Context/Background : International-level swimmers periodize their training to qualify for major championships, then improve further at these events. However, the effects of various factors that could affect performance progressions have not been described systematically. Purpose: To quantify the pattern of change in performance between season best qualifying time and the major championships of the year and to assess the influence of time between performance peaks, ranking at the major events, stroke, event distance, sex, age, and country. Methods: A total of 7832 official competition times recorded at 4 FINA World Championships and 2 Olympic Games between 2011 and 2017 were compared with each swimmer’s season best time prior to the major event of the year. Percentage change in performance was related with the time elapsed between season best and major competition, race event, sex, age, and country using linear mixed modeling. Results: Faster performance (−0.79% [0.67%]; mean [SD]) at the major competition of the year occurred in 38% of all observations vs 62% no change or regression (1.10% [0.88%]). The timing between performance peaks (<34 to >130 d) had little effect on performance progressions (P = .83). Only medal winners (−0.87% [0.91%]), finalists (−0.16% [0.97%]), and US swimmers (−0.44% [1.08%]) progressed between competitions. Stroke, event distance, sex, and age had trivial impact on performance progression. Conclusions: Performance progressions at Olympic Games and World Championships were not determined by timing between performance peaks. Performance progression at a major competition appears necessary to win a medal or make the final, independent of race event, sex, and age.
Iñigo Mujika, Luis Villanueva, Marijke Welvaert, and David B. Pyne
Peter Ibbott, Nick Ball, Marijke Welvaert, and Kevin G. Thompson
Purpose: To assess pacing strategies using prescribed and self-selected interset rest periods and their influence on performance in strength-trained athletes. Methods: A total of 16 strength-trained male athletes completed 3 randomized heavy strength-training sessions (5 sets and 5 repetitions) with different interset rest periods. The interset rest periods were 3 min (3MIN), 5 min (5MIN), and self-selected (SS). Mechanical (power, velocity, work, and displacement), surface electromyography (sEMG), and subjective (rating of perceived exertion) and readiness-to-lift data were recorded for each set. Results: SS-condition interset rest periods increased from sets 1 to 4 (from 207.52 to 277.71 s; P = .01). No differences in mechanical performance were shown between the different interset rest-period conditions. Power output (210 W; 8.03%) and velocity (0.03 m·s−1; 6.73%) decreased as sets progressed for all conditions (P < .001) from set 1 to set 5. No differences in sEMG activity between conditions were shown; however, vastus medialis sEMG decreased as the sets progressed for each condition (1.75%; P = .005). All conditions showed increases in rating of perceived exertion as sets progressed (set 1 = 6.1, set 5 = 7.9; P < .001). Participants reported greater readiness to lift in the 5MIN condition (7.81) than in the 3MIN (7.09) and SS (7.20) conditions (P < .001). Conclusions: Self-selecting interset rest periods does not significantly change performance compared with 3MIN and 5MIN conditions. Given the opportunity, athletes will vary their interset rest periods to complete multiple sets of heavy strength training. Self-selection of interset rest periods may be a feasible alternative to prescribed interset rest periods.
Rebekah D. Alcock, Gregory C. Shaw, Nicolin Tee, Marijke Welvaert, and Louise M. Burke
The urinary excretion of hydroxyproline (Hyp), abundant in collagen protein, may serve as a biomarker of habitual collagen intake, assisting with investigations of current interest in the role of dietary collagen intake in supporting the synthesis of collagenous body tissues. This study investigated the time course of urinary Hyp excretion in “free-living,” healthy, active males following the ingestion of a standardized bolus (20 g) of collagenous (gelatin and a hydrolyzed collagen powder) and dairy (calcium caseinate and hydrolyzed casein) proteins. The excretion of Hyp was assessed over a 24-hr period, separated into three collection periods: 0–6, 6–12, and 12–24 hr. Hyp was elevated for 0–6 hr after the consumption of collagen-containing supplements (gelatin 31.3 ± 8.8 mmol/mol and hydrolyzed collagen 33.7 ± 22.0 mmol/mol vs. baseline: gelatin 2.4 ± 1.7 mmol/mol and hydrolyzed collagen 2.8 ± 1.5 mmol/mol; p < .05), but not for the dairy protein supplements (calcium caseinate 3.4 ± 1.7 mmol/mol and hydrolyzed casein 4.0 ± 3.7 mmol/mol; p > .05). Therefore, urinary Hyp reflects an acute intake of collagenous protein, but is not suitable as a biomarker for quantifying habitual collagen intake, provided through regular dietary practices in “free-living,” healthy, active males.
