Impact of Caffeine Intake on 800-m Running Performance and Sleep Quality in Trained Runners

nd: Ca eine ingestion improves athletic performance, but impairs sleep quality. We aimed to analyze the e ect of ca eine intake on 800-m running performance, sleep quality (SQ), and nocturnal cardiac autonomic activity (CAA) in trained runners. Methods: Fifteen male middle-distance runners participated in the study (aged 23.7 8.2 years). In a randomized and comparative crossover study design, the athletes ingested a placebo (PL) or ca eine supplement (CAF; 6 mg kg 1 ) one hour before an 800-m running time-trial test in the evening. During the night, CAA and SQ were assessed using actigraphy and a sleep questionnaire. A second 800-m running test was performed 24 h after the rst. Time, heart rate, rating of perceived exertion, and blood lactate concentration were analyzed for each running test. Results: No signi cant di erences in CAA and performance variables were found between the two conditions. However, CAF impaired sleep e ciency (p=0.003), actual wake time (p=0.001), and the number of awakenings (p=0.005), as measured by actigraphy. Also, CAF impaired the questionnaire variables of SQ (p=0.005), calm sleep (p=0.005), ease of falling asleep (p=0.003), and feeling refreshed after waking (p=0.006). Conclusion: The supplementation with ca eine (6 mg kg 1 ) did not improve the 800-m running performance, but did impair the SQ of trained runners. Keywords:actigraphy; athletic; co ee; ergogenic aid; supplement 1. Introduction Scientists and coaches are continually looking for techniques to develop more e ective and e cient methods to improve exercise performance [1]. One of the popular methods commonly used by athletes to

mg kg 1 ) did not improve the 800-m running performance, but did impair the SQ of trained runners. Keywords:actigraphy; athletic; co ee; ergogenic aid; supplement 1. Introduction Scientists and coaches are continually looking for techniques to develop more e ective and e cient methods to improve exercise performance [1]. One of the popular methods commonly used by athletes to maximize their physical performance is the intake of legal ergogenic aids [2]. In this way, ca eine is frequently used in sport as an ergogenic aid to improve athletic performance and endurance [3]. In fact, it has been reported that 74% of elite athletes may use ca eine as an ergogenic aid prior to or during a competition [4]. Ca eine is a xanthine alkaloid that increases central nervous activity by the blockade of central and peripheral adenosine receptors [5]. This stimulant action produces a greater recruitment of motor units [6], improves the Na + –K + pump response [7], and increases the rate of calcium release from the sarcoplasmic reticulum [8] and the mobilization of free fatty acids [9]. Also, ca eine enhances adrenaline secretion [10] and reduces ratings of perceived exertion [11]. Therefore, ca eine is administered in order to improve sport performance. Previous studies that analyzed the e ect of ca eine ingestion on runners have shown improvements in running performance compared to placebo [12,13]. It had previously been reported that compared to placebo, the intake of 4.5 mg kg 1 of ca eine increased exercise distance by 2–3 km when running at 85% maximum oxygen uptake until exhaustion [10]. Regarding middle-distance races, compared to Nutrients2019,11, 2040; doi:10.3390 /nu11092040 /journal/nutrients

Nutrients2019,11, 2040 2 of 10 placebo, 1500-m [13] or one-mile [14] running performances are improved by 1.3–1.9% after 150–200 mg and 3 mg kg 1 of ca eine intake, respectively. However, another study found similar 800-m running performance in amateur runners after placebo or 5.5 mg kg 1 of ca eine administration [15]. Thus, there is con icting evidence in relation to the e ectiveness of ca eine as an ergogenic aid to improve middle-distance race performance in athletes. On the other hand, ca eine intake can impair sleep [16], which is considered the most important method for recovery from daily load [17]. Sleep assists in the recovery of the nervous and metabolic cost imposed by the waking state [18]. However, ca eine typically prolongs sleep latency, reduces total sleep time and sleep e ciency, and worsens perceived sleep quality (SQ) [16], particularly if it is administered close to bedtime. Moreover, vigorous-intensity exercise completed close to bedtime increases the latency time and impairs SQ [19]. Therefore, the use of ca eine as an ergogenic aid in a competition performed close to bedtime may decrease SQ and the recovery process, which may decrease athlete performance on the following day. There are some sports modalities, such as athletics, where the athlete needs to perform in quali cation races over consecutive days. Some of these races are performed at the end of the evening, and the rest time between the rst race (e.g., a semi- nal) and the following one (e.g., the nal) may be very short. For example, during the Athletics World Championships of 2019, the quali cation and the semi- nal race of the 800-m event were separated by 24 h. Thus, the administration of ca eine before a quali cation race performed in the evening may a ect the recovery process and performance in the races on the following day due to sleeping problems. However, there are no studies that have analyzed the e ect of ca eine administration to aid performance in a race close to bedtime on SQ and on the running performance the following day. Therefore, the

