Journal Logo

Clinical Sciences: Clinical Investigations

Changes in physical strain and physical capacity in men with spinal cord injuries


Author Information
Medicine & Science in Sports & Exercise: May 1996 - Volume 28 - Issue 5 - p 551-559
  • Free


As a result of a lower-body dysfunctioning, individuals with spinal cord injuries (SCI) depend on their upper-body during almost all activities of daily living (ADL). It is suggested that individuals with long-standing SCI consequently have a relatively inactive lifestyle(4), resulting, as in the able-bodied population, in a decreased physical capacity and in a concomitant increased risk of medical complications, such as cardiovascular disease and obesity(10,12).

A decreased physical capacity in combination with the use of a relatively small upper-body muscle mass may increase the physical strain during ADL. The high physical strain during ADL (13,17) may cause fatigue and discomfort which, together with the high incidence of upper-extremity injuries in wheelchair users (8,21), reduces the possibilities and incentive to be more active. Several authors suggest that wheelchair users with SCI are at high risk of ending up in this debilitative cycle (12,14). Based on cross-sectional data, Sawka et al. (23) estimated the decrement in ˙VO2peak in healthy active wheelchair users with various disabilities to be approximately 3 ml·kg-1·min-1 per decade of life, which is similar to the decrement (4 ml·kg-1·min-1) reported among able-bodied persons (5). It should be noted here that the suggested decline in ˙VO2peak may be too optimistic, since Sawka et al. (23) only compared healthy active students with healthy domiciliary wheelchair users.

The suggested debilitative cycle may be aggravated by factors that restrict the activity level or influence the physical condition, such as a higher age, obesity, upper-extremity injuries, periods of illness, or rehospitalization(14). In addition, the lesion level determines the mobility to an important extent. Persons with quadriplegia may lead a more inactive life than those with paraplegia due to a lower physical capacity(16) and a concomitant restricted mobility, leading to a larger decline in physical capacity.

Maintaining or improving the physical capacity may impede or even reverse this debilitative cycle. However, the physical strain of normal ADL in individuals with SCI was found to be insufficient in magnitude or duration to maintain or improve the physical capacity (13,17). In contrast, training regimens have been shown to increase the physical capacity of persons with SCI and to reduce the strain during submaximal arm crank exercise or wheelchair ambulation (12,15). Several authors assert that a reduced physical capacity would also coincide with higher strain levels during other ADL (11,14). In a previous cross-sectional study it was shown that the physical strain during standardized ADL was indeed inversely related to the physical capacity(18). Until now, no longitudinal studies have investigated changes in physical capacity in relation to changes in strain during ADL among persons with SCI.

The first purpose of this study, therefore, was to verify the hypothesis that persons with SCI demonstrate a significant decline in physical capacity over a period of 3 yr. Based on the above-mentioned cross-sectional data, a decline of circa 1 ml·kg-1·min-1 could be expected in this time period. Activity-restricting factors could possibly increase this amount considerably. The relatively short period of 3 yr was chosen as a first part of a continuing study, which will continue to evaluate changes over future years. The second purpose was to investigate whether personal factors(age, lesion level, time since injury, body mass, skinfolds), behavioral factors (activity level, smoking), and health status were related to the change in physical capacity. The third goal of this study was to investigate whether the hypothesis that longitudinal changes in physical capacity coincide with inverse changes in physical strain during ADL could be confirmed.


Experimental Design

Subjects were tested on two occasions: the first test (T1) was conducted during the first half of 1990 and the second (T2) between January and May 1993, with a mean time of 34.5 ± 1.5 months between tests. On each occasion, the subjects completed an interviewer-administered questionnaire eliciting personal, behavioral, and health status factors. They were also evaluated on body mass and skinfold thicknesses and performed a series of ADL tasks between 10 a.m. and 3 p.m., followed by a maximal wheelchair exercise test at least 2 h after a light lunch and after a 1-h rest. Subjects were asked to refrain from smoking and ingestion of caffeine and alcohol at least 2 h prior to each testing session.


