INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A PROBABLE FUTURE STEP FOR space exploration is to send a manned spaceflight to Mars. However, many physiological decrements, such as losses in muscle strength (7), bone density (3, 19, 25), balance (23), aerobic capacity (3), and orthostatic tolerance (8), occur during long-duration spaceflights and may have adverse effects on crew safety and performance. On Earth, blood pressures are greater in the feet than at the heart or head during upright posture because of gravity's effects on columns of blood in the body (13). During exposure to microgravity, all gravitational blood pressures disappear. The lack of gravitational blood pressures that occurs because of microgravity may compromise blood vessel function in gravity and cause orthostatic intolerance (29). These detrimental effects of spaceflight due to microgravity must be counteracted to bring astronauts back to Earth safely after a long spaceflight.

Presently, exercise protocols and equipment for astronauts in space are unresolved (3, 10). Prior studies and pilot work in our laboratory indicate that, because of the design of the current treadmill and bungee cord systems, all exercise in space to date has lacked sufficient mechanical and physiological loads to maintain preflight musculoskeletal mass, strength, and aerobic capacity (4, 9, 27, 29). Discomfort from shoulder strap and waist belt compression limits bungee cord harness systems to 60-70% body wt (BW) in microgravity (7).

Cycle ergometry has been effective in stressing the cardiovascular system during spaceflight (9, 18). However, exercise protocols have not included sufficient footward force [which we will call ground reaction force (GRF)] at the feet to maintain bone density (25). GRFs at the feet and rates of force generation are both factors in maintaining bone density on Earth (5, 24, 30). Also, Hargens et al. (12) stressed the need for eccentric skeletal muscle exercise during spaceflight. Eccentric exercise is more effective than concentric exercise for increasing strength without an appreciable increase in energy cost (1, 6, 26). Running has a larger eccentric component than cycle ergometry, and, on Earth, running generates two to three times the BW during the stance phase (21). Therefore, running may be a time-effective and energy-efficient countermeasure for maintaining muscular strength and bone density during spaceflight if adequate GRFs can be obtained.

Resistive exercise may also be incorporated for musculoskeletal maintenance, but, because time is limited during spaceflights, it is important to stress as many systems as possible with one exercise device. Over the past several years, lower body negative pressure (LBNP) exercise (14, 20) was developed for exercise in microgravity. It allows running and therefore eccentric exercise (16), as well as simulated gravitational blood pressures, within one device. It is less costly than a human-rated centrifuge apparatus, and it may help to create musculoskeletal and cardiovascular strains equal to or greater than those experienced on Earth. No device presently available during spaceflight simulates these gravitational blood pressures in space during exercise.

The ability of the LBNP device to maintain exercise fitness during bed rest was studied by Lee et al. (17). That study examined the efficacy of 30 min of supine LBNP treadmill running for maintaining heart rate (HR) and respiratory responses after 5 days of bed rest. Subjects in that study maintained upright exercise capacity after 5 days of bed rest using this LBNP exercise protocol. However, Lee et al. did not analyze oxygen consumption (O₂), gait mechanics, or footward forces during the exercise protocol and, therefore, did not substantiate that these variables are similar for both supine LBNP and upright exercise on a treadmill. Murthy et al. (20) examined the utility of LBNP exercise for simulating upright exercise. The study was limited to plantarflexing and dorsiflexing the foot as exercise but did not analyze walking or running. That study also used a waist seal that required 100 mmHg in supine LBNP to generate one BW of footward force. With the use of this device, HR was much higher than during upright exercise.

Therefore, the purpose of the present study was to determine whether the kinematics, musculoskeletal loading, and metabolic rate during supine walking and running on a vertical treadmill within LBNP are similar to those on a level treadmill in an upright posture in Earth gravity (1 G). We hypothesized that gait mechanics, GRFs, O₂, and HR during supine LBNP exercise would approximate those during upright 1-G treadmill exercise.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Eight healthy subjects [5 men and 3 women; age 30 ± 2 (SD) yr; height 169 ± 8 cm; weight 67.8 ± 12.0 kg] participated in the study after giving their informed, written consent. This protocol was approved by the Human Research Experiments Review Board at National Aeronautics and Space Administration Ames Research Center. Subjects maintained normal daily activities and refrained from caffeine, alcohol, medications, and strenuous exercise 24 h before the study.

