International Journal of Sports Physiology and Performance, 2015, 10, 528  -534

http://dx.doi.org/10.1123/ijspp.2014-0196
© 2015 Human Kinetics, Inc. Brief Report

Cardiac Parasympathetic Activity and Race Performance:

An Elite Triathlete Case Study
Jamie Stanley, Shaun D’Auria, and Martin Buchheit

The authors examined whether changes in heart-rate (HR) variability (HRV) could consistently track adaptation to training and
race performance during a 32-wk competitive season. An elite male long-course triathlete recorded resting HR (RHR) each
morning, and vagal-related indices of HRV (natural logarithm of the square root of mean squared differences of successive
R–R intervals [ln rMSSD] and the ratio of ln rMSSD to R–R interval length [ln rMSSD:RR]) were assessed. Daily training
load was quantified using a power meter and wrist-top GPS device. Trends in HRV indices and training load were examined
by calculating standardized differences (ES). The following trends in week-to-week changes were consistently observed: (1)
When the triathlete was coping with a training block, RHR decreased (ES –0.38 [90% confidence limits –0.05;–0.72]) and ln
rMSSD increased (+0.36 [0.71;0.00]). (2) When the triathlete was not coping, RHR increased (+0.65 [1.29;0.00]) and ln rMSSD
decreased (–0.60 [0.00;–1.20]). (3) Optimal competition performance was associated with moderate decreases in ln rMSSD
(–0.86 [–0.76;–0.95]) and ln rMSSD:RR (–0.90 [–0.60;–1.20]) in the week before competition. (4) Suboptimal competition
performance was associated with small decreases in ln rMSSD (–0.25 [–0.76;–0.95]) and trivial changes in ln rMSSD:RR (–0.04
[0.50;–0.57]) in the week before competition. To conclude, in this triathlete, a decrease in RHR concurrent with increased ln
rMSSD compared with the previous week consistently appears indicative of positive training adaptation during a training block.
A simultaneous reduction in ln rMSSD and ln rMSSD:RR during the final week preceding competition appears consistently
indicative of optimal performance.

Keywords: monitoring, cardiac parasympathetic function, triathlon

Knowing how an athlete is responding during key precompeti-
tion phases of training and preevent taper is of particular importance
to elite athletes and coaches. Longitudinal monitoring of heart-rate
(HR) variability (HRV) (eg, natural logarithm of the square root of
mean squared differences of successive R–R intervals [ln rMSSD])
has potential for better prescription of training loads1 and predict-
ing maladaptation2 or aerobic performance.3 However, equivocal
findings may be due to methodological inaccuracies associated
with the large day-to-day variation in HRV.2–4 Furthermore, correct
practical interpretation of changes in HRV requires understanding of
sympathetic versus vagal inputs, which can be achieved by normal-
izing HRV data for the prevailing HR (eg, ratio of ln rMSSD to R–R
interval [ln rMSSD:RR]).3,4 For example, a trend toward reduced
ln rMSSD with concurrent reduction in ln rMSSD:RR suggests
vagal saturation, while concurrent increase in ln rMSSD:RR sug-
gests increased sympathetic activity.3 Such responses demonstrate
how interpreting HRV alone may inaccurately diagnose adaptive
responses to training.4
Despite the potential of these HRV indices for monitoring
training adaptation,1,3,5 such observations are currently limited to
isolated events in elite rowers.3,6 Whether such HRV patterns are
reproducible and consistent throughout repeated events for athletes
of different sports remains unknown. Therefore, the aim of this
study was to monitor a long-course triathlete during a 32-week
competitive season to determine whether trends in vagal-related
indices of HRV with respect to training stimulus and competition
performance were reproduced and consistent.
Methods
Subjects
A professional male (28-y-old) triathlete was monitored throughout
a 32-week competitive season that included 5 triathlons (1.9-km
swim, 90.1-km cycle, 21.1-km run). At the start of the monitoring
period his height was 1.79 m, body mass 71.8 kg, maximal aerobic
capacity 72.2 mL · kg–1 · min–1, and peak power output 438 W.7
Maximal values were obtained during a test to exhaustion (25-W
increments every 60 s starting at 125 W). Training was periodized
with the goal of peaking for each successive event. The study pro-
cedure was approved by the human research ethics committee at
the University of Queensland.

