Rev Bras Med Esporte _ Vol. 13, Nº 3 – Mai /Jun, 2007 173e 1. Laboratório de Pesquisas em Educação Física, UNESP, Bauru, SP. 2. Laboratório de Biodinâmica, UNESP, Rio Claro, SP. 3. Universidade Federal do Mato Grosso do Sul, UFMS, Campo Grande, MS. Received in 3/2/05. Final version received in 5/9/05. Approved in 12/9/05. Correspondence to: Rafael Deminice, Rua Arnaldo Victaliano, 971, Igua- temi – 14091-220 – Ribeirão Preto, SP, Brasil. E-mail: deminice@ig.com.br Validity of 30 minutes test (T-30) in aerobic capacity, stroke parameters and aerobic performance determination of trained swimmers Rafael Deminice1, Marcelo Papoti2, Alessandro Moura Zagatto2,3 and Milton Vieira do Prado Júnior1 ORIGINAL ARTICLE Keywords: Swimming. Anaerobic threshold. Stroking parameters. 30 minutes test. Performance. ENGLISH VERSION tion of the aerobic training intensity(5-6) and performance prediction of long distance events(2). However, the determination of the anaer- obic threshold using the blood lactate concentration needs specif- ic equipment and has a financial acquisition and operational cost, which is not possible for most swimming teams in Brazil. More- over, it consists of an invasive test which requires care with hy- giene and safety(2), limiting thus, its use in most clubs and fitness centers. Therefore, many researchers try to make less costly and of easy application assessment protocols available, which also pre- cisely and reliably evaluate and monitor training(1-2,6-7). Olbrecht et al.(6) have developed the T-30 test, which consists in moving the longest distance in 30 minutes at regular rhythm from beginning to end of the test. The mean velocity of the T-30 test (VT-30) has been highly correlated with the anaerobic velocity thresh- old(6,8-11) and with swimming performance(12), which is non-invasive and of easy application. Nevertheless, the swimming mechanics also plays a crucial role in the myriad of determinant factors of swimming performance and should be considered in the assessments. It has been demon- strated that the dislocation velocity in swimming is the product of the stroke rate (SR) by the stroke length (SL) and variations in the swimming velocity by training and lack of training mainly occur by modifications in SR and SL(13-14). For this reason, these variables have been the aim of studies on elite swimming(13,15), educational status(16), disabled subjects(17) and for technical analysis between swimmers and triathletes(18). Costill et al.(7) presented the stroke index (SI) as the product of the swimming velocity by the distance completed per stroke cy- cle, and found significant correlations among oxygen uptake (VO2), swimming velocity and this variable. These authors demonstrate that the swimmer’s energy cost in crawl depends on the tech- nique of his/her stroke. Keskinen and Komi(19) demonstrated that the relationship between SR and SL is influenced by the increase of the effort intensity. When the swimming velocity is lower than the anaerobic threshold intensity, the swimmers are able to con- trol the velocity and simultaneously keep the stroke length steady. However, when the effort is performed at intensities above the anaerobic threshold, a progressive reduction in SL is observed, this fact being relied on the development of local muscle fatigue. Dek- erle et al.(8) highlighted that the swimmer should be able to choose the SR corresponding to the lowest energy cost during his event, suggesting hence a relationship between physiological and techni- cal parameters in swimming. Langeani et al.(20) demonstrated the occurrence of increase and sudden decrease of SR and SL, re- spectively, following the lactacidemia behavior in progressive ex- ercise. Thus, it was shown that dynamic lactate balance may be ob- served during extensive exercise at intensities corresponding to the Lan(6,21); hence, this metabolic balance must reflect in the be- havior of technical parameters in swimming. However, studies ABSTRACT The aim of the present study was to verify the use of the 30 minutes velocity (VT-30), stroke rate (SR), stroke length (SL) and stroke index (SI), obtained in T-30 test, as non-invasive methods for trained swimmer’s aerobic performance and technical determi- nation. Fourteen swimmers accomplished three efforts along 400 m (85, 90 and 100% of the maximum effort) for anaerobic thresh- old speed determination (ATS) corresponding to 3.5 mM lactate fixed concentration, as well as a 30 minutes maximal effort (VT- 30). SR, SL, and SI were calculated within the 10 m midsection of the swimming pool (clean swim) for T-30 test (SRT-30, SLT-30 and SIT-30) and progressive test. Through the relation between ATS and stroke parameters in progressive test, stroke rate threshold (SRT), stroke length threshold (SLT) and stroke index threshold (SIT) were determined. The time to complete 400 m at maximal effort was considered as performance parameter (P400). No significant difference was found between the ATS (1.29 ± 0.07 m.s-1) and VT- 30 (1.29 ± 0.08 m.s-1), along with a correlation (r = 0.90). The val- ues of SRT (33.6 ± 4.14 cycles/min) and SRT-30 (34.9 ± 3.53 cy- cles /min) and of SLT (2.09 ± 0.20 m/cycle) and SLT-30 (2.09 ± 0.20 m/cycle) also had no significant differences. Significant correla- tions (p < 0.05) were also found between VT-30 and P400 (r = 0.95); SRT and SRT-30 (r = 0.73); SLT and SLT-30 (r = 0.89) and SIL and SIT-30 (r = 0.94). It was concluded that the VT-30 shows reli- ability for training monitoring, performance prediction and techni- cal parameters determination in swimmers. INTRODUCTION The monitoring and assessment of physiological variables and performance in sports training may be factors which determine success of elite swimmers. It is known nowadays that the lactaci- demia technique has been a trustworthy and sensitive assessment, prescription and alterations instrument derived from the training state of this modality(1-3). An analysis of the relationship between blood lactate concentration ([LAC]) versus swimming velocity (V), represented by curves obtained through incremental tests, show improvement, stability or degradation of the aerobic capacity of the swimmer(3). The determination of the anaerobic threshold (Lan) through the use of the blood lactate concentration identifies the highest exer- cise intensity in which the ATP resynthesis is performed by the aerobic metabolism(4), which is represented by a parameter of aer- obic capacity. Lan has been used for training monitoring(3), prescrip- 174e Rev Bras Med Esporte _ Vol. 13, Nº 3 – Mai /Jun, 2007 which investigated the T-30 in order to relate the mechanical and physiological parameters with the aerobic capacity test and perfor- mance of swimmers are still insufficient in the literature. There- fore, the aim of the present study was to verify the use of the VT- 30 and stroke parameters (SR, SL and SI) obtained with the T-30 performance as non-invasive instruments in the aerobic capacity assessment, swimming technique and performance prediction of swimmers. MATERIAL AND METHODS Participants Fourteen swimmers (9 males and 5 females), aged range of 15.9 ± 1.9 years, members of the swimming team of Bauru-SP, volun- tarily participated in the present study. Prior to the study, the vol- unteers signed a written consent form approved by the Ethics Committee of the UNESP, Rio Claro. The athletes performed regular training and have participated in state and national competitions for over three years. General char- acteristics of the participants are present in table 1. Blood samples 25 µl of blood were collected from the earlobe for lactate con- centration measurement [LAC]. The samples were stored in 1.5 ml Eppendorf tubes containing 50 µl of sodium fluoride at 1% (NaF). The homogeneity was analyzed in an electrochemical YSI lactate meter; model 1500 Sport (YSI, Ohio, USA). Lactate concentrations were expressed in mM. Determination of the stroke parameters (SR, SL and SI) In order to determine the stroke rate (SR), stroke length (SL) and stroke index (SI), in the VLan and T-30 tests, an S-VHS Pana- sonic M9000 camera was used. The camera was placed parallel to the swimming pool lanes, registering only the swimming in the ten central meters of the pool (figure 1). In order to have only the 10 meters of clean swim registered (with no influence of the im- pulse of the laps and exits), colored balloons were set in the lanes in which the participants performed the swims 7.5 meters away from the exit boarder, as shown in figure 1. TABLE 1 General characteristics of the studied athletes (n = 14) Weight Height Wingspan Performance 400 m (kg) (cm) (cm) (m.s-1) Mean 62.5 171.1 175.2 1.38 SD 09.1 007.7 009.5 0.09 Procedures The study was performed in a semi-Olympic swimming pool (25 x 12 meters), of SESI, Bauru-SP (Brazil), with water temperature at 27oC ± 1oC. Two tests in crawl with 48-hour interval between them were performed. The swimmers previously performed a standardized warm-up period of approximately 1000 m in crawl and at intensity subjectively determined by the athletes and coaches as ‘easy’. Determination of the anaerobic threshold velocity (VLan) and maximal performance in 400 m crawl (P400) In order to determine the VLan, the protocol validated by Pereira et al.