LBNP exercise. The LBNP exercise device employed in the present study used the negative pressure to pull subjects, who are suspended supine, inward against a treadmill that resides inside the chamber. GRFs are generated, which equal the product of the pressure differential and the cross-sectional area at the level of the flexible waist seal of the LBNP chamber (Fig. 1) (14, 31). GRFs can be comfortably raised by increasing the suction pressure within the chamber. One BW of GRF can be generated by using negative pressures of only 50-60 mmHg by radially increasing the flexible surface area of the waist seal. This equals approximately one-half of the pressure previously needed to generate one BW using LBNP (20). The subjects walk or run on a treadmill that is positioned vertically within the chamber. This orientation avoids the effects of gravity in the z axis and allows the negative pressure within the chamber solely to determine GRF, which is analogous to how LBNP exercise would operate in microgravity.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Exercise on a treadmill during lower body negative pressure (LBNP) exposure. Footward force (Gz) is produced by the suction force of LBNP. Ground reaction force (GRF) during treadmill exercise in LBNP while supine and in microgravity equals the product of the body cross-sectional area (Axy) and the pressure differential (P) across the LBNP chamber, with additional inertial forces equaling body mass (M) multiplied by footward acceleration (az) at the body center of mass (cm). The subject is suspended at the ankles and thighs by a pulley system, and the hands are holding the suspension cables at waist level. Gx, Earth gravity.

Testing protocol. At least five days before testing, subjects came to the laboratory to self-select their walking and running speeds and to become comfortable with ambulating on the treadmill (Aerobics, Little Falls, NJ). On the day of testing, reflective markers were placed on the shoulder, hip, knee, ankle, and fifth metatarsal. Two reference markers were positioned in the camera view for calibration. Subjects were instrumented for HR measurement.

Metabolic data. O₂ was measured and averaged every 15 s by using turbine volumetry (model S-301, Pnueumoscan, K.L. Engineering, Sylman, CA) and gas analysis (Applied Electrochemistry, Ametek, Thermox Instruments Division, Sunnyvale, CA). Data averaged over the fifth minute of exercise were used for statistical analysis. HR was measured both by electrocardiogram leads placed on the skin and a telemetric HR monitor (Polar CIC, Port Washington, NY) positioned around the chest.

Kinetics/force data. The exact level of LBNP necessary to generate one BW of GRF was determined for each subject by using a force insole (Electronic Quantification, Plymouth Meeting, PA) inserted in the left shoe. The force measured by the force insole was calibrated in the z axis with an AMTI force plate (model OR6-5-1, Biomechanics Platform, Advanced Mechanical Technology, Newton, MA). Insole GRF data were assessed at a sampling rate of 500 Hz during the third minute of each exercise period to correspond to gait kinematics during the third minute.

Movement kinematics. Analysis of movement kinematics was accomplished by using video analysis techniques. Subjects were videotaped with a Minolta C-570 camcorder during the third minute of each exercise period at a frame rate of 60 Hz to obtain five consecutive strides of a sagittal plane view. The video data were digitized by using a Peak Performance motion analysis system (Englewood, CO) and filtered to smooth the data by using a second-order Butterworth filter (15).

Variables. The metabolic variables measured were HR (beats/min) and O₂ (ml · min¹ · kg¹; l/min). Force variables measured were integrated GRF (N · s), rate of force development (N/s), and peak force (N). Kinematic variables measured were stance time, swing time, step frequency, step length, step length relative to leg length, and maximum rise distance of the foot. Maximum knee, hip, and ankle angles were also measured during the stance and swing phases.

Statistical analysis. Statistical differences were assessed by using a repeated-measures ANOVA with Tukey-Kramer post hoc tests. Significance was noted at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Metabolic data. Metabolic variables were not significantly different between upright gait and supine LBNP gait. HR (Table 1) was not significantly different between upright (136 ± 7 beats/min, overall mean ± SE) and LBNP exercise (134 ± 8 beats/min). As expected, HR was significantly higher during running than walking. O₂ (Table 1) was not significantly different between upright and LBNP exercise. Again, as expected, O₂ was significantly higher for running than for walking.

Kinetics/force data. There were few observed differences in kinetic variables between upright and LBNP gait. Figures 2 and 3 illustrate walking and running GRF raw data, respectively, for one subject during supine LBNP and upright 1-G exercise. Table 2 outlines the GRF results. GRFs integrated over each stride were the same for LBNP and upright exercise. Rate of force development was the same for upright and LBNP conditions. Peak impact and push-off GRFs were similar for walking in LBNP and upright walking, but peak GRF during running was 17% less during LBNP than in the upright condition. As expected, all kinetic variables were significantly different between running and walking.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Raw GRF data during supine LBNP treadmill walking (A) and upright treadmill walking in gravity on the ground (1 G; B) expressed as %body weight (%BW). Note the similarity in the waveforms of the two conditions.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Raw GRF data during supine LBNP treadmill running (A) and upright 1-G treadmill running (B) expressed as %BW.

