One of the most valuable exercises athletes use for developing an aerobic base is maximal steady state (MSS) exercise, generally determined by maximal lactate steady state (MLSS) by way of blood lactate concentration (HLa) (8,19,28). Maximal lactate steady state is the highest point of equilibrium between HLa production and removal (9). Increasing workload above MLSS results in a rate of HLa production that exceeds its disappearance, and therefore HLa levels continue to increase, and exercise generally lasts less than 30 minutes before fatigue occurs (9,18,31). Although MLSS exercise intensity has been shown to be unrelated to high-intensity exercise performance (peak workload of incremental test) (5), it has been shown to be related to endurance exercise performance (8,26,28) and has been used as a tool to monitor training intensity (29,30). Performing MSS exercises twice per week has been shown to improve lactate threshold, run time at MSS, and maximal oxygen uptake (10). Maximal steady state intensity exercise has been termed heavy-intensity endurance training (5) and generally lasts from 30 to 60 minutes.
Although the determination of MSS by way of HLa during an incremental test can be performed on athletes, athletes are often hesitant to have repeated venipunctures of fingers or earlobes for blood collection. Other techniques have been used in an attempt to predict MSS, such as heart rate, minimal lactate, ventilation, time trials, and pace, with mixed results (14,19,22,24,27,31,34). Thus, a noninvasive laboratory method of determining MSS exercise intensity would be beneficial.
Near-infrared spectroscopy (NIRS) has recently been used to determine total muscle oxygen saturation at rest and during exercise. This device measures hemoglobin oxygenation in tissue using spectrophotometric principles. The absorption changes of oxygenated and deoxygenated hemoglobin allow for the calculation of percent oxygen saturation (StO2) in the tissue. Although NIRS measures oxygenation and deoxygenation of hemoglobin, it may also include in its value oxygen bound to myoglobin, thus reflecting total oxygen tissue saturation. Even though NIRS does not distinguish among oxygen bound to hemoglobin, deoxyhemoglobin, and myoglobin, it does measure total oxygen saturation in the tissue. A functional percent saturation can be obtained when oxygenated and deoxygenated hemoglobins are included in the tissue saturation value. More information pertaining to the theoretical basis of NIRS can be obtained in Myers et al. (25).
Near-infrared spectroscopy has been used to examine muscle oxygenation during cycling (1,2,3,7,13,16), rowing (13), arm cranking (7), and speed skating (15). In most of these studies, the quadriceps muscle group (specifically the vastus lateralis muscle) was examined. A relationship (r > 0.90) has been reported between the breakpoint workload of HLa and the breakpoint workload of muscle StO2 during incremental cycling exercise (1,16,33). However, only 1 published study with runners reported a similar breakpoint determined with NIRS (1). In the study by Austin et al. (1), muscle oxygenation was determined at lactate threshold (r = 0.87); however, StO2 breakpoint was observed in only 11 of 23 runners. No use of this breakpoint was made to establish MSS for the runners. Therefore, the purpose of this study was to examine the use of NIRS to detect a breakpoint in StO2 of the muscle during incremental running exercise and whether the exercise intensity at that breakpoint could then be used to determine MSS exercise intensity. We hypothesized that the StO2 breakpoint would occur in running as it has been shown to in cycling and that the StO2 breakpoint could be used to determine MSS exercise intensity.
Experimental Approach to the Problem
Initially, subjects performed an incremental running test so that lactate threshold could be determined. Muscle oxygenation was also obtained during this test for the determination of an StO2 breakpoint. The subjects then performed a number of 30 minute runs until the subjects had performed the highest-intensity steady state run and a run just above this intensity, that is, a run above steady state as indicated by lactate values, which increased throughout the run. The running speed during the highest-intensity steady state exercise was then compared with that at lactate threshold and StO2 breakpoint to determine whether any differences in the speeds occurred or if the speeds were similar.
Well-trained distance runners and triathletes (men = 9, age = 32 ± 6 yr, O2max = 64.9 ± 4.9 ml·kg−1·min−1; women = 7, age = 31 ± 9 yr, O2max = 50.8 ± 7.0 ml·kg−1·min−1) volunteered for this study. The experimental procedures were explained to all subjects, and informed consent was obtained as outlined by the Institutional Review Board of the University of Wisconsin-Milwaukee. Data collection occurred during the spring precompetition training period. All subjects had been training at least 5 years for endurance events and had been training at least 2 months since their last recovery microcycle when the study was conducted. All data were collected in an exercise physiology laboratory.
