Caffeine Use or Napping to Enhance Repeated Sprint Performance After Partial Sleep Deprivation: Why Not Both?

Abstract

Purpose:
To compare the effect of a 20-minute nap opportunity (N20), a moderate dose of caffeine (CAF; 5 mg·kg-1), or a moderate dose of caffeine before N20 (CAF+N) as possible countermeasures to the decreased performance and the partial sleep deprivation-induced muscle damage.

Methods:
Nine male, highly trained judokas were randomly assigned to either baseline normal sleep night, placebo, N20, CAF, or CAF+N. Test sessions included the running-based anaerobic sprint test, from which the maximum (Pmax), mean (Pmean), and minimum (Pmin) powers were calculated. Biomarkers of muscle, hepatic, and cardiac damage and of enzymatic and nonenzymatic antioxidants were measured at rest and after the exercise.

Results:
N20 increased Pmax compared with placebo (P < .01, d = 0.75). CAF+N increased Pmax (P < .001, d = 1.5; d = 0.94), Pmin (P < .001, d = 2.79; d = 2.6), and Pmean (P < .001, d = 1.93; d = 1.79) compared with placebo and CAF, respectively. Postexercise creatine kinase increased whenever caffeine was added, that is, after CAF (P < .001, d = 1.19) and CAF+N (P < .001, d = 1.36). Postexercise uric acid increased whenever participants napped, that is, after N20 (P < .001, d = 2.19) and CAF+N (P < .001, d = 2.50) and decreased after CAF (P < .001, d = 2.96).

Conclusion:
Napping improved repeated-sprint performance and antioxidant defense after partial sleep deprivation. Contrarily, caffeine increased muscle damage without improving performance. For sleep-deprived athletes, caffeine before a short nap opportunity would be more beneficial for repeated sprint performance than each treatment alone.

Full text

Caffeine Use or Napping to Enhance Repeated Sprint
Performance After Partial Sleep Deprivation: Why Not Both?
Mohamed Romdhani, Nizar Souissi, Imen Moussa-Chamari, Yassine Chaabouni, Kacem Mahdouani,
Zouheir Sahnoun, Tarak Driss, Karim Chamari, and Omar Hammouda
Purpose:To compare the effect of a 20-minute nap opportunity (N20), a moderate dose of caffeine (CAF; 5 mg·kg
−1
), or a
moderate dose of caffeine before N20 (CAF+N) as possible countermeasures to the decreased performance and the
partial sleep deprivation–induced muscle damage. Methods:Nine male, highly trained judokas were randomly assigned to
either baseline normal sleep night, placebo, N20, CAF, or CAF+N. Test sessions included the running-based anaerobic sprint
test, from which the maximum (P
max
), mean (P
mean
), and minimum (P
min
) powers were calculated. Biomarkers of muscle,
hepatic, and cardiac damage and of enzymatic and nonenzymatic antioxidants were measured at rest and after the exercise.
Results:N20 increased P
max
compared with placebo (P<.01, d= 0.75). CAF+N increased P
max
(P<.001, d= 1.5; d= 0.94),
P
min
(P<.001, d= 2.79; d= 2.6), and P
mean
(P<.001, d= 1.93; d= 1.79) compared with placebo and CAF, respectively.
Postexercise creatine kinase increased whenever caffeine was added, that is, after CAF (P<.001, d= 1.19) and CAF+N
(P<.001, d= 1.36). Postexercise uric acid increased whenever participants napped, that is, after N20 (P<.001, d= 2.19) and
CAF+N (P<.001, d= 2.50) and decreased after CAF (P<.001, d= 2.96). Conclusion:Napping improved repeated-sprint
performance and antioxidant defense after partial sleep deprivation. Contrarily, caffeine increased muscle damage without
improving performance. For sleep-deprived athletes, caffeine before a short nap opportunity would be more beneficial for
repeated sprint performance than each treatment alone.
Keywords:sleep restriction, daytime sleep, psychostimulant, RSA, inflammation, antioxidant status
Several studies have reported that physical performances are
affected by partial sleep deprivation (PSD).
1–6
PSD reduces aerobic
intermittent,
1,2
time to exhaustion,
3,4
short-term all-out,
5
and
repeated sprint
6
performances. Several studies have found that
PSD caused by awakening at the end of the night had more
deleterious effects than PSD caused by a late bedtime on repeated
sprint performance during the postlunch dip (PLD).
1,3,6
Morning
performances were unaffected by PSD
1,5
, however, exercise per-
formed during or after the PLD would be affected by the PSD.
3–6
The interaction between the 2 process models of sleep regulation is
believed to be behind the previously mentioned outcomes.
7
Indeed,
sleep pressure increases exponentially with increased wakefulness
duration (homeostatic process S). Also, the PLD corresponds to the
secondary peak in sleepiness (circadian process C). Especially after
suboptimal nocturnal sleep, sleepiness increases and performance
decreases during the PLD.
8
Thus, PLD seems to be the perfect time
to nap, because sleep propensity is higher and sleep latency is
shorter.
9
Furthermore, PSD increases muscle, hepatic, and cardiac
damage and decreases antioxidant defense after a single bout of
exercise,
3,6,10
which would potentially affect recovery. Therefore,
upcoming training and competition would be potentially affected
by the incomplete recovery process.
10
Because PSD is very common
among athletes before major competitions,
11
the need to find
strategies to reduce the negative effects of PSD on performance
and the increased cellular damage is paramount. Therefore, the
primary intent would be totake a nap or consume a psychostimulant.
It has been reported that even a short nap could be an efficient
countermeasure to sleep loss–induced sleepiness.
4,8,12–16
Naps of no
longer than 20 minutes can restore cognitive impairment caused by
more than 16 hours of extended wakefulness.
17
From several studies,
20 minutes is regarded as the perfect duration of a nap, because it is
long enough to reduce sleepiness and short enough to prevent sleep
inertia.
4,8,13,16–18
Sleep inertia is the groggy feeling that occurs at
awakening.
19
Its severitydepends on several factors such as the sleep
stage at awakening, the length of prior wakefulness, and the time of
day (for an extensive review, cf Hilditch et al
19
).
