Effect of 15% body weight support on exercise capacity of adults without impairments. (Research Report).

Over the past 2 decades, the conceptual framework guiding rehabilitation of people with neurologic conditions has undergone changes. With the emergence of the systems model of motor control, the historical focus on the neuromuscular system, in our view, has been replaced with an emphasis on how multiple systems (neuromuscular, musculoskeletal, cardiorespiratory) interact with each other and with the environment to affect functional outcomes. (1) We believe this broader perspective has led to the need for adapting evaluative techniques not traditionally used in neurologic physical therapy. One such technique, and the topic of this study, is the testing of exercise capacity (ie, cardiovascular endurance).

We believe it is becoming clear that recovery of patients with neurologic injuries cannot be attributed solely to improved neuromuscular function. Roth and colleagues (2) determined that only 2% to 36% of the variance in disability following stroke is explained by neurologic impairment. Gresham et al proposed that "much of the disability of stroke victims appears to be due to coexisting cardiovascular disease." (3)(p490) Approximately 75% of people with stroke also have cardiac disease, (4) and people in the chronic phase of recovery (at least 6 months post-stroke) have abnormally low exercise capacity. (5-7) There is some evidence of compromised cardiorespiratory fitness in people with other neurologic diagnoses, including post-polio syndrome, (8) Parkinson disease, (9) Guillain-Barre syndrome, (10) traumatic brain injury, (11,12) cerebral palsy, (13) multiple sclerosis, (14) and spinal cord injuries, (15,16)

Investigations of exercise testing of people with neurologic impairments have been restricted mainly to individuals with chronic neuromotor deficits. We contend that there is a need to develop procedures that will permit examination of patients with recently acquired disability, because it is typically this group that is actively engaged in rehabilitation. As stated by Noonan and Dean, "physical therapists ... need to assume a role in refining existing exercise tests and measures and to assume a leadership role in developing new tests and measures." (17)(p796) We believe the challenge is to design safe and efficacious methods of testing, given the high probability of motor and postural impairments in patients with neurologic injuries. Maximal oxygen consumption (V[O.sub.2]max) is generally accepted as the definitive index of exercise capacity, (18) as well as the best measure of the functional limit of the cardiovascular system. (19)

Although most investigators studying oxygen consumption (V[O.sub.2]) of people with neurologic pathology have used cycle and wheelchair ergometers, (7,20-27) we argue that treadmill walking is the testing method of choice for several reasons. To measure V[O.sub.2]max, approximately 50% of the total muscle mass must be recruited (19); this condition is much more likely to be met while walking than while cycling, particularly in a person who is deconditioned. (19) For many patients with neurologic impairments, the standing posture is used for most mobility tasks. Both measured V[O.sub.2]max values and aerobic training are task-specific, that is, specific to the exercise method used (eg, treadmill walking versus cycle ergometry) (28) or the task being done. (19,29)

Corcoran and Brengelmann (30) noted that, although the treadmill is the ideal tool for studying exercise capacity after stroke, feelings of anxiety and insecurity elicited during treadmill walking limit the use of this testing mode. However, Macko and colleagues (5) reported successful use of a self-selected, low-speed treadmill protocol to test the peak heart rate (HRpeak) of 30 patients who were, on average, 2 years post-stroke. Exercise capacity was not measured. Macko et al concluded that, although there is a need for testing in the early post-stroke period, use of their protocol may not be feasible due to physical limitations in the subacute phase.

Body weight support (BWS) systems have been developed to offset a percentage of body mass and to provide external balance support, thereby permitting treadmill walking of people in the early stages of neurologic recovery.(31,32) The typical system consists of a vest similar to a parachute harness that is attached to an overhead support. Unweighting of a prescribed amount of body mass is achieved by vertical displacement through the supporting frame using a weight and pulley or a pneumatic system. Several studies have been conducted to determine the level of unweighting needed to approximate a normal gait pattern in people with neurologic injuries. In 1989, Visintin and Barbeau (33) reported that individuals with paraparesis attained a more symmetrical gait pattern and a decrease in ankle clonus when treadmill walking with 40% BWS compared with treadmill walking with full weight bearing. More recently, Hesse and colleagues (34) observed a reduction in activation of the soleus and vastus lateralis muscles during the stance phase of gait in 11 subjects with hemiparesis when 45% to 60% BWS was used and therefore recommended an upper limit of 30% BWS for gait retraining.

In 1999, Hesse et al (35) compared treadmill walking with 0%, 15%, and 30% BWS in 18 patients with longstanding hemiparesis and found a decrease in electromyographic activity of antigravity leg muscles with increasing BWS. Consistent with this finding, Harkema et al (36) observed that the amplitude of electromyographic signals from the ankle muscles during BWS-facilitated locomotion was correlated with peak limb load (r=.57-.96). In a study by Hassid et al (37) of the effects of 15%, 30%, and 50% BWS on hemiparetic gait, 15% BWS was found to optimize the symmetry of loading, suggesting to the authors that 15% BWS provided optimal step-related sensory feedback to the locomotor networks of the brain stem and spinal cord. Miyai et al (38) pilot tested the treadmill training protocol using 0%, 10%, 20%, and 30% BWS for people with Parkinson disease and observed that subjects were most comfortable walking with 20% BWS and most uncomfortable with 30% BWS. As a result of these findings, Miyai et al used 10% and 20% BWS in the gait training protocol for these subjects.

