Cardiovascular responses to isokinetic trunk exercise.

C Peel, PhD, PT, is Associate Professor and Chair, Department of Physical Therapy, University of Texas Medical Branch, Galveston, TX 77550. She was Assistant Professor, Department of Physical Therapy, University of Texas Southwestern Medical Center, Dallas, TX 75235, when this study was conducted. Address all correspondence to Dr Peel at 9 Maple Ln, Galveston, TX 77551 (USA).

MJ Alland, BS, PT, is Staff Physical Therapist, John Peter Smith Hospital, 1500 S Main St, Fort Worth, TX 76104. She was a research assistant in the Department of Physical Therapy, University of Texas Southwestern Medical Center, when this study was conducted.

This paper was presented at the Sixty-Fifth Annual Conference of the American Physical Therapy Association, Nashville, Tenn, June 11-15, 1989.

This study was approved by the University of Texas Southwestern Medical Center Institutional Review Board.

This article was submitted September 1, 1989, and was accepted April 30, 1990

During the past several years, devices have been developed that allow patients to perform isokinetic (constant velocity) movements of trunk flexion and extension. The patient usually stands with the lower extremities supported and moves forward and backward against resistance arms that are positioned over the upper chest and back. The maximal speed of motion is determined by adjusting a setting on the ergometer, with available settings on most ergometers ranging from 0[degrees] to 180[degrees]/sec. Most models are computerized and display values on a computer monitor for the forces exerted during flexion and extension. These devices were designed to evaluate patients with back dysfunction and to provide a mode of therapeutic exercise for persons who need reconditioning.

Heart rate (HR) and blood pressure (BP) responses during exercise reflect the stress placed on the cardiovascular system. Knowledge of typical responses to various types of exercise is important so that programs can be modified for persons with heart disease. Isokinetic exercise is used extensively in physical therapy clinics for rehabilitation of persons with musculoskeletal injuries. Patients with musculoskeletal injuries may have either symptomatic or asymptomatic cardiovascular disease. There is a need for research in this area because studies documenting cardiovascular responses to strengthening exercises using isokinetic devices are limited in number.

Vinson et al[1] compared HR and BP responses during isokinetic extremity exercise with those measured during maximal exercise performed on a treadmill. The isokinetic exercise test used in this study was designed to measure muscular endurance of the knee flexor and extensor muscles of one extremity. The test consisted of one exercise period, which averaged 48 seconds, and subjects exerted maximal effort during this period. Peak systolic blood pressure (SBP) and diastolic blood pressure (DBP) values were similar to maximal values measured during treadmill exercise. Peak HR values for isokinetic exercises ranged from 70% to 88% of maximal HR values recorded during the treadmill test.

In a similar study, HR and SBP values were measured during a 20-minute isokinetic exercise training session and compared with values measured during maximal exercise using a lower extremity cycle ergometer.[2] The isokinetic exercise sessions consisted of exercise periods of 30 seconds followed by rest intervals of 60 seconds. During the 20-minute exercise session, unilateral shoulder exercise was performed for 10 minutes and unilateral knee exercise for 10 minutes. Exercise was performed at varied speeds ranging from 60[degrees] to 300[degrees]/sec. Peak SBP values were higher during isokinetic exercise compared with maximal values measured during cycling exercise. Peak HR values recorded during isokinetic exercise averaged 84% of maximal HR values for cycling exercise.

Greer et al[3] documented HR and BP responses to unilateral elbow flexion and extension performed on an isokinetic dynamometer. The protocol consisted of one exercise period, with subjects performing 10 repetitions in a 15-second period. When subjects exerted maximal effort, average increases in SBP, DBP, and HR were 27 mm Hg, 20 mm Hg, and 21 bpm, respectively. The duration of exercise was considerably shorter and a smaller muscle mass was exercised in the study of Greer et al,[3] as compared with the studies of Vinson et al[1] and Negus et al.[2] Although Vinson et al[1] reported no change in DBP, an increase in DBP was reported by Greer et al.[3] In addition, the increases in HR and SBP were less than those reported by Vinson et al[1] and Negus et al.[2]

Because isokinetic trunk ergometers are fairly new, limited information is available on the cardiovascular responses to this form of exercise. In the only study identified,[4] a prototype of a commercially available trunk flexion-extension ergometer was used. Heart rate and BP responses were measured in six healthy subjects and six subjects with low back dysfunction. Subjects participated in four exercise periods with resting intervals of varied duration between the exercise periods. Subjects performed a set number of repetitions at four speeds, ranging from 60[degrees] to 150[degrees]/sec. The duration of each exercise period varied from approximately 10 to 30 seconds. Peak HR and SBP averaged 167 bpm and 159 mm Hg, respectively, for the healthy subjects and 136 bpm and 145 mm Hg, respectively, for the subjects with low back dysfunction. The lower responses for the subjects with low back abnormalities were attributed to a lower work output, possibly because of pain and deconditioning.

