 40 , 54 ] during tapering, the reduction in
load was mainly due to the decrease in volume partially coun-
terbalanced by the increase in intensity. Intensity of training and
the number of workouts per week were kept nearly at the same
level as during the overload phase to ensure positive eff ects of
the taper and a possible increase in performance [ 40 , 49 , 54 ] . The
training log was sent weekly with an Excel-sheet by e-mail to
the main investigator for analysis. Feedback with light modifi ca-
tions to the training program was then provided by the investi-
gators. Training loads from week 10–7 prior to the marathon
were averaged to one mean value “normal training”. Training
loads from week 6–4 prior to the marathon were averaged to
one mean value “overload”. During the taper phase, summarized
loads of the fi rst week were recorded as “TP1” value and simi-
larly as “TP2” for the 2
nd week. The last tapering week was
recorded without the load corresponding to the marathon race
(i. e., 6 days adjusted to 7 by linear extrapolation).
Submaximal running test (T1)
The subjects started with a general standardized warm-up for
5 min at 8 km · h − 1 . Next, a blood sample was drawn from the
earlobe and analysed for pre-testing lactate ([La]). The runners
then continued to run at 9 km · h
− 1 for 5 min followed by resting
for 30 s. This procedure was repeated with 11 km · h
− 1 , 13 km · h − 1
and 15 km · h − 1 . During the 30-s rest, a blood sample was taken
from the earlobe for [La] measurement, and RPE was indicated
by the subject using a scale from 6 to 20 [ 5 ] . [La] was analysed
with a standard analyser (Hitado Super GL; Dr. Müller Gerätebau
GmbH, Freital, Germany) using 10-μl open-end capillaries. Res-
piratory gas exchanges were measured breath-by-breath during
the entire submaximal running test (Jaeger Oxycon Pro; Jaeger,
Hoechberg, Germany). Average O
2 consumption of the last 3 min
of each running stage was used for the calculation of the indi-
vidual relationship between oxygen uptake and running speed,
as usually performed in our laboratory [ 58 ] .
The analyser was calibrated before each use with 2 samples of a
known concentration. Calibration procedures were performed
before each test, according to the manufacturer’s recommenda-
tions and also for the location of the laboratory at 950 m above
sea level. The respiratory analysis system was calibrated fi rst
using a gas of known O
2 and CO
2 concentrations and then using
ambient air, with partial O
2 composition being assumed to be
20.9 %. Calibration of the turbine fl ow-meter of the Oxycon Pro
was performed with an automated program.
V
˙O
2peak -Test (T1)
15 min after cessation of the submaximal running test, the sub-
jects performed a V
˙
O 2peak test on a treadmill. Starting at 7 km · h
− 1
running velocity was increased by 0.5 km · h
− 1 every 30 s until
voluntary exhaustion [ 19 ] . During the tests, gas exchange data
were collected continuously and recorded as means for every
30 s. The following variables were measured and analysed: oxy-
gen uptake (V
˙ O
2 ), respiratory exchange ratio (RER), ventilation
(V
˙ E), breathing frequency (BF), maximal lactate ([La]
max ) and
maximal velocity. V
˙ O
2peak was defi ned as the highest mean V
˙ O
2
value obtained for any continuous period of 30 s. Maximal HR
(HR
max ) (Suunto dual belt; Helsinki, Finland) was defi ned as the
highest value during the test.
Time to exhaustion test at 95 % vV
˙O
2peak (T2
- T5)
In this test, the athletes ran at an individual running speed
corresponding to 95 % of the velocity associated with V
˙ O
2peak
(vV
˙ O
2peak ) until exhaustion. vV
˙
O 2peak was calculated by the indi-
vidual relationship of oxygen uptake and running speed by
extrapolation of the 4 values recorded at submaximal speeds
(9, 11, 13 and 15 km · h
− 1 ) and the V
˙ O
2peak recorded during the
incremental test as described previously [ 58 ] . Of interest is that
this protocol is the one recommended by Swiss Olympics and
used in most Swiss elite athletes. The testing protocol was
planned with a running time that should not exceed 20 min for
not interfering with the training program. As T
lim at vV
˙
O
2peak is
related to the anaerobic threshold [ 3 ] and endurance time plot-
ted against velocity exhibits a hyperbolic shape, it can be pre-
dicted that running time to exhaustion at 95 % vV
˙ O
2peak should
be around 10 ± 5 min in well-trained runners. The subjects
started running at 60 % vV
˙
O 2peak and ran continuously for 10 min
on the treadmill followed by 30 s of rest (for blood sampling and
RPE collection). They then continued running for eight min at 80 %
vV
˙ O
2peak . After 5 min of rest, the subjects then performed a maxi-
mal running test to exhaustion at 95 % vV
˙ O
2peak . V
˙ O
2peak was
defi ned as the highest 30 s-mean V
˙ O
2 value obtained at exhaus-
tion. Running energy cost (EC, ml · min
− 1. ∙ km − 1 ) was calculated
from the V
˙ O
2 of the last 5 min during the stage of 60 % vV
˙
O
2peak . In
all sessions, the subjects were encouraged to perform to their best
eff ort and required to immediately sit for 10 min at T
lim test cessa-
tion. The time duration between the end of exercise and sitting
was less than 5 s.