There have been no reports describing the use of BWS to assist in the measurement of V[O.sub.2]max. In 2 reports, however, the effects of various levels of BWS on submaximal treadmill test results were documented. Mangione et al (39) applied 0%, 20%, and 40% BWS to reduce the ground reaction forces during submaximal treadmill testing of 27 people with osteoarthritis of the knee. They found an inverse relationship between the degree of unweighting and V[O.sub.2] at a given submaximal workload (eg, after 6 minutes of treadmill walking with 0%, 20%, and 40% BWS, V[O.sub.2] was 12.2 [+ or -] 2.5, 10.6 [+ or -] 2.6, and 9.3 [+ or -] 1.8 mL/kg/min, respectively). This result is consistent with evidence indicating that use of muscles at submaximal workloads is less with BWS than with full weight bearing (40) and that submaximal V[O.sub.2] levels are proportional to the muscle mass used to perform a task.(41) Danielsson and Sunnerhagen (42) found lower submaximal V[O.sub.2] levels while subjects walked on a treadmill with 30% BWS than when they walked unsupported. They studied 9 patients who were more than 6 months post-stroke and 9 subjects without disabilities.

For meaningful interpretation of the measurements of exercise capacity in people with pathology, we believe that comparison with normative reference values is desirable. Thus, we contend that V[O.sub.2]max values obtained using BWS treadmill testing should be comparable to those obtained under standard testing conditions. The purpose of our study was to investigate the effects of BWS on the exercise capacity of adults without impairments by determining whether the results obtained with 15% BWS are comparable to those obtained with full weight bearing. In addition, we wanted to determine whether the presence of the harness support alone, without BWS, would affect test results. If BWS can be used to obtain valid measurements, testing protocols could be developed for people with neurologic conditions.

We used 15% BWS for 2 principal reasons. First, exercise capacity would be a useful measure of initial and outcome status of cardiovascular function for people with moderate to mild gait and balance disturbances who are engaged in gait training and other dynamic activities that could affect cardiovascular function. Body weight support in the range of 10% to 20% has been shown to assist in safe treadmill walking for this patient group? (38,43) In a clinical trial of the effects of BWS gait training in 100 people between 1 and 5 months post-stroke, Visintin et al (43) reported that 30% to 40% BWS was required to achieve proper weight shift and weight bearing onto the hemiparetic limb during the loading phases of gait, but after 3 weeks of training the majority of the subjects were training with between 0% and 20% BWS. Second, research findings suggest that low percentages of unweighting enhance the gait pattern of people with neurologic conditions without altering the maximum amount of muscle mass used, and thus the peak V[O.sub.2] levels attained should be comparable. (35,37) Therefore, we anticipated that the V[O.sub.2]max values attained using 15% BWS would be similar to those measured in the standard test and in the test using the harness support without BWS.

Method

Subjects

Volunteers over 40 years of age were recruited from the Halifax regional municipality by word of mouth. We interviewed the subjects to screen for cardiovascular disease, musculoskeletal abnormalities, or pulmonary disease that would preclude maximal exercise. The subjects reported having various levels of physical fitness, but none had participated in physical training during the 6 months preceding the study. Potential subjects were asked to complete the revised Physical Activity Readiness Questionnaire (rPAR-Q). (44) The Physical Activity Readiness Questionnaire (PAR-Q) was developed to identify individuals for whom physical activity may be contraindicated. (45) Later, Cardinal and colleagues (46) determined that the rPAR-Q was a valid exercise screening tool for older adults. Seventeen subjects with no positive responses to the rPAR-Q questions were given a detailed explanation of the study and signed an informed consent form that was approved by the university's research ethics committee.

Two subjects, a 60-year-old man and a 61-year-old woman, did not complete the study, and we referred them to a cardiologist due to abnormal electrocardiographic responses during their first exercise test. Of the remaining 15 subjects, 10 had never smoked and 5 were former smokers (mean of 7.0 pack-years, where pack-year is the product of packs per year and years of smoking; mean years since quitting smoking = 31, SD = 16, range = 10-57). Other characteristics of the subjects are summarized in Table 1. The Physical Activity Questionnaire was used as a general measure of the physical activity level of each subject. (47) This questionnaire asks the subject to indicate the frequency and duration of participation over the past year in 11 forms of physical activity. A score for each activity is derived from the product of the length of time of participation per session (in hours), the number of sessions per week, and the number of seasons of participation per year. A total activity score is the sum of the individual scores and can be categorized as: score>18 = very active, 1-18 = active, and 0 = inactive.

Study Protocol

All testing took place in the exercise testing laboratory of a tertiary health care facility. Each subject participated in a familiarization session and performed 3 maximal treadmill exercise tests, with an interval of 48 hours between tests. The sequence of experimental conditions was randomized for each subject. The conditions were: (1) no BWS--standard exercise test, (2) 0% BWS--harness support with no body mass displaced, and (3) 15% BWS--harness support with 15% of body mass displaced.

Familiarization Session

One week prior to the initial exercise test, each subject visited the laboratory to become familiar with the testing equipment, the exercise protocol, and the unweighting procedure and to practice breathing with the respiratory mouthpiece, headgear, and noseclip in place. Each subject was fitted with a harness. The subjects reported feeling comfortable walking on the treadmill with and without BWS after less than 5 minutes of practice, consistent with a previous report that only 1 to 2 minutes is required for most individuals to become accustomed to treadmill walking. (48)

