Acute and chronic effects of exercise using an Exercycle in healthy, untrained adults.

L Cunningham, DPE, FASCM, is Professor of Physical Education, Fitchburg State College, 160 Pearl St, Fitchburg, MA 01420-2697 (USA). Address all correspondence to Dr Cunningham.

R Cantu, MD, FASCM, is Director, Service of Sports Medicine, Chairman, Department of Surgery, and Chief, Neurosurgical Service, Emerson Hospital, Concord, MA 01742.

The study protocol was approved by the Human Studies Committee at Fitchburg State College.

This study was supported in part by a grant from the Exercycle Corporation, Woonsocket, RI.

This article was submitted January 17, 1989, and was accepted March 12, 1990.

A variety of aerobic exercise equipment has recently been introduced for cardiopulmonary rehabilitation and adult fitness. Assessment of the relative efficacy of these devices and the ability to make intelligent decisions about their use requires objective evaluation of the acute and chronic physiologic effects induced by exercising on these devices.

The Exercycle, (Exercycle Corp, 667 Providence St, PO Box 1349, Woonsocket, RI 02895.) a motor-driven exercise machine that allows passive or resisted movements of both the upper and lower body, is one such device. The mechanical action of the Exercycle consists of cranking movements by the lower body and push-pull movements by the upper extremities and upper body. Exercise may be performed at either 60 or 90 rpm. The exercise primarily requires concentric contractions of the muscles of the chest, back, upper arm, and lower limbs. The exercise can be modified by the user to require eccentric contractions. The resistance provided by the device can be increased by manually adjusting a setting (arbitrary units of 1-6) built into a monitor attached to the device. Power output, as measured in watts or kilogram-meters per minute for each of the arbitrary settings, is not known. Thus, the Exercycle is an exercise machine and not an ergometer. The Exercycle is primarily a cardiopulmonary exercise device as opposed to a device that can be used to develop muscular fitness.

The Exercycle can be used for passive movement or resisted movement, depending on the health and aerobic capacity of the patient. The motor-driven mechanical action of the Exercycle allows the upper and lower body limb segments to be passively moved through a limited range of motion, thus providing patients with very low levels of exercise.

We raised three basic questions related to use of the Exercycle by healthy, untrained adults. First, because the Exercycle is motorized and provides both passive and resisted movement of limb musculature, can the cardiopulmonary system be adequately stressed to promote aerobic fitness? Second, does the Exercycle provide submaximal and maximal physiologic responses to graded exercise similar to traditional or standard lower extremity (LE) cycle ergometry? A larger mass of muscle theoretically may be used during exercise with the Exercycle than with LE cycle ergometry, resulting in a higher peak oxygen consumption (VO.sub.2] value and a lower heart rate (HR) at any comparable [VO.sub.2] level during exercise.[2] Third, are there gender differences in the training response? Do male subjects respond differently than female subjects to graded exercise and training on the Exercycle?

To investigate these questions, 20 healthy adults trained with the Exercycle for 6 weeks at a moderate intensity based on American College of Sports Medicine (ACSM) guidelines.[1] Our rationale for using only healthy adults was twofold. First, the enhancement of cardiopulmonary fitness for healthy adults is a goal of many physical therapy programs. Second, we believed that, prior to designing studies investigating the use of the Exercycle for various patient populations, the safely and efficacy of the Exercycle should be determined using healthy subjects as controls. Each subject was tested before and after training on the Exercycle and an LE cycle ergometer.

Method

Subjects

Twenty healthy, untrained subjects (9 male, 11 female), aged 18 to 53 years (X = 33.9), completed an 18-session training program and the final testing in 6 1/2 to 8 weeks. The characteristics of the subjects are summarized in Table 1. Each subject signed a statement of informed consent. All subjects were sedentary, as defined by not having participated in a formal exercise program for at least 3 months prior to the start of the study.

Procedure

All testing was completed in the Cardiovascular Laboratory at Burbank Hospital, Fitchburg, Mass. Metabolic and ventilatory data were obtained every 30 seconds with an automated Metabolic Measurement Cart.* Expired volume was calibrated with 1- and 3-L syringes, and the gas analyzers were calibrated with known gas concentrations. Electrocardiograms were obtained, and HRs were determined from a bipolar lead (CM5) using a Quinton stress test monitor.**

Subjects reported to the laboratory in the afternoon dressed in shorts and exercise shoes. Resting metabolic and HR data were obtained while the subjects were seated on either the Exercycle or the Monark cycle ergometer.*** (* SensorMedics Corp, 16030 S State College Blvd, Anaheim, CA 92806.

