Endurance exercise training in healthy subjects typically results in improvements in exercise tolerance. The peak work rate, ˙VO2(8,12), and cardiac output (31) are all increased, as is the time to exhaustion. End-exercise ˙VO2 and blood lactate levels, on the other hand, are both reduced during heavy submaximal exercise (8). These results are thought to be the result of a combination of central (cardiac) and peripheral (skeletal muscle) adaptations, which result in greater capacity to deliver O2 from the lungs to the mitochondria of the contracting muscles. These improvements are observed whether the subjects are young(22) or elderly (5), male or female(5), and can be elicited in the presence of cardiovascular(17) and/or pulmonary disease (7).
Conversely, spinal cord injury (SCI) leads to rapid atrophy of the muscles of the afflicted limbs, with reduced resting blood flow(34). Methods to produce functional electrical stimulation (FES) of the afflicted limb muscles in patients with SCI have been recently been developed (26). The goal has been to maintain or increase muscle mass and blood flow so as to reduce the incidence of pressure ischium ulcers, maintain muscle mass for subsequent ambulation, and for aesthetic reasons. Functional electrical stimulation training does improve FES exercise tolerance in these patients, manifest both as peak˙VO2 and work rate(11,15,20,27,28).
Changes in exercise capacity and the associated metabolic responses at peak incremental exercise after training in able-bodied volunteers appear to be dependent on the ability to improve cardiac output (31), as well as on any changes in the aerobic capacity of the contracting muscles themselves (14). We have previously reported a lower-than-expected heart rate response during FES leg cycle exercise in patients with SCI, given the exhaustive nature of the exercise, as evidenced by the respiratory exchange ratio (RER) being above 1.0 and elevation of blood lactate (4). It is unclear what the consequences of a blunted heart rate response to exercise would be on exercise capacity (as peak˙VO2 and work rate) in this patient population. A reduced rise in heart rate may also compromise the ability of the contracting muscles to increase and/or sustain aerobic metabolism during transitions to and from submaximal exercise. A noninvasive method for examining muscle aerobic capacity and its sensitivity to O2 delivery during submaximal exercise is to determine the kinetics, or rate of adjustment, of pulmonary gas exchange(˙VO2, ˙VCO2, and ˙VE) following the onset, and in recovery from, constant work rate exercise. In general, the kinetics, particularly of ˙VO2, follow changes in chronic usage of the muscles, speeding with training (8,12) and slowing with bed rest (9). However, the kinetics of˙VO2 during exercise are sensitive not only to the inherent oxidative characteristics of the muscles, but also are sensitive to changes in O2 delivery, e.g., slowing of the kinetics of circulatory adjustments with β-blockade (25) or reducing arterial O2 content with hypoxia (33) or carbon monoxide breathing(19) results in slower kinetics for ˙VO2. In contrast, during recovery from exercise there is good reason to believe that a slower adjustment of the circulation compared to the onset of exercise would result in a relative hyperemia of the contracting muscles(3). In this case, recovery ˙VO2 kinetics would be less sensitive to, or even independent of, O2 delivery, and would therefore be a better reflection of the inherent oxidative characteristics of the contracting muscles.
The purposes of the present study were to a) assess the trainability of a group of patients with SCI performing chronic FES leg cycle exercise, b) evaluate if the training adaptations include improvement in peripheral utilization of O2, as indicated by a speeding of the kinetics of pulmonary gas exchange (especially ˙VO2), and c) test if the magnitude of training adaptations were correlated with heart rate responses during exercise.
Subjects. Nine male volunteers with SCI were enrolled in the project after giving written informed consent. The project was approved by the Research and Development Committee at the West Los Angeles VA Medical Center. Prior to testing, medical screening was performed on each subject, which included physical exam including a sensory/motor neurological assessment, x-rays of the chest, spine, and lower limbs, computerized tomography and magnetic resonance imaging of the lower extremities, blood chemistry and urinalysis, and a resting and stress test ECG during arm crank ergometer testing. All subjects were ASIA Class A (i.e., complete sensory and motor block below the level of the injury)(1). Two subjects(nos. 1 and 2 in Table 1) had complete quadriplegia, with upper extremity weakness and trunk instability. These two subjects required assistance with daily activities (e.g., transfers, bathing, etc.). The other seven subjects were paraplegics with complete use of their upper body musculature. These subjects did not require daily assistance in their lifestyles. Subjects are described in Table 1.