Stanley is with the School of Human Movement and Nutrition Sciences,
University of Queensland, Brisbane, Australia. D’Auria is with the Triathlon
Program, Queensland Academy of Sport, Brisbane, Australia. Buchheit is
with the Sport Science Unit, Myorobie Association, Montvalezan, France.
Address author correspondence to Jamie Stanley at j.stanley@uq.edu.au.
Experimental Overview
Training and racing were monitored using a power meter (Riken,
Quarq Technology, USA) for cycling and a wrist-top GPS (Ambit,
Suunto Oy, Finland) for running and processed in dedicated soft-
ware (WKO+ v 3.0, PeaksWare, USA). Every swimming session
was recorded and processed manually. HR was recorded on waking

528

in a seated position for 5 minutes using an HR monitor (Ambit,
Suunto Oy, Finland).
Data Analysis
Daily training load was expressed as a training-stress score, acute
training load, and chronic training load as described by Allen and
Coggan.8 Daily waking resting HR (RHR) and HRV data were pro-
cessed7 and analyzed as reported elsewhere.2,3 Responses to training
were assessed via retrospective analysis of training logs. A “coping”
response to training was deemed only if all weekly programmed
sessions were completed successfully. A race performance within
10% of the winner’s overall time was classified as optimal based
on this triathlete’s initial world ranking (>300). The winner’s per-
formance was assumed comparable between events because each
was either an Olympian or ex–world champion.
Statistical Analysis
Weekly data are expressed as means (Monday–Sunday) and 90%
confidence intervals (CI) unless stated otherwise. Week-to-week
Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4
differences in HR-based variables and training load are expressed as
standardized mean difference (ES) and assessed using an approach
based on magnitudes of change.9 The threshold values for ES sta-
tistics were ≤0.2 trivial, >0.2 small, >0.6 moderate, >1.2 large, and
>2.0 very large.9
Results
Weekly mean ± SD training duration for the 32-week monitoring
period was 14 ± 5 hours. Training comprised 24.6% swimming,
48.3% cycling, and 27.1% running based on the weekly total
training-stress score. The weekly mean ± SD for RHR was 52.3 ± 2.5
beats/min, ln rMSSD was 4.0 ± 0.2 milliseconds, and ln rMSSD:RR
(×103) was 3.5 ± 0.1. Figure 1 depicts longitudinal changes in race
performance relative to the winner, ratings of perceived fatigue,
HR-based measures, and training load. The triathlete improved his
world ranking from >300 to 69th after the fifth competition.
Table 1 displays weekly changes in perceived fatigue, HR-based
measures, and contextual interpretation.4 The following trends in
week-to-week changes were consistently observed: (1) When the
triathlete was “coping,” small decreases in RHR (ES –0.38 [90%
confidence limits –0.05;–0.72]) and small increases in ln rMSSD
(+0.36 [0.71;0.00]) were observed. (2) When the triathlete was
not coping, RHR moderately increased (+0.65 [1.29;0.00]) and ln
rMSSD moderately decreased (–0.60 [0.00;–1.20[). (3) Optimal
competition performance was associated with moderate decreases
in ln rMSSD (–0.86 [–0.76;–0.95]) and ln rMSSD:RR (–0.90
[–0.60;–1.20]) in the week before competition. (4) Suboptimal
competition performance was associated with small decreases in ln
rMSSD (–0.25 [–0.76;–0.95]) and trivial changes in ln rMSSD:RR
(–0.04 [0.50;–0.57]) in the week before competition.
Discussion
This study provides novel and practical findings regarding the use of
HR-based indexes for monitoring training adaptation. First, varia-
tions in training load induce changes in HR-based indices, which
may be dissociated from the triathlete’s training status. Second,
IJSPP Vol. 10, No. 4, 2015
Monitoring Training Adaptation With HRV   529
consistent trends in HR-based indices were observed when the
triathlete was coping versus not coping during a training block and
during the week before an optimal versus suboptimal competition
performance.
In the current study, the triathlete cycled through 4 precompe-
tition training blocks of progressively increased training load (eg,