Resting HR and HRV
The runners sat down and rested for 10 min in a quiet environ-
ment in the laboratory prior to the T
lim test. For both HR and HRV
measurements, the fi rst 2 min of recording were disregarded
due to lack of stability, and the last (stable) eight minutes of the
resting phase were analysed. While this 8-min period for HRV
analysis is slightly longer than that usually employed in the lit-
erature (i. e., 5-min), the recording length is unlikely a major
issue when dealing exclusively with time-domain HRV indices
as in the present study [ 22 ] . Additionally, if we consider the pos-
sible small fl uctuations in ANS activity over time, a longer
recording period may refl ect a more accurate representation of
the actual cardiac autonomic activity. The Suunto T6 watch has
recently shown good validity compared to a mobile ECG-system
[ 59 ] . Standard custom software from the company was used to
generate average values for HR and RR-intervals, which were
exported into an Excel fi le. HRV was analysed with Kubios HRV
(Version 2.0, University of Kuopio, Finland).
Post-exercise HRR and parasympathetic reactivation
RR intervals were recorded, and HRR was assessed during the
recovery period after cessation of the T
lim test and analysed as
follows: (1) the fi rst 30 s (from the 10
th to the 40
th s) of HRR via
semi-logarithmic regression analysis (T30) as proposed by Imai
et al. [ 29 ] ; (2) the time constant of the HR decay (HRRτ) by fitting
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
the 10-min post-exercise HR recovery into a first-order expo-
nential decay curve; and (3) the absolute diff erence between the
final 5-s averaged HR at test completion and the HR recorded at
60 s and 600 s during recovery (HRR
60 s , HRR 600 s ). Post-exercise
HRV was assessed as described previously [ 21 ] . Calculated val-
ues were (1) the time course of RMSSD on successive
30-s segments (RMSSD
30 s ) (moving window) and (2) the
RMSSD from the 5
th to the 10
th min during seated recovery
(RMSSD
5–10 min ). While we acknowledge that the period of anal-
ysis diff ered in length between resting (i. e., 8 min) and post-
exercise HRV (i. e., 5 min) conditions, this is unlikely to aff ect to
present results since those data were not compared directly.
RMSSD refl ects parasympathetic activity of the cardiac auto-
nomic nervous system [ 53 ] and is therefore frequently used to
monitor changes in ANS response to training [ 46 ] . To standardize
testing conditions (i. e., to avoid breathing perturbations), par-
ticipants were not allowed to speak or drink during the 10-min
recovery period. Respiratory rate was spontaneous for practical-
ity during fi eld-based measurements, and because there is little
diff erence in parasympathetic-related HRV indices during con-
trolled or spontaneous breathing [ 4 ] . Importantly also, RMSSD
has much greater reliability than other spectral indices [ 1 ] , par-
ticularly during ‘free-running’ ambulatory conditions [ 41 ] .
Statistical analysis
Data in the text and in the tables are presented as mean ± SD.
Each variable was tested with one-way repeated measure analy-
sis of variance (RM ANOVA) completed by post-hoc Tukey test to
locate statistical diff erences (SigmaPlot 11.0; Systat Software,
Inc, San Jose, CA). Two-way repeated measures trials [normal
training; overload; TP1; TP2] x time [20 × 30-s windows] ANOVA
was used for the analysis of RMSSD
30 s . Diff erences were consid-
ered statistically signifi cant when P ≤ 0.05. Pearson correlation
coeffi cients were calculated to test for signifi cant associations
between performance, HRR and HRV parameters. Pearson cor-
relations (r, 90 % confi dence limits, CL) between per cent changes
in HRV, HRR, training load, performance and related variables
were calculated for each training phase.
Results
▼
Marathon results
Eight runners successfully completed the marathon at the end of
the third taper week. 6 athletes reached their personal best time.
The average time was 169.5 ± 9.9 min. One athlete was forced to
drop out halfway due to gastrointestinal disturbances.
Training intervention
Training load data are shown in
●
▶
Table 1 . The compliance to
training guidelines was satisfying as shown by the per cent
changes in training load (23 % increase during overload; reduc-
tion compared to the overload period was 12 % (TP1), 22 % (TP2)
and 64 % during the last tapering week).
