RESNA 26th International Annual Confence

Technology & Disability: Research, Design, Practice & Policy

June 19 to June 23, 2003
Atlanta, Georgia


Katharine J. Fronczak, Michael L. Boninger M.D., Aaron L. Souza, MS & Rory A. Cooper, Ph.D.
Dept. Rehab. Science &Technology, University of Pittsburgh, Pittsburgh, PA 15261
Dept. PM&R, University of Pittsburgh Medical Center, PA 15261
Human Engineering Research Laboratories, Highland Drive VA Medical Center, Pittsburgh, PA


Manual wheelchair users (MWU) are found to be at a high risk of developing carpal tunnel syndrome (CTS). The objective of this paper was to examine pushrim biomechanics longitudinally and its relation to the progression of median nerve damage. Twenty-three MWU with a spinal cord injury completed nerve conduction studies (NCS), at two separate times over a range of two to six years. A biomechanical analysis of wheelchair propulsion was completed at the first visit. Backward linear regression models found a significant relationship between median motor amplitude and non-planar moment peak (r2 = 0.79) and between median motor latency and resultant force (F), rate of rise of F and weight (r2 = 0.87 & 0.84). Propulsion biomechanics collected at the start of the study was able to predict progression of median nerve injury. Altering injurious biomechanics should prevent median nerve injury and wrist and hand pain.


Carpal tunnel syndrome (CTS) is caused by injury to the median nerve and is associated with high-force, high-repetition tasks commonly experienced during manual wheelchair propulsion (1). Incidence of CTS in individuals with a spinal cord injury (SCI) exceeds 50 to 60 percent (2). Individuals with a SCI find it hard to tolerate CTS, because they rely on their hands and arms for daily activities (1). Repetitive strain injuries in manual wheelchair users (MWU) have been linked in the past with high peak resultant forces and high stroke frequency (1). Boninger et al. examined 34-experienced MWU's median nerve function and the kinetics while propelling in their manual wheelchair. Correlations were found between stroke frequency and rate of rise of forces applied to the pushrim and mean median sensory amplitude as well as latency. Previous studies that found an association between the biomechanics of wheelchair propulsion and median nerve function have mentioned that longitudinal studies are needed to gain further insight into cause and effect. Can the way a subject propels a wheelchair predict their risk of injury? The purpose of this was to examine longitudinal biomechanical data and its connection to CTS in MWU.


Subjects. Twenty-three experienced manual wheelchair users (MWU) with a spinal cord injury (SCI) with an average age, weight, height and years post SCI (36.6 ± 9.8 yrs, 752.7 ± 169.8 N, 175.3 ± 9.9 cm & 11.8 ± 6.5) were recruited for this study. Written informed consent was obtained prior to data collection.

Nerve Conduction Study Data: Each subject underwent two (1st & last visit) NCS with no missing data that was similar to the methods of Boninger et al's paper (1). Median sensory amplitude (MSA) and latency (MSL) and median motor amplitude (MMA) and latency (MML) of the wrist were the specific NCS used to determine the progression of median mononeuropathy (MM).

Kinetic Data Collection: The SMARTwheel (3-D force sensing pushrim) was utilized to measure the three-dimensional forces Fx, Fy, and Fz (3,5). After each subject's wheelchair was fitted with two SMARTWheels, the wheelchair was strapped down onto an independent two-drum dynamometer (6). The Camber, pushrim diameter, and tire size were left unchanged. The subject's acclimated themselves to the testing environment by propelling their own wheelchair for 3 to 10 minutes before data collection started. The subjects were asked