Journal Logo

Applied Sciences: Physical Fitness And Performance

Physiological responses to upper body exercise on an arm and a modified leg ergometer

KANG, JIE; CHALOUPKA, EDWARD C.; MASTRANGELO, M. ALYSIA; ANGELUCCI, JOHN

Author Information
Medicine & Science in Sports & Exercise: October 1999 - Volume 31 - Issue 10 - p 1453
  • Free

Abstract

Stationary arm cranking is a mode of exercise commonly used for fitness testing and aerobic conditioning in individuals with lower extremity and spinal cord injuries (2,11,12,24). This exercise is often performed on a mechanically braked ergometer, which permits accurate measurement of power output (1). The Monark (Varberg, Sweden) arm ergometer (AE) is one such device that is widely used.

In many clinical settings, a modified Monark leg ergometer (LE) is often used to replace an AE for upper-body testing and conditioning (3,7,8,10,27). As such, questions arise as to whether upper body exercise on an AE and LE will elicit different physiological and perceptual responses even though the exercise is performed at a similar power output. There are several mechanical distinctions between an AE and an LE. An AE contains a shorter crank arm as well as a smaller flywheel as compared with an LE. An AE also contains a smaller crank sprocket as compared with LE, which will lead to a shorter displacement per crank revolution. To date, it has not been documented how these mechanical differences would affect physiological and perceptual responses during upper body exercise.

The present investigation was undertaken to compare cardiorespiratory, metabolic, and perceptual responses during arm cranking on an AE and an LE. The comparison was made by having subjects perform steady-state arm crank exercise on an AE and an LE at the same crank frequency and equivalent power outputs.

METHODS

Subjects

Seventeen male and seven female volunteers served as subjects. None of these subjects was engaged in any type of competitive sport. The subjects were informed of the purpose of the investigation and gave their written consent to participate. All experimental procedures were evaluated and approved by the Rowan University Institutional Review Board for Human Subjects experimentation. The physical and physiological characteristics of the subjects are presented in Table 1.

T1-15
Table 1:
Physical characteristics of subjects.

Experimental Design

Each subject completed a maximal oxygen uptake (V̇O2max) test on a treadmill and two experimental trials. The two experimental trials were presented in a counterbalanced order with each subject randomly assigned to a predetermined sequence. During each experimental trial, the subjects performed two 8-min steady-state arm crank exercises on an AE or an LE. The maximal test and two experimental trials were separated by at least 2 d rest and completed within a 3-wk period. All tests were conducted with subjects in a 4-h postabsorptive state.

Testing Procedures

V̇O2max test.

The V̇O2max test was used to document subjects fitness level and to establish anchors for rating perceived exertion. The test was administered on a motor-driven treadmill using the Costill/Fox running protocol (6). After a 10-min warm-up consisting of walking at 3–4 mph and 10% elevation, the subject began to run at a given speed and 0% elevation. The speed selected for this protocol was 3.5 m·s−1 (7.8 mph) for female subjects and 4 m·s−1 (8.9 mph) for male subjects. During the test, the speed was kept constant. However, treadmill elevation was increased by 2% every 2 min until the subject could no longer continue. Each subject was verbally encouraged to continue exercising until exhaustion. Oxygen uptake (V̇O2) and respiratory exchange ratio (RER) were measured every 30 s. Heart rate (HR) was measured during the last 15 s of each min. Maximal oxygen uptake was determined by averaging the two consecutive highest measures. To ensure that a true maximal oxygen uptake had been attained, at least two of the following three criteria were met: an increase in V̇O2 less than 100 mL·min−1 despite an increase in work load (leveling-off criterion), an HR considered to be maximal for a subject’s age, or an RER greater than 1.15 (20).

During this V̇O2max test, the subject also received the instructions regarding the definition of perceived exertion and the use of the perceived exertion rating scale. The instructions emphasized that the perceptual ratings should reflect sensations of exertion, stress, and/or discomfort in the limbs and respiratory system (the written copy of the instructions available from author). In addition, rating standards or anchors were established with the low standard being set equal to the feeling of exertion during walking at 3 mph and 0% and the high standard being equivalent to the sensation experienced during maximally exhaustive exercise. A rating of 7 was assigned to the low standard and a 19 to the high standard.

