Services on Demand
- Cited by SciELO
- Access statistics
Print version ISSN 1517-8692On-line version ISSN 1806-9940
Rev Bras Med Esporte vol.12 no.6 Niterói Nov./Dec. 2006
Critical velocity as a noninvasive method to estimate the lactate minimum velocity on cycling*
La velocidad crítica como un método no invasivo para estimar la velocidad de lactato mínimo en el ciclismo
Wolysson Carvalho Hiyane; Herbert Gustavo Simões; Carmen Sílvia Grubert Campbell
The lactate minimum velocity (LMV) represents the equilibrium point between blood lactate (lac) production and removal. With the purpose of analyzing the validity of critical velocity (CV) as a non-invasive method to estimate the LMV on outdoor cycling, 15 cyclists (67.9 ± 5.7 kg; 1.70 ± 0.1 m; 26.7 ± 4.2 years) performed all-out tests on distances of 2, 4 and 6 km on velodrome. The CV was identified by distance-time model from combinations of 2 and 4 km (CV2/4), 2 and 6 km (CV2/6), 4 and 6 km (CV4/6) and 2, 4 and 6 km (CV2/4/6). The LMV was identified during 6 x 2 km incremental bouts after a lactic acidosis induced by the all-out 2 km. The lower lac during test identified the LMV visually (LMVv) and by applying a polynomial function (LMVp). No differences were observed between LMVv (33.3 ± 2.5 km.h1) and LMVp (33.1 ± 2.6 km.h1). Apart from CV4/6 (34.6 ± 3.5 km.h1), the values of CV2/4 (38.0 ± 2.2 km.h1), CV2/6 (36.1 ± 2.4 km.h1) and CV2/4/6 (36.1 ± 2.5 km.h1) differed from LMVp and LMVv (P < 0,001). The authors concluded that, besides being ~1 km/h above the LMV, the CV determined through predictive series of longer duration (4 and 6 km approximately 6 and 10 min) did not differ statistically from LMV and presented a high correlation and agreement to each other. However, it is necessary to investigate whether the CV reflects the balance between lac production and removal during long-term exercise on outdoor cycling.
Keywords: Anaerobic threshold. Indirect methods. Predictive bouts. Maximal lactate steady state.
La velocidad de lactato mínimo (VLM) representa el punto de equilibrio entre la producción y la remoción de lactato sanguíneo (lac). Con el objetivo de analizar la validez de la velocidad crítica (VC) como método no invasivo de estimar la VLM en el ciclismo "outdoor", 15 ciclistas (67,9 ± 5,7 kg; 1,70 ± 0,1 m; 26,7 ± 4,2 años) percorrieron distancias de 2, 4 y 6 km en velódromo en el menor tiempo posible. La VC fue identificada por el modelo distancia-tiempo a partir de las combinaciones de series predictivas de 2 y 4 km (VC2/4), 2 y 6 km (VC2/6), 4 y 6 km (VC4/6) y 2, 4 y 6km (VC2/4/6). Para la identificación de VLM fue realizada una serie de 2km a máxima velocidad, seguida de 6 series incrementales de 2km con 1 minuto de pausa para dosaje de lac. La VLM fue identificada visualmente (VLMv) y aplicando función polinomial (VLMp). No se observaron diferencias entre VLMv (33,3 ± 2,5km.h1) y VLMp (33,1 ± 2,6km.h1). A excepción de VC4/6 (34,6 ± 3,5km.h1), los valores de VC2/4 (38,0 ± 2,2km.h1), VC2/6 (36,1 ± 2,4km.h1) y VC2/4/6 (36,1 ± 2,5km.h1) difirieron de VLMp y VLMv. Concluimos que, a pesar de ser ~1km/h por encima de VLM, la VC identificada a partir de series predictivas de mayor duración (4 y 6km - aproximadamente 6 y 10 min) no tienen diferencia estadística y presentan alta correlación y concordancia con VLM. A pesar de esto, es necesario investigar si la VC representa un equilibrio entre remoción y producción de lac durante los ejercicios de larga duración en ciclismo outdoor.
Palabras-clave: Límite anaeróbico. Métodos indirectos. Series predictivas. Máxima fase estable de lactato.
Davis and Gass(1), observed during incremental test performed after high intensity exercise that the blood lactate concentrations decreased in the first incremental loads until a minimum point and returned to increase in the subsequent loads. These authors concluded that the exercise intensity corresponding to the balance point between blood lactate production and removal could be identified during incremental bouts performed after induction of metabolic acidosis. Later, this protocol was improved and called minimum lactate(2) (ML). Moreover, several subsequent studies(3-9) were conducted in order to verify its validity.