V
˙O
2peak Test (T1)
The V
˙O
2peak was 60.9 ± 2.8 ml . min − 1 · kg − 1 , vV
˙O
2peak was 17.7 ±
1.0 km · h − 1 , [La] max was 5.7 ± 1.5 mmol · l − 1 and HR
max was
182 ± 13 bpm. Calculated individual speed parameters were
10.7 ± 0.6 km · h − 1 for the 10-min stage at 60 % vV
˙O
2peak ,
14.2 ± 0.8 km · h − 1 for the 8-min stage at 80 % vV
˙O
2peak and
16.9 ± 1.0 km · h − 1 for the fi nal phase at 95 % vV
˙O
2peak .
T
lim test at 95 % vV
˙O
2peak (T2
- T5)
●
▶
Table 1 shows the results of the repeated T
lim test for each
training period (T
lim , HR peak , V
˙ E peak , BF peak , [La] max ). HR peak did
not change during overload (P = 0.995) and at TP1, but an increase
was found at TP2 (P = 0.024). A signifi cant time vs. percent
Normal training Overload Taper 1
st week Taper 2
nd week
training load 1 826 ± 293 2 227 ± 460 ## 1 938 ± 241 ^^ 1 686 ± 227 ^^ *
T lim (s) 603 ± 105 614 ± 132 618 ± 132 727 ± 185 # ^ *
V
˙O
2peak (ml · min − 1 · kg − 1 ) 59.5 ± 2.9 60.3 ± 4.1 58.9 ± 2.8 60.4 ± 2.6
V
˙E
peak (l · min − 1 ) 143.0 ± 18.2 141.8 ± 15.4 137.8 ± 19.0 # 139.4 ± 18.0
BF peak (min − 1 ) 57.6 ± 7.7 57.0 ± 8.5 56.0 ± 9.5 55.6 ± 7.6
[La] peak (mmol · l − 1 ) 6.8 ± 2.3 6.4 ± 2.4 5.8 ± 1.8 # 5.6 ± 1.5 ##
EC 60 % vV
˙O
2peak (ml · min − 1 · km − 1 ) 201.9 ± 11.8 200.4 ± 11.4 195.1 ± 14.9 202.8 ± 12.8 *
HR Rest (bpm) 52.7 ± 9.0 52.9 ± 10.5 50.9 ± 10.1 52.3 ± 7.9
HR peak (bpm) 178.0 ± 13.0 177.7 ± 12.3 176.3 ± 12.5 179.6 ± 11.3 *
HR 60 s (bpm) 119 ± 21.3 118.9 ± 22.6 115.7 ± 21.9 123.4 ± 19.6 *
HR 600 s (bpm) 88.2 ± 10.6 88.2 ± 11.0 88.8 ± 12.4 91.9 ± 12.0
HR 5–10 min (bpm) 89.0 ± 11.3 89.1 ± 11.1 88.1 ± 12.9 93.0 ± 11.7 *
HRR 60 s (bpm) 59.0 ± 12.3 58.8 ± 12.9 60.7 ± 13.6 56.1 ± 11.4 *
HRR 600s (bpm) 89.8 ± 10.9 89.4 ± 9.1 87.6 ± 8.2 87.7 ± 9.7
T30 (s) 126.6 ± 52.5 115.4 ± 49.8 116.9 ± 62.5 126.6 ± 69.1
HRRτ (s) 55.0 ± 16.2 54.7 ± 21.4 53.7 ± 18.2 55.9 ± 17.9
RMSSD Rest (ms) 63.5 ± 24.4 67.7 ± 34.2 66.8 ± 29.2 64.5 ± 29.7
RMSSD 5–10 min (ms) 8.1 ± 2.7 8.6 ± 2.9 8.1 ± 3.7 7.6 ± 3.3 ^
Values are mean ± SD. T
lim : running time to exhaustion (s) at 95 % of the velocity associated with V
˙O
2peak . V
˙E
peak : peak ventilation;
BF
peak = maximum breathing frequency. [La]
peak : peak lactate. EC: energy cost. HR
peak : peak heart rate; HR
60 s : hear t rate 60 s after
exercise. HR
600 s : hear t rate 600 s after exercise. HRR
60 s : number of hear t beats recovered within 60 s after exercise. HRR
600 s : number
of heart beats recovered within 600 s after exercise. T30: semi-logarithmic regression analysis from the 10
th to the 40
th s of heart rate
post exercise. HRRτ: time constant of heart rate recovery. RMSSD: square root of the mean squared diff erences between successive
RR-intervals
# : P < 0.05, # : P < 0.01 for diff erences with normal training
^:P < 0.05, ^^: P < 0.01 for diff erences with overload
*:P < 0.05, **:P < 0.01 for diff erences with taper 1
st week
Table 1 Performance and heart
rate recovery variables with
respect to the diff erent training
phases.
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
change eff ect was observed for T
lim between normal training and
TP2 (P = 0.019), overload and TP2 (P = 0.035), as well as between
TP1 and TP2 (P = 0.044).