days 1–21, 56–63, 119–140, 182–210; Figure 1).10 Coping with
prescribed training load during precompetition training blocks is
fundamental for inducing positive adaptations and making per-
formance gains (ie, overload or functional overreaching). Data
from elite rowers,3 and elite2 and well-trained5 triathletes suggest
that an ~4% to 9% increase in vagal-related HRV during precom-
petition training blocks may be required for optimal competition
performance after a taper. Consistent with these data, we observed
increases in ln rMSSD of up to 10% during precompetition training
blocks (Figure 1[d]). Furthermore, we observed that when the tri-
athlete was coping with the prescribed training load, week-to-week
changes in HR-based indices consistently demonstrated decreases
in RHR with concurrent increases in ln rMSSD (Figure 1, Table
1). By contrast, when the triathlete was not coping, week-to-week
changes in HR-based indices consistently demonstrated increases
in RHR with concurrent decreases in ln rMSSD (Figure 1, Table 1).
Based on these patterns, stagnation/decrease in ln rMSSD during
a precompetition training block may suggest insufficient training
stimulus—confirmation requires data from the off-season or a
sustained period of reduced training that was not captured in this
study. While daily variations exist due to acute changes in training
load and recovery kinetics,1 our data suggest that weekly changes
in HR-based measures assessed in parallel with the training context
can provide practically useful information regarding adaptation to
training.
For positive adaptations initiated in precompetition training
blocks to be translated into optimized competition performance,
training load must be reduced for a variable period of time before
competition.11 Accordingly, training load was reduced before each
competition (eg, d 28–42, 70, 147–161, 217–224; Figure 1[f], 2[d]).
Preceding both optimal overall race performances (races 4 and 5;
Figure 1[a]), moderate decreases in ln rMSSD and ln rMSSD:RR
were observed (Figure 2[b], [c]). Because these changes were
associated with an increased RHR, possible HRV saturation can
be excluded, suggesting a likely increase in sympathetic activity
probably resulting from both the reduced training volume and
prerace anxiety.3,4,12 By contrast, small decreases in ln rMSSD and
trivial change in ln rMSSD:RR were observed preceding subopti-
mal competition performances (Figure 2[b], [c]). The suboptimal
performance in race 1 was likely due to poor training adaptation
(wk 2 and 4), reduced training load (loss of fitness), and elevated
fatigue and sickness during the preceding training block. Conse-
quently, race 2 was also a suboptimal performance, irrespective of
HR-based indices, suggesting that the triathlete had adapted well (ie,
increased ln rMSSD, d 49–63, Figure 1; wk 8 and 9, Table 1) and
was ready to race (wk 10, Table 1). The suboptimal performance in
race 3 was likely due to an inappropriate taper because the triathlete
experienced a substantial increase in perceived fatigue despite a
stable training load (d 133–140, Figure 1; wk 20–21, Table 1) yet
appeared to have coped well with the preceding training block (d
112–132, Figure 1). While our data largely confirm those reported
in elite rowers3 during the lead-up to a race, these examples of
suboptimal competition performance highlight the importance
of interpreting HR-based indices with knowledge of the training
 530  Stanley, D’Auria, and Buchheit
Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4

Figure 1(A–C) — Changes in (A) half-ironman competition performance (overall time relative to winner), (B) perceptions of general fatigue, and
(C) resting heart rate (RHR) for the entire 32-week competition season. The horizontal dashed line represents the threshold between good and poor
competition performance. The gray shaded area for general fatigue, RHR, ln rMSSD, and ln rMSSD:RR represents the smallest worthwhile change and
was calculated from 0.5 times the individual coefficient of variation over the entire 32-week competition season.9 Arrows indicate day of competition.