At T1, 44 male wheelchair users with nonacute SCI volunteered to participate after having signed an informed consent statement. Three subjects were not able to perform the tests at T2 due to medical problems (pressure sores, cirrhosis, and epicondylitis medialis, respectively). Three refused to participate again and one subject could not be located. The final number of subjects used in the analyses, therefore, was 37, 23 of whom had a complete lesion. All subjects lived independently (more or less) at home and used a handrim wheelchair as their usual way of locomotion. The subjects were grouped into subjects with quadriplegia (lesion level: C4-C8; N = 8), with high-level paraplegia (T1-T5; N = 5), with mid-level paraplegia(T6-T10; N = 10), and with low-level paraplegia (T11-L5; N= 14). No significant differences in age and time since injury (TSI) were found among the four subject groups (Table 1).

Personal and Behavioral Factors and Health Status

Body mass was determined on a Berkel scale (type 3000) with the subject sitting in light clothing. Skinfolds were determined at the triceps, biceps, suprailiac, and subscapular sites and summed (Σ 4SF).

A questionnaire was used to assess personal characteristics, type of wheelchair, activity level (defined as the hours of weekly active-sport participation), and smoking behavior (smoking or nonsmoking). In addition, subjects were asked to indicate whether they had suffered from an illness during the year before T2, whether they had complaints concerning shoulders, arms, hands, back, or neck, whether they had suffered from pressure sores or urinary tract infections, and whether they had been rehospitalized during the year before T2. To exclude minor illnesses such as colds, illnesses were only included if subjects had visited a physician and/or had been absent from employment or sport participation due to the illness. All these (dichotomous: yes = 1, no = 0) factors were considered to be potentially activity-restricting. An overall health status classification was made summing the occurrence of illnesses and complaints: those subjects with no (group 1), with 1 or 2 (group 2), or with more than two complaints or illnesses or a chronic disorder (group 3). Although this classification included the incidence of pressure sores, urinary tract complications, and rehospitalization during the year preceding T2, these factors were also included in the analyses as separate variables. Additionally, the changes in body mass, in Σ 4SF, in hours of weekly sport participation, and in health status classification (possible range: -2 to +2) between T1 and T2 were determined and used as separate variables.

Physical Capacity

To assess physical capacity, all subjects performed an identical graded maximal wheelchair exercise test in their daily-use wheelchair on a motor-driven treadmill (Enraf Nonius, model 3446, belt width 1.25 m, length 3.0 m) at T1 and T2. A more detailed description of this test and results at T1 have been described previously (16). In short, the test consisted of 3-min exercise periods followed by 2-min active rest periods, during which the workload was reduced. The power output (PO) during the first exercise period equaled the rolling resistance (Fd) times the belt velocity. During every next exercise period, PO was increased by imposing an additional resisting force (Fad) on the wheelchair by means of a pulley system, whereas the exercise velocity was held constant throughout the test.

Both at T1 and T2, Fd was determined in a drag test according to Woude et al. (27). Although the mean Fd at T1 (7.8± 3.0 N, range 1.9-17.3 N) was not significantly different from that at T2 (8.1 ± 3.8 N, range 3.1-15.2 N), the individual Fd could have changed (either reduced or increased) substantially at T2 due to various causes (change in wheelchair characteristics or body mass, replaced wheelchair, etc.). To correct for these possible changes in Fd, Fad was adjusted on the second exercise period at T2 to achieve the same PO. Since increments in PO were equal for the tests at T1 and T2, the PO levels for each individual were identical from exercise period 2 on.

During the exercise periods, expired gas was collected and analyzed with an Oxycon Ox4 (Mijnhardt, The Netherlands) with a 30-s sample period. The Oxycon, which was calibrated before each session with reference gas mixtures, determined the oxygen consumption (˙VO2; STPD). Parameters of physical capacity were ˙VO2peak' defined as the highest 30-s value during the test, and POmax, defined as the PO level when˙VO2peak was attained, and were expressed both in absolute values and relative to body mass. Heart rate (HR) was continuously recorded with a Sport Tester PE3000 (Polar Electro, Finland), a light-weight telemetric HR monitor. Every 5 s the Sport Tester calculated an average HR on a time-based algorithm using the last 15 R-R-intervals. HRmax was defined as the highest HR recorded during the test.

The ˙VO2peak of one subject was not determined during T1 due to technical problems, whereas another subject (age 71 yr) did not perform the maximal exercise test at T2 for safety reasons (recent history of ischemic infarction).