Exercise Testing Protocol

Each subject did 3 graded exercise tests. The graded exercise tests were conducted at the same time of the day in a temperature-controlled laboratory with the temperature maintained at 22 [degrees] [+ or -] 2 [degrees] C and a relative humidity of 45% to 60%. Subjects were requested to avoid food and smoking for at least 2 hours before testing, to refrain from drinking caffeinated beverages for at least 6 hours, and avoid heavy exertion or exercise for 12 hours. A progressive exercise test was conducted using a calibrated motorized treadmill (Quinton model 18-60 *) in accordance with American College of Sports Medicine (ACSM) criteria. (49) The Naughton-Balke protocol (2.5% grade increment/2 minutes at a constant speed of 1.3 m/s) was used for all tests. Testing was preceded by a 3-minute warm-up at level grade using a speed of 0.9 m/s and was followed by a 2-minute cool-down at level grade and a speed of 1.0 m/s. Subjects were requested to avoid using the handrails of the treadmill for support because such support can increase the total treadmill time and reduce submaximal values of V[O.sub.2]. (50) Termination of testing was done in accordance with ACSM guidelines. (49) Subjects were instructed to use the "thumbs down" signal to indicate their desire to terminate the test. Our criteria for subjects reaching a maximal effort were attainment of at least 2 of the following: (1) increase in V[O.sub.2] of less than 100 mL in the final minute of exercise (V[O.sub.2] plateau), (2) maximal heart rate (HRmax) within 10 bpm of age-predicted HRmax (220 -- age), and (3) peak respiratory exchange ratio (RERpeak) of greater than 1.10. (51)

Expired gas was analyzed using open-circuit spirometry using a SensorMedics 2900 metabolic measurement cart ([dagger]) to determine V[O.sub.2], carbon dioxide production (VC[O.sub.2]), minute ventilation (VE), respiratory exchange ratio (PER), and tidal volume (VT). Expired volumes were passed through a 3-1 mixing chamber where the percentages of [O.sub.2] and C[O.sub.2] were analyzed by a mass spectrometer (accuracy of [+ or -] 0.02%). Calibration of the analyzer was done using standard gases (26% [O.sub.2]/74% [N.sub.2] and 16% [O.sub.2]/4% C[O.sub.2]/80% [N.sub.2]) and verified before each test. Calibration of the volume-measuring system was done using a 3-L syringe prior to each test. Subjects wore a noseclip and breathed room air through a one-way directional valve system attached to a mouthpiece. Maximal values for exercise variables were averaged over the last 30 seconds.

Electrical activity of the heart was monitored using a 10-lead electrocardiogram. Skin sites were abraded with fine sandpaper and cleaned with alcohol to remove surface epidermis and oil in an effort to minimize impedance. To ensure good contact, each electrode was tapped vigorously after placement while monitoring the corresponding lead on the oscilloscope. In addition, each lead had some slack, and this was checked prior to application of the BWS harness to avoid undue tension on the leads during the exercise tests. Heart rate was obtained from the electrocardiograph recording. Resting heart rate (HRrest) was determined after the subject had rested for 10 minutes while seated in a chair placed on the treadmill belt. This measurement was taken just prior to the exercise test with the respiratory mouthpiece, headgear, and noseclip in place. The HRmax was the average HR during the last 30 seconds of exercise. Maximal oxygen pulse ([O.sub.2] pulsemax) was calculated using the formula: [O.sub.2] pulsemax = V[O.sub.2]max/HRmax. Right brachial artery systolic blood pressure (SBP) and diastolic blood pressure (DBP) was measured using a calibrated mercury sphygmomanometer (Baumanometer ([double dagger])). Resting SBP and DBP were measured subsequent to determining HRrest. Blood pressure was measured every 2 minutes during exercise testing and every minute during recovery until it returned to baseline. Maximal rate-pressure product (RPPmax), an index of myocardial oxygen consumption, (52) was calculated as the product of HRmax and peak SBP divided by 100.

In the 0% and 15% BWS conditions, the Pneuweight Unweighting System, ([sections]) consisting of a supporting frame, a compressor, and a 0.7-kg harness, was used. The experimental setup is shown in Figure 1. According to the manufacturer, the overhead frame provides vertical displacement of a prescribed amount of weight using pneumatic pressure generated by the compressor. The Pneuweight Unweighting System accommodates the 5-cm vertical displacement of the center of gravity that occurs in the normal gait cycle, thus, in theory, permitting a normal gait pattern. The chest strap of the harness is placed around the subject's torso at the level of the xiphoid process, and leg straps are placed around each upper thigh for additional support. The harness is then attached to the frame by 2 clips. In the 15% BWS condition, the unweighting dial was set to allow displacement of 15% of body mass. Periodically, the unweighting mechanism was checked by placing a 5-kg weight, which was attached to the overhead support, on the treadmill and adjusting the unweighting dial until the weight was lifted just off the surface. This procedure was repeated using a 10-kg weight. During these trials, recalibration of the unweighting mechanism was not required because the actual extent of unweighting coincided with that registered by the unweighting dial.

Exercise time (ie, the time from the initiation to termination of the exercise protocol excluding the warm-up and cool-down) was recorded. Predicted maximal oxygen consumption (V[O.sub.2]max [pred]) was calculated in milliliters per kilogram per minute using the ACSM formula for treadmill walking: V[O.sub.2]max (pred)= [(3.5 mL/kg/min) + (m/min X 0.1) + (grade [fraction] X m/min X 1.8)]. (49) Measured V[O.sub.2]max was expressed as a percentage of V[O.sub.2]max (pred). Exertion on the treadmill was recorded at the end of each stage of exercise and at peak exercise using the modified Rating of Perceived Exertion (RPE) on a scale of 0 ("nothing at all") to 10 ("very, very strong"). (53) Subjects were asked to point with their index finger to the number indicative of their RPE. In addition, at the conclusion of each testing session, each subject was asked to identify the reason for termination of the test and to respond to the open-ended question: "How did you feel during and after the exercise test?" At the end of the final session, the subjects were also asked: "Please comment on the harness support system. Did you prefer to do the exercise test with or without the support?" The responses were documented verbatim.

Data Analysis

One-way analysis of variance (ANOVA) for repeated measures using the within-subject factor of experimental condition and Bonferroni post hoc testing were applied to detect differences in the dependent variables across the 3 experimental conditions. To ascertain the potentially confounding effect of order of the experimental conditions, the ANOVAs of all dependent variables were repeated using testing order as a between-subject factor. All statistical tests were performed with the alpha level set at .05.

