| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Reports |
ME Mendelsohn, MSc, PhD, is a postdoctoral fellow at the Toronto Rehabilitation Institute, Toronto, Ontario, Canada
DM Connelly, BScPT, PhD, is Assistant Professor, School of Physical Therapy, The University of Western Ontario, Room 1588, Elborn College, 1201 Western Rd, London, Ontario, Canada N6G 1H1, and Associate Scientist, Lawson Health Research Institute, St Joseph's Health Care London, London, Ontario, Canada
TJ Overend, BScPT, PhD, is Associate Professor, School of Physical Therapy, The University of Western Ontario, and Associate Scientist, Lawson Health Research Institute, St Joseph's Health Care London
RJ Petrella, MD, PhD, is Professor, Departments of Family Medicine, Medicine (Division of Cardiology), Physical Medicine, and Rehabilitation, Schulich School of Medicine, and School of Kinesiology, The University of Western Ontario, and Beryl and Richard Ivey Research Chair, Aging, Rehabilitation, and Geriatric Care, Lawson Health Research Institute, St Joseph's Health Care London
Address all correspondence to Dr Connelly at: dconnell{at}uwo.ca
Submitted May 31, 2007;
Accepted February 3, 2008
 | Abstract |
|---|
Subjects and Methods: The study participants were 18 older adults (mean age=82 years, SD=5; 3 women, 15 men) who were living in long-term care centers and who completed 2 test sessions of a standardized exercise protocol 1 week apart. The exercise protocol included a warm-up period, three 4-minute stages of exercise at incremental workload levels, and a cool-down period. The breath-by-breath metabolic data from the portable metabolic unit, heart rate, MET values from the exercise machine, Borg Rating of Perceived Exertion, and watts were recorded continuously throughout the exercise protocol.
Results: The concurrent validity of the MET values from the exercise machine and the portable metabolic unit ranged from very good to excellent on both day 1 and day 2 (r=.85–.97). The test-retest reliability of subjects' heart rate responses and MET values from the portable metabolic unit was moderate to high across submaximal exercise intensities (intraclass correlation coefficients [2,1]=.85–.91).
Discussion and Conclusion: The exercise machine software-generated MET values were representative of directly measured oxygen consumption values across a range of submaximal intensities during all-extremity semirecumbent exercise in older adults with multisystem impairments.
| Introduction |
|---|
 TopAbstract Introduction MethodResultsDiscussionConclusionReferences |
|---|
Exercise that can be performed by older adults with multiple impairments will be important for addressing morbidity related to frailty with aging, including dependency, disability, falls, hospitalization, admission to long-term care centers, and associated mortality.2–5 There is evidence that exercise training, largely resistance exercise, in older adults in institutions is important for maintaining physical performance.6 The few available studies on aerobic exercise training in older adults (>80 years of age) have shown increased peak oxygen consumption (
O2) and decreased systolic blood pressure (BP)7 as well as a relative increase in peak O2 (14% for men and 13% for women compared with control groups).8 Residents of assisted-living facilities (mean age=85.5 years, SD=6.6) who exercised 
9 minutes per week using an all-extremity semirecumbent machine had significantly improved mobility (P=.002) compared with those who exercised for shorter durations.9
All-extremity semirecumbent exercise machines, such as the BioStep Clinical Pro Semi-Recumbent Elliptical Cross Trainer* and the NuStep TRS 4000 Recumbent Cross Trainer,
provide a new mode of seated aerobic exercise training with both arms and legs contributing to work output. These exercise trainers are ideal for accommodating impairments of the lower or upper extremities, weight-bearing ability, balance, coordination, or gait in older adults.10 However, research addressing the validity of software-generated aerobic fitness values and the reliability of these values during all-extremity exercise is limited.11 Therefore, this study was undertaken to investigate the concurrent validity of values for software-generated metabolic equivalents of task (MET
) from an all-extremity semirecumbent exercise machine (NuStep) and directly measured values for MET from a portable metabolic unit (K4b2
) across a range of submaximal exercise intensities. A second purpose was to determine the test-retest reliability of heart rate (HR) responses and O2 in older adults between standardized sessions of submaximal all-extremity aerobic exercise.
| Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
=.05, power=80%). The primary variable was the correlation coefficient (Pearson r), indicating the minimum expected agreement between software-generated NuStep and directly-measured K4b2 MET values. Thirteen subjects were required to declare a Pearson r correlation coefficient of .70 to be statistically significant. However, with an expected dropout rate of about 30% in this older population, a minimum sample size of 17 subjects was required. The demographic characteristics of the subjects (Tab. 1), including their age, height, body mass, type of gait aid, date of admission to long-term care center, number of health conditions, and daily medications, were obtained from a review of their medical charts. Subjects were visited individually in their rooms to complete the MMSE. Informed written consent was obtained from all subjects prior to participation.
