Effect of long-haul transmeridian travel on recovery and performance in international level swimmers

Abstract

This study holistically examined the effects of long-haul transmeridian travel (LHTT) on physiological, perceptual, sleep and performance markers in nine international level swimmers preparing for the 2019 FINA World Long Course Championships in Gwangju, South Korea. Baseline (BL) measurements were taken over two days during the week before a long-haul eastward flight across eight time-zones. Following the flight, measurements were taken over a six-day holding camp in Japan (C1-C6), and over four days at the competition venue in Gwangju before the Championships commenced (PR1-PR4). Salivary cortisol (sCort), immunoglobulin A (sIgA), alpha-amylase (sAA) concentrations and perceptual measures via the Liverpool John Moore's University Jetlag Questionnaire were assessed. Sleep was monitored using wrist activity monitors and self-report sleep diaries. Performance was assessed via squat jump (SJ), countermovement jump (CMJ) and a 4 × 100 m swim test. Participants perceived themselves to be significantly more fatigued and jet lagged than BL for five- and nine-days post-travel, respectively. Morning sCort decreased by 70% on C1 and remained significantly lower than BL until C6 (p < 0.05). Sleep ratings improved significantly in comparison to BL from C5 onwards (p < 0.05). Compared with BL, there was no significant change in swim performance or SJ height following travel; however, there was a 3.8 cm improvement (p < 0.001) in CMJ height on C5. It took ten days for elite swimmers to perceive themselves recovered from jet lag following LHTT in an eastward direction across eight time-zones. LHTT did not negatively affect sleep or physical performance in the swimmers in comparison to BL.

Keywords

Aquatic sport cortisol countermovement jump immunoglobulin A jet lag sleep travel fatigue

Introduction

Elite athletes frequently embark on long-haul transmeridian travel (LHTT) for training and major international competitions. LHTT across three or more time-zones typically causes travel fatigue and jet lag which can not only be disruptive to athletes’ training and lifestyle, but it is associated with physical and physiological decrements, cognitive and psychomotor disturbances, disrupted sleep, and increased risk of illness in elite athletes.^1,2 These issues are all known to be associated with under-recovery and reduced athlete performance;³ however, the direct effects of LHTT travel on elite athlete performance remain unclear.⁴

The symptoms of travel fatigue and jet lag are numerous and complex and often difficult to separate.⁵ Travel fatigue is associated with acute symptoms of general fatigue, disorientation and weariness that occur during or immediately following long-haul (LH) travel in any direction.² Travel fatigue is caused by challenges of LH travel, including stressful situations, such as flight delays, disrupted sleep duration and quality,^6,7 limited access to preferred food and fluid choices and changes to normal scheduling of mealtimes.⁸ Poor air quality, low relative humidity and “cabin altitude” on the aircraft itself can also contribute to travel fatigue and be a risk to athlete health.^1,2 Recovery from travel fatigue is expected to take a day or two with adequate rest, sleep, and nutrition,² or longer, if allowed to accumulate over a season.⁹

Rapid LH travel across three or more time-zones has been shown to cause misalignment between the body clock and destination time, which can cause an array of symptoms classified as “jet lag”.¹⁰ Jet lag symptoms include poor sleep, gastrointestinal disturbance, negative subjective changes, and impaired physical and cognitive performance.^2,10 Elite athletes consistently report feeling jet lagged for several days following LH travel.^11–14 Symptoms of jet lag are generally more severe and last longer following an eastward flight in comparison to a westward flight^10,15 and when greater number of time-zones are crossed.² The human body clock needs time to gradually adjust and consequently the symptoms of jet lag will persist until circadian rhythms shift to a new environment.¹⁰

This study aimed to assess the impact of LH eastward travel across eight time-zones on physical performance of a group of international level swimmers preparing to compete at the 2019 FINA World Long Course Championships in Gwangju, South Korea. To gain an understanding of the underlying mechanisms affecting recovery from LHTT, this novel study also carried out a comprehensive evaluation of changes in perceived wellness, sleep and biomarkers associated with circadian rhythm, stress, and immunity.

Methods

Participants

All ten swimmers of the Irish National Team travelling to compete at the 2019 FINA World Long Course Championships in Gwangju, South Korea volunteered to participate in the study; however, one swimmer was excluded because the swimmer did not wear an activity monitor. Data from seven male (mean ± SD: age 22.6 ± 2.9 years, height 185.7 ± 6.1 cm, body mass 81.5 ± 6.2 kg) and two female swimmers (mean ± SD: age 17.9 ± 0.2 years, height 166.6 ± 11.2 cm, body mass 63.3 ± 16.5 kg) were included in the final analysis. Prior to data collection, all swimmers provided written informed consent. This study was approved by the human research ethics committee of the University of Limerick and conformed to the declaration of Helsinki pertaining to research using human subjects.

Swimmers’ chronotype was identified using the Horne and Ostberg Morningness Eveningness Questionnaire (MEQ), a 19-item self-report questionnaire based on individuals’ activity preferences.¹⁶ The questions yielded MEQ scores ranging from 16–96. Lower scores (≤ 41) indicated an evening type (EType), higher scores (≥ 59) indicated a morning type (MType) and scores between 42–58 indicated neither type (NType). Swimmers’ MEQ scores were 53.4 ± 6.7 (mean ± SD). Based on their MEQ scores, two of the swimmers were MType and seven swimmers were NType.

Experimental design

Data were obtained from swimmers over an 18-day period before the 2019 FINA World Long Course Swimming Championships in Gwangju, South Korea, which included a trip from Ireland to Japan, a six-day holding camp in Japan and a flight to South Korea five days before the World Championships commenced (Figure 1). Baseline data were collected over two days during the week before leaving Ireland (BL1-BL2). Data were collected every day during the 6-day camp in Hamamatsu, Japan (C1-C6) and for the 4 days prior to the start of the World Championships in Gwangju, South Korea (PR1-PR4).