Post-exercise HRR and parasympathetic reactivation
All HRR and HRV values are presented in
●
▶
Table 1 . A signifi -
cantly slower HRR was found at TP2 compared with TP1. Aver-
age HR from the 5
th to the 10
th min during exercise recovery was
higher (P = 0.037) in TP2 than in TP1. HRR
60 s (P = 0.017) was sig-
nifi cantly lower at TP2 than TP1. RMSSD
5–10 min was reduced
(P = 0.045) at TP2 compared with the overload measurement.
Parasympathetic reactivation (RMSSD
30 ) for most subjects had a
peak between the fi rst and third min of the recovery phase
(
●
▶
Fig. 2 ). A signifi cant RMSSD
30 measure vs. time interaction was
reported with diff erences located between 90 s – 120 s between
overload and TP2 (
●
▶
Fig. 2 ) .
Changes between variables
In the present study, there was no signifi cant association
between training load and any other recorded variable during
overload and tapering. Of interest is the very large and negative
relationship between the changes in HRR
60s and T
lim from the
end of the overload training to the 2
nd week of tapering (r = − 0.84,
90 % CL ( − 0.50; − 0.96); P = 0.005) (
●
▶
Fig. 3a ). A very large corre-
lation was also observed between the changes in HRRτ and in
T
lim for the period between overload and TP2 (r = 0.69, 90 %
CL(0.17;0.91); P = 0.039) (
●
▶
Fig. 3b ). There was, however, no sig-
nifi cant correlation between RMSSD
5–10min and either training
load (r = − 0.09, P = 0.82) or T lim (r = − 0.21, P = 0.59). There was no
signifi cant relationship when considering the other variables.
Discussion
▼
The main fi ndings of this study were:
1. Indices of parasympathetic reactivation were sensitive to
training load manipulation, as evidenced by the signifi cant
decreases in HRR and RMSSD observed during the tapering
period.
2. Performance was only signifi cantly improved (e. g. increase in
T
lim ) after the second week of taper, and these changes were
very largely correlated with those in HRR. In contrast, changes
in performance did not correlate with changes in post-exer-
cise vagal-related indices.
Our results support the view that in endurance athletes known
to have a high parasympathetic activity [ 18 , 50 , 51 ] , an effi cient
tapering phase is associated with decreased parasympathetic
activity and/or increased sympathetic tone [ 28 , 30 , 37 , 46 ] , as
evidenced here by the slower HRR indices/lower RMSSD values.
These fi ndings suggest that enhanced performance is not neces-
sarily associated with faster HRR [ 32 , 33 ] . In endurance athletes,
a more balanced sympathovagal activity seems to be favourable
to aerobic performance [ 28 , 30 , 37 ] .
Training intervention
In the present study, the 18 % improvement in T
lim during taper-
ing clearly shows that the latter was effi cient. Tapering aims to
reduce fatigue while maintaining fi tness during the last weeks
prior a main competition [ 39 , 40 , 54 ] . In these lines, the improved
recovery of overload-induced fatigue during the second tapering
week could partly explain the signifi cant increase in T
lim at the
end of the tapering phase.
*
overload
taper 1st week
taper 2nd week
normal training
RMSSD (ms)
14
12
10
8
6
4
15
45
75
105
135
165
195
225
255
285
time (s)
315
345
375
405
435
465
495
525
555
585
Fig. 2 Root mean square of successive diff erences of RR-intervals meas-
ured on successive 30-s segments (RMSSD
30s ) during the 10 min recovery
period after running to exhaustion at 95 % of the velocity associated to
V
˙O
2peak . Values of repeated trials (normal training, overload, taper 1
st week,
taper 2
nd week) are plotted without SD for clarity. * P = 0.05 for group vs.
time interaction between overload and taper 2
nd week (90–120 s)
20
a
b
r=–0.84, 90% CL (–0.50;–0.96)
P=0.005
10
0
relative change HRR60s (%)
–10
–20
–30
–20
60
40
20
relative change HRRT (%)
0
–20
–20 0 20
relative change time to exhaustion (%)
40 60 80
020
relative change time to exhaustion (%)
40 60 80
r=0.69, 90% CL (0.17;0.91)
P=0.039
Fig. 3 a Relationship between the relative change in running time to
exhaustion (T
lim ; s) and relative change in heart rate recovery during the
fi rst 60 s (HRR
60s ; bpm) during the fi rst 2 weeks of tapering. b Relation-
ship between the relative change in running time to exhaustion (T
lim ; s)
and relative change in the time constant of the heart rate decay (HRRτ; s)
during the fi rst 2 weeks of tapering.
