SEATED POSTURAL STABILITY OF WHEELCHAIR PASSENGERS IN MOTOR VEHICLES

Derek G. Kamper+, Maureen A. Linden, Thomas C. Adams, Steven I. Reger, Ph.D. Cleveland Clinic Foundation, Dept. of Physical Medicine and Rehabilitation, Cleveland, OH + Biomedical Engineering Center, Ohio State University, Columbus, OH

ABSTRACT

The seated postural stabilities during driving maneuvers of quadriplegic, paraplegic, and able- bodied individuals were compared. Testing, conducted with the subject seated in a wheelchair in a 22-ft. van, consisted of left turns performed at two different speeds resulting in sustained centrifugal acceleration levels of either 0.2g or 0.4g. Subject response was captured in video recordings while vehicle acceleration was measured with a triaxial accelerometer. The quadriplegic subjects lost stability for all test runs, while the paraplegics withstood all the 0.2g and some of the 0.4g runs. The able-bodied subjects maintained balance for all runs. Better stability in the driving maneuvers correlated with an ability to withstand greater center of gravity displacement in static testing. For this small study seat cushion type had no significant effect on response.

BACKGROUND

While much clinical effort is expended to properly seat wheelchair users, the issues addressed primarily pertain to the static situation in which the client is at rest. New concerns may arise when the wheelchair user enters a dynamic environment, such as that inside a motor vehicle. Forces experienced inside a vehicle may greatly challenge the seated postural stability of a passenger or driver, thereby leading to fatigue, discomfort, or the jeopardizing of safety. For individuals with reduced postural control, such as those with SCI, difficulties may arise even during normal driving conditions. When Sprigle and Linden tested the response of quadriplegic subjects to various driving maneuvers, they found that the subjects had trouble staying upright at very low acceleration levels1-2. However, they were unable to collect any kinematic data and did not examine any other disability groups. The intent of this study was to begin to gather data, including kinematic information, on the effects of injury level and seat surface on postural control during driving maneuvers. Also, the authors wished to examine the efficacy of use of a static quantification of stability in predicting responses in the dynamic situation.

APPROACH

This study assessed postural control without use of the arms, thereby simulating events in which balance must be maintained while the arms are occupied with other tasks. Thus, instability was defined as the assumption of a posture from which a subject could not return to his resting posture without use of his arms. Balance was considered lost when a subject had to use his arms to prevent himself from becoming unstable. A means for quantifying seated stability also had to be established. For the static environment, movement of the center of gravity, cg, was employed. The same wheelchair used in the vehicle testing was mounted to four load cells, as shown in figure 1. The subject rotated his trunk as far forward as possible up to the point of instability, and the corresponding cg movement was calculated. The protocol was then repeated with the subject leaning laterally.

Figure 1: Apparatus for measuring maximum cg displacements which subject can withstand before losing stability.

In the dynamic environment degrees of stability were roughly quantified by conducting the left turn maneuvers at different speeds; this resulted in different lateral vehicle accelerations and, thus, larger disturbing forces being applied to the subject. Loss or maintenance of balance for a run at a given acceleration level was recorded. Also, if instability resulted, the amount of time for which the subject remained upright during the turn was computed.

METHODS

After the procedure for measuring balance in the static environment was completed, as outlined in the previous section, the experiments in the vehicle were conducted. The independent variables were the subject's functional level, the type of seat cushion, and the level of lateral vehicle acceleration. Chosen values are specified in table 2. Three runs were conducted for each combination. Two male subjects from each of the three functional groups, quadriplegic, high paraplegic, and able-bodied, participated in this study. Subjects were instructed to keep their arms crossed in front of them while trying to maintain an upright posture during the turning maneuvers. Runs were conducted both with the subject seated on a 3" foam cushion with a plywood base and on a high profile ROHO cushion. The seats were incorporated into a standard manual wheelchair with a sling back. The wheelchair, facing the front of the vehicle, was rigidly mounted in the bay of a 22-ft van to precluded possible confounding from wheelchair slide and tip. Constant radius turning maneuvers were performed at two different speeds to create sustained 0.2g and 0.4g acceleration inputs to the wheelchair- subject system for 8 and 5 seconds, respectively. A sample curve of lateral vehicle acceleration vs. time for a 0.4g turn is shown in figure 2. The 0.2g and 0.4g acceleration levels were chosen to represent the range seen in normal driving conditions as determined by on-the-road testing by this research group3. A video camera mounted on the van recorded subject response. Reflective markers were placed at the positions of the subject's ASISs, sternum, and coracoid processes. Images of initial position, voluntary response, and maximal displacement were digitized and analyzed for angles and displacements. Vehicle acceleration was measured with a triaxial accelerometer. A ramp voltage which was simultaneously sampled and displayed on a multimeter allowed for synchronization of acceleration and video data.