Shona L. Halson, Renee N. Appaneal, Marijke Welvaert, Nirav Maniar, and Michael K. Drew
Purpose: Psychological stress is reported to be an important contributor to reduced sleep quality and quantity observed in elite athletes. The purpose of this study was to explore the association between psychological stress and sleep and to identify if specific aspects of sleep are disturbed. Methods: One hundred thirty-one elite athletes (mean [SD], male: n = 46, age 25.8 [4.1] y; female: n = 85, age 24.3 [3.9] y) from a range of sports completed a series of questionnaires in a 1-month period approximately 4 months before the 2016 Rio Olympic Games. Questionnaires included the Pittsburgh Sleep Quality Index; Recovery-Stress Questionnaire; Depression, Anxiety, and Stress Scale (DASS 21); and Perceived Stress Scale (PSS). Results: Regression analysis identified the PSS and DASS stress as the main variables associated with sleep. A PSS score of 6.5 or higher was associated with poor sleep. In addition, a PSS score lower than 6.5 combined with a DASS stress score higher than 4.5 was also associated with poor sleep. Univariate analyses on subcomponents of the Pittsburgh Sleep Quality Index confirmed that PSS is associated with lower sleep quality (t 99 = 2.40, P = .018), increased sleep disturbances (t 99 = 3.37, P = .001), and increased daytime dysfunction (t 99 = 2.93, P = .004). DASS stress was associated with increased sleep latency (t 94 = 2.73, P = .008), increased sleep disturbances (t 94 = 2.25, P = .027), and increased daytime dysfunction (t 94 = 3.58, P = .001). Conclusions: A higher stress state and higher perceived stress were associated with poorer sleep, in particular increased sleep disturbances and increased daytime dysfunction. Data suggest that relatively low levels of psychological stress are associated with poor sleep in elite athletes.
Alice M. Wallett, Amy L. Woods, Nathan Versey, Laura A. Garvican-Lewis, Marijke Welvaert, and Kevin G. Thompson
Studies examining pacing strategies during 4000-m cycling time trials (TTs) typically ensure that participants are not prefatigued; however, competitive cyclists often undertake TTs when already fatigued. This study aimed to determine how TT pacing strategies and sprint characteristics of cyclists change during an intensified training period (mesocycle). Thirteen cyclists regularly competing in A- and B-grade cycling races and consistently training (>10 h/wk for 4  y) completed a 6-wk training mesocycle. Participants undertook individually prescribed training, using training stress scores (TrainingPeaks, Boulder, CO), partitioned into a baseline week, a build week, 2 loading weeks (designed to elicit an overreached state), and 2 recovery weeks. Laboratory-based tests (15-s sprint and TT) and Recovery-Stress Questionnaire (RESTQ-52) responses were repeatedly undertaken over the mesocycle. TT power output increased during recovery compared with baseline and loading weeks (P = .001) with >6-W increases in mean power output (MPO) detected for 400-m sections (10% bins) from 1200 to 4000 m in recovery weeks. Decreases in peak heart rate (P < .001) during loading weeks and postexercise blood lactate (P = .005) during loading week 2 and recovery week 1 were detected. Compared with baseline, 15-s sprint MPO declined during loading and recovery weeks (P < .001). An interaction was observed between RESTQ-52 total stress score with a 15-s sprint (P = .003) and with a TT MPO (P = .04), indicating that participants who experienced greater stress during loading weeks exhibited reduced performance. To conclude, intensified endurance training diminished sprint performance but improved 4000-m TT performance, with a subtle change in MPO evident over the last 70% of TTs.
Michael J. Davies, Bradley Clark, Laura A. Garvican-Lewis, Marijke Welvaert, Christopher J. Gore, and Kevin G. Thompson
Purpose: To determine if a series of trials with fraction of inspired oxygen (FiO2) content deception could improve 4000-m cycling time-trial (TT) performance. Methods: A total of 15 trained male cyclists (mean [SD] body mass 74.2 [8.0] kg, peak oxygen uptake 62  mL·kg−1·min−1) completed six 4000-m cycling TTs in a semirandomized order. After a familiarization TT, cyclists were informed in 2 initial trials they were inspiring normoxic air (NORM, FiO2 0.21); however, in 1 trial (deception condition), they inspired hyperoxic air (NORM-DEC, FiO2 0.36). During 2 subsequent TTs, cyclists were informed they were inspiring hyperoxic air (HYPER, FiO2 0.36), but in 1 trial, normoxic air was inspired (HYPER-DEC). In the final TT (NORM-INFORM), the deception was revealed and cyclists were asked to reproduce their best TT performance while inspiring normoxic air. Results: Greater power output and faster performances occurred when cyclists inspired hyperoxic air in both truthful (HYPER) and deceptive (NORM-DEC) trials than NORM (P < .001). However, performance only improved in NORM-INFORM (377 W; 95% confidence interval [CI] 325–429) vs NORM (352 W; 95% CI 299–404; P < .001) when participants (n = 4) completed the trials in the following order: NORM-DEC, NORM, HYPER-DEC, HYPER. Conclusions: Cycling performance improved with acute exposure to hyperoxia. Mechanisms for the improvement were likely physiological; however, improvement in a deception trial suggests an additional placebo effect. Finally, a particular sequence of oxygen deception trials may have built psychophysiological belief in cyclists such that performance improved in a subsequent normoxic trial.