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
Post-exercise HRR during overload and tapering
Post-exercise HRR is induced by a combination of sympathetic
withdrawal and parasympathetic reactivation [ 47 ] . Central neu-
ral control causes the parasympathetic system to be inhibited in
direct proportion to sympathetic activation [ 43 ] . In the present
study, we did not observe any signifi cant change in HRR during
the overload period, which contrasts with the conclusions by
Daanen et al. [ 17 ] , where HRR was reported to be related to
changes in training load. These results also contrast with the
fi ndings from Borresen and Lambert [ 6 ] , who found a decreased
HRR with an increase in training load. However, the authors
speculated that their subjects had reached a state of short-term
overtraining, which is improbable in the present study given
their running performance. In contrast, present results support
the idea of Lambert et al. [ 32 ] and Lehmann et al. [ 36 ] , where a
faster HRR can be observed with acute fatigue (overload).
Following the second week of taper however, there was a sig-
nifi cant reduction in HRR, which is consistent with the results
reported by Houmard et al. [ 27 ] , who reported a slower HRR
after a 10-day reduction of training load. While additional phys-
iological measures (e. g., drugs, muscle nerve activity) would be
required to draw defi nitive conclusions, this suggests that either
sympathetic activity increased and/or that post-exercise para-
sympathetic reactivation was reduced after 2 weeks of tapering.
Possible explanations for this altered sympathovagal balance
include changes in training intensity distribution related to the
tapering phase: while high volumes of moderate intensity exer-
cise may induce rapid increases in post-exercise parasympa-
thetic activity (within 24 h), high-intensity exercise generally
leads to prolonged reduction in parasympathetic activity (48–
72 h) [ 52 ] . To our knowledge, the present study is the fi rst to
report that this reduction in HRR may be associated with an
improvement in T
lim during tapering in endurance athletes. In
line with this observation, was also the very large correlation
between changes in HRRτ and T
lim during the taper (
●
▶
Fig. 3b ).
In fact, only one subject exhibited both an increased HRR
60 s and
an increase in performance during tapering, whereas all of the
athletes had a decreased HRR
60 s along with an increase in per-
formance (
●
▶
Fig. 3a ). One may then speculate that a more bal-
anced ANS activity (inferred by a slower HRR) is actually
favourable for marathon performance [ 46 ] . Since T lim was longer
after the second week of tapering, it could be argued that the
changes observed in HRR were more related to changes in rela-
tive exercise intensity, exercise duration, and/or energy contri-
bution, rather than ANS activity per se. While we cannot rule
out the possibility that the longer time to exhaustion was associ-
ated with greater sympathetic activity (higher central com-
mand, greater peak HR), which could have, in turn, lowered
parasympathetic reactivation, post-exercise blood lactate was
actually lower after the second week of taper, which would be
expected to be related to faster, not slower HRR. In fact, para-
sympathetic reactivation has been shown to be largely corre-
lated with muscle metaborefl ex activation and associated
systems stress metabolites accumulation in the blood, with the
greater the anaerobic contribution, the slower the post-exercise
parasympathetic reactivation [ 13 ] . It is also worth noting that
exercise intensity (which remained the same throughout the
diff erent testing sessions, i. e., 95 % of vV
˙O
2peak ), instead of exer-
cise duration, is likely the stronger determinant of cardiac para-
sympathetic reactivation [ 52 ] . Importantly, there was no
diff erence in any of the other potential confounding factors
between each testing session (e. g., nutrition status, peak venti-
lation, peak breathing frequency), which increases the confi -
dence in the interpretation of the observed changes. The use of a
standardized submaximal exercise (with respect to both dura-
tion and intensity [ 14 ] ) might nevertheless allow a better exam-
ination of the changes in ANS activity per se, which should be
the focus of further research. We nevertheless believe that, irre-
spective of the underlying mechanisms responsible for this
reduced HRR following tapering, these changes are likely of
interest for practitioners, who may use HRR as an indirect meas-
ure of training adaptation.
Post-exercise parasympathetic modulation during
overload and tapering
It is well known that long-term aerobic training increases para-
sympathetic activity and reduces sympathetic activity at rest
and during submaximal exercise [ 11 , 12 ] . The short-term eff ects
of endurance training likely mirror a dose-response relationship
between training load and HRV components in recreational
marathon runners [ 37 ] . It was found that increased training
loads during a 6-month training period were related to a shift
toward a sympathetic predominance in supine resting position.
In an earlier study with world-class rowers, performance
required adaptations in the neural regulation of the cardiovascu-
lar system that were the opposite of those brought about by
moderate-intensity training [ 28 ] . After the second tapering
week, RMSSD
5–10 min was lower and RMSSD
30 s was reduced
between 90 and 120 s of recovery (
●
▶
Table 1 ,
●
▶
Fig. 2 ). These
results confi rm the data collected in elite rowers [ 46 ] , and sug-
gest a reduction in post-exercise parasympathetic reactivation
at the end of the taper, as already inferred from the HRR results.
Interestingly also, the lack of association between T
lim and HRV
changes support previous observations showing that the respective
link between HRR, HRV, training load and performance diff er [ 12 ] .