** Model Q3000, Quinton Instrument Co, 2121 Terry Ave, Seattle, WA 98121-2791.

*** Monark-Crescent AB, Varberg, Sweden, imported by Universal Fitness Products, 50 Commercial St, Plainville, NY 11803.)

The testing protocol used for either exercise device was a 2-minute-per-stage continuous test. The work load for the cycle ergometer test was increased by 30 W every 2 minutes until the subject was unable to maintain the required work output. Female subjects started at 30 W, and male subjects started at 90 W.

The Exercycle protocol was begun at a work-load setting of 1 for female subjects and at a setting of 3 for male subjects at the slow speed (60 rpm). The subsequent settings used in the testing protocol were 3,4, 4.5, 5, 5.5, and 6. The unit attached to the Exercycle contained a meter with a needle and lighted zone for visual inspection. Each subject was instructed to keep the needle within this zone at all times to standardize the resistance at each setting. The resistance provided is curvilinear when [VO.sub.2] is plotted as a function of work setting, as demonstrated by the results of a pilot study of four female and four male subjects and as shown in this study (Fig. 2). That is, settings 1 through 3 exhibited small increases in 702 and settings 4 through 6 showed larger increases. Subjects were encouraged to use both their upper and lower body and to exert enough force to maintain the needle within the lighted area at each setting up to maximal effort. The test-retest data for peak HR, minute ventilation (VE), and [VO.sub.2], were determined to be reliable from Exercycle testing in the pilot study. Correlation coefficients using the Pearson product-moment method of .880,.905,.926, and .972 were obtained for peak HR, VE, [VO.sub.2] (in milliliters per kilogram per minute), and [VO.sub.2] (in liters per minute), respectively.

Criteria used for determining a maximal effort were a plateau in [VO.sub.2] or an increase of less than 250 mL [times] [min.sup. -1] with increased levels of work, a respiratory exchange ratio (RER) greater than 1.0, and achievement of an HR within 10% of the predicted age-adjusted maximum value. The two highest 30-second [VO.sub.2] values were averaged for the peak [VO.sub.2] determination for both the cycle ergometer and Exercycle protocols. Data from the second minute of each stage were used for analysis of submaximal physiologic responses to graded exercise. The test sessions on each exercise machine were completed within 4 days of each other for most of the subjects. The Exercycle test always preceded the cycle ergometer test. All subjects participated in at least one practice training session on each testing device prior to the initial testing.

On completion of the Exercycle testing, an individualized exercise prescription was written based on ACSM guidelines.[1] Heart rate and the rating of perceived exertion (RPE)[3] were monitored during each training session by a technician. Because untrained subjects who exhibited low initial levels of aerobic fitness (peak [VO.sub.2] values of 29 and 35 mL [times] [kg.sup.-1] [times] [min.sup. -1] for female and male subjects, respectively, on the Exercycle test) participated in this study, a conservative approach to training was taken as might be used in a supervised physical therapy session. Each training session consisted of a 5-minute warm-up, a cardiopulmonary segment where the HR and RPE were within each subject's target zone (mean percentage of peak [VO.sub.2] of 61.5%), and a 5-minute cool-down. The cardiorespiratory segment of the session was begun conservatively at 15 minutes and progressed by 1 minute per session to a total of 32 minutes over the 18 training sessions. The average training HR was 144 (78% of maximum HR) and 142 bpm (77% of maximum HR) for male and female subjects, respectively. The average exercise time with the HR in the training zone was 25.6 minutes for the male subjects and 24.0 minutes for the female subjects over the 18 sessions. The average RPE was 12.5 and 11.7 for the male and female subjects, respectively. Male subjects trained at an average peak [VO.sub.2] of 62%, and female subjects trained at an average peak [VO.sub.2] of 61%, as estimated from the exercise HR based on the Exercycle pretraining measurements. Energy expenditure for male and female subjects during a training session was estimated at 244 and 138 kcal [times] [session.sup.-1], respectively. The difference in energy expenditure was suggested to be due to fitness levels and body weight because exercise time and intensity were comparable between genders.

Data Analysis

All data were expressed as means and standard errors of the means. A three-factor mixed design with repeated measures on two factors or a two-factor design with repeated measures on one factor were used where appropriate, to analyze the data. The three factors were 1) training (pretraining-posttraining measurements), 2) mode of testing (Exercycle-cycle ergometer), and 3) gender (male-female). This statistical design allowed us to interpret both the combined (main effects) and group (simple effects) results. Linear-regression equations were developed for each subject for each test for the HR-[VO.sub.2] relationship. The submaximal HR data were analyzed based on this linear model. A two-way analysis of variance was used to compute intraclass correlation coefficients to determine reliability of these data.[4] If an F ratio was significant, the Newman-Keuls post hoc test was used to determine mean differences at the .05 level of significance.

