Progressive exercise testing in closed head-injured subjects: comparison of exercise apparatus in assessment of a physical conditioning program.

Progressive Exercise Testing in Closed Head-Injured Subjects: Comparison of Exercise Apparatus in Assessment of a Physical Conditioning Program In healthy sedentary subjects, physical conditioning programs result in marked changes in the cardiorespiratory responses to exercise without changes in exercise efficiency.[1-4] Exercise testing on a bicycle ergometer has revealed a decrease in exercise tolerance in acute and chronic head-injured adults.[5] It is unclear whether this impairment in exercise performance is due to underlying neuromuscular deficits or to physical deconditioning. Two purposes of our study were 1) to determine whether exercise performance in chronic closed head-injured subjects improves after a physical conditioning program and 2) to determine whether an improvement in exercise performance is associated with improvement in either exercise efficiency or cardiorespiratory fitness. Another purpose of our study was to determine the most suitable type of exercise apparatus for assessing exercise performance in closed head-injured adults. Progressive exercise tests were performed on a treadmill, a bicycle ergometer, and mechanical stairs before and after a physical conditioning program.

Although many studies [1,6-14] have examined healthy subjects' cardiorespiratory response to walking, running, and cycling, few studies have examined cardiorespiratory performance during stair climbing. The paucity of information about physiologic functions during stair climbing has been largely due to the lack of a stationary apparatus for stair climbing comparable to the treadmill for walking or to the bicycle ergometer for cycling. Recently, stationary stair-climbing devices, which allow a subect to climb stairs at varying rates (ie, stepping speeds) while remaining in a fixed position, have become commercially available for use in rehabilitation and physical conditioning programs.

A stair-climbing ergometer (*1) has been used to study stair climbing in healthy subjects. [12-14] Verstraete and Ben-Ezra (R Verstraete, V Ben-Ezra; personal communication) found that exercise testing on the treadmill, bicycle ergometer, and mechanical stairs yielded comparable values of maximal oxygen consumption ([Vo.sub.2max]). The physiologic response to stair climbing has not been studied in closed head-injured subjects.

Method

Subjects

Twelve individuals with closed head injuries (3 women, 9 men), between 26 and 48 years of age (X = 31), participated in the study. The subjects were selected from consecutive admissions to The Transitional Learning Community (Galveston, Tex). All subjects were at least 1 year postin-jury. Table 1 shows the specific type of problem the subjects incurred as well as their specific residual neuro-logical deficits.

The residual deficiencies noted in Table 1 were documented by a licensed physical therapist through physical examination and use of past medical records. For the purposes of this study, motor slowing and incoordination were defined as problems of abnormal timing observed during performance of gross motor tasks and were evaluated through use of timed rapid alternating movements of the upper and lower limbs. For the purposes of this study, 12 or fewer movements in a 10-second period was categorized as slow. This performance represents 50% or less of that achieved by healthy subjects of similar age in our laboratory. Spasticity was determined by presence of hyperactive deep tendon reflexes, ankle clonus, and limb reaction to manual stretch. All subject ambulated independent of lateral support. Three subjects wore dynamic ankle-foot orthoses. Despite their locomotor defects, all subjects were capable of performing progressive exercise tests on the treadmill, bicycle ergometer, and mechanical stairs with minimal to no assistance. For the purposes of this study, minimal assistance was operationally defined as follows: 1) providing a manual cue to the subject's back in order that the walking position in the upper third of the treadmill would be maintained, 2) manually assisting a subject's foot to seat fully on a stair to avoid tripping, and 3) assisting the subject's foot to seat in the toe clip of the bicycle. The assistance was provided only if the subject was in danger of slipping out of position on the testing apparatus. This assistance was provided for three or four subjects and only at the highest levels of power output. None of the subjects had a medical history of respiratory diseases, gross anatomic chest or spinal deformities (eg, kyphosis, scoliosis), or neurologic respiratory muscular defects (eg, paralytic diaphragm). Informed consent was obtained from each subject.

