Critical velocity estimates lactate minimum velocity in youth runners

In order to investigate the validity of critical velocity (CV) as a noninvasive method to estimate the lactate minimum velocity (LMV), 25 youth runners underwent the following tests: 1) 3,000m running; 2) 1,600m running; 3) LMV test. The intensity of lactate minimum was defined as the velocity corresponding to the lowest blood lactate concentration during the LMV test. The CV was determined using the linear model, defined by the inclination of the regression line between distance and duration in the running tests of 1,600 and 3,000m. There was no significant difference (p=0.3055) between LMV and CV. In addition, both protocols presented a good agreement based on the small difference between means and the narrow levels of agreement, as well as a standard error of estimation classified as ideal. In conclusion, CV, as identified in this study, may be an alternative for noninvasive identification of LMV.


Introduction
Blood lactate responses ([lac]) during exertion tests have been the focus of several studies (review of Faude, Kindermann, & Meyer, 2009), being the maximal lactate steady state (MLSS) considered the gold standard of aerobic capacity, since it represents the highest intensity of exercise in which [lac] remains in equilibrium during an exercise with constant workload (Beneke, 2003). The MLSS delimits an intensity of exercise with a stable physiological ratio between lactate and pyruvate, oxygen pressure, carbonic acid (HCO 3 -), excessive basicity, oxygen uptake (O 2 ), respiratory exchange ratio (RER), ventilation (VE), and the ventilatory equivalent of oxygen (VE/O 2 ) and carbon dioxide (VE/CO 2 ) (Baron et al., 2003). Thus, the MLSS has been considered the optimal intensity for exercise prescription when training aims benefits associated with the improvement of aerobic capacity (Baron et al., 2008).
However, despite its accuracy on aerobic fitness evaluation, the MLSS protocol is time consuming, and depends upon evaluators with the ability to perform blood sampling and lactate analysis using expensive equipment. This, in turn, reduce the accessibility to such protocol (Franken, Zacca, & Castro, 2011), therefore derailing its application in large samples (Hiyane, Simões, & Campbell, 2006). In this scenario, the lactate minimum test (LM) proposed by Tegtbur, Busse and Braumann (1993), which is characterized by the equilibrium point between the production and removal of blood lactate during an incremental test after performing a high intensity exercise, appears as an alternative when it regards MLSS, since several studies show that the lactate minimum velocity (LMV) agrees with the intensity of MLSS, with the convenience of being performed in a single test session Puga, Kokubun, Simões, Nakamura, & Campbell, 2012;Sotero et al., 2007;Sotero et al., 2009).
However, the use of LMV protocol also characterizes itself as an invasive and expensive procedure, differing from MLSS only when it comes to numbers of sessions in which the volunteers are submitted. One alternative would be the use of indirect methods to identify velocities similar to the LMV and MLSS, such as the critical velocity (CV), which has been used in several studies (review of Leclair, Mucci, Mcgawley, & Berthoin, 2008). Theoretically, CV has been suggested as an intensity of physical exercise that can be sustained for a long period of time without exhaustion (Monod & Scherrer, 1965), being characterized as a noninvasive low cost method that can be easily applied to evaluate aerobic capacity and to identify the intensity for exercise prescription (Leclair et al., 2008).
On the other hand, no study has been found regarding this subject (CV vs. LMV) in youth runners, while other studies have obtained positive results with adult runners  and athletes from other modalities (Altimari, Altimari, Gulak, & Chacon-Mikahil, 2007;Hiyane et al., 2006). Therefore, it is of interest better understand and compare both methods in adolescents, since this population tends to present lower blood lactate concentrations, due to lower enzymes concentration of glycolytic and higher of aerobic pathways (Dotan et al., 2012). In this sense and due to the need of utilizing non invasive and low cost methods with the capability of evaluate aerobic capacity and to identify intensity for exercise prescription in running, the aim of the present study was to compare the CV and LMV in youth runners.

Methods
The present study was approved by the ethics committee of the Catholic University of Brasília (UCB -nº 019/2004). All participants were instructed not to perform exercise and not to drink caffeine beverages during the 24 hours that preceded the experimental procedures. After being informed of the risks and benefits of the study and having signed an informed consent letter, 25 medium and long distance youth runners (table 1) from the Joaquin Cruz Institute were submitted to three experimental sessions performed in a 400m athletics track, with a minimum of 48 hours between them. In the period of collection, all athletes were national sporting level, and were in pre competition period with a training volume between 40-50 miles per week.