Results

Table 2 shows the simple effects for selected peak physiologic variables for both male and female subjects. Because changes in weight did occur, the most valid criterion for analyzing aerobic power was absolute peak [VO.sub.2] (in liters per minute). Exercycle testing revealed that the male subjects' peak [VO.sub.2] improved by 14% (ie, from 3.12 [plus or minus] 0.14 to 3.55 [plus or minus] 0.19 L [times] [min.sup.-1]). When [VO.sub.2] was expressed relative to body weight, a 20% gain was noted (ie, from 35.2 [plus or minus] 1.5 to 42.3 [plus or minus] 1.8 mL [times] [kg.sup.-1] [times] [min.sup. -1]). A statistically significant improvement in peak [VO.sub.2] of 7% (ie, from 1.89 [plus or minus] 0.05 to 2.03 [plus or minus] 0.06 L [times] [min.sup. -1]) was found for female subjects after Exercycle testing. When expressed relative to body weight, a 9% improvement was noted (ie, from 28.2 [plus or minus] 1.3 to 30.8 [plus or minus] 1.4 mL [times] [kg.sup.-1] [times] [min.sup. -1]). Significant changes were also noted after cycle ergometer testing. Peak [VO.sub.2] increased 6% (ie, from 2.95 [plus or minus] 0.19 to 3.13 [plus or minus] 0.22 L [times] [min.sup.-1]) for the male subjects. Relative to body weight, the male subjects' peak [VO.sub.2] increased 12% (ie, from 33.3 [plus or minus] 1.4 to 37.3 [plus or minus] 1.9 mL [times] [kg.sup.-1] [times] [min.sup.-1]). The female subjects demonstrated a 6% increase in peak [VO.sub.2] (ie, from 1.71 [plus or minus] 0.04 to 1.82 [plus or minus] 0.05 L [times] [min.sup. -1]) for cycle ergometer testing. Relative to body weight, the female subjects' pretraining-posttraining cycle ergometer tests revealed a change of 8% for peak 90, (ie, from 25.5 [plus or minus] 1.1 to 27.6 [plus or minus] 1.5 mL [times] [kg.sup.-1] [times] [min.sup.-1]). The changes in absolute peak [VO.sub.2] with training and the interaction for training x gender are shown in Figure 3.

The Exercycle showed a larger peak [VO.sub.2] than when testing was completed on the cycle ergometer for main effects (P [less than] .05). The difference was 10% higher for Exercycle testing (2.53 [plus or minus] 0.17 L [times] [min.sup.-1]) compared with cycle ergometer testing (2.3 [plus or minus] 0.16 L [times] [min.sup.-1]). Peak HR, VE, and RER were not significantly different for any main effect.

Figure 2 shows the [VO.sub.2] response as a function of Exercycle work level. Each data point represents the mean [VO.sub.2] of approximately 15 female and 10 male subjects from the pretraining test. The [VO.sub.2]-work-load response was shown to be nonlinear over the entire range of work-load settings (ie, 1-6). However, the increases in [VO.sub.2] for both male and female subjects are shown to be near-linear from work loads 3 to 5.5.

Figure 4 shows a comparison of the HR-[VO.sub.2] relationship to graded exercise between the Exercycle and the LE cycle ergometer. There were no significantly different HRs between testing modes at any given [VO.sub.2] level for either male or female subjects (P [greater than] .05). This finding indicates a similar submaximal cardiopulmonary response between the Exercycle and the cycle ergometer.

Figure 5 illustrates the pretraining-posttraining response to graded exercise for LE cycle ergometer and Exercycle testing. A small reduction in HR at any given [VO.sub.2] level was noted for both male and female subjects at the conclusion of the 6-week cardiorespiratory training program when testing was completed on the cycle ergometer. The differences did not reach statistical significance (P [greater than] .05). When testing was completed on the Exercycle, however, significantly lower HRs were found at most [VO.sub.2] levels, as shown in Figure 5b (P [less than] .05). This finding would support the notion of exercise specificity and would suggest some peripheral adaptation had occurred to the training stimulus specific to the Exercycle protocol.