Procedure

Each subject performed 1-minute progressive exercise tests on the treadmill, (*2) the bicycle ergometer, (*3) and the mechanical stairs (StairMaster) upon entry into a 1-year residential transitional rehabilitation program. The tests were repeated approxmately 3 months later following participation in a physical conditioning program. The mean time period between the two sets of evaluations was 12 weeks (s = 2). During the initial and 3-month evaluations, the three exercise tests were performed at random, at least 2 days apart, over a 2-week period. The test-retest repeatability of response to exercise on the same apparatus was not determined; however, all subjects were familiar with the use of the equipment, having performed on each apparatus several times prior to initial testing. The protocols used for testing on each apparatus are shown in Figure 1. At the beginning of all exercise tests, baseline recordings at rest were obtained for 3 minutes. The treadmill protocol consisted of having the subjects walk at a fixed speed of 1.8, 2.5, 3.0, or 3.5 mph, depending on their physical condition. The initial grade of 1% was progressively increased by 2% increments each minute to a maximum grade of 25%. If the subject reached the maximum grade, the grade then remained at 25% while the speed was progressively increased each minute by 0.5 mph. the bicycle ergometer protocol consisted of progressive increases in power output of 150 kpm/min at a cycling speed of 60 rpm. The seat level on the bicycle ergometer remained constant for all tests for each subject. During stair climbing, the stepping speed began at approximately 25 steps/min and was progressively increased by 6 to 8 steps/min. Recovery from exercise was monitored for 4 minutes with the subject performing at a power output equal to the initial power output during the progressive exercise.

The actual stepping speed during stair climbing was a function of the resistance placed on the mechanical gears of the apparatus and was regulated by a brake lever. The progressive release of the brake lever was measured with a metal ruler mounted parallel to the brake lever track. The initial brake lever position was set at the 20-mm mark and increased 5 mm each minute.

The speed of stair climbing at the different lever positions for individuals of different body weights was determined in a separate series of experiments. Ten healthy subjects with body weights ranging from 43 to 100 kg exercised on the mechanical stairs at lever positions of 20, 30, 40, 50, and 60 mm. Stepping speed was counted at each lever position. Over this weight range, no differences were apparent in stepping speed at a given lever position. Furthermore, the relationship between stepping speed and lever position was linear. Therefore, we believed that it was appropriate at any given lever position to use the mean stepping speed of the healthy subjects as the stepping speed for the closed head-injured subjects.

During all exercise tests, the electrocardiograph (*3) was continuously monitored with standard limb leads, and the ECG results were used for the determination of heart rate (HR). Exercise testing was terminated when the subject reached the maximum predicted HR (220 bpm -- age) or was unable to continue because of exhaustion. Each subject's electrocardiogram was monitored for signs of ischemia (eg, ST-segment depression, T-wave changes) and arrhythmias; however, no subjects displayed these findings during exercise testing.

During exercise testing, subjects inhaled room air through a mouth-piece attached to a low-resistance 2-way valve. (*4) Expired air was collected into a metabolic cart (*5) for measurement of oxygen consumption ([Vo.sub.2]), minute ventilation ([V.sub.E]), tidal volume ([V.sub.T]), respiratory rate (RR), and oxygen pulse. The oxygen pulse was used as an indirect measure of cardiac stroke volume, where oxygen pulse equals [Vo.sub.2]/HR. [3] Calibrations of carbon dioxide and oxygen were performed with known concentrations of fixed gases with a confidence limit of [+ or -] 0.01%. Volume measurements were calibrated with a 3.0-L syringe.

Following the completion of the initial progressive exercise test, each subject received an initial physical therapy assessment within the 2-week evaluation period. After the evaluation period, each subject was enrolled in a physical conditioning program consisting of a 50-minute exercise protocol performed 5 days a week. The specific content of the conditioning program for each individual was based on the evaluation data and was selected from the major categories of flexibility activities, muscle performance activities, and aerobic activities.

Table 2 summarizes the specific activities from which each program was selected. The 50-minute protocol was divided into 50% aerobic activities, 20% flexibility activities, and 30% muscle performance activities. This pattern was followed by all subjects each day, and the order was randomized by the subject. Compliance was monitored by on-site physical therapists and their assistants as well as by written documentation kept by the subjects.