Simões et al.(5), analyzing the relation between ML and other protocols proposed for identification of the maximal lactate steady phase (MLSP) in runners, did not find statistically significant differences between the minimum lactate velocity (VLM) and the velocities corresponding to the individual anaerobic threshold and to the steady concentration of 4 mM of blood lactate. These authors also observed that the MLV did not differ from the running velocity in which blood lactate steady phase was observed during long duration exercise.
MacIntosh et al.(4) verified in a study with 14 cyclists/triathletes of both sexes, that the ML protocol was valid for prediction of physical exercise intensities corresponding to the MLSP. Bacon and Kern(9) also evidenced that the MLSP and MLV intensities, identified during running tests in physically active individuals, were not different from each other.
Although it is an interesting method for identification of an exercise intensity which represents the MLSP, the determination of the MLV depends on costly equipment and evaluators skilled in blood collection and lactate dosing, which makes its wide application not viable.
An alternative would be the use of indirect methods(10) for identification of velocities similar to the MLV and MLSP. Among these methods we find the determination of the critical velocity (CV) CV has been suggested as the intensity of physical exercise which can be sustained for an extensive period of time without exhaustion(11). Moreover, its determination involves non-invasive and inexpensive procedures which may be easily applied on field tests.
The VC validity in estimating the MLSP is still very controversial. Wakayoshi et al.(12) observed that the CV could identify the MLSP in swimmers. In this study, the participants performed repeated series of 400 meters in velocities corresponding to 98, 100 and 102% of the CV, being 4 series at each intensity. It was verified that in exercises performed at 100% of the CV the lactate concentrations stabilized, while at 102% of the CV the concentrations increased between the first and fourth series. It was concluded hence, that the CV could represent the MLSP. Kokubun(13) evidenced similar results in his study with 48 swimmers from both sexes. The study showed that during 5 series of 400 meters with 30 seconds of pause no alterations in the lactate concentrations were observed when the series were performed at 100% of the CV. On the other hand, when these series were performed at intensities higher than the CV, accumulation of blood lactate was observed.
Although some studies have established that the CV may estimate the MLSP in swimming, studies performed with other exercise modalities show that the critical velocity-power overestimates the intensities corresponding to individual anaerobic threshold, MLSP and minimum lactate(8,14). In a study performed with 20 runners, Simões et al.(5) observed that the CV (292,1 ± 17,5 m.min-1) overestimated the MLV (281,0 ± 14,8 m.min-1). Similarly, Denadai, Gomide and Greco(15) verified that the CV determined in running in soccer players overestimated the MLSP.
As far as we know, no study has compared the CV determined on outdoor cycling with other protocols of aerobic evaluation such as anaerobic threshold, ML and MLSP, where the majority of the studies were performed in cycle ergometer with power measured in Watts. McLellan and Cheung(14) verified in 14 individuals that the CP determined in cycle ergometer by the power- 1/time linear relation overestimated in 13% the MLSP. However, the CV may also suffer influence of the mathematical model applied(16-17) and the predictive series combinations used for its determination. Nevertheless, there are few studies investigating the effects of different combinations of predictive series over the CV identification as well as its comparison with other protocols which propose the identification of the MLSP, especially on outdoor cycling.
Considering that laboratory tests usually lose in specificity, we propose in this study the utilization of tests which reach for the highest specificity in the sport, namely track tests with the utilization of the cyclist's own bicycle. Therefore, the aim of the present study was to compare the values of CV identified by different combinations of predictive series on field tests performed on cycling as well as the CV values identified by different combinations of predictive series with velocities of minimum lactate determined by visual inspection and by the application of polynomial function.
Fifteen cyclists with 6,3 ± 3,2 years of practice participated in the research, whose characteristics are represented in table 1. The participants answered a questionnaire of anamnesis and signed a free and clarified consent form about the study's procedures. Each participant was instructed to remain hydrated as well as to have the last meal 3 hours prior to the tests sessions. Ingestion of alcoholic beverages and intense physical exercises were not allowed during the 24 hours prior to the tests. The methods used in the present study were approved by the Ethics in Human Research Committee of the Catholic University of Brasília.
All tests were performed in a velodrome in of 400 meters in Brasília DF, with the volunteers using their own bicycles. The procedures were always conducted at the same time of the day and consisted of 3 tests with the purpose to measure performance in the 2, 4 and 6 km distances, besides an incremental bout after hyperlactatemia induction for identification of the MLV. All tests were performed within a 2 weeks period. The tests were randomly applied, except for the 6 km performance, which was the first one to be applied. Unfavorable climate conditions such as rain and gusty winds were avoided. The mean velocity as well as the correct measurement of the velodromein were taken by a cyclecomputer (ASSIZE, CYCLOCOMPUTER).