|
Instrumentation
The NuStep is a hybrid semirecumbent bicycle and stair climber machine (Fig. 1) with a range of 10 workload levels. Each arm handle is coupled with a contralateral foot pedal, allowing the user to exercise the upper and lower extremities by using a reciprocal forward and backward motion. Exercise parameters displayed continuously on the NuStep liquid crystal display (LCD) panel include stepping rate, workload level, and accumulated exercise duration. To obtain continuous feedback for HR during exercise, subjects wore an HR monitor (Polar Vantage NV||) around the chest at the level of the xiphisternum. The Polar Vantage NV chest strap relayed HR via telemetry, and the HR was displayed on the NuStep LCD panel. The NuStep LCD touch screen panel could be manipulated to show instantaneous MET values, watts, total number of steps taken, or expended calories. The NuStep instantaneous MET values were software generated from power output (watts) and subject body mass.
|
O2 and carbon dioxide production, calculating the respiratory exchange ratio (RER), and reporting HR throughout exercise. The K4b2 is a lightweight (925-g) unit that is worn on a subject's torso and that uses telemetry to send data from the exercising subject. Subjects wear a face mask equipped with a turbine flowmeter to allow real-time collection of O2 data. Previous research comparing the K4b2 with the Douglas bag method over a range of cycling exercise intensities, from rest to 250 W,17 showed no significant differences between K4b2 and Douglas bag O2 values; these data suggested that the K4b2 portable metabolic system was acceptable for measuring O2 values over this range of exercise intensity. The K4b2 was calibrated according to the manufacturer's instructions prior to each exercise test session and set up for transfer of collected data to a laptop computer via telemetry. Data collection with the K4b2 began at the start of the warm-up period and ended at the completion of the cool-down period. The rating of perceived exertion (RPE) has been used to monitor exercise intensity for middle-aged adults when the HR response is blunted by HR-controlling medications18 and for older adults (>60 years of age) during walking or treadmill or cycle ergometer exercise.19–24 During maximal graded cycle ergometer exercise, older women (>70 years of age) were able to quantify their sense of effort very well by using the RPE (RPE-HR relationship, r=.95).24 In another study,23 the RPE was accurately used by a group of women (>70 years of age) to regulate exercise intensity during a training program.
Submaximal Exercise Protocol
The exercise protocol for this study was modified from the Young Men's Christian Association (YMCA) protocol.25 Each of the 3 constant workload levels was lengthened from 3 to 4 minutes to allow more time for the older adults in the present study to reach steady-state HR and O2 levels.26 Warm-up and cool-down phases remained the same as those in the YMCA protocol.25
Subjects completed a practice exercise session on the NuStep prior to 2 exercise test sessions. During the practice session, subjects wore the K4b2 unit and the Polar Vantage NV chest strap so that they could become familiar with the equipment, exercise protocol, and exercise motion and so that we could determine the workloads to be used for the 2 standardized exercise test sessions. Subjects were seated comfortably on the NuStep with a 5-degree knee bend at maximal leg extension and a 5-degree elbow bend at maximal arm extension. The seat position and arm length were noted so that we could standardize subject positioning during the exercise test sessions. Resting HR and BP values were obtained, and age-predicted maximal HR (HRmax) was calculated with the equation 208–(0.7 x age).27 A metronome provided audible cues for a rate of 60 arm and leg push-pull cycles per minute during exercise. Subjects also watched the stepping rate displayed on the NuStep LCD panel so that they could monitor and adjust their tempo of push-pull cycles.
During the practice session, subjects completed 2-minute bouts of exercise on the NuStep at progressively higher workloads (range=1–8) to determine their HR responses and RPE scores at each workload. At the highest workload attained, subjects completed 4 minutes of exercise without exceeding 75% of the age-predicted HRmax to confirm their ability to complete this level of exercise. At the end of the practice session, a workload was determined for each of the three 4-minute stages of exercise in preparation for the 2 exercise test sessions. The workload for stage 1 (workload 1) was the highest workload given an RPE of 3 by the subject. The stage 3 workload (workload 3) was the highest workload for which the subject could complete 4 minutes of exercise. The stage 2 workload (workload 2) was selected to represent a workload halfway between those selected for stages 1 and 3. For each subject, age-predicted HRmax calculations, seat position and arm length, and workload for each of the 3 stages of exercise from the practice session were transferred to day 1 and day 2 data collection sheets to standardize the exercise protocol and subject positioning for testing. Within 10 days of the practice session, subjects completed 2 standardized exercise test sessions at the selected workloads.