Figure 1.

Schematic representation of study design and total daily training load (mean ± SE) and training load in the morning (TL AM) and evening (TL PM) measured at baseline (BL), during the holding camp (C1-C6) and in the days leading up to start of world championships (PR1-PR4). * indicates statistical significance (p < 0.05) in comparison to total training load on C3 as determined by multi-level regression analysis. LJMUQ, Liverpool John Moore's University Jet Lag Questionnaire; SJ, squat jumps; CMJ, countermovement jumps.

The trip from Dublin, Ireland to Hamamatsu, Japan in July required an eight-hour eastward time zone change. On the morning of departure, the swimmers convened at Dublin airport approximately 2–3 h before the first flight. The first leg was from Dublin, Ireland to Frankfurt, Germany (flight departure time = 12:10 local-time, LT; travel duration = 2 h 05 min). In Frankfurt, there was a 2 h 55 min layover before the second flight from Frankfurt to Haneda, Tokyo (flight departure time = 18:10 LT; travel duration = 11 h 05 min). The final leg was a 20 min train ride from Haneda airport to catch a bullet train from Shinagawa to Hamamatsu (train departure time = 14:00 LT; travel duration = 1 h 22 min), followed by a 40 min bus journey. The swimmers arrived at their accommodation at 16:10 LT. Total travel time from their flight departure from Dublin to when they arrived at their destination was approximately 20 h.

Following a 6-day training camp in Hamamatsu, the swimmers returned to Haneda airport in Tokyo by bus and bullet train and then took a flight to Seoul, South Korea (flight departure time = 12:05 LT; flight duration = 2 h 30 min). On arrival to Seoul, they took a train from the airport to Gwangju (train departure time = 16:00 LT; travel duration = 2 h 40 min), the location at which the World Championships took place. There were four days (PR1-PR4) before the World Championships commenced. The journey from Hamamatsu to Gwangju took approximately 12 h and did not require a time zone change.

Prior to departure, general travel and sleep recommendations were provided to the swimmers and coaches via workshop and infographics, but there was no monitoring of these recommendations during travel. The swimmers did not attempt to shift their circadian rhythm prior to departure through adjustment of sleep/wake schedule or by using melatonin. The swimmers refrained from melatonin and hypnotic medications for the duration of the study period.

Training schedules

The swimmers trained at an intensity, frequency, and duration, which was set at the discretion of the coaching staff and individualised for each swimmer. Swimmers completed between 9 and 11 training sessions during the holding camp in Hamamatsu (C1-C6) and between 4 and 5 sessions during the pre-race period in Gwangju (PR1-PR4). Training sessions consisted of a swim or a gym session or a combination of both. Morning training sessions commenced between 08:00 and 10:00 Japanese Standard Time (JST) and evening training sessions between 15:30 and 18:00 JST.

The session-RPE method was used to calculate training load as proposed by Foster et al.¹⁷ Training loads (arbitrary units [AU]) were calculated by multiplying each swimmer's total session duration (min) by their rating of perceived exertion provided approximately 15 min after each training session. Due to swimmers using various means of recording training load while training at home, BL training data were not recorded. The highest training load occurred on C3. In comparison to C3, there was a significant reduction (p < 0.001) in total daily training load on C4 to PR4 (Figure 1), as the athletes tapered for competition. The swimmers did not train on travel days.

Daily schedules were set at the discretion of the coaching and management staff. All athletes underwent similar eating and sleeping schedules as they all shared rooms and ate together at the same restaurant at set mealtimes.

Methodology

Performance markers

Swimmers performed two squat jumps (SJ), followed by two counter-movement jumps (CMJ) with 120 s between each jump after a dry-land warm up and before swim training on BL1 (AM), C1 (AM), C2 (AM) and C5 (AM), using protocols described elsewhere.¹⁸ Jump tests were performed using the same equipment and where possible the same athlete testing order. Swimmers were instructed to perform each jump at maximum effort. Feedback on jump height was provided by the researcher to the swimmer immediately after each jump and the highest of the two jumps was recorded. The swimmers executed the jumps in their normal gym footwear. All swimmers were familiar with the jump tests as they were utilised regularly as part of their monitoring programme. Jump height was recorded using the Optojump system (Optojump, Microgate, Bolzano, Italy). The Optojump system was used as it is fully portable and therefore ideal for use on an international training camp. It has been demonstrated that a systematic difference in jump height exists between the Optojump and gold-standard force plate due to the design of the Optojump.¹⁸ However, the test-retest reliability of the Optojump system is excellent, with intraclass correlation coefficients of 0.997–0.998, low coefficient of variation (CV, 2.7%) and low random errors (±2.81 cm).¹⁸

A 4 × 100 m swim test was the measure of sport-specific performance. This test was chosen as it has been used to assess and monitor swimmer readiness in various British and Irish high-performance systems for several years. All swimmers in the present study were familiar with the 4 × 100 m swim test as it was utilised regularly as part of their monitoring programme. Swim tests were performed in a 50 m indoor swimming pool on BL1 (PM), C1 (AM), C3 (AM), C5 (PM) and PR1 (AM). All starts were an in-water push start from the wall. The swim coach recorded performance times using a stopwatch, which was started on the coaches call and stopped when the swimmers’ hand touched the wall. The 4 × 100 m repetitions were performed at increasing swim speeds with a 1 min 40 s turn-around time. Swim speeds were paced by the swimmer and performed at a subjective swim speed to elicit heart rates 60, 50, 40 and 30 beats below maximum (BBM). On completion of each 100 m repetition the swimmer measured heart rate manually at carotid pulse for a period of 10 s and gave this value to the coach to record. The relationship between swim time per 100 m and self-reported heart rate for the four stages of the swim test were analysed by linear regression. The test was deemed valid if there was a linear relationship (r² > 0.9) between 100 m swim time and heart rate. The key performance indicator was swim time per 100 m at 40 BBM, and this value was determined by linear interpolation.