Experimental trials.

For each of the experimental trials, subjects reported to the laboratory at least 4 h after their last meal. During each trial, subjects performed two 8-min arm crank exercises on either an AE (Monark 881) or an LE (Monark 868) at two different power outputs in an ascending order. The LE was adapted for upper body use and the adapting was made according to Bohannon (3). As a result of adaptation, subjects were able to perform arm cranking on the AE and LE with the same body position. Both the ergometers were calibrated on a regular basis. The two power outputs selected for the two exercise bouts were 50 and 75 W for male subjects and 25 and 50 W for female subjects. These power outputs represent intensities that commonly fall into the range of 40 to 85% of arm V̇O2peak (14,15). The lower intensities were chosen for female subjects because female subjects generally have lower upper-body aerobic power. During each exercise bout, the height of the ergometer crank sprocket was adjusted so that the crank axle was at shoulder level and the elbow was extended but not locked when the handgrip was farthest from the body. The crank rate was kept constant at 50 rev·min−1 with the use of a metronome. The brake resistance was set at 1, 2, and 3 kg at power outputs of 25, 50, and 75 W, respectively, for the AE and 0.5, 1, and 1.5 kg at power outputs of 25, 50, and 75 W, respectively, for the LE. During the last 2 min of each exercise bout, V̇O2, expiratory ventilation (V̇E), RER, HR, and ratings of perceived exertion for the overall body (RPE) were measured every min and the average of the 2-min values for each variable was used for statistical analysis. A 10-min recovery interval separated the two exercise bouts during each experimental trial.

Measurements

Oxygen uptake, RER, and V̇E were determined using a two-way T-shaped breathing valve (Hans Rudolph 2700, Kansas City, MO) and a Cardio-Pulmonary Exercise System (Q-Plex1, Quinton Instruments, Inc., Seattle, WA). The Q-Plex1 calibrations included a carbon dioxide infrared absorption sensor (0–10% measurement range), an oxygen zirconia oxide sensor (10–35% measurement range), and a pneumotachometer (0–12 L·s−1 flow range). Heart rate was determined using a 12-lead electrocardiogram tracing apparatus (EK-10, Siemens Burdick, Inc., Milton, WI). Ratings of perceived exertion for overall body were determined using the Borg 15-category scale (4). The rating scale was in full view of the subject during the entire test. As the respiratory valve prohibited a verbal rating response, a finger signal was employed (23). Subjects were instructed to extend the index finger when the appropriate numerical rating was called by the investigator. The technique made it possible to measure RPE during arm exercise where the hands were firmly gripped around the crank handles. The RPE for arm was not measured. However, it can be reasonably assumed that this more localized perception of exertion will change in proportional to RPE for overall body. In addition, caloric expenditure during each exercise bout was calculated based upon V̇O2 adjusted for RER (17). The caloric expenditure in conjunction with power output was then used to determine gross efficiency (GE) using the following equation:MATH 1

Statistical Analysis

All statistical analyses were performed separately for male and female subjects. The dependent variables measured during experimental trials were analyzed using a two-way (ergometer mode × power output) analysis of variance with repeated measures. Analysis of variance was also followed by post hoc comparisons using Scheffé procedure. In addition, a linear regression analysis was used to determine the relation between oxygen uptake and power output for AE and LE. For all statistical analyses, a probability level of 0.05 was established to determine statistical significance.

RESULTS

V̇O2max Test

Cardiorespiratory and metabolic responses including V̇O2, RER, and HR during the V̇O2max test for both male and female subjects are presented in Table 2.

T2-15
Table 2:
Physiological responses during V̇O2max test.

Experimental Trials

In male subjects, significant main effects of exercise mode and power output were observed for all the dependent variables including V̇O2, HR, RER, V̇E, and RPE (Table 3). Post hoc comparisons further indicated that V̇O2, HR, RER, V̇E, and RPE were greater (P < 0.05), whereas GE was lower (P < 0.05) during arm crank exercise on an AE than an LE at both 50 and 75 W (Table 4. Post hoc comparisons also revealed that V̇O2 HR, RER, V̇E, and RPE were greater (P < 0.05) during arm crank exercise at 75 than 50 W when exercise was performed on either an AE or an LE (Table 4). Nevertheless, no difference in GE was found between 50 and 75 W for both modes of ergometer (Table 4).