All-out tests of 2, 4 and 6 km
After a 10-minute warm-up pedaling in their own bicycles between 90 and 100 rpm, the volunteers completed 2, 4 and 6 km all-out tests, always at different days and with an interval of at least 24 h. These tests were chosen as predictive series of CV since they were finished between approximately 1 to 10 minutes, as proposed by Poole(18).
Determination of the critical velocity
The CV was determined in all participants from the distance-time linear model(12). Linear regression between the completed distance (km) and the time used in order to complete this distance (h) was performed. The inclination of the distance-time regression line was defined as critical velocity (CV) (figure 1).
Having the 2, 4 and 6 km predictive series performance as starting point, combinations for the determination of four indices of different critical velocities were performed. The CV2/4 (2 and 4 Km series), CV2/6 (2 and 6 Km series), CV4/6 (4 and 6 Km series) and CV2/4/6 (2, 4 and 6 Km series) were determined through the distance-time linear model. Such model also allows the identification of the anaerobic work capacity, despite not being object of this study as a parameter.
Identification of the equilibrium point between blood lactate production and removal
It consisted of the ML test application using a MacIntosh et al.(4) modified protocol, being one 2 km all-out series used for blood lactate increase, followed by 8 minutes of recovery with a blood collection in the 7th minute. In the 8th minute an incremental bout consisting of 6 series of 2 km in progressive intensities, with 1 minute pause for capillary blood collection was applied. The intensity of the first series corresponded to 5 Km/h below the mean velocity obtained in 6 km bout previously performed, with increases of 1 km/h at each series of 2 km. The mean velocity of each series was controlled by a cyclecomputer (ASSIZE, CYCLOCOMPUTER) as well as by sound stimulus (whistle). The MLV identification was visually inspected (MLVv, figure 2-A), as well as by polynomial function of second degree for mathematical adjustment of the blood lactate response (MLVp, figure 2-B).
Blood collections and analyses
After local asepsis, a small incision on the earlobe was done with disposable material for collection of 25 µL of capillary blood using heparinized and calibrated capillary tubes. The collections occurred in the intervals between exercise series during the minimum lactate test and stored in Eppendorfs tubes containing 50 µL of sodium fluorite 1%. During the collection procedures, all materials were disposable in order to avoid any kind of contamination. Besides that, the first blood drop was discarded in order to avoid blood and sweat mixture. The samples were analyzed through electroenzymatic method, using a lactate analyzer (Yellow Springs instruments 2.700 STAT).
The data were expressed in mean ± standard deviation (SD). The results of MLVv, MLVp and CV determined by different combinations of predictive series were compared using variance analysis for repeated measures and Tukey test as well as post hoc and the correlations between CV and MLV were determined using the Pearson correlation coefficient. The significance level accepted was p < 0,05. Moreover, the agreement level between LMV and CV determined by different methods was analyzed through the Bland-Altman technique(19).
No statistically significant differences were verified between MLVp and MLVv (table 2) nor between their respective lactate concentrations (table 3) (p > 0,05). The intensities (Km/h-1) corresponding to CV2/4, CV2/6 and CV2/4/6 were statistically different from the MLV identified both visually and by polynomial function (p < 0,001). However, the CV4/6 did not differ from the MLVp and MLVv (table 2).
Relative intensities concerning the CV identified by different methods expressed in % of the MLVv and MLVp, are presented in table 3. The blood lactate concentrations [Lac] corresponding to the MLVv and MLVp are also represented in table 3.
The Pearson correlations between the minimum lactate velocities (MLVp and MLVv) and the variables CV2/4, CV2/6, CV4/6, CV2/4/6 are presented in table 4.
The application of the technique proposed by Bland-Altman showed high level of agreement between MLV and CV determined by different methods (figure 3 A-E), with the mean of the differences (and 95% of the reliability interval) of 2,8 (0,8 to 4,8); 4,6 (0,2 to 9,4) and 2,7 (0,7 to 4,7) km*h-1 respectively between MLV and CV2/6, MLVv and CV2/4, and MLVv and CV2/4/6. However, the best agreement occurred between MLVv and MLVp 0,2 (2,4 to 2) km*h-1, as well as between MLVp and CV4/6 1,2 (2,4 to 4,8) km*h-1.