These data for peak [VO.sub.2] were determined to be reliable across the testing sessions. An intraclass R of .926 for absolute peak [VO.sub.2] (in liters per minute) and an R of .978 for peak [VO.sub.2] relative to body weight (in milliliters per kilogram per minute) were computed. These values indicate that the bulk of the variance is attributable to the four testing trials and that only a small amount of the variance is due to error.

Discussion

Training

Exercycle training was effective for improving aerobic capacity and the cardiac response to submaximal exercise for individuals with a relatively low initial level of aerobic fitness. improvement in absolute peak [VO.sub.2] of 7% to 14% was found in just 6 weeks of training at a moderate exercise intensity (62% of peak [VO.sub.2] and 78% of peak HR). This finding is comparable to most training studies that report increases in maximal oxygen consumption ([VO.sub.2] max) for healthy individuals in the range of 10% to 20%.[1] The improvement of aerobic power and change in HR response to graded exercise indicates that the program used was effective in improving cardiopulmonary fitness. These data must be interpreted with caution, however, because the design of the study did not include a control group. It is noteworthy that not a single musculoskeletal complaint was voiced by any of the 20 individuals who completed the training program.

Exercycle training was effective for promoting change in the submaximal HR response to graded exercise. As shown in Figure 5b, HR was significantly reduced after training when testing was completed on the Exercycle. A significant reduction in HR at any given [VO.sub.2] level was not noted after cycle ergometer testing was completed. The greater reduction in HR that occurred with Exercycle testing for both male and female subjects suggests that a peripheral adaptation occurred to the training stimulus.[5] That is, the muscle mass related to the action of the Exercycle became specifically trained over the 18 training sessions. The reduction in HR is a well-documented training effect[5] and provides additional support for the Exercycle as an effective cardiopulmonary training device.

The pooled data suggest that Exercycle training provided a crossover effect of an increase in absolute peak [VO.sub.2] during cycle ergometer testing of 9.3%. Using cycling, running, or leg-press training, other researchers[6-8] have reported crossover effects in the range of 8% to 15% following training of 20 to 24 weeks' duration. Because the two modes of testing used in our study involved similar LE musculature, the approximately 9% crossover of aerobic power achieved in only 6 weeks could be expected based on the concept of exercise specificity. Apparently, the muscle mass of two legs is large enough to stimulate a central circulatory adaptation, as suggested by Clausen[5] and Shephard et al.[9] Lower extremity exercise, therefore, may have predominated during the workouts, resulting in an aerobic training adaptation that was very similar between the two testing devices. Previous studies[10-12] have demonstrated that when upper and lower body movements are combined, approximately 80% to 90% of the peak [VO.sub.2] is contributed by the musculature of the lower body. Further studies would be required to fully determine the exact nature of the training adaptation to Exercycle training.

Comparison of Testing Modes

The submaximal physiologic response to graded exercise revealed when testing was completed with the Exercycle compares well with that revealed by standard LE cycle ergometry. As shown in Figure 4, the HR-[VO.sub.2] submaximal response to graded exercise revealed by Exercycle testing was similar to the cycle ergometer testing response. it should be noted that the Exercycle is not an ergometer because power output units are unknown. When cardiopulmonary training is the goal, the Exercycle is best used with HR monitoring.

Peak aerobic power from the pooled data of all tests (main effects) was approximately 10% higher after Exercycle testing was completed than after cycle ergometer testing was completed. This finding differs from the results of studies by Nagle and colleagues[10] and Hagen et al,[12] who compared a combined upper and lower body air-braked ergometer with an LE cycle ergometer. They reported equivalent peak [VO.sub.2] values between testing modes. In our study, the differences in aerobic power between combined upper and lower body exercise and LE cycle ergometry might be due to the low levels of both aerobic and muscular fitness of the subjects. Most of the subjects reported lower limb discomfort as the primary limitation to maximal exercise during LE cycle ergometer testing. Another explanation may have been the order of testing. The Exercycle testing always preceded the cycle ergometer testing, both before and after training. We would argue that peak [VO.sub.2] was not influenced in a systematic manner by this factor. The testing sessions were conducted within four days of each other for most subjects, which allowed enough recovery between tests, yet was not long enough to be influenced by changes in training or detraining. Furthermore, peak VE, RER, and HR did not differ between the pretraining and posttraining testing sessions, suggesting similar maximal efforts were given by the subjects. The practical significance of a larger peak [VO.sub.2] elicited by Exercycle testing is unclear.