Perceptual markers

The Liverpool John Moore's University Jet Lag Questionnaire (LJMUQ) previously described by Waterhouse et al.¹⁹ was completed AM and PM at BL and during the holding camp (C1 – C6) and AM only during the four days before the World Championships (PR1 – PR4). Only jet lag and fatigue were assessed in both AM and PM questionnaires, all other components were assessed either in the AM (sleep) or PM (function, diet, and bowel movement). Following a method outlined elsewhere,²⁰ data were pooled and summed into five components (jet lag, sleep, fatigue, function, diet and bowel movement ratings). The greater the summed value of each component, the worse the related symptoms. The LJMUQ has been demonstrated to be highly internally reliable (Cronbach's alpha = 0.85) and internally consistent;²¹ however, it has not been validated against objective measures of circadian disruption.

Sleep markers

Sleep actigraphy data were collected for eight nights that comprised of a two-night baseline (BL1 and BL2) and six nights on arrival to Japan (C1-C6). Sleep/wake behaviour was monitored using self-report sleep diaries and wrist activity monitors (Philips Respironics, Bend, OR). Activity monitors have been found to be a valid method of monitoring sleep in elite athletes, with very good agreement rates (81–90%) with polysomnography, the gold standard for measuring sleep.²² Swimmers were instructed not to remove the activity monitor except when showering or swimming. Data derived from the sleep diaries and activity monitor were used to determine the amount and quality of sleep, using the same methods previously described to quantify the sleep/wake behaviour in elite swimmers.²³

Physiological markers

Swimmers’ saliva samples were collected immediately on rising in the morning (AM) and just before bedtime (PM) at BL and C1 – C6, and AM only on PR1 – PR4. Swimmers were required to avoid eating, brushing their teeth, chewing gum and physical activity for at least one hour before providing the saliva sample. Salivary samples were collected using an oral fluid collector kit (Soma Bioscience, Wallingford, UK), in accordance with manufacturer's guidelines. Samples were analysed using a lateral flow device (LFD) with separate cartridges used to analyse salivary immunoglobulin A (sIgA)/salivary cortisol (sCort) and salivary alpha-amylase (sAA). The LFD has been validated against ELISA analysis (r = 0.89, p < 0.01 and CV = 9.4%),²⁴ and provides the practitioner with a portable and time-efficient method of analysis suitable for “field-based” support.

Statistical analysis

Data are presented as mean ± SEM. Descriptive statistics were calculated for all variables, using MLwiN software (version 3.04, Centre for Multilevel Modelling, University of Bristol). Data were analysed using a multilevel modelling approach.²⁵ Multi-level modelling was used as there were multiple testing time points and the study sought to investigate both between and within subject variability. A two-level model was conducted, study phase (level 1) and swimmers (level 2); and two levels were created for each variable separately. This was a random intercept model where repeated observations and participants were assumed to be normally distributed. These assumptions were assessed using a Q-Q plot and it was confirmed that these assumptions were acceptably normal.²⁶ Analysis was used to identify changes in mean values of all variables across the identified time points. No significant difference was found between BL1 and BL2, therefore BL1 and BL2 were averaged to give a mean baseline (BL). Significance was set at p < 0.05.

Results

Performance markers

CMJ height, SJ height and swim time for 100 m at 40 BBM, are presented in Figure 2(a)–(c), respectively. There was no significant change in time to swim 100 m at 40 BBM or SJ height in comparison to BL following LHTT. There was a significant 3.8 cm (8.5%; beta = 2.9 [95% CI 1.3 to 4.5] cm; p < 0.001) (estimated parameter [95% CI lower limit to upper limit]), where CI = confidence interval) improvement in CMJ height on C5 in comparison to BL. During the World Championships, three of the nine swimmers achieved a PB time. The team averaged times that were 101.6 ± 1.2% of pre-Championship PB times, scoring of 830 ± 48 FINA points, mean ± SD.

Figure 2.

Performance data. Counter-movement jump (CMJ) Height (a), squat jump (SJ) Height (b) and Swim time for 100 m at 40 beats below maximum (BBM) (c), measured at baseline (BL), during the holding camp (C1-C6) and in the days leading up to start of World Championships (PR1). Data presented as mean ± SE. * indicates statistical significance (p < 0.05) in comparison to morning scores at BL determined by multi-level regression analysis.

Perceptual markers

Self-reported jet lag and fatigue AM and PM values are presented in Figure 3(a) and (b), respectively. Jet lag AM was worst on the morning after arrival (C1), remained significantly elevated in comparison to BL for nine days (C1-PR2, p < 0.05) and normalised on PR3, ten days post-travel. Jet lag PM values were significantly higher during C1-C6 in comparison to BL (p ≤ 0.01).

Figure 3.

Liverpool John Moore's University Jetlag Questionnaire. Perceived jet lag (a) measured at baseline (BL) and post-travel in the morning (JL AM) and evening (JL PM) and fatigue (b) values measured at baseline (BL) and post-travel in the morning (Fatigue AM) and evening (Fatigue PM), during 6-day holding camp (C1 – C6) and during the 4 days before World Championships began (PR1 – PR4). Higher values mean worse symptoms. Data presented as mean ± SE. * indicates statistical significance (p < 0.05) in comparison to morning scores at BL determined by multi-level regression analysis. ^ indicates statistical significance (p < 0.05) in comparison to evening scores at BL determined by multi-level regression analysis.

Perceived fatigue was significantly higher than BL in the AM on C1, C3 and C4 (p ≤ 0.04) and in the PM on C1-C5 (p ≤ 0.03).

Function-related values were significantly higher than BL on C2 (beta = 2.06 [95% CI 0.13 to 3.99]; p = 0.02). Meal-related values were significantly higher on C1 (beta = 1.52 [95% CI −0.12 to 3.16]; p = 0.03) and C2 (beta = 2.08 [95% CI 0.44 to 3.73]; p = 0.006). There were no significant differences in bowel-related ratings in comparison to BL.