The present study investigated the effects of different combinations of predictive series in the identification of the CV on cycling as well as analyzed the CV validity on estimating the equilibrium point between blood lactate production and removal (identified by the ML test). The main results of the present study were that the majority of the CV indices identified with different combinations of predictive series overestimated the MLV identified both visually and by application of polynomial function. Nevertheless, it was observed that the combination of longer predictive series (4 and 6 km) may produce CV values that do not differ from the MLV (table 2), with high correlation and agreement between these variables when Pearson correlation and Bland-Altman technique were applied (table 4 and figure 3 A-E).
Moreover, the present study showed that the minimum lactate velocities identified by visual inspection and by polynomial function were not statistically different (table 2), and the blood lactate concentrations corresponding to the MLVv and MLVp were respectively of 3,5 ± 2,3 and 3,1 ± 2,1 (table 3).
Polynomial function is a new method used for the MLCV identification from a response adjustment of the blood lactate during the test. The polynomial function of second order originates an equation which may be derived to mathematically identify the equilibrium point between blood lactate production and removal. Although differences between the MLVv and MLVp were not observed, the utilization of the MLVp should be stimulated since this new methodology avoids misinterpretation which may occur when only visual inspection is used. The Bland-Altman technique showed that the mean of the residual scores between MLVv and MLVp was close to zero (figure 3A) and that the MLV values either visually determined or by polynomial adjustment were within the agreement limits, suggesting that the MLVp is an interesting option. Besides that, further studies have been conducted in our laboratories in which a considerable decrease in the incremental bouts during the test for determination of the MLV has been possible through the application of polynomial function.
Although several studies have demonstrated that the MLV represents the intensity of MLSP(2,4,7,9), the determination of the MLV and the MLSP involves invasive procedures as well as costly equipment. The utilization of the CV in order to estimate the MLV and consequently the MLSP would largely facilitate the procedure. Nonetheless, in the present study the CV overestimated the MLV between 3 to 15% (table 3). Such difference found between the CV and MLV indices may be explained by the influence of the exercise time of the predictive series which originated the CV indices. The CV identification is dependent on the duration of the predictive series, being the CV indices inversely proportional to their duration, which was confirmed in the present study. The CV identified from the combination of the 4 and 6 km series produced indices which were close to each other and did not statistically differ from the MLV, suggesting that longer distances produce critical velocities which would theoretically represent the equilibrium point between blood lactate production and removal. Moreover, the Bland- Altman technique showed that the mean of the residual scores between MLVv and CV4/6 was close to zero suggesting that the agreement level between MLVv and CV4/6 (longer series) is higher than between MLVv and CV2/4/6, CV2/4 and CV4/6 (figure 3, B-E), whose combinations of predictive series include the 2 km bout (shortest duration).
Bishop et al.(16) verified that the combination of loads that allow exhaustion time between 68 and 193 seconds determines higher indices of critical power (201 W) if compared with the loads that allow exhaustion time between 193 and 485 seconds (164 W). Jenkins et al.(20) verified that the critical power presented different values when the 3 lowest (268 W) and the 3 highest (321 W) loads were selected, the latter resulting in higher values of critical power.
According to Poole(18), predictive bouts that are able to be completed between 1 and 10 minutes should be chosen in order to determine the critical power/critical velocity. In the present study these recommendations were followed; however, only when the 4 and 6 km bouts were used (which had duration between 6 and 10 minutes) it was possible to identify CV indices which were similar to the MLV. Despite not being statistically different from the MLV as well as the Bland and Altman technique has confirmed the acceptable level of agreement between the variables, the CV4/6 corresponded approximately to 104% of the MLV. It would be necessary to perform rectangular tests in intensities below, above and at the CV, analyzing the blood lactate response in these intensities in order to really assert that the CV identifies an exercise intensity similar to the MLSP, which was not done in the present study. Therefore, further studies should be conducted in order to affirm whether the MLSP and CV intensities are really similar (and not different from each other) when predictive series with performance times higher than 6 minutes are applied, as in the present study.
It would be interesting to standardize a protocol of CV determination which could estimate the MLSP. The results of the Bland and Altman technique application showed agreement level acceptable between MLVv and the remaining studied parameters (figure 3). Nevertheless, the CV4/6 was the only predictive bout which did not differ in relation to the minimum lactate velocities (table 2) besides presenting the best agreement (figure 3A-E). Thus, the utilization of the CV by coaches and athletes should be cautiously conducted. Despite being a non-invasive method of easy application, low cost and suitability for evaluation of a large number of individuals, the selection of the predictive series should be careful in order to determine values of CV which are close to the intensity corresponding to the equilibrium point between lac production and removal, or MLSp.