Movement kinematics. Overall, the kinematic variables measured were very similar, but subtle yet significant differences were detected between upright and supine LBNP exercise. Table 3 outlines the kinematic results. Maximum rise distance of the foot was significantly higher during upright running (0.12 ± 0.02 m) than running in the LBNP chamber (0.08 ± 0.014 m; P < 0.05). This variation was not significant for walking. However, maximum rise of the foot was significantly higher for running than for walking (0.10 ± 0.013 vs. 0.07 ± 0.003 m, P < 0.05). Knee flexion during the swing phase was significantly less during LBNP walking (57.6 ± 2.6°) than during upright walking (71.8 ± 1.9°; P < 0.05). This variation was not significant for running. Knee flexion angle during the swing phase was also significantly less for walking (64.7 ± 2.4°) than for running (81.7 ± 2.5°; P < 0.05). Hip flexion angle during the stance phase was less for walking upright (19.7 ± 1.0°) than in LBNP (28.9 ± 0.8°; P < 0.05), but running had greater hip flexion angle upright (26.1 ± 1.9°) than in LBNP (20.5 ± 1.4°; P < 0.05). As expected, step frequency, step length, stride time, and stance time were significantly different between walking and running (P < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Despite some minor differences in gait mechanics and GRF profiles between supine LBNP and upright gait, exercise within LBNP is physically demanding and remarkably similar to upright exercise against gravity. Integrated GRFs and rate of force generation within LBNP are the same as those determined in upright conditions. Peak GRFs within LBNP are similar or close to the values produced during upright gait (5) and can be equaled by adjusting the level of negative pressure within the chamber. O₂ and HR are very similar for LBNP and upright gait, which is an improvement over previous devices (20). This study also used a newly devised waist seal with a larger diameter, which required pressures of only ~50-60 mmHg to generate one BW of footward force (28). This lower pressure elicits HR responses that are closer to upright conditions than what Murthy et al. (20) observed. The present study extended the Murthy et al. study by having subjects perform treadmill exercise upright and within the LBNP chamber on a treadmill to determine whether gait kinematics, O₂, HR, and GRFs are similar for both conditions using a more rigorous cardiovascular exercise. The speed of gait changes many observable gait and metabolic characteristics. For example, step length, step frequency, and hip and knee joint angles increase with increased speed (22). HR and O₂ also increase with increases in speed (2). Many of these expected changes were observed in this study. However, there were some differences observed between supine LBNP and upright gait that require explanation.

The kinematic differences observed during LBNP exercise, such as the decrease in knee flexion, hip flexion, and maximum rise of the foot during the swing phase of gait, are probably due to the leg and back suspension system and the waist seal in the LBNP. This suspension system is necessary to counteract Earth gravity (with respect to the subject; Fig. 1) but will not be necessary in microgravity. As the leg extends through the push-off phase, there is resistance against the suspension cords as the contralateral leg starts to bend during the stance phase. When one knee is flexed and the other extended, there is maximal counterresistance against the suspension system. In microgravity, there will be no need to suspend the subjects against gravity, and the suspension system will be eliminated. Although there are some small kinematic and kinetic differences between upright and supine LBNP gait, they do not translate into significant metabolic differences.

Kinetics/force data. It is thought that rate of force development and the magnitude of load bearing are both necessary stimuli for maintaining bone density (24, 30). Hargens et al. (12) and Dudley et al. (6) suggested that eccentric exercise is needed during spaceflight to maintain muscular strength. On Earth, running is known to incorporate both concentric and eccentric muscle contractions (16). The rate of force development and the magnitude of the peak GRFs for walking and running were well within normal ranges in this study (5). The present results suggest that running within LBNP supplies an eccentric exercise component, as is proposed for maintaining musculoskeletal structure and function during spaceflight. If it is necessary to increase the magnitude of GRF produced within the chamber, the negative pressure or waist seal cross-sectional area can be increased to raise GRF levels (11). The shape of the GRF curves produced within LBNP are also nearly identical to those produced during upright walking and running, indicating that the timing of GRF production within the LBNP is very similar to that of upright gait.

Movement kinematics. The small reduction in range of motion observed during LBNP exercise is probably due to the leg suspension system and the waist seal in the LBNP chamber. The decrease in knee flexion, hip flexion, and maximum rise of the foot during the swing phase of gait is also likely due to the subject suspension system in the LBNP. The suspension cords and the back support, which extend below the hips, decrease overall hip flexion and extension during gait. At the most flexed position of the knee, there is the most resistance from the suspension system. At maximum knee flexion and foot rise, it is easier for the subject to limit knee flexion than to exert force against the suspension cords.

Limitations of LBNP treadmill exercise. LBNP exercise may not stimulate the vestibular system in the same way that upright exercise in gravity does. LBNP exercise will simulate the inertial accelerations of gait but will not produce the static acceleration that gravity imposes on our vestibular system over long periods of time. Supine exercise was used in this study because it is the best simulation of microgravity on Earth. The subject suspension system was necessary to perform supine treadmill exercise, yet this system influenced our results in ways that would not occur in microgravity, where no suspension system would be necessary. Treadmill exercise does not provide normal visual flow, but virtual environment systems could be designed to provide this type of stimulation.