All subject performed all flexibility tasks, selected their choice of equipment for muscle performance tasks, and combined outdoor walking with two additional aerobic tasks (ie, exercise on the treadmill, bicycle ergometer, or mechanical stairs). Each subject selected the additional aerobic tasks in random order, rotating among the three tasks throughout each week. The rotation continued for the entire study period. The level of activity during the aerobic tasks was designed to maintain HR at 60% to 80% of the maximum predicted HR for the individual.

Data Analysis

The data were analyzed to compare exercise performance on the three testing devices and to determine the effect of the 3-month physical conditioning program on exercise performance. The subjects' [Vo.sub.2], HR, [V.sub.E], RR, [V.sub.T], and oxygen pulse were analyzed at each of the three stages (rest, [Vo.sub.2 max], and recovery) as a two-factor factorial experiment with repeated measures on both factors. Power output was analyzed in the same way only at [Vo.sub.2 max]. The factors were defined as equipment (treadmill, bicycle ergometer, and mechanical stairs) and physical conditioning (prior to and after). Effects and interactions were tested at the .05 level of significance.

To determine whether the physical conditioning program was associated with a change in the efficiency of exercise, the estimated slopes of the [Vo.sub.2]-power output regression data were computed and analyzed as a 2-factor factorial experiment with repeated measures on both factors. The factors were defined as equipment (treadmill, bicycle ergometer, and mechanical stairs) and physical condition (before and after 3-month physical conditioning program). The Duncan multiple-range test was used for mean separation. Effects and interactions were tested at the .05 level of significance. [15]

The power output on the treadmill was calculated as

wt X (sin *) X B

where wt is the subject's body weight (in kilograms), * is the grade of the treadmill (in percent), and B is the treadmill speed (in meters per minute). [16] Power output on the mechanical stairs was calculated as

ht X SS X wt

where ht is the height of one stair (in meters) and SS is the stepping speed (in steps per minute). [1] Power output readings on the bicycle ergometer were verified using standard calibration techniques provided by the manufacturer.

The Metropolitan height-weight tables were used to calculate ideal body weight (IBW) of each subject. [17] As determined by a Student's paired t test, the weight of the subjects remained the same between the initial and 3-month tests (p [is greater than] .05).

Results

Figure 2 compares [Vo.sub.2 max] (in liters per minute), maximum HR, and maximum [V.sub.E] (in liters per minute) on initial exercise testing on the treadmill, bicycle ergometer, and mechanical stairs for all subjects. On both the initial and 3-month exercise tests, [Vo.sub.2 max] was significantly greater on the treadmill and on the mechanical stairs than on the bicycle ergometer. Maximum HR was significantly greater on the mechanical stairs than on the bicycle ergometer. Maximum [V.sub.E] was significantly greater on the mechanical stairs and on the bicycle ergometer than on the treadmill.

Figure 3 compares maximum power output of [Vo.sub.2 max] on the initial and 3-month exercise tests on the treadmill, bicycle ergometer, and mechanical stairs for all subjects. On each apparatus, the maximum power output and [Vo.sub.2 max] were significantly greater at the 3-month test than at the initial test. Using a formula based on height and age for prediction of [Vo.sub.2 max] for treadmill exercise in sedentary men and women, [18,19] the predicted [Vo.sub.2 max] for each of the 12 subjects was calculated. A comparison of predicted and actual performance of [Vo.sub.2 maxh is provided in Table 3. The mean of the predicted [Vo.sub.2 max] levels for the entire group was 2,859 mL/min (s = 617), whereas the actual mean of the group when tested on the treadmill was 2,134 mL/min (s = 510), only 74% of the mean predicted level. This percentage rose to 85% of the predicted [Vo.sub.2 max] after participation in the 3-month physical conditioning program (Tab 3).

