The characteristic low “sitting” position of competitive speed skating has been shown to result in a right shifted heart rate-˙VO2 curve compared with running or cycling(26,27). The impact of modifying hip and knee angles during treadmill in-line speed skating on heart rate, oxygen uptake, and blood lactate concentration has been described by Rundell(26). In that study (26), skating in the low posture resulted in lower peak ˙VO2, higher heart rate, lower submaximal ˙VO2, and higher submaximal blood lactate concentrations than skating in a more upright position. Peak oxygen uptake was approximately 8% lower and submaximal blood lactate concentration at a fixed O2 uptake (50 ml·kg-1·min-1) was approximately two-fold higher during low skating (26). This study extended previous work comparing speed skating(4,19,30,33) and low walking(27) with cycle ergometry. The compromised oxygen uptake and higher blood lactate values are thought to be a consequence of reduced blood flow to the exercising muscle in the skating position. The low posture and the long duty cycle of the muscle during skating forces a static contraction of the quadriceps (4,19). As originally hypothesized by Snyder and Foster (29) and supported by Rundell (26,27), the low skating position may compromise blood flow to the working muscles, limit oxygen delivery, and increase the dependence on anaerobic energy production during speed skating. However, at present no data directly address tissue oxygenation during skating.
The ability to measure tissue oxygenation using near infrared spectroscopy(NIRS) was originally developed by Millikan (22). Duel wavelength spectrophotometry in the near infrared region provides an opportunity to measure hemoglobin/myoglobin (Hb/Mb) deoxygenation in the capillary bed of an exercising muscle (2,16). Hb/Mb deoxygenation can be measured at 760 nm, while HbO2 and MbO2 show peak absorbance at 850 nm. The 760 nm signal provides a relative estimate of deoxygenation trends (2,3) and the sum of the signals at 760 and 850 nm provides a reference to blood volume changes. Two recent studies suggest strong relationships between tissue deoxygenation, oxygen uptake, and lactate threshold during cycle ergometry(1) and rowing (2). Belardinelli et al. (1) found a progressive decrease in muscle oxygenation that plateaued at approximately 80% ˙VO2max using NIRS during incremental cycle ergometry. Chance et al. (2) evaluated recovery time for Hb/Mb desaturation in the muscle capillary bed of exercising competitive rowers. This study demonstrated a near perfect relationship between Hb/Mb re-oxygenation recovery time and blood lactate at 70, 85, and 100% maximal voluntary exercise.
The physiological differences between speed skating in the sitting position over a more upright position (26) provide a unique opportunity to evaluate muscle tissue deoxygenation. If the restricted blood flow hypothesis (29) is correct, one would expect increased Hb/Mb deoxygenation concomitant with anaerobic energy production and inconsistent with whole body oxygen uptake. Support for this assumption can be found in the work of Stringer et al. (31) who found that acidification of muscle capillary blood as a result of lactic acidosis accounted for virtually all of the O2Hb dissociation above lactate threshold. The purpose of this study was to evaluate the effect of speed skating position on Hb/Mb oxygen dissociation in various muscle groups during treadmill in-line skating. The results may provide insight to performance and training of speed skaters and may also contribute to the body of clinical knowledge concerning blood flow and peripheral arterial insufficiency.
Subjects. Eight top U.S. male short track speed skaters gave their informed written consent to participate in this study. All subjects were residents at the U.S. Olympic Training Center at Lake Placid and successfully compete at the National and/or International level. Subjects were tested during July 1995. Subject characteristics (mean ± SD) were 18.6± 3.66 yr of age, 69.9 ± 7.86 kg body weight, and 6.9 ± 1.91% body fat (estimated using the sum of 7 skinfolds)(18).
Testing procedures. Each subject performed four randomized trials of treadmill in-line skating in the upright (US) and low (LS) positions at 2.7 and 3.1 m·s-1 (4% grade). Each trial was 5 min in duration with 20 min recovery between trials. The treadmill used in this study had a skating surface of 2.44 × 3.05 m. Respiratory gases were continuously monitored during each test using open-circuit spirometry (Sensor Medics 2900, Yorba Linda, CA). Calibration was performed using standard gases (26% O2 balance N2, and 16% O2 4% CO2 balance N2). Heart rate (HR) was continuously monitored using Polar Vantage XL (Port Washington, NY) heart rate monitors, and steady state HR was recorded during the last min of each trial. At the completion of each 5-min trial, an arterialized blood sample was taken via fingerstick for blood lactate (LA) determination (YSI 2300 Stat Plus). Calibration was verified using 2.5 and 5 mM standard lactate samples, and recalibration was performed after every four samples.