Incremental treadmill running was conducted according to the modified protocol of Snyder et al. (31). Exercise stages of 6 minutes in duration were performed (21) with respiratory gas exchange measures (RAMS M-100 Mass Spectrometry, Marquette Electronics, Milwaukee, WI), heart rate (Polar Heart Watch, Kempele, Finland), and muscle oxygenation from the gastrocnemius muscle (InSpectra, Hutchinson Technology Incorporated, Hutchinson, MN) obtained during the last minute of each workload. Before the subjects warmed up on the treadmill, the InSpectra probe was attached over the midpoint of the belly of the gastrocnemius muscle with a shield with an adhesive back. An athletic wrap was loosely wrapped around the lower leg and then taped to ensure placement of the probe (Figure 1). No probes came loose during the testing. Muscle StO2 was obtained every 3 seconds based on a ratio of the second derivative of the changes in oxyhemoglobin and deoxyhemoglobin concentrations measured spectrophotometrically at 720 and 760 nm (25). A 25-mm probe that detects 95% of its signal from a depth of 0 to 23 mm was used (1, 25). Finally, a blood sample from a fingertip (50 μL) was obtained at the end of each exercise stage and placed in a lysing agent for blood lactate analysis (YSI Sport 1500, Yellow Springs, OH).
Mancini et al. (23) have previously determined that NIRS is a valid technique for determining muscle oxygenation, whereas Austin et al. (1) reported that measurement of muscle oxygenation of the gastrocnemius muscle by the InSpectra during running was highly reliable at lactate threshold (r = 0.87) and maximal effort (r = 0.88). The InSpectra probe was calibrated to an StO2 measurement equivalent to 49 ± 2% before each use. All other equipment (RAMS M-100 Mass Spectrometer, YSI Sport 1500, etc.) were calibrated as per the manufactures' specifications.
Throughout the incremental exercise test, the motorized treadmill was set at a 1% elevation to compensate for the absence of wind resistance (18). Each exercise test was individualized such that the fourth stage of the test was run at a speed that the subjects thought they could maintain for an hour. The other stages were increments of 0.22 m·s−1 slower or faster. The first 3 stages were 0.66, 0.44 and 0.22 m·s−1 slower than stage 4, and all stages after stage 4 were 0.22 m·s−1 faster than the previous one. Beneke et al. (4-6) have used this increment in speed for women and a slightly greater increment in speed for men (0.24-0.26 m·s−1). In our laboratory, we have found that the lower speed increments work equally well for men and women, and thus the speed increments were the same for all subjects (31). This exercise protocol has been used successfully in our laboratory for years. The exercise test continued through successive stages until the investigators and subjects were certain that the subject was past lactate threshold. Because athletes typically overestimate their abilities, the lactate threshold generally occurred in the fourth stage, and the test was completed at the end of the fifth stage as designed and previously used (31). After a recovery period of approximately 10 minutes, the subject performed a maximal exercise to exhaustion to determine aerobic capacity (i.e., O2 max). During this test, the treadmill stayed at a constant speed (stage 4 of incremental test), with a percent grade increase of 2% at minutes 2 and 4 followed by percent grade increases of 1% each minute thereafter until maximal effort occurred. Test duration for all subjects was 6 to 9 minutes.
The data from the incremental exercise tests, especially HLa and StO2, were graphed vs. speed for the determination of breakpoints. Blood lactate breakpoint workload was defined as an increase in HLa of greater than 1.0 mM for the next workload. Muscle oxygenation breakpoint workload (or velocity) was defined as the workload before a decrease in StO2 of greater than 15% that lead to a continuous decrease. Running speed at 4 mM was also determined by interpolating the running speed where the lactate curve crossed the 4 mM value.
After the performance of the incremental exercise test and the determination of the exercise intensity at which a breakpoint in HLa concentration occurred, the athletes performed 2 to 5 30-minute constant load exercise bouts to determine MLSS exercise. All of the exercise bouts were separated from each other by at least 48 hours. The day before the constant load exercise, the subjects were to perform only light exercise and consume their usual food intake. Exercise activities and food intake before testing were not monitored, but the subjects were reminded to conform and asked to verify that they complied. MLSS workload was defined as the greatest exercise intensity in which HLa changed less than 1.0 mM between minutes 10 and 30 of the 30 minute constant load exercise and the lactate value at minute 30 was greater than 4.0 mM (4,6,19). The first constant load exercise was performed at the exercise intensity of the lactate breakpoint, and exercise intensity was increased or decreased 0.22 m·s−1 on subsequent trials as necessary (Figure 2). During the constant load exercises, muscle oxygen saturation, oxygen uptake, and heart rate were obtained continuously, whereas blood for lactate determination and rating of perceived exertion were obtained at minutes 10 and 30.