Results

The order of testing conditions did not affect the results of the exercise tests. The requirements for the designation of a "maximal effort" were met with all tests, and in 27 of the 45 tests, all 3 criteria were achieved. The HRmax criterion was achieved by all subjects, but 3 subjects (two 61-year-old men and one 71-year-old man) were unable to attain an RERpeak greater than 1.10 and 3 subjects (one 43-year-old woman, one 46-year-old man, and one 56-year-old man) did not attain a V[O.sub.2] plateau during the last minute of exercise. In all tests, the subjects voluntarily requested termination of the tests. In the no BWS and 0% BWS conditions, the reason for termination was consistent within individual subjects, 6 subjects offering dyspnea as the main reason and the remaining 9 subjects claiming general fatigue. However, in the 15% BWS condition, 4 subjects who claimed general fatigue stopped due to dyspnea, yielding a total of 10 subjects in this experimental condition whose reason for termination was respiratory difficulty.

The values for V[O.sub.2]max, exercise time, and HRmax achieved by each of the 15 subjects for each testing condition are illustrated in Figure 2. Visual observation of these graphs suggests consistent trends in the data--that neither V[O.sub.2]max nor HRmax appeared to be affected by the testing condition but exercise time was longer in the 15% BWS condition. The results of the inferential analyses confirmed these trends. Relative and absolute V[O.sub.2]max, maximal carbon dioxide production (VC[O.sub.2]max), RERpeak, HRmax, [O.sub.2] pulsemax, RPPmax, peak minute ventilation (VEpeak), peak tidal volume (VTpeak), peak respiratory rate (RRpeak), and peak modified Rating of Perceived Exertion (RPEpeak) were not influenced by the testing condition (Tab. 2). However, exercise time was longer in the 15% BWS condition than in the either the no BWS condition or the 0% BWS condition by averages of 12.4% and 13.8%, respectively. Peak tidal volume was lower in the 15% BWS condition than in the no BWS condition. In addition, the percentages of predicted V[O.sub.2]max were lower in the 15% BWS condition than in the other 2 conditions. In the no BWS condition, the mean V[O.sub.2]max was 32.9 mL/kg/min (SD=7.4) for the female subjects and 37.6 mL/kg/min (SD=5.0) for the male subjects.

[FIGURE 2 OMITTED]

For comparative purposes, V[O.sub.2] values during treadmill testing across the testing conditions were normalized by expressing V[O.sub.2] values as percentages of the highest value of V[O.sub.2]max obtained during the standard no BWS test. Similarly, exercise time was normalized by expressing it as a percentage of the exercise time recorded during the no BWS test (Fig. 3). As shown in Figure 3, the curves for the 3 testing conditions were virtually superimposed until 25% of the total exercise time had expired, at which point the curve for the 15% BWS condition shifted to the right.

[FIGURE 3 OMITTED]

All subjects reported that the harness was comfortable to wear. Nine subjects, however, preferred testing without BWS because the harness was perceived to be mildly restrictive of respiration toward the end of the test. This group included the 4 subjects whose reason for terminating the exercise test changed from fatigue in the no BWS and 0% BWS conditions to dyspnea in the 15% BWS condition. In addition, the VTpeak values for these 4 subjects were lowest in the 15% BWS condition, and VEpeak and V[O.sub.2]max values were also lowest in that condition for 2 female subjects (aged 48.7 and 60.5 years). The remaining 6 subjects preferred using the harness because of the security it provided during testing, 2 of whom stated that the harness also provided back support, which they felt enhanced their performance. However, their test results did not reflect this observation.

Discussion

Interpretation of data obtained from maximal exercise tests is dependent on subjects making a maximal effort. Although every subject in our study met the required 2 of 3 criteria for designation of V[O.sub.2]max in all tests, only 9 subjects (60%) met all 3 criteria. Despite the recommendation of Londeree and Moeschberger (54) that HRmax not be used as an absolute criterion of V[O.sub.2]max due to the high intersubject variability of HRmax (SD = 11 bpm), all subjects met this criterion. Although it is not uncommon for subjects to fail to demonstrate a plateau in V[O.sub.2], (51) 12 (80%) of the subjects met this criterion. That 3 subjects over 60 years of age did not achieve an RERpeak greater than 1.10 is consistent with previously reported reductions in RERpeak values at maximal effort in people over 60 years of age. (55)

We found further assurance that the tests represented maximal efforts in that the V[O.sub.2]max values that we observed were comparable to those observed in previous studies. Recently, Jackson and colleagues (56) conducted maximal exercise treadmill tests with 160 men without impairments over the age of 55 years ([bar]X = 58.0, SD = 3.0) and reported a mean V[O.sub.2]max of 33.2 mL/kg/min (SD=6.0). In addition, the extent of variability in the V[O.sub.2]max measurements can be compared using the coefficient of variation, a dimensionless number expressing the standard deviation as a proportion of the mean. (57) The coefficients of variation for V[O.sub.2]max were 18.5% in our study and 18.1% in the study by Jackson et al. (56) The intrasubject differences in V[O.sub.2]max values on repeated testing under the 3 testing conditions varied, on average, by 2%. This variability is within the reported 2% to 4% of variability for repeated measurements of V[O.sub.2]max among people without impairments. (41)

As we anticipated, neither the presence of the harness support in the 0% BWS condition nor the harness plus unweighting in the 15% BWS condition affected the endpoint values of the main respiratory gas exchange variables measured during exercise testing. At any given submaximal stage of exercise, the energy expenditure and cardiorespiratory responses were less in the unweighted condition. We believe this was because V[O.sub.2] is proportional to the muscle mass recruited; therefore, a longer exercise time was required to elicit a maximal response in the 15% BWS condition. Because V[O.sub.2]max levels attained were not different from those of the standard test, we assume that the application of 15% BWS did not alter the total muscle mass recruited at peak effort below the threshold required to attain true V[O.sub.2]max values. This assumption, in our view, is consistent with previous investigations in which the researchers concluded that low percentages of unweighting do not alter the peak muscle mass recruited. (34,36)