Leg exercise training with FES. Exercise was performed on a REGYS1 FES leg cycle ergometer and computer system (Therapeutic Alliances, Inc., Fairborn, OH), as previously described (4). The exercise was elicited by sequential stimulation of the quadriceps and gluteal muscle groups via carbon-filled silastic surface electrodes which were placed over the motor points for each muscle. The current was varied automatically by the computer over the range of 10-132 mA at the minimum level necessary to maintain a pedalling frequency of 50 rpm. Limits in the design of the system do not permit recording of the actual current used during a study.
The training protocol has been described elsewhere(15) and consisted of a minimum of 24 exercise sessions, which were intended to be performed three times per week. At each session, FES leg cycle exercise was performed for 30 min. The initial work rate was unloaded cycling. Once the subject could complete three consecutive 30-min sessions at a given work rate, the work rate was increased by 1/8 kp (6.1 W at 50 rpm). If during the training session the subject could not tolerate the work rate (i.e., rpm fell below 35), then the work rate was reduced to the previous level and exercise continued until a total of 30 min was accomplished. Prior to the pretraining testing, an initial phase of training was accomplished to bring each subject up to a common minimum level of conditioning. This phase consisted initially of FES leg lifts with light weight to strengthen the quadriceps. This was subsequently followed by intermittent exercise at unloaded cycling, during which the periods of exercise were gradually lengthened until the subject could perform 30 min of continuous exercise at unloaded cycling on two separate occasions. Once the subject could perform 30 min of continuous exercise, the pretraining testing was performed.
Exercise tests. Each subject performed both incremental and constant work rate tests on a cycle ergometer prior to and following the training period. In the incremental test the work rate was increased by 1/8 kp every 5 min until the exercise could no longer be sustained, owing to muscle fatigue. The constant work rate tests were maintained for at least 10 min. Prior to training the subjects were only capable of “cycling” with no breaking force applied to the flywheel of the ergometer (unloaded cycling). This was also set as the initial training work rate. Post training the subjects performed two such tests: one at the same absolute work rate (i.e., unloaded) and one at the same relative work rate (i.e., the end-training intensity). The constant work rate protocol began with two minutes assisted cycling, during which the pedals were turned at 40-45 rpm by a technician while FES was applied to the muscles. This was followed by 10 min of unassisted, unloaded cycling produced by FES. Recovery began with assisted pedalling without FES for the first two minutes, followed by rest for several minutes. The assisted pedalling during recovery was included to minimize blood pooling by maintaining an active muscle pump for the first 2 min.
Pulmonary gas exchange (˙VO2, ˙VCO2, and minute ventilation ˙VE) was measured breath-by-breath throughout each testing protocol using a Medical Graphics CPX metabolic cart (Medical Graphics Corporation, Minneapolis, MN). The gas analyzers and pneumotachograph were calibrated before each exercise test. In addition, the system was validated at regular intervals using a metabolic simulator (MGC)(16). Heart rate was continuously monitored via a six-lead ECG; this was printed out every 30 s in addition to being stored in the breath-by-breath gas exchange file.