On day 1 of testing, subjects were seated in the standardized position and wore the K4b2 unit and the Polar Vantage NV chest strap. Subjects completed a 3-minute warm-up period at workload 1, three 4-minute stages at the preselected incremental workloads, and a cool-down period at workload 1. Subjects were encouraged regularly to maintain a push-pull tempo of 60 cycles per minute and to use the stepping rate displayed on the NuStep LCD panel to monitor and adjust their tempo of push-pull cycles. Exercise termination criteria included volitional stopping, an inability to maintain a cadence of 60 cycles per minute, an HR in excess of 75% of the age-predicted HRmax, or any of the ACSM guidelines for stopping an exercise test.14
At the onset of the warm-up period, the K4b2 was turned on for continuous data collection throughout the exercise protocol and the cool-down period for each subject. Data from the NuStep, including displayed HR, MET values, and watts, and subjects' reported RPE scores were manually recorded on a data collection sheet during testing. In the third and fourth minutes of an exercise stage, NuStep MET values and watts were recorded every 6 seconds. During testing, subjects provided an RPE score at the end of the second minute in each 4-minute exercise stage. Heart rate, as displayed on the NuStep LCD panel, was recorded for each minute of warm-up, exercise, and cool-down periods. For the cool-down period, subjects were instructed to exercise slowly for 5 minutes at workload 1. If the HR was greater than 5 beats above the resting HR after 5 minutes, then the cool-down period was extended. Blood pressure was recorded after the completion of the cool-down period for safety and as a measure of recovery. The same exercise protocol, equipment, and individualized positioning were used for subjects during day 2 testing.
Data Analysis
Descriptive analyses were performed for subjects' age, body mass index, MMSE score, walking aids, duration of residency, chronic health conditions, and medications. Data reduction was required to generate single data points for K4b2 and NuStep MET values and for NuStep watts and HR. The last 30 seconds of breath-by-breath K4b2 O2 data in the third and fourth minutes of each 4-minute exercise stage were analyzed to produce 2 O2 values (milliliters of oxygen per kilogram per minute) per 4-minute exercise stage for each subject. The mean of these 2 O2 values was used as the K4b2 MET value (1 MET=O2/3.5) per exercise stage. The RER values for the subjects were generated by the K4b2 software following the completion of testing. The NuStep MET values and watts in the third and fourth minutes of exercise were averaged to produce one MET value and one watt value per 4-minute exercise stage. The HR value for each exercise stage was the average of the HR values in the third and fourth minutes.
To determine whether resting HR, BP, and RER values were different on day 1 and day 2, we performed 3 separate one-way repeated-measures analyses of variance (ANOVAs). Six separate one-way repeated-measures ANOVAs were performed to determine whether watts, HR, percentage of HRmax (%HRmax) achieved, RPE, K4b2 MET values, and NuStep MET values by exercise stage were different on day 1 and day 2. To investigate whether watts, HR, %HRmax achieved, RPE, K4b2 MET values, and NuStep MET values collected on day 1 across the three 4-minute stages of exercise were different from those collected on day 2, we performed 6 separate 2-way repeated-measures ANOVAs. A Tukey post hoc test was used to identify the location of specific differences when the ANOVA omnibus F statistic was significant.
The relationship between group mean MET values from the K4b2 and those from the NuStep for each 4-minute exercise stage was determined by use of Pearson r correlation coefficients on day 1 and day 2. The strength of the correlation was characterized as described by Colton28: 0 to .25=no relationship, .25 to .50=fair, .50 to .75=moderate to good, and greater than .75=very good to excellent.
The test-retest reliability of NuStep MET values, K4b2 MET values, HR, and RPE was determined by use of intraclass correlation coefficients (ICC [2,1]) calculated from the results of the separate 2-way repeated-measures ANOVAs. The ICCs were characterized as described by Vincent,29 who suggested that ICCs below .80 are of questionable importance for physiological data, whereas values of .80 to .89 reflect moderate reliability and values greater than .90 indicate high reliability.