Physical Strain During ADL Tasks

At T1, physical strain was estimated while subjects performed a series of standardized tasks, such as making transfers, negotiating obstacles, and doing household tasks, the results of which were described previously(18). At T2, subjects performed a selection of these tasks in the same order: transfers from a wheelchair to a toilet, a shower wheelchair, and a shower seat, and the negotiating of a curb (0.08 m and 0.12 m). The time to complete the tasks was the same for T1 and T2.

The subjects were free in the way of performing the task and free to use some assistive devices (hinged support bars, stirrup grips, and grab rails) at T1. The way of performance (the use of assistive devices, movement order) at T1 was recorded on videotape and the subjects were asked to perform, if possible, the task at T2 in the same way as at T1.

During these tasks, HR was recorded with the Sport Tester with a 5-s storing interval. A previous study (19) showed that HR responses to these tasks were reproducible in these subjects. Physical strain was defined as the highest HR above resting HR provoked by the task, expressed relative to the individual heart rate reserve (HRR). The HRR, defined as the difference between the lowest HR at rest (while sitting) and the maximal HR determined during the graded exercise test, was determined at both T1 and T2 and was not significantly different (P > 0.05; pairedt-test) between T1 (mean HRR: 109 ± 24 bpm) and T2 (108± 26 bpm). To isolate the HR response to a certain task, a resting period of at least 2 min before and after each task was imposed.


The longitudinal change in physical capacity was determined by subtracting the values for ˙VO2peak and POmax at T1 from those at T2. Subjects were classified as having either an improved or a reduced physical capacity if the (absolute) subtracted value exceeded 4.3 W (POmax) or 0.11 l·min-1 (˙VO2peak). These values denote the standard error of measurement (SEM) of these parameters determined previously with the same test mode and protocol and 12 of the subjects from the present study (20). A Student's paired t-test compared physical capacity, physical strain during the tasks, body mass, Σ 4SF, and activity level at T1 and T2. Stepwise multiple regression analyses determined the contributions to changes in physical capacity made by the independent variables body mass, Σ 4SF, activity level, health status(and the possible changes of these factors), age, lesion level (each lesion level was assigned an arbitrary number from 1 to 22, C4 being 1 and L5 being 22), TSI, smoking behavior (smoking = 1, nonsmoking = 0), and the incidence of pressure sores, urinary tract infections, and rehospitalizations during 1 yr before T2.

Pearson correlations were determined to assess the relation between changes in physical capacity and changes in physical strain during the ADL tasks. If these relations were significant (one-tailed), a regression equation was calculated to describe the relation. All results were considered significant at P < 0.05.


Personal and Behavioral Factors and Health Status

Mean body mass increased significantly between T1 and T2(Table 1). Twenty-two subjects (59%) showed an increase in body mass of more than 2 kg, while only five displayed a decrease of more than 2 kg. The increase in body mass coincided with a significant mean increase inΣ 4SF of 20 mm. The average hours of sport participation were not significantly different between T1 and T2 for the whole subject group, whereas the subjects with quadriplegia showed a significant increase. Twenty subjects had a different wheelchair at T2.

The number of subjects with pressure sores, urinary tract complications, and rehospitalizations during 1 yr before T2 was respectively eight, seven, and eight. Almost half (17 = 46%) of all subjects had more than two complaints or illnesses during the year preceding T1, whereas only 12 had no complaints or illnesses. At T2, the number of subjects without complaints or illnesses decreased to 10, while the number of subjects with more than two complaints remained unchanged. The number of smokers was similar at T1 (N = 13) and T2 (N = 14).

Change in Physical Capacity

Twelve subjects (34%) showed an improvement (>0.11 l·min-1) whereas six showed a decrease in absolute˙VO2peak. More than half (21 = 60%) of all subjects improved POmax (>4.3 W), while in only five subjects (14%) a serious decline was observed. The largest increase in ˙VO2peak (from 2.2 to 3.1 l·min-1) was achieved by a young (16 yr at T1) man with paraplegia who did not participate in sport activities at T1 while at T2 he was involved in a weekly strength training of 4-5 h. The largest decrease(from 2.6 to 2.3 l·min-1) was seen in a subject with paraplegia who was a member of the Dutch national wheelchair basketball team at T1 but who reduced his sports activities considerably after T1. A dramatic decrease(from 60.7 to 43.4 W) in POmax was also observed in a man with high-level paraplegia who was forced to reduce his sport participation(wheeling, wheelchair tennis) from 2 h at T1 to 0 at T2 as a result of upper-extremity complaints.