Sleep markers

The swimmers subjectively rated their best night's sleep on C5, and sleep ratings were significantly better than BL from C5 (beta = −5.07 [95% CI −8.64 to −1.51]; p = 0.002) onwards (p < 0.05) (Figure 4).

Figure 4.

Daily subjective values relating sleep-related factors measured at baseline (BL), during 6-day holding camp (C1 – C6) and during the 4 days before world championships began (PR1 – PR4). Higher values means worse symptoms. Data presented as mean ± SE. * indicates statistical significance (p < 0.05) in comparison to scores at BL determined by multi-level regression analysis.

Table 1 presents sleep/wake data as recorded by activity monitor. The swimmers tended to go to bed significantly earlier (p ≤ 0.004) and get up significantly earlier (p ≤ 0.04) during the camp in comparison to BL values recorded at home. The earliest bedtime (beta = −2.64 [95% CI −3.16 to −2.12] h; p = 0.004), shortest sleep latency (beta = −35.2 [95% CI −52.7 to −17.8] min; p < 0.001), longest sleep duration (beta = 1.1 [95% CI 0.02 to 2.19] h; p = 0.02) and best sleep efficiency (beta = 7.8 [95% CI −1.3 to 17.0] %; p = 0.046) was on C1. After this, sleep duration did not differ significantly in comparison to BL except on C5, where sleep duration was significantly higher than BL (beta = 0.94 [95% CI −0.1 to 2.0] h; p = 0.04). Sleep latency during the camp tended to be shorter than BL, and significantly so on C1, C3, C5 and C6 (p ≤ 0.01).

Table 1.

Sleep/wake behaviours at baseline and during 6-day holding camp as recorded by actiwatch activity monitor.

	Bedtime (hh:mm), LT	Get up Time (hh:mm), LT	Fragment Index (units)	Total Sleep Time (h)	Sleep Latency (min)	Sleep Efficiency (%)
BL	23:22 (00:14)	08:34 (00:31)	26.1 (3.0)	7.5 (0.5)	39.9 (8.7)	82.1 (3.0)
C1	20:45 (00:09)*	06:14 (00.07)*	26.8 (2.5)	8.6 (0.2)*	5.1 (2.6)*	89.5 (1.4)*
C2	22:17 (00:12)*	07:03 (00:17)*	26.4 (2.4)	6.8 (0.7)	28.2 (10.6)	77.8 (4.7)
C3	22:08 (00:13)*	06:17 (00:04)*	29.4 (3.2)	6.5 (0.6)	15.8 (5.3)*	79.2 (5.2)
C4	22:21 (00:09)*	06:21 (00:03)*	29.6 (4.2)	6.6 (0.4)	27.1 (4.9)	82.4 (3.5)
C5	22:32 (00:14)*	08:02 (00:05)*	23.8 (2.2)	8.3 (0.3)*	22.1 (4.1)*	87.7 (1.9)
C6	22:39 (00:08)*	07:24 (00:02)*	20.8 (2.2)	7.7 (0.2)	17.7 (4.3)*	87.8 (1.8)

BL, baseline; C, camp, C1 is camp day 1 and so on; * indicates statistical significance (p < 0.05) in comparison to BL; all values are expressed as mean (SE); all times are expressed as local-time (LT).

Physiological markers

sCort, sAA and sIgA (AM and PM) concentrations at each time point are presented in Figure 5(a)–(c), respectively. In comparison to BL, there was a 70% decrease in sCort AM on C1 (beta = -5.49 [95% CI −8.11 to −2.86] ng.ml⁻¹; p < 0.001). sCort remained significantly lower than BL from C1 to C6 (p < 0.05). On the morning of PR1, sCort AM increased to levels similar to BL (beta = 0.75 [95% CI −1.87 to 3.38] ng.ml⁻¹; p = 0.29). There was a further increase in sCort AM on PR2 (63%; beta = 5.24 [95% CI 2.62 to 7.87] ng.ml⁻¹; p < 0.001) and PR3 (44%; beta = 3.56 [95% CI 0.83 to 6.29] ng.ml⁻¹; p = 0.01) in comparison to BL but sCort normalised again on the morning of PR4 (beta = 1.48 [95% CI −1.25 to 4.2] ng.ml⁻¹; p = 0.14). sCort PM were significantly higher than BL on C3 (beta = 1.57 [95% CI −2.03 to 5.17] ng.ml⁻¹; p = 0.03) and C6 (beta = 3.38 [95% CI −4.05 to 10.82] ng.ml⁻¹; p < 0.001) only.

Figure 5.

Salivary data. Morning and evening sCort (a) sAA (b) and sIgA (c) values measured at baseline (BL), during 6-day holding camp (C1 – C6), and 4 days before World Championships began (PR1 – PR4). Data presented as mean ± SE. * indicates statistical significance (p < 0.05) in comparison to morning scores at baseline (BL) determined by multi-level regression analysis. ^ indicates statistical significance (p < 0.05) in comparison to evening scores at baseline determined by multi-level regression analysis.

sAA AM was significantly lower than BL on six of the ten mornings after arrival (Figure 5(b), p ≤ 0.04). There were no significant differences in sAA PM values during C1-C6 in comparison to BL.

sIgA AM was significantly higher than BL on C6 (beta = 178.4 [95% CI 18.6 to 338.1] µg.ml⁻¹; p = 0.01) and PR2 (beta = 171.1 [95% CI 11.4 to 330.8] µg.ml⁻¹; p = 0.02). sIgA PM was significantly higher than BL on C2 (beta = 119.4 [95% CI −6.5 to 245.3] µg.ml⁻¹; p = 0.03) only.

