- Email: [email protected]
Effect of seat cushion on dynamic stability in sitting during a reaching task in wheelchair users with paraplegia
274
PROSTHETICS/ORTHOTICS/DEVICES
Effect of Seat Cushion on Dynamic Stability in Sitting During a Reaching Task in Wheelchair Users With Paraplegia Rachid Aissaoui, PhD, Chantal Boucher, BSc, Daniel Bourbonnais, PhD, Miche`le Lacoste, OT, Jean Dansereau, PhD ABSTRACT. Aissaoui R, Boucher C, Bourbonnais D, Lacoste M, Dansereau J. Effect of seat cushion on dynamic stability in sitting during a reaching task in wheelchair users with paraplegia. Arch Phys Med Rehabil 2001;82:274-81.
© 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
Objectives: To examine the effects of seat cushions on dynamic stability in sitting during a controlled reaching task by wheelchair users with paraplegia. Design: A randomized, controlled test. Setting: Rehabilitation center. Participants: Nine wheelchair users with paraplegia. Interventions: Three types of cushions—an air flotation, a generic contoured, and a flat polyurethane foam—were tested during a controlled reaching task in ipsilateral and contralateral directions, at 45° from the sagittal plane in the anterolateral direction. Center of pressure (COP) coordinates were monitored by using a pressure measurement system as well as a force platform under seat. Main Outcome Measures: Trajectory of COP, maximal distance covered by COP, maximal velocity of COP; and the index of asymmetry between right and left maximal pressure under ischial tuberosities. Results: The generic contoured cushion allowed the COP to cover significantly (p ⬍ .02) a larger distance (81 ⫾ 28mm) when compared with the air flotation (63 ⫾ 25mm) or the flat foam (61 ⫾ 29mm) cushions. The COP velocity was significant ( p ⬍ .05) for the generic contoured cushion (.14 ⫾ .05m/s) versus the air flotation (.10 ⫾ .04m/s) or the flat-foam (.10 ⫾ .03m/s) cushions. The index of asymmetry was higher for the generic contoured and the flat foam cushions. During reaching, maximal pressure under ipsilateral ischial tuberosity was significantly higher for the flat foam (275 ⫾ 70mmHg) and the generic contoured (235 ⫾ 81mmHg) cushions, when compared with the air flotation cushion (143 ⫾ 51mmHg). Conclusion: Seat cushions can significantly affect sitting balance during reaching tasks. This study provided an objective method to assess the dynamic stability of wheelchair users when they perform activities of daily living requiring reaching. These findings have implications for wheelchair seating recommendations, especially seat cushion selection. Key Words: Paraplegia; Pressure; Rehabilitation; Wheelchairs.
YNAMIC SITTING REFERS to the continuous process of D postural changes during sitting. Sitting posture is usually unstable without additional external support because the hip
From the De´partement de Ge´nie Me´canique, E´cole Polytechnique de Montre´al (Aissaoui, Boucher, Lacoste, Dansereau), Research Center, Montreal Rehabilitation Institute (Bourbonnais), Montreal, Quebec, Canada. Accepted in revised form May 23, 2000. Supported by the Natural Sciences and Engineering Research Council of Canada. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the author(s) is/are associated. Reprint requests to Rachid Aissaoui, PhD, Chaire Industrielle CRSNG sur les aides techniques a¯ la posture, Dept de Ge´nie Me´canique, Ecole Polytechnique de Montreal, CP 6079, SVCL Centre-ville, Montreal, Que H3C 3A7, Canada. e-mail: Rachid. [email protected]. 0003-9993/01/8202-5983$35.00/0 doi:10.1053/apmr.2001.19473
Arch Phys Med Rehabil Vol 82, February 2001
joints are in an intermediate position with respect to the range of motion and because the trunk cannot be locked relative to the thighs by ligamentous restraint. As a result, muscle activity is necessary to maintain the trunk segment in an upright posture when sitting without additional stabilizers.1 The capacity to maintain balance and posture in sitting is a prerequisite to the activities of daily living (ADLs), and deficits in sitting balance control can severely limit task performance.2,3 Research related to the quantification of sitting stability has been oriented toward applications such as postural sway measurements4-7 and functional reach,3,8-13 as well as wheelchair stability.14-18 However, ways of measuring sitting stability vary between investigators and applications. Greater stability during quiet sitting has been associated with a smaller radius of motion at the level of the upper trunk and neck segment.4-7 In wheelchair seating, static stability has been measured by the angle at which the front or side wheel of the wheelchair loses contact with the ground on a tilting platform.15,16 During dynamic perturbation of a wheelchair user, a sagittal acceleration was imposed on paraplegic athletes and the stability of the wheelchair user was measured by a damping factor14 (the ratio between the measured and imposed acceleration to the trunk segment). Recently, Kamper et al18 imposed a lateral tilting perturbation to individuals with spinal cord injury (SCI) and measured the displacement of center of pressure (COP) at ground level. Kamper18 found that the maximal COP displacement tolerated was highly related to dynamic stability. As far as functional reach is concerned, however, the manner in which the body is stabilized while performing reaching tasks is not well understood. The influence of the lower legs in maintaining sitting balance during forward reaching movements has been investigated in healthy subjects, and it was found that thigh and foot support permitted larger forward excursions of the center of gravity.8,10 In terms of seating aids, a forward seat inclination has been reported to influence trunk posture during forward reaching.9 Curtis et al12 compared the effect of using a chest belt on the area of functional reach in the sagittal and transverse planes, and showed that the area of functional reach increased for subjects with paraplegia when using a chest belt. Allison and Singer19 have shown that, during a lateral reaching task, the lateral displacement of a subject’s COP increased significantly when using trunk orthoses. Dean and Shepherd3 evaluated the effect of a 2-week task-related training program for hemiplegic adults with the aim of increasing their reaching distance; they found that after training, subjects were able to reach faster and longer. In terms of body movement, it was found that this improvement in reaching represented a gain in terms of sitting stability. In fact, both the
275
DYNAMIC STABILITY IN SITTING, Aissaoui
displacement13 and the peak velocity20 of the COP in the anteroposterior direction were larger for able-bodied subjects during controlled bimanual forward reaching when compared with a group of individuals with SCI. Moreover, Seelen20 has shown that a clear improvement in controlling sitting posture corresponded to an increase of COP displacement in high and low SCI groups after a period of rehabilitation program. However, the effect of seat cushions on sitting balance has not been examined from a quantitative viewpoint in previous studies. In fact, in most of these studies, either a hard seat was used or the subject sat directly on a rigid force-platform. This lack of relevant studies can be explained by to the fact that it is difficult to estimate the COP location at the body-seat interface. The stability provided by seat cushions is considered to be the most important characteristic of sitting support after pressure distribution.17,21,22 The effect that cushions have on stability, though real or perceived, can be critical especially for users with poor trunk control. Persons with spinal lesions often have an acute awareness of their trunk stability that is far more subtle and complex than position measurement systems are able to detect.22 In a clinical setting, compromises between cushion stability and pressure distribution capacity often have to be made. This is usually performed in practice by accepting a small increase in the accepted pressure over bony areas.22 Furthermore, it is suggested that maintaining postural stability requires controlling both displacement and velocity of COP.18,23,24 The purpose of this study was to examine the effect of seat cushion on dynamic stability in sitting during a controlled reaching task performed by individuals with paraplegia. In this study we hypothesized that sitting balance is influenced by the seat cushion when performing reaching movement tasks, and that the seat cushion is considered stable when it allows the COP to move rapidly and to cover a larger distance. METHOD Subjects The subject population for this study included 9 wheelchair users with paraplegia by SCI. They were recruited from the Rehabilitation Institute of Montreal. To participate, subjects had to meet the following criteria: (1) diagnosis of paraplegia resulting in complete spinal lesion for at least 12 months; (2) no pressure ulcer for at least 1 year; (3) no orthopedic problem that would interfere with the ability to perform reaching task while seated; (4) the range of spinal lesion was limited to T3–L2 to
still have minimal trunk control12; (5) ability to give informed consent; and (6) ability to sit for a period of 1.5 hours. All of the subjects were right-handed. Their anthropometric characteristics as well as the clinical description of their seated posture are presented in table 1. All procedures were performed with the approval of the Rehabilitation Institute of Montreal’s Ethics Committee. Experimental Protocol Subjects sat on a stool (fig 1). Footrest height was adjusted to have the thigh segment horizontal. The feet were placed parallel, and the distance between the medial borders of the feet was about 15cm. The seat angle was fixed at 0°, while the seat to back angle was fixed at 95°. The distance between the popliteal fossea and the front edge of the seat was adjusted to 4cm. Arm length was measured by using a measuring tape, in a seated position with the upper limb extended in horizontal plane and pointing in a forward direction. The distance between the acromion process and the tip of the second finger was measured. The subject started with his/her right hand resting on a pressure-sensitive starting switch (button I) that was placed on the right side of the stool (fig 1). A first alarm indicated to the subject to prepare to reach, whereas a second alarm randomly activated within a maximum period of 3 seconds signaled the start of the hand movement. The subject reached a second switch (button II) in 2 directions: 45° ipsilateral and 45° contralateral from the sagittal plane (fig 1). No instructions about movement speed were given. Reaching distance was standardized to 130% of arm length, whereas the height of the switch button II was standardized to 100% of the glenohumeral joint height as measured from the floor in the seated posture. The subject maintained the switch button II pressed for at least 2 seconds and then released it to return finally to press the switch button I, which ended the trial. During reaching, the hand not used was placed on the opposite thigh segment to ensure stability. A belt was fixed on the lateral side of the seat support and placed on the thigh segment to secure the subject to the stool. The belt did not interfere with the reaching movement. Subjects performed 3 trials for each condition (left and right hand), and with 3 randomly selected seat cushions a 3-inch air-flotationa (ROHO), a 2-inch polyurethane HR35 flat-foamb (flat foam), and a 3-inch polyurethane HR45 generic contoured foam (ISCUS).b The air-flotation cushion is used by 51% of the