All lesion groups had higher values for POmax (W) at T2 than at T1, which was significant for the group with quadriplegia, for those with mid-level paraplegia, and for the whole group (Table 1). However, these differences were no longer significant when expressed relative to body mass. ˙VO2peak(l·min-1) showed a similar tendency, but the difference between T2 and T1 was not significant. The group with high-level paraplegia showed a significant reduced ˙VO2peak, whereas the other groups tended to have increased values.

The regression analysis, performed to predict the change in˙VO2peak (Δ˙VO2peak) and in POmax(ΔPOmax), indicated that the activity level was the most important predictor of absolute Δ˙VO2peak (R2 = 0.22): 1-h sport participation was associated with a 0.03 l·min-1 increase (Table 2). After statistically adjusting for activity level, TSI accounted for an additional 10% of the variance. A weak relation was observed between TSI and relative Δ˙VO2peak: each year was related to a 0.16 ml·kg-1·min-1 decrease (Table 2). This relation was probably caused by five recently (<2 yr at T1) injured subjects who showed a large increase in˙VO2peak. Including only subjects with less recent injuries in the analyses resulted in a nonsignificant relation. Subjects who had been rehospitalized in the year preceding T2 had on average a 12 W less positive change in POmax than nonrehospitalized subjects. Additionally, a 1-kg increase in body mass was associated with a 0.02 W·kg-1 reduction in POmax.

Physical Strain During ADL Tasks

Due to the inability of some subjects at T1 and T2 and in some cases lack of time at T1, not all tasks were performed by all subjects, resulting in a varying number of subjects used in the analyses. For example, 11 subjects were not able to ascend both curbs.

The average change in physical strain during the tasks varied from a significant decrease of 8.5% HRR (negotiating the 8-cm curb) to a significant increase of 7% HRR (transfer to the toilet) (Table 3). The strain during all other tasks showed no significant changes. An important result is the occurrence of large individual changes (in both directions) in strain between T1 and T2, ranging from a reduction of 33% HRR during the transfer to the shower seat to an increase of 40% HRR during the transfer to the toilet.

A noticeable decline was seen in one subject with quadriplegia who was able to ascend the 8-cm curb and perform all transfers independently at T1, but who was not able to perform any task anymore at T2. In contrast, two other subjects with quadriplegia improved their functional ability and were able to perform almost all transfers without help at T2.

Relationship Change in Physical Capacity and Physical Strain

The change in physical strain during the ADL tasks showed inverse relations with Δ˙VO2peak and ΔPOmax(Table 4), which was significant for the relation betweenΔ˙VO2peak(ml·kg-1·min-1) and change in strain during the transfers to the shower wheelchair and shower seat and the 0.08-m curb ascent and between ΔPOmax(W·kg-1) and change in strain during the 0.08-m curb ascent.

Table 5 presents the significant regression equations describing the change in strain associated with Δ˙VO2peak andΔPOmax. A decline of 1 ml·kg-1·min-1 in ˙VO2peak or a 0.1 W·kg-1 reduction in POmax was related to a mean increase of 1.7-2.6% HRR in strain during all tasks.

The subject with quadriplegia who could no longer perform any task at T2 showed a decrease of 5 W (from 30.2 to 25.5 W: 16%) in POmax and a decrease of 0.11 l·min-1 (from 0.96 to 0.85 l·min-1: 11%) in ˙VO2peak. In contrast, the subjects who improved functional ability and were able to perform almost all tasks without help at T2 displayed increases in POmax of 10 W (from 18.3 to 28.3 W: 54%) and 8 W (from 31.0 to 39.1 W: 26%), respectively.


Longitudinal Change in Physical Capacity

Based on cross-sectional data, Sawka et al. (23) estimated the decrement in ˙VO2peak in healthy wheelchair users with various disabilities to be about 3 ml·kg-1·min-1 per decade, which is similar to values reported among able-bodied persons (5). However, the study by Sawka et al. (23) included only healthy active subjects, which may have influenced the relations studied due to a positive selection. It could be expected that a longitudinal study with a more heterogeneous (including less healthy and less active persons) group of persons would show more pronounced decrements in physical capacity. However, the hypothesized decrease in physical capacity could not be observed after the 3-yr period among the subjects of the present study, which may have been due to the relatively short period and the rather low mean age (37 yr). In addition, very small decrements might not be observable due to confounding factors and measurement resolution. Nonetheless, at least a tendency toward a decrease could have been expected. On the contrary, the results show a tendency toward an increased physical capacity, being significant for POmax (W).