Near Infrared Spectroscopy. Duel wavelength near infrared spectroscopy (NIRS; RUN-MAN, NIM, Philadelphia, PA) was used to measure hemoglobin/myoglobin (Hb/Mb) deoxygenation and blood volume in the capillary bed of five muscle groups (vastus lateralis, vastus medialis, rectus femoris, biceps femoris, and gluteus maximus) during exercise. Linearity of NIRS has been validated (1-3). The signal from five NIRS probes placed on each muscle was verified prior to exercise to ensure proper placement. The probe consists of two tungsten lamps that emit white light and two photodetectors. The average tissue depth of light penetration is 2.5-3.0 cm (2). Wavelengths of 760 and 850 nm were monitored to detect relative change in oxygenation of Hb and Mb. Oxygen desaturation of muscles was estimated by changes in the difference in signal strength at 760 and 850 nm (760-850 nm). The sum of the wavelengths (760 + 850 nm) estimated total Hb and Mb present, thus reflecting changes in Hb concentration. Total Hb/Mb concentration was used to estimate changes in blood volume as previously done (1-3).Figure 1 represents typical desaturation curves during US and LS. Since NIRS does not indicate absolute values of Hb/Mb oxygenation, values are presented as change in optical density (ΔOD) relative to maximal hyperemia recorded within the first min of exercise recovery as previously done (1) and as percent of maximal deoxygenation determined during cuff ischemia.
Joint angle measurements. Knee and trunk angles were determined for each skating position (US and LS) by video analysis as previously done(26). The skaters were filmed during treadmill skating tests at an angle perpendicular to the skating direction using a tripod mounted normal speed Panasonic AG450 video camera (Matsushita Electric Industrial Co., Okayama, Japan) operating at 60 Hz with a shutter speed of 1/205. A Panasonic 7500 VCR connected to a Panasonic AG-A750 video editor interfaced with the PEAK5 Motion Analysis system (PEAK Performance Technologies Inc., Englewood, CO) was used for joint angle determination. Preextension angles were measured as previously done(4,26,37). Knee angle in the skating position was considered the angle between the upper and lower skating leg. Trunk angle was measured using the angle presented by the line between the hip joint and the middle of the neck with respect to the line between the hip joint and the knee joint.
Cuff ischemia. NIRS measurements were obtained during cuff ischemia to obtain an estimate of maximal desaturation for the four leg muscles examined during skating. An ischemic value was not obtained for the gluteus maximus. NIRS probes were placed on each muscle to record muscle tissue deoxygenation while a cuff pressure of 300 torr was maintained on the upper left thigh for 8 min. The athlete remained seated for the duration of the procedure. In each case, the deoxygenation signal plateaued at a minimum level prior to release of cuff pressure.
Statistical analysis. Statistical comparisons of physiological responses between LS and US at the two different speeds were made using repeated measures ANOVA. Tukey's post-hoc analysis was performed when significant F-ratios were found. Pearson product moment correlations were used to identify significant relationships. Multiple regression analysis was performed to identify significant relationships between physiological data and NIRS and joint angle measurements. AP-value of < 0.05 was considered significant for all comparisons.
Two-dimensional video analysis during treadmill skating allowed an estimate of differences in knee and trunk angles during US and LS. Significant differences were identified between US and LS for knee and trunk angles(Fig. 2, P < 0.05). Pre-extension knee angles were 118 ± 4.1 and 105 ± 3.7 for US and LS, respectively. Respective trunk angles for US and LS were 111 ± 5.7 and 58 ± 5.9.
Table 1 summarizes physiological variables measured at 2.7 and 3.1 m·s-1 (4% grade) during US and LS. Significant differences were noted between US and LS at each speed for all measured variables except ˙VO2 at 3.1 m·s-1 (P < 0.05). Significant differences were noted between speeds for US and LS for all variables except R (P < 0.05). Heart rate and ˙VE were not different between LS at 2.7 m·s-1 and US at 3.1 m·s-1, even though ˙VO2 was higher and blood lactate was lower for US at 3.1 m·s-1 (P < 0.05). Changes between US and LS at 2.7 and 3.1 m·s-1 in blood lactate and˙VO2 as a function of heart rate are presented inFigure 3.