race performed in the evening may a ect the recovery process and performance in the races on the following day due to sleeping problems. However, there are no studies that have analyzed the e ect of ca eine administration to aid performance in a race close to bedtime on SQ and on the running performance the following day. Therefore, the aim of the present study was to analyze the e ect of ca eine intake one hour (19:00 h) before an 800-m race (20:00 h) on actigraphic SQ, subjective SQ, and nocturnal cardiac autonomic activity (CAA), and on the 800-m performance performed 24 h later in trained middle-distance athletes. We hypothesized that the pre-exercise ingestion of 6 mg of ca eine per kg of an athletes's body mass would impair SQ through subjective and actigraphic impairment, but it would not a ect the race performance on the following day. 2. Methods 2.1. Design A randomized and comparative crossover study was conducted to test the e ects of ca eine intake or placebo before an 800-m running time trial on actigraphic SQ, the subjective quality of sleep, nocturnal autonomous cardiac activity, countermovement jump (CMJ), and the 800-m performance of athletes at international and national levels. Athletes reported to their usual o cial athletics track four times over two consecutive weeks. The testing sessions were developed during two consecutive Friday and Saturday evenings in March. Two weeks before the study, the athletes had nished their winter season, performing in the National Indoor Championships. Therefore, the study was developed in a general period training phase. Upon arrival at the athletics track, runners were given a ca eine or a placebo supplement—placebo (PL) or ca eine (CAF) in randomized order—in experimental Sessions 1 and 3, while no supplements were taken in the experimental Sessions 2 and 4. Forty- ve minutes (min) after the intake of the supplements in Sessions 1 and 3, or 45 min after the runners arrived at the athletics track, the participants started the testing session. An 800-m running time-trial test was performed in each testing session. Performance (time, CMJ height),

3, while no supplements were taken in the experimental Sessions 2 and 4. Forty- ve minutes (min) after the intake of the supplements in Sessions 1 and 3, or 45 min after the runners arrived at the athletics track, the participants started the testing session. An 800-m running time-trial test was performed in each testing session. Performance (time, CMJ height), physiological (peak and mean heart rate and blood lactate concentration), and subjective (rating of perceived exertion) variables were collected during the testing session. The sessions were carried out at 20:00 h and under similar environmental conditions (20–22 C). In addition, we used actigraphy to monitor the night after PL or CAF ingestion to assess SQ and a sleep questionnaire and to analyze the autonomic modulation.

Nutrients2019,11, 2040 3 of 10 2.2. Participants Fifteen male runners in mid-level events participated in the study (age: 23.7 8.2 years; height: 177.4 9.0 cm; weight: 64.6 9.8 kg). Runners performed 9.0 1.8 h per week of training and had at least six years of middle-distance training experience. They were of national and international standard at the 800-m level and their best time at that distance ranged between 1:46.72–2:04.10. Eleven of the runners were of Caucasian race, two were from North Africa (Maghreb race), one was from South America (Latino race), and another was from Central Africa (Black race). All the subjects gave their signed and informed consent, and the study was approved (CE031909) by the Ethics Committee in Institutional Sciences of the University and was in accordance with the Declaration of Helsinki. The subjects were asked to maintain their usual diet and hydration status and not to ingest ca eine or alcohol at least 24 h before each test session or to carry out exhaustive training in the 48 h prior the rst and third testing sessions. 2.3. Procedures Athletes ingested a placebo (sucrose) or ca eine supplement (6 mg kg 1 ) in capsules of the same size, color, and smell in a typical double-blind trial, with a 50% chance of ingesting the actual active or placebo substance, avoiding any e ects of session or time on the results. The blinding e cacy was checked after the participants had nished their participation. In addition, participants were issued with nutritional guidelines to ensure that they followed a similar diet in the 48 h before each condition session. This diet was the same that runners usually used during competition. The last meal was eaten by runners 3 h before the test. Furthermore, 24 h before each experimental session, ca eine ingestion was restricted. In addition, a ca eine consumption questionnaire [20] was administered to the runners, which showed that all the runners were daily consumers of ca eine (between 250–572 mg of ca eine day 1 ) according to classi cation proposed elsewhere [21]. Also, all the runners were used