The effects of 15% BWS observed in our study are analogous to the results of investigations of the effect of handrail support on treadmill exercise testing. Measured V[O.sub.2]max is not different with or without handrail support, but submaximal V[O.sub.2] levels are reduced and total exercise time is increased when handrail support is permitted. (58,59) The reduction in aerobic demands of walking with handrail support results in a shift of the exercise time/V[O.sub.2] curve to the right, (58) similar to the right shift observed in our study in the 15% BWS test (Fig. 3). Thus, there appears to be a common trend in the effects of unweighting of 15% of body mass and the use of handrail support on the response to treadmill exercise testing.

The ACSM formula for estimating V[O.sub.2]max values during treadmill walking in the absence of direct measurement was derived by relating mechanical measures of work rate (ie, treadmill speed and elevation) and their metabolic equivalents. (49) Application of this formula resulted in overestimates of the V[O.sub.2]max achieved, which also occurred in a previous study (18) where steady-state requirements of the last treadmill stage attained almost always overestimated the achieved V[O.sub.2]max. The substantial overestimation in the 15% BWS test (134%) was anticipated because peak treadmill speed and elevation, variables used in the formula,' were greatest in the unweighted condition. The ACSM formula, in our opinion, is inappropriate for the test conditions in our study. A revision to the ACSM formula has recently been suggested for exercise protocols with relatively small workload increments between stages, (18) such as the Naughton-Balke protocol used in our study. Use of this formula in the standard no BWS test reduced the overestimation of V[O.sub.2]max from 119% to 103%. Interestingly, application of another formula that Foster et al (18) adapted for use when handrail support was permitted during testing decreased the overestimate of V[O.sub.2]max for the 15% BWS test from 134% to 103%. This finding corroborates the above-stated suggestion that the testing conditions imposed by 15% BWS test appear to be somewhat comparable to those encountered with a standard maximal treadmill test allowing handrail support.

Values for the respiratory variables VEpeak, VTpeak, and RRpeak attained in the standard no BWS test are consistent with those of previous reports. (60) The finding of lower VTpeak values in the 15% BWS condition than in the no BWS and 0% BWS conditions suggests that restriction in chest wall excursion is due to a combination of the circumferential pressure exerted on the thorax by the chest strap of the harness and the upward force imposed by the vertical displacement of body mass.

This reduction in VTpeak without concomitant changes in the VEpeak implies a compensatory increase in respiration rate. Two of the subjects who preferred the no BWS condition recorded their lowest values for V[O.sub.2]max, VTpeak, and VEpeak in the 15% BWS condition and terminated that testing session because of dyspnea. Although definitive conclusions cannot be drawn from data on 2 subjects, this suggests that caution may need to be exercised in the use of BWS for individuals with compromised respiratory function. The effect of the particular harness used in this study on the respiratory variables is unknown. Harnesses that have a pelvic strap rather than a chest strap may be less restrictive. Further investigation of the combined effects of unweighting and harness design on respiratory function is warranted.

The exercise testing protocol and application of the BWS harness were well-tolerated by the subjects, and unweighting of 15% of body mass did not affect the endpoint values of the principal respiratory gas exchange variables. Thus, comparability of the results to normative reference values obtained under the standard, full weight-bearing testing condition was preserved. These findings support the use of 15% BWS to assist in treadmill exercise testing of people whose neuromuscular limitations preclude standard exercise testing. However, a less challenging testing procedure than the one used in this study would be required for people with neurologic impairments because the treadmill speed and grade increments of the Naughton-Balke protocol would be too demanding. A more appropriate protocol might be an individualized, low-speed method such as that used by Macko and colleagues (5) to measure HRpeak of patients in the chronic, post-stroke period. These investigators studied the use of an initial treadmill tolerance test without incline to identify the target speed for subsequent maximal-effort graded treadmill testing. They also recognized the need for testing early post-stroke but questioned the feasibility of using their protocol, given the physical limitations of people in the subacute phase. A study of the safety and efficacy of their customized treadmill protocol, using 15% BWS to provide external support and to assist in walking, in measuring exercise capacity early after stroke is needed.

Limitations

The protocol used in our study was appropriate for measuring exercise capacity of adults without impairments but would need to be adapted, in terms of treadmill speed and grade, for use with people who have neurologic impairments. Our conclusions may have limited application because of the small sample size. Furthermore, because the subjects were relatively inexperienced with treadmill walking, our results may have been affected by learning. However, we believe the possibility of a learning effect is unlikely because the order of testing did not affect the results. In addition, our results findings regarding respiratory variables are limited because only one of several harness designs available was used in this study.

Conclusions

Unweighting of 15% of body mass had no effect on the endpoint values of the principal respiratory gas exchange variables measured during V[O.sub.2]max treadmill testing of 15 adults without neurologic impairments over the age of 40 years. Further development of BWS-assisted testing protocols is warranted for measuring energy expenditure and exercise capacity of people with neurologic impairments whose neuromuscular limitations preclude standard exercise testing. Our findings in this preliminary study of comparable respiratory endpoints attained with or without use of 15% BWS are important because they allow limited comparisons of test results obtained for people with neurologic impairments. Reduction in VTpeak values with unweighting suggests that caution may need to be exercised when using BWS with individuals who manifest compromised respiratory function. Further study of the effects of varying percentages of unweighting and various-harness designs on respiratory gas exchange variables of people with and without pathology would extend the clinical usefulness of this technique.