Napping has recently received increasing attention in sport
performance.
4,8,16,20–23
Results range from improving various
physical performances
4,8,16,20–22
to neutral effects.
23
It has been
suggested that the beneficial effect of napping depends on athletes’
need for sleep.
20
Napping would improve performance only for
athletes with suboptimal sleep. Likewise, napping reduces sleep
loss–induced pain sensitivity
18
and normalizes the effect of PSD on
cortisol,
24
norepinephrine,
18
and interleukin-6.
18,24
However, only
a few studies have examined the effect of napping on biochemical
response to exercise.
16,21
The latter investigations reported
Chamari and Hammouda participated equally in the study. Romdhani and Souissi are
with the High Inst of Sport and Physical Education, Ksar-Said, Manouba University,
Manouba, Tunisia; and Physical activity, Sport and health, UR18JS01, National
Observatory of Sports, Tunis, Tunisia. Moussa-Chamari is with the Physical Educa-
tion Dept, College of Education, Qatar University, Doha, Qatar. Chaabouni and
Mahdouani are with the Dept of Biochemistry, CHU Ibn Jazzar, Kairouan, Tunisia;
and the Laboratory of Analysis, Treatment and Valorization of Pollutants of the
Environment and Products (LATVEP), Faculty of Pharmacy, University of Monastir,
Monastir, Tunisia. Sahnoun is with the Laboratory of Pharmacology, Faculty of
Medicine, University of Sfax, Sfax, Tunisia. Driss is with the Research Center on
Sport and Movement (Centre de Recherches sur le Sport et le Mouvement, CeRSM),
UPL, UFR STAPS, Université Paris Nanterre, Nanterre, France. Chamari is with
Aspetar, Qatar Orthopaedic and Sports MedicineHospital, Doha, and ''OPS Research
Lab'', CNMSS, Tunis, Tunisia. Hammoudais with the High Inst of Sport and Physical
Education, University of Sfax, Sfax, Tunisia. Romdhani (romdhaniroma@gmail
.com) is corresponding author.
1
International Journal of Sports Physiology and Performance, (Ahead of Print)
https://doi.org/10.1123/ijspp.2019-0792
© 2021 Human Kinetics, Inc. ORIGINAL INVESTIGATION

conflicting results regarding the effects of napping on sleep loss–
induced inflammation after exercise. Therefore, the effect of napping
on the biochemical response to exercise warrants further
investigation.
On the other hand, caffeine is the most widely consumed
psychostimulant in the world.
25
It offsets physical and cognitive
degradation with sleep loss.
25
However, data on the effects of
caffeine ingestion on short-term high-intensity performance are
controversial.
25
It has been reported that a moderate dose of
caffeine increased,
26
decreased,
27
or had no effects
28
on repeated
sprint exercise (cf McLellan et al
25
for an extensive review).
Despite the large number of studies, the effects of caffeine on
the biochemical responses to exercise have shown contradictory
findings. Caffeine may have anti-inflammatory
29
or prooxidant
30,31
properties, and some studies have shown neutral effects of caffeine
on biomarkers of muscle injury.
32
Thus, the effect of caffeine
ingestion on repeated sprint performance and the associated bio-
chemical response to the exercise warrants more studies.
When multicomponent interventions were studied, the com-
bination of caffeine with a short nap resulted in better psychocog-
nitive performance than each intervention alone.
13–15
Orally
ingested caffeine goes into action ∼30 minutes after ingestion
and has a long half-life (2–10 h).
33
Thus, caffeine should be
ingested shortly before a 20-minute nap to prevent it from dis-
turbing sleep and to maximize the gains obtained from nap.
13,14
To
the best of the authors’knowledge, the effects of a combination of
caffeine and a short nap were not tested for physical performance.
Therefore, the current study aimed to investigate the effect of a 20-
minute nap (N20), a moderate dose of caffeine (CAF; 5 mg·kg
−1
),
and the combination of N20 with CAF (CAF+N) as possible
countermeasures to performance decrement during the PLD and
biochemical disruption after PSD.
Methods
Participants
First, 13 highly trained male judokas fulfilled the inclusion criteria
(nonhabitual nappers, nonsmokers, and caffeine naïve, ie, ≤80 mg
of caffeine·d
–125
), volunteered to participate, and completed the
protocol. Only 9 participants’data are included in the analysis
(18.78 [1.09] y, 170.83 [5.87] cm, 68.44 [6.19] kg, body mass
index = 23.42 [1.24] kg·m
−2
). They were regularly engaged in
∼2 hours per day, 5 days per week of training (including high-
intensity intermittent training) and participated in competitions for
at least 5 years. Participants were recruited according to their
chronotype, explored via a morningness/eveningness questionnaire.
34
Only athletes with moderate and intermediate chronotypes (ie, those
who scored between 31 and 69 on the morningness/eveningness
questionnaire) were recruited. During the month preceding the experi-
ment, sleep diaries were collected (retiring and rising time, time in bed,
sleep latency, and waking frequency and duration). Only participants
who scored ≤5 on the Pittsburg Sleep Quality Index were recruited, to
avoid the involvement of participants with poor sleep.
35
The present
study was conducted according to the ethical guidelines of the
Declaration of Helsinki (64th World Medical Association General
Assembly, Fortaleza, Brazil, October 2013). The Manouba Univer-
sity Review Board approved the protocol (P-SC no. 009/15). All
participants were informed about the study design and potential risks
and signed an informed consent before the assessments began.
Participants were also informed about their right to withdraw
from the study at any time without any penalty.
Experimental Design
Before starting the main experimental protocol, participants under-
went 2 habituation sessions, in which they were familiarized with
the experimenters, laboratory, materials, napping room, tests, and
questionnaires (Figure 1).