Muscle tissue deoxygenation was significantly greater during LS than during US at both speeds for all muscles examined (Table 2,P < 0.05). Deoxygenation for all muscles during LS was significantly greater at the faster speed (P < 0.05), while significant differences in deoxygenation between speeds during US were noted only for the vastus lateralis and rectus femoris (P < 0.05). Multiple regression analyses using the sum of the five muscle tissue deoxygenation ODs as the dependent variable and knee and hip angles as independent variables yielded a significant multiple R (0.65, F = 10.5, P < 0.05; Fig. 4) with hip angle being the strongest predictor of deoxygenation. The sum of tissue deoxygenation was significantly correlated with heart rate (R = 0.64, P < 0.05;Fig. 5A). Separate multiple regression analyses using either blood lactate or oxygen uptake as the dependent variable and tissue deoxygenation for each of the five muscles as independent variables were performed. A significant multiple R for blood lactate (0.95, F = 44.2, P < 0.05) with the vastus lateralis and the rectus femoris having the highest beta values and the vastus medialis having the lowest beta. A multiple R of 0.48 (F = 1.6) for oxygen uptake was not significant. Figure 6 represents the relationships between the sum of delta ODs for the five muscles and blood lactate and oxygen uptake.
Muscle tissue blood volume change was significantly lower during LS than for US for all combinations of speed and position for all muscles examined except biceps femoris and gluteus maximus at 2.7 m·s-1 LS and 3.1 m·s-1 US (Table 3, P < 0.05). No difference in blood volume change was noted between US at 2.7 and 3.1 m·s-1. The rectus femoris, biceps femoris, and gluteus maximus demonstrated significantly less blood volume change for LS between 2.7 and 3.1 m·s-1 (P < 0.05). Multiple regression analyses using the sum of the five muscle tissue blood volume changes as the dependent variable and knee and hip angles as independent variables yielded a significant multiple R (0.70, F = 13.8, P < 0.05;Fig. 7,) with hip angle being the strongest predictor of blood volume change. The sum of tissue blood volume change was significantly correlated to heart rate (R = 0.63, P < 0.05;Fig. 5B). Separate multiple regression analyses using either blood lactate or oxygen uptake as dependent variables and tissue blood volume change for each of the five muscles as independent variables were performed. A significant multiple R for blood lactate was identified (0.82,F = 11.0, P < 0.05). The vastus lateralis and the gluteus maximus demonstrated the highest beta values, and the vastus medialis showed the lowest beta value. A multiple R of 0.63 (F = 3.5,P < 0.05) was found for oxygen uptake (the vastus lateralis demonstrated the highest beta). Figure 8 represents relationships between the sum of delta blood volumes for the five muscles and blood lactate and oxygen uptake. The Pearson product moment correlation between the sums of blood volume change and deoxygenation was significant (R =-0.66, P < 0.05).
Table 4 summarizes results from 8 min of resting ischemia. Deoxygenation resulting from ischemia was significantly different than skating deoxygenation measurements except for the vastus medialis during LS at both speeds and the rectus femoris during LS at 3.1 m·s-1(P < 0.05). After 6 min of 300 torr cuff induced ischemia, deoxygenation and blood volume values were stable.Figure 9 demonstrates differences in deoxygenation and blood volume during cuff ischemia. The vastus lateralis and rectus femoris demonstrated a decrease in blood volume, while the vastus medialis and biceps femoris demonstrated a hyperemic response to ischemia. Differences in skating deoxygenation represented as percent maximal deoxygenation during ischemia are presented in Figure 10.
Numerous studies have identified lower peak ˙VO2 values during speed skating than during cycle ergometry(4,19,33,35,36) or treadmill running (7). In addition, Rundell(26) found an 8% decrease in peak ˙VO2 and a two-fold increase in submaximal blood lactate concentration when comparing skating in the low “sitting” position with skating in an upright position. Moreover, submaximal ˙VO2 was depressed when skaters assumed the characteristic low sitting position. NIRS provides a noninvasive method of evaluating the unique metabolic consequences of speed skating posture. Recently, the NIRS signal has been found to correlate highly with blood flow and O2 consumption during exercise(17). This group (17) estimated a blood flow index, an O2 supply index, and an O2 consumption index from NIRS signals during venous occlusion imposed at rest and immediately following handgrip exercise. These indexes compared favorably with estimated forearm blood flow by strain-gauge plethysmography (R = 0.99) and invasive forearm O2 consumption (R = 0.98).