Discussion

This novel study investigated the effects of eastward travel across eight time-zones on a comprehensive set of physiological, perceptual, sleep and performance markers in a group of international level swimmers prior to competing at the 2019 FINA World Long Course Championships in South Korea. LHTT did not have any detrimental effect on direct (4 × 100 m swim test) or indirect measures (CMJ and SJ) of performance; however, athletes perceived themselves as “jet lagged” for up to nine days following travel. There was also disturbance to sCort which normalised by day eight post-travel. Objective and subjective sleep and other perceptual markers all normalised or improved in comparison to BL by day six post-travel. In the final days before competition, elevated sCort, sAA and sIgA indicated possible increases in psychological stress. There was indication of improved physical performance in comparison to BL in the days leading up to competition, indicating that the swimmers had recovered sufficiently from LHTT for the performance benefits of the pre-competition taper to occur.

Performance

It might be expected that jet lag and travel fatigue would have a detrimental effect on physical performance;² however, this is not supported by the present study which finds no negative changes in swim time at 40 BBM, SJ or CMJ height on days following LHTT in comparison to BL. This is despite the test time (approx. 10 am JST) corresponding to an unfavourable body clock time (approx. 2 am at home) for physical performance on the days following travel.²⁷ While LHTT has previously been shown to detrimentally effect upper limb grip strength in elite male gymnasts¹² and elite female soccer players²⁸ the effect of LHTT on more complex tasks associated with sports performance and real-life competitive performance is less clear.⁴

The reasons for disruption of perceptual and physiological markers associated with LHTT not influencing subsequent sports performance in elite athletes are unknown. In the present study, swimmers performed a standard training warm-up prior to testing, and it has been suggested by others that a sufficient warm-up may mask the negative effects of LHTT on physical performance.¹¹ CMJ height was significantly higher than BL on C5. This was followed by a non-significant improvement of swim time at 40 BBM of 2.2% (p = 0.05) on the morning of PR1. These performance improvements coincided with reduced training load from C4 onwards due to final competition taper and perceived fatigue levels returning to BL from C5 onwards. Mujika et al. also found improved swimming performance of 2.2% following the three-week tapering period before the Sydney Olympics. These data seem to indicate that the athletes in the present study had recovered sufficiently from the flight so that the desired performance effect of the taper could occur. Overall, this team of swimmers were deemed by coaching and management staff to have performed successfully at the FINA World Championships.

Perceptual markers

Analysis of the perceptual markers collected revealed that the athletes perceived themselves as jet lagged the morning after arrival to their destination and, on average, it took 10 days for perceived jet lag values to return to BL levels. The dissipation of perceived jet lag in these swimmers was slightly longer than the expected timeframe of one day for every time zone crossed, as previously reported,²⁹ but the findings are consistent with other studies on elite athletes from individual sports.^11,12

This study found that perceived morning and evening fatigue were elevated for three and five days after travel, respectively and slightly longer than reported in previous studies.^11,14 In agreement with other studies,^11,14 other perceptual markers associated with jet lag and evaluated in LJMJQ such as “function”, “meals” and “bowel” were all observed to have normalised by day three.

Sleep

The present group of athletes perceived sleep as “worse than normal” at BL and the four days following LHTT. Although it was expected that sleep might be disturbed on the nights following LHTT due to a combination of change in time zone² and an unfamiliar sleep environment,³⁰ the reporting of “worse than normal” sleep during BL was unexpected. The perception of poor sleep during this period may be associated with high training loads³⁰ or psychological anxiety and stress³¹ associated with preparing for LHTT for major competition. There was a significant improvement in perceived sleep on most nights from C5 onwards in comparison to BL and is aligned with improved fatigue scores from C6 onwards.

Collection of objective sleep data by actigraphy provided further insight into sleep experienced by athletes during the period before and after LHTT. The average BL sleep duration of 7.5 h is higher than the 6.8–7.1 h per night reported for elite athletes from individual and team sports,³² but less than the 8 h of sleep per night recommended for athletes to reduce risk of sustaining an injury.³³ Although average BL sleep latency of 39.9 min were within ranges reported for competitive swimmers,²³ it should be noted that this is longer than reported in elite athletes from individual and team sports (16–22 min).³² BL sleep efficiency at 82.1% is lower than values reported in elite athletes (83.7–91.1%)³² but better than that reported in competitive swimmers (70–77%).²³ It should also be noted that this group of swimmers had later bedtimes and get up times compared to those reported in individual athletes.³²

In agreement with other studies on elite athletes,^6,28,34,35 this study found that the swimmers had a greater propensity to sleep on the night of arrival (C1) with significantly earlier bedtime, shorter sleep onset latency, extended sleep duration and improved sleep efficiency in comparison to BL. It is likely that sleep deprivation experienced by the swimmers during the journey that involved multiple flight connections masked any sleep disturbance due to circadian disruption.

Interindividual variation in the timing of minimum core temperature (CBTmin) has been reported,² but assuming the CBTmin normally occurs at approximately 04:00,¹⁵ following an eastward flight across 8 time-zones the expectation is that CBTmin should occur at approximately midday. Thus, there may be increased sleepiness during the daytime and difficulty falling to sleep at night-time. However contrary to this; the athletes in the present study woke earlier and went to bed earlier during the camp in comparison to their home environment (BL). It should be considered that swimmers’ wake times on camp may have at least been partially influenced by their training schedule and pool availability. Furthermore, the swimmers generally trained twice daily on camp, including in the morning time when their circadian cycle was most likely at its lowest point. It is possible that the requirement to train twice daily, led to an accumulation of fatigue in these swimmers which superseded the peak of the circadian cycle in the evening time, resulting in earlier bedtimes and shorter sleep latency on camp in comparison to BL. This was confirmed in Figure 3; where perceived fatigue was similar or higher in the evening in comparison to the morning time from C2 to C5.