Participants performed 5 test sessions 1 week apart in a
randomized and counterbalanced order. They were assigned to
either baseline normal sleep night, placebo (PLA), 20-minute nap
with PLA ingestion (N20), intake of 5 mg·kg
−1
of caffeine without
napping (CAF), and intake of 5 mg·kg
−1
of caffeine before N20
(CAF+N; Figure 1). During each session, participants came to the
laboratory at ∼2000 hours, ate a standardized dinner at ∼2030 hours,
and performed ∼90 minutes of free activity until ∼2200 hours, when
they were asked to go bed (ie, all lights and devices off). Except for
normal sleep night, when participants had a full sleep night (ie, time
in bed between 2200 hours and 0630 hours), all sessions were
undertaken after a PSD night (ie, time in bed between 2200 hours
and 0230 hours). They ate a standardized breakfast at 0700 hours.
After this, they were asked to not consume food, but water was
allowed ad libitum. They stayed awake until ∼1200 hours doing
passive activities (eg, watching television, playing cards or video
Figure 1 —Simplified experimental protocol. All times are expressed in local time (GMT+1 h). CAF indicates 5 mg·kg
−1
of caffeine; CAF+N,
5 mg·kg
−1
of caffeine + 20-minute nap; N20, 20-minute nap with placebo; NSN, normal sleep night; PLA, partial sleep deprivation with placebo; RAST,
the running-based anaerobic sprint test.
(Ahead of Print)
2Romdhani et al

games, or reading). At 1200 hours, they ate a standardized isocaloric
lunch and reclined in a comfortable armchair. At 1400 hours,
participants in the nap condition entered the room that was condu-
cive to sleep (comfortably warm, fully dark, and quiet). At this time,
volunteers ingested 5 mg·kg
−1
of pure powdered caffeine for CAF
+N or cellulose and starch–based PLA for N20.
13,14
The desired
amount of caffeine was measured with a specific Shimadzu (Shi-
madzu Corporation, Kyoto, Japan) electronic weighing machine
(±0.1 mg) and put into capsules to match PLA in weight, color, and
smell. After this, participants wore earplugs and eye masks and got
into bed, and they were allowed to 20-minute nap opportunity (from
1410 hours to 1430 hours). Participants in the no-nap condition
ingested caffeine or PLA at the same time (ie, 1400 hours). Constant
supervision was performed via an infrared camera connected, in real
time, to the experimenter’s computer. When the nap period elapsed,
participants were awakened by an alarm placed next to the bed. At
awakening, participants subjectively rated their sleep on a 100-mm
visual analog scale ranging from 0 (no sleep at all) to 100 (deep,
uninterrupted sleep). Four participants’data were excluded from the
analysis because they were unable to fall asleep (scored 0 on the
visual analog scale). After the awakening, 30 minutes was allowed
for participants to overcome any sleep inertia.
8,16
Participants spent
the same period in the PLA and normal sleep night conditions
(ie, from 1400 hours to 1500 hours) watching a neutral documentary
while reclining in a comfortable armchair. Finally, the running-
based anaerobic sprint test (RAST) started at 1500 hours, followed
by a postexercise (ie, 5 min of passive recovery) blood sampling.
Participants completed the protocol during the postseason
period (April and May). At this time of year, daytime lasted
∼1306 hours (±21 min), because the sun rose at 0546 hours
(±12 min) and set at 1852 hours (±14 min). Participants simulta-
neously completed the nap in separate rooms, and they were
instructed to perform the same passive activities on sleep depriva-
tion nights. Laboratory conditions were set at temperature ∼25°C
(±1.8°C), humidity ∼35% (±3.2%), and luminosity (1) ∼2000 lux
during tests and (2) <5 lux during sleep.
Protocols
The Running-Based Anaerobic Sprint Test. The RAST was
developed by Draper and Whyte.
36
Its performance was correlated
to the Wingate test, and it can predict short-distance performance.
37
It is composed of 6 maximal all-out repeated sprints of ∼6to7
seconds interspersed with 10-second recovery periods, which is
similar to the effort during judo combat.
38
The RAST protocol
was followed according to the guidelines developed by the Uni-
versity of Wolverhampton, United Kingdom, and as described by
Romdhani et al.
6,16
Rating of Perceived Exertion. The CR-10 psychophysiological
scale allowed experimenters to assess the exertion experienced by
the athletes during the RAST.
39
Blood Sampling. Blood samples were collected and analyzed as
described by Romdhani et al.
6,16
All methods used in the sample
analysis are presented in Table 1.
Statistical Analyses. The statistical tests were processed in
GraphPad Prism 6 (GraphPad Software, San Diego, CA). All
values in the text, figures, and tables are reported as mean (SD).
The Shapiro–Wilks test revealed that data were normally distrib-
uted, and therefore parametric tests were used. For subjective sleep
quality, RAST performances, and [La], a 1-way analysis of vari-
ance (ANOVA) with repeated measures (5 conditions) was used.
For the remaining biochemical parameters, a 2-way ANOVA with
repeated measures was used (5 conditions ×2“timing”before and
after exercise). To assess the ANOVA practical significance, partial
eta squared (η2
p) was calculated. When the ANOVA indicated a
significant main effect or an interaction, the Bonferroni post hoc
test was used. In addition, the effect size (d) was calculated for
pairwise comparisons according to Cohen.
40
The magnitude of dwas
classified under the following thresholds: small (0.2 ≤d≥0.5),
moderate (0.5 ≤d≤0.8), and large (d≥0.8).
40
Further, mean differ-
ence (MD) and the 95% confidence interval (95% CI) were provided
for pairwise comparison. The level of significance was set at P<.05.
Results
Subjective Sleep Quality During the Nap
The 1-way ANOVA showed a significant main effect of condition
on subjective sleep quality during the nap (F
2,7
= 41.1; P<.001;
η2
p=.83). There was no significant difference between N20 (35.67
Table 1 Different Methods Used in Blood Analysis
Biochemical
variables Method
[La] Lactate oxidase peroxidase method (intraassay and interassay CVs were 0.9% and 1.9%, respectively)
GLC Glucose hexokinase method (intraassay and interassay CVs were 0.94% and 1.3%, respectively)
AST Kinetic method at 340 nm
CK Kinetic method at 340 nm
LDH NADH consumption method (intraassay and interassay CVs were 3.7% and 4.31%, respectively)
UA Colorimetric endpoint (intraassay and interassay CVs were 1.55% and 1.76%, respectively)
GPx Spectrophotometric method based on Paglia and Valentine method (1967; with kit from Randox Lab [Randox Laboratories
Ltd., Crumlin County Antrim, UK]; Ransel RS. 505). The intraassay and interassay CVs were 7.3% and 4.86%, respectively.