test. Furthermore, 24 h before each experimental session, ca eine ingestion was restricted. In addition, a ca eine consumption questionnaire [20] was administered to the runners, which showed that all the runners were daily consumers of ca eine (between 250–572 mg of ca eine day 1 ) according to classi cation proposed elsewhere [21]. Also, all the runners were used to ingesting ca eine (6 mg kg 1 ) as an ergogenic aid prior to competition. 2.4. Testing Session During the rst visit, body composition was evaluated using a bioimpedance segmental analyzer (Tanita BC-601, Tanita Corp, Tokyo, Japan) following previous recommendations [22]. In addition, 45 min after supplement ingestion, participants performed their traditional competitive warm-up of 15 min duration, including running at low intensity, joint mobility, dynamic stretching, and progressive running sets. After warm-up, a CMJ test was carried out. Two minutes later (~60 min after supplement ingestion), an 800-m time-trial test was performed. Finally, 2 min after the end of the running test, a blood lactate concentration analysis and another CMJ test were carried out. The mean and peak heart rates (Polar RS800, Polar Electro Oy, Kempele, Finland) were recorded during the 800-m running time trial. In addition, ratings of perceived exertion (RPE) were determined using the 10-point Borg scale [23] following the 800-m time trial. The 800-m times were recorded using a Geonaute chronometer Onstart 710 (Decathlon, Villeneuve-d'Ascq, France) by two of the researchers, and the mean of these values was used for analysis. Capillary blood samples (5 L) were collected by nger prick 2 min after the end of the running test and analyzed for blood lactate concentration ([La–]) using a Lactate Pro analyzer (Lactate Pro, Arkay, Inc., Kyoto, Japan). Countermovement jump heights were performed using a contact platform (Ergotester, Globus, Codogne, Italy). The participants executed two submaximal trials to ensure proper execution of the jumps with 1-min rest between trials. The CMJ height was measured before warm-up and prior to the 800-m time trial, and performed at the center of the platform with the feet placed shoulder-width apart in the standing position. Participants were asked to

contact platform (Ergotester, Globus, Codogne, Italy). The participants executed two submaximal trials to ensure proper execution of the jumps with 1-min rest between trials. The CMJ height was measured before warm-up and prior to the 800-m time trial, and performed at the center of the platform with the feet placed shoulder-width apart in the standing position. Participants were asked to jump as high as possible with a rapid self-selected countermovement. The depth of the countermovement was self-selected, and participants were asked to try to land close to the take-o point. Each individual's best performed was used for data analysis. The same testing procedure was applied in each testing session.

Nutrients2019,11, 2040 4 of 10 2.5. Actigraphic Quality of Sleep, Subjective Quality of Sleep, and Autonomous Nocturnal Cardiac Activity Between the end of testing session and the time to go to bed, the athletes had to do their normal life and record any activity in a diary. Participants were instructed to measure actigraphic sleep quality and nocturnal cardiac autonomic activity (Heart Rate Variability-HRV) during sleep after each day with a training session day. Actigraphic sleep quality was recorded using an actiwatch activity monitoring system (Cambridge Neurotechnology, Cambridge, UK), which measures activity by means of a piezoelectric accelerometer. The movement of the non-dominant wrist of each participant was monitored. A low actigraphic sensitivity threshold (80 counts per epoch) was selected, and the data recorded by the actigraph were analyzed with Actiwatch Sleep Analysis Software. Each subject received a sleep diary to record bedtime, wake-up time, hours napping, hours without wearing the actigraph, and the number of nocturnal awakenings. Data analysis started with the onset of nocturnal rest (bedtime) and ended with the onset of daytime activity (wake time). The following sleep parameters were measured: (I) sleep e ciency (%): percentage of time spent asleep; (II) time in bed (min); (III) actual sleep time (min); (IV) actual wake time (min); (V) number of awakenings; (VI) average time of each awakening(min); and (VII) latency. Together with the actigraph, during the night, each subject wore an H7 strap Heart monitor (Polar Electro, Kempele, Finland) to evaluate HRV. Variables of cardiac autonomic activity were analyzed for the 4-h period of sleep starting 30 min after the reported bedtime [20]. The R–R series were analyzed using Kubios HRV software (version 2.0, Biosignal Analysis and Medical Imaging Group, University of Kuopio, Finland). The following HRV variables were assessed: (I) low-frequency (LF) band/ high-frequency (HF) band ratio; (II) total power (TP); (III) percentage of di erences between adjacent normal R–R intervals more than 50 ms (pNN50); (IV) square root of the mean of the sum of the squared di erences between adjacent normal R–R intervals (RMSSD); (V) standard deviation of all normal N–N intervals (SDNN); (VI) mean heart