The belief that the requisite position of speed skating results in reduced blood flow and increased dependence on anaerobic energy production is an intriguing hypothesis supported by a plethora of coincidental data(4-6,10-14,23,25-27,29,30,33-35). The hypothesis assumes that the static nature of the sitting position, in combination with the long contractile cycle of the skating muscles, limits oxygen delivery by impeding blood flow. The static nature of the position may create a situation whereby intramuscular pressure exceeds perfusion pressure, thus limiting flow to the exercising muscle. This scenario could result in both increased cardiac afterload and decreased venous return which would dramatically affect cardiac output (8). Static knee extensions have decreased cardiac output by approximately 40%(21) over dynamic extensions, and reduced blood flow to the working muscles has been tightly coupled to reduced peak ˙VO2(9,32). The depressed cardiac output during static contractions has been thought to result from a decreased venous return owing to the ineffective muscle pump (28). During speed skating, it seems reasonable to assume that the quadriceps could generate sufficient intramuscular tension because of the static nature of the position that peripheral resistance impacts perfusion. Petrofsky et al.(24) found that intramuscular pressure exceeded perfusion pressure in cat muscle at tensions above 50% of maximal. Although static contractions during skating may only approach 20-30% of maximal contractions (34), local pressure gradients may hinder blood flow. In addition, the higher submaximal blood lactate concentration measured during speed skating suggests an increased recruitment of low-oxidative fast-twitch fibers. Blood flow apportioned to these fibers could compromise delivery to the more oxidative muscle fibers. If this assumption is correct, then the corresponding differences between skating up and skating low would be reflected in a lower ˙VO2 (during LS) without the hypothesized reduced blood flow in effect. Since oxygen uptake reflects the mass-averaged response of a heterogeneous mass of muscle fibers, this assumption could provide an alternative explanation for the reduced blood flow hypothesis.
In this study the dramatic increase in Hb/Mb deoxygenation and the attenuated changes in exercise hyperemia during skating in the sitting position, paralleled by an almost two-fold increase in post-skate blood lactate concentration but not whole body ˙VO2, supports the reduced blood flow hypothesis and is in agreement with the work of Homma et al.(17). The large increase in blood lactate concentration when the skater assumes the sitting position suggests that oxygen delivery does not meet the bioenergetic requirements of the exercising tissue. Although the lack of any significant correlation between whole body ˙VO2 and exercising tissue deoxygenation is not consistent with the work of Belardinelli et al. (1), the mode of exercise and the fitness level of subjects studied could explain this inconsistency. Belardinelli et al. (1) used untrained subjects during incremental cycle ergometry, while the subjects in this study were highly trained speed skaters. Also Belardinelli et al. (1) saw little change in tissue deoxygenation above the lactate threshold. In our study the magnitude of change in tissue deoxygenation between US (at or below lactate threshold) and LS (above lactate threshold) was the same. Potential blood flow changes between skating positions and/or increased recruitment of low- oxidative fast-twitch muscle fibers make this study uniquely different than any previous studies incorporating NIRS. The increased deoxygenation during LS could have been a consequence of decreased blood flow and/or lactic acid facilitated dissociation (Bohr effect) of the O2Hb complex. Strong arguments can be presented for both possibilities. In fact, the consequential metabolic stress from insufficient oxygen delivery coincident with the recruitment of low-oxidative fast-twitch muscle fibers could result in increased lactate production and subsequently facilitate O2 dissociation from the hemoglobin molecule. Our observation of increased deoxygenation at the tissue level without a corresponding increase in oxygen uptake strongly suggests compromised blood flow to the working muscles. Further, the diminished exercise induced hyperemia during LS (compared with US; compare Table 3, Fig. 8) could be a result of reduced blood flow during LS.