context.4 Furthermore­, these data demonstrate the importance of Practical Applications

adapting (coping) well during precompetition training blocks,
because if the work has not been done, an athlete’s fitness becomes Endurance sports such as long-course triathlon require consis-
the limitation despite HR-based indices suggesting a readiness for tently high training loads, so monitoring adaptation to training
optimal performance. is fundamental for achieving performance gains and minimizing

IJSPP Vol. 10, No. 4, 2015

 Monitoring Training Adaptation With HRV   531
Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4

Figure 1(D–F) — Changes in (D) natural logarithm of the square root of the mean of the sum of the squares of differences between adjacent normal
R–R intervals measured on waking (ln rMSSD), (E) ratio of ln rMSSD to mean R–R interval (ln rMSSD:RR), and (F) distribution of acute training
load8 and chronic training load8 for the entire 32-week competition season. The horizontal dashed line represents the threshold between good and poor
competition performance. The gray shaded area for general fatigue, RHR, ln rMSSD, and ln rMSSD:RR represents the smallest worthwhile change and
was calculated from 0.5 times the individual coefficient of variation over the entire 32-week competition season.9 Arrows indicate day of competition.

overtraining or injury risk during precompetition training blocks and coping during training blocks and an optimal versus suboptimal
optimizing the precompetition taper. The current data suggest that precompetition taper. It must be noted that in practice, acute varia-
provided the training context is known, HR-based indices provide tions in training load may substantially influence short-term HR
consistent trends that can discriminate between coping versus not responses.6 Therefore, while weekly trends may flag important

IJSPP Vol. 10, No. 4, 2015

 Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4

532
Table 1  Contextual Interpretation and Mechanisms of Changes in Heart Rate Measures and Perceived Fatigue Compared With the Previous Week
Week Training contexta (TSS) ATL Fatigue RHR HRV HRV:RR Likely mechanism4 Practical interpretation4 Alternative interpretation
2 Load block (642.9)*§ ↑↑ ↑↑ ↓ ↑ ↓ Increased PA Coping well Adequate training reduction
3 Load block (1030.3) ↓↓ ↓↓ ↔ ↔ ↔ Unknown and highly individual
4 Load block (671.2)* ↔ ↑↑ ↑↑ ↔ ↑↑ Increased SA/reverse saturation Accumulated fatigue
5 Taper (384.3) ↓↓↓↓ ↓↓ ↑ ↓↓ ↓↓ Increased PA/saturation Coping well
6 Taper+race (704.3) ↔ ↔ ↔ ↔ ↔ Unknown and highly individual
7 Recovery week (499.5) ↓↓ ↓↓↓ ↓↓ ↑↑ ↑ Increased SA Accumulated fatigue
8 Load block (1004.0) ↑↑↑↑ ↔ ↓↓↓ ↑↑↑ ↑ Increased SA Ready to perform
9 Load block (1011.8) ↑ ↑↑ ↓ ↓↓ ↓↓↓ Increased PA/saturation Coping well
10 Taper+race (724.5) ↔ ↓↓↓ ↑ ↓↓ ↓↓ Increased SA Coping well Ready to perform
11 Recovery week (138.4) ↓↓↓↓ ↑↑ ↑↑ ↓ ↑ Increased SA Ready to perform
12 Load block (941.7) ↑↑ ↓↓↓ ↔ ↑↑ ↑↑↑ Increased SA Ready to perform
13 Load block (983.9) ↑↑↑↑ ↑↑↑ ↔ ↓ ↓ Increased PA Coping well
14 Load block (1162.6) ↑↑↑↑ ↑↑ ↓ ↔ ↓ Increased PA Coping well
15 Recovery week (553.6) ↓↓↓↓ ↔ ↑ ↓ ↔ Increased SA Accumulated fatigue
16 Load block (703.4)* ↓↓ ↔ ↔ ↓ ↓↓ Increased PA Unknown and highly individual
17 Load block (1296.9) ↑↑↑↑ ↓↓↓ ↔ ↑↑ ↑ Increased SA Accumulated fatigue
18 Load block (850.4)*§ ↔ ↑↑ ↑ ↓↓ ↓ Increased PA Coping well Adequate training reduction
19 Load block (870.9) ↓↓ ↓↓↓ ↓↓↓ ↑↑ ↓ Increased PA/saturation Coping well
20 Load block (1073.2) ↑↑↑ ↑↑ ↑↑↑ ↓ ↑ Increased SA Accumulated fatigue
21 Load block+taper (588.2)*§ ↓↓↓↓ ↑ ↑↑↑ ↓↓↓ ↑↑ Increased SA Accumulated fatigue