Breakpoint speed for both lactate and muscle oxygen saturation were determined from the graphed incremental test data of each subject by 2 individuals who were blinded to the subject characteristics of each test (Figure 3). A change in linearity when plotting StO2 or HLa vs. running speed was defined as the breakpoint. Because running speed at 4.0 mM blood lactate (9,21) has also been used to determine MSS, this measurement was obtained for comparison purposes. If a discrepancy occurred between the 2 reviewers (5 of 64 tests), they discussed their reasoning with each other and determined a compromise; either one of the reviewers' values or the mean of the two values was used as the breakpoint. Speed at MSS, lactate breakpoint, 4.0 mM lactate, and muscle oxygen saturation breakpoint were determined and compared with a repeated measures one-way analysis of variance. In addition, a Bland and Altman (12) analysis was performed to examine the agreement between the criterion running speed (MLSS) and the speeds predicted by muscle oxygenation (StO2) and blood lactate (HLa). A p < 0.05 level was used for all comparisons.
Of the 16 subjects, all variables could be determined in 12. Of the remaining 4 subjects, 1 subject did not reach a MSS, 2 subjects displayed either an HLa or an StO2 breakpoint but not both, and 1 subject did not demonstrate a breakpoint for both HLa and StO2. Thus, the results of 2 subjects could not be used because of their HLa results and 2 subjects because of their StO2 results. Therefore, the remaining analyses reflect data from the 12 subjects from whom complete data were obtained.
Mean running speed determined from the incremental test for HLa breakpoint (12.76 ± 1.63 km·h−1), StO2 breakpoint (12.84 ± 1.58 km·h−1), 4 mM lactate (13.49 ± 1.71 km·h−1), and MSS (13.04 ± 2.03 km·h−1) were similar (Table 1), especially MSS, StO2 breakpoint, and HLa breakpoint, although none of the running speeds were significantly different from each other (Figure 4). Both StO2 breakpoint and HLa breakpoint were highly related to MSS (r = 0.92 for StO2, r = 0.95 for HLa). Likewise, individual subjects varied little in the running speed determined by MSS, StO2 breakpoint, HLa breakpoint, and 4 mM HLa (Figure 5). Finally, the Bland and Altman analysis of agreement between the MLSS and the StO2 breakpoint speeds resulted in a mean difference of 0.14 ± 0.36, whereas the mean difference between MLSS and HLa breakpoint speeds was 0.19 ± 0.43.
Individual differences from the MSS speed were very similar. The breakpoint for StO2 speed was 0.45 ± 0.40 km·h−1 different from the mean MSS speed, 0.63 ± 0.39 km·h−1 different from HLa breakpoint speed, and 0.53 ± 0.40 km·h−1 different from 4 mM HLa speed (Table 2). The distribution of absolute differences from MSS for the 3 predictors (StO2 breakpoint, HLa breakpoint, and 4 mM HLa speed) was also similar (Figure 6).
The purposes of this study were to determine whether NIRS could detect a breakpoint in StO2 of the muscle during incremental running exercise and whether the exercise intensity at that breakpoint could be used to predict MSS exercise intensity as well as HLa does. Twelve endurance trained subjects performed both incremental and steady state runs with complete comparisons. Exercise speed predicted by StO2 and HLa were similar. Thus, muscle StO2 predicted MSS as well as HLa.