T3-15
Table 3:
Analysis of variance on physiological and perceptual responses during submaximal arm crank exercise on AE and LE.
T4-15
Table 4:
Post-hoc comparison of physiological and perceptual responses during submaximal arm crank exercise on AE and LE.

In female subjects, a significant main effect of exercise mode was only observed for HR and RPE (Table 3). Post hoc comparisons also revealed that HR and RPE were greater (P < 0.05) during arm crank exercise on an AE than an LE at both 25 and 50 W (Table 4). With respect to other dependent variables, the differences between AE and LE varied at different power outputs. As shown in Table 4, V̇O2 RER, and V̇E were greater (P < 0.05) and GE was lower (P < 0.05) during arm crank exercise on an AE than an LE at 50 W, whereas these differences did not exist during exercise at 25 W. The results of analysis of variance on the main effect of power output and subsequent post hoc comparisons were similar to those of male subjects (Table 3 and 4), except for GE as a significant main effect of power output was not observed (Table 3).

The power output-V̇O2 relationships during arm crank exercise on an AE and an LE were examined using a regression analysis. The results of the analysis showed that at a given level of V̇O2, the corresponding power output was consistently lower during arm crank exercise on an AE as compared with an LE (Fig. 1). The magnitude of the difference in power output between AE and LE ranged from 5 to 10 W (or 30 to 60 kgm·min−1) and was directly proportional to the level of V̇O2.

F1-15
Figure 1:
The linear relationships between power output and oxygen uptake during arm crank exercise on AE and LE.

DISCUSSION

The primary goal of the present study was to compare cardiorespiratory, metabolic, and perceptual responses to upper body exercise on an AE and an LE. This was accomplished by having subjects perform arm crank exercise on an AE and an LE at the same crank frequency and equivalent power outputs. It was found that in male subjects, V̇O2, HR, RER, V̇E, and RPE were higher, whereas GE was lower, during arm cranking on an AE than an LE at both 50 and 75 W. In female subjects, similar differences in these same variables between the two ergometers were also observed when exercise was performed at 50 W. However, some of the variables such as V̇O2, RER, V̇E, and GE did not differ between the two ergometers when exercise was performed at 25 W.

The greater cardiorespiratory and metabolic responses such as V̇O2, HR, RER, and V̇E during arm crank exercise on an AE than an LE at a given power output may be attributable to the mechanical differences between AE and LE. As mentioned earlier, the crank arm length is shorter on an AE (0.14 m) than an LE (0.17 m). This shorter crank arm length would provide subjects a shorter crank lever. As a result, a greater muscular effort would have to be produced to maintain the same motive torque necessary for crank rotation and such a greater muscular effort may have resulted in a greater cardiorespiratory and metabolic responses during arm crank exercise on an AE than an LE. The diameter of the flywheel is also smaller on an AE (0.28 m) than an LE (0.56 m). It may be expected that the smaller flywheel on the AE would rotate faster and thus create more friction in the bearings. However, as the smaller flywheel seen in an AE is associated with a smaller crank sprocket, our measurement has revealed that for a given crank revolution, the smaller flywheel on an AE rotates only 2.4 times, whereas the larger flywheel on an LE travels 3.6 times. On the other hand, the larger flywheel of an LE possesses a larger momentum due to not only a higher speed but also a greater mass. It is possible that this larger momentum of the flywheel itself may have reduced muscular effort of the subject, thereby leading to a lower cardiorespiratory and metabolic responses during arm crank exercise on an LE as compared with an AE. In equating the total power output between AE and LE, we have doubled the brake resistance on an AE to compensate for the loss of displacement per crank revolution according to the manufacturer’s instruction (18). It may be hypothesized that such a doubling of the brake resistance would result in a disproportionally greater muscle recruitment and thus contribute to a greater cardiorespiratory and metabolic responses during arm cranking on an AE than an LE. However, this hypothesis remains to be tested as there can be no fundamental difference in muscular effort in producing power output with a higher force or a higher velocity, so long as the crank frequency and power output are held constant.