Data analysis. Gas exchange variables from the constant work rate tests were prepared for curve fitting by aligning the start and end of exercise, and interpolating the breath-by-breath data to create evenly spaced data points once per second. The responses of ˙VO2,˙VCO2, and ˙VE for each subject both during exercise and in recovery were then fit to the following exponential model,equation 1
where ˙Vx stands for ˙VO2, ˙VCO2, or˙VE, REF represents the average over the 2 min of assisted exercise for the transition to work, and the last minute of exercise for the transition into recovery, A1 is the increase (exercise) or decrease (recovery) from baseline, TD is a time delay relative to the start of exercise or recovery, and τ is the time constant. The kinetics were described as the mean response time (MRTon for exercise, MRToff for recovery), which is equal to the sum of τ + TD. Although recovery included 2 min of assisted cycling at the beginning, this did not have a visible effect on the gas exchange responses or heart rate in recovery. We, thus, believed that modeling recovery with the single exponential function above was justifiable.
Statistics. Analysis of variance with repeated measures was used to compare the end-exercise values and MRTon and MRToff for˙VO2, ˙VCO2, and ˙VE before and after training. Results were compared both for the same absolute work rate (unloaded cycling) and the same relative work rate (training work rate). Significance was declared for P < 0.05. For significant results from ANOVA, post hoc examination of specific differences was performed using Duncan's multiple range test. Relationships between variables were examined using linear correlational analysis, with subsequent testing of the correlation coefficient r for significant differences from 0. Data are presented mean± SD.
Training. The training sessions/week averaged 2.1 ± 0.4. Eight of the nine subjects were able to increase their training work rate, from unloaded cycling to a mean work rate of 10.2 ± 5.3 W; one subject, however, remained at unloaded cycling. There were no complications reported during any of the training sessions.
Peak exercise. One subject was not able to be tested before training commenced; thus, the number of subjects whose peak exercise responses were compared pre- and post-training was eight. Training significantly increased peak ˙VO2 from 1.28 ± 0.31 to 1.42 ± 0.34 l·min-1 (P < 0.05), which was associated with a significant increase in peak work rate achieved (from 9.9 ± 5.6 W to 14.5 ± 5.6 W)(Table 2). Heart rate was unaffected, both at rest (69.3 ± 11.5 pre, 66.4 ± 5.6 post, NS) or peak exercise (134.3 ± 26.7 pre, 131.5 ± 23.8 post, NS)(Table 2). However, heart rate for any given˙VO2 during the incremental test was reduced. This is shown inFigure 1 for the subject with the highest exercise capacity (top panel) and lowest capacity (lower panel). Peak O2 pulse(as peak ˙VO2/peak heart rate) was significantly increased with training, from 9.5 ± 0.7 before training to 10.7 ± 1.5 after training, P < 0.01. For both pre- and post-training, the peak˙VO2 achieved by each subject was highly correlated with the peak heart rate obtained (r = 0.97, pre-training, and r = 0.85 post-training,P < 0.01 for both) (Fig. 2). While both the improvement in peak ˙VO2 and any changes in peak heart rate were variable among subjects, the order of subjects for both peak ˙VO2 and peak heart rate remained the same after training.
Constant work rate exercise. Figure 3 shows the responses of one representative subject (no. 3) for ˙VO2,˙VCO2, and ˙VE for rest, 10 min of unloaded cycling, and recovery. End-exercise levels for ˙VO2, ˙VCO2, and˙VE are summarized for all of the subjects inTable 3. While there was a tendency for end-exercise responses of ˙VCO2 and ˙VE during the unloaded cycling to be lower after training, this did not reach statistical significance. However, the RER was significantly lower after training at the same absolute work rate (unloaded cycling)(1.15 ± 0.12 pre vs 1.06 ± 0.07 post, P < 0.05) but not for the same relative work rate (1.07± 0.09, NS). Heart rate rose from 69.3 ± 11.5 for the assisted cycling baseline to 106.3 ± 17.8 b·min-1 during the pre-training unloaded cycling. After training, it rose from 66.4 ± 5.6 to 109.7 ± 26.1 b·min-1 at this work rate. At the same relative work rate, however, it rose from 64.0 ± 2.2 to 125.3 ± 28.6 b·min-1. There were no significant differences either in the resting, or in the end-exercise values, as a result of training.