Grassi et al. (16), Austin et al. (1), and Soller et al. (33) have previously used NIRS to determine HLa breakpoint. Grassi et al. (16) used mountain climbers who were cycling and were able to obtain similar breakpoints in 4 of the 5 subjects. In 1 subject, StO2 breakpoint occurred before the HLa breakpoint. HLa and StO2 breakpoints were also significantly correlated at r = 0.95. Austin et al. (1) examined both runners and cyclists performing their specialized activity. StO2 breakpoints were determined in 11 of the 23 (48%) runners and 18 of the 21 (86%) cyclists. Breakpoints for StO2 and HLa were significantly related for the runners (r = 0.99) and the cyclists (r = 0.99). Thus, both Grassi et al. (16) and Austin et al. (1) observed significant correlations between StO2 and HLa breakpoints and reported the identity of the StO2 breakpoint while cycling. Soller et al. (33) used healthy men and women on a cycling ergometer with lactate and muscle StO2 data available for 7 of the 10 subjects. In the present study, the StO2 and HLa breakpoints were similar to those observed by Grassi et al. (16) and Austin et al. (1). However, in the present study, a much higher percentage of StO2 breakpoints was determined than by Austin et al. (1). The methods used by Austin et al. (1) during the incremental running test could contribute to this difference. In our study, we used 6 minute exercise stages during the incremental exercise. Kuipers et al. (21) have previously demonstrated that 6 minute bouts were necessary to obtain accurate HLa during a running test. Austin et al. (1) used 3 minute exercise stages during their incremental exercises, which could have resulted in a more steady decline in StO2 and a more steady increase in HLa and not necessarily a steady state level at each exercise intensity. Therefore, a breakpoint would be much more difficult to determine. Given that 12.5% of the subjects in the present study had no StO2 breakpoint, 7 minute exercise stages could possibly be used to enhance breakpoint determination, although others have found exercise StO2 plateaus in much shorter time periods (16). Possibly, the use of smaller speed increments would also increase the ability to determine StO2 breakpoint with a running test.
Test interruptions for the collection of data, specifically HLa, are another possible technique problem that could influence the results of this study. Beneke et al. (6) have observed that when test interruptions occur, MLSS is achieved at a higher workload and relative intensity. Test interruptions may also affect muscle oxygenation and alter the data such that the breakpoints may not be noticeable. In the present study, we did not include test interruptions in either the incremental or steady state exercise bouts; rather, we collected the blood sample while the subject continued to run. Austin et al. (1) did not report whether they interrupted the exercise tests for blood sampling. With cycling, shorter exercise stages appear more feasible given that both Austin et al. (1), who used 3 minute stages, and Grassi et al. (16), who used 4 minute stages, reported at least an 80% prediction rate. Other factors that could influence the determination of StO2 breakpoint might be adipose tissue thickness, blood hemoglobin concentrations, skin temperature, capillary density of the muscle fibers, and muscle fiber composition (1,16). Although we did not measure any of these variables, most of the variables would appear to be similar across the subjects in this study because of the endurance trained status of the runners and triathletes. The depth of penetration of the light might also affect StO2 breakpoint determination. Future research is necessary to determine whether these variables affect muscle oxygen saturation of homogeneous and nonhomogeneous populations.
Maximal steady state speed and HLa breakpoint have been associated with the fastest training pace for aerobic activities and the pace that can lead to endurance improvements (9,10,19,28). Increasing exercise intensity above MSS results in elevated blood lactate levels and exercise generally lasts less than 30 minutes before fatigue occurs (8,9,19,31). The criteria for determining MSS is generally understood to be 30 minutes of exercise with less than a 1 mM increase in HLa from minutes 10 to 30. Exactly how long one can perform at MSS has not been determined, although a few studies have shown that subjects can perform longer than 30 minutes (11,17,20). Examination of the muscle StO2 and HLa data during the 30 minute exercises to determine MSS in the present study show that, at the MSS workload, HLa was increasing minimally, whereas at the workload above MSS, HLa increased greatly. Conversely, StO2 was relatively constant for both the MSS workload and the workload above MSS (Figures 7 and 8). Thus, one would expect that, without an increase in HLa and with oxygen readily available in the tissue, as occurs at MSS, continued exercise could be performed. Further studies are necessary to determine the length of time athletes can perform an activity at MSS and what the most important causative agents of fatigue at this exercise intensity are. Similarly, Soller et al. (33) recently differentiated the contribution of hydrogen ion concentration and muscle oxygen saturation on threshold determination during incremental exercise. Hydrogen ion concentration was strongly correlated with lactate threshold. Further work to differentiate the effects of hydrogen ions and oxygen is needed.
Although StO2 was able to predict MSS speed in this study as well as HLa, the methods used are not without drawbacks. Specifically, the NIRS unit used was not portable and is relatively expensive. More portable and less expensive units are available, and technology should soon be available to allow for field tests to be performed (32). Also, in this study, individual plots of StO2 were examined, and an objective assessment was made by 2 reviewers. A computer program to assess the changes in linearity may or may not be helpful in determining StO2 breakpoint.