SOD SOD activity in erythrocytes was measured by the rate of inhibition of 2-(4-iodophenyl)-3-(4-nitrophenol)-5-phenyltetrazolium
chloride reduction. The kit used in this method was from Randox Lab (Ransod, RX MONZA). About 0.5 mL of whole blood was
centrifuged and then separated from the plasma. Erythrocytes were washed 4 times with 3 mL of 0.9% NaCl solution and centrifuged
after each wash. Then, 2.0 mL with cold redistilled water was added to the resulting erythrocytes, mixed, and left to stand at +4°C for
15 min. A 25-fold dilution of lysate was then added. The intraassay and interassay CVs were 5.96% and 4.64%, respectively.
Abbreviations: AST, aspartate aminotransferase; CK, creatine kinase; CV, coefficient of variation; GLC, plasma glucose; GPx, glutathione peroxidase; [La], plasma lactate;
LDH, lactate dehydrogenase; SOD, superoxide dismutase; UA, uric acid.
(Ahead of Print)
Caffeine and Napping for Short-Term Performance 3

[14.21]) and CAF+N (28.97 [9.24]) in the subjective sleep scores
(P= .667). However, sleep scores for N20 and CAF+N were
statistically higher than those for PLA and CAF (P<.001 for all).
RAST Performances
There was a significant main effect on P
max
(F
2,7
= 35.5; P<.001;
η2
p=.91), P
mean
(F
2,7
=195.6; P<.001; η2
p=.97), and P
min
(F
2,7
=46.4; P<.001; η2
p=.93). P
max
increased after N20 (P<.01,
d=0.75, MD = −82.4 W; 95% confidence interval [CI], −177.3 to
12.5 W) and CAF+N (P<.001, d=1.5, MD=−101.1 W; 95% CI,
−163.7 to −38.6 W) compared with PLA (Figure 2A). Furthermore,
P
max
increased after CAF+N compared with N20 (P<.001, d=0.89,
MD = −82.4 W; 95% CI, −259to24.8W)andCAF(P<.001,
d= 0.94, MD = −129.3 W; 95% CI, −302.2 to −42.4 W). P
min
increased after CAF+N compared with PLA (P<.001, d=2.79,
MD = −111.3 W; 95% CI, −185 to −37.6 W), N20 (P<.01,
d= 1.31, MD = −69.1 W; 95% CI, −123.9 to 5.6 W), and CAF
(P<.001, d=2.6, MD=−125.8 W; 95% CI, −181 to −70.5 W).
P
mean
increased (Figure 2B) after CAF+N compared with PLA
(P<.001, d=1.93, MD=−89.5 W; 95% CI, −165.1 to −13.9 W),
N20 (P<.001, d=1.2,MD=−75.6 W; 95% CI, −132.5 to 17.1 W),
and CAF (P<.001, d=1.79, MD=−95 W; 95% CI, −112 to
−77.9 W) (Figure 2C). Rating of perceived exertion and Fatigue
index were unaffected by neither condition.
Plasma Lactate
There was a significant main effect on plasma lactate ([La];
F
2,7
=32.03; P<.001; η2
p=.92). [La] increased after N20 (P= .008,
d=1.48, MD = −1.15 mmol·L
−1
; 95% CI, −2.46 to 0.15 mmol·L
−1
)
and CAF+N (P<.001, d= 2.53, MD =−3.16 mmol·L
−1
;
95% CI, −4.9 to −1.3 mmol·L
−1
) compared with PLA (Figure 2D).
Biomarkers of Muscle and Hepatic Damage
and Glucose
ANOVA revealed a significant interaction between exercise
and condition on aspartate aminotransferase (AST; F
4,32
=8.75;P<
.001; η2
p=.68), creatine kinase (CK; F
4,32
= 11.53; P= .009; η2
p=
.63), lactate dehydrogenase (LDH; F
4,32
= 9.73; P= .011; η2
p=.61),
and plasma glucose (GLC; F
4,32
= 13.77; P= .004; η2
p=.79). Postex-
ercise AST increased after all sessions (Figure 3A), with lower levels
after N20 compared with PSD (P<.001, d= 2.14, MD = 18.22 U·L
−1
;
95% CI, 7.4 to 29.04 U·L
−1
). Postexercise CK (Figure 3B)increased
only after CAF (P<.001, d= 1.19, MD =−153.8 U·L
−1
; 95% CI,
−222.5 to −85.05 U·L
−1
) and CAF+N (P<.001, d=1.36,
MD = −103.7 U·L
−1
; 95% CI, −172.4 to −34.94 U·L
−1
). Postexercise
LDH increased after CAF+N (P<.001, d= 1.72, MD = −126.4
UI·L
−1
; 95% CI, −199.2 to −53.7 UI·L
−1
) compared with rest
(Figure 3C). Postexercise GLC increased whenever caffeine was
added (Figure 3D), whether after CAF (P<.001, d=1.59,
MD = −1.39 mmol·L
−1
; 95% CI, −2.2 to −0.5 mmol·L
−1
)orafter
CAF+N (P<.001, d= 2.86, MD =−2.3 mmol·L
−1
; 95% CI, −3.1 to
−1.5 mmol·L
−1
).
Enzymatic and Nonenzymatic Antioxidants
There was a significant interaction between exercise and condition
on superoxide dismutase (SOD; F
4,32
= 8.75; P<.001; η2
p=.68),
Figure 2 —Individual maximum (P
max
; A), minimum (P
min
; B), and mean (P
mean
; C) powers and [La] concentrations (D) after different protocol
sessions. CAF □indicates 5 mg·kg
−1
of caffeine; CAF+N ▪, 5 mg·kg
−1
of caffeine + 20-minute nap; N20 •, 20-minute nap with placebo ingestion;
NSN ▴, baseline normal sleep night; PLA ○, partial sleep deprivation with placebo ingestion; [La], plasma lactate; PSD, partial sleep deprivation.