Table 1.
Subject Characteristics

                Age (y)                 Height (cm)

Subjects   N    [bar]X   SD     Range   [bar]X   SD    Range

Female      8   55.1     13.6   43-82   166.0    6.8   159-175
Male        7   56.4      9.4   46-72   177.2    6.8   168-183
Total      15   55.2     11.3           171.4    8.6

           Mass (kg)               BMI (a)

Subjects   [bar]X   SD     Range   [bar]X   SD    Range

Female     70.5     11.2   60-92   25.6     4.5   22-36
Male       74.9     11.3   58-93   24.0     3.1   20-29
Total      72.6     11.1           24.8     3.9

           PAQ (b)

Subjects   [bar]X   SD     Range

Female     29.4      7.4    6-45
Male       25.6     10.5   10-36
Total      27.6     14.2

(a) BMI=body mass index (mass/[height.sup.2]).
(b) PAQ=Physical Activity Questionnaire. (47)
Table 2.
Physiologic Variables at Peak Exercise Intensity
Testing Conditions (N = 15) (a)

                               Test

                               No BWS

Variable                       [bar]X     SD      Range

V[O.sub.2] max (mL/kg/min)       35.1     6.6     24-45
V[O.sub.2] max (% predicted)     83.8     5.0     82-157
V[O.sub.2] max (L/min)            2.53    0.54   1.4-3.3
V[CO.sub.2] max (L/min)          40.1     7.4     27-55
RERpeak                           1.15    0.27   1.0-1.3
HRmax (bpm)                     177.8     3.2    163-211
[O.sub.2] pulsemax (mL/beat)     14.3     3.1      8-18
RPPmax                          338.5    53.4    238-407
VEpeak (L/min)                   88.4    19.4     49-123
VTpeak (L)                        2.57    0.54   1.9-3.9
RRpeak (breaths/min)             34.4     8.1     24-43
Exercise time (min)              15.9     3.0     10-21
RPEpeak (0-10)                    7.9     1.2      6-9

                               Test

                               0% BWS

Variable                       [bar]X     SD       Range

V[O.sub.2] max (mL/kg/min)       34.9     6.2     24-45
V[O.sub.2] max (% predicted)     83.4     4.7     80-156
V[O.sub.2] max (L/min)            2.52    0.54   1.4-3.2
V[CO.sub.2] max (L/min)          39.9     7.0     27-52
RERpeak                           1.15    0.31   1.0-1.3
HRmax (bpm)                     175.9    32.2    160-213
[O.sub.2] pulsemax (mL/beat)     14.4     3.1      8-19
RPPmax                          334.0    38.3    290-398
VEpeak (L/min)                   89.8    20.5     49-127
VTpeak (L)                        2.51    0.50   1.7-3.7
RRpeak (breaths/min)             35.8     8.5     25-44
Exercise time (min)              16.1     3.0     10-21
RPEpeak (0-10)                    7.5     1.5      6-10

                               Test

                               15% BWS

Variable                       [bar]X     SD          Range

V[O.sub.2] max (mL/kg/min)       34.4     6.2         23-44
V[O.sub.2] max (% predicted)     74.4     9.3 (b)     89-154
V[O.sub.2] max (L/min)            2.49    0.54       1.4-3.3
V[CO.sub.2] max (L/min)          39.1     7.7         25-56
RERpeak                           1.14    0.23       1.0-1.3
HRmax (bpm)                     175.6     3.0        160-207
[O.sub.2] pulsemax (mL/beat)     14.2     3.0          8-18
RPPmax                          339.7    48.4        285-411
VEpeak (L/min)                   85.8    20.9         46-111
VTpeak (L)                        2.36    0.46 (c)   1.8-3.4
RRpeak (breaths/min)             35.4     8.5         20-50
Exercise time (min)              18.1     3.7 (d)     12-26
RPEpeak (0-10)                    7.3     2.2          6-9

(a) V[O.sub.2] max=maximal oxygen consumption, V[CO.sub.2] max=maximal
carbon dioxide production, RERpeak=peak respiratory exchange ratio,
HRmax=maximal heart rate [O.sub.2] pulsemax=maximal oxygen pulse,
RPPmax=maximal rate-pressure product, VEpeak=peak minute ventilation,
Vzpeak=peak tidal volume, RRpeak=peak respiratory rate, RPEpeak=peak
modified rating of Perceived Exertion.

(b) P<.01, F test, df=14,2; denotes significant differences from no
BWS test and between 0% BWS and 15% BWS tests.

(c) P<.001, F test, df=14,2; denotes significant differences from no
BWS test.

(d) P<.0001, F test, df=14,2; denotes significant differences from no
BWS test and between 0% BWS and 15% BWS tests.

* Quinton Fitness Equipment, 3303 Monte Villa Pkwy, Bothell, WA 98021-8906.

([dagger]) SensorMedics, 22705 Savi Ranch Pkwy, Yorba Linda, CA 92687.

([double dagger]) WA Baum Co Inc, 620 Oak St. Copaigue, NY 11726.

([sections]) Pneumex Inc, 804 Airport Way., Sandpoint, ID 83864.

References

(1) Horak FB. Assumptions underlying motor control for neurologic rehabilitation. In: Lister MJ, ed. Contemporary Management of Motor Control Problems: Proceedings of the II STEP Conference. Alexandria, Va: Foundation for Physical Therapy 1992:11-27.

(2) Roth EJ, Heinemann AW, Lovell LL, et al. Impairment and disability: their relation during stroke rehabilitation. Arch Phys Med Rehabil. 1998;79:329-335.

(3) Gresham GE, Phillips TF, Wolf PA, et al. Epidemiologic profile of long-term stroke disability: the Framingham study. Arch Phys Med Rehabil. 1979;60:487-491.

