Effect of fatigue on three-dimensional kinematic characteristics of elite and sub-elite swimmers during maximal intermittent exercise.
Exercise-induced fatigue constitutes an important limiting factor for sports performance. In swimming biomechanics, studies have examined the effect of fatigue-induced changes in kinematic variables associated with swim performance. However, as the vast majority of those studies have tested swimmers over race distances, there is a lack of data regarding changes in kinematic parameters during maximal intermittent training sets (a common method used by coaches in order to create physiological adaptations to delay the onset of fatigue). Furthermore, little is known with regard to possible different technical adjustments between swimmers of different skill level during this type of training sets. The identification of distinct kinematic parameters in different skill levels will advance the knowledge and understanding of the kinematical/technical adaptations under fatigue conditions. Therefore, the aim of the present study is to investigate the effect of fatigue on the three-dimensional (3D) kinematic characteristics of elite and sub-elite swimmers during a maximal intermittent training set. Sixteen male sprint swimmers of national/international level volunteered to participate in the study. The swimmers, were assigned to elite or sub-elite group based on their skill/performance level: elite, 8 swimmers whose personal best time in 100-m front crawl equal at least that of the 90% of the World Record; sub-elite, 8 swimmers whose personal best time in the 100-m front crawl slower that 90% of the World Record). The swimmers performed 12 maximal 50-m trials (word to rest ration 1:1.5) which were recorded with the use of 6 synchronised cameras (4 underwater and 2 above water) sampling at a frequency of 25 fields/second. A non-breathing stroke cycle through a 6.75m³ calibrated volume was analysed every second trial, starting from the first one (trials 1,3,5,7,9 and 11). Anthropometric data were obtained with the used of the e-Zone method. All selected stroke cycles were processed using APAS software for the acquisition of the 3D coordinates. The following kinematic variables were calculated: average horizontal velocity (V), stroke length (SL), stroke frequency (SF), stroke index (SI), relative duration (percentile of stroke cycle duration) of stroke phases (entry, pull, push, release, recovery), shoulder and hip roll angles, absolute and relative (percentile of V) intracycle velocity variations (IVV) in horizontal (IVVx and IVVx%), vertical (IVVy and IVVy%) and lateral (IVVz and IVVz%) axes. Two-way ANOVAs (trial x skill level) with repeated measures was performed to compare elite and sub-elite groups across the trials. V and SL decrease significantly (5.3% and 4.9% respectively) throughout the trials as a result of fatigue, while SF remained unchanged. Elite swimmers displayed significantly larger V, SL and SI values across the trials, while SF values were not different between groups. As fatigue developed, the elite swimmers managed to maintain a more constant V and SI by minimising the reductions in their SL. For all swimmers the duration of entry phase remained constant throughout the trials , the duration of pull and push phases increased in trials 5 and 7 before they decreased in trials 9 and 11, while, conversely, the duration of the release and recovery phase decreased in trials 5 and 7 and decreased in trials 9 and 11. Across the trials, the elite swimmers spent significantly longer time in pull and push phase and shorter time in entry, release and recovery phased. Also, as the trials progressed, the elite swimmers adjusted their technique in order to increase the time spent in pull and push phases by reducing the time spent in entry, release and recovery phases. The shoulder and hip roll angles increased for all swimmers during the trials, with the sub-elite swimmers rolling their hips significantly more, whereas no differences between groups in shoulder roll angles were observed. The sub-elite swimmers also significantly increased their hip roll in trials 9 and 11. IVVx decreased during the trials, while IVVx%, IVVy%, IVVz and IVVz% remained unchanged. The elite group displayed significantly larger IVVx, IVVx% and IVVy, while showed smaller IVVz%. In conclusion, the development of fatigue affected significantly performance-related kinematic parameters of all swimmers. The most marked fatigue-induced kinematic changes were observed in the sub-elite swimmers, while the elite swimmers maintained better their technique by making technical adjustments to maintain their V. It is suggested that coaches should monitor the technical elements of their swimmers during maximal intermittent training sets n order to prevent serious impairments in technique. Especially for sub-elite swimmers, this type of training sets should not exceed 300-400 m at a time.