+,++,+++
Significant difference in comparison with NSN session at P<.05, P<.01, and P<.001, respectively; **
,
***Significant difference in comparison
with PSD session at P<.01 and P<.001, respectively;
##,###
Significant difference in comparison with N20 values at P<.01 and P<.001, respectively;
ααα
Significant difference in comparison with CAF values at P<.001. Bars represent the group means and SD.
(Ahead of Print)
4Romdhani et al

glutathione peroxidase (F
4,32
= 4.89; P<.01; η2
p=.45), and uric
acid (UA; F
4,32
= 11.53; P= .009; η2
p=.63). SOD increased after
N20 (P<.001, d= 1.68, MD = −310.9 unit per gram of hemoglobin
(U·g HB
−1
); 95% CI, −563.2 to −58.6 U·g HB
−1
) compared with
preexercise levels (Figure 4A). However, postexercise SOD
decreased after CAF session (P<.01, d= 0.77, MD = 294.9 U·g
HB
−1
; 95% CI, 42.6 to 547.2 U·g HB
−1
) compared with resting
values. Glutathione peroxidase (Figure 4B) increased after N20
compared with resting values (P<.01, d= 1.06, MD = 14.1 U·g
HB
−1
; 95% CI, −26.8 to −1.4 U·g HB
−1
) and compared with CAF
(P<.05, d= 0.65, MD = 13.7 U·g HB
−1
; 95% CI, 0.2 to 27.3 U·g
HB
−1
). Postexercise nonenzymatic antioxidant UA increased
whenever participants napped (Figure 4C), that is, after N20
(P<.01, d= 2.19, MD = −54.6 mmol·L
−1
; 95% CI, −93.9 to −15.4
mmol·L
−1
) and CAF+N (P<.001, d= 2.5, MD = −71.8 mmol·L
−1
;
95% CI, −111.1 to −32.6 mmol·L
−1
), and deceased after CAF
(P<.001, d= 2.96, MD = 72.8 mmol·L
−1
; 95% CI, 33.6 to
112.1 mmol·L
−1
).
Figure 3 —AST (A), CK (B), LDH (C), and GLC (D) before ▪and after ▪the exercise during different protocol sessions. AST indicates aspartate
amino transferase; CAF, 5 mg·kg
−1
of caffeine; CAF+N, 5 mg·kg
−1
of caffeine + 20-minute nap; CK, creatine kinase; LDH, lactate dehydrogenase; GLC,
plasma glucose; N20, 20-minute nap with placebo ingestion; NSN, baseline normal sleep night; PLA, partial sleep deprivation with placebo ingestion;
PSD, partial sleep deprivation;
•
,
••
,
•••
Significant effect of exercise at P<.05, P<.01, and P<.001, respectively; *
,
**
,
***Significant difference in
comparison with PSD at P<.05, P<.01, and P<.001, respectively;
#,##,###
Significant difference in comparison with N20 at P<.05, P<.01, and P<.001,
respectively;
α,αα,ααα
Significant difference in comparison with CAF at P<.05, P<.01, and P<.001, respectively. Data are presented as Mean (SD).
Figure 4 —SOD (A), GPx (B), and UA (C) during different protocol sessions. CAF indicates 5 mg·kg
−1
of caffeine; CAF+N, 5 mg·kg
−1
of
caffeine + 20-minute nap; GPx, glutathione peroxidase; N20, 20-minute nap with placebo ingestion; NSN, baseline normal sleep night; PLA, partial sleep
deprivation with placebo ingestion; SOD, superoxide dismutase; UA, uric acid;
•
,
••
,
•••
Significant effect of exercise at P<.05, P<.01, and
P<.001, respectively; *
,
**
,
***Significant difference in comparison with PSD at P<.05, P<.01, and P<.001, respectively;
#,##,###
Significant
difference in comparison with N20 at P<.05, P<.01, and P<.001, respectively;
α,αα,ααα
Significant difference in comparison with CAF at P<.05,
P<.01, and P<.001, respectively.
(Ahead of Print)
Caffeine and Napping for Short-Term Performance 5

Discussion
The current investigation studied the effects of a 20-minute nap
(N20), the ingestion of 5 mg·kg
−1
of caffeine (CAF), and their
combination (CAF+N) on repeated sprint performance (RAST)
and associated biochemical responses in PSD athletes. The main
findings were that the combination of CAF+N resulted in the
highest RAST performances compared with PLA, CAF, and
N20. Interestingly, CAF increased muscle damage and decreased
antioxidant defense without improving performance. Analyzing
the effect of PSD compared with normal sleep is beyond the scope
of this study, and it is discussed elsewhere.
6
The short nap increased P
max
slightly but significantly, con-
firming the findings of earlier studies that reported improved all-out
8
and repeated
16,22
sprint and endurance
4,20
performance after a brief
nap. Also, the current results confirm literature findings about the
limited effects of caffeine on short-term high-intensity exer-
cise.
25,27,28
Although the ergogenic effect of caffeine is higher in
caffeine–naïve participants,
25
the current results showed that
repeated sprint after CAF was no better than after PLA. Because
the effect of caffeine is dose dependent,
25
it could be possible that the
dosage of caffeine used was not high enough to improve repeated
sprint performance in sleep-deprived participants. To the best of the
authors’knowledge, the current study is the first to compare the
effect of a short nap with a moderate dose of caffeine on physical
performance outcomes. N20 was more effective than CAF at least
for P
max
, which generally corresponds to the first sprint of the
RAST. It is possible that napping is more efficient than caffeine at
offsetting the PLD sleep drive buildup and improving repeated
sprint performance of sleepy athletes. Indeed, it has been reported
that the wakefulness-promoting effects of caffeine do not replace the
restorative effects of sleep to the same extent.
41
The combination CAF+N was far more beneficial for repeated
sprint than CAF or N20 conditions alone. Likewise, previous
literature has shown that the combination of caffeine with a short
nap was clearly superior to either caffeine alone or napping alone
in maintaining alertness
13–15
and driving stimulator tasks
14,15
and
memory search tasks.
