Data collected from a maximum exercise test can be used to assess an individual's level of aerobic fitness. A maximum exercise test with concomitant measurement of oxygen consumption (o2) is one method used to guide exercise prescription. However, assessment of maximum o2 (o2max) using traditional modes of testing may prove to be difficult in a variety of populations such as the elderly (19), deconditioned individuals (19), or those with brain injury such as stroke due to balance deficits, gait impairments, and decreased coordination (4). For these individuals, cycle ergometer (14,17) or the use of a combined arm and leg ergometer (8,10) may be beneficial for exercise testing when treadmill testing is not feasible. However, appropriate exercise testing protocols should be used to obtain accurate values for cardiorespiratory fitness.
For example, Siconolfi and colleagues (19) suggested that even for healthy individuals some testing protocols may not provide an accurate assessment of cardiorespiratory fitness. In their study, 13 of 32 participants who were 39 years of age and older demonstrated difficulty performing 2 minutes of exercise during the Astrand-Rhyming cycle test. These 13 participants were unable to complete the test because pedaling against the resistance was too difficult; therefore, the test values were underestimated for o2max in this sample. Arm ergometry is another viable option for exercise testing; however, arm cranking can produce peak o2 values that are up to 30%-35% less than those of treadmill testing (18). The American College of Sports Medicine's Exercise Management for Persons with Chronic Diseases and Disabilities recommends using a seated stepper to implement exercise prescription for individuals with neuromuscular disorders such as traumatic brain injury or stroke (12). However, in order to allow for appropriate exercise programming using a seated stepper, a valid and reliable exercise test for the total body recumbent stepper (TBRS) would be valuable.
A seated stepper has been used in an exercise intervention in a variety of settings including rehabilitation, community fitness centers, and retirement communities. The NuStep TRS 4000 (NuStep, Inc., Ann Arbor, MI) (Figure 1), or TBRS, consists of 10 settings (1-10) with load 1 (50 W) being the least amount of resistance. The TBRS uses reciprocal movements of both upper and lower extremities. Because the TBRS is a seated stepper, it is low impact, which is ideal for the elderly and individuals with arthritis or joint replacements of the lower extremities (10). Since the individual using the TBRS grasps the arm pole in a position halfway between wrist pronation and supination, this also accommodates those with limited ranges of motion (12). For individuals who experience difficulty maintaining proper foot placement during exercise, wide footplates with posterior and lateral borders support the feet and foot straps hold the foot securely in place.
No maximal exercise test exists for the TBRS. The purpose of this study was to assess the validity and reliability of the TBRS exercise test (TBRS-XT) for maximum exercise testing in a healthy population. It was hypothesized that the maximum TBRS-XT would be a valid and reliable test for determining o2max.
Experimental Approach to the Problem
This was a within-subject design to examine the validity and reliability of the TBRS-XT that terminated at o2max or when the participant reached volitional fatigue. The variables (maximal heart rate, o2max, respiratory exchange ratio (RER), and maximal exercise time) from TBRS-XT were compared to the Bruce protocol treadmill test.
Twenty-two healthy participants (13 men, 9 women) with a mean age of <27.9 ± 6.0 years volunteered to participate in the study. Eighteen participants were tested for the validity of the TBRS-XT, and 5 participants participated in reliability testing. All participants reported no neuromuscular, cardiopulmonary, metabolic, or musculoskeletal impairments that would limit their performance. Participant responses from the nonexercise estimation of o2max (6) placed 19 individuals in the high and 3 in the moderate physical exercise categories. Institutionally approved informed consent was obtained in writing before participation in the study. Data collection was performed at the Georgia Holland Cardiopulmonary and Neuromuscular Lab at the University of Kansas Medical Center.
Eighteen participants performed 2 maximum graded exercise tests in random order on separate days to test the validity of the TBRS-XT. Exercise testing sessions were separated by at least 48 hours but no more than 14 days apart and were controlled for time of day. Participant demographics are presented in Table 1. Reliability testing consisted of 2 paired test-retest sessions that were separated by no more than 5 days and were also controlled for time of day. Five participants, including one who participated in the validity testing, performed the 2 maximum graded exercise tests using the TBRS-XT for the purpose of assessing reliability. Participants were instructed not to eat food 2 hours before testing, but were allowed to hydrate with water ad libitum. Additionally, participants were asked to avoid caffeine and vigorous physical activity for at least 6 and 24 hours, respectively, before testing. Once informed consent was obtained, participants were asked to complete a nonexercise estimation of o2max to place them in a fitness level based on the subject's exercise activity (6).