We originally hypothesized that isokinetic trunk exercise would produce SBP responses of a similar or greater magnitude than those typically seen with isokinetic extremity exercise. This hypothesis was based on the results of a study of cardiovascular responses to isometric exercise, which showed a direct relationship between size of active muscle mass and BP.[5] We also hypothesized that BP responses would be higher during exercise at slower speeds compared with faster speeds. The muscle tension generated during exercise at slower speeds is usually higher than tension generated when exercising at faster speeds. As muscles generate high levels of tension, blood flow to those muscles is impaired. A compensatory mechanism to maintain blood flow is an increase in arterial pressure. Therefore, the purposes of this study were 1) to document the amount of work performed (ie, HR and BP) during a typical isokinetic trunk exercise session and 2) to determine the differences in work performed (ie, HR and BP) during isokinetic trunk exercise performed at speeds of 60[degrees], 90[degrees], and 120[degrees]/sec.

Method

Subjects

Twelve subjects (6 male, 6 female) volunteered to participate in the study. The subjects had normal resting HR and BP levels, were within 20% of ideal body weight, and were able to perform isokinetic trunk exercise without discomfort. Subjects who could perform the selected exercise without experiencing low back pain were chosen to represent the population that uses this form of exercise in our clinic. The purpose of performing isokinetic trunk exercise is to increase the strength and endurance of trunk muscles. We believe that prior to initiating this form of exercise training, patients should demonstrate normal movement patterns in trunk flexion and extension and be able to perform the activity without pain.

Several of the subjects participated regularly in recreational sports, which included running, soccer, and softball. None of the subjects performed regular resistive trunk exercises. Subjects ranged in age from 22 to 44 years (X = 26.9; SD = 6.4). Body weight averaged 75.4 kg (SD = 10.8) for the male subjects and 57.8 kg (SD = 5.7) for the female subjects. Written informed consent was obtained from all subjects.

Experimental Design

The experimental protocol consisted of a repeated-measures design, with two within-subjects factors (ie, speed of exercise and time within exercise sessions). Each subject attended four exercise sessions, which were conducted on separate days. The first session served as an orientation and practice session. The following three sessions were test sessions, with tests performed in a random order at speeds of 60[degrees], 90[degrees], and 120[degrees]/sec. At the practice session, subjects performed trunk flexion and extension at one of the three speeds, which was randomly selected. The practice session was conducted using the same measures and procedures as in the three test sessions. The data collected during the practice session were used to determine reliability. During each test session, the subjects participated in five 30-second exercise periods, with 60 seconds of rest between each exercise period. Heart rate, BP, and peak work were measured for each exercise period.

Procedure

The study was performed at the Cybex Center for Education and Research, using the Cybex [registered trademark] Trunk Flexion/Extension Unit.* The device was calibrated at the beginning of the study and periodically throughout the study, using procedures recommended by the manufacturer. At the initial session, subjects were positioned on the dynamometer, and the machine settings were recorded to replicate the position for the following test sessions. Subjects were positioned in standing with a padded strap placed over the pelvis for stabilization. Positioning devices were placed posteriorly in the popliteal area and anteriorly above and below the knee joints. These devices placed the lower extremities in approximately 10 to 15 degrees of knee flexion. The standing platform of the ergometer was moved so that the axis for trunk flexion and extension was located at approximately the L5-S1 vertebral level. The axis was aligned with the spinal column at the level that corresponded to the top of the pelvis. Padded bars were positioned over the mid-scapular area and across the upper chest.

Heart rate was measured using a radiotelemetry monitor.** An elastic strap, containing two electrodes and a transmitter, was strapped to the subject's chest. The receiver, contained in a wrist-watch-type device, was attached to the upper chest stabilization device. Blood pressure was measured by auscultation using a standard adult BP cuff and aneroid manometer.*** (* Cybex, Div of Lumex Inc, 2100 Smithtown Ave, Ronkonkoma, NY 11779.