Data were analyzed by use of SPSS-PC (version 15.0).# The level of significance was set at P<.05 for all statistical tests.
| Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
|
|
|
|
|
O2, or K4b2 MET values, across the 3 stages of exercise (Tab. 4). In addition, RPE scores reported by subjects at stages 1, 2, and 3 of exercise were moderate to high: ICC (2,1)=.88, .87, and .93, respectively. | Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
O2, were not different from the directly measured K4b2 MET values during submaximal exercise in our sample of frail older adults living in long-term care centers and that this group of older adults with multisystem impairments showed reproducible exercise responses (HR and K4b2 MET values) during 2 test sessions with a standardized submaximal 3-stage exercise protocol. These results are important for physical therapists and rehabilitation professionals using the NuStep to assess submaximal aerobic performance and to monitor and prescribe exercise training. Also of clinical importance is that because the NuStep is ergonomically designed and has a range of workload levels to accommodate a broad range of people with variations in height, weight, range of motion, and strength and mobility impairments, no equipment modifications were necessary to enable our group of older adults with multiple health conditions to complete submaximal all-extremity exercise at some workload. A limitation of the present study was that the complete range of available workloads was not tested. Our subjects exercised up to a mean workload of 5 (range=3–8) out of 10 workloads. The group mean age-predicted %HRmax achieved at the highest workload tested was 61% (SD=10%). Only 1 of our 18 subjects was tested at workload 8, possibly because of low levels of aerobic fitness, higher ratings of perceived exertion secondary to a lack of regular aerobic exercise, the cumulative effects of multiple chronic conditions, and the presence in many subjects of chronic cardiovascular conditions (78% of subjects were prescribed β-blockers). Future studies addressing the reliability and validity of exercise responses in older adults at higher NuStep workload levels will need to recruit older subjects with higher functional abilities.
The combined contribution of arms and legs to work output during exercise on the NuStep facilitated exercise participation in our group of older adults with multisystem impairments. Similarly, given that the HR could not be relied on solely to reflect exercise intensity because most subjects were taking β-blockers, the RPE provided a mode for monitoring subjects during exercise and for guiding the choice of workloads. The RPE provided by exercising older adults was shown to be a reliable measure of exercise intensity, with moderate to high ICCs.
Although no randomized clinical trials have examined the therapeutic effects of all-extremity exercise, 2 studies supported the use of combined upper- and lower-limb exercises.31,32 Loudon et al31 compared peak O2 values from maximal treadmill and all-extremity ergometer (Pro II Power Trainer**) tests in women aged 30 to 60 years who were healthy and found that all-extremity exercise was a reliable and valid mode of testing aerobic fitness and a viable alternative to using a treadmill. Hill et al32 demonstrated that the all-extremity Power Trainer safely measured cardiopulmonary fitness in people with movement impairments following a stroke. Finally, Zehr33 suggested that rhythmic movements of the upper and lower limbs on the NuStep during supported sitting elicited neural responses similar to those elicited by walking, and Ferris et al34 reported that rhythmic upper-limb movements resulted in increased lower-limb muscle recruitment, which may benefit people with impairments of the lower or upper limbs or both. These preliminary findings, combined with our results, suggest that all-extremity semirecumbent exercise machines may be a viable exercise option for older adults and other people with health and functional impairments.
| Conclusion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
O2. The exercise responses of frail older adults were reliable across the range of workloads tested in the present study, suggesting that direct physiological assessments are not needed for monitoring subjects when the NuStep is used for exercise testing or training. The present study, therefore, provides preliminary evidence to support the use of the NuStep all-extremity semirecumbent exercise machine for aerobic exercise by older adults who have multiple health conditions and who have difficulty with traditional exercise modalities. The results of this study are important for physical therapists and rehabilitation professionals in hospitals and long-term care centers using these exercise machines to maintain or improve quality of life for frail older adults.
| Footnotes |
|---|
This study was approved by the Research Ethics Board for the Review of Health Sciences Research Involving Human Subjects at The University of Western Ontario.
Data from this research were presented by Dr Mendelsohn at the Aging, Rehabilitation, and Geriatric Care Annual Research Day, Parkwood Hospital, St Joseph's Health Care London; November 9, 2006; London, Ontario, Canada. Data from this research also were presented in: Mendelsohn ME, Connelly DM, Overend TJ, Petrella RJ. Submaximal aerobic fitness testing of frail, older adults using a semi-recumbent all-extremity exercise machine [abstract]. J Am Geriatr Soc. 2006;54:S109.
* Biodex Medical Systems, 20 Ramsay Rd, Shirley, NY 11967-4704. 
NuStep Inc, 5111 Venture Dr, Suite 1, Ann Arbor, MI 48108.
Cosmed USA Inc, 2211 N Elston Ave, Suite 305, Chicago, IL 60614.
|| Polar Electro Inc, 1111 Marcus Ave, Suite M15, Lake Success, NY 11042-1034.
# SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.
** SCIFIT Systems Inc, 5151 S 110th E Ave, Tulsa, OK 74146.
| References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. A Billinger, B. Y Tseng, and P. M Kluding Modified Total-Body Recumbent Stepper Exercise Test for Assessing Peak Oxygen Consumption in People With Chronic Stroke Physical Therapy, October 1, 2008; 88(10): 1188 - 1195. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
1 MET=3.5 mL O2·kg–1·min–1.