There is evidence of an insignificant reduction in sleep duration and efficiency on C2, C3 and C4, but these improved by C5 and C6. In agreement with the subjective sleep scores, these may indicate that it took approximately five days for sleep to normalise in this group of athletes following LH eastward travel. This is in keeping with the timeline of “partial” jet lag recovery following LH travel across eight time-zones.¹⁵

Physiological markers

Cortisol (Cort) serves as both a major circadian signal and an indicator of stress throughout the body.³⁶ Cort follows a predictive diurnal rhythm with relatively high values in the morning time, followed by a decline throughout the day reaching the lowest point at around midnight.³⁷ In the present study, on the morning after travel (C1) there was a 70% decrease in sCort AM in comparison to BL, so that sCort AM on C1 was similar to sCort PM at BL. This was unsurprising considering at morning time in Japan, the biological clock remains at the point of origin (i.e. night-time) and confirms that LHTT causes desynchronization of sCort. It was only on the eighth day (PR1) that sCort AM were found to be similar to BL. It was previously reported that sCort remains disrupted for up to 11 days in elite athletes following eastward travel across eight time-zones.^11,12 These data suggest that sCort is a convenient and cost-effective method to indicate resynchronisation of circadian rhythm in elite athletes following LHTT.

In the present study, sCort reached its highest levels two to three days before competition, despite the lowest levels of training load. These findings are supported by previous studies which found an increase in Cort during the final taper phase³⁸ and competition phase³⁹ despite considerable reductions in training load. This may indicate an induced psychological stress response in anticipation of upcoming major competition.³⁹

sCort PM was between 6–137% higher in comparison to BL on the days following arrival to Japan, with the highest values on C3 and C6. This might be due to circadian disruption; but it is also expected that sCort PM will be much more responsive to acute stress in the evening time in comparison to the morning.³⁶ Elevated sCort PM on C3 might indicate an acute physical stress response to a relatively high training load and on C6 elevated psychological stress in anticipation of travel the following morning.

Production of alpha-amylase (AA) is regulated by the autonomic nervous system and is strongly influenced by physical and psychological stress.⁴⁰ Consistent with previous research,⁴¹ this study found that sAA concentration has a diurnal rhythm with lowest sAA values observed in the morning and highest values in the evening. On days following LHTT there was no significant difference in sAA PM concentrations in comparison to BL. However, on most mornings, sAA AM was significantly lower than BL. Little is known about the effect of LHTT on sAA;⁴² however, the present findings indicate that the physical and psychological stresses associated with LHTT² were not sufficient to cause an elevation of sAA following LHTT. It has previously been suggested that AA may be a marker of host defence (antimicrobial) protein activity thus it is possible that reduced sAA AM following LHTT indicated suppressed immunity.⁴³ Although none of the athletes in the present study reported upper respiratory tract infection (URTI), reduced sAA may explain the increased incidence of URTI reported in elite athletes following LHTT.¹

It is possible that sAA was elevated in the days leading up to travel (BL) and competition (PR2 and PR3) due to elevated psychological stress associated with preparation for travel and major competition. Reduced morning sAA on C1 and C3-PR1 might be due to enforced rest and recovery occurring during travel and taper. This is supported by Sinnott-O’Connor et al. who found that sAA levels in elite Paralympic swimmers were reduced during taper in comparison to intense training phase.³⁹ Poor perceived sleep and elevated sAA during the days before travel and competition may be associated with elevated psychological stress associated with preparing for LHTT and major competition.

sIgA concentrations in the present study are higher than those reported in healthy males,⁴⁴ and male competitive judo athletes,³⁸ but similar to those observed in trained triathletes⁴⁵ and elite female rowers.⁴⁶ It should be noted that large within and between subject variations in sIgA have been reported in sedentary and trained individuals.⁴⁶ The findings from the current study revealed that sIgA concentrations at BL were highest in the morning time with a 48% decline in the evening time indicating that the diurnal rhythm of sIgA is consistent with previous findings.⁴⁴ Furthermore, there was no significant change in sIgA concentration or diurnal pattern in the days following LHTT in comparison to BL. Thus, if suppressed immunity was experienced by these athletes following LHTT, it was not accompanied by suppressed sIgA. These findings are supported by a previous study which reported no change in sIgA concentrations in masters level triathletes following a LH flight in a north-westerly direction.⁴⁵ These data may suggest that suppressed immunity following LH travel is associated with suppressed sAA levels, but not reduced sIgA.

sIgA values on the mornings of C6 and PR2 were significantly higher than BL values. In support of this, increased concentrations of sIgA have been found during the final taper period in competitive judo athletes³⁸ and during the competitive period in elite Paralympic swimmers.³⁹ Engaging in chronic intense training can lead to suppression of sIgA, leaving an “open window” of depressed mucosal immunity in athletes.⁴⁷ It is reasonable to assume that the decrease in training load from C4 onwards in the present study might lead to increased sIgA levels in the final phase of taper. An alternative explanation for variance in sIgA in the days preceding competition could be linked to the changes in acute psychological strain.⁴⁸ Unfortunately psychological assessment, which would have allowed better understanding of changes in swimmers’ psychological strain in the lead up to competition, did not take place.

Study limitations

The number of athletes participating in this study was limited because of the competition level and the prohibitive cost of transporting individuals on a LH flight. However, it presented the advantage of a homogeneous sample, possessing a high level of skill and fitness and a unique opportunity to observe an elite athlete group in their final preparations for major competition.