** Computer Instruments Corp, 100 Madison Ave, Hempstead, NY 11550.

*** JA Preston Corp, 60 Page Rd, Clifton, NJ 07012.)

At the beginning of each test session, body weight was measured and recorded. After the measurement devices were in place, the subject was positioned on the dynamometer, and resting HR and BP were measured.

One investigator (MJA) served as timer, gave instructions to begin and end exercise, and read the HR monitor at the end of the exercise period as the subject stopped moving. Subjects were instructed to push as hard and as fast as they could and to pace themselves so that they could continue for the 30-second period. Subjects moved from the standing position to approximately 80 degrees of trunk flexion and then returned to the starting position. Subjects were monitored to ensure that they moved through the 80-degree arc of motion during each repetition. The number of repetitions for each exercise period differed depending on the speed of testing. Subjects performed approximately 10, 16, and 22 repetitions during each 30-second exercise period for the test speeds of 60[degrees], 90[degrees], and 120[degrees]/sec, respectively. The number of repetitions varied slightly between subjects because of the variability in the time to reverse movements from flexion to extension and from extension to flexion. The variability in number of repetitions was never more than [plus or minus] 1 repetition per 30-second exercise period.

A second investigator (CP) measured BP. When performing pilot studies, we determined that it was impossible to accurately measure BP during exercise with either a manual or an automatic measurement system. The noise of the ergometer interfered with auscultation, and associated muscle activity interfered with using a transducer to sense the brachial pulse. Consequently, BP measurements were made during the initial 20 seconds of each 60-second recovery interval. The cuff was inflated during the final 5 seconds of each exercise period. As the subject returned to the vertical position, the cuff was deflated. Systolic blood pressure was obtained within the initial 10 seconds of the recovery period, and DBP was obtained within the initial 20 seconds of recovery. The first Korotkoff sound was used to estimate SBP, and the fourth Korotkoff sound was used to estimate DBP.

During the rest interval, peak work performed for trunk flexion and extension for the exercise period was measured from a bar graph displayed on the computer screen. The bar graph displayed work in foot-pounds (1 ft [times] lb = 1.356 N [times]m) for each repetition of flexion and for each repetition of extension. Peak work was defined as the sum of the highest values recorded for a flexion movement and an extension movement during the 30-second period.

During both exercise and recovery periods, subjects were monitored for signs or symptoms of exercise intolerance. Subjects were asked to report any feelings of discomfort or any unusual feelings. Subjects were observed for changes in skin color and for excessive sweating.

Reliability

Measurements of HR, BP, and peak work were made by the same investigators throughout the study. The data collected at the practice sessions were used to estimate the reliability of subject performance and measurement. Practice session data for each subject were matched with the test session data collected at the same speed. Reliability was estimated using intraclass correlation coefficients (ICC[3,1]).[6] This version of the ICC was selected because our interest was limited to the reliability of the two raters used in the study, rather than the reliability of other randomly selected raters. The values for peak work were used to determine whether subjects performed in a similar manner at the practice and test sessions. The ICC for peak work was .87. Heart rate and BP data for the practice and test sessions were used to estimate the reliability of these measurements. Intraclass correlation coefficients for HR, SBP, and DBP were .95, .60, and .64, respectively.

Data Analysis

Descriptive statistics were calculated for peak work, HR, SBP, and DBP at each time period of measurement (rest, exercise periods 1-5) for exercise at each of the three test speeds (60[degrees], 90[degrees], 120[degrees]/sec). An analysis of variance (ANOVA) was used to determine whether there were significant differences in the dependent variables (peak work, HR, SBP, DBP) resulting from the independent variables (time during exercise session, speed of exercise). For significant results of the ANOVA, Tukey's method of follow-up testing was used to determine specific pair-wise differences. An alpha level of .05 was used to determine statistical significance.

Results

Average peak work performed for the five exercise periods at the three test speeds is shown in Figure 2. The results of the ANOVA are described in Table 1. The interaction between speed and time and the main effect for time were not significant, but the main effect for speed of exercise was significant. Peak work was highest at the slowest speed and lowest at the fastest speed, with significant differences between the three speeds. As indicated by the nonsignificant effect of time, peak work performed did not change significantly from the beginning of exercise period 1 to the end of exercise period 5.

Means and standard deviations for HR, SBP, and DBP at each test speed for rest and each exercise period are presented in Table 2. Even though peak work differed between the three test speeds, the interaction between speed and time and the main effect for speed were not significant for HR, SBP, or DBP.