(4) Roth E. Heart disease in patients with stroke, part 1: classification and prevalence. Arch Phys Med Rehabil. 1993;74:752-760.

(5) Macko RF, Katzel LI, Yataco A, et al. Low-velocity graded treadmill stress testing in hemiparetic stroke patients. Stroke. 1997;28:988-992.

(6) Potempa K, Braun LT, Tinknell T, Popovich J. Benefits of aerobic exercise after stroke. Sports Med. 1996;21:337-346.

(7) Bachynski-Cole M, Cumming GR. The cardiovascular fitness of disabled patients attending occupational therapy. Occup Ther J Res. 1985;5:233-242.

(8) Dean E, Ross J. Modified aerobic walking program: effect on patients with postpolio syndrome symptoms. Arch Phys Med Rehabil. 1988;69: 1033-1038.

(9) Proms EJ, Stanley RK, Jankovic J, MacNeill B. Cardiovascular and metabolic responses of upper- and lower-extremity exercise in men with Parkinson's disease. Phys Ther. 1996;76:34-40.

(10) Pitetti KH, Barrett PJ, Abbas D. Endurance exercise training in Guillain-Barre syndrome. Arch Phys Med Rehabil. 1993;74:761-765.

(11) Kinney LaPier TL, Sirotnak N, Alexander K. Aerobic exercise for a patient with chronic multisystem impairments. Phys Ther. 1998;78: 417-424.

(12) Jankowski LW, Sullivan SJ. Aerobic and neuromuscular training: effect on the capacity, efficiency, and fatigability of patients with traumatic brain injuries. Arch Phys Med Rehabil. 1990;71:500-504.

(13) Tobimatsu Y, Nakamura R, Kusano S, Iwasaki Y. Cardiorespiratory endurance in people with cerebral palsy measured using an arm ergometer. Arch Phys Med Rehabil. 1998;79:991-993.

(14) Petajan JH, Gappmaier E, White AT, et al. Impact of aerobic training on fitness and quality of life in multiple sclerosis. Ann Neurol. 1996;39:432-441.

(15) Pare G, Noreau L, Simard C. Prediction of maximal aerobic power from a submaximal exercise test performed by paraplegics on a wheelchair ergometer. Paraplegia. 1993;31:584-592.

(16) Fletcher BJ, Dunbar SB, Felner JM. Exercise testing and training of physically disabled men with clinical evidence of coronary artery disease. Am J Cardiol. 1994;73:170-174.

(17) Noonan V, Dean E. Submaximal exercise testing: clinical application and interpretation. Phys Ther. 2000;80:782-807.

(18) Foster C, Crowe AJ, Daines E, et al. Predicting functional capacity during treadmill testing independent of exercise protocol. Med Sci Sports Exerc. 1996;28:752-756.

(19) Rowell LB. Human cardiovascular adjustments to exercise and thermal stress. Physiol Rev. 1974;54:75-159.

(20) Brinkmann JR, Hoskins TA. Physical conditioning and altered self-concept in rehabilitated hemiplegic patients. Phys Ther. 1979;59: 859-865.

(21) Hjeltnes N. Capacity for physical work and training after spinal injuries and strokes. Scand J Soc Med. 1982;29(suppl):245-251.

(22) Monga TN, Deforge DA, Williams J, Wolfe LA. Cardiovascular responses to acute exercise in patients with cerebrovascular accidents. Arch Phys Med Rehabil. 1988;69:937-940.

(23) Hoskins TA. Physiologic responses to known exercise loads in hemiparetic patients [abstract]. Arch Phys Med Rehabil. 1975;56:544.

(24) Moldover JR, Daum MC, Downey JA. Cardiac stress testing of hemiparetic patients with a spine bicycle ergometer: preliminary study. Arch Phys Med Rehabil. 1984;65:470-473.

(25) King ML, Gurracini M, Lennihan L, et al. Adaptive exercise testing for patients with hemiparesis. J Cardiopulm Rehabil. 1989;9:237-242.

(26) Potempa K, Lopez M, Braun LT, et al. Physiological outcomes of aerobic exercise training in hemiparetic stroke patients. Stroke. 1995; 26:101-105.

(27) Bjuro T, Fugl-Meyer AR, Grimby G, et al. Ergonomic studies of standardized domestic work in patients with neuromuscular handicap. Scand J Rehabil Med. 1975;7:106-113.

(28) Holmer I, Astrand PO. Swimming training and maximal oxygen uptake. J Appl Physiol. 1972;33:510-513.

(29) McArdle WD, Margel JR, Delio DJ, et al. Specificity of run training on V[O.sub.2]max and heart rate changes during running and swimming. Med Sci Sports. 1978;10:16-20.

(30) Corcoran PJ, Brengelmann GL. Oxygen uptake in normal and handicapped subjects in relation to speed of walking beside a velocity-controlled cart. Arch Phys Med Rehabil. 1970;51:78-87.

(31) Visintin M, Barbeau H. The effects of body weight support on the locomotor pattern of spastic paretic patients. Can J Neurol Sci. 1989;16: 315-325.

(32) Wernig A, Muller S. Laufband locomotion with body weight support improved walking in persons with severe spinal cord injuries. Paraplegia. 1992;30:229-38.

(33) Visintin M, Barbeau H. The effects of parallel bars, body weight support and speed on the modulation of the locomotor pattern of spastic paretic gait: a preliminary communication. Paraplegia. 1994;32: 540-53.

(34) Hesse S, Helm B, Krajnik J, et al. Treadmill training with partial body weight support: influence of body weight release on the gait of hemiparetic patients. J Neurol Rehabil. 1997;11:15-20.

(35) Hesse S, Konrad M, Uhlenbrock D. Treadmill walking with partial body weight support versus floor walking in hemiparetic subjects. Arch Phys Med Rehabil. 1999;80:421-427.