In contrast to a general lack of change in steady state values for pulmonary gas exchange, both MRTon and MRToff for˙VO2, ˙VCO2, and ˙VE were significantly reduced (faster responses) after training at the same absolute work rate(P < 0.05)(Table 3). This speeding of the responses was also evident at the same relative work rate, where MRTon for ˙VO2, and MRToff for all three pulmonary variables, were also significantly reduced (P < 0.05)(Table 3).
Predicting training variability. While group mean improvements in peak ˙VO2 and MRTon and MRToff for ˙VO2 were significant following training, there was great individual variability in the extent of the improvements. To explore this variability in responses among the subjects further, we examined the relationships between several end-points and evidence of training adaptations. The pre-training MRTon and MRToff for ˙VO2 were not significantly related to the pre-training peak ˙VO2, nor to either peak heart rate or end-constant-work rate heart rate. Further, the changes in both MRTon and MRToff following training were not related to the improvements in peak ˙VO2, nor to peak or end-exercise heart rate. Neither the pre-training peak ˙VO2 nor the relative improvement in peak˙VO2 were significantly correlated with post injury time.
The present study confirms the observations of others(11,15,20,28) that the peak work rate achieved by untrained patients with SCI performing FES exercise is very low(generally 5-15 W), and that FES training can produce modest but significant improvements. In contrast, increases in peak ˙VO2 of 22-58% have been previously reported for patients with SCI undergoing FES cycle exercise training (15,20,28). This is somewhat more than the 11% increase found here. The lower percent improvement in our study could be due to at least two factors. First, our subjects entered the exercise training program with higher initial peak ˙VO2 values than in the other studies. In part, this may be explained by our initial testing being performed after the subjects could complete 30 min of continuous cycling. Thus, significant improvement in exercise capacity are likely to have already taken place in our subjects. Other studies conducted the pre-training testing earlier in the conditioning sequence(15,20,28). However, our subjects still have higher peak ˙VO2 values even when compared with values in the other studies taken at comparable time points in the training. In the previous studies, as well as in the present one, the percent improvement with FES training appears to be inversely related to the initial aerobic capacity of the patients performing FES leg cycle exercise, i.e., the higher the aerobic capacity before training, the lower the percent improvement. This suggests that there may be a maximal level of exercise tolerance and metabolic rate that can be achieved with FES exercise in these patients. If true, then the closer a subject is to this potential ceiling before training, the less will be the improvement. A second explanation for the smaller training responses observed in our study is that in the previous FES training studies, the subjects trained three times per week, compared to the twice-per-week sessions that our subjects were able to make. The changes in peak ˙VO2 seen here, while small in absolute terms, are similar to those found in able-bodied subjects training for 3-5 d·wk-1 for 7-8 wk (13-15%)(8,13). More frequent training appears to lead to greater improvements in peak ˙VO2 in able-bodied volunteers (24% when training 6 d·wk-1 for 9 wk). It is possible, therefore, that our subjects might have been able to exhibit greater improvements in peak˙VO2 if training sessions had been more frequent. Even so, our results have clinical importance, for they suggest that even with only twice per week training, patients with SCI can undergo significant training adaptations.
As previously noted (4), the kinetics of˙VO2, ˙VCO2 and ˙VE during exercise and in recovery in patients with SCI are among the longest ever reported for a patient population. Similarly long kinetics for PCr and Pi during recovery from tetanic exercise have been reported for patients with SCI as well(21). Patients with congestive heart failure had an average half-time of 72 sec (MRToff of 104 s) for ˙VO2 during recovery from incremental exercise. Patients with cyanotic congenital heart disease have similarly slowed kinetics in the transition from rest to unloaded cycling, with a mean half-time of 63 s, or an equivalent time constant of 91 s (32). Our patients with SCI demonstrated comparable kinetics for ˙VO2 prior to training (mean MRToff of 102 s). But these become much faster after training (mean MRToff of 82 s). As we previously observed (4), MRTon for˙VO2, ˙VCO2 and for ˙VE in untrained patients with SCI shows greater intersubject variability than MRToff, and this remains true after training as well. Interestingly, we found that the training decreased the intersubject variability of both MRTon and MRToff for all three gas exchange variables (Table 3). However, even after training, MRTon remained longer than MRToff. This discrepancy in kinetics has been noted in able bodied subjects performing heavy exercise (i.e., above the lactate threshold LT)(24,29). It has been hypothesized to be due to the O2 debt (or excess post-exercise O2 consumption EPOC) reflecting restoration of muscle PCr stores and venous O2 content, while the onset kinetics (or O2 deficit) include additional energetic contribution from“anaerobic” lactic acid production (37). Thus, the kinetics of ˙VO2 during exercise would be slower than the kinetics in recovery.