Avish P. Sharma, Philo U. Saunders, Laura A. Garvican-Lewis, Brad Clark, Marijke Welvaert, Christopher J. Gore, and Kevin G. Thompson
Purpose: To determine the effect of altitude training at 1600 and 1800 m on sea-level (SL) performance in national-level runners. Methods: After 3 wk of SL training, 24 runners completed a 3-wk sojourn at 1600 m (ALT1600, n = 8), 1800 m (ALT1800, n = 9), or SL (CON, n = 7), followed by up to 11 wk of SL racing. Race performance was measured at SL during the lead-in period and repeatedly postintervention. Training volume (in kilometers) and load (session rating of perceived exertion) were calculated for all sessions. Hemoglobin mass was measured via CO rebreathing. Between-groups differences were evaluated using effect sizes (Hedges g). Results: Performance improved in both ALT1600 (mean [SD] 1.5% [0.9%]) and ALT1800 (1.6% [1.3%]) compared with CON (0.4% [1.7%]); g = 0.83 (90% confidence limits −0.10, 1.66) and 0.81 (−0.09, 1.62), respectively. Season-best performances occurred 5 to 71 d postaltitude in ALT1600/1800. There were large increases in training load from lead-in to intervention in ALT1600 (48% [32%]) and ALT1800 (60% [31%]) compared with CON (18% [20%]); g = 1.24 (0.24, 2.08) and 1.69 (0.65, 2.55), respectively. Hemoglobin mass increased in ALT1600 and ALT1800 (∼4%) but not CON. Conclusions: Larger improvements in performance after altitude training may be due to the greater overall load of training in hypoxia compared with normoxia, combined with a hypoxia-mediated increase in hemoglobin mass. A wide time frame for peak performances suggests that the optimal window to race postaltitude is individual, and factors other than altitude exposure per se may be important.
Alice M. Wallett, Naroa Etxebarria, Nicole A. Beard, Philo U. Saunders, Marijke Welvaert, Julien D. Périard, Andrew J. McKune, and David B. Pyne
Purpose: The risk of exercise-induced endotoxemia is increased in the heat and is primarily attributable to changes in gut permeability resulting in the translocation of lipopolysaccharides (LPS) into the circulation. The purpose of this study was to quantify the acute changes in gut permeability and LPS translocation during submaximal continuous and high-intensity interval exercise under heat stress. Methods: A total of 12 well-trained male runners (age 37  y, maximal oxygen uptake [VO2max] 61.0 [6.8] mL·min−1·kg−1) undertook 2 treadmill runs of 2 × 15-minutes at 60% and 75% VO2max and up to 8 × 1-minutes at 95% VO2max in HOT (34°C, 68% relative humidity) and COOL (18°C, 57% relative humidity) conditions. Venous blood samples were collected at the baseline, following each running intensity, and 1 hour postexercise. Blood samples were analyzed for markers of intestinal permeability (LPS, LPS binding protein, and intestinal fatty acid–binding protein). Results: The increase in LPS binding protein following each exercise intensity in the HOT condition was 4% (5.3 μg·mL−1, 2.4–8.4; mean, 95% confidence interval, P < .001), 32% (4.6 μg·mL−1, 1.8–7.4; P = .002), and 30% (3.0 μg·mL−1, 0.03–5.9; P = .047) greater than in the COOL condition. LPS was 69% higher than baseline following running at 75% VO2max in the HOT condition (0.2 endotoxin units·mL−1, 0.1–0.4; P = .011). Intestinal fatty acid–binding protein increased 43% (2.1 ng·mL−1, 0.1–4.2; P = .04) 1 hour postexercise in HOT compared with the COOL condition. Conclusions: Small increases in LPS concentration during exercise in the heat and subsequent increases in intestinal fatty acid–binding protein and LPS binding protein indicate a capacity to tolerate acute, transient intestinal disturbance in well-trained endurance runners.