(36) Harkema SJ, Hurley SL, Patel UK, et al. Human lumbosacral spinal cord interprets loading during stepping. J Neurophysiol. 1997;77: 797-811.

(37) Hassid E, Rose D, Commisarow J, et al. Improved gait symmetry in hemiparetic stroke patients induced during body weight-supported treadmill stepping. J Neurol Rehabil. 1997;11:21-26.

(38) Miyai IM, Fujimoto Y, Ueda Y, et al. Treadmill training with body weight support: its effect on Parkinson's disease. Arch Phys Med Rehabil. 2000;81:849-852.

(39) Mangione KK, Axen K, Haas F. Mechanical unweighting effects on treadmill exercise and pain in elderly people with osteoarthritis of the knee. Phys Ther. 1996;76:387-394.

(40) Finch L, Barbeau H, Arsenault B. Influence of body weight support on normal human gait: development of a gait retraining strategy. Phys Ther. 1991;71:842-856.

(41) Taylor HL, Buskirk E, Henschel A. Maximal oxygen uptake as an objective measure of cardiorespiratory performance. J Appl Physiol. 1955;8:73-80.

(42) Danielsson A, Sunnerhagen KS. Oxygen consumption during treadmill walking with and without body weight support in patients with hemiparesis after stroke and in healthy subjects. Arch Phys Med Rehabil. 2000;81:953-957.

(43) Visintin M, Barbeau H, Korner-Bitensky N, Mayo N. A new approach to retrain gait in stroke patients through body weight support and treadmill stimulation. Stroke. 1998;29:1122-1128.

(44) Thomas HL, Reading J, Shepherd RJ. Revision of the Physical Activity Readiness Questionnaire (PARQ). Can J Sports Sci. 1992;17: 73-80.

(45) Chishom DM, Collis ML, Kulak LL, et al. Physical activity readiness. Br Col Med J. 1975;17:375-378.

(46) Cardinal BJ, Esters L, Cardinal MK. Evaluation of the Revised Physical Activity Readiness Questionnaire in older adults. Med Sci Sports Exer. 1996;28:468-472.

(47) Jagal SB, Krieger N, Darlington G. Past and recent physical activity and risk of hip fracture. Am J Epidemiol. 1993;138:107-118.

(48) Ebbeling CB, Ward A, Puleo EM, et al. Development of a single-stage submaximal treadmill walking test. Med Sci Sports Exerc. 1991;23: 966-973.

(49) American College of Sports Medicine. Guidelines for Exercise Testing and Prescription. 5th ed. Baltimore, Md: Williams and Wilkins; 1995.

(50) Manfree MJ, Yu GH, Varma AA, et al. The effect of limited handrail support on total treadmill time and prediction of V[O.sub.2]max. Clin Cardiol. 1994;17:445-450.

(51) Howley ET, Bassett DR Jr, Welch HG. Criteria for maximal oxygen uptake: review and commentary. Med Sci Sports Exerc. 1995;27: 1292-1301.

(52) Kitamura K, Jorgensen CR, Gobel FL, et al. Hemodynamic correlates of myocardial oxygen consumption during upright exercise. J Appl Physiol. 1972;32:516-522.

(53) Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982; 14:377-381.

(54) Londeree BR, Moeschberger ML. Influence of age and other factors on maximal heart rate. J Cardiopulm Rehabil. 1984;4:44-49.

(55) Sidney KH, Shephard RJ. Maximum and submaximum exercise tests in men and women in the seventh, eighth, and ninth decades of life. J Appl Physiol. 1977;43:280-287.

(56) Jackson AS, Beard EF, Wier LT, et al. Changes in aerobic power of men, ages 25-70 yr. Med Sci Sports Exerc. 1995;27:113-120.

(57) Afifi AA, Azen SP. Statistical Analysis. New York, NY: Academic Press Inc; 1979.

(58) McConnell TR, Foster C, Conlin NC, Thompson NN. Prediction of functional capacity during treadmill testing: effect of handrail support. J Cardiopulm Rehabil. 1991; 11:255-260.

(59) Myers J, Buchanan N, Walsh D, et al. Comparison of the ramp versus standard exercise protocols. J Am Coll Cardiol. 1991;17: 1334-1342.

(60) Blackie SP, Fairbarn MS, McElvaney NG, et al. Normal values and ranges for ventilation and breathing pattern at maximal exercise. Chest. 1991;100:136-142.

M MacKay-Lyons, PT, PhD, is Assistant Professor, School of Physiotherapy, Dalhousie University, 5869 University Ave, Halifax, Nova Scotia, Canada, B3H 3J5 ([e-mail]). Address all correspondence to Dr MacKay-Lyons.

L Makrides, PhD, is Professor and Director, School of Physiotherapy, Dalhousie University, and Director, Atlantic Health and Wellness Institute, Halifax, Nova Scotia, Canada.

S Speth, MSc (Kin), is Exercise Physiologist, Atlantic Health and Wellness Institute, Halifax, Nova Scotia, Canada.

Dr MacKay-Lyons and Dr Makrides provided concept/research design, writing, project management, and consultation (including review of manuscript before submission). Dr MacKay-Lyons and Ms Speth provided subjects and dam collection and analysis. Dr Makrides provided fund procurement, facilities/equipment, institutional liaisons, and clerical support.

This study was approved by the Research Ethics Committee of Dalhousie University.

This study was supported by the Hazel Lloyd Foundation and by a Doctoral Fellowship in Applied Cardiovascular and Cerebrovascular Health Research granted to Dr MacKay-Lyons by the Heart and Stroke Foundation of Canada.

An abstract of this paper was presented at the 45th Annual Meeting of the American College of Sports Medicine, June 8, 1998, Orlando, Fla.

This article was submitted November 23, 1999, and was accepted April 12, 2001.