The swim test used in this study was not validated in peer-reviewed published literature; however, preliminary investigations on a variation of this test (4 × 200 m) suggest typical error of measurement (TEM) of 0.16 s and smallest worthwhile change (SWC) 0.62 s (unpublished, Hollis 2019, personal communication). Since the TEM is less than the SWC, the test may be considered useful in its ability to detect real-life changes in fitness.⁴⁹ It is acknowledged that ideally a swim test with proven validity and reliability would have been used to assess performance; however, the test in the present study was chosen as it has been habitually used to assess and monitor swimmer readiness and is one that is familiar to the swimmers included in this study. The introduction of a different performance test is likely to have reduced compliance with the study particularly during this critical period of final preparation for World Championships. For a similar reason, it was deemed not appropriate to administer the LJMJQ at the proposed frequency of five times daily.¹⁹ This has been a common issue in studies involving elite athletes.^11,14,34,35

As noted in previous studies on elite athletes,^11,12 this study found a high degree of individual variability in the types and severity of jet lag symptoms exhibited and varying rates of recovery following LH travel. The characteristics that contribute to this variation are sex, age, chronotype, training status and travel experience.⁵ The age (21.5 ± 3.2 y), chronotype profiles (average MEQ score = 53.4 ± 6.7) and training status of this elite athlete population within this study were relatively homogenous. Other possible reasons are variations of athletes’ exposure to light, timing of physical activity, sleep, and diet; all of which have been described as possible interventions for jet lag and travel fatigue.^2,15,50 Although the swimmers in the present study received general travel guidelines designed specifically for their journey and followed similar training, meal and sleep schedules, no attempt was made by the researchers to control these variables.

The performance data in this study seemed to indicate that the athletes had recovered sufficiently from LHTT so that the desired performance effect of the pre-competition taper could occur; however, salivary markers in the final days before competition may indicate acute psychological strain experienced by the swimmers in anticipation of major competition. Future research might consider the use of psychological assessment such as REST-Q or BRUMS to understand the stress associated with the preparation for LHTT and major international competition. At the team's request, objective sleep monitoring and the evening version of the LJMJQ were withdrawn in the days leading up to competition (PR1-PR4). Thus, it is not known if there were any changes in objective sleep markers or perceptual markers associated with “function”, “meals” and “bowel” during this period.

Conclusion

The present study is the first comprehensive presentation and evaluation of ecologically valid perceptual, sleep, physiological and performance data from a group of international level athletes before and after LHTT, and in the lead up to a major international competition. It took ten days for elite swimmers to perceive themselves to be fully recovered from jet lag following LHTT in an eastward direction across eight time-zones. sCort, a physiological marker of circadian rhythm, took eight days to normalise. This was preceded by improved sleep and other perceptual markers such as fatigue by day six post-travel. The study also found that LHTT did not detrimentally affect physical performance, sleep or physiological markers associated with stress or immunity in this group of swimmers.

In the approach to competition there was indication that the swimmers had recovered sufficiently from LHTT for the performance benefits of the taper to occur. However, elevated sCort, sAA and sIgA during the final days before competition may indicate psychological stress in anticipation of competition.

The data and findings from this study indicate that athletes engaging in LHTT in an eastward direction across eight time-zones for major competition should be allowed ten days to recover from jet lag and travel fatigue. There is suggestion that special consideration should be given to strategies to reduce sleep disruption in the days leading up to LHTT and to manage psychological stress in the approach to major competition.

Footnotes

Acknowledgements

We thank the coaches, practitioners, team manager and athletes from the Swim Ireland Performance Team for their participation in, and help with, the study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iDs

Antonia Rossiter

Alan M. Nevill

References

Schwellnus

Derman

Jordaan

, et al. Elite athletes travelling to international destinations >5 time zone differences from their home country have a 2–3-fold increased risk of illness. Br J Sports Med 2012; 46: 816–821.

Waterhouse

Reilly

Atkinson

, et al. Jet lag: trends and coping strategies. Lancet 2007; 369: 1117–1129.

Soligard

Schwellnus

Alonso

, et al. How much is too much? (part 1) international olympic committee consensus statement on load in sport and risk of injury. Br J Sports Med 2016; 50: 1030–1041.

Rossiter

Warrington

Comyns

. Effects of long-haul travel on recovery and performance in elite athletes. A systematic review. J Strength and Conditioning Res. Epub ahead of print Mar 18 2021. DOI: 10.1519/JSC.0000000000004021.

Bin

Postnova

Cistulli

. What works for jetlag? A systematic review of non-pharmacological interventions. Sleep Med Rev 2019; 43: 47–59.

Lastella

Roach

Sargent

. Travel fatigue and sleep/wake behaviors of professional soccer players during international competition. Sleep Health 2019; 5: 141–147.

Roach

Matthews

Naweed

, et al. Flat-out napping: the quantity and quality of sleep obtained in a seat during the daytime increase as the angle of recline of the seat increases. Chronobiol Int 2018; 35: 872–883.

Reilly

Waterhouse

Burke

, et al. International association of athletics federations. Nutrition for travel. J Sports Sci 2007; 25: S125–S134.

Samuels

. Jet lag and travel fatigue: a comprehensive management plan for sport medicine physicians and high-performance support teams. Clin J Sport Med 2012; 22: 268–273.

10.

Simmons

McGrane

Wedmore

. Jet lag modification. Curr Sports Med Rep 2015; 14: 123–128.

11.

Bullock

Martin

Ross

, et al. Effect of long haul travel on maximal sprint performance and diurnal variations in elite skeleton athletes. Br J Sports Med 2007; 41: 569–573.

12.

Lemmer

Kern

Nold

, et al. Jet lag in athletes after eastward and westward time-zone transition. Chronobiol Int 2002; 19: 743–764.

13.

Fullagar

Skorski

Duffield

, et al. Impaired sleep and recovery after night matches in elite football players. J Sports Sci 2016; 34: 1333–1339.

14.

Fowler

Duffield

Howle

, et al. Effects of northbound long-haul international air travel on sleep quantity and subjective jet lag and wellness in professional Australian soccer players. Int J Sports Physiol Perform 2015; 10: 648–654.

15.

Roach

Sargent

. Interventions to minimize jet lag after westward and eastward flight. Front Physiol 2019; 31: 927.

16.

Horne

Ostberg

. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol 1976; 4: 97–110.

17.

Foster

Hector

Welsh

, et al. Effects of specific versus cross-training on running performance. Eur J Appl Physiol Occup Physiol 1995; 70: 367–372.

18.