Table 1. Characteristics of Wheelchair Users
Subject
Gender
Age (yr)
1 2 3 4 5 6 7 8 9
Woman Man Man Man Man Man Woman Man Woman
39 25 20 65 59 42 52 47 46
Diagnosis
Arm-Length (cm)
Sitting Hip/ Thigh Width* (cm)
ISCUS Size (cm)
ROHO Pressure† (mmHg)
Pelvic Obliquity
Posterior Pelvic Tile
Transverse Pelvic Rotation
T5 paraplegia T12 paraplegia T8 paraplegia T12 paraplegia T3–T5 paraplegia T4–T5 paraplegia T4–T5 papaplegia T4–T6 papaplegia Postmyelitis paraplegia
66 83.5 76 76 78 70 66 71 70.5
30.5 30.5 32 40.5 43 37 33 35.5 43
38 38 38 40 45.5 38 35.5 35.5 45.5
45 34 36 46 40 30 24 45 32
No No No No Yes (R) Yes (L) Yes (R) Yes (L) Yes (L)
No No No Yes No No No Yes No
No Yes (R) Yes (L) Yes (L) Yes (R) No Yes (R) Yes (L) Yes (L)
Abbreviations: R, right; L, left. * The subject remained sitting, but raised his/her arms slightly and the horizontal distance across the widest part of the hips or thighs was measured by an anthropometer.f † The ROHO cushion was pressurized until a clearance of 1.5cm was reached under both right and left ischial tuberosity, which were located by palpation. Pressure under ROHO cushion was measured by a hand-held sphygmomanometer.g
Arch Phys Med Rehabil Vol 82, February 2001
276
DYNAMIC STABILITY IN SITTING, Aissaoui
Fig 1. Experimental setup. Schematic representation of subject seated on a stool, facing investigators (top); from above (bottom). Abbreviations: FSA, force sensing array.
spinal cord–injured population, whereas its use in the general wheelchair user population is actually around 30%.25 Garber and Dyerly25 reported that patients using air flotation cushion may feel unstable or experience difficulty transferring to another surface. In our study, 5 of the 9 subjects used the air flotation cushion as their regular cushion. The ability of the custom-contoured cushion to improve posture and balance, as well as to reduce pressure has already been shown.26-28 It should be noted that the contoured cushion used in this study had a generic concave shape with fixed maximal depth of 55mm with respect to the lateral edge of the cushion. This maximal depth corresponds approximately to the average contour of the SCI group in the recent study of Brienza and Karg.28 Arch Phys Med Rehabil Vol 82, February 2001
Instrumentation Pressure measurement system. During the reaching task, the pressure distribution at the body-seat interface was recorded at 5Hz by using a force sensing arrayc (FSA), which consists of 225 force sensors organized in a flexible mat of 15 ⫻ 15. Before each subject’s session, the FSA mat was calibrated up to its maximum of 300mmHg. The trajectory of the COP was computed from the pressure distribution data for each trial. A median low-pass filter was used to filter raw displacement data of COP over time. An interpolation by a cubic spline function was used on the filtered COP data to estimate the magnitude of the velocity vector by using a central
DYNAMIC STABILITY IN SITTING, Aissaoui
difference method. The maximal covered distance (MCD), by the COP, the maximal velocity (MV) of the COP, as well as the surface area under the curve defined by the velocity-distance relationship were computed (fig 2). An index of asymmetry between right and left maximal pressure under the ischial tuberosities regions was also calculated at each sample. The maximal value of the index of asymmetry (MIA) was computed for each reaching task trial.
277
Seat force platform. An AMTId force-platform was used to collect force and moment data under seat at 50Hz. Force data were filtered by means of a low-pass median filter. The COP coordinates under the seat were determined from the forceplatform data. For each trial, the MCD, MV, as well as the surface under the curve defined by the velocity-distance relationship, were estimated. During the experiment, the FSA mat and the AMTI were synchronized.
Fig 2. (A) COP trajectory of a typical subject during right hand reaching in the ipsilateral and contralateral side. (B) Phase diagram plot of the COP (instantaneous velocity in m/s versus instantaneous displacement mm). Abbreviations: AP, anteroposterior direction in mm; ML, mediolateral direction in mm; FF, flat foam.
Arch Phys Med Rehabil Vol 82, February 2001
278
DYNAMIC STABILITY IN SITTING, Aissaoui
Statistical Analysis For each condition the 3 trials were averaged for the selected dependent variables. The selected variables were the MCD, MV, the surface area in the forward reaching (SF) and, for the pressure data, the MIA, as well as the maximal pressure at right and left ischial tuberosity regions (max PR, max PL). Each dependent variable was analyzed separately, in an analysis of variance (ANOVA) (2 factors: cushion ⫻ direction) with repeated measures on both independent variables, to determine if significant differences existed between the cushions. If indicated, a Bonferroni test was performed to determine which pairs of means were significantly different (p ⬍ .05). STATISTICAe software was used to perform the statistical analysis. RESULTS All subjects were able to perform reaching tasks at 130% of their arm length except for 1 subject who was only able to reach at 115%. This subject has poor trunk control and his paraplegia was caused by myelitis. Figure 2A shows a COP trajectory for a typical subject during a reaching task with the right hand in the 2 directions and for 3 cushions. The COP coordinates were computed from pressure distribution data. There were 3 distinct COP trajectories for the 3 cushions in the ipsilateral direction, though this was less apparent in the contralateral direction (fig 2A). However, it is difficult to associate sitting stability with a specific COP trajectory. In the phase diagram in figure 2B, which represents the instantaneous velocity with respect to the cumulative distance covered by the COP, it is clear that the COP covers a larger distance with higher speed for the contoured foam, compared with either the flat-foam or the air flotation cushions. The pattern was similar in the 2 directions but with different amplitudes. In the ipsilateral direction, the upper limb is outside the body during reaching, contributing largely to the displacement of the total body center of mass, which is reflected by the large displacement of the COP. In general, the COP reached a MV at middistance of a reaching target. This was caused by the acceleration and deceleration process of total body center of mass. There was more fluctuation in the backward movement than in the forward. Thus, the MV, MCD, SF, as well as the MIA parameters were evaluated in the forward direction of the reaching movement only. In figure 2B, the COP covered a distance of approximately 130mm, and reached a velocity of 0.2m/s for the ISCUS cushion in the ipsilateral direction. These values represent approximately a gain of 38%, 100%, and 64%, respectively, for the distance covered, the velocity, and the SF parameters with respect to the ROHO cushion. The flat-foam cushion behaved
similarly to the ROHO cushion in term of distance, but stayed in between the ISCUS and the ROHO cushions in term of velocity. This trend was generally present for all subjects. COP Parameters From the FSA System Table 2 shows the mean average of the COP parameters as estimated from the FSA system for the right hand. During the forward motion of the reaching task with the right hand, the MCD parameter was larger for the generic contoured cushion (81 ⫾ 28mm) when compared with the air flotation (63 ⫾ 25 mm) and flat-foam (61 ⫾ 29mm) cushions. ANOVA revealed that there was a cushion effect (F ⫽ 3.99; df ⫽ 2,16; p ⬍ .05) and a direction effect (F ⫽ 18.65; df ⫽ 1,8; p ⬍ .01) when reaching with the right hand. There was significant interaction between the effects of cushion and direction on MCD (F ⫽ 6.24; df ⫽ 2,16; p ⬍ .01). The same separate effects caused by cushion and direction were present for the left hand, except that there was no interaction between them. The generic contoured cushion had a significant effect (p ⬍ .02) on the MCD parameter compared with the flat-foam and the air flotation cushions. The average MV of the COP was significantly higher (p ⬍ .05) for the generic contoured (.14 ⫾ .05m/s) than for the air flotation or the flat-foam cushions (.10 ⫾ .04m/s, .10 ⫾ .03 m/s, respectively). There were significant cushion effect (F ⫽ 6.97; df ⫽ 2,16; p ⬍ .01) and direction effect (F ⫽ 9.76; df ⫽ 1,8; p ⬍ .02) on the MV parameter because of the generic contoured cushion. Similar effects on the MV parameter were found for the left hand. There was a significant effect on the direction factor (p ⬍ .05) and a moderate effect (p ⬍ .06) on the cushion factor with respect to the SF parameter, when reaching with the right hand for the generic contoured cushion. In summary, for all of the MCD, MV, and SF parameters, the generic contoured cushion was significantly different from both the air flotation and the flat foam cushions. The foam cushions (the generic contoured and the flat) exhibited more asymmetry than the air flotation cushion. In fact, during the reach, the MIA reached .67 ⫾ .20 and .65 ⫾ .18, respectively, for the generic contoured and flat foam cushions in the ipsilateral direction, whereas it did not exceed .51 ⫾ .15 for the air flotation. The air flotation cushion had a significant effect (p ⬍ .02) on the MIA parameter. It should be noted that a MIA value of 0.5 means that the pressure under the right ischial tuberosity region is 3 times higher than the left one. Max PR was significantly higher for the flat foam cushion (275 ⫾ 70mmHg) and the generic contoured (235 ⫾ 81mmHg) compared with the air flotation cushion (143 ⫾ 51mmHg).