(22) was used. In this protocol the swimmers were submitted to three progressive efforts of 400 meters at intensities correspond- ing to 85, 90, and 100% of maximal velocity for the distance. A three-minute interval was performed between each swim. The three trials were initiated with exits from the water. The partici- pants were verbally encouraged during the whole test and received visual information for swimming intensity control. Blood samples were collected (25 µl from the earlobe) one minute after the end of each swim and one, three and five minutes after the end of the test for lactacidemia analysis. For each swim, the mean velocity and blood lactate concentration were calculated. The anaerobic threshold velocity (VLan) was assumed as the swimming velocity corresponding to the lactate steady concentration of 3.5 mM in the lactate versus velocity ratio by exponential growth curve ad- justment(22). The 400 m effort performed at 100% was assumed as the 400 m maximal performance parameter (P400). Determination of the mean velocity in 30 minutes (VT-30) In the T-30 test, the athletes were told to swim the longest dis- tance possible in 30 minutes. The VT-30 was determined by the ratio between the distance swum (m) and the swimming time (1800 s). Blood collections were performed at one, three and five min- utes after the end of the test for lactacidemia analysis. Figura 1 – Modelo representativo dos procedimentos de determinação dos parâmetros de braçada nos 10m de nado limpo: balões coloridos (BC); posicionamento da câmera de vídeo (CV); perspectiva de registro (––); margem de saída (MS). The images analysis was performed with the aid of the Studio DC10 Plus software. The images recorded by the camera at 30 Hz were digitalized and analyzed shot by shot at 0.03 s for accurate determination of the time spent in order to perform the 10 m clean swimming as well as four complete stroke cycles. The stroke rate (SR) was determined by the adapted four-cycle method by Kennedy et al.(23). The SR was corresponding to the four stroke cycle (sc) by the time spent ratio in order to complete it in the ten meters of clean swimming (equation 1). The swimming velocity (V) was determined by the ratio between ten meters of clean swimming by the time spent in order to complete it (equa- tion 2). The stroke length in 10 m was determined by the ratio between V and SR (equation 3). The stroke index (SI) was calculat- ed according to Costill et al.(7), through the product of the swim- ming velocity of clean swimming by the stroke length (equation 4). All stroke parameters were calculated in 400 m during the pro- gressive test and at each four hundred meters in the T-30 test. SR = 4 sc/time for 4 sc x 60 s (equation 1) V = 10 m/time for 10 m (equation 2) SL = V/SR (equation 3) SI = V x SL (equation 4) Rev Bras Med Esporte _ Vol. 13, Nº 3 – Mai /Jun, 2007 175e The stroke rate in 30 minutes (SRT-30), stroke length in 30 min- utes (SLT-30) and stroke index in 30 minutes (SIT-30) were deter- mined through the mean of the SR, SL and SI in 30 minutes val- ues. The values corresponding to the stroke rate threshold (SRLan), stroke length threshold (SLLan) and stroke index threshold (SILan) were determined by linear interpolation from the ratio between anaerobic threshold velocity and stroke parameters obtained in the progressive test, as seen in figure 2. DISCUSSION The greatest advantage of using indirect methods in the training routine of swimmers is mainly concerned with low cost and easy application. Although the T-30 is a widely used methodology in the determination of VLan in swimming, it presents some limitations, since the swimmers should be instructed to swim the longest dis- tance within the pre-set time (30 minutes), which many times is influenced by the degree of motivation. Moreover, this methodolo- gy does not consider the participation of the anaerobic metabo- lism involved. Thus, it is possible that a given swimmer presents improvement in the mean velocity obtained in the T-30 (V-T30) as a result of application of training sessions with the aim to develop lactate tolerance. However, the anaerobic training effects over the V-T30 seem to be modest concerning aerobic training effects(10). In the present study, the V-T30 was not significantly different from the VLan. This finding corroborates the results by Olbrecht et al.