Similar ˙VO2 values for US and LS, in spite of higher heart rate and two-fold higher blood lactate for LS, coupled with the 59% and 49% increases in delta OD (at 2.7 and 3.1 m·s-1, respectively) provides strong evidence for reduced perfusion of the exercising muscle during LS. The higher heart rate during LS may be the consequence of increased catecholamine production because of the static nature of the position(28) and/or a compensatory attempt by the heart to maintain cardiac output in spite of a falling stroke volume(20). The large rise in tissue deoxygenation when the skater assumes the low sitting position suggests a facilitated unloading of O2 to meet increased ATP demand. Conversely, O2 demand may not appreciably change, and the increased deoxygenation signal is a consequence of similar quantitative O2 extraction under conditions of reduced perfusion. No definitive change in whole body O2 uptake (< 2%) albeit the 49% increase in delta OD at the faster speed supports reduced perfusion.
If the lower skating position reduces blood flow, then oxygen demand for aerobic energy production must be met by increased unloading of oxygen from hemoglobin. This concept is supported by Stringer et al.(31) who found that femoral vein PO2 did not change upon reaching a “floor” value (≈20 torr), in spite increased HbO2 desaturation at exercise intensities above lactate threshold. It was concluded (31) that HbO2 desaturation at intensities below lactate threshold were a result of muscle end-capillary PO2, while lactic acidosis accounted for virtually all of the HbO2 desaturation at intensities above lactate threshold (Bohr effect). The greater than two-fold increase in blood lactate concentrations when the skaters assumed the low position and the strong relationship of blood lactate, but not oxygen uptake, to deoxygenation (R = 0.95, compareFig. 6) suggests that HbO2 dissociation is in part facilitated by lactic acidosis. The blood lactate concentrations during US at the two different speeds represent intensities just below and at or slightly above lactate threshold (compare Table 1). The significant increase in O2 desaturation at these intensities (during US) for the vastus lateralis and the rectus femoris (but not other muscles examined) is similar to the response observed by Belardinelli et al.(1), who found a rapid desaturation in the vastus lateralis associated with lactate threshold. The high blood lactate values and the increase in tissue desaturation during LS at both speeds supports the role of lactic acidosis facilitated desaturation. Moreover, the significantly greater deoxygenation during LS at 2.7 m·s-1 than during US at 3.1 m·s-1, albeit higher ˙VO2 during US dissociates Hb/Mb desaturation from whole body ˙VO2 and supports lactic acid facilitated uncoupling of the HbO2 complex.
Although one would hypothesize that a hyperemic response would be evident during LS if blood flow was restricted, we found only slight changes in blood volume (no hyperemia) during exercise in the low position. Since cuff ischemia in the forearm has been shown to result in a hyperemia(3), we expected to observe a consistent hyperemic response during cuff ischemia of the leg and during LS. Hyperemia during cuff ischemia was evident in the vastus medialis and biceps femoris in our subjects, but not in the vastus lateralis or rectus femoris. Given the number of deep vessels in the quadricep muscle, the 300 torr cuff pressure may not have completely impeded arterial flow.
The minor blood volume change during LS was similar to the cuff ischemia response noted with our subjects. This small change in blood volume during LS could have been a result of several factors. Although we did not investigate pressure response in this study, the rapid increase in blood volume during US could be a function of the hydrostatic pressure in the upright position resulting in venous pooling, which would be attenuated during LS. In addition, a reduction in arterial flow because of the static nature of LS, without compromising the more peripheral venous effluent flow, could be responsible for the difference in blood volume change between the two skating positions. Increased heart rate, blood lactate, and deoxygenation during LS, in spite of similar ˙VO2 values provides support for this observation. Although blood flow through the exercising muscles was not measured, change in blood volume inferred from the sum of signal change at 760 and 850 nm provides insight to flow dynamics. The observed high rate of blood volume increase at the onset of exercise during US and the 10-50% lower rate of change during LS could be interpreted as compromised flow, whereby diminished flow rate would be reflected in an incomplete perfusion of the exercising muscle. Conversely, during US flow should not be influenced by position (as the static deep sitting during LS) and should allow the muscle bed to become well perfused and should be reflected in an increase in total Hb/Mb signal. The significant relationship between blood volume and positional joint angles supports this hypothesis.
Resolution between Hb and Mb saturation cannot be distinguished by NIRS owing to overlap of the Hb and Mb spectra. However, current evidence by Wang et al. (38) using proton 1H-NMR suggests that 75% of the NIRS signal is a result of HbO2 → HB while 25% of the signal is due to MbO2 → Hb. When US and LS deoxygenation is represented as percent of ischemic deoxygenation, our data suggest that only Hb deoxygenation occurred during US, while the added stress of skating low invoked dissociation of the higher oxygen affinity MbO2 complex. However, limitations of this study prevent us from confirming MbO2 dissociation.