The increase in POmax and the tendency toward an increase in˙VO2peak may be the result of an increased active muscle mass. However, the change in POmax was no longer significant when expressed relative to body mass. A similar tendency was seen for the change in˙VO2peak. Changes in physical capacity relative to body mass not only reflect changes in absolute physical capacity, but also in body composition and the relative amount of tissue that can be used to perform work. Hence, the significant increase in absolute POmax combined with the nonsignificant change in relative POmax suggest that the increase in body mass was at least to a certain extent the result of an expanded amount of nonexercising tissue (adipose tissue). This is supported by the large increase in skinfolds.

It was hypothesized that the group subjects with quadriplegia would show a larger decrement in physical capacity than those with paraplegia due to a more restricted mobility, and consequently a more inactive lifestyle. This hypothesis could not be confirmed, which suggests that persons with quadriplegia are equally capable of maintaining or even improving their physical capacity. In fact, the group with quadriplegia was the only group that showed a significant increase in the amount of sport participation(predominantly due to the rising popularity of quad rugby, a relatively new team sport designed for persons with quadriplegia), which may have prevented a decline in physical capacity. However, a significant decline in˙VO2peak was observed in the group with highlevel paraplegia, which may have been due to a tendency toward a lower sport participation in this group compared to the other groups.

Activity level accounted for 22% of the variance inΔ˙VO2peak, indicating that those with a higher activity level were more likely to maintain or improve ˙VO2peak than inactive individuals, which is in agreement with studies among able-bodied persons(5). Cross-sectional studies support, in part, the relation found in the present study: sedentary persons with SCI have been found to have lower levels of physical capacity than more active persons with SCI (3,7,30). Although the differences in physical capacity may be the result of the different activity level, it may also be that fitter persons are more likely to participate in sports activities. However, several studies demonstrate that training regimens can improve the physical capacity of both persons with paraplegia and those with quadriplegia (6,10,22,24).

The physical capacity was inversely related with the time since injury. However, this appeared to be primarily due to some subjects who had their SCI less than 2 yr at T1. Five of these nine subjects showed essential improvements in physical capacity. Major progress in physical capacity can be expected after the acute phase of injury. Moreover, the functional ability of persons with SCI seems to improve especially during the first 4 yr after discharge from a rehabilitation hospital (28,29). These large improvements during the first few years after rehabilitation may in part reflect the adaptation of the upper-body to the new situation while taking over the mobility functions of the lower-body. After this period, large improvements may predominantly be established through physical exercise.

Individuals with SCI have been shown to be more prone to medical complications than the able-bodied population. High incidences of pressure sores, urinary tract complications, cardiorespiratory problems, and upper-extremity injuries have been reported among persons with SCI(2,8,25). In agreement, an important part of our subject group had been rehospitalized (N = 8) and had pressure sores or urinary tract complications during the year preceding T2, whereas only 10 subjects had not experienced any serious medical problem (including chronic problems) the year prior to T2. The results of this study may even represent an optimistic view, since the three subjects who dropped out due to medical problems were not considered. In addition, the subject group at T1 may have been a positive selection: those with severe medical problems may have been less willing or were unable to volunteer in this study. Moreover, only individuals capable of manually propelling their wheelchair were included.

The incidence of medical problems that result in prolonged periods of bed rest (such as pressure sores) or inactivity (upper-extremity complications) can cause important reductions in physical capacity. Although the incidence of pressure sores and urinary tract complications were not significantly independently related to Δ˙VO2peak or ΔPOmax, those with severe medical complications resulting in rehospitalization had on average a 12-W less positive change in POmax, which underlines the important influence of medical complications on the maintenance of physical capacity among individuals with SCI and stresses the need for more research to prevent medical secondary complications in these individuals.