Table 2. COP Parameters From the FSA System for Right-Hand Reaching Air Flotation
Generic Contoured
Flat-Foam
Parameters
I
C
I
C
I
C
MCD (mm) MV (m/s) SF (mm ⫻ m/s) MIA max PR (mmHg) max PL (mmHg)
63 ⫾ 25 .10 ⫾ .04 4.1 ⫾ 2.8 .51 ⫾ .15* 143 ⫾ 51‡ 124 ⫾ 36†
31 ⫾ 18 .06 ⫾ .04 1.6 ⫾ 2.0 .47 ⫾ .18 140 ⫾ 62‡ 146 ⫾ 38†
81 ⫾ 28* .14 ⫾ .05† 7.0 ⫾ 5.3† .67 ⫾ .20 235 ⫾ 81 171 ⫾ 68
32 ⫾ 15 .07 ⫾ .03 1.8 ⫾ 1.6 .65 ⫾ .27 223 ⫾ 69 183 ⫾ 76
61 ⫾ 29 .10 ⫾ .03 3.9 ⫾ 3.0 .65 ⫾ .18 275 ⫾ 70 183 ⫾ 83
29 ⫾ 23 .06 ⫾ .04 1.7 ⫾ 2.3 .60 ⫾ .37 252 ⫾ 86 229 ⫾ 88
NOTE. Data presented as mean ⫾ standard deviation (SD). Abbreviations: I, ipsilateral; C, contralateral. * p ⬍ .02. † p ⬍ .05. ‡ p ⬍ .01.
Arch Phys Med Rehabil Vol 82, February 2001
279
DYNAMIC STABILITY IN SITTING, Aissaoui Table 3. COP Parameters From Seat Platform for Right Hand Reaching Air Flotation
MCD (mm) MV (m/s) SF (mm ⫻ m/s)
Generic Contoured
Flat-Foam
I
C
I
C
I
C
50 ⫾ 18 .14 ⫾ .05 4.3 ⫾ 2.7
28 ⫾ 12 .10 ⫾ .03 1.9 ⫾ 1.2
64 ⫾ 21* .18 ⫾ .07† 6.6 ⫾ 4.6
34 ⫾ 12 .13 ⫾ .03† 2.9 ⫾ 1.7
50 ⫾ 23 .14 ⫾ .04 3.9 ⫾ 2.6
29 ⫾ 16 .11 ⫾ .03 2.4 ⫾ 2.4
Data presented as mean ⫾ SD. * p ⬍ .01. † p ⬍ .05.
COP Parameters From the Seat Force Platform Results from the seat force platform exhibited the same pattern as the pressure measurement system used in this study (FSA system). The data were similar but not equal to those issued from the FSA. In fact, both the cushion and the direction factors had a significant effect on MCD, MV, and SF parameters. Moreover, the generic contoured cushion behaved differently from a statistical viewpoint than the air flotation and flat foam cushions as indicated by the higher value of all the parameters (table 3). DISCUSSION This study investigated the effects of seat cushion on dynamic stability during a controlled reaching task. The major finding of this study was that paraplegic subjects were able to increase both the distance covered and the velocity of the COP when the generic contoured cushion was used during reaching. This result was confirmed by data gathered from both the seat force platform and the pressure measurement system. The MCD value found in this study compared well with the data of Kamper et al,18 who found that the maximal lateral displacement of the COP for subjects with paraplegia was about 73.2mm. In this last case, however, the displacement represented the limit of the COP movement that the subject could maintain without the use of the upper extremities when leaning to the right side. It should be noted that the seat cushion in the earlier study18 was made of 2 layers of flat foam (5cm of HR70 topped with 2.5cm of HR32). The MV parameter found in this study, as estimated from the seat platform (MV ⫽ .18 ⫾ 0.7m/s), was similar to the mean value obtained by Seelen20 (MV ⫽ .17m/s) for his low thoracic paraplegic group. The ability of the COP to cover a large distance with high velocity appears to be linked to the increase in stability during reaching. This conclusion was confirmed in recent studies for subjects with hemiplegia3 and paraplegia.13 The arguments in favor of using COP as the outcome measure instead of speed of the reach are based on the fact that performing a reaching task requires not only adequate coordination of the trunk and upper limbs, but also the control of load in the lower limbs. The speed of the reach represented the average of the hand velocity during the pointing task (ie, the distance divided by the time duration). Because in our study the distance was standardized to 130% of the arm length, any variation in speed would be caused by the variation in the hand time movement. However, the duration of the voluntary movement of upper limb and hand constituted only part of the duration of the movement of the body’s center of gravity. Indeed, the movement began before the execution of the task and lasted a long time after reaching the target.29 These postural adjustments precede, accompany, and follow the movement; they are specific to
the intended movement and would counteract the disturbance.29 Hand movement time during reaching can be used as an indicator of the improvement of reaching task only when the maximum reaching distance is allowed to vary, or when the subjects are instructed to go faster.3 However, when the reaching distance was standardized to a percentage of the arm length, time movement did not appear to be an indicator of the performance. Lino et al30 have studied the effect of seat contact area (as measured by the percentage of the ischiofemoral distance in contact with the seat) on the velocity of the reaching task. In this study,30 subjects were asked to point to a target at their maximal velocity. The time duration of the movement was unaffected by the variation of the seat contact from 100% to 30% of the ischiofemoral distance, whereas the peak vertical force under seat increased significantly. Also, in a recent study, Dean et al31 found that reach direction had no significant effect on the time taken to complete the reaching task movement in a group of elderly subjects. But the peak vertical ground reaction forces through each foot were significantly affected by reach direction. One would assume that if a high load is measured under the ipsilateral foot during the reaching task, the COP under seat is moving through the loaded foot, which means that the COP covers a larger distance. Unfortunately, Dean31 did not report on the COP data under seat. Finally, Seelen et al32 found, after a rehabilitation period, a clear improvement in the control of sitting posture in a group of patients with low thoracic SCI during bimanual forward reaching task. In fact, when spinal bracing was stopped, these patients showed higher levels of COP displacements. The COP is generally viewed as the neuromuscular response to imbalances of the body’s center of gravity.33 This means that during a reaching task, the COP is continuously tracking the body’s center of gravity; therefore, the horizontal distance between the center of gravity and the COP should always be minimized to ensure stability. However, in this study we did not assess the kinematics of the body’s center of gravity; further study must be performed to verify the assumption that the increase in stability corresponds to the decrease of the horizontal distance between the COP and the center of gravity. In this sense, a seating device that allows the COP to cover a fairly large distance with higher velocity should be considered stable. This characteristic of the cushion should then improve wheelchair users’ stability during ADLs. In our study, higher pressure was recorded when the foam cushions were used. However, these peak pressures corresponded to the instant at middistance of reaching target, and did not represent static pressure data. In a similar dynamic movement, Kernozek and Lewin34 found that peak pressure under the ischial tuberosity region increased by 42% during Arch Phys Med Rehabil Vol 82, February 2001
280
DYNAMIC STABILITY IN SITTING, Aissaoui