Maximum Treadmill Exercise Test
The Bruce protocol was used for treadmill testing. Participants' heart rate, RER, and o2 were recorded continuously, and calculation of expiratory gases was performed every 30 seconds throughout the test. The treadmill exercise test was considered maximum when the patient reached at least 2 of the 3 criteria: 1) maximum heart rate was within 10 beats of predicted maximum, 2) plateau of o2max (within 0.57 L·min−1), and 3) RER ≥1.10, as suggested by Loudon and colleagues (10).
Maximum Total Body Recumbent Stepper Exercise Test
Participants were acclimated to the TBRS before the exercise test. The seat was adjusted either forward or backward to allow for a slight bend in the knee with extension, which would prevent hyperextension or the knee from “locking” in full extension. Differences in arm lengths were accounted for by modifying the length of the adjustable arm poles. These values were recorded for reuse in subsequent testing sessions. Individualization allowed participants to adequately and appropriately extend and flex the elbow and the shoulder joints during the exercise session. The maximum TBRS-XT (Table 2) consisted of 2-minute stages and concomitant increases in resistance until the participant reached test termination criteria. Exercise testing began with a warm-up stage at load 1 (50 W). Preliminary study suggested that loads 2 and 3 unnecessarily prolonged test duration >16-18 minutes (data not shown). Therefore, participants exercised at loads 4-10, increasing in resistance with test progression. At each load, participants were asked to maintain a stepping cadence of 115 steps per minute but were allowed a range of 110-120 steps per minute. If participants were unable to maintain a stepping cadence at 110 steps per minute with verbal cues and encouragement for participation, the test was terminated. No tests were terminated based on this criterion. Test termination criteria for the TBRS-XT used identical criteria mentioned above in the treadmill protocol (10).
Heart rate was continuously monitored throughout each exercise testing session using the Polar Vantage XL heart rate monitor (Polar Electro, Kempele, Finland). The maximum treadmill graded exercise test was performed on a Quinton Q55 (Quinton Instrument Co., Seattle, WA) and the TBRS protocol was performed on the NuStep TRS 4000 (NuStep). To ensure reliability of the TBRS-XT, the NuStep was calibrated during assembly using the Lode Portable Calibrator 2000 (Lode BV, Groningen, the Netherlands). A ParvoMedics TrueOne 2400 open circuit spirometry (ParvoMedics, Sandy, UT) was used for collection and analysis of expired gases continuously using a 2-way rebreather valve (Hans Rudolph, Kansas City, MO) and a noseclip. The sampling technique was a 30-second averaging of the data. All equipment was calibrated according to the manufacturer's recommendations before testing procedures. Participants rated perceived exertion at each stage of exercise testing using Borg's 6-20 scale.
Statistical analyses of data were conducted using a simple linear regression in which the response variable was the o2max from the Bruce protocol and the prediction variable was the o2max from the TBRS-XT. Since the Bruce protocol is considered gold standard, it was chosen as the response variable in order to predict measures from the TBRS-XT (prediction variable). For the regression equation, a 95% prediction interval was used to assess the strength of the prediction of o2max and maximum heart rate from the Bruce protocol. Pearson's correlation coefficient was used to assess the relationship between o2max and maximum heart rate for the TBRS-XT and Bruce protocol. Paired Student t tests were conducted to determine whether a significant difference existed between o2max and maximum heart rate values for the TBRS-XT and Bruce protocol. One goal of this study was to assess the clinical predictability of TBRS-XT for the Bruce protocol. Therefore, using linear regression, we calculated a predicted Bruce protocol using a linear combination of slope and intercept for each TBRS-XT for both o2max and maximum heart rate. Paired Student's t tests were conducted to determine whether a significant difference between o2max and maximum heart rate values for the predicted Bruce protocol (using the TBRS-XT) and the Bruce protocol on the treadmill.
To determine the reliability for each testing protocol, an intraclass correlation coefficient (ICC) was calculated for the paired TBRS-XT and Bruce protocol. One-way analysis of variance F statistic was used to test for statistical significance (α ≤ 0.05).