Twenty of our subjects had a different wheelchair at T2. The mean rolling resistance of these 20 wheelchairs, however, was not significantly (pairedt-test; P > 0.05) different between T1 (7.4 ± 3.4 N) and T2 (6.9 ± 2.9 N). In addition, the resisting force added to the wheelchair adjusted for individual changes in rolling resistance, resulting in identical (submaximal) PO levels at T1 and T2. However, differences in wheelchair configuration may have altered the fitting of the wheelchair to the user, which can affect the efficiency of propulsion(26), and concomitantly POmax. This could explain the significant increase in POmax (W) without a significant change in˙VO2peak. Although this confounding factor could have been avoided by using a standardized wheelchair, we decided to let the subjects use their daily-use wheelchairs, which would provide us with more accurate information on their POmax during daily life. In contrast to POmax,˙VO2peak is, within certain limits, independent of the wheelchair configuration or the exercise mode used, and was consequently not influenced by the changed wheelchairs.

POmax may also have been enhanced due to an improved propulsion technique as a result of a continuing adjustment of the upper-extremities to the task they were not made for: ambulation. An improved propulsion technique may lead to an increased efficiency. Cross-sectional studies have shown that wheelchair-dependent individuals display significantly higher mechanical efficiency values than able-bodied persons (1), suggesting that efficiency might be improved by habituation. To investigate whether our subjects had an improved propulsion efficiency at T2, the gross efficiency(GE) during submaximal (at 62.2 ± 5.5%˙VO2peak at T1) levels of the graded exercise test was determined at identical PO levels during T1 and T2. The mean GE was higher (P < 0.01) at T2 (10.0 ± 2.5%) than at T1 (9.3 ± 2.8%), connoting that an increased GE may have accounted, at least in part, for the significant improvement in POmax(W) without a significant change in ˙VO2peak. In addition, a possibly augmented contribution of the anaerobic energy system may also have enhanced POmax without altering the ˙VO2peak.

In spite of a relatively small and very heterogeneous subject group, 32-42% of the variance in Δ˙VO2peak and ΔPOmax could be explained by the independent factors used in this study. One explanation for these percentages not being higher may be that activity level was defined as the self-reported average hours of sport activities per week. The intensity of sport activities, which may play the most important role in the maintenance of the physical capacity, was not considered in the analyses. Moreover, normal daily activity, either at work or during leisure time, was not included. Although it has been suggested that physical strain during daily activities of persons with SCI is insufficient in magnitude or duration to maintain or improve the physical capacity (13,17), it might be possible that even lower mean strain levels in combination with the high strain during specific tasks (17) contribute to the maintenance of the physical capacity in those persons with a (very) low physical capacity. Additional factors that may have influenced the relations studied are, among other things, seasonal influences and medical complications just prior to T1 or T2.

Relation Between Change in Physical Capacity and Change in Physical Strain

The results of the present study show an inverse relation between change in physical capacity and change in physical strain during a number of the tasks studied, which confirms the hypothesis often posed in the literature(11,14). The relations only reached significance when physical capacity was expressed relative to body mass. It was already mentioned that this parameter not only reflects changes in absolute physical capacity, but also changes in nonexercising tissue (e.g., adipose tissue). The significant increase in body mass and skinfold thicknesses indicates that the workload of activities during which height differences have to be overcome(negotiating a curb) or the whole body has to be lifted and moved (making transfers) has increased. Hence, a combination of changes in physical capacity and changes in body mass is a better predictor of physical strain than the single parameters. The implication is that not only physical capacity should be maintained or improved, but that special attention has to be given to the avoidance or reduction of excessive body mass to avert physical strain during ADL from becoming too high. It has often been mentioned by individuals with long-standing SCI that body mass gains indeed had made their performance of ADL more difficult and that this hindered their maintenance of independence(2,9).

The regression equations revealed that rather large improvements are necessary to induce substantial reductions in strain during the tasks studied: a large increase of 5 ml·kg-1·min-1 in˙VO2peak (observed in some subjects) was associated with a reduction of 8% HRR in strain during the transfer to the shower wheelchair. Hence, it may be that minor changes in physical capacity do not alter the ability to perform ADL tasks of those with normal to high capacities. However, this study shows that even small changes in POmax could be associated with drastic changes in functional ability of those with a low POmax(15-40 W): an improvement of 5-10 W can result in total independence during ADL, whereas a decrease of 5-10 W can result in the opposite: total dependency. Not only persons with quadriplegia are at increased risk of becoming more dependent with time, but also older persons, which is supported by surveys showing that these groups were in greater need of help(9). Since the life expectancy of individuals with SCI has increased considerably, special attention has to be given to the older individuals (especially with high lesions) to prevent serious decrements in their mobility and freedom of motion.