manual wheelchair propulsion to reach a value of 272mmHg/s in individuals with paraplegia, when they sat on a Jay Active cushion. Nevertheless, we noticed in our study a hysterisis in COP trajectory between forward and backward reaching movements. This was because of the stability provided by the cushions. It is well known that gel, fluid-filled cushions, and simple air pillows behave fairly elastically when sat on, and create a feeling of instability during acceleration and deceleration of the human body. Because they are elastic, they tend to transmit the energy of dynamic loading to the body via the gluteal tissues for dissipation. Visoelastic materials such as foam have the capacity to absorb energy on impact, which results in a real and perceived sense of stability.22 In this study, the generic contoured foam (ISCUS) seemed more stable than the flat-foam and the Roho cushions. The concave shape of the contoured foam cushion could have an effect on wheelchair users’ stability in terms of lowering the center of gravity. In fact, the location of the center of gravity is different between able-bodied subjects and wheelchair users. Individuals with paraplegia have a larger upper extremity mass compared with their lower extremities. The center of gravity of individuals with paraplegia has been estimated to be about 5% of the body length (3– 4 vertebral levels) higher than that of able-bodied subjects.35 This change in center of gravity would contribute to a change in their limits of stability. CONCLUSION This study provides an objective method to assess the dynamic stability of wheelchair users when they perform ADLs such as reaching tasks. The relationship between the instantaneous velocity and the distance covered by the COP, at the body-seat interface, gives insight information about dynamic stability. It has been shown here that a seat cushion is said to be stable when it allows the COP to cover a larger distance with higher speed during voluntary reaching tasks. These findings have implications for wheelchair seating recommendations and especially for seat cushion selection. Acknowledgment: The authors thank Margot Lacroix for her assistance in revising the text. References 1. Zacharkow D. Posture: sitting, standing, chair design and exercise. Springfield (IL): Charles C Thomas; 1988. p 50-76. 2. Riley PO, Benda BJ, Gill-Body KM, Krebs DE. Phase plane analysis of stability in quiet standing. J Rehabil Res Dev 1995; 32:227-35. 3. Dean CM, Shepherd RB. Task-related training improves performance of seated reaching tasks after stroke. Stroke 1997; 28:722-8. 4. McClenaghan BA. Sitting stability of selected subjects with cerebral palsy. Clin Biomech 1989;4:213-6. 5. Reid DT, Sochaniwskyj A, Milner M. An investigation of postural sway in sitting of normal children and children with neurological disorders. Phys Occup Ther Pediatr 1991;11:19-35. 6. Sochaniwskyj A, Koheil R, Bablich K, Milner M, Lotto W. Dynamic monitoring of sitting posture for children with spastic cerebral palsy. Clin Biomech 1991;6:161-7. ¨ , C¸akci A. Angular 7. Dursun E, Hamamci N, Do¨nmez S, Tu¨zu¨nalp O biofeedback device for sitting balance of stroke patients. Stroke 1996;8:1354-7. 8. Chari VR, Kirby RL. Lower limb influence on sitting balance while reaching forward. Arch Phys Med Rehabil 1986;67: 730-3. Arch Phys Med Rehabil Vol 82, February 2001
9. Bendix T, Jessen F, Krohn L. Biomechanics of forward-reaching movements while sitting on fixed forward or backward-inclining or tiltable seats. Spine 1988;13:193-6. 10. Son K, Miller JAA, Schultz AB. The mechanical role of the trunk and lower extremities in a seated weight-moving task in the sagittal plane. J Biomech Eng 1988;110:97-103. 11. Moore S, Brunt D. Effects of trunk support and target distance on postural adjustments prior to a rapid reaching task by seated subjects. Arch Phys Med Rehabil 1991;72:638-41. 12. Curtis KA, Kindlin CM, Reich KM, White DE. Functional reach in wheelchair users: the effects of trunk and lower extremity stabilization. Arch Phys Med Rehabil 1995;76:360-7. 13. Seelen HAM, Potten YJM, Huson A, Spaans F, Reulen JPH. Impaired balance control in paraplegic subjects. J Electromyogr Kinesiol 1997;7:149-60. 14. Bernard P-L, Peruchon E, Micallef J-P, Hertog C, Rabischong P. Balance and stabilization capability of paraplegic wheelchair athletes. J Rehabil Res Dev 1994;31:287-96. 15. Cooper RA, Stewart KJ, VanSickle DP. Evaluation of methods for determining rearward static stability of manual wheelchairs. J Rehabil Res Dev 1994;31:144-7. 16. Kirby RL, Sampson MT, Thoren FAV, MacLeod DA. Wheelchair stability: effect of body position. J Rehabil Res Dev 1995;32:36772. 17. Chesney DA, Axelson PW. Measuring functional changes: practical methodologies for use in clinics. Proceedings of the 13th International Seating Symposium; 1997; Pittsburgh. 1997; p 132-8. 18. Kamper D, Barin K, Parnianpour M, Reger S, Weed H. Preliminary investigation of the lateral postural stability of spinal cordinjured individuals subjected to dynamic perturbations. Spinal Cord 1999;37:40-6. 19. Allison GT, Singer KP. Assisted reach and transfers in individuals with tetraplegia: towards a solution. Spinal Cord 1997; 35:217-22. 20. Seelen HAM. Reorganisation of postural control in spinal cord injured persons [dissertation]. Maastricht (Netherlands): Univ of Maastricht; 1997. 21. Staarink HAM. Sitting posture, comfort and pressure. Delft: Delft Univ Pr; 1995. p 169-89. 22. Ferguson-Pell MW. Seat cushion selection. J Res Rehabil Dev 1990;(Suppl 2):49-73. 23. Pai Y-C, Patton J. Center of mass velocity-position predictions for balance control. J Biomech 1997;30:347-54. 24. Wolff DR, Rose J, Jones VK, Bloch DA, Oehlert JW, Gamble JG. Postural balance measurements for children and adolescents. J Orthop Res 1998;16:271-5. 25. Garber SL, Dyerly LR. Wheelchair cushions for persons with spinal cord injury: an update. Am J Occup Ther 1991;45: 550-4. 26. Sprigle SH, Faisant TE, Chung K-C. Clinical evaluation of custom-contoured cushions for the spinal cord injured. Arch Phys Med Rehabil 1991;71:655-8. 27. Rosenthal MJ, Felton RM, Hileman DL, Lee M, Friedman M, Navach JH. A wheelchair cushion designed to redistribute sites of sitting pressure. Arch Phys Med Rehabil 1996;77:278-82. 28. Brienza DM, Karg PE. Seat cushion optimization: a comparison of interface pressure and tissue stiffness characteristics for spinal cord injured and elderly patients. Arch Phys Med Rehabil 1998; 79:388-94. 29. Do MC, Bouisset S, Moynot C. Are paraplegics handicapped in the execution of a manual task? Ergonomics 1985;28:136575. 30. Lino F, Bouisset B, Duchene JL. Effect of seat cushion area on the velocity of a pointing task. In: Bellotti P, Cappozzo A, editors. Biomechanics: Proceedings of the VIIIth Meeting of the European Society of Biomechanics; 1992, June 21-24. Rome: Universita´ Sapienza; 1992. 31. Dean CM, Shepherd RB, Adams RD. Sitting balance II: reach direction and thigh support affect the contribution of the lower limbs when reaching beyond arm’s length in sitting. Gait Posture 1999;10:147-53.
DYNAMIC STABILITY IN SITTING, Aissaoui
32. Seelen HAM, Potten YJM, Drukker J, Reulen JPH, Pons C. Development of new muscle synergies in postural control in spinal cord injured subjects. J Electromyogr Kinesiol 1998;8:23-34. 33. Winter DA. Biomechanics and control of human movement. New York: John Wiley; 1990 p 93-6. 34. Kernozek TW, Lewin JE. Seat interface pressure of individuals with paraplegia: influence of dynamic wheelchair locomotion compared with static seated measurements. Arch Phys Med Rehabil 1998;79:313-6. 35. Duval-Beaupere G, Robain G. Upward displacement of the center of gravity in paraplegic patients. Paraplegia 1991;29:309-17.
281
Suppliers a. HIGH PROFILE威 Cushion; ROHO Inc, 100 N Florida Ave, Belleville, IL 62221-5429. b. ISCUS and flat-foam cushion; Orthofab Inc, 10370 Louis-H Lafontaine, Anjou, Que H1J 2T3, Canada. c. Vista Medical Ltd, 120 Maryland St, Winnipeg, Man R3G 1L1, Canada. d. Advanced Mechanical Technology Inc, 151 California St, Newton, MA 02158. e. STATISTICA 5.0; StatSoft, Inc, 2300 E 14th St, Tulsa, OK 74104. f. SERTIX, Inc, 1 Madison St, East Rutherford, NJ 07073. g. AMG Medical, Inc, 8505 Dalton, Montreal, Que H4T 1V5, Canada.