The lack of correlation between improvements in peak ˙VO2 and improvements in either MRTon or MRToff with training is consistent with the belief that different mechanisms and/or limitations determine each one in our patients with SCI. Our data are consistent with the view that peak ˙VO2 is determined primarily by peak cardiac output(31). This is reflected in the patients with SCI by peak heart rate (Fig. 2). The strong correlation between peak heart rate and peak ˙VO2, both before and after training, suggests that the exercise capacity of these patients is particularly sensitive to the ability (or inability) to increase heart rate during exercise. This hypothesis is upheld in other studies in patients with SCI, which have noted lower peak˙VO2's prior to training which were consistently associated with peak heart rates that were at the low end of what we observed here(15,20). The low peak heart rates observed in some of our patients were not apparently due to a lack of effort. The exercise stimulus was externally produced, and consequently the exercise was not stopped volitionally, but rather when the contracting muscles could no longer support the exercise at the minimum required rpm. Furthermore, the mean pre- and post-training RER at peak exercise were 1.16 ± 0.08 and 1.10± 0.09, respectively, with only one value less than 1.00 (0.96), also consistent with good “effort.” It is presently unclear what the reason is for the limited heart rate response in these patients. Stimuli below the level of the lesion that are perceived as noxious, including FES, can cause spinal-cord-level vasoconstrictor reflexes and acute hypertension(2). This in turn elicits a bradycardia via normal baroreceptor feedback. This autonomic dysreflexia has been reported in patients with SCI performing FES leg extension exercise(2). While dysreflexia could be involved in the present study, our patients did not exhibit any clinical symptoms associated with this phenomenon.
It is unlikely that the reduced lower extremity muscle mass that accompanies SCI is the reason for the inter-subject variability in peak heart rate and ˙VO2. Examination of the data in Table 1 reveals that the subjects with the highest peak ˙VO2 and heart rate (subject 7) and lowest values (subject 9) were similar in body weight and height. Further, in normal subjects performing voluntary exercise with smaller muscle mass (arm cycling), peak heart rates are similar to those achieved during exercise with larger mass (leg cycling) (6).
An alternative explanation for the low ˙VO2 and heart rate in some of our subjects might be early fatigue of the contracting muscles. This fatigue could be related to early onset of muscle acidosis, possibly due to the extreme deconditioning and/or FES method of muscle stimulation. Fatigue would compromise the ability of the muscles to increase aerobic metabolism. In this case, the coupling of heart rate to metabolic rate (as ˙VO2) might even be normal, but because the ability to increase aerobic metabolism is compromised, so also will be the rise in heart rate. Consistent with this mechanism, the slope of the ˙VO2/HR relationship for subject 7, with the lowest peak HR (lower panel in Fig. 1) is 30 ml·beat-1, suggesting heart rate and ˙VO2 were increasing in a normal fashion until exercise was terminated. For comparison, subject 9, with the highest peak heart rate, had a slope of 19 ml·beat-1, suggesting near normal increase in heart rate, but reduced stroke volume and/or extraction of O2 from the venous blood. However, our present data do not allow us to distinguish among these explanations for the reduced peak heart rate and associated peak˙VO2 seen in some of the subjects with SCI.