The greater cardiorespiratory and metabolic responses during exercise on an AE were found to be associated with higher RPE. Such an association supports physiological- perceptual congruity. It is conceivable that the greater RPE on an AE is due to a greater level of metabolic demand. It can be further speculated that the greater RPE may have been ultimately mediated by greater sensory inputs that are both central, i.e., ventilation, and peripheral, i.e., blood H+, in origin (19).

GE was found to be lower during exercise on an AE than an LE when the power output was equated. Apparently, this lower GE on an AE is attributable to a greater muscular effort exerted in performing a given power output as discussed earlier. When comparing two different power outputs within each mode of ergometry, we found that GE remained unchanged, although there is a trend toward a greater GE at a higher power output. We (15), as well as others (9,21), have previously found a progressive increase in GE with increments in exercise intensity, and the finding has been attributed to a decrease in the proportion of unmeasured work that comprises the total energy expenditure. Failure to show a significant increase in GE in the present study may be due to the fact that energy expended for unmeasured work increases proportionally with an increase in exercise intensity. Indeed, we did observe some extraneous movement of both the upper and lower body during exercise on an AE as well as an LE at the higher power output in some of our subjects.

An LE is often used to replace an AE for upper body exercise in clinical settings (3,7,8,10,27). If both an AE and an LE are used interchangeably, caution must be made in setting up the target exercise intensity because at a given power output, exercise on an LE will elicit comparatively lower physiological and perceptual responses as compared to exercise on an AE. We have made additional analyses to determine what adjustment in power output should be made to achieve an equivalent V̇O2 during exercise on an LE as compared with an AE. As depicted in Figure 1, it appears that an additional 5–10 W (30–60 kgm·min−1) should be added when using an LE for upper body exercise instead of an AE. It should be emphasized, however, that our analysis was made based upon only three intensity levels. The results of the analysis could be more conclusive if a greater number of intensities had been used. In light of the fact that arm crank exercise on LE elicits lower cardiorespiratory and metabolic responses, using an LE for upper body exercise may be viewed as a safer approach especially for individuals at high risk for developing cardiovascular and pulmonary diseases. In addition, we have noted that a number of studies that compared physiological responses between upper and lower body exercise have used the data derived from arm cranking on an AE and leg cycling on an LE, respectively (13,16,22,26). Based upon our current findings, it is likely that the physiological differences between upper and lower body exercise revealed in these studies could be attributable to not only the difference between upper and lower body musculature, but also the difference between the two ergometers per se. FIGURE 2

F2-15
Figure 2:
Oxygen uptake during arm crank exercise on AE and LE at 25 and 50 W in female subjects.

It is of interest to note that in our regression equations, the slopes (i.e., 3.6 mL·kgm−1 on AE and 3.3 mL·kgm−1 on LE) were greater than the slope used as an ACSM metabolic factor for arm ergometry (i.e., 3.0 mL·kgm−1). This difference may be due to the fact that our analysis has used the data of multistage steady-state exercise, whereas the ACSM slope was derived from the data of an incremental exercise. The greater slopes found presently indicate that as power output increases, V̇O2 measured in a steady state would increase to a greater extent than V̇O2 measured during an incremental exercise test. Such a discrepancy deserves further evaluation as our finding could raise a question as to whether the use of the ACSM equation would lead to a potential underestimation of metabolic cost during arm ergometry.