Figure 2: Sample vehicle lateral acceleration profile. Darker curve displays moving time average of raw data.

RESULTS

Static The limits of cg excursion before loss of stability are listed in table 1. DXcg represents the displacement of the subject's center of gravity frontwards as the subject leaned in that direction. DYcg signifies the movement of the cg in the lateral direction as subject leans to his right. The able- bodied subjects, when leaning forward, were limited by the chest contacting the thighs rather than loss of stability.

Table 1: Cg movement subject can withstand before losing stability

Functional Subject Level DXcg DYcg MB C5-6 0.3" 0.8"

TT C7 0.2" 1.0"

JW T4-5 1.3" 2.3"

TC T5 1.1" 2.2"

TA AB 4.5" 8.9"

MP AB 7.5" 6.2"

AB : able-bodied subject

Dynamic The results of the runs in the vehicle are summarized in table 2. The one set of tests conducted at the 0.4g level with the able-bodied subject on the ROHO produced results so similar to those observed with the foam cushion that no further testing was done with the able-bodied individuals on ROHO cushions. The paraplegics, while able to withstand all of the 0.2g turns, seemingly were pushed to their stability limit during the 0.4g turns. In the turns in which balance was lost, control was sustained until at least 50% of the turn was completed. One of the quadriplegics, MB, was also able to retain a stable posture through part of the turn, but only at the 0.2g level. On average he lost equilibrium 40% of the way through the turn. Table 2: Number of runs for which subject remained stable Functional Foam Cushion ROHO Cushion Subject Level 0.2g 0.4g 0.2g 0.4g MB C5-6 0 0 0 0

TT C7 0 0 0 0

JW T4-5 3 2 3 2

TC T5 3 0 3 2

TA AB 3 3 - 3

MP AB 3 3 - -

AB : able-bodied subject

Preliminary analysis of the kinematic data did not reveal any striking distinctions in body posture among the three subject groups or with use of the different cushions. Actually, initial postural response to the turn was quite similar in 5 of the 6 participants. These subjects compensated for the centrifugal forces by rotating their shoulders and upper torso clockwise 5-15 into the left turn. Figure 3 depicts this body position in an image capturing one of the quadriplegic subjects just before he enters the turn.

DISCUSSION

While the sample size for this study was too small to allow generalization of the results, some trends became apparent which warrant further investigation. The three disability groups had very distinct responses both in the static and dynamic tests. The paraplegic subjects displayed a greater ability to maintain equilibrium while shifting their center of gravities in the static testing than the quadriplegic subjects. While the increased cg displacement was small when compared to the values attained by the able-bodied participants, it translated into significantly improved postural control during the vehicle testing in comparison to the quadriplegic subjects. Our results showed no significant overall differences between responses obtained with the subject seated on the ROHO or foam cushions. Preliminary investigation of body angle and displacement data suggested some similarity in seated postural control strategy among the different subject groups. Further testing may lead to the identification of representative kinematic responses.

Figure 3: MB entering 0.2g left turn. Arrow represents direction of centrifugal acceleration.

REFERENCES

1) Sprigle, S. and Linden, M., "Accelerations Experienced by Wheelchair Users with SCI in a Moving Van", Proceedings of the Rehabilitation Engineering Society of North America 1994 Annual Conference, Nashville, Tennessee, 55-57, 1994.

2) Linden, M. and Sprigle, S., "Development of Instrumentation and Protocol to Measure the Dynamic Environment of a Modified Van", Journal of Rehabilitation Research and Development, 33(1), 1996.

3) Adams, T., presentation to the SAE Wheelchair Restraint Task Group, Dearborn, MI, August 25, 1995.

ACKNOWLEDGMENTS

The authors wish to acknowledge project funding through TCRP Project C-1, administered by the Transportation Research Board and through the Center for Automotive Research at the Ohio State University.

Derek Kamper U1-106 9500 Euclid Avenue Cleveland, OH 44195 (216) 445-5064

Postural stability