Conclusion
▼
The present results show that in well-trained marathon runners,
the signifi cant increase in running performance following taper-
ing was associated with slower HRR. Therefore, fi tness improve-
ments may not always be associated with increased HRR (or
conversely). The present results suggest that the interpretation
of changes in HRR should be made in relation to the specifi c
training phases of an endurance program (e. g. base training;
tapering). One may also recommend using HRR – and for practi-
cal reasons the simplest variable, HRR
60 s – instead of post exer-
cise HRV for monitoring acute or short-term changes in cardiac
autonomic function, and possibly performance capacity during
taper. Further studies on HRR and HRV indices comparing
endurance, power-sprint, glycolytic (anaerobic) or intermittent
athletes during tapering are required to confi rm the present
results.
Acknowledgements
▼
The authors would like to thank the subjects for their participa-
tion and Dr. Adrian Bürgi (SFISM, Switzerland) for providing
excellent support. No confl ict of interest or source of funding.
The authors have no disclosures to declare. There is no confl ict of
interest or source of funding for any of the authors.
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
References
1 Al Haddad H , Laursen P B , Chollet D , Ahmaidi S , Buchheit M . Reliabil-
ity of resting and postexercise heart rate measures . Int J Sports Med
2011 ; 32 : 598 – 605
2 Atlaoui D , Pichot V , Lacoste L , Barale F , Lacour J R , Chatard J C . Heart rate
variability, training variation and performance in elite swimmers . Int
J Sports Med 2007 ; 28 : 394 – 400
3 Billat V , Renoux J C , Pinoteau J , Petit B , Koralsztein J P . Times to exhaus-
tion at 100 % of velocity at VO2max and modelling of the time-limit/
velocity relationship in elite long-distance runners . Eur J Appl Physiol
1994 ; 69 : 271 – 273
4 Bloomfi eld D M , Magnano A , Bigger J T Jr , Rivadeneira H , Parides M , Stein-
man R C . Comparison of spontaneous vs. metronome-guided breathing
on assessment of vagal modulation using RR variability . Am J Physiol
2001 ; 280 : H1145 – H1150
5 Borg G . Perceived exertion as an indicator of somatic stress . Scand
J Rehabil Med 1970 ; 2 : 92 – 98
6 Borresen J , Lambert M I . Changes in heart rate recovery in response to
acute changes in training load . Eur J Appl Physiol 2007 ; 101 : 503 – 511
7 Borresen J , Lambert M I . Autonomic control of heart rate during and
after exercise: measurements and implications for monitoring train-
ing status . Sports Med 2008 ; 38 : 633 – 646
8 Borresen J , Lambert M I . The quantifi cation of training load, the train-
ing response and the eff ect on performance . Sports Med 2009 ; 39 :
779 – 795
9 Bosquet L , Montpetit J , Arvisais D , Mujika I . E ff ects of tapering on per-
formance: a meta-analysis . Med Sci Sports Exerc 2007 ; 39 : 1358 – 1365
10 Buchheit M . The two-hour marathon: through a highly individualized
training process? J Appl Physiol 2011 ; 110 : 282 – 283 , discussion 294
11 Buchheit M , Chivot A , Parouty J , Mercier D , Al Haddad H , Laursen P B ,
Ahmaidi S . Monitoring endurance running performance using cardiac
parasympathetic function . Eur J Appl Physiol 2010 ; 108 : 1153 – 1167
12 Buchheit M , Gindre C . Cardiac parasympathetic regulation: respec-
tive associations with cardiorespiratory fi tness and training load . Am
J Physiol 2006 ; 291 : H451 – H458
13 Buchheit M , Laursen P B , Ahmaidi S . Parasympathetic reactivation after
repeated sprint exercise . Am J Physiol 2007 ; 293 : H133 – H141
14 Buchheit M , Mendez-Villanueva A , Quod M J , Poulos N . Bourdon P. Deter-
minants of the variability of heart rate measures during a competitive
period in young soccer players . Eur J Appl Physiol 2010 ; 109 : 869 – 878
15 Buchheit M , Millet G P , Parisy A , Pourchez S , Laursen P B , Ahmaidi S .
Supramaximal training and postexercise parasympathetic reactiva-
tion in adolescents . Med Sci Sports Exerc 2008 ; 40 : 362 – 371
16 Buchheit M , Papelier Y , Laursen P B , Ahmaidi S . Noninvasive assessment
of cardiac parasympathetic function: postexercise heart rate recovery
or heart rate variability? Am J Physiol 2007 ; 293 : H8 – H10
17 Daanen H A , Lamberts R P , Kallen V L , Jin A , Van Meeteren N L . A s y s -
tematic review on heart-rate recovery to monitor changes in training
status in athletes . Int J Sports Physiol 2012 ; 7 : 251 – 260
18 Dixon E M , Kamath M V , McCartney N , Fallen E L . Neural regulation of
heart rate variability in endurance athletes and sedentary controls .