The individual scores for o2max values (L·min−1) and heart rate (b·min−1) for each test are shown in Appendix 1. For the o2max prediction interval, the slope and intercept point estimates were 0.96 and 0.56, respectively, with an R2 of 0.851 and the SEE of 0.307. For the maximum heart rate prediction interval, the slope and intercept point estimates were 16.25 and 0.95, respectively, with an R2 of 0.0.93 and the SEE of 3.48 (Figures 2 and 3).
Pearson's correlation coefficient for o2max and maximum heart rate (r = 0.92 and 0.96, respectively) between the Bruce protocol and TBRS-XT protocol indicates a strong relationship for the group. Data were then analyzed separately for men and women. Pearson's correlation coefficient also revealed a strong relationship for o2max (men: r = 0.88, women: r = 0.96) and maximum heart rate (men: r = 0.97, women: r = 0.92) between the Bruce protocol and the TBRS-XT protocol.
Group mean o2max values were 3.67 ± 1.07 L·min−1 (1.97-4.77 L·min−1) and 3.13 ± 0.80 L·min−1 (1.72-4.20 L·min−1) for the Bruce protocol and TBRS-XT, respectively. A statistically significant difference was observed between o2max values for the Bruce protocol and the TBRS-XT, where p < 0.01. Group mean maximum heart rate for the Bruce protocol was 188 ± 13 b·min−1, while the TBRS-XT elicited 181 ± 13 b·min−1 with a statistically significant difference observed between values (p < 0.01). Significant differences in o2max and maximum heart rate were seen for both men and women (p < 0.01). Adjusted TBRS-XT data for both o2max and maximum heart rate values using the slope and intercept showed no significant differences (p = 0.40) for either maximum heart rate or o2max. Table 3 shows o2max, heart rate, RER, and metabolic equivalents (METS).
The ICC for the Bruce protocol and TBRS-XT proved highly reliable (r = 0.99, p < 0.001 and r = 0.98, p < 0.001, respectively). In a review article, Atkinson and Nevill (2) suggest using 20-30 participants for assessing the reliability of variables. However, even with the sample size of 5 subjects, statistically significant values were obtained.
In this study, we evaluated the validity and reliability of a novel maximum exercise test using the TBRS-XT. Our results indicate that the TBRS-XT is a valid, reliable, and alternative exercise test for assessing cardiorespiratory fitness and predicting o2max. The validity for group o2max and heart rate (R2 = 0.85 and 0.93, respectively) between tests suggested a strong relationship. The current study had a higher o2max in the Bruce protocol (3.67 ± 1.07) than the TBRS-XT (3.13 ± 0.80 L·min−1). The lower values for o2max and heart rate for the TBRS-XT may be attributed to the fact that participants' body weight was supported by the TBRS seat in addition to localized fatigue of the upper and lower extremities, which according to American College of Sports Medicine guidelines (1) may account for a 5%-20% reduction in o2max values. In this study, a 12% decrease in o2max values existed between the Bruce protocol and the TBRS-XT, which was statistically significant. Therefore, based on the inability to reach o2max and maximum heart rate, the TBRS-XT may be used to assess and measure peak o2 rather than o2max.
Pollock and colleagues (16) found significant differences in o2max measures among 3 different protocols (Bruce protocol, Balke for the treadmill, and a bike test). Maximum heart rates were significantly lower with the bike protocol (177 ± 11 b·min−1) compared to the 2 treadmill tests (181 ± 9 and 182 ± 9 b·min−1 for the Bruce and Balke protocols, respectively.) No significant differences for heart rate existed between the Bruce and Balke treadmill tests. In addition, o2 max values for the group were significantly lower in the bike test compared to the Bruce protocol. The bike test elicited a o2max of 36.6 ± 4.9 ml·kg−1·min−1, while values from the Bruce protocol were 40.7 ± 4.4 ml·kg−1·min−1. The results found by Pollock et al. concur with our results that a seated modality elicits a lower maximum heart rate and o2, similar to findings of other studies (5,11,15,13) for cycle and treadmill testing.