In conclusion, NIRS was used to predict MSS running speed in 12 distance trained runners and triathletes. Percent muscle oxygen saturation breakpoint, as determined by NIRS during an incremental exercise test, predicted MSS running speed as well as the traditional HLa methods currently used in this group of aerobically trained athletes. The ability of StO2 to predict MSS in anaerobically trained athletes and in heterogeneous groups of subjects is yet to be determined.
From a practical perspective, obtaining a blood sample from an athlete to determine lactic acid levels is undesirable for the athlete and pathologically dangerous for the coach or sports medicine professional. Use of muscle oxygenation to determine MSS exercise intensity in this study was as effective as blood lactate and much more pleasurable for the athletes. Thus, repeated uses of this exercise test and protocol during a training season would not be seen negatively by the athletes and could, in the long run, result in more accurate testing results.
A portion of this work was previously presented at an American College of Sports Medicine Meeting. This study was supported in part by a grant from Hutchinson Technology, Inc., Hutchinson, MN. No professional relations existed between the authors and Hutchinson Technology, Inc. The results of the present study do not constitute endorsement of the product by the authors or the NSCA.
1. Austin, KG, Daigle, KA, Patterson, P, Cowman, J, Chelland, S, and Haymes, EM. Reliability of near-infrared spectroscopy for determining muscle oxygen saturation during exercise. Res Q Exerc Sport
76: 440-449, 2005.
2. Belardinelli, R, Barstow, TJ, Porszasz, J, and Wasserman, K. Changes in skeletal muscle oxygenation during incremental exercise measured with near infrared spectroscopy. Eur J Appl Physiol
70: 487-492, 1995.
3. Belardinelli, R, Barstow, TJ, Porszasz, J, and Wasserman, K. Skeletal muscle oxygenation during constant work rate exercise. Med Sci Sports Exerc
27: 512-519, 1995.
4. Beneke, R. Methodological aspects of maximal lactate steady-state: implications for performance testing. Eur J Appl Physiol
89: 95-99, 2003.
5. Beneke, R, Hutler, M, and Leithauser, RM. Maximal lactate-steady-state independent of performance. Med Sci Sports Exerc
32: 1135-1139, 2000.
6. Beneke, R, Hutler, M, vonDuvillard, SP, Sellens, M, and Leithauser, RM. Effect of test interruptions on blood lactate during constant workload testing. Med Sci Sports Exerc
35: 1626-1630, 2003.
7. Bhambhani, Y, Maikala, R, and Buckley, S. Muscle oxygenation during incremental arm and leg exercise in men and women. Eur J Appl Physiol
78: 422-431, 1998.
8. Billat, V, Lepretre, PM, Heugas, AM, Laurence, MH, Salim, D, and Koralsztein, JP. Training and bioenergetic characteristics in elite male and female Kenyan runners. Med Sci Sports Exerc
35: 297-304, 2003.
9. Billat, VL, Sirvent, P, Py, G, Koralsztein, JP, and Mercier, J. The concept of maximal lactate steady state. A bridge between biochemistry, physiology and sport science. Sports Med
33: 407-426, 2003.
10. Billat, V, Sirvent, P, Lepretre, P-M, and Koralsztein, JP. Training effect on performance, substance balance and blood lactate concentration at maximal lactate steady state in master endurance-runners. Pflug Arch
447: 875-883, 2004.
11. Bishop, D, Jenkins, DG, McEniery, M, and Carey, MF. Relationship between plasma lactate parameters and muscle characteristics in female cyclists. Med Sci Sports Exerc
32: 1088-1093, 2000.
12. Bland, JM and Altman, DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet
13. Chance, B, Dait, MT, Zhang, C, Hamaoka, T, and Hagerman, F. Recovery from exercise-induced desaturation in the quadriceps muscles of elite competitive rowers. Am J Physiol
262: C766-C775, 1992.
14. Dekerle, J, Pelayo, P, Clipet, B, Depretz, S, Lefevre, T, and Sidney, M. Critical swimming speed does not represent the speed at maximal lactate steady state. Int J Sports Med
26: 524-530, 2005.