Cardiovasc Res 1992 ; 26 : 713 – 719
19 Esteve-Lanao J , Foster C , Seiler S , Lucia A . Impact of training intensity
distribution on performance in endurance athletes . J Strength Cond
Res 2007 ; 21 : 943 – 949
20 Foster C . Monitoring training in athletes with reference to overtraining
syndrome . Med Sci Sports Exerc 1998 ; 30 : 1164 – 1168
21 Goldberger J J , Le F K , Lahiri M , Kannankeril P J , Ng J , Kadish A H . A s s e s s -
ment of parasympathetic reactivation after exercise . Am J Physiol
2006 ; 290 : H2446 – H2452
22 Hamilton R M , McKechnie P S , Macfarlane P W . Can cardiac vagal tone be
estimated from the 10-second ECG? Int J Cardiol 2004 ; 95 : 109 – 115
23 Harriss D J , Atkinson G . Ethical standards in sport and exercise science
research: 2014 update . Int J Sports Med 2013 ; 34 : 1025 – 1028
24 Hautala A J , Kiviniemi A M , Tulppo M P . Individual responses to aerobic
exercise: the role of the autonomic nervous system . Neurosci Biobe-
hav Rev 2009 ; 33 : 107 – 115
25 Hedelin R , Kentta G , Wiklund U , Bjerle P , Henriksson-Larsen K . Short-
term overtraining: eff ects on performance, circulatory responses, and
heart rate variability . Med Sci Sports Exerc 2000 ; 32 : 1480 – 1484
26 Hedelin R , Wiklund U , Bjerle P , Henriksson-Larsen K . Cardiac autonomic
imbalance in an overtrained athlete . Med Sci Sports Exerc 2000 ; 32 :
1531 – 1533
27 Houmard J A , Kirwan J P , Flynn M G , Mitchell J B . E ff ects of reduced train-
ing on submaximal and maximal running responses . Int J Sports Med
1989 ; 10 : 30 – 33
28 Iellamo F , Legramante J M , P igozzi F , Spataro A , Norbiato G , Lucini D ,
Pagani M . Conversion from vagal to sympathetic predominance with
strenuous training in high-performance world class athletes . Circula-
tion 2002 ; 105 : 2719 – 2724
29 Imai K , Sato H , Hori M , Kusuoka H , Ozaki H , Yokoyama H , Ta keda H ,
Inoue M , Kamada T . Vagally mediated heart rate recovery after exercise
is accelerated in athletes but blunted in patients with chronic heart
failure . J Am Coll Cardiol 1994 ; 24 : 1529 – 1535
30 Iwasaki K , Zhang R , Zuckerman J H , Levine B D . Dose-response relation-
ship of the cardiovascular adaptation to endurance training in healthy
adults: how much training for what benefi t? J Appl Physiol 2003 ;
95 : 1575 – 1583
31 Kiviniemi A M , Hautala A J , Kinnunen H , Tulppo M P . Endurance training
guided individually by daily heart rate variability measurements . Eur
J Appl Physiol 2007 ; 101 : 743 – 751
32 Lamberts R P , Swart J , Capostagno B , Noakes T D , Lambert M I . Heart
rate recovery as a guide to monitor fatigue and predict changes in
performance parameters . Scand J Med Sci Sports 2009 ; 20 : 449 – 457
33 Lamberts R P , Swart J , Noakes T D , Lambert M I . Changes in heart rate
recovery after high-intensity training in well-trained cyclists . Eur
J Appl Physiol 2009 ; 105 : 705 – 713
34 Le Meur Y , Pichon A , Schaal K , Schmitt L , Louis J , Gueneron J , Vidal P P ,
Hausswirth C . Evidence of parasympathetic hyperactivity in function-
ally overreached athletes . Med Sci Sports Exerc 2013
35 Lee C M , Wood R H , Welsch M A . I n fl uence of short-term endurance exer-
cise training on heart rate variability . Med Sci Sports Exerc 2003 ;
35 : 961 – 969
36 Lehmann M , Foster C , Dickhuth H H , Gastmann U . Autonomic imbalance
hypothesis and overtraining syndrome . Med Sci Sports Exerc 1998 ;
30 : 1140 – 1145
37 Manzi V , Castagna C , Padua E , Lombardo M , D’Ottavio S , Massaro M ,
Volterrani M , Iellamo F . Dose-response relationship of autonomic nerv-
ous system responses to individualized training impulse in marathon
runners . Am J Physiol 2009 ; 296 : H1733 – H1740
38 Mourot L , Bouhaddi M , Perrey S , Cappelle S , Henriet M T , Wolf J P , Rouil-
lon J D , Regnard J . Decrease in heart rate variability with overtraining:
assessment by the Poincare plot analysis . Clin Physiol Funct Imaging
2004 ; 24 : 10 – 18
39 Mujika I , Padilla S . Scientifi c bases for precompetition tapering strate-
gies . Med Sci Sports Exerc 2003 ; 35 : 1182 – 1187
40 Mujika I , Padilla S , Pyne D , Busso T . Physiological changes associated
with the pre-event taper in athletes . Sports Med 2004 ; 34 : 891 – 927
41 Penttila J , Helminen A , Jartti T , Kuusela T , Huikuri H V , Tulppo M P , Cof-
feng R , Scheinin H . Time domain, geometrical and frequency domain
analysis of cardiac vagal outfl ow: eff ects of various respiratory pat-
terns . Clin Physiol 2001 ; 21 : 365 – 376
42 Pichot V , Busso T , Roche F , Garet M , Costes F , Duverney D , Lacour J R ,
Barthelemy J C . Autonomic adaptations to intensive and overload
training periods: a laboratory study . Med Sci Sports Exerc 2002 ; 34 :
1660 – 1666
43 Pierpont G L , Voth E J . Assessing autonomic function by analysis of heart
rate recovery from exercise in healthy subjects . Am J Cardiol 2004 ;
94 : 64 – 68
44 Plews D J , Laursen P B , Kilding A E , Buchheit M . Heart rate variability in
elite triathletes, is variation in variability the key to eff ective training?