Figure 4 compares [Vo.sub.2]/kg (in milliliters per minute per kilogram of body weight) at selected stepping speeds for all subjects while stair climbing before and following the 3-month physical conditioning program. At any given stepping speed, comparison of [Vo.sub.2]/kg values from the initial and 3-month tests shows no apparent differences. The estimated slopes of the [Vo.sub.2]-power output relationships on the treadmill, bicycle ergometer, and mechanical stairs were determined prior to and following physical conditioning. No significant differences in the estimated slopes on a particular apparatus were found before and after physical conditioning, suggesting that exercise efficiency remained constant.

For the group as a whole, HR at rest and on the fourth minute of recovery from exercise on a given apparatus decreased by an average of 7 bpm after 3 months of physical conditioning (p [is less than] .05), whereas oxygen pulse and [Vo.sub.2] remained unchanged. On a given apparatus, maximum HR was unchanged on the initial and 3-month exercise tests; however, maximum oxygen pulse was significantly increased.

At rest and during the fourth minute of recovery, [V.sub.E], RR, and [V.sub.T] were unchanged after 3 months of physical conditioning. On a given apparatus, maximum [V.sub.E] was significantly increased on the 3-month test. The increase in maximum [V.sub.E] was due to an increase in [V.sub.T].

Figure 5 shows HR and [Vo.sub.2] at rest and at a stepping speed of 25 steps/min in two subjects who were 176% and 133% of IBW and in an age- and height-matched subject weighing 85% of IBW. The two overweight subjects had higher resting HRs and reached nearly maximum predicted HR at this slow stepping speed. Both subjects had to stop exercising at this stepping speed because of exhaustion. Power outputs for the overweight subjects climbing at 25 steps/min were calculated to be 495 and 400 kpm/min, respectively. In comparison, the 85% IBW subject's power output at 25 steps/min was only 274 kpm/min.

Discussion

One purpose of our study was to compare exercise performance on the treadmill, bicycle ergometer, and mechanical stairs in closed head-injured subjects. Most subjects were able to exercise on the mechanical stairs without difficulty; however, the two overweight subjects developed dangerously high HRs at very low stepping speeds. The only other notable problems occurred in thos subjects with severe ataxia or spasticity of lower limb muscles. These subjects usually required minimal assistance during exercise (eg, help directing or guiding a foot to the next step).

The results suggest that maximal exercise performance in the closed head-injured subject is more accurately assessed using a treadmill or mechanical stairs than a bicycle ergometer. On either the initial or 3-month exercise tests, [Vo.sub.2 max] was significantly greater on the treadmill and on the mechanical stairs than on the bicycle ergometer. In general, there were no statistically significant differences for a given variable during maximal exercise on the treadmill and mechanical stairs, except for maximum [V.sub.E].

The results of this study agree with those reports for healthy subjects (R Verstraete, V Ben-Ezra; personal communication). [7,13] Progressive exercise tests performed on the treadmill and mechanical stairs by 38 fire fighters yielded comparable values of [Vo.sub.2 max], maximum [V.sub.E], and maximum HR (R Verstraete, V Ben-Ezra; personal communication). [13] Respective values of [Vo.sub.2 max], maximum [V.sub.E], and maximum HR on the mechanical stairs were 7%, 3.5%, and 2% less than values obtained on the treadmill. In another study, [7] healthy subjects demonstrated higher [Vo.sub.2 max] values on the treadmill than on the bicycle ergometer. This finding may have been due to the fact that a smaller muscle mass was used during cycling than during walking or running.

The lower [Vo.sub.2 max] on the bicycle ergometer when compared with that on the treadmill or the mechanical stairs in the closed head-injured subjects may also have been due to two other factors. Many closed head-injured subjects have difficulty with initiation, attention, memory, and concentration. When exercising on the bicycle ergometer, the subjects could stop pedaling whenever they desired, and most required a great deal of verbal encouragement to maintain a constant cycling speed. In contrast, rate of exercise was not controlled by the subjects while performing on the treadmill or mechanical stairs. The inability of subjects to directly control the rate and duration of exercise on the treadmill and mechanical stairs may explain the higher [Vo.sub.2 max] values obtained on these two types of equipment. In addition, those subjects with hemiplegia may have predominantly used the unaffected extremity to achieve given power outputs on the bicycle ergometer. This excessive mechanical load may have led to early fatigue of the less involved limb and a cessation fo exercise. The number of subjects with hemiplegia in this study was too small to determine the validity of this latter hypothesis.