3,000m running performance test
The participants performed a 3,000m running test in a 400m athletics track in order to obtain the mean velocity (mV 3,000m) of the test. The volunteers were guided to run the distance as fast as possible. The result obtained was used to calculate the intensity of the stages in the lactate minimum incremental tests and the linear regression equation to obtain the CV.

1,600m running performance test
The participants performed a 1,600m running test in order to obtain the mean velocity (mV 1,600m) of the test. The volunteers were guided to run the distance as fast as possible. The result obtained was used in the linear regression equation to obtain the CV.

Incremental test for determination of lactate minimum velocity
Determination of LMV was performed according to the incremental test proposed by Simões et al. (1998), in which the participants ran 500m at maximum speed in order to induce hyperlactatemia, followed by 10min of recovery and 6 incremental sets of 800m at the intensities of 83, 86, 89, 92, 95 and 98% of the mV 3,000m. The velocities during the incremental tests were controlled by a sonoric stimulus at each 100m. Pauses with 1min of duration between each set were executed in order to collect 25µl of blood from the ear lobe using procedure gloves, disposable lancets and calibrated and heparinized glass capillars. Afterwards, the samples were stored in Eppendorf microtubes containing 50µl of NaF at 1%. Blood lactate responses were analyzed through the electroenzymatic method (Yellow Springs 2700, STAT, OH, EUA). The procedures used to identify the LMV are shown in Figure 1. The running velocity corresponding to the lowest concentration of [lac] during the incremental test was determined by visual inspection (Tegtbur et al., 1993).
After adjust the optimal distance to reach the previously described recommendations (Greco, 2000;Housh et al., 1990;Kranenburg & Smith, 1996), CV was determined through the inclination of the regression line between the 1,600m and 3,000m performance tests results and their respective durations. In this equation, the inclination of the regression line indicates the intensity of velocity correspondent to the CV, where the same can be obtained by the following equation CV (m·s -1 ) = (2 nd distance -1 st distance) / (2 nd duration -1 st duration). Figure  2 shows an example where the inclination of the regression line represents the intensity corresponding to the CV.

Statistical analysis
After verification of data normality through skewness and kurtosis, a descriptive analysis (mean ± standard deviation) was performed. In order to compare CV and LMV, a paired Student's t-test was applied. In addition, effect size was tested using Cohen's d test. Linear regression was adjusted by gender and performed to verify the degree of association between the protocols (CV and LMV), as well as the variance analysis to confirm the hypothesis of regression. Bland and Altman's technique (Bland & Altman, 1999) was used to attest the level of agreement between the different tests. Lastly, the standard error of estimate (SEE) between the protocols was calculated. The level of significance adopted was 5% (p< .05) and all analyses were performed using the Statistical Package for the Social Sciences (SPSS) 18.0.

Results
The results from the present study show no significant difference (p= .305) between the LMV and CV. In addition, both protocols presented a good agreement based on the small difference between means and the narrow levels of agreement [0.2 (1.9) km·h -1 ]  ( Figure 3). Furthermore, the difference between means presented a small effect size (d= .123) and a SEE below 2.0% (Table 2). Linear regression, adjusted by gender, between the CV and LMV presented a significant association (r 2 = 0.397, p= .004), besides of an F value of 7.245, significantly for p= .004 (Figure 4).