13
Carrier et al
42
reported that caffeine
ingestion before daytime sleep reduced the duration and the
intensity of slow wave sleep. In addition, caffeine before a PLD
nap blunted sleep inertia and promoted the benefits of napping.
13
Mechanistically, caffeine acts as a nonselective, competitive
antagonist to adenosine receptors.
33
Adenosine is a sleep-pro-
moting substance that increases during wakefulness and modi-
fies the neuronal activities in key brain areas that regulate sleep
homeostasis.
43
Caffeine exerts its alerting effects by binding to
adenosine receptors in the central nervous system and antago-
nizing adenosine’s effects. In addition, napping depresses the
homeostatic buildup of sleep drive (ie, decreases adenosine
concentration). It has been reported that sleep pressure dissipates
at a faster rate during sleep than its accumulation during wake-
fulness.
12
Therefore, it is possible that the double challenge to
adenosine system (ie, the combined effects of napping and
caffeine ingestion) dampened the sleep loss–induced homeo-
static component of sleep drive buildup (process S) more than
each intervention alone.
Plasma lactate showed a similar trend to RAST performance
(Figure 2D). Compared with PLA, postexercise [La] decreased
after CAF sessions, increased slightly after N20, and increased
notably after CAF+N. Thus, [La] increased relatively to exercise
load and was not influenced by caffeine ingestion per se. However,
postexercise GLC levels increased after caffeine ingestion with or
without napping (Figure 3D). Indeed, it has been shown that
caffeine intake was associated with increased catecholamine levels,
which increase glycogenolysis and thus GLC.
31
In this regard, it
has been consistently reported that caffeine ingestion increases
energy availability and energy expenditure during exercise.
43
Nevertheless, despite the greater energy availability after caffeine
ingestion, repeated sprint performances were not any better than
those after PLA.
Postexercise CK increased whenever caffeine was added, with
or without napping, even when performance declined. Therefore,
the increased CK could be attributed to caffeine ingestion. Accord-
ingly, earlier studies have concluded that caffeine intake may
increase the risk of muscle damage in athletes.
31,44
In that regard,
Tauler et al
31
reported an increased inflammatory response to
exercise through higher interleukin-6 and interleukin-10 after
the ingestion of 6 mg·kg
–1
of caffeine. In addition, Bassini-Ca-
meron et al
44
reported that AST, LDH, and CK increased in
response to exercise after the ingestion of a similar dosage of
caffeine (ie, 5 mg·kg
–1
) compared with PLA. However, in the
present study, the increase in LDH and AST was not caused by
caffeine ingestion. LDH increased concomitantly with the increase
in exercise performance. Also, AST did not increase further after
caffeine ingestion. However, CK increased only when caffeine was
added regardless of napping or exercise performance. With this in
mind, caffeine before repeated sprints should be ingested with
caution.
The present findings showed a depressed antioxidative
capacity through lower SOD and UA after CAF. During and
after exercise, the generation of reactive oxygen species in-
creases dramatically due to the high metabolic rate and the
increased oxygen uptake. This break in the balance between
the generation and removal of reactive oxygen species creates a
situation known as oxidative stress. Without adequate antioxi-
dant defense, protein, lipid, and nucleic acid may be oxidized,
which causes degenerative diseases. SOD is the first defensive
line against oxidative stress. Also, UA represents more than
50% of plasmatic antioxidant capacity.
45
Interestingly, napping
reversed the prooxidant effect of caffeine. In fact, enzymatic
(SOD) and nonenzymatic (UA) antioxidants increased after
napping with or without caffeine ingestion. This finding indi-
cates, for the first time, that napping may produce a good defense
to exercise-induced oxidative damage, even with the consumption of
a prooxidant (ie, caffeine). Therefore, the positive effects of napping
not only improve exercise performance but also reduce the negative
effects of oxidative stress.
The current study was designed to mimic real-life situations in
which athletes and coaches do not perform objective sleep mea-
surement. After suboptimal nocturnal sleep or PSD, a moderate
dose of caffeine before a 20-minute nap in the PLD was beneficial
for repeated sprint performance. However, participants in the
present study were caffeine naïve. Thus, the current findings apply
only to habitually low caffeine consumers (<80 mg·d
–1
). Further-
more, sleep quality during the nap was measured only subjectively,
which could limit the present findings. Of note, caffeine ingested
shortly before a daytime nap did not lower sleep quality, at least
from a subjective perspective. We could not confirm whether
participants’sleep during the nap was objectively unaffected by
caffeine ingestion, especially in caffeine-naïve participants. In
addition, we studied only the effects of a single dose of caffeine
and only 1 nap opportunity. It is possible that higher doses of
caffeine or longer naps result in better performance benefits. Finally,
(Ahead of Print)
6Romdhani et al

in the present study, sleep deprivation was caused by early awaken-
ing (toward the end of the night). The effects of such a combination
on performance after PSD caused by late bedtime are unknown.
Therefore, further investigations are warranted.
Practical Applications
It is common for athletes to start a competition or training session
with sleep debt. The combination of caffeine ingestion and a short
afternoon nap could improve repeated sprint performance and
antioxidant mechanisms and reduce muscle damage. Nevertheless,
we advise athletes to be aware of the effects of caffeine ingestion
after PSD and before repeated sprint exercise, because it results in
increased muscle damage and would probably affect recovery
without improving performance.
Conclusions
A short nap is beneficial for partially sleep-deprived, caffeine-naïve
athletes. Not only are repeated sprint performances improved, but
also exercise-induced oxidative damage is reduced compared with
the same exercise performed without napping. Postexercise muscle
damage with or without napping was greater after caffeine ingestion,
and performance was not improved. Nevertheless, when caffeine
was consumed before a shortnap, repeated sprint performances were
better than after caffeine ingestion or a short nap alone.
Acknowledgments
The authors express sincere gratitude to the participants, Mr Firas Ben Slama,
and Dr Olfa Abene for their time and effort throughout the experiment.