Maximal oxygen consumption in the closed head-injured subjects was 74% of the predicted value on the initial exercise tests. The decreased [Vo.sub.2 max] indicates a marked imairment in physical fitness. These findings are supported by previous investigators who report a higher HR at rest and at given power outputs, as well as a lower oxygen pulse at given power outputs in head-injured subjects than in healthy controls. [5] The impairment in physical fitness on the initial exercise testing might be attributable to underlying neuromuscular abnormalities or to physical deconditioning. [18-20] The underlying neuromuscular abnormalities could include not only impairments in muscular strength and coordination but also abnormal brain-stem function regulating cardiovascular and pulmonary reflexes. The significant increase in [Vo.sub.2 max] on the 3-month testing, however, reveals the presence of a large reversible component to this exercise limitation. The failure to detect a change in exercise efficiency between the initial and 3-month tests indicates that this improvement in exercise performance resulted primarily from physical conditioning.

Clinical Implications

The results of progressive exercise tests can be used by physical therapists to establish a safe work load for persons needing a conditioning program. A retest, at a specified time after the intervention with a physical activity program, can be used to validate the specific program contents. The results of this study indicate that progressive exercise tests in closed head-injured subjects can be used to 1) quantify the degree of physical deconditioning, 2) help devise the level of physical activity during a physical conditioning program, and 3) assess the outcome of such a program.

Conclusions

Following participation in a 3-month physical conditioning program, the [Vo.sub.2 max] of the closed head-injured subjects rose from 74% to 85% of the predicted value. The improvement in [Vo.sub.2 max] reveals the presence of a reversible component to the impairment in physical fitness. This increase in physical fitness may have resulted from an improvement in exercise efficiency or cardiorespiratory fitness. Comparison of [Vo.sub.2] at given power outputs on a given apparatus on the initial and 3-month tests showed no statistically significant differences, suggesting that exercise efficiency had not changed. In contrast, the exercise tests performed after 3 months of physical conditioning were associated with significant reductions in HR at rest and after 4 minutes of recovery from exercise, along with significant increases in maximum oxygen pulse. These results indicate that the improvement in physical fitness was associated with an improvement in cardiovascular function. Similar results have been described in healthy sedentary subjects who participated in a physical conditioning program. [20]

Finally, in our 12 closed head-injured subjects, 2 were 176% and 133% of IBW. We noted that these two subjects had difficulty performing on the mechanical stairs. They approached maximum predicted Hrs at very low stepping speeds. During stair climbing, power output is directly proportional to body weight. [1] In overweight subjects, even low stepping speeds can result in dangerously high power outputs, rapidly resulting in exhaustion. The use of stair climbing as a part of a physical conditioning program in overweight subjects should be carefully monitored.

Further research in exercise testing of persons with noncardiopulmonary pathological conditions is needed. Future studies might explore 1) whether the achieved level of physical fitness is retained when closed head-injured subjects assume the independent living phase of their rehabilitation program, 2) the effects of asymmetrical lower limb muscle performance on bicycle pedaling, and 3) the effect of body weight on the physiologic response to stair climbing.

C*1) StairMaster Southwest, PO Box 6574, Tulsa, OK 74156.

(*2) Universal, Subsidiary of KIDDE, PO Box 1270, Cedar Rapids, IA 52406.

(*3) Schwinn Bicycle Co, 1856 N Kostner Ave, Chicago, IL 60639.

(*4) Hewlett-Packard, PO Box 42816, 10535 Harwin St, Houston, TX 77036.

(*5) Hans Rudolph, 7200 Wyandotte, Kansas City, MO 64114.

(*6) Model 9000, Gould Inc, 3631 Perkins Ave, Cleveland, OH 44114.