There was a progressive and significant increase in HR during exercise, as indicated by a significant main effect for time. Heart rate values increased from a resting average of 88 bpm to average values ranging from 144 bpm to 153 bpm at the end of exercise period 5. The main effect for time during exercise is graphically shown in Figure 3. Follow-up testing revealed significant increases from rest to exercise period 1, from exercise period 1 to exercise period 3, and from exercise period 2 to exercise period 5.

As shown in Table 2, SBP values at rest and during exercise were similar. The speed x time interaction and the main effects of both speed and time were not significant. Diastolic blood pressure values decreased from resting levels to the first exercise period. This change was confirmed by a significant main effect for time. Follow-up testing revealed a significant decrease in DBP from rest to exercise period 1, with no significant differences between values for exercise periods 1 through 5. A summary of BP values during the exercise sessions is shown in Figure 4, indicating the main effect for time for DBP.

Discussion

The results on the reliability of peak work performed indicate that the subjects were fairly consistent in their performance when tests were conducted within a 1-week period. The reliability estimate may have been higher if we could have used a value for average work performed per repetition during each 30-second exercise period rather than peak work. The highest work value-peak work-was selected because this value could be accurately measured from the computer screen. A computer-generated value for average work probably would have been a more stable estimate, but was not possible to obtain with the measurement system used in our study.

The reliability was good for the measurement of HR, but only moderate for the measurement of BP. Part of the variability in HR and BP responses from one exercise session to another could be attributed to differences in subject performance as indicated by peak work performed. The lower reliability estimates for BP compared with HR may be attributed to a higher measurement error. The measurement error associated with our method of measuring HR has been estimated to be [plus or minus] bpm.[7] Sources of error when measuring BP include an inappropriate rate of cuff deflation and sound artifact created by upper extremity movements and environmental noise. We attempted to minimize these sources of error by adjusting the rate of cuff deflation according to the HR and by manually supporting the upper extremity during the measurement. Two sources of error that could not be eliminated were upper extremity isometric activity and noise generated by the ergometer. In addition, the study did not include an evaluation of the ergometer to verify its ability to provide constant-velocity motion.

The inverse relationship between peak work performed and speed of exercise was expected. When exercising at a slow speed (ie, 60[degrees]/sec) compared with a faster speed (ie, 120[degrees]/sec), the additional time allows muscles to generate higher levels of tension and produce higher torque and work output values.[8] With many forms of exercise, the cardiovascular responses are directly related to the work load performed.[9] Although peak work differed between exercise at the three speeds, there were no significant differences in the HR and BP responses. Similar responses may be explained by the constant duration of exercise for each test speed and by similar subjective levels of exertion. Although we did not ask subjects to rate perceived exertion during exercise, they appeared to exert their best efforts during the three tests. Levels of perceived exertion correlate closely with HR during aerobic or endurance activities.[10] Additional research of the relationships among speed of movement, work performed, and cardiovascular responses, using a set number of repetitions, rather than time, as an independent variable, is needed.

The increases in HR were similar to those reported for other studies involving lower extremity dynamic resistive exercise.[1,2] For our sample, the average peak HR for the first exercise period was 65% of predicted maximal HR. By the fifth exercise period, average peak HR had increased to 77% of predicted maximal HR. Although the peak HR responses were similar to values recommended to elicit cardiovascular training adaptations, the elevations were not sustained during the exercise sessions. During the 1-minute rest intervals, HR decreased considerably, often approximating baseline values.

The nonsignificant changes in SBP were surprising. Studies of isokinetic exercise of unilateral knee flexion and extension, performed for 30- to 50-second periods, have documented SBP values similar to those occurring during maximal cycling and treadmill exercise.[1,2] A study that documented cardiovascular responses to isokinetic trunk exercise performed for periods of 15 to 20 seconds also demonstrated increases in SBP.[4] Our results may be partially explained by the fact that the BP values were obtained during the initial 20 seconds after exercise, rather than during exercise. We believe that BP values recorded immediately after exercise are an accurate reflection of the values occurring during exercise. Most of the subjects who participated in this isokinetic trunk exercise study also participated in a similar study of knee flexion and extension. Systolic blood pressure values for these subjects that were recorded immediately after 30-second periods of unilateral isokinetic knee exercise were similar to those reported in the literature.[1,2]

Factors that affect SBP include total peripheral resistance and cardiac output. Increases in SBP with exercise usually result from increases in cardiac output.[11] Total peripheral resistance typically decreases with dynamic exercise because of vasodilation in large areas of active muscle.[11] We can speculate that cardiac output increased with isokinetic trunk exercise, as indicated by the increases in HR. The movement performed involved a large muscle mass, which included not only trunk flexors and extensors but also hip flexors and extensors. Vasodilation of this large muscle mass would most likely produce a decrease in total peripheral resistance. An increase in cardiac output that is similar to a decrease in peripheral resistance would produce no changes in SBP.