Gender Differences

The male subjects showed a greater adaptation to training than the female subjects. This finding is clearly shown in Figure 3. Other than a greater caloric output per exercise session for the male subjects, the intensity, frequency, and duration of the training sessions were similar for both genders. it has been suggested that, when total energy expenditure is held constant, improvement of aerobic power will be similar.[1] Had the duration of training been extended for female subjects to the average total energy expenditure of the male subjects, then comparable percent changes in [VO.sub.2] max might have been noted. Another factor may have been the amount of upper body exercise involved during training. Interviews at the completion of the study revealed that female subjects who made the greatest improvement in aerobic fitness used their upper body to a greater extent than those who made only minimal improvement, despite training at an appropriate percentage of maximum HR. All except one male subject stated that the upper extremities were used extensively during the training sessions. This extensive use of the upper extremities was particularly evident toward the end of the study as the subjects' aerobic fitness improved. Apparently, most women were able to obtain a target HR by using the LEs alone. The physiologic mechanisms to explain why a smaller mass of muscle used during exercise elicits higher HR responses are suggested to be related primarily to higher levels of plasma catecholamines and lower stroke volume, as compared with exercise involving a larger muscle mass.[2,8,9] We would recommend use of the upper body musculature during cardiopulmonary training with the Exercycle based on these observations. Further studies would be warranted to clarify these gender differences.

Conclusion

We evaluated the Exercycle as a cardiopulmonary training device to be used by physical therapists. We found that 1) the Exercycle promoted improvement in peak [VO.sub.2] and reduced HR response to submaximal exercise following 6 weeks of training, 2) the submaximal response to graded exercise revealed by the Exercycle compares with that revealed by LE cycle ergometry, and 3) male subjects demonstrated a greater response to Exercycle training than did female subjects. We conclude that the Exercycle can be used as an effective cardiopulmonary training device with healthy, untrained adults. Further studies using the Exercycle are warranted for specific patient populations such as those in cardiac, pulmonary, stroke, and orthopedic rehabilitation programs.

Acknowledgments

We are indebted to Lisa Conrad and Tracy Steeves for their assistance in the testing and training portions of this study, the Fitchburg State College Health Center, and the Clinical Laboratory and Respiratory Therapy Department at Burbank Hospital. We also thank Dr Steven Siconolfi of the Lyndon B Johnson Space Center for reviewing the manuscript and assisting with the statistical design and Dr Michael Sawka of the US Army Institute of Environmental Medicine for his review of the manuscript.

References

1. The recommended quantity and quality of exercise for developing and maintaining fitness in healthy adults. Med Sci Sports. 1978;10:8-10,

2. Sawka MN. Physiology of upper body exercise. Exerc Sport Sci Rev. 1986;14:175-211.

3. Borg G, Linderholm H. Perceived exertion and pulse rate during graded exercise in various age groups. Acta Med Scand 1973;35:236-243.

4. Kerlinger FN. Foundations of behavioral Research: Educational, Physiological, and Sociological Inquiry. 2nd ed. New York, NY: Holt, Rinehart & Winston General Book; 1973:442-451.

5. Clausen JP. Effect of physical training on cardiovascular adjustments to exercise in man. Physiol Rev. 1977;57:779-815.

6. Pollock ML, Dimmick J, Miller HS, et al. Effects of mode of training on cardiovascular function and body composition of adult men. Med Sci Sports, 1975;7:139-145.

7. Wilmore JH, Davis JA, O'Brien RS, et al. Physiological alterations consequent to 20-week conditioning programs of bicycling, tennis and jogging. Med Sci Sports Exerc, 1980; 12:1-8.

8. Pels AE, Pollock ML, Dohmeier TE, et al. Effects of leg press training on cycling, leg press, and running peak cardiorespiratory measures. Med Sci Sports Exerc 1987;19:66-70.

9. Shephard RJ, Bouhlel E, Vandewalle H, Monod H. Muscle mass as a factor limiting physical work. J Appl Physiol. 1988;64:1472-1479.

10. Nagle FJ, Richie JP, Giese MD. [VO.sub.2] max responses in separate and combined arm and leg air-braked ergometer exercise. Med Sci Sports Exerc, 1984;16:563-566.

11. Toner MM, Sawka MN, Levine L, Pandolf KB. Cardiorespiratory responses to exercise distributed between the upper and lower body. J Appl Physiol 1983;54:1403-1407.

12. Hagen RD, Gettman LR, Upton SJ, et al. Cardiorespiratory responses to arm, leg, and combined arm and leg work on an air-braked ergometer. Journal of Cardiac Rehabilitation. 1983;3:689-695.