(6) who in their study used the two-velocities test (2x400 m) to determine the aerobic capacity, where the VLan was assumed as the swimming velocity corresponding to the 4 mM steady concen- tration. In order to verify the VLan in the present investigation, we used pauses of only 3 minutes between efforts, with the purpose to optimize the time in the tests. Therefore, the lactate steady concentration of 3.5 mM was adopted(22,24), not the 4.0 mM as it is usually used(4,25). The utilization of this concentration clashes with Heck et al.(4), who suggest the steady concentration of 3.5 mM only for protocols with stages with duration up to 3 min, which is lower than the ones used in this study (4 to 5 min). The use of the 4 mM steady concentration suggested by these authors when the stages duration is of 5 minutes seems to overestimate the VLan in swimming, if the pause between incremental efforts is small. This fact is probably due to the existence of residual effects of metab- olism and fatigue specific to the previous stages, once the incre- mental tests for determination of the VLan, may be considered dependent protocols(24). In the present study, the correlation tests outcomes show that the best VLan predictor was the VT-30 (r = 0.90), which did not present differences between the two variables. This result con- firms the findings in the literature, reinforcing the possibility to use the T-30 test as a determinant index of aerobic capacity in swim- ming(2,6,8-11). Besides that, the mean concentration of peak lactate found at the end of the T-30 test was of 3.76 ± 1.65 mM, a value very close to the lactate concentration used in the test for VLan determination. Dekerle et al.(8) observed peak lactate concentra- tion of 3.65 ± 1.58 in the T-30 test, a value also close to 3.5 mM. Olbrecht et al.(6) while using the steady concentration of 4 mM did not find significant difference either, besides finding a high correla- tion between the VLan and mean velocity in 30 minutes, with peak lactate concentration of 4.01 ± 0.75 mM. The VT-30 also acted as a good 400 meters performance predictor, showing a high correla- tion (r = 0.95). Pioneering studies have used the swimming final time or veloc- ity based on the total final time by the swimming distance ratio for stroke parameters determination (SR and SL)(13-14), a procedure which considers the influence of the exits and laps impulse at each swimming segment. In order to calculate the SR, SL and SI in the present study, we used the clean swimming velocity (with no in- fluence of the exits and laps impulse). The use of this method allows the real calculation of the swimmer’s technical skill, since it decreases the interference by particularities in the exits and laps of the athletes. Craig et al.(14) demonstrated that when calculating the stroke length for uniform laps, the values decrease in 5%. In our study, the determination of the stroke thresholds were performed from the plotting of the linear ratio between anaerobic threshold velocity and stroke parameters (figure 2). Keskinen and Komi(19) while studying different ratios between the stroke param- eters in different exercise intensities, reported that the swimming Figura 2 – Exemplo de regressão linear para determinação da freqüência de braçada de limiar (fBLan), comprimento de braçada de limiar (CBLan) e índice de braçada de limiar (IBLan) no teste progressivo Statistical analysis The t-Student test for dependent samples and the Pearson cor- relation test were used to compare and verify possible associa- tions of V, SR, SL and SI derived from the VLan and T-30 tests, respectively. The Pearson correlation test was also used to verify associations among the parameters obtained from the VLan and T- 30 tests with the maximal performance of 400 m crawl. In all cas- es the significance level was pre-set for p ≤ 0.05. RESULTS The results are expressed in mean and standard deviation. The