Although an inverse relation between the change in physical capacity and strain during the standardized tasks was obvious, the correlation coefficients were less strong than expected and not always significant, which may have been due to the relatively small and very heterogeneous subject group. Another explanation may be that not only physical capacity determined during wheelchair propulsion regulates the strain during a task, but also factors such as technique of task execution (movement order, speed, use of assistive devices), the occurrence of muscle spasms, wheelchair characteristics, and anthropometrics. In addition, the inclusion of subjects with incomplete lesions may have disturbed the relations studied (18). Physical capacity as determined in the present study only reflects the upper-body capacity, while some subjects with incomplete lesions may be capable of using (parts of) their lower-body to perform the tasks. Moreover, the external workload during a task such as making a transfer is, in contrast to, for example, wheelchair propulsion or arm cranking, difficult to determine and even more difficult to control. Although the subjects were informed about the manner in which they had executed the task at T1 and were asked to execute the task in that way at T2, differences in performance were inevitable. As a consequence, changes in upper-body capacity will not always coincide with inverse changes in physical strain.


In summary, this study shows that the hypothesized decrement in physical capacity of individuals with SCI did not occur after the 3-yr period. Although the period of study was relatively short, at least a tendency toward a decrease in physical capacity could be expected. On the contrary, a tendency toward an increased physical capacity was found, which was even significant for POmax. The individual change in physical capacity was positively associated with the activity level and inversely related with time since injury, an increase in body mass, and the occurrence of rehospitalization. The change in physical capacity, expressed relative to body mass, was related with inverse changes in physical strain while making transfers and negotiating curbs. Hence, maintenance or improvement of physical capacity, which can in part be achieved through regular sport participation and improved medical care, together with avoidance or reduction of excessive body mass, may prevent restrictions in mobility and stimulate a more independent lifestyle.