While peak ˙VO2 is predominantly dependent on peak cardiac output, MRT for ˙VO2 is determined by muscle aerobic capacity(10,36), and modifiable to some variable extent by O2 delivery by the circulation (25,33). Regarding the former, muscles of the affected limbs in patients with SCI show disproportionate loss of Type I (slow oxidative) muscle fibers, reduced capillaries per fiber, and lower succinate dehydrogenase activity(23,30). This is consistent with a reduced capacity for aerobic metabolism. Chronic electrical stimulation of the anterior tibialis muscle in patients with SCI has been shown to reverse these changes(23). The faster MRTon and MRToff for˙VO2 in the present study resulting from FES exercise training suggests that similar changes in the stimulated muscles may have taken place as a result of the FES training, i.e., increased proportion of Type I fibers, increased mitochondria and capillarity. The extent to which compromised circulatory O2 delivery to the stimulated muscles contributes to the slow ˙VO2 kinetics in these patients with SCI is presently unknown.
In able-bodied subjects, endurance training leads to a reduction in end-exercise ˙VO2 (8) and ˙VCO2(35) when the given work rate is above the LT. In the present study, end-exercise levels of ˙VO2, ˙VCO2 and˙VE did not change significantly for the same absolute work rate(unloaded cycling). This has been observed previously for patients with SCI(11). Although the end-exercise RER was above 1.0 both before and after training in the present study, it was significantly lower during the post testing. This suggests that there may have been a reduction in lactic acidosis, or in some other stimulus which produced hyperventilation, following training. Both the faster ˙VO2 kinetics and the reduced RER following training are consistent with improved O2 delivery to, and/or enhanced aerobic capacity of, the contracting muscles during the exercise.
Changes in the heart rate response during exercise following FES training have been variable. In other studies, the increased peak ˙VO2 following FES training was associated with an increase in peak heart rate(15,20), without any changes in stroke volume or arteriovenous O2 difference (15). In contrast, we saw no increase in peak heart rate following FES training in the present study. However, there was a reduction in heart rate for submaximal work during the incremental exercise (Fig. 1), as also noted by Faghri et al. (11), and peak O2 pulse was significantly increased with training (Table 2). This reduction in heart rate for a given ˙VO2 (and thus increased O2 pulse) in our subjects is suggestive of an increase in stroke volume and/or a greater O2 extraction (wider arteriovenous O2 difference) following training (11). However, end-exercise heart rate during the same absolute work rate (unloaded cycling) was not reduced in our subjects. These results, together with those of other studies(11,15,20), imply that training does not result in a consistent adaptation of the heart rate response to FES exercise in patients with SCI, irrespective of the mechanism(s) responsible for the heart rate attenuation seen in these subjects.
It has recently been demonstrated, however, that even when able-bodied volunteers perform FES exercise with epidural anesthesia, the work rate required to elicit the same ˙VO2 (and heart rate) as during voluntary cycle exercise is much less, revealing a large inefficiency (high˙VO2 for the work rate) for FES exercise (18). This inefficiency is on the same order of magnitude as that seen in patients with SCI performing FES exercise, and is not affected by training(11,20,27). This suggests that the inefficiency of FES exercise may be more related to the manner in which the muscle motor (and fiber) unit recruitment was accomplished than to inherent differences between able-bodied and SCI subjects.
We have, in this study, been able to demonstrate improvements in gas exchange kinetics as a result of exercise training in patients with SCI performing leg cycle ergometer exercise induced by FES. The exaggerated metabolic responses to FES cycle exercise in patients with SCI, however, may be more a function of fiber recruitment profile than pathology. Peak˙VO2 is directly related to the peak heart rate achieved. Exercise training leads to increased peak ˙VO2 and work rate in most of the subjects, and the kinetics of adjustment of pulmonary gas exchange(˙VO2, ˙VCO2 and ˙VE) are speeded, both during exercise and in recovery. Finally, improvements in gas exchange kinetics occur independently of increases in peak ˙VO2, suggesting different mechanisms are involved in regulating each response. Future research is necessary to identify the mechanism(s) for the reduced heart rate response during FES exercise and also to better describe the training adaptations and the mechanisms that limit them in these patients.