A case comparison . Eur J Appl Physiol 2012 ; 112 : 3729 – 3741
45 Plews D J , Laursen P B , Kilding A E , Buchheit M . Evaluating training adap-
tation with heart rate measures: A methodological comparison . Int
J Sports Physiol Perform 2013 ; 8 : 688 – 691
46 Plews D J , Laursen P B , Stanley J , Kilding A E , Buchheit M . Training adapta-
tion and heart rate variability in elite endurance athletes: opening the
door to eff ective monitoring . Sports Med 2013 ; 43 : 773 – 781
47 Savin W M , Davidson D M , Haskell W L . Autonomic contribution to
heart rate recovery from exercise in humans . J Appl Physiol 1982 ;
53 : 1572 – 1575
48 Seiler S , Haugen O , Kuff el E . Autonomic recovery after exercise in
trained athletes: intensity and duration eff ects . Med Sci Sports Exerc
2007 ; 39 : 1366 – 1373
49 Shepley B , MacDougall J D , Cipriano N , Sutton J R , Tarnopolsky M A , Coates
G . Physiological eff ects of tapering in highly trained athletes . J Appl
Physiol 1992 ; 72 : 706 – 711
50 Shin K , Minamitani H , Onishi S , Yamazaki H , Lee M . The power spectral
analysis of heart rate variability in athletes during dynamic exercise–
Part I . Clin Cardiol 1995 ; 18 : 583 – 586
51 Shin K , Minamitani H , Onishi S , Yamazaki H , Lee M . The power spectral
analysis of heart rate variability in athletes during dynamic exercise–
Part II . Clin Cardiol 1995 ; 18 : 664 – 668
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.

Training & Testing
Hug B et al. Tapering for Marathon and … Int J Sports Med
52 Stanley J , Peake J M , Buchheit M . Cardiac parasympathetic reactivation
following exercise: implications for training prescription . Sports Med
2013 [Epub ahead of print]
5 3 Task_Force . Heart rate variability. Standards of measurement, physi-
ological interpretation, and clinical use . Eur Heart J 1996 ; 17 : 354 – 381
54 Thomas L , Busso T . A theoretical study of taper characteristics to opti-
mize performance . Med Sci Sports Exerc 2005 ; 37 : 1615 – 1621
55 Uusitalo A L , Uusitalo A J , Rusko H K . Endurance training, overtraining
and barorefl ex sensitivity in female athletes . Clin Physiol 1998 ; 18 :
510 – 520
56 Uusitalo A L , Uusitalo A J , Rusko H K . Exhaustive endurance training for
6-9 weeks did not induce changes in intrinsic heart rate and cardiac
autonomic modulation in female athletes . Int J Sports Med 1998 ; 19 :
532 – 540
57 Uusitalo A L , Uusitalo A J , Rusko H K . Heart rate and blood pressure vari-
ability during heavy training and overtraining in the female athlete .
Int J Sports Med 2000 ; 21 : 45 – 53
58 Wehrlin J P , Hallen J . Linear decrease in.V
∙O2max and performance with
increasing altitude in endurance athletes . Eur J Appl Physiol 2006 ;
96 : 404 – 412
59 Weippert M , Kumar M , Kreuzfeld S , Arndt D , Rieger A , Stoll R . C o m p a r i -
son of three mobile devices for measuring R-R intervals and heart rate
variability: Polar S810i, Suunto t6 and an ambulatory ECG system . Eur
J Appl Physiol 2010 ; 109 : 779 – 786
Downloaded by: Bibliothèque Cantonale et Universitaire. Copyrighted material.