In our female subjects, some of the variables such as V̇O2, RER, V̇E, and GE did not differ significantly between the two ergometers when exercise was performed at 25 W. As there was a trend toward a greater V̇O2, a greater RER, a greater V̇E, and a lower GE during exercise on an AE than an LE, which was consistent with the differences observed at both 50 and 75 W in male subjects and at 50 W in female subjects, it is tempting to speculate that the similarity in these variables between AE and LE may have been due to a smaller sample size of our female subjects. Our further analysis revealed that the effect sizes [(mean 1 − mean 2)/pooled standard deviation] for V̇O2, V̇E, and GE were 0.7, 0.7, and 0.6, respectively. These effect size values suggest that a larger sample size is needed to detect the difference between means (5). In light of the fact that the metabolic differences found in male and female subjects in the present study occurred at higher levels of power output, i.e., 50 and 75 W, it is also possible that the metabolic differences between the two ergometers as demonstrated presently is intensity-dependent. This speculation is consistent with the observation that the magnitude of the difference in V̇O2 between AE and LE is directly proportional to the level of power output as shown in Figure 2.

The displacement per crank revolution on an LE is equal to 6.0 m (1). According to the instructions provided by the manufacturer, for a given power output, the brake resistance on an AE is exactly double what is required on an LE if the crank frequency is maintained at 50 rev·min−1 (18). This leads to our conclusion that the crank displacement on an AE is one-half as much as on an LE, i.e., 3.0 m. Interestingly, Swain and Leutholtz (25) have calculated displacement per crank revolution based upon the number of teeth on the crank sprocket. They found that the displacement per crank revolution on AE was 2.4 rather than 3.0 m, suggesting that to achieve a given power output, the brake resistance should be 2.5 times higher on an AE as compared with an LE. Although the displacement value reported by Swain and Leutholtz appears to be accurate, the finding is inconsistent with what is provided by the manufacturer. An explanation for such an inconsistency is not readily apparent. Nevertheless, it can be concluded that the magnitude of the cardiorespiratory and metabolic differences found in the present study could have been greater if the displacement value reported by Swain and Leutholtz (25) had been used.

In the present study, a steady-state in V̇O2 was observed in all trials (Figs. 2 and 3). However, comparisons between AE and LE at 75 W in male subjects and 50 W in female subjects should be viewed with caution as this apparent steady-state in V̇O2 was accompanied by a RER exceeding 1. Although the V̇O2 obtained on an AE at 75 W in male subjects and at 50 W in female subjects were equivalent to 70 and 90% of the arm V̇O2peak, respectively, which we documented previously using similar subjects (14), it is still possible that during these trials, some subjects may have reached their peak of circulatory and respiratory function and begun to accumulate an oxygen deficit.

F3-15
Figure 3:
Oxygen uptake during arm crank exercise on AE and LE at 50 and 75 W in male subjects.

In conclusion, upper body exercise elicits greater cardiorespiratory, metabolic, and perceptual responses on an AE than an LE at the same power output when power output is computed according to the manufacturer’s instructions. This finding suggests that when both an AE and an LE are used interchangeably for upper body exercise, an adjustment in power output should be made in order to achieve similar physiological and perceptual responses.

The authors would like to thank the subjects for their time and effort.

This study was supported by a Faculty Separately Budgeted Research Grant from Rowan University.