References
1. Mejri MA, Hammouda O, Zouaoui K, et al. Effect of two types of
partial sleep deprivation on Taekwondo players’performance during
intermittent exercise. Biol Rhythm Res. 2014;45(1):17–26. doi:10.
1080/09291016.2013.787686
2. Romdhani M, Hammouda O, Smari K, et al. Total sleep deprivation
and recovery sleep affect the diurnal variation of agility performance:
the gender differences [published online ahead of print May 30,
2018]. J Strength Cond Res. doi:10.1519/jsc.0000000000002614
3. Mougin F, Bourdin H, Simon-Rigaud ML, Nguyen NU, Kantelip JP,
Davenne D. Hormonal responses to exercise after partial sleep
deprivation and after a hypnotic drug-induced sleep. J Sports Sci.
2001;19(2):89–97. doi:10.1080/026404101300036253
4. Daaloul H, Souissi N, Davenne D. Effects of napping on alertness,
cognitive, and physical outcomes of karate athletes. Med Sci Sports
Exerc. 2019;51(2):338–345. PubMed ID: 30239491 doi:10.1249/
MSS.0000000000001786
5. Souissi N, Chtourou H, Aloui A, et al. Effects of time-of-day and
partial sleep deprivation on short-term maximal performances of judo
competitors. J Strength Cond Res. 2013;27(9):2473–2480. PubMed
ID: 23974210 doi:10.1519/JSC.0b013e31827f4792
6. Romdhani M, Hammouda O, Chaabouni Y, et al. Sleep deprivation
affects post-lunch dip performances, biomarkers of muscle damage
and antioxidant status. Biol Sport. 2019;36(1):55–65. doi:10.5114/
biolsport.2018.78907
7. Borbely AA. Two process model of sleep regulation. Hum Neurobiol.
1982;1(3):195–204. PubMed ID: 7185792
8. Waterhouse J, Atkinson G, Edwards B, Reilly T. The role of a
short post-lunch nap in improving cognitive, motor, and sprint
performance in participants with partial sleep deprivation. J Sports
Sci. 2007;25(14):1557–1566. PubMed ID: 17852691 doi:10.1080/
02640410701244983
9. Monk TH. The post-lunch dip in performance. Clin Sports Med.
2005;24(2):e15–e23. doi:10.1016/j.csm.2004.12.002
10. Rae DE, Chin T, Dikgomo K, et al. One night of partial sleep
deprivation impairs recovery from a single exercise training session.
Eur J Appl Physiol. 2017;117(4):699–712. PubMed ID: 28247026
doi:10.1007/s00421-017-3565-5
11. Juliff LE, Halson SL, Peiffer JJ. Understanding sleep disturbance
in athletes prior to important competitions. JSciMedSport. 2015;
18(1):13–18. PubMed ID: 24629327 doi:10.1016/j.jsams.2014.02.007
12. Achermann P, Dijk DJ, Brunner DP, Borbély AA. A model of human
sleep homeostasis based on EEG slow-wave activity: quantitative
comparison of data and simulations. Brain Res Bull. 1993;31(1–2):97–
113. PubMed ID: 8453498 doi:10.1016/0361-9230(93)90016-5
13. Hayashi M, Masuda A, Hori T. The alerting effects of caffeine, bright
light and face washing after a short daytime nap. Clin Neurophysiol.
2003;114(12):2268–2278. PubMed ID: 14652086 doi:10.1016/
S1388-2457(03)00255-4
14. Reyner LA, Horne JA. Suppression of sleepiness in drivers; combination
of caffeine with a short nap. Psychophysiology. 1997;34(6):721–725.
PubMed ID: 9401427 doi:10.1111/j.1469-8986.1997.tb02148.x.
15. Horne JA, Reyner LA. Counteracting driver sleepiness: effects of
napping, caffeine, and placebo. Psychophysiology. 1996;33(3):306–
309. PubMed ID: 8936399 doi:10.1111/j.1469-8986.1996.tb00428.x
16. Romdhani M, Souissi N, Chaabouni Y, et al. Improved physical
performance and decreased muscular and oxidative damage with
postlunch napping after partial sleep deprivation in athletes [pub-
lished online ahead of print February 4, 2020]. Int J Sports Physiol
Perform. 15(6):874–889. doi:10.1123/ijspp.2019-0308
17. Milner CE, Cote KA. Benefits of napping in healthy adults: impact of
nap length, time of day, age, and experience with napping. J Sleep Res.
2009;18(2):272–281. PubMed ID: 19645971 doi:10.1111/j.1365-2869.
2008.00718.x
18. Faraut B, Nakib S, Drogou C, et al. Napping reverses the salivary
interleukin-6 and urinary norepinephrine changes induced by sleep
restriction. J Clin Endocrinol Metab. 2015;100(3):E416–E426.
PubMed ID: 25668196 doi:10.1210/jc.2014-2566
19. Hilditch CJ, Dorrian J, Banks S. A review of short naps and sleep
inertia: do naps of 30 min or less really avoid sleep inertia and slow
wave sleep? Sleep Med. 2017;32:176–190. PubMed ID: 28366332
doi:10.1016/j.sleep.2016.12.016
20. Blanchfield AW, Lewis-Jones TM, Wignall JR, Roberts JB, Oliver
SJ. The influence of an afternoon nap on the endurance performance
of trained runners. Eur J Sport Sci. 2018;18(9):1177–1184. PubMed
ID: 29851569 doi:10.1080/17461391.2018.1477180
21. Hammouda O, Romdhani M, Chaabouni Y, Mahdouani K, Driss T,
Souissi N. Diurnal napping after partial sleep deprivation affected
hematological and biochemical responses during repeated sprint.
Biol Rhythm Res. 2018;49(6):927–939. doi:10.1080/09291016.2018.
1429553
22. Boukhris O, Abdessalem R, Ammar A, et al. Nap opportunity
during the daytime affects performance and perceived exertion in
5-m shuttle run test. Front Physiol. 2019;10:1–8. doi:10.3389/fphys.