IJSPP Vol. 10, No. 4, 2015

22 Taper+race (683.1) ↓↓ ↓↓ ↔ ↔ ↑ Increased SA Ready to perform
23 Taper+race (625.8) ↓↓ ↑↑ ↑ ↓↓ ↓↓ Increased SA Coping well Ready to perform
24 Recovery week (157.0) ↓↓↓ ↑ ↔ ↔ ↔ Detraining
25 Load block (853.0) ↑ ↓ ↓↓ ↑↑ ↔ Increased PA Coping well
26 Load block (1170.1) ↑↑↑↑ ↓↓ ↓ ↔ ↓ Increased PA Coping well
27 Load block (1285.0) ↑↑↑↑ ↔ ↓↓↓ ↑↑↑↑ ↑↑↑ Increased PA Accumulated fatigue Functional overreaching
28 Recovery week (958.4)* ↓ ↑↑ ↑↑ ↓↓ ↔ Increased SA Accumulated fatigue
29 Load block (1208.5) ↑↑ ↑↑ ↓↓ ↑↑ ↔ Increased PA Coping well
30 Load block+taper (1100.7) ↓ ↓ ↑↑↑ ↓↓ ↑ Increased SA Accumulated fatigue
31 Taper (928.6) ↓ ↔ ↓↓ ↑↑ ↓ Increased PA Coping well
32 Taper+race (817.3) ↓↓↓ ↓ ↑ ↓↓ ↓↓ Increased SA Coping well Ready to perform
Abbreviations: TSS, total training-stress score for the
week8; ATL,
acute training load8;
fatigue, perception of general fatigue; RHR, resting heart rate (beats/min); HRV, natural logarithm of the square root of the mean of the
sum of the square differences between adjacent normal R–R intervals (ms); HRV:RR, ratio of HRV to R–R interval; PA, parasympathetic activity; SA, sympathetic activity. Standardized week-to-week differences: ↔ trivial,
↑ small increase, ↑↑ moderate increase, ↑↑↑ large increase, ↑↑↑↑ very large increase, ↓ small decrease, ↓↓ moderate decrease, ↓↓↓ large decrease, ↓↓↓↓ very large decrease.
aBased on training logs.
*Missed training session(s) that week (ie, not coping; see Methods). §Upper respiratory tract infection.

Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4
Figure 2(A–B) — Individual and mean (± 90% CI) changes with respect
to the preceding week for the entire 32-week competition season during the
weeks preceding the 5 competitions. Upward triangles represent changes
preceding an optimal competition performance; downward triangles repre-
sent changes preceding suboptimal competition performance. Changes in
the weekly mean (A) resting heart rate (RHR) and (B) resting vagal-related
heart-rate variability (ln rMSSD).8 The hatched shaded area represents a
trivial difference.9
(adverse) responses, practitioners should focus on interpreting
the overall trends throughout a training block when assessing its
success. How this information can be used to direct daily training
prescription is an area of future research.
Conclusion
This case study demonstrates the utility of HR-based indices for
monitoring adaptation to training. HR-based indices displayed con-
sistent trends in response to changes in training that can discriminate
between coping versus not coping during a training block and an
optimal versus suboptimal precompetition taper.
Acknowledgments
This study was supported by the Centre of Excellence for Applied Sport
Science Research and the Triathlon Program at the Queensland Academy
of Sport.
IJSPP Vol. 10, No. 4, 2015
Monitoring Training Adaptation With HRV   533
Figure 2(C–D) — Individual and mean (±90% CI) changes with respect
to the preceding week for the entire 32-week competition season during the
weeks preceding the 5 competitions. Upward triangles represent changes
preceding an optimal competition performance; downward triangles rep-
resent changes preceding suboptimal competition performance. Changes
in the weekly mean (C) ratio of resting vagal-related heart-rate variability
to R–R interval (ln rMSSD:RR) and (D) acute training load.8 The hatched
shaded area represents a trivial difference.9
References
1. Stanley J, Peake J, Buchheit M. Cardiac parasympathetic reactivation
following exercise: implications for training prescription. Sports Med.
2013;43:1259–1277. PubMed doi:10.1007/s40279-013-0083-4
2. Plews DJ, Laursen P, Kilding A, Buchheit M. Heart rate variability
in elite triathletes, is variation in variability the key to effective train-
ing?: a case comparison. Eur J Appl Physiol. 2012;112:3729–3741.
PubMed doi:10.1007/s00421-012-2354-4
3. Plews DJ, Laursen PB, Stanley J, Kilding AE, Buchheit M. Training
adaptation and heart rate variability in elite endurance athletes: open-
ing the door to effective monitoring. Sports Med. 2013;43:773–781.
PubMed doi:10.1007/s40279-013-0071-8
4. Buchheit M. Monitoring training status with hr measures: do all roads
lead to Rome? Front Physiol. 2014;5:73. PubMed
5. Le Meur Y, Pichon A, Schaal K, et al. Evidence of parasympa-
thetic hyperactivity in functionally overreached athletes. Med
Sci Sports Exerc. 2013;45:2061–2071. PubMed doi:10.1249/
MSS.0b013e3182980125
 534  Stanley, D’Auria, and Buchheit
6. Plews DJ, Laursen PB, Le Meur Y, Hausswirth C, Kilding AE, Buchheit
M. Monitoring training with heart-rate variability: how much compli-
ance is needed for valid assessment? Int J Sports Physiol Perform.
2014;9(5):783–790. PubMed http://dx.doi.org/10.1123/IJSPP.2013-0455
7. Stanley J, Peake J, Coombes J, Buchheit M. Central and peripheral
adjustments during high-intensity exercise following cold water
immersion. Eur J Appl Physiol. 2014;114:147–163. PubMed
doi:10.1007/s00421-013-2755-z
8. Allen H, Coggan AR. Training With a Powermeter. Boulder, CO: Velo
Press; 2006.
9. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progres-
sive statistics for studies in sports medicine and exercise science.
Downloaded by ETSU on 09/20/16, Volume 10, Article Number 4
IJSPP Vol. 10, No. 4, 2015
Med Sci Sports Exerc. 2009;41:3–13. PubMed doi:10.1249/
MSS.0b013e31818cb278
10. Issurin VB. New horizons for the methodology and physiology
of training periodization. Sports Med. 2010;40:189–206. PubMed
doi:10.2165/11319770-000000000-00000
11. Mujika I, Padilla S, Pyne D, Busso T. Physiological changes associated
with the pre-event taper in athletes. Sports Med. 2004;34:891–927.
PubMed doi:10.2165/00007256-200434130-00003
12. Plews DJ, Laursen PB, Kilding AE, Buchheit M. Heart-rate variability
and training-intensity distribution in elite rowers. Int J Sports Physiol
Perform. 2014;9(6):1026–1032. PubMed http://dx.doi.org/10.1123/
ijspp.2013-0497