Hass and colleagues (7) used the TBRS as a mode of exercise training in a randomized study measuring o2max. In the Hass et al. study, o2max was measured using the Bruce and modified Bruce protocols, depending on participants' age, to determine baseline o2max values, and then the maximum TBRS-XT was performed. Participants were asked to maintain a step rate of 120 steps per minute, while the intensity increased every 2 minutes until participants reached exhaustion. In the present study, participants were asked to maintain 115 steps per minute, and the intensity increased every 2 minutes. However, loads 2 and 3 were eliminated from our protocol for 2 reasons: the number of watts for loads 1, 2, and 3 (50, 52, and 55 W) were similar across the bandwidth of the TBRS, and this resulted in maximum exercise tests longer than 16 and 18 minutes in pilot testing. Hass and colleagues reported the mean maximum TBRS-XT time for the exercise group at 16.2 ± 4.1 versus 10.9 minutes on the treadmill. In the present study, mean exercise duration for the TBRS-XT was not significantly different from that of the Bruce protocol (Table 3).
After baseline o2max testing, Hass and colleagues (7) had participants begin a 12-week exercise training program using the TBRS. After the training period, a significant increase in o2max was observed with both the Bruce protocol and TBRS-XT. Hass and colleagues did not mention validation of the TBRS-XT that was used and if any significant differences were observed between the Bruce protocol and maximum TBRS-XT. For the exercise group, the mean o2max values reported by Hass et al. were higher with the Bruce (or modified Bruce) protocol 2.3 ± 0.06 L·min−1 than with the TBRS 2.1 ± 0.5 L·min−1 for the exercise group, which is similar to the results presented in the current study.
Loudon and colleagues (10) recruited 33 women of various ages and fitness levels to engage in a discontinuous “all-extremity maximum exercise test” and a maximum treadmill test using the Bruce protocol. The all-extremity device used by Loudon et al. was a combined arm and leg cycle ergometer and similar to the TBRS-XT, which is an all-extremity recumbent stepper. While both modes of exercise use the upper and lower extremities, Loudon et al. found no significant difference in o2max and heart rate values between the 2 modes of testing, which differs from the results of the present study. Loudon and colleagues suggested that use of an all-extremity device may be appropriate for assessing cardiorespiratory fitness in women with varying fitness levels.
The TBRS allows movement of bilateral upper and lower extremities, which may be beneficial to individuals who have limitations of motion secondary neuromuscular deficit (12). Recently, Hill and colleagues (9) used an all-extremity device similar to the one used by Loudon and colleagues (10) for exercise testing in individuals in the subacute phase of stroke recovery. Hill et al. (9) suggested that use of the all-extremity ergometer as an exercise modality would be appropriate to use in subacute poststroke rehabilitation settings, whereby increasing functional capacity and/or mobility skills would be beneficial. The authors of the current study agree that using a seated recumbent stepper that incorporates use of the upper and lower extremities such as the TBRS would be an appropriate mode of exercise testing for individuals with neuromuscular deficits (3).
Limitations of this study are the small sample size and physical activity level. Most of the participants were actively engaging in regular, vigorous physical activity. A heterogeneous sample for fitness levels would be desirable so the TBRS-XT could be applied to a variety of fitness levels. Also, maintaining a consistent stepping cadence of 115 steps per minute for the duration of the TBRS-XT was difficult, and the use of a metronome may have helped participants maintain a consistent cadence. Participants who are not accustomed to exercise on the TBRS may have experienced localized leg fatigue (1) similar to that of cycle ergometry, which would cause them to reach volitional fatigue more quickly. However, if the lower extremities experience fatigue, the upper extremities can assist in performance and disperse the load across other muscle groups.
A strong correlation exists between the TBRS-XT and the Bruce protocol for maximum heart rate and o2max. Therefore, the TBRS-XT may be a valid and reliable alternative method for assessing cardiorespiratory fitness in a young, healthy population. A future study to determine the percentage of difference between the Bruce protocol and the TBRS-XT would be advantageous for individuals using the TBRS-XT. Furthermore, comparing a seated modality such as a cycle protocol to the TBRS-XT would beneficial and recommended to determine whether any differences exist in o2max or heart rate between the 2 modalities.
The results of this study may be applicable to those individuals who perform graded exercise tests for which the TBRS-XT may be a valid and reliable alternative method for maximum effort exercise testing. For individuals who train or exercise on a recumbent stepper, peak o2 values obtained from the TBRS-XT may help guide exercise intensity during training. It is recommended that future studies using the TBRS-XT focus on other populations that would benefit from using a recumbent stepper for exercise testing or prescription.