VLan (1.29 ± 0.07 m.s-1) was not significantly different from the VT-30 (1.29 ± 0.08 m.s-1) and they were highly correlated (0.90). Moreover, 3.76 ± 1.65 mM of blood lactate were verified after the T-30. Significant differences between SRLan and SRT-30 and between SLLan and SRT-30 were not found, contrary to what occurred be- tween SILan and SIT-30 (table 2). Additionally, correlations of 0.73, 0.89 and 0.94 for SR, SL and SI values respectively, derived from the VLan and T-30 tests were verified. The P400 (1.38 ± 0.09 m.s-1) presented significant correlations with the VLan (0.94) and VT-30 (0.95). TABLE 2 Mean values and SD of stroke rate threshold (SRLan), stroke rate in 30 minutes (SRT-30), stroke length threshold (SLLan), stroke length in 30 minutes (SLT-30), stroke index threshold (SIBLan) and stroke index in 30 minutes (IBT-30) SR (cycles/min) SL (m/cycle) SI SRLan SRT-30 SLLan SLT-30 SILan SIT-30 Mean 33.9 34.9 2.09 2.09 2.44 2.53* SD 4.14 3.53 0.20 0.20 0.29 0.32* * p < 0.05 concerning SILan. 176e Rev Bras Med Esporte _ Vol. 13, Nº 3 – Mai /Jun, 2007 velocity and the stroke rate were kept practically steady until the anaerobic threshold intensity was reached. However, a significant decrease in the stroke length curve was observed when this thresh- old intensity was surpassed. In a recent study, Langeani et al.(20) found an increase and a dramatic decrease in the stroke rate and length respectively, in progressive exercise of 6 increments which corresponded with the blood lactate curve behavior, with high cor- relations between lactate threshold velocities and stroke parame- ters threshold (V-LT vs V-SRT, r = 0.98; V-LT vs V-ST, r = 0.96), suggesting the use of these parameters as alternatives to the in- vasive lactacidemia tests for the VLan determination. The relationship between swimming velocity and stroke param- eters observed in this study presented a linear behavior (r2 = 0.99), clashing with the findings by Keskinen and Komi(19). A possible ex- planation for these differences may be the use of only 3 points in the determination of the stroke parameters concerned with the VLan, and may not reliably reflect the behavior of the swimming mechanical characteristics due to the increase of the exercise in- tensity. The non-difference found from the SRLan and SLLan values with SRT-30 and SLT-30, respectively, besides the significant correla- tions found between these parameters, corroborate the hypothe- sis that the SR and the SL are directly related with fatigue in swim- ming(7,15). These findings show that stroke length is spontaneously kept at steady intensities. They also suggest the existence of a reflex technical balance of the lactate dynamic balance observed during prolonged exercise at intensities corresponding to the VLan. Thus, SR and SL corresponding to the VLan and/or the V-T30 may be useful parameters for the control and training prescription in- tensities and evaluation of the swimming mechanics. Improvement in the swimming mechanics, especially in aerobic bouts, will be probably reflected in changes in these parameters, being able to influence the increase of the swimming velocity during the com- petition(8). The majority of studies which try to relate swimming mechani- cal parameters and physiological aspects have been using the anaer- obic threshold determined through the ratio between blood lactate concentration versus swimming velocity from incremental swims(11,19-20), a methodology also used in the present study. Perei- ra et al.(22) highlighted that incremental protocols may fail or indi- cate an unsuitable training intensity. Therefore, the maximal swim- ming velocity which can be kept with the maximal balance of lactate production and removal, determined from the maximal lactate steady state protocol(4) (MLSS), may represent the most suitable intensity to control and improve the swimming technique during the aerobic training. Thus, further studies are necessary in order to relate the MLSS and swimming technical parameters. The results of the present study suggest the use of the T-30 as a non-invasive and low cost instrument in aerobic capacity assess- ment, determination of parameters concerned with the swimming technique as well as prediction of 400 m performance in trained swimmers. ACKNOWLEDGMENTS The authors thank the collaboration of the swimming coach André Bar- bosa Velosa from SESI Prata-Unimed,Bauru for having made his athletes available; Prof. Dr. Sérgio Tossi Rodrigues, for having made the images analysis possible in his laboratory and Marina Álvares Denardi, for the spell- ing review. All the authors declared there is not any potential conflict of inter- ests regarding this article. 