1. American Spinal Injury Association. Standards for Neurological, and Functional Classification of Spinal Cord Injury
. Chicago: American Spinal Injury Association, 1992, pp. 1-26.
2. Ashley, E. A., J. J. Laskin, L. M. Olenik, et al. Evidence of autonomic dysreflexia during functional electrical stimulation in individuals with spinal cord injuries. Paraplegia
3. Barstow, T. J., N. Lamarra, and B. J. Whipp. Modulation of muscle and pulmonary oxygen uptakes by circulatory dynamics. J. Appl. Physiol.
4. Barstow, T. J., A. M. E. Scremin, D. L. Mutton, C. F. Kunkel, T. G. Cagle, and B. J. Whipp. Gas exchange kinetics during functional electrical stimulation induced leg exercise (FESILE) cycling in spinal cord injured patients. Med. Sci. Sports Exerc.
5. Belman, M. J. and G. A. Gaesser. Exercise training below and above the lactate threshold in the elderly. Med. Sci. Sports Exerc.
6. Casaburi, R., T. J. Barstow, T. Robinson, and K. Wasserman. Dynamic and steady-state ventilatory and gas exchange responses to arm exercise. Med. Sci. Sports Exerc.
7. Casaburi, R., A. Patessio, F. Ioli, S. Zanaboni, C. F. Donner, and K. Wasserman. Reductions in exercise lactic acidosis and ventilation as a result of exercise training in patients with obstructive lung disease. Am. Rev. Respir. Dis.
8. Casaburi, R., T. W. Storer, I. Ben-Dov, and K. Wasserman. Effect of endurance training on possible determinants of VO2 during heavy exercise. J. Appl. Physiol.
9. Convertino, V. A., D. J. Goldwater, and H. Sandler. VO2 kinetics of constant-load exercise following bed-rest-induced deconditioning.J. Appl. Physiol.
10. Dudley, G. A., P. C. Tullson, and R. L. Terjung. Influence of mitochondrial content on the sensitivity of respiratory control.J. Biol. Chem.
11. Faghri, P. D., R. M. Glaser, and S. F. Figoni. Functional electrical stimulation leg cycle ergometer exercise: training effects on cardiorespiratory responses of spinal cord injured subjects at rest and during submaximal exercise. Arch. Phys. Med. Rehabil.
12. Hagberg, J. M., R. C. Hickson, A. A. Ehsani, and J. O. Holloszy. Faster adjustment to and recovery from submaximal exercise in the trained state. J. Appl. Physiol.
13. Henritze, J., A. Weltman, R. L. Schurrer, and K. Barlow. Effects of training at and above the lactate threshold on the lactate threshold and maximal oxygen uptake. Eur. J. Appl. Physiol.
14. Holloszy, J. O. Biochemical adaptations in muscle: effects of exercise on mitochrondrial oxygen uptake and respiratory enzyme activity in skeletal muscle. J. Biol. Chem.
15. Hooker, S. P., S. F. Figoni, M. M. Rodgers, et al. Physiologic effects of electrical stimulation leg cycle exercise training in spinal cord injured persons. Arch. Phys. Med. Rehabil.
16. Huszczuk, A., B. J. Whipp, and K. Wasserman. A respiratory gas exchange simulator for routine calibration in metabolic studies. Eur. Respir. J.
17. Kellerman, J. The role of exercise therapy in patients with impaired ventricular function and chronic heart failure. J. Cardiovasc. Pharmacol.
18. Kjaer, M., G. Perko, N. H. Secher, et al. Cardiovascular and ventilatory responses to electrically induced cycling with complete epidural anaesthesia in humans. Acta Physiol. Scand.
19. Koike, A., K. Wasserman, K. D. McKenzie, S. Zanconato, and D. Weiler-Ravell. Evidence that diffusion limitation determines O2
uptake kinetics during exercise in man. J. Clin. Invest.