The authors acknowledge Dr. Karen Kuphal for her editorial expertise and NuStep, Inc. for providing calibration of the total body recumbent stepper.
1. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription
Philadelphia: Johnson and Napora, 2000. pp. 91-114.
2. Atkinson, G and Nevill, AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med
26: 217-238, 1998.
3. Billinger, SA, Tseng, BY, and Kluding, PM. Use of a recumbent stepper to obtain vo2
peak in people with chronic stroke. [abstract]. J Geriatr Phys Ther
4. Dobrovolny, CL, Ivey, FM, Rogers, MA, Sorkin, JD, and Macko, RF. Reliability of treadmill exercise testing in older patients with chronic hemiparetic stroke. Arch Phys Med Rehabil
84: 1308-1312, 2003.
5. Fernhall, B and Kohrt, W. The effect of training specificity on maximal and submaximal physiological responses to treadmill and cycle ergometry. J Sports Med Phys Fitness
30: 268-275, 1990.
6. Foster, C, Thompson, NN, and Bales, S. Functional translation of exercise responses during combined arm-leg ergometry. Cardiology
78: 150-155, 1991.
7. Hass, CJ, Garzarella, L, de Hoyos, DV, Connaughton, DP, and Pollock, ML. Concurrent improvements in cardiorespiratory and muscle fitness in response to total body recumbent stepping in humans. Eur J Appl Physiol
85: 157-163, 2001.
8. Hagan, RD. Cardiorespiratory responses to arm, leg and combined arm and leg work on an air-braked ergometer. J Cardiac Rehabil
3: 689-695, 1983.
9. Hill, DC, Ethans, KD, MacLeod, DA, Harrison, ER, and Matheson, JE. Exercise stress testing in subacute stroke patients using a combined upper- and lower-limb ergometer. Arch Phys Med Rehabil
86: 1860-1866, 2005.
10. Loudon, JK, Cagle, PE, Figoni, SF, Nau, KL, and Klein, RM. A submaximal all-extremity exercise test to predict maximal oxygen consumption. Med Sci Sports Exerc
30: 1299-1303, 1998.
11. Moody, DL, Killias, J, and Buskirk, ER. Evaluation of aerobic capacity in lean and obese women with four test procedures. J Sports Med Phys Fitness
9: 1-9, 1969.
12. Palmer-McLean, K, and Harbst, K. Stroke and Brain Injury. In: ACSM: Exercise Management for Persons with Chronic Diseases and Disabilities
. L. Robertson, D. Bihler, J. Davis, and K. Bernard, eds. Champaign, IL: Human Kinetics, 2003. pp. 238-246.
13. Pannier, JL, Vrijens, J, and Van Cauter, C. Cardiorespiratory response to treadmill and bicycle exercise in runners. Eur J Appl Physiol Occup Physiol
43: 243-251, 1980.
14. Pitetti, KH, Vaughan, RH, and Snell, PG. Estimation of VO2
max from heart rates during submaximal work on the Schwinn Air-Dyne ergometer. Med Sci Sports Exerc
19: S64, 1987.
15. Pollock, ML, Dimmick, J, Miller, HS, Kendrick, Z, and Linnerud, AC. Effects of mode of training on cardiovascular function and body composition of adult men. Med Sci Sports
7: 139-145, 1975.
16. Pollock, ML, Foster, C, Schmidt, D, Hellman, C, Linnerud, AC, and Ward, A. Comparative analysis of physiologic responses to three different maximal graded exercise test protocols in healthy women. Am Heart J
103: 363-373, 1982.
17. Ponchiatti-Mulcare, JA, Mathews, T, Glaser, RM, and Gupta, SC. Maximal aerobic exercise of individuals with multiple sclerosis using three modes of ergometry. Clin Kinesiol
49: 4-13, 1995.
18. Rimmer, JH and Nicola, T. Stroke. In: ACSM's Resources for Clinical Exercise Physiology
J. Meyers, W. Herbert, and R. Humphrey, eds. Philadelphia: Lippincott, Williams & Wilkins, 2002. pp. 3-15.
19. Siconolfi, SF, Cullinane, EM, Carleton, RA, and Thompson, PD. Assessing VO2
max in epidemiologic studies: modification of the Astrand-Rhyming test. Med Sci Sports Exerc
14: 335-338, 1982.