A factor to consider in comparing the results of our study with the results obtained by Stray-Gundersen et al[4] in their study of isokinetic trunk exercise response is the design of the ergometers. The prototype used in the previous study[4] provided a different method of lower extremity stabilization. Subjects were positioned in their normal standing postures, with one strap positioned below the knees. in our study, subjects were positioned in slight knee flexion, with the extremities securely stabilized with positioning devices above and below the knees. The positioning devices were adequate to allow subjects to remain in the upright position with slight knee flexion for extended periods. Differences in the amount of lower extremity stabilization between the two ergometers may produce differences in the amount of lower extremity muscle activity required to maintain the standing position. We can speculate that differences in associated muscle activity between exercising on the two ergometers may produce differences in venous return and consequently affect SBP responses during exercise.

During the recovery intervals, several of the subjects expressed feelings of dizziness and light-headedness. In addition to these symptoms, two subjects experienced nausea and were unable to complete the final exercise period. After these subjects were assisted from the ergometer to a supine position, the SBP values for both subjects were between 80 and 90 mm Hg. The symptoms occurred during recovery rather than during exercise. During exercise, lower extremity muscles are probably partially active, whereas during recovery, these muscles tend to relax. Although we did not record BP during the recovery intervals, we speculate that BP may decrease gradually during this period.

Clinical implications

Our original concerns were that the cardiovascular responses would be exaggerated for isokinetic trunk exercise and that activities of this nature might be unsafe for persons with cardiac disease. The results of our study demonstrate that exercising in this manner may be unsafe for a different reason. Based on the BP responses and symptoms noted in our subjects, we believe that during rest intervals patients should be observed closely and the stabilization devices should be loosened to allow lower extremity muscle activity. The specific exercise protocol used in this study resembles an exercise training session more than a test session. In a clinical setting, patients often are supervised less closely during exercise training compared with during testing. Our results indicate that observation and supervision are important during training sessions.

Conclusions

Cardiovascular responses to isokinetic trunk exercise performed for five 30-second exercise periods with 1-minute rest intervals included 1) a progressive increase in HR during exercise, 2) nonsignificant changes in SPB, and 3) a significant decrease in DBP from rest to the first period of exercise. Peak work performed differed between three test speeds, but the cardiovascular responses were similar. Subjective complaints of dizziness, light-headedness, and nausea occurred in some subjects during the rest intervals. The results suggest that patients should be observed for signs and symptoms of exercise intolerance when performing this type of exercise training.

References

1. Vinson S, Lindsay B, Mettler P, Murray T. Cardiovascular response to isokinetic endurance exercise. Phys Ther. 1987;67:777. Abstract.

2. Negus RA, Rippe JM, Freedson P, Michaels J. Heart rate, blood pressure, and oxygen consumption during orthopedic rehabilitation exercise. Journal of Orthopaedic and Sports Physical Therapy. 1987;8:346-350.

3. Greer M, Dimick S, Bums S. Heart rate and blood pressure response to several methods of strength training. Phys Ther. 1984;64: 179-183.

4. Stray-Gundersen J, Snell PG, Smith SS, et al. The systemic and myocardial oxygen demand associated with an isokinetic trunk testing protocol. Med Sci Sports Exerc. 1985; 17:207. Abstract.

5. Mitchell JH, Payne FC III, Saltin B, Schibye B. The role of muscle mass in the cardiovascular response to static contraction. J Physiol Lond. 1980;309:45-54.

6. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-426.

7. Leger L, Thivierge M. Heart rate monitors: validity, stability and functionality. The Physician and Sportsmedicine. May 1988;16:143-151.

8. Smith S, Mayer TG, Gatchel RJ, Becker TJ. Quantification of lumbar function, part 1: isometric and multispeed isokinetic trunk strength measures in sagittal and axial planes in normal subjects. Spine. 1985;10:757-764.

9. Astrand PO, Rodahl K. Textbook of Work Physiology. 2nd ed. New York, NY: McGraw-Hill Book Co; 1977.

10. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2-3:92-98.

11. Smith EE, Guyton AC, Manning RD, White RJ. Integrated mechanisms of cardiovascular response and control during exercise in the normal human. Prog Cardiovasc Dis. 1976;18:421-443.