REFERENCES

1. Åstrand, P. O. Applied work physiology. In:Textbook of Work Physiology, 2nd Ed., New York: McGraw-Hill, 1977, pp. 449–459.
2. Balady, G. J. Types of exercise: arm-leg and static-dynamic. Cardiol. Clin. 11:297–308, 1993.
3. Bohannon, R. W. Adapting a bicycle ergometer for arm crank ergometry. Phys. Ther. 66:363–363, 1986.
4. Borg, G. A. V. Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc. 14:377–381, 1982.
5. Cohen, J. The Analysis of variance and covariance. In:Statistical Power Analysis for the Behavioral Science, Hillsdale, NJ: Erlbaum, 1988, pp. 273–403.
6. Costill, D. L., and E. L. Fox. Energetics of marathon running. Med. Sci. Sports 1:81–86, 1969.
7. Dicarlo, S. E. Improved cardiopulmonary status after a two month program of graded arm exercise in a patient with C6 quadriplegia. Phys. Ther. 62:456–459, 1982.
8. Franklin, B. A., L. Vander, D. Wrisley, et al. Aerobic requirements of arm ergometry: implications for exercise testing and training. Physician Sportsmed. 11:81–90, 1983.
9. Gaesser, G. A., and G. A. Brooks. Muscular efficiency during steady-rate exercise: effect of speed and work rate. J. Appl. Physiol. 38:1132–1139, 1975.
10. Glaser, R. M., M. N. Sawka, M. F. Bruce, et al. Physiological responses to maximal effort wheelchair and arm crank ergometry. J. Appl. Physiol. 48:1060–1064, 1980.
11. Glaser, R. M. Exercise and locomotion for the spinal cord injured. In: Exercise and Sports Science Reviews, R. L. Terjung (Ed.). New York: Macmillan, 1985, pp. 263–303.
12. Graves, J. E., and M. L. Pollock. Exercise testing in cardiac rehabilitation. Cardiol. Clin. 11:253–266, 1993.
13. Hooker, S. P., C. L. Wells, M. M. Manore, S. A. Philip, and N. Martin. Differences in epinephrine and substrate responses between arm and leg exercise. Med. Sci. Sports Exerc. 22:779–784, 1990.
14. Kang, J., E. C. Chaloupka, M. A. Mastrangelo, M. S. Donnelly, W. P. Martz, and R. J. Robertson. Regulating exercise intensity by ratings of perceived exertion during arm and leg ergometry. Eur. J. Appl. Physiol. 78:241–246, 1998.
15. Kang, J., R. J. Robertson, F. L. Goss, et al. Metabolic efficiency during arm and leg exercise at the same relative intensities. Med. Sci. Sports Exerc. 29:377–382, 1997.
16. Keyser, R. E., D. Mor, and F. F. Andres. Cardiovascular responses and anaerobic threshold for bicycle and arm ergometer exercise. Arch. Phys. Med. Rehabil. 70:687–691, 1989.
17. McArdle, W. D., F. I. Katch, and V. L. Katch. Measurement of human energy expenditure. In:Exercise Physiology Textbook, 2nd Ed, Philadelphia: Lea & Febiger, 1986, pp. 121–130.
18. Monark-Crescent, A. B. Instructional Manual for Monark Ergometer and Exercise Cycles. Varberg, Sweden, 1980, pp. 1–27.
19. Noble, B. J., and R. J. Robertson. Respiratory-metabolic mediators of exertional perceptions. In:Perceived Exertion, Champaign, IL: Human Kinetics, 1996, pp. 107–120.
20. Powers, S. K., and E. T. Howley. Work test to evaluate cardiorespiratory fitness. In:Exercise Physiology Textbook, 3rd Ed. Dubuque, IA: Brown & Benchmark, 1997, pp. 265–281.
21. Powers, S. K., R. E. Beadle, and M. Mangum. Exercise efficiency during arm ergometry: effects of speed and work rate. J. Appl. Physiol. 56:495–499, 1984.
22. Powers, S. K., S. Dodd, J. Woodyard, R. E. Beadle, and G. Church. Hemoglobin saturation during incremental arm and leg exercise. Br. J. Sports Med. 18:212–216, 1984.
23. Robertson, R. J., J. E. Falkel, A. L. Drash, et al. Effect of blood pH on peripheral and central signals of perceived exertion. Med. Sci. Sports Exerc. 18:114–122, 1986.
24. Sawka, M. N. Physiology of upper body exercise. In:Exercise and Sports Science, Rev. Vol. 14. K. B. Pandolf (Ed.). New York: Macmillan, 1986, pp. 175–211.
25. Swain, D. P., and B. C. Leutholtz. Using the cycle ergometry equations. In:Metabolic Calculations: Simplified. Baltimore: Williams & Wilkins, 1997, pp. 22–30.
26. Taguchi, S., and S. M. Horvath. Metabolic responses to light arm and leg exercise when sitting. Eur. J. Appl. Physiol. 56:53–57, 1987.
27. Zwiren L. D., and O. Bar-Or. Responses to exercise of paraplegics who differ in conditioning level. Med. Sci. Sports Exerc. 7:94–98, 1975.
Keywords:

ERGOMETER; OXYGEN UPTAKE; VENTILATION; HEART RATE; GROSS EFFICIENCY; RATINGS OF PERCEIVED EXERTION

© 1999 Lippincott Williams & Wilkins, Inc.