Arch Phys Med Rehabil Vol 82, February 2001
PROSTHETICS/ORTHOTICS/DEVICES
Effect of Seat Cushion on Dynamic Stability in Sitting During a Reaching Task in Wheelchair Users With Paraplegia Rachid Aissaoui, PhD, Chantal Boucher, BSc, Daniel Bourbonnais, PhD, Miche`le Lacoste, OT, Jean Dansereau, PhD ABSTRACT. Aissaoui R, Boucher C, Bourbonnais D, Lacoste M, Dansereau J. Effect of seat cushion on dynamic stability in sitting during a reaching task in wheelchair users with paraplegia. Arch Phys Med Rehabil 2001;82:274-81.
© 2001 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
Objectives: To examine the effects of seat cushions on dynamic stability in sitting during a controlled reaching task by wheelchair users with paraplegia. Design: A randomized, controlled test. Setting: Rehabilitation center. Participants: Nine wheelchair users with paraplegia. Interventions: Three types of cushions—an air flotation, a generic contoured, and a flat polyurethane foam—were tested during a controlled reaching task in ipsilateral and contralateral directions, at 45° from the sagittal plane in the anterolateral direction. Center of pressure (COP) coordinates were monitored by using a pressure measurement system as well as a force platform under seat. Main Outcome Measures: Trajectory of COP, maximal distance covered by COP, maximal velocity of COP; and the index of asymmetry between right and left maximal pressure under ischial tuberosities. Results: The generic contoured cushion allowed the COP to cover significantly (p ⬍ .02) a larger distance (81 ⫾ 28mm) when compared with the air flotation (63 ⫾ 25mm) or the flat foam (61 ⫾ 29mm) cushions. The COP velocity was significant ( p ⬍ .05) for the generic contoured cushion (.14 ⫾ .05m/s) versus the air flotation (.10 ⫾ .04m/s) or the flat-foam (.10 ⫾ .03m/s) cushions. The index of asymmetry was higher for the generic contoured and the flat foam cushions. During reaching, maximal pressure under ipsilateral ischial tuberosity was significantly higher for the flat foam (275 ⫾ 70mmHg) and the generic contoured (235 ⫾ 81mmHg) cushions, when compared with the air flotation cushion (143 ⫾ 51mmHg). Conclusion: Seat cushions can significantly affect sitting balance during reaching tasks. This study provided an objective method to assess the dynamic stability of wheelchair users when they perform activities of daily living requiring reaching. These findings have implications for wheelchair seating recommendations, especially seat cushion selection. Key Words: Paraplegia; Pressure; Rehabilitation; Wheelchairs.
YNAMIC SITTING REFERS to the continuous process of D postural changes during sitting. Sitting posture is usually unstable without additional external support because the hip
From the De´partement de Ge´nie Me´canique, E´cole Polytechnique de Montre´al (Aissaoui, Boucher, Lacoste, Dansereau), Research Center, Montreal Rehabilitation Institute (Bourbonnais), Montreal, Quebec, Canada. Accepted in revised form May 23, 2000. Supported by the Natural Sciences and Engineering Research Council of Canada. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the author(s) is/are associated. Reprint requests to Rachid Aissaoui, PhD, Chaire Industrielle CRSNG sur les aides techniques a¯ la posture, Dept de Ge´nie Me´canique, Ecole Polytechnique de Montreal, CP 6079, SVCL Centre-ville, Montreal, Que H3C 3A7, Canada. e-mail: Rachid. [email protected]. 0003-9993/01/8202-5983$35.00/0 doi:10.1053/apmr.2001.19473
Arch Phys Med Rehabil Vol 82, February 2001
joints are in an intermediate position with respect to the range of motion and because the trunk cannot be locked relative to the thighs by ligamentous restraint. As a result, muscle activity is necessary to maintain the trunk segment in an upright posture when sitting without additional stabilizers.1 The capacity to maintain balance and posture in sitting is a prerequisite to the activities of daily living (ADLs), and deficits in sitting balance control can severely limit task performance.2,3 Research related to the quantification of sitting stability has been oriented toward applications such as postural sway measurements4-7 and functional reach,3,8-13 as well as wheelchair stability.14-18 However, ways of measuring sitting stability vary between investigators and applications. Greater stability during quiet sitting has been associated with a smaller radius of motion at the level of the upper trunk and neck segment.4-7 In wheelchair seating, static stability has been measured by the angle at which the front or side wheel of the wheelchair loses contact with the ground on a tilting platform.15,16 During dynamic perturbation of a wheelchair user, a sagittal acceleration was imposed on paraplegic athletes and the stability of the wheelchair user was measured by a damping factor14 (the ratio between the measured and imposed acceleration to the trunk segment). Recently, Kamper et al18 imposed a lateral tilting perturbation to individuals with spinal cord injury (SCI) and measured the displacement of center of pressure (COP) at ground level. Kamper18 found that the maximal COP displacement tolerated was highly related to dynamic stability. As far as functional reach is concerned, however, the manner in which the body is stabilized while performing reaching tasks is not well understood. The influence of the lower legs in maintaining sitting balance during forward reaching movements has been investigated in healthy subjects, and it was found that thigh and foot support permitted larger forward excursions of the center of gravity.8,10 In terms of seating aids, a forward seat inclination has been reported to influence trunk posture during forward reaching.9 Curtis et al12 compared the effect of using a chest belt on the area of functional reach in the sagittal and transverse planes, and showed that the area of functional reach increased for subjects with paraplegia when using a chest belt. Allison and Singer19 have shown that, during a lateral reaching task, the lateral displacement of a subject’s COP increased significantly when using trunk orthoses. Dean and Shepherd3 evaluated the effect of a 2-week task-related training program for hemiplegic adults with the aim of increasing their reaching distance; they found that after training, subjects were able to reach faster and longer. In terms of body movement, it was found that this improvement in reaching represented a gain in terms of sitting stability. In fact, both the
275
DYNAMIC STABILITY IN SITTING, Aissaoui
displacement13 and the peak velocity20 of the COP in the anteroposterior direction were larger for able-bodied subjects during controlled bimanual forward reaching when compared with a group of individuals with SCI. Moreover, Seelen20 has shown that a clear improvement in controlling sitting posture corresponded to an increase of COP displacement in high and low SCI groups after a period of rehabilitation program. However, the effect of seat cushions on sitting balance has not been examined from a quantitative viewpoint in previous studies. In fact, in most of these studies, either a hard seat was used or the subject sat directly on a rigid force-platform. This lack of relevant studies can be explained by to the fact that it is difficult to estimate the COP location at the body-seat interface. The stability provided by seat cushions is considered to be the most important characteristic of sitting support after pressure distribution.17,21,22 The effect that cushions have on stability, though real or perceived, can be critical especially for users with poor trunk control. Persons with spinal lesions often have an acute awareness of their trunk stability that is far more subtle and complex than position measurement systems are able to detect.22 In a clinical setting, compromises between cushion stability and pressure distribution capacity often have to be made. This is usually performed in practice by accepting a small increase in the accepted pressure over bony areas.22 Furthermore, it is suggested that maintaining postural stability requires controlling both displacement and velocity of COP.18,23,24 The purpose of this study was to examine the effect of seat cushion on dynamic stability in sitting during a controlled reaching task performed by individuals with paraplegia. In this study we hypothesized that sitting balance is influenced by the seat cushion when performing reaching movement tasks, and that the seat cushion is considered stable when it allows the COP to move rapidly and to cover a larger distance. METHOD Subjects The subject population for this study included 9 wheelchair users with paraplegia by SCI. They were recruited from the Rehabilitation Institute of Montreal. To participate, subjects had to meet the following criteria: (1) diagnosis of paraplegia resulting in complete spinal lesion for at least 12 months; (2) no pressure ulcer for at least 1 year; (3) no orthopedic problem that would interfere with the ability to perform reaching task while seated; (4) the range of spinal lesion was limited to T3–L2 to
still have minimal trunk control12; (5) ability to give informed consent; and (6) ability to sit for a period of 1.5 hours. All of the subjects were right-handed. Their anthropometric characteristics as well as the clinical description of their seated posture are presented in table 1. All procedures were performed with the approval of the Rehabilitation Institute of Montreal’s Ethics Committee. Experimental Protocol Subjects sat on a stool (fig 1). Footrest height was adjusted to have the thigh segment horizontal. The feet were placed parallel, and the distance between the medial borders of the feet was about 15cm. The seat angle was fixed at 0°, while the seat to back angle was fixed at 95°. The distance between the popliteal fossea and the front edge of the seat was adjusted to 4cm. Arm length was measured by using a measuring tape, in a seated position with the upper limb extended in horizontal plane and pointing in a forward direction. The distance between the acromion process and the tip of the second finger was measured. The subject started with his/her right hand resting on a pressure-sensitive starting switch (button I) that was placed on the right side of the stool (fig 1). A first alarm indicated to the subject to prepare to reach, whereas a second alarm randomly activated within a maximum period of 3 seconds signaled the start of the hand movement. The subject reached a second switch (button II) in 2 directions: 45° ipsilateral and 45° contralateral from the sagittal plane (fig 1). No instructions about movement speed were given. Reaching distance was standardized to 130% of arm length, whereas the height of the switch button II was standardized to 100% of the glenohumeral joint height as measured from the floor in the seated posture. The subject maintained the switch button II pressed for at least 2 seconds and then released it to return finally to press the switch button I, which ended the trial. During reaching, the hand not used was placed on the opposite thigh segment to ensure stability. A belt was fixed on the lateral side of the seat support and placed on the thigh segment to secure the subject to the stool. The belt did not interfere with the reaching movement. Subjects performed 3 trials for each condition (left and right hand), and with 3 randomly selected seat cushions a 3-inch air-flotationa (ROHO), a 2-inch polyurethane HR35 flat-foamb (flat foam), and a 3-inch polyurethane HR45 generic contoured foam (ISCUS).b The air-flotation cushion is used by 51% of the