Glatthorn

Gouge

Nussbaumer

, et al. Validity and reliability of optojump photoelectric cells for estimating vertical jump height. J Strength Cond Res 2011; 25: 556–560.

19.

Waterhouse

Edwards

Nevill

, et al.

Do subjective symptoms predict our perception of jet-lag?

Ergonomics 2000; 43: 1514–1527.

20.

Thompson

Batterham

Jones

, et al. The practicality and effectiveness of supplementary bright light for reducing jet-lag in elite female athletes. Int J Sports Med 2013; 34: 582–589.

21.

Ledger

Bin

Nour

, et al. Internal consistency and convergent and divergent validity of the liverpool jetlag questionnaire. Chronobiol Int 2020; 37: 218–226.

22.

Sargent

Lastella

Halson

, et al. The validity of activity monitors for measuring sleep in elite athletes. J Sci Med Sport 2016; 19: 848–853.

23.

Sargent

Halson

Roach

. Sleep or swim? Early-morning training severely restricts the amount of sleep obtained by elite swimmers. Eur J Sport Sci 2014; 14: S310–S315.

24.

Coad

Mclellan

Whitehouse

, et al. Validity and reliability of a novel salivary immunoassay for individual profiling in applied sports science. Res Sports Med 2015; 23: 140–150.

25.

Rasbash

Steele

Browne

, et al. A User's Guide to MLwiN, v3.03. Centre for Multilevel Modelling, University of Bristol, 2019.

26.

Goldstein

. Multilevel modelling of survey data. Statistician 1991; 40: 235–244.

27.

Reilly

Waterhouse

. Sports performance: is there evidence that the body clock plays a role? Eur J Appl Physiol 2009; 106: 321–332.

28.

Hill

Fields

, et al. Effects of jet lag on factors related to sport performance. Can J Appl Physiol 1993; 18: 91–103.

29.

Loat

Rhodes

. Jet-lag and human performance. Sports Med 1989; 8: 226–228.

30.

Erlacher

Ehrlenspiel

Adegbesan

, et al. Sleep habits in German athletes before important competitions or games. J Sports Sci 2011; 29: 859–866.

31.

Hausswirth

Louis

Aubry

, et al. Evidence of disturbed sleep and increased illness in overreached endurance athletes. Med Sci Sports Exerc 2014; 46: 1036–1045.

32.

Lastella

Roach

Halson

, et al. Sleep/wake behaviours of elite athletes from individual and team sports. Eur J Sport Sci 2015; 15: 94–100.

33.

Milewski

Skaggs

Bishop

, et al. Chronic lack of sleep is associated with increased sports injuries in adolescent athletes. J Pediatr Orthop 2014; 34: 129–133.

34.

Kölling

Treff

Winkert

, et al. The effect of westward travel across five time zones on sleep and subjective jet-lag ratings in athletes before and during the 2015’s world rowing junior championships. J Sports Sci 2017; 35: 2240–2248.

35.

Fullagar

Duffield

Skorski

, et al. Sleep, travel, and recovery responses of national footballers during and after long-haul international air travel. Int J Sports Physiol Perform 2016; 11: 86–95.

36.

Spencer

Chun

Hartsock

, et al. Glucocorticoid hormones are both a major circadian signal and major stress signal: how this shared signal contributes to a dynamic relationship between the circadian and stress systems. Front Neuroendocrinol 2018; 49: 52–71.

37.

Pruessner

Wolf

Hellhammer

, et al. Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sci 1997; 61: 2539–2549.

38.

Papacosta

Gleeson

Nassis

. Salivary hormones, IgA, and performance during intense training and tapering in judo athletes. J Strength Cond Res 2013; 27: 2569–2580.

39.

Sinnott-O’Connor

Comyns

Nevill

, et al. Salivary biomarkers and training load during training and competition in paralympic swimmers. Int J Sports Physiol Perform 2018; 13: 839–843.

40.

Chatterton Jr

Vogelsong

, et al. Salivary alpha-amylase as a measure of endogenous adrenergic activity. Clin Physiol 1996; 16: 433–448.

41.

Nater

Rohleder

Schlotz

, et al. Determinants of the diurnal course of salivary alpha-amylase. Psychoneuroendocrinology 2007; 32: 392–401.

42.

Strahler

Skoluda

Kappert

, et al. Simultaneous measurement of salivary cortisol and alpha-amylase: application and recommendations. Neurosci Biobehav Rev 2017; 83: 657–677.

43.

Pumpa

McKune

Harnett

. A novel role of probiotics in improving host defence of elite rugby union athlete: a double blind randomised controlled trial. J Sci Med Sport 2019; 22: 876–881.

44.

Gleeson

. The effect of single and repeated bouts of prolonged cycling and circadian variation on saliva flow rate, immunoglobulin A and alpha-amylase responses. J Sports Sci 2004; 22: 1015–1024.

45.

Stevens

Thornton

Fowler

, et al. Long-haul northeast travel disrupts sleep and induces perceived fatigue in endurance athletes. Front Physiol 2018; 9: 1826.

46.

Nehlsen-Cannarella

Nieman

Fagoaga

, et al. Saliva immunoglobulins in elite women rowers. Eur J Appl Physiol 2000; 81: 222–228.

47.

Papacosta

Nassis

. Saliva as a tool for monitoring steroid, peptide and immune markers in sport and exercise science. J Sci Med Sport 2011; 14: 424–434.

48.

Bosch

Ring

de Geus

, et al. Stress and secretory immunity. Int Rev Neurobiol 2002; 52: 213–253.

49.

Pyne

. Interpreting the results of fitness testing. In: Proceedings of the International Science and Football Symposium. (ed T Flanagan), Melbourne: Victorian Institute of Sport, 2003.

50.

Janse van Rensburg

DCC

Jansen van Rensburg

Fowler

, et al. How to manage travel fatigue and jet lag in athletes? A systematic review of interventions. Br J Sports Med 2020; 54: 960–968.