References

[1] Astrand P, Rodahl K. Textbook of Work Physiology. New York, NY: McGraw-Hill Book Co; 1977

[2] Shepard RJ. Physiology and Biochemistry of Exercise. New York, NY: Praeger Publishers; 1982:371-383

[3] Wasserman K, Whipp BJ. Exercise physiology in health and disease. Am rev Respir Dis. 1975;112:219-249

[4] Whipp BJ, Wasserman K. Efficiency of muscular work. J Appl Physiol. 1969;26:644-648

[5] Becker E, Bar-Or O, Mendleson L, et al. Pulmonary functions and responses to exercise of patients following cranio-cerebral injury. Scand J Rebabil Med. 1978;10:47-50

[6] Shepard RJ. the relative merits of the step test, bicycle ergometer, and treadmill in the assessment of cardio-respiratory fitness. Arbeit-physiologie. 1966;23:219-230

[7] Hermansen L, Saltin B. Oxygen uptake during maximal treadmill and bicycle exercise. J Appl Physiol. 1969;26:31-37

[8] Kamon E, Pandolf KB. Maximal aerobic power during laddermill climbing, uphill running, and cycling. J Apply Physiol. 1972;32:467-473

[9] Nagle FJ, Balke B, Baptista G, et al. Comparability of progressive treadmill, bicycle and step test based on oxygen uptake responses. Med Sci Sports. 1971;3:149-154

[10] Swinberg CR, Waefield JM, Jones PW. Performance, ventilation, and oxygen consumption in three different types of exercise test in patients with chronic obstructive lung disease. Thorax. 1985;40:581-586

[11] Johnson AN, Cooper DE, Edwards RHT. Exertion of stair climbing in normal subjects and in patients with chronic obstructive bronchitis. Thorax. 1977;32:711-716

[12] Verstraete R, Ben-Ezra V. Submaximal cardiovascular response to positive stair climbing and treadmill walking in cardiac patients. Med Sci Sports Exerc. 1987;19:S57

[13] Ben-Ezra V, Verstraete R. Step ergometry: task specific training for fire fighters. Med Sci Sports Exerc. 1987;19:S62

[14] rosentswieg J, Verstraete D, Bassett G. Stair climbing as a training modality for women. Med Sci Sports Exerc. 1986;18:S28

[15] Steel RGD, Torrie JH. Principles and Procedures of Statistics. New York, NY: McGrawHill Inc; 1960

[16] Mathews DK, Fox EL. The Physiological Basis of Physical Education and Athletics. Philadelphia, Pa: WB Saunders Co; 1971:46-48

[17] 1983 Metropolitan height and weight tables. Statistical Bulletin Metropolitan Insurance Company. 1984;64:2-9

[18] Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and normographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J. 1973;85:546-562

[19] Taylor HL, Buskirk E, Henschel A. Maximal oxygen intake as an objective measure of cardio-respiratory performance. J Appl Physiol. 1955;8:73-80

[20] Davis JA, Frank MH, Whipp BJ, et al. Anaerobic threshold alterations caused by endurance training in middle-aged men. J Appl Physiol. 1979;46:1039-1946

M Hunter, MD, is a pulmonary physician, 1612 E Cardwell, Brownfield, TX 79316. He was a Pulmonary Fellow in the Department of Internal Medicine, The University of Texas Medical Branch, Galveston, TX, when this study was conducted.

J Tomberlin, MA, is Director of Physical Therapy Services, The Transitional Learning Community, PO Box 1528, Galveston, TX 77550.

C Kirkikis, BA, is Clinical Coordinator, Gastroenterology Division, Louisiana State University Medical Center, 1440 Canal St, New Orleans, LA 70112. She was Clinical Coordinator, Pulmonary Division, The University of Texas Medical Branch, Galveston, TX, when this study was conducted.

S Kuna, MD, is Associate Professor, Pulmonary Division (E-61), Departments of Internal Medicine, The University of Texas Medical Branch, Galveston, TX 77550 (USA). Address all correspondence to Dr Kuna.