2019.00779
23. Suppiah HT, Yong LC, Choong G, Chia M. Effects of a short daytime
nap on shooting and sprint performance in high-level adolescent
athletes. Int J Sports Physiol Perform. 14(1):76–82. doi:10.1123/
ijspp.2018-0107
(Ahead of Print)
Caffeine and Napping for Short-Term Performance 7

24. Vgontzas AN, Pejovic S, Zoumakis E, et al. Daytime napping after a
night of sleep loss decreases sleepiness, improves performance, and
causes beneficial changes in cortisol and interleukin-6 secretion. Am J
Physiol Endocrinol Metab. 2006;292(1):E253–E261. PubMed ID:
16940468 doi:10.1152/ajpendo.00651.2005
25. McLellan TM, Caldwell JA, Lieberman HR. A review of caffeine’s
effects on cognitive, physical and occupational performance. Neurosci
Biobehav Rev. 2016;71:294–312. doi:10.1016/j.neubiorev.2016.09.001
26. Lee CL, Cheng CF, Lin JC, Huang HW. Caffeine’s effect on
intermittent sprint cycling performance with different rest intervals.
Eur J Appl Physiol. 2012;112(6):2107–2116. PubMed ID: 21960086
doi:10.1007/s00421-011-2181-z
27. Paton CD, Hopkins WG, Vollebregt L. Little effect of caffeine
ingestion on repeated sprints in team-sport athletes. Med Sci Sports
Exerc. 2001;33(5):822–825. PubMed ID: 11323555 doi:10.1097/
00005768-200105000-00023
28. Crowe MJ, Leicht AS, Spinks WL. Physiological and cognitive
responses to caffeine during repeated, high-intensity exercise. Int J Sport
Nutr Exercise Metab. 2006;16(5):528–544. PubMed ID: 17240784
doi:10.1123/ijsnem.16.5.528
29. Alavi Y, Mirdar S, Rngrazan MD. Effect of different caffeine doses
on exercise-induced oxidative stress in active men. Int J Sci Basic
Appl Res. 2015;4(11):667–672.
30. Olcina GJ, Mu ˜noz D, Timo´n R, et al. Effect of caffeine on oxidative
stress during maximum incremental exercise. J Sports Sci Med.
2006;5(4):621–628. PubMed ID: 24357958
31. Tauler P, Martínez S, Moreno C, Monjo M, Martínez P, Aguilo´A.
Effects of caffeine on the inflammatory response induced by a 15-km
run competition. Med Sci Sports Exerc. 2013;45(7):1269–1276.
PubMed ID: 23299767 doi:10.1249/MSS.0b013e3182857c8a
32. Marco Machado, Paulo Vinícios C, Zovico D, et al. Caffeine does not
increase resistance exercise-induced microdamage. J Exerc Sci Fit.
2008;6(2):115–120.
33. Fredholm BB, Bättig K, Holmén J, Nehlig A, Zvartau EE. Actions of
caffeine in the brain with special reference to factors that contribute to
its widespread use. Pharmacol Rev. 1999;51(1):83–133. PubMed ID:
10049999
34. Horne JA, Ostberg O. A self-assessment questionnaire to determine
morningness-eveningness in human circadian rhythms. Int J Chron-
obiol. 1976;4(2):97–110. PubMed ID: 1027738
35. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The
Pittsburgh Sleep Quality Index: a new instrument for psychiatric
practice and research. Psychiatry Res. 1989;28(2):193–213. PubMed
ID: 2748771 doi:10.1016/0165-1781(89)90047-4
36. Draper PN, Whyte G. Anaerobic performance testing. Peak Perform.
1997;96:3–5.
37. Zagatto AM, Beck WR, Gobatto CA. Validity of the running
anaerobic sprint test for assessing anaerobic power and predicting
short-distance performances. J Strength Cond Res. 2009;23(6):1820–
1827. PubMed ID: 19675478 doi:10.1519/JSC.0b013e3181b3df32
38. Franchini E, Del Vecchio FB, Matsushigue KA, Artioli GG. Physiolog-
ical profiles of elite judo athletes. Sports Med. 2011;41(2):147–166.
PubMed ID: 21244106 doi:10.2165/11538580-000000000-00000
39. Foster C. Monitoring training in athletes with reference to over-
training syndrome. Med Sci Sports Exerc. 1998;30(7):1164–1168.
PubMed ID: 9662690 doi:10.1097/00005768-199807000-00023.
40. Cohen J. A power primer. Psychological Bulletin. 1992;112(1):155–
159. PubMed ID: 19565683 doi:10.1037/0033-2909.112.1.155
41. Wyatt JK, Cajochen C, Ritz-De Cecco A, Czeisler CA, Dijk DJ. Low-
dose repeated caffeine administration for circadian-phase-dependent
performance degradation during extended wakefulness. Sleep. 2004;
27(3):374–381. PubMed ID: 15164887 doi:10.1093/sleep/27.3.374.
42. Carrier J, Paquet J, Fernandez-Bolanos M, et al. Effects of caffeine on
daytime recovery sleep: a double challenge to the sleep-wake cycle in
aging. Sleep Med. 2009;10(9):1016–1024. PubMed ID: 19342294
doi:10.1016/j.sleep.2009.01.001
43. Urry, E, & Landolt, HP. Adenosine, caffeine, and performance: from
cognitive neuroscience of sleep to sleep pharmacogenetics. In: Sleep,
Neuronal Plasticity and Brain Function. Berlin, Heidelberg:
Springer; 2014:331–366. doi:10.1007/7854_2014_274
44. Bassini-Cameron A, Sweet E, Bottino A, Bittar C, Veiga C, Cameron
LC. Effect of caffeine supplementation on haematological and
biochemical variables in elite soccer players under physical stress
conditions. Br J Sports Med. 2007;41(8):523–530. doi:10.1136/
bjsm.2007.035147
45. Wayner DDM, Burton GW, Ingold KU, Barclay LRC, Locke SJ. The
relative contributions of vitamin E, urate, ascorbate and proteins to the
total peroxyl radical-trapping antioxidant activity of human blood
plasma. Biochim Biophys Acta. 1987;924(3):408–419. doi:10.1016/
0304-4165(87)90155-3
(Ahead of Print)
8Romdhani et al