Discussion
The main finding of the present study was that CV seemed to estimate LMV in youth runners, presenting an association even when adjusted by gender, a good level of agreement, low effect size and a standard error of estimate below 2% between values. This indicates that CV can be an easily applied and is low cost alternative to evaluate male and female youth runners.
The findings of the present study agree with the ones in Hiyane's et al. (2006) in trained cyclists. Likewise, Altimari et al. (2007) found a positive association between the studied methods in adolescent swimmers. However, other studies have shown that CV overestimated LMV in trained canoeists (Nakamura et al., 2006) and adult runners . Simões et al. (2005) showed that CV overestimated LMV in adult runners. However, they reported a significant positive association between the methods and concluded that CV is a valid method to predict and evaluate performance. One reason that could explain this overestimation is the use of inappropriate test distances, which are crucial to determine CV (Franken et al., 2011).
The mean durations of the tests performed in the present study were 6.4min (1,600m) and 13.1min (3,000m), agreeing with the recommendations (Greco, 2000;Kranenburg & Smith, 1996). However, in order to efficiently perform the tests, it is important to know how trained the volunteers are and/or perform pilot studies, since short duration tests (less than 3min) can overestimate critical power (CP) or CV (Bishop, Jenkins, & Howard, 1998). In addition, in long duration tests (more than 20min) other factors contribute to exhaustion, such as thermoregulation, substrate depletion and motivation (Greco, 2000). Finally, the present study followed Housh's et al. (1990) recommendations, in which the distances chosen should respect a 5min interval.
In the present study only two coordinates to calculate CV were used, agreeing with previous studies (de Lucas et al., 2012;Pacheco et al., 2006;Penteado et al., 2014;Silva et al., 2005;Simões et al., 2005). Several studies have shown that two coordinates are enough to determine CV in adult trained runners (de Lucas et al., 2012;Simões et al., 2005), moderately trained adults (Smith, Kendall, Fukuda, Cramer, & Stout, 2011), physically active adults (Pacheco et al., 2006;Silva et al., 2005), adult cyclists (Hiyane et al., 2006) and youth swimmers (Altimari et al., 2007). Housh et al. (1990), for instance, submitted 12 young adults to four loads until exhaustion and CP was determined with two, three and four coordinates. They reported that CP determined by two coordinates was strongly associated with the values assessed by four coordinates (r= .80 -.99).
The identification of CV, as performed in the present study, allows the evaluation of aerobic capacity and identify the intensity for exercise prescription (Leclair et al., 2008). Furthermore, through CV it is possible to perform an evaluation in the competition venue or during training sessions (Kranenburg & Smith, 1996) without the need of expensive lab equipments. Regarding intensity, Toubekis and Tokmakidis (2013) suggest that running at CV displays characteristics a "very heavy" intensity, where VO 2 is close to maximum (VO 2 max) and [lac]    long distance runners with a CV corresponding to 86% of the VO 2 max velocity (vVO 2 max) were capable of exercising at an intensity of 90% of vVO 2 max without achieving VO 2 max and without cardiovascular drift, which is frequently observed in high intensity aerobic exercise. Similar findings were reported by Bull, Housh, Johnson, & Rana (2008). In this scenario, de Lucas et al. (2013) reported that CP is the physiological index that estimates the limits between "heavy" and "severe" exercises in trained subjects. These authors evaluated the physiological responses and time to exhaustion in acute sessions until exhaustion at CP and 5% above CP, and showed that VO 2 , VE, and [lac] obtained at the end of the 5% above CP exhaustion trial were not significantly different from the maximal variables. The physiological end values during the CP test were significantly lower than when compared to the incremental test, and time to exhaustion at CP was significantly higher than 5% above.
Curiously, it seems that exercise prescription using intermittent running protocols produce a better physiological balance and higher training volume when compared to continuous running protocols. Penteado et al. (2014) compared level of tolerance and physiological responses in running tests at CV until exhaustion between intermittent and continuous protocols and found that heart rate, perceived exertion and [lac] at the end of both exhaustion tests were not significantly different when compared with incremental treadmill test values. However, time to exhaustion was twice longer in the intermittent test when compared to continuous, and only the continuous session showed an increase 9.0±0.8 mmol·l −1 of [lac] at the end of exercise.
It is worth highlighting that the present study has some limitations. One of them was not comparing CV with the intensity of MLSS, which is considered the gold standard in aerobic capacity evaluation from [lac] responses (Beneke, 2003). However, several studies have demonstrated that there is no difference between MLSS and LMV (MacIntosh, Esau, & Svedahl, 2002;Pardono et al., 2008Pardono et al., , 2009Puga et al., 2012;Sotero et al., 2007Sotero et al., , 2009. Another limitation was not assessing the participant's maturational state. In this matter, Frainer, Oliveira and Pazin (2006) verified no associations between sexual maturation, age and growth indexes with performance of running.

Conclusion
The combination of the predictive sets performed (1,600 and 3,000m), in order to obtain the CV proposed by the present study, presented values that did not differ from the ones obtained through LMV. Therefore, the test proposed in the present study is valid and CV did in fact estimate LMV in youth runners. This finding is important since CV is a low cost and non-invasive method of evaluating aerobic capacity and to identify the intensity for exercise prescription.