1. Brown, D. D., R. G. Knowlton, J. Hamill, T. L. Schneider, and R. K. Hetzler. Physiological and biomechanical differences between wheelchair-dependent and able-bodied subjects during wheelchair ergometry.Eur. J. Appl. Physiol. 60:179-182, 1990.
2. Cushman, L. A. and J. Hassett. Spinal cord injury: 10 and 15 years after. Paraplegia 30:690-696, 1992.
3. Davis, G. M. and R. J. Shephard. Cardiorespiratory fitness in highly active versus inactive paraplegics. Med. Sci. Sports Exerc. 20:463-468, 1988.
4. Dearwater, S. R., R. E. Laporte, R. J. Robertson, G. Brenes, L. L. Adams, and D. Becker. Activity in the spinal cord-injured patient: an epidemiologic analysis of metabolic parameters. Med. Sci. Sports Exerc. 18:541-544, 1986.
5. Dehn, M. M. and R. A. Bruce. Longitudinal variation in maximal oxygen intake with age and activity. J. Appl. Physiol. 33:805-807, 1972.
6. DiCarlo, S. E., M. D. Supp, and H. C. Taylor. Effect of arm ergometry training on physical work capacity of individuals with spinal cord injuries. Phys. Ther. 63:1104-1107, 1983.
7. Eriksson, P., L. Löfström, and B. Ekblom. Aerobic power during maximal exercise in untrained and well-trained persons with quadriplegia and paraplegia. Scand. J. Rehabil. Med. 20:141-147, 1988.
8. Gellman, H., I. Sie, and R. L. Waters. Late complications of the weight-bearing upper extremity in the paraplegic patient. Clin. Orthop. Relat. Res. 233:132-135, 1988.
9. Gerhart, K. A., E. Bergstrom, S. W. Charlifue, R. R. Menter, and G. G. Whiteneck. Long-term spinal cord injury: functional changes over time. Arch. Phys. Med. Rehabil. 74:1030-1034, 1993.
10. Glaser, R. M. Arm exercise training for wheelchair users. Med. Sci. Sports Exerc. 21:S149-S157, 1989.
11. Glaser, R. M., M. N. Sawka, R. J. Durbin, D. M. Foley, and A. G. Suryaprasad. Exercise program for wheelchair activity. Am. J. Phys. Med. 60:67-75, 1981.
12. Hjeltnes, N. Capacity for physical work and training after spinal injuries and strokes. Scand. J. Soc. Med. Suppl. 29:245-251, 1982.
13. Hjeltnes, N. and Z. Vokac. Circulatory strain in every day life of paraplegics. Scand. J. Rehabil. Med. 11:67-73, 1979.
14. Hoffman, M. D. Cardiorespiratory fitness and training in quadriplegics and paraplegics. Sports Med. 3:312-330, 1986.
15. Hooker, S. P. and C. L. Wells. Effects of low- and moderate-intensity training in spinal cord-injured persons. Med. Sci. Sports Exerc. 21:18-22, 1989.
16. Janssen, T. W. J., C. A. J. M. van Oers, A. P. Hollander, H. E. J. Veeger, and L. H. V. van der Woude. Isometric strength, sprint power, and aerobic power in individuals with a spinal cord injury.Med. Sci. Sports Exerc. 25:863-870, 1993.
17. Janssen, T. W. J., C. A. J. M. van Oers, L. H. V. van der Woude, and A. P. Hollander. Physical strain in daily life of wheelchair users with spinal cord injuries. Med. Sci. Sports Exerc. 26:661-670, 1994.
18. Janssen, T. W. J., C. A. J. M. van Oers, H. E. J. Veeger, A. P. Hollander, L. H. V. van der Woude, and R. H. Rozendal. Relationship between physical capacity and physical strain during standardized tasks in men with spinal cord injuries. Paraplegia 32:844-859, 1994.
19. Janssen, T. W. J., C. A. J. M. van Oers, L. H. V. van der Woude, and A. P. Hollander. Reliability of heart rate responses to nonsteady-state activities of daily living in men with spinal cord injuries.Scand. J. Rehabil. Med. 26:71-78, 1994.
20. Janssen, T. W. J., C. A. J. M. van Oers, A. P. Hollander, and L. H. V. van der Woude. Reliability of responses to a maximal wheelchair exercise test in men with a spinal cord injury. In: Physical Strain and Physical Capacity in Men with Spinal Cord Injuries, T. W. J. Janssen, Ph.D. dissertation, Amsterdam: Vrije Universiteit, 1994, pp. 25-33.
21. Nichols, P. J. R., P. A. Norman, and J. R. Ennis. Wheelchair user's shoulder? Scand. J. Rehabil. Med. 11:29-32, 1979.
22. Pollock, M. L., H. S. Miller, A. C. Linnemd, E. Laughridge, E. Coleman, and E. Alexander. Arm pedaling as an endurance training regimen for the disabled. Arch. Phys. Med. Rehabil. 55:418-424, 1974.
23. Sawka, M. N., R. M. Glaser, L. L. Laubach, O. Al-Samkari, and A. G. Suryaprasad. Wheelchair exercise performance of the young, middle-aged, and elderly. J. Appl. Physiol. 50:824-828, 1981.
24. Sedlock, D. A., R. G. Knowlton, and P. I. Fitzgerald. The effects of arm crank training on the physiological responses to submaximal wheelchair ergometry. Eur. J. Appl. Physiol. 57:55-59, 1988.
25. Stotts, K. M. Health maintenance: paraplegic athletes and non-athletes. Arch. Phys. Med. Rehabil. 67:109-114, 1986.
26. van der Woude, L. H. V., H. E. J. Veeger, and R. H. Rozendal. Ergonomics of wheelchair propulsion: a prerequisite for optimum wheeling conditions. Adapt. Phys. Activ. Q. 6:109-132, 1989.
27. van der Woude, L. H. V., G. De Groot, A. P. Hollander, G. J. van Ingen Schenau, and R. H. Rozendal. Wheelchair ergonomics and physiological testing of prototypes. Ergonomics 29:1561-1573, 1986.
28. Welch, R. D., S. J. Lobley, S. B. O'Sullivan, and M. M. Freed. Functional independence in quadriplegia: critical levels. Arch. Phys. Med. Rehabil. 67:235-240, 1986.
29. Yarkony, G. M., E. J. Roth, A. W. Heinemann, L. Lovell, and Y. Wu. Functional skills after spinal cord injury rehabilitation: three-year longitudinal follow-up. Arch. Phys. Med. Rehabil. 69:111-114, 1988.
30. Zwiren, L. D. and O. Bar-Or. Responses to exercise of paraplegics who differ in conditioning level. Scand. J. Rehabil. Med. 9:107-113, 1975.


©1996The American College of Sports Medicine