assessed: (I) low-frequency (LF) band/ high-frequency (HF) band ratio; (II) total power (TP); (III) percentage of di erences between adjacent normal R–R intervals more than 50 ms (pNN50); (IV) square root of the mean of the sum of the squared di erences between adjacent normal R–R intervals (RMSSD); (V) standard deviation of all normal N–N intervals (SDNN); (VI) mean heart rate; and (VII) mean R–R intervals. Participants were also instructed to evaluate their subjective sleep quality in the morning after awakening using the Karolinska Sleep Diary [24], which analyzes the following questions: (I) sleep quality (very well [5] to very poorly [1]); (II) calm sleep (very calm [5] to very restless [1]); (III) ease of falling asleep (very easy [5] to very di cult [1]); (IV) amount of dreaming (much [3] to none [1]); (V) ease of waking up (very easy [5] to very di cult [1]); (VI) feeling refreshed after awakening (completely [3] to not at all [1]); (VII) slept throughout the time allotted (yes [5] to woke up much too early [1]). 2.6. Statistical Analysis Statistical analysis of data was performed with SPSS 21.0 software (SPSS 21.0, Chicago, IL, USA) in a Windows environment. Descriptive data are presented as mean SD and range. For inferential analysis, a Shapiro–Wilk W-test was performed to establish the normality of the sampling distribution, and Mauchly's W-test analyzed the sphericity between measurements. In addition, analysis of variance for repeated measures (ANOVA) was calculated (general linear model) to analyze the e ects of ca eine intake on performance over 800 m, and a paired sample T-test or the nonparametric equivalent (Wilcoxon test) was used to compare the e ect of ca eine on heart rate variability and SQ. E ect size (ES) was calculated using partial eta-squared ( 2p) for variance analysis and Cohen's d to indicate the standardized di erence between two means. Threshold values for ES were 0.1 (small), 0.3 (moderate), 1.2 (large), and 2.0 (very large) [25]. The level of signi cance was set at p 0.05. 3. Results Table conditions (placebo and ca eine). No signi cant e ects were

using partial eta-squared ( 2p) for variance analysis and Cohen's d to indicate the standardized di erence between two means. Threshold values for ES were 0.1 (small), 0.3 (moderate), 1.2 (large), and 2.0 (very large) [25]. The level of signi cance was set at p 0.05. 3. Results Table conditions (placebo and ca eine). No signi cant e ects were found in performance (Figure). Signi cant e ects were observed in the variable CMJ (F=4.564;p=0.008) with a large e ect size ( 2p=0.28); the pair comparison showed a signi cant di erence between the CMJ results (D) on days

Nutrients2019,11, 2040 5 of 10 1 and 2 when participants took ca eine (mean di erences= 6.51, t= 3.14,p=0.020). However, no signi cant e ects were found in in any other variable.Nutrients 2019, 11, x FOR PEER REVIEW 5 of 10 Figure 1. Time in 800 m (s). T1: First test 1; T2: second test; PLA: Placebo; CAF: Caffeine. Table 1. Results of 800-m running time trial test variables. Placebo Caffeine ANOVA Test 1 Test 2 Test 1 Test 2 mean SD mean SD mean SD mean SD F p η²p Time in 800 m (s) 122.6 5.6 123.8 6.2 122.3 5.1 123.3 5.4 2.317 0.12 0.15 RPE (A.U) 8.4 1.1 8.2 1.0 8.3 0.9 8.1 0.9 0.142 0.934 0.01 mean HR in 800 m (bpm) 170.4 9.8 171.4 10.1 172.7 10.6 173.2 9.2 0.625 0.525 0.06 peak HR in 800 m (bpm) 185.8 9.1 184.5 10.1 188.3 8.2 185.5 10.5 0.889 0.395 0.08 CMJ (Δ cm) −10.2 8.8 −6.8 4.8 −13.3 8.7 −6.8 5.9 4.564 0.008 0.28 Lactate (mmoL/L) 19.1 4.7 19.0 4.2 20.1 4.6 17.8 4.4 0.979 0.413 0.07 RPE: Rate of perceived exertion; CMJ: countermovement jump. Concerning the SQ results, actigraphic analysis showed significant differences between conditions (placebo versus caffeine) in sleep efficiency (p = 0.003; ES = 0.71), actual wake time (p = 0.001; ES = –1.18), and number of awakenings (p = 0.005; ES = –0.96) (Figure 2 and Table 2). In addition, the Karolinska sleep questionnaire showed significant differences between conditions, favoring placebo in SQ (p = 0.005; ES = 1.11), calm sleep (p = 0.005; ES = 1.11), ease of falling asleep (p = 0.003; ES = 1.38), and feeling refreshed after waking (p = 0.006; ES = 1.11) (Table 2). Table 2. Sleep quality results. Placebo Caffeine Effect Size (ES) 95% CI for ES Mean SD Mean SD p Lower Upper Actigraphic sleep quality Latency (min) 6.15 2.79 6.77 2.32 0.290 −0.31 −0.86 0.25 Sleep efficiency (%) 92.2 3.0 86.4 5.5 0.003 0.71 0.27 0.91 Time in bed (min) 470.2 118.3 461.2 128.2 0.641 0.13 −0.42 0.68 Actual sleep time (min) 434.8

The study investigates caffeine's impact on running performance and sleep quality in trained athletes.