1. Denadai BS. Avaliação aeróbia: determinação indireta do lactato sanguíneo. Mo- trix: Rio Claro; 2000. 2. Maglischo EW. Nadando ainda mais rápido. São Paulo: Ed. Manole; 1999. 3. Pyne DB, Lee H, Swanwick KM. Monitoring the lactate threshold in world-ranked swimmers. Med Sci Sports Exerc. 2001;33:291-7. 4. Heck H, Mader A, Hess G, Mucke S, Muller R, Hollman W. Justification of 4- mmol/l lactate threshold. Int J Sports Med. 1985;6:117-30. 5. Madsen O, Lohberg M. The lowdown on lactates. Swimming Technique. 1987; 24:21-6. 6. Olbrecht J, Madsen Ø, Mader A, Liesen H, Hollmenn W. Relationship between swimming velocity and lactic acid concentration during continuous and intermit- tent training exercises. Int J Sports Med. 1985;6:74-7. 7. Costill D, Kovaleski J, Porter D, Kirwan J, Fielding R, King D. Energy expenditure during front crawl swimming: predicting success in middle-distance events. Int J Sports Med. 1985;6:266-70. 8. Dekerle J, Sidney M, Hespel JM, Pelayo P. Validity and reliability of critical speed, critical stroke rate, and anaerobic capacity in relation to front crawl swimming performances. Int J Sports Med. 2002;23:93-8. 9. Weiss M, Bouws NE, Weicker. Comparison between the 30-minutes-test and 300 m-step-test according to Simon in the women’s national swimming team. Int J Sports Med. 1998;9:379 (Abstract). 10. Olbrecht J. The science of winning: planning, periodizing and optimizing swim training. Sponson; 2000. 11. Deminice R, Prado Júnior MV, Papoti M, Zagatto AM. Utilização de métodos não-invasivos como indicadores da capacidade aeróbia e da performance em natação competitiva. Rev Bras Ciên e Mov. 2003;(edição especial):S130. 12. Dos Santos ILG, Papoti M, Gobatto CA, Zagatto AM. Utilização de protocolos não-invasivos para estimar a performance de 200 e 400 metros em natação. Rev Bras Ciên e Mov. 2004;(edição especial):S200. 13. Craig JRAB, Pendergast DR. Relationships of stroke rate distance per stroke and velocity in competitive swimming. Med Sci Sports Exerc. 1979;11:278-83. 14. Craig JRAB, Skehan PL, Pawelczyk J, Boomer WL. Velocity, stroke rate, and distance per stroke during elite swimming competition. Med Sci Sports Exerc. 1985; 17:625-34. 15. Wakayoshi K, Acquisto LJD, Cappaert JM, Troup JP. Relationship between oxy- gen uptake, stroke rate and swimming velocity in competitive swimming. Int J Sports Med. 1995;16:19-23. 16. Pelayo P, Wille PL, Sidney M, Berthoin S, Lavoiej M. Swimming performances and stroking parameters in non skilled grammar school pupils: relation with age, gender and some anthropometrics characteristics. J Sports Med Phys Fitness. 1997;37:187-93. 17. Pelayo P, Sidney M, Moretto M, Wille PL, Chollet D. Stroking parameters in top- level swimmers with a disability. Med Sci Sports Exerc. 1998;31:1839-43. 18. Toussant HM. Differences in propelling efficiency between competitive and tri- athlon swimmers. Med Sci Sports Exerc. 1990;22:409-15. 19. Keskinen KL, Komi PV. Stroking characteristics of front crawl swimming during exercise. J Appl Biomech. 1993;9:219-26. 20. Langeani AK, Belli T, Baldissera V, Ribeiro LFP, Costa PHL. Identificação do li- miar de lactato através de parâmetros da mecânica de nado em atletas adoles- centes. Motriz. 2003;9:S61. 21. Stegmann H, Kindermann W. Comparison of prolonged exercise tests at the individual anaerobic threshold and the fixed anaerobic threshold of 4 mmol.l-1 lactate. Int J Sports Med. 1982;3:105-10. 22. Pereira RR, Papoti M, Zagatto AM, Gobatto CA. Validação de dois protocolos para determinação do limiar anaeróbio em natação. Motriz. 2002;8:63-8. 23. Kennedy P, Brown P, Chengalur SN, Nelson RC. Analysis of male and female Olympic swimmers in the 100-meter events. Int J Sport Biomech. 1990;6:187- 97. 24. Kiss MAPDM, Fleishmann E, Cordani IK, Kalinovsky F, Costa R, Oliveira FR, et al. Validade da velocidade de limiar de lactato de 3,5mmol/L-1 identificada atra- vés de teste em pista de atletismo. Rev Paul Educ Fis. 1995;9:16-25. 25. Sjodin B, Jacobs I. Onset of blood accumulation and marathon running perfor- mance. Int J Sports Med. 1981;12:23-6. REFERENCES