We concluded that despite being 1 km/h above the MLV, the CV identified from the predictive bouts of longest duration (4 and 6km approximately 6 and 10 min) does not statistically differ from the MLV. Nevertheless, it is necessary to investigate whether the CV represents a balance between lac production and removal during long duration exercises.
1. Davis HA, Gass GC. Blood lactate concentrations during incremental work before and after maximum exercise. Br J Sports Med. 1979;13:165-9. [ Links ]
2. Tegtbur UWE, Busse MW, Braumann KM. Estimation of an individual equilibrium between lactate production and catabolism during exercise. Med Sci Sports Exerc. 1993;25:620-7. [ Links ]
3. Campbell CSG, Simões HG, Denadai BS. Influence of glucose and caffeine administration on identification of maximal lactate steady state. Med Sci Sports Exerc. 1998;30:327. [ Links ]
4. MacIntosh BR, Esau S, Svedahl K. The lactate minimum test for cycling: estimation of the maximal lactate stady state. Can J Appl Physiol. 2002;27:232-49. [ Links ]
5. Simões HG, Campbell CSG, Baldissera V, Denadai BS, Kokubun E. Determinação do limiar anaeróbio por meio de dosagens glicêmicas e lactacidêmicas em testes de pista para corredores. Rev Paul Educ Fís. 1998;12:17-30. [ Links ]
6. Simões HG, Campbell CSG, Kokubun E, Denadai BS, Baldissera V. Blood glucose responses in humans mirror lactate responses for individual anaerobic threshold and for lactate minimum in track test. Eur J Appl Physiol. 1999;80:34-40. [ Links ]
7. Simões HG, Kushnick MR, Nakamura A, Katsanos CS, Baldissera V, Moffatt RJ. Blood glucose threshold and the metabolic responses to incremental exercise tests with and without prior lactic acidosis induction. Eur J Appl Physiol. 2003; 89:603-11. [ Links ]
8. Simões HG, Denadai BS, Baldissera V, Campbell CSG. Relationships and significance of lactate minimum, critical velocity, heart rate deflection and 3000 m track-tests for running. J Sports Med Phy Fitn. In press, 2005. [ Links ]
9. Bacon L, Kern M. Evaluating a test protocol for predicting maximum lactate steady state. J Sports Med Phys Fitness. 1999;39:300-8. [ Links ]
10. Denadai BS. Avaliação aeróbia: determinação indireta da resposta do lactato sanguíneo. Rio Claro: Motrix, 2000. [ Links ]
11. Monod H, Scherrer J. The work capacity of a synergic muscular group. Ergonomics. 1981;24:329-38. [ Links ]
12. Wakayoshi K, Yoshida T, Udo M, Kasai T, Moritani T, Mutoh Y, et al. A simple method for determining critical speed as swimming fatigue threshold in competitive swimming. Int J Sports Med. 1992;13:367-71. [ Links ]
13. Kokubun E. Velocidade crítica como estimador do limiar anaeróbio na natação. Rev Paul Educ Fís. 1996;10:5-20. [ Links ]
14. McLellan TM, Cheung KSY. A comparative evaluation of the individual anaerobic threshold and the critical power. Med Sci Sports Exerc. 1992;24:543-50. [ Links ]
15. Denadai BS, Gomide EB, Greco CC. The relationship between onset of blood lactate accumulation, critical velocity, and maximal lactate steady state in soccer players. J Strength Cond Res. 2005;19:364-8. [ Links ]
16. Bishop D, Jenkins DG, Howard A. The critical power is dependent on the duration of the predictive exercise tests chosen. Int J Sports Med. 1998;19:125-9. [ Links ]
17. Bull AJ, Housh TI, Johnson GO, Perry SR. Effect of mathematical modeling on the estimation of critical power. Med Sci Sports Exerc. 2000;32:526-30. [ Links ]
18. Poole DC. Correspondence. Med Sci Sports Exerc. 1986;18:703-4. [ Links ]
19. Martelli Filho JA, Maltagliati LA, Trevisan F, Gil CTLA. Novo método estatístico para análise da reprodutibilidade. R Dental Press Ortodon Ortop Facial. 2005;10: 122-9. [ Links ]
20. Jenkins DG, Kretek K, Bishop D. The duration of predicting trials influences time to fatigue at critical power. J Sci Med Sport. 1998;1:213-8. [ Links ]
Received in 4/7/05. Final version received in 21/7/06. Approved in 25/7/06.
All the authors declared there is not any
potential conflict of interests regarding this article.
* Programa de Pós-Graduação Stricto-Senso em Educação Física. Universidade Católica de Brasília-UCB, Brasília-DF, Brasil.