15. Foster, C, Rundell, KW, Snyder, AC, Stray-Gundersen, J, Kemkers, G, Thometz, N, Broker, J, and Knapp, E. Evidence for restricted muscle blood flow during speed skating. Med Sci Sports Exerc
31: 1433-1440, 1999.
16. Grassi, B, Quaresima, V, Marconi, C, Ferrari, M, and Cerretelli, P. Blood lactate accumulation and muscle deoxygenation during incremental exercise. J Appl Physiol
87: 348-355, 1999.
17. Harnish, CR, Swensen, TC, and Pate, RR. Methods for estimating the maximal lactate steady state in trained cyclists. Med Sci Sports Exerc
33: 1052-1055, 2001.
18. Jones, AM and Doust, JH. A 1% treadmill grade most accurately reflects the energy cost of outdoor running
. J Sports Sci
14: 321-327, 1996.
19. Jones, AM and Doust, JH. The validity of the lactate minimum test for determination of the maximal lactate steady state. Med Sci Sports Exerc
30: 1304-1313, 1998.
20. Kenefick, RW, Mattern, CO, Mahood, NV, and Quinn, TJ. Physiological variables at lactate threshold under-represent cycling time-trial intensity. J Sports Med Phys Fitness
42: 396-402, 2002.
21. Kuipers, H, Rietjens, G, Verstappen, F, Schoenmakers, H, and Hoffman, G. Effects of stage duration in incremental running
tests on physiological variables. Int J Sports Med
24: 1-6, 2004.
22. MacIntosh, BR, Esau, S, and Svedahl, K. The lactate minimum test for cycling: Estimation of the maximal lactate steady state. Can J Appl Physiol
27: 232-249, 2002.
23. Mancini, DM, Bolinger, L, Li, H, Kendrick, K, Chance, B, and Wilson, JR. Validation of near-infrared spectroscopy in humans. J Appl Physiol
77: 2740-2747, 1994.
24. McGehee, JC, Tanner, CJ, and Houmard, JA. A comparison of methods for estimating the lactate threshold. J Strength Cond Res
19: 553-558, 2005.
25. Myers, DE, Anderson, LD, Seifert, RP, Ortner, JP, Cooper, CE, Beilman, GJ, and Mowlem, JD. Noninvasive method for measuring local hemoglobin oxygen saturation in tissue using wide gap second derivative near-infrared spectroscopy. J Biomed Optics
10: 034017-1-034017-18, 2005.
26. Nicholson, RM and Sleivert, GG. Indices of lactate threshold and their relationship with 10-km running
velocity. Med Sci Sports Exerc
33: 339-342, 2001.
27. Perrey, S, Grappe, F, Girard, A, Bringard, A, Groslambert, A, Bertucci, W, and Rouillon, JD. Physiological and metabolic responses of triathletes to a simulated 30-min time-trial in cycling at self-selected intensity. Int J Sports Med
24: 138-143, 2003.
28. Pringle, JS and Jones, AM. Maximal lactate steady state, critical power and EMG during cycling. Eur J Appl Physiol
88: 214-216, 2002.
29. Smith, CG and Jones, AM. The relationship between critical velocity, maximal lactate steady-state velocity and lactate turnpoint velocity in runners. Eur J Appl Physiol
85: 19-26, 2001.
30. Snyder, AC and Foster, C. Skating. In: Nutrition in Sports
of the IOC Encyclopaedia of Sports Medicine Series
. Maughan, R, ed. Oxford, UK: Blackwell Science, Inc., 2000. pp. 646-655.
31. Snyder, AC, Woulfe, T, Welsh, R, and Foster, C. A simplified approach to estimating the maximal lactate steady state. Int J Sports Med
15: 27-31, 1994.
32. Soller, BR, Yang, Y, Lee, SM, Wilson, C, and Hagan, RD. Noninvasive determination of exercise-induced hydrogen ion threshold through direct optical measurement. J Appl Physiol
104: 837-844, 2008.
33. Soller, BR, Yang, Y, Soyemi, OO, Ryan, KL, Rickards, CA, Walz, JM, Heard, SO, and Convertino, VA. Noninvasively determined muscle oxygen saturation is an early indicator of central hypovolemia in humans. J Appl Physiol
104: 475-481, 2008.
34. Vobejda, C, Fromme, K, Samson, W, and Zimmerman, E. Maximal constant heart rate: a heart rate based method to estimate maximal lactate steady state in running
. Int J Sports Med
7: 368-372, 2005.