The strong relationship between blood lactate concentration and desaturation suggests that the Bohr effect may be critical to O2 dissociation at high exercise intensities where oxygen delivery does not meet oxygen demand. The high blood lactate coincident to the low sitting position may be a result of increased recruitment of the fast-twitch fiber population in concert with compromised blood flow. This suggestion is supported by data from ice hockey players (15) demonstrating rapid muscle glycogen depletion in both fast-twitch and slow-twitch muscle fibers during high speed interval skating (analogous to LS), but muscle glycogen depletion primarily in slow-twitch fibers during steady skating (analogous to US). Mechanical compression in the sitting position clearly affects arterial flow. Whether blood flow and lactate production are related to altered recruitment patterns such that a given flow is apportioned among a greater muscle mass(consisting of a higher percentage of low oxidative fast-twitch fibers) has yet to be elucidated.
In summary, we have extended our previous work (26) by using NIRS to demonstrate increased Hb/Mb oxygen desaturation during speed skating in the low `sitting' position. Hb/Mb oxygen desaturation was highly related to blood lactate concentration but not whole body ˙VO2. Moreover, blood volume differences between skating upright and low lend further support to the reduced blood flow hypothesis of speed skating.
1. Belardinelli, R., T. J. Barstow, J. Porszasz, and K. Wasserman. Changes in skeletal muscle oxygenation during incremental exercise measured with near infrared spectroscopy. Eur. J. Appl. Physiol.
2. Chance, B., M. T. Dait, C. Zhang, T. Hamaoka, and F. Hagerman. Recovery from exercise-induced desaturation in the quadriceps muscles of elite competitive rowers. Am. J. Physiol.
262 (Cell Physiol:31): C766-C775, 1992.
3. Chance, B., S. Nioko, J. Kent, et al. Time-resolved spectroscopy of hemoglobin and myoglobin in resting and ischemic muscle.Anal. Biochem.
4. Deboer, R. W., G. J. C. Ettema, B. G. M. Faessen, et al. Specific characteristics of speed skating: implications for summer training.Med. Sci. Sports Exerc.
5. Degroot, G., G. J. Van Ingen Schenau, and R. W. Deboer. Evaluation of speed skating capacity on the basis of tests. In:Physiological, Biomechanical and Technical Aspects of Speed Skating
. P. Rispens and R. Lamberts (Eds.). Gronigen: Private Press, 1985, pp. 39-48.
6. De Koning, J. J., F. C. Bakker, G. De Groot, and G. J. Van Ingen Schenau. Longitudinal development of young talented speed skaters: physiological and anthropometric aspects. J. Appl. Physiol.
7. Ekblom, B. The effect of physical training on oxygen transport system in man. Acta Physiol. Scand.
8. Ekblom, B. and L. Hermansen. Cardiac output in athletes.J. Appl. Physiol.
9. Erney, T. P., G. M. Mathien, and R. L. Terjung. Muscle adaptations in trained rats with peripheral arterial insufficiency. Am. J. Physiol.
10. Foster, C., N. N. Thompson, and A. C. Synder. Ergometric studies with speed skaters: evolution of laboratory methods.J. Strength Cond. Res.
11. Foster, C., M. Green, A. C. Snyder, and N. N. Thompson. Physiological responses during simulated competition. Med. Sci. Sports Exerc.
12. Foster, C., A. C. Snyder, N. N. Thompson, and K. Kuettel. Normalization of the blood lactate
profile in athletes. Int. J. Sports Med.
13. Foster, C., N. N. Thompson, and A. C. Synder. In:Physiology of Speed Skating
. Milwaukee, WI: University of Wisconsin Medical School, 1990, pp.6-13.
14. Geysel, J. S. M., G. Bomhoff, J. Van Velzen, G. De Groot, and G. J. van Ingen Schenau. Bicycle ergometry and speed skating performance. Int. J. Sports Med.
15. Green, H. G. Glycogen depletion pattern during continuous and intermittent ice skating. Med. Sci. Sports Exerc.
16. Hampson, N. B. and C. A. Piantadosi. Near-infrared monitoring of human skeletal muscle oxygenation during forearm ischemia.J. Appl. Physiol.