20. Krauss, J. C., R. A. Robergs, J. L. Depaepe, et al. Effects of electrical stimulation and upper body training after spinal cord injury. Med. Sci. Sports Exerc.
21. Levy, M., T. Kushnir, J. Mizrahi, and Y. Itzchak.In vivo
31P NMR studies of paraplegics' muscles activated by functional electrical stimulation. Magn. Reson. Med.
22. Mahon, A. D. and P. Vaccaro. Ventilatory threshold and VO2max
changes in children following endurance training. Med. Sci. Sports Exerc.
23. Martin, T. P., R. B. Stein, P. H. Hoeppner, and D. C. Reid. Influence of electrical stimulation on the morphological and metabolic properties of paralyzed muscle. J. Appl. Physiol.
24. Paterson, D. H. and B. J. Whipp. Asymmetries of oxygen uptake transients at the on- and offset of heavy exercise in humans. J. Physiol.
25. Petersen, E. S., B. J. Whipp, J. A. Davis, D. J. Huntsman, H. V. Brown, and K. Wasserman. Effects of B-adrenergic blockade on ventilation and gas exchange during exercise in humans. J. Appl. Physiol.
26. Petrofsky, J. S. and C. A. Phillips. Use of functional electrical stimulation for rehabilitation of spinal cord injured patients.Cent. Nerv. Syst. Trauma
27. Petrofsky, J. S. and R. Stacy. The effect of training on endurance and the cardiovascular responses of individuals with paraplegia during dynamic exercise induced by functional electrical stimulation.Eur. J. Appl. Physiol.
28. Pollack, S. F., K. Axen, N. Spielholz, N. Levin, F. Haas, and K. T. Ragnarsson. Aerobic training effects of electrically induced lower extremity exercise in spinal cord injured people. Arch. Phys. Med. Rehabil.
29. Ren, J. M., S. Broberg, and K. Sahlin. Oxygen deficit is not affected by the rate of transition from rest to submaximal exercise.J. Physiol.
30. Round, J. M., F. M. D. Barr, B. Moffat, and D. A. Jones. Fibre areas and histochemical fibre types in the quadriceps muscle of paraplegic subjects. J. Neurol. Sci.
31. Saltin, B., B. Blomqvist, J. H. Mitchell, R. L. Johnson, Jr., K. Windenthal, and C. B. Chapman. Response to submaximal and maximal exercise after bed rest and training. Circulation
32. Sietsema, K. E., D. M. Cooper, J. K. Perloff, et al. Dynamics of oxygen uptake during exercise in adults with cyanotic congenital heart disease. Circulation
33. Springer, C. S., T. J. Barstow, K. Wasserman, and D. M. Cooper. Oxygen uptake and heart rate responses during hypoxic exercise in children and adults. Med. Sci. Sports Exerc.
34. Taylor, P. N., D. J. Ewins, B. Fox, D. Grundy, and I. D. Swain. Limb blood flow, cardiac output and quadriceps muscle bulk following spinal cord injury and the effect of training for the Odstock functional electrical stimulation standing system. Paraplegia
35. Taylor, R. and N. L. Jones. The reduction by training of CO2 output during exercise. Eur. J. Cardiol.
36. Whipp, B. J. and M. Mahler. Dynamics of pulmonary gas exchange during exercise. In: Pulmonary Gas Exchange
, J. B. West(Ed.). New York: Academic Press, 1980, pp. 33-96.
37. Whipp, B. J., S. A. Ward, and D. A. Paterson. Dynamic asymmetries of ventilation and pulmonary gas exchange during on- and off-transients of heavy exercise in humans. In: Control of Breathing and Its Modeling Perspective
, Y. Honda, et al. (Eds.). New York: Plenum Press, 1992.
EXERCISE; EXERTION; REHABILITATION; OXYGEN UPTAKE; PARAPLEGIA; QUADRIPLEGIA; SPINAL CORD INJURY©1996The American College of Sports Medicine