Table 1. Characteristics of Wheelchair Users
Subject
Gender
Age (yr)
1 2 3 4 5 6 7 8 9
Woman Man Man Man Man Man Woman Man Woman
39 25 20 65 59 42 52 47 46
Diagnosis
Arm-Length (cm)
Sitting Hip/ Thigh Width* (cm)
ISCUS Size (cm)
ROHO Pressure† (mmHg)
Pelvic Obliquity
Posterior Pelvic Tile
Transverse Pelvic Rotation
T5 paraplegia T12 paraplegia T8 paraplegia T12 paraplegia T3–T5 paraplegia T4–T5 paraplegia T4–T5 papaplegia T4–T6 papaplegia Postmyelitis paraplegia
66 83.5 76 76 78 70 66 71 70.5
30.5 30.5 32 40.5 43 37 33 35.5 43
38 38 38 40 45.5 38 35.5 35.5 45.5
45 34 36 46 40 30 24 45 32
No No No No Yes (R) Yes (L) Yes (R) Yes (L) Yes (L)
No No No Yes No No No Yes No
No Yes (R) Yes (L) Yes (L) Yes (R) No Yes (R) Yes (L) Yes (L)
Abbreviations: R, right; L, left. * The subject remained sitting, but raised his/her arms slightly and the horizontal distance across the widest part of the hips or thighs was measured by an anthropometer.f † The ROHO cushion was pressurized until a clearance of 1.5cm was reached under both right and left ischial tuberosity, which were located by palpation. Pressure under ROHO cushion was measured by a hand-held sphygmomanometer.g
Arch Phys Med Rehabil Vol 82, February 2001
276
DYNAMIC STABILITY IN SITTING, Aissaoui
Fig 1. Experimental setup. Schematic representation of subject seated on a stool, facing investigators (top); from above (bottom). Abbreviations: FSA, force sensing array.
spinal cord–injured population, whereas its use in the general wheelchair user population is actually around 30%.25 Garber and Dyerly25 reported that patients using air flotation cushion may feel unstable or experience difficulty transferring to another surface. In our study, 5 of the 9 subjects used the air flotation cushion as their regular cushion. The ability of the custom-contoured cushion to improve posture and balance, as well as to reduce pressure has already been shown.26-28 It should be noted that the contoured cushion used in this study had a generic concave shape with fixed maximal depth of 55mm with respect to the lateral edge of the cushion. This maximal depth corresponds approximately to the average contour of the SCI group in the recent study of Brienza and Karg.28 Arch Phys Med Rehabil Vol 82, February 2001
Instrumentation Pressure measurement system. During the reaching task, the pressure distribution at the body-seat interface was recorded at 5Hz by using a force sensing arrayc (FSA), which consists of 225 force sensors organized in a flexible mat of 15 ⫻ 15. Before each subject’s session, the FSA mat was calibrated up to its maximum of 300mmHg. The trajectory of the COP was computed from the pressure distribution data for each trial. A median low-pass filter was used to filter raw displacement data of COP over time. An interpolation by a cubic spline function was used on the filtered COP data to estimate the magnitude of the velocity vector by using a central
DYNAMIC STABILITY IN SITTING, Aissaoui
difference method. The maximal covered distance (MCD), by the COP, the maximal velocity (MV) of the COP, as well as the surface area under the curve defined by the velocity-distance relationship were computed (fig 2). An index of asymmetry between right and left maximal pressure under the ischial tuberosities regions was also calculated at each sample. The maximal value of the index of asymmetry (MIA) was computed for each reaching task trial.
277
Seat force platform. An AMTId force-platform was used to collect force and moment data under seat at 50Hz. Force data were filtered by means of a low-pass median filter. The COP coordinates under the seat were determined from the forceplatform data. For each trial, the MCD, MV, as well as the surface under the curve defined by the velocity-distance relationship, were estimated. During the experiment, the FSA mat and the AMTI were synchronized.
Fig 2. (A) COP trajectory of a typical subject during right hand reaching in the ipsilateral and contralateral side. (B) Phase diagram plot of the COP (instantaneous velocity in m/s versus instantaneous displacement mm). Abbreviations: AP, anteroposterior direction in mm; ML, mediolateral direction in mm; FF, flat foam.
Arch Phys Med Rehabil Vol 82, February 2001
278
DYNAMIC STABILITY IN SITTING, Aissaoui
Statistical Analysis For each condition the 3 trials were averaged for the selected dependent variables. The selected variables were the MCD, MV, the surface area in the forward reaching (SF) and, for the pressure data, the MIA, as well as the maximal pressure at right and left ischial tuberosity regions (max PR, max PL). Each dependent variable was analyzed separately, in an analysis of variance (ANOVA) (2 factors: cushion ⫻ direction) with repeated measures on both independent variables, to determine if significant differences existed between the cushions. If indicated, a Bonferroni test was performed to determine which pairs of means were significantly different (p ⬍ .05). STATISTICAe software was used to perform the statistical analysis. RESULTS All subjects were able to perform reaching tasks at 130% of their arm length except for 1 subject who was only able to reach at 115%. This subject has poor trunk control and his paraplegia was caused by myelitis. Figure 2A shows a COP trajectory for a typical subject during a reaching task with the right hand in the 2 directions and for 3 cushions. The COP coordinates were computed from pressure distribution data. There were 3 distinct COP trajectories for the 3 cushions in the ipsilateral direction, though this was less apparent in the contralateral direction (fig 2A). However, it is difficult to associate sitting stability with a specific COP trajectory. In the phase diagram in figure 2B, which represents the instantaneous velocity with respect to the cumulative distance covered by the COP, it is clear that the COP covers a larger distance with higher speed for the contoured foam, compared with either the flat-foam or the air flotation cushions. The pattern was similar in the 2 directions but with different amplitudes. In the ipsilateral direction, the upper limb is outside the body during reaching, contributing largely to the displacement of the total body center of mass, which is reflected by the large displacement of the COP. In general, the COP reached a MV at middistance of a reaching target. This was caused by the acceleration and deceleration process of total body center of mass. There was more fluctuation in the backward movement than in the forward. Thus, the MV, MCD, SF, as well as the MIA parameters were evaluated in the forward direction of the reaching movement only. In figure 2B, the COP covered a distance of approximately 130mm, and reached a velocity of 0.2m/s for the ISCUS cushion in the ipsilateral direction. These values represent approximately a gain of 38%, 100%, and 64%, respectively, for the distance covered, the velocity, and the SF parameters with respect to the ROHO cushion. The flat-foam cushion behaved
similarly to the ROHO cushion in term of distance, but stayed in between the ISCUS and the ROHO cushions in term of velocity. This trend was generally present for all subjects. COP Parameters From the FSA System Table 2 shows the mean average of the COP parameters as estimated from the FSA system for the right hand. During the forward motion of the reaching task with the right hand, the MCD parameter was larger for the generic contoured cushion (81 ⫾ 28mm) when compared with the air flotation (63 ⫾ 25 mm) and flat-foam (61 ⫾ 29mm) cushions. ANOVA revealed that there was a cushion effect (F ⫽ 3.99; df ⫽ 2,16; p ⬍ .05) and a direction effect (F ⫽ 18.65; df ⫽ 1,8; p ⬍ .01) when reaching with the right hand. There was significant interaction between the effects of cushion and direction on MCD (F ⫽ 6.24; df ⫽ 2,16; p ⬍ .01). The same separate effects caused by cushion and direction were present for the left hand, except that there was no interaction between them. The generic contoured cushion had a significant effect (p ⬍ .02) on the MCD parameter compared with the flat-foam and the air flotation cushions. The average MV of the COP was significantly higher (p ⬍ .05) for the generic contoured (.14 ⫾ .05m/s) than for the air flotation or the flat-foam cushions (.10 ⫾ .04m/s, .10 ⫾ .03 m/s, respectively). There were significant cushion effect (F ⫽ 6.97; df ⫽ 2,16; p ⬍ .01) and direction effect (F ⫽ 9.76; df ⫽ 1,8; p ⬍ .02) on the MV parameter because of the generic contoured cushion. Similar effects on the MV parameter were found for the left hand. There was a significant effect on the direction factor (p ⬍ .05) and a moderate effect (p ⬍ .06) on the cushion factor with respect to the SF parameter, when reaching with the right hand for the generic contoured cushion. In summary, for all of the MCD, MV, and SF parameters, the generic contoured cushion was significantly different from both the air flotation and the flat foam cushions. The foam cushions (the generic contoured and the flat) exhibited more asymmetry than the air flotation cushion. In fact, during the reach, the MIA reached .67 ⫾ .20 and .65 ⫾ .18, respectively, for the generic contoured and flat foam cushions in the ipsilateral direction, whereas it did not exceed .51 ⫾ .15 for the air flotation. The air flotation cushion had a significant effect (p ⬍ .02) on the MIA parameter. It should be noted that a MIA value of 0.5 means that the pressure under the right ischial tuberosity region is 3 times higher than the left one. Max PR was significantly higher for the flat foam cushion (275 ⫾ 70mmHg) and the generic contoured (235 ⫾ 81mmHg) compared with the air flotation cushion (143 ⫾ 51mmHg).
Table 2. COP Parameters From the FSA System for Right-Hand Reaching Air Flotation
Generic Contoured
Flat-Foam
Parameters
I
C
I
C
I
C
MCD (mm) MV (m/s) SF (mm ⫻ m/s) MIA max PR (mmHg) max PL (mmHg)
63 ⫾ 25 .10 ⫾ .04 4.1 ⫾ 2.8 .51 ⫾ .15* 143 ⫾ 51‡ 124 ⫾ 36†
31 ⫾ 18 .06 ⫾ .04 1.6 ⫾ 2.0 .47 ⫾ .18 140 ⫾ 62‡ 146 ⫾ 38†
81 ⫾ 28* .14 ⫾ .05† 7.0 ⫾ 5.3† .67 ⫾ .20 235 ⫾ 81 171 ⫾ 68
32 ⫾ 15 .07 ⫾ .03 1.8 ⫾ 1.6 .65 ⫾ .27 223 ⫾ 69 183 ⫾ 76
61 ⫾ 29 .10 ⫾ .03 3.9 ⫾ 3.0 .65 ⫾ .18 275 ⫾ 70 183 ⫾ 83
29 ⫾ 23 .06 ⫾ .04 1.7 ⫾ 2.3 .60 ⫾ .37 252 ⫾ 86 229 ⫾ 88
NOTE. Data presented as mean ⫾ standard deviation (SD). Abbreviations: I, ipsilateral; C, contralateral. * p ⬍ .02. † p ⬍ .05. ‡ p ⬍ .01.
Arch Phys Med Rehabil Vol 82, February 2001
279
DYNAMIC STABILITY IN SITTING, Aissaoui Table 3. COP Parameters From Seat Platform for Right Hand Reaching Air Flotation
MCD (mm) MV (m/s) SF (mm ⫻ m/s)
Generic Contoured
Flat-Foam
I
C
I
C
I
C
50 ⫾ 18 .14 ⫾ .05 4.3 ⫾ 2.7
28 ⫾ 12 .10 ⫾ .03 1.9 ⫾ 1.2
64 ⫾ 21* .18 ⫾ .07† 6.6 ⫾ 4.6
34 ⫾ 12 .13 ⫾ .03† 2.9 ⫾ 1.7
50 ⫾ 23 .14 ⫾ .04 3.9 ⫾ 2.6
29 ⫾ 16 .11 ⫾ .03 2.4 ⫾ 2.4
Data presented as mean ⫾ SD. * p ⬍ .01. † p ⬍ .05.
COP Parameters From the Seat Force Platform Results from the seat force platform exhibited the same pattern as the pressure measurement system used in this study (FSA system). The data were similar but not equal to those issued from the FSA. In fact, both the cushion and the direction factors had a significant effect on MCD, MV, and SF parameters. Moreover, the generic contoured cushion behaved differently from a statistical viewpoint than the air flotation and flat foam cushions as indicated by the higher value of all the parameters (table 3). DISCUSSION This study investigated the effects of seat cushion on dynamic stability during a controlled reaching task. The major finding of this study was that paraplegic subjects were able to increase both the distance covered and the velocity of the COP when the generic contoured cushion was used during reaching. This result was confirmed by data gathered from both the seat force platform and the pressure measurement system. The MCD value found in this study compared well with the data of Kamper et al,18 who found that the maximal lateral displacement of the COP for subjects with paraplegia was about 73.2mm. In this last case, however, the displacement represented the limit of the COP movement that the subject could maintain without the use of the upper extremities when leaning to the right side. It should be noted that the seat cushion in the earlier study18 was made of 2 layers of flat foam (5cm of HR70 topped with 2.5cm of HR32). The MV parameter found in this study, as estimated from the seat platform (MV ⫽ .18 ⫾ 0.7m/s), was similar to the mean value obtained by Seelen20 (MV ⫽ .17m/s) for his low thoracic paraplegic group. The ability of the COP to cover a large distance with high velocity appears to be linked to the increase in stability during reaching. This conclusion was confirmed in recent studies for subjects with hemiplegia3 and paraplegia.13 The arguments in favor of using COP as the outcome measure instead of speed of the reach are based on the fact that performing a reaching task requires not only adequate coordination of the trunk and upper limbs, but also the control of load in the lower limbs. The speed of the reach represented the average of the hand velocity during the pointing task (ie, the distance divided by the time duration). Because in our study the distance was standardized to 130% of the arm length, any variation in speed would be caused by the variation in the hand time movement. However, the duration of the voluntary movement of upper limb and hand constituted only part of the duration of the movement of the body’s center of gravity. Indeed, the movement began before the execution of the task and lasted a long time after reaching the target.29 These postural adjustments precede, accompany, and follow the movement; they are specific to
the intended movement and would counteract the disturbance.29 Hand movement time during reaching can be used as an indicator of the improvement of reaching task only when the maximum reaching distance is allowed to vary, or when the subjects are instructed to go faster.3 However, when the reaching distance was standardized to a percentage of the arm length, time movement did not appear to be an indicator of the performance. Lino et al30 have studied the effect of seat contact area (as measured by the percentage of the ischiofemoral distance in contact with the seat) on the velocity of the reaching task. In this study,30 subjects were asked to point to a target at their maximal velocity. The time duration of the movement was unaffected by the variation of the seat contact from 100% to 30% of the ischiofemoral distance, whereas the peak vertical force under seat increased significantly. Also, in a recent study, Dean et al31 found that reach direction had no significant effect on the time taken to complete the reaching task movement in a group of elderly subjects. But the peak vertical ground reaction forces through each foot were significantly affected by reach direction. One would assume that if a high load is measured under the ipsilateral foot during the reaching task, the COP under seat is moving through the loaded foot, which means that the COP covers a larger distance. Unfortunately, Dean31 did not report on the COP data under seat. Finally, Seelen et al32 found, after a rehabilitation period, a clear improvement in the control of sitting posture in a group of patients with low thoracic SCI during bimanual forward reaching task. In fact, when spinal bracing was stopped, these patients showed higher levels of COP displacements. The COP is generally viewed as the neuromuscular response to imbalances of the body’s center of gravity.33 This means that during a reaching task, the COP is continuously tracking the body’s center of gravity; therefore, the horizontal distance between the center of gravity and the COP should always be minimized to ensure stability. However, in this study we did not assess the kinematics of the body’s center of gravity; further study must be performed to verify the assumption that the increase in stability corresponds to the decrease of the horizontal distance between the COP and the center of gravity. In this sense, a seating device that allows the COP to cover a fairly large distance with higher velocity should be considered stable. This characteristic of the cushion should then improve wheelchair users’ stability during ADLs. In our study, higher pressure was recorded when the foam cushions were used. However, these peak pressures corresponded to the instant at middistance of reaching target, and did not represent static pressure data. In a similar dynamic movement, Kernozek and Lewin34 found that peak pressure under the ischial tuberosity region increased by 42% during Arch Phys Med Rehabil Vol 82, February 2001
280
DYNAMIC STABILITY IN SITTING, Aissaoui
manual wheelchair propulsion to reach a value of 272mmHg/s in individuals with paraplegia, when they sat on a Jay Active cushion. Nevertheless, we noticed in our study a hysterisis in COP trajectory between forward and backward reaching movements. This was because of the stability provided by the cushions. It is well known that gel, fluid-filled cushions, and simple air pillows behave fairly elastically when sat on, and create a feeling of instability during acceleration and deceleration of the human body. Because they are elastic, they tend to transmit the energy of dynamic loading to the body via the gluteal tissues for dissipation. Visoelastic materials such as foam have the capacity to absorb energy on impact, which results in a real and perceived sense of stability.22 In this study, the generic contoured foam (ISCUS) seemed more stable than the flat-foam and the Roho cushions. The concave shape of the contoured foam cushion could have an effect on wheelchair users’ stability in terms of lowering the center of gravity. In fact, the location of the center of gravity is different between able-bodied subjects and wheelchair users. Individuals with paraplegia have a larger upper extremity mass compared with their lower extremities. The center of gravity of individuals with paraplegia has been estimated to be about 5% of the body length (3– 4 vertebral levels) higher than that of able-bodied subjects.35 This change in center of gravity would contribute to a change in their limits of stability. CONCLUSION This study provides an objective method to assess the dynamic stability of wheelchair users when they perform ADLs such as reaching tasks. The relationship between the instantaneous velocity and the distance covered by the COP, at the body-seat interface, gives insight information about dynamic stability. It has been shown here that a seat cushion is said to be stable when it allows the COP to cover a larger distance with higher speed during voluntary reaching tasks. These findings have implications for wheelchair seating recommendations and especially for seat cushion selection. Acknowledgment: The authors thank Margot Lacroix for her assistance in revising the text. References 1. Zacharkow D. Posture: sitting, standing, chair design and exercise. Springfield (IL): Charles C Thomas; 1988. p 50-76. 2. Riley PO, Benda BJ, Gill-Body KM, Krebs DE. Phase plane analysis of stability in quiet standing. J Rehabil Res Dev 1995; 32:227-35. 3. Dean CM, Shepherd RB. Task-related training improves performance of seated reaching tasks after stroke. Stroke 1997; 28:722-8. 4. McClenaghan BA. Sitting stability of selected subjects with cerebral palsy. Clin Biomech 1989;4:213-6. 5. Reid DT, Sochaniwskyj A, Milner M. An investigation of postural sway in sitting of normal children and children with neurological disorders. Phys Occup Ther Pediatr 1991;11:19-35. 6. Sochaniwskyj A, Koheil R, Bablich K, Milner M, Lotto W. Dynamic monitoring of sitting posture for children with spastic cerebral palsy. Clin Biomech 1991;6:161-7. ¨ , C¸akci A. Angular 7. Dursun E, Hamamci N, Do¨nmez S, Tu¨zu¨nalp O biofeedback device for sitting balance of stroke patients. Stroke 1996;8:1354-7. 8. Chari VR, Kirby RL. Lower limb influence on sitting balance while reaching forward. Arch Phys Med Rehabil 1986;67: 730-3. Arch Phys Med Rehabil Vol 82, February 2001
9. Bendix T, Jessen F, Krohn L. Biomechanics of forward-reaching movements while sitting on fixed forward or backward-inclining or tiltable seats. Spine 1988;13:193-6. 10. Son K, Miller JAA, Schultz AB. The mechanical role of the trunk and lower extremities in a seated weight-moving task in the sagittal plane. J Biomech Eng 1988;110:97-103. 11. Moore S, Brunt D. Effects of trunk support and target distance on postural adjustments prior to a rapid reaching task by seated subjects. Arch Phys Med Rehabil 1991;72:638-41. 12. Curtis KA, Kindlin CM, Reich KM, White DE. Functional reach in wheelchair users: the effects of trunk and lower extremity stabilization. Arch Phys Med Rehabil 1995;76:360-7. 13. Seelen HAM, Potten YJM, Huson A, Spaans F, Reulen JPH. Impaired balance control in paraplegic subjects. J Electromyogr Kinesiol 1997;7:149-60. 14. Bernard P-L, Peruchon E, Micallef J-P, Hertog C, Rabischong P. Balance and stabilization capability of paraplegic wheelchair athletes. J Rehabil Res Dev 1994;31:287-96. 15. Cooper RA, Stewart KJ, VanSickle DP. Evaluation of methods for determining rearward static stability of manual wheelchairs. J Rehabil Res Dev 1994;31:144-7. 16. Kirby RL, Sampson MT, Thoren FAV, MacLeod DA. Wheelchair stability: effect of body position. J Rehabil Res Dev 1995;32:36772. 17. Chesney DA, Axelson PW. Measuring functional changes: practical methodologies for use in clinics. Proceedings of the 13th International Seating Symposium; 1997; Pittsburgh. 1997; p 132-8. 18. Kamper D, Barin K, Parnianpour M, Reger S, Weed H. Preliminary investigation of the lateral postural stability of spinal cordinjured individuals subjected to dynamic perturbations. Spinal Cord 1999;37:40-6. 19. Allison GT, Singer KP. Assisted reach and transfers in individuals with tetraplegia: towards a solution. Spinal Cord 1997; 35:217-22. 20. Seelen HAM. Reorganisation of postural control in spinal cord injured persons [dissertation]. Maastricht (Netherlands): Univ of Maastricht; 1997. 21. Staarink HAM. Sitting posture, comfort and pressure. Delft: Delft Univ Pr; 1995. p 169-89. 22. Ferguson-Pell MW. Seat cushion selection. J Res Rehabil Dev 1990;(Suppl 2):49-73. 23. Pai Y-C, Patton J. Center of mass velocity-position predictions for balance control. J Biomech 1997;30:347-54. 24. Wolff DR, Rose J, Jones VK, Bloch DA, Oehlert JW, Gamble JG. Postural balance measurements for children and adolescents. J Orthop Res 1998;16:271-5. 25. Garber SL, Dyerly LR. Wheelchair cushions for persons with spinal cord injury: an update. Am J Occup Ther 1991;45: 550-4. 26. Sprigle SH, Faisant TE, Chung K-C. Clinical evaluation of custom-contoured cushions for the spinal cord injured. Arch Phys Med Rehabil 1991;71:655-8. 27. Rosenthal MJ, Felton RM, Hileman DL, Lee M, Friedman M, Navach JH. A wheelchair cushion designed to redistribute sites of sitting pressure. Arch Phys Med Rehabil 1996;77:278-82. 28. Brienza DM, Karg PE. Seat cushion optimization: a comparison of interface pressure and tissue stiffness characteristics for spinal cord injured and elderly patients. Arch Phys Med Rehabil 1998; 79:388-94. 29. Do MC, Bouisset S, Moynot C. Are paraplegics handicapped in the execution of a manual task? Ergonomics 1985;28:136575. 30. Lino F, Bouisset B, Duchene JL. Effect of seat cushion area on the velocity of a pointing task. In: Bellotti P, Cappozzo A, editors. Biomechanics: Proceedings of the VIIIth Meeting of the European Society of Biomechanics; 1992, June 21-24. Rome: Universita´ Sapienza; 1992. 31. Dean CM, Shepherd RB, Adams RD. Sitting balance II: reach direction and thigh support affect the contribution of the lower limbs when reaching beyond arm’s length in sitting. Gait Posture 1999;10:147-53.
DYNAMIC STABILITY IN SITTING, Aissaoui
32. Seelen HAM, Potten YJM, Drukker J, Reulen JPH, Pons C. Development of new muscle synergies in postural control in spinal cord injured subjects. J Electromyogr Kinesiol 1998;8:23-34. 33. Winter DA. Biomechanics and control of human movement. New York: John Wiley; 1990 p 93-6. 34. Kernozek TW, Lewin JE. Seat interface pressure of individuals with paraplegia: influence of dynamic wheelchair locomotion compared with static seated measurements. Arch Phys Med Rehabil 1998;79:313-6. 35. Duval-Beaupere G, Robain G. Upward displacement of the center of gravity in paraplegic patients. Paraplegia 1991;29:309-17.
281
Suppliers a. HIGH PROFILE威 Cushion; ROHO Inc, 100 N Florida Ave, Belleville, IL 62221-5429. b. ISCUS and flat-foam cushion; Orthofab Inc, 10370 Louis-H Lafontaine, Anjou, Que H1J 2T3, Canada. c. Vista Medical Ltd, 120 Maryland St, Winnipeg, Man R3G 1L1, Canada. d. Advanced Mechanical Technology Inc, 151 California St, Newton, MA 02158. e. STATISTICA 5.0; StatSoft, Inc, 2300 E 14th St, Tulsa, OK 74104. f. SERTIX, Inc, 1 Madison St, East Rutherford, NJ 07073. g. AMG Medical, Inc, 8505 Dalton, Montreal, Que H4T 1V5, Canada.
Arch Phys Med Rehabil Vol 82, February 2001