17. Homma, S., H. Eda, S. Ogasawara, and A. Kagaya. Near-infrared estimation of O2
supply and consumption in forearm muscles working at varying intensity. J. Appl. Physiol.
18. Jackson, A. S. and M. L. Pollock. Generalized equation for predicting body density of men. Brit. J. Nutrition
19. Kandou, T. W. A., I. L. D. Houtman, E. Van Der Bol, R. W. Deboer, G. Degroot, And G. J. Van Ingen Schenau. Comparison of physiology and biomechanics of speed skating with cycling and skateboard exercise.Can. J. Sport Sci.
20. Koike, A., K. Wasserman, D. K. McKenzie, S. Zanconato, and D. Weiler-Ravell. Evidence that diffusion limitation determines oxygen uptake kinetics during exercise in humans. J. Clin. Invest.
21. Lewis, S. F., P. G. Snell, W. F. Taylor, et al. Role of muscle mass and mode of contraction in circulatory responses to exercise.J. Appl. Physiol.
22. Millikan, G. A. Experiments in muscle haemoglobin.Proc. R. Soc. Lond. B Biol. Sci.
23. Nemoto, I., K. Iwaoka, K. Funato, N. Yoshioka, and M. Miyashita. Aerobic threshold, anaerobic threshold and maximal oxygen uptake of Japanese speed skaters. Int. J. Sports Med.
24. Petrofsky, J. S., D. Hanpeter, and C. A. Phillips. Intramuscular pressure during isometric exercise in fast and slow muscles of the cat (Abstract). Physiologist
25. Rundell, K. W. Effects of drafting during short track speed skating. Med. Sci. Sports Exerc.
26. Rundell, K. W. Compromised oxygen uptake in speed skaters during treadmill in-line skating. Med. Sci. Sports Exerc.
27. Rundell, K. W. and L. P. Pripstein. Physiological responses of speed skaters to treadmill low walking and cycle ergometry.Int. J. Sports Med.
28. Sawka, M. N. Physiology of upper body exercise. In:Exercise and Sport Sciences Reviews
. Vol. 14. K.B. Pandolf (Ed.). Baltimore: Williams & Wilkins, 1986, pp. 75-211.
29. Snyder, A. C. and C. Foster. Physiology and nutrition for skating. In:Perspectives in Exercise Science And Sports Medicine
, Vol. 7, D. R. Lamb, H.G. Knuttgen, and R. Murray (Eds.). Carmel. IN:Cooper Publishing Group, 1994, pp. 181-219.
30. Snyder, A. C., K. P. O'Hagen, P. S. Clifford, M. D. Hoffman, and C. Foster. Exercise responses to in-line skating: comparisons to running and cycling. Int. J. Sports Med.
31. Stringer, W., K. Wasserman, R. Casaburi, J. Porszasz, K. Maehara, and W. French. Lactic acidosis as a facilitator of oxyhemoglobin dissociation during exercise. J. Appl. Physiol.
32. Sundberg, C. J., O. Eiken, A. Nygren, and L. Kaijser. Effects of ischaemic training on local aerobic muscle performance in man.Acta Physiol. Scand.
33. Van Ingen Schenau, G. J., F. C. Bakkwe, G. Degroot, and J. J. Dekoning. Supramaximal cycle tests do not detect seasonal progression in performance in groups of elite speed skaters. Eur. J. Appl. Physiol.
34. Van Ingen Schenau, G. J., J. J. Dekoning, and G. Degroot. Assimulation of speed skating performances based on a power equation.Med. Sci. Sports Exerc.
35. Van Ingen Schenau, G. J., R. W. Deboer, J. S. M. Geijsel, and G. Degroot. Supramaximal test results of male and female speed skaters with particular reference to methodological problems. Eur. J. Appl. Physiol.
36. Van Ingen Schenau, G. J., G. Degroot, and A. P. Hollander. Some technical, physiological and anthropometric aspects of speed skating. Eur. J. Appl. Physiol. Occup. Physiol.
37. Van Ingen Schenau, G. J. The influence of air friction in speed skating. J. Biomechanics
38. Wang, D. J., Z. Wang, E. Noyszewski, et al. Correlation of optical and 1H NMR of Hb and Mb deoxygenation in canine gastrocnemius(Abstract). Soc. Magn. Reson. Med. 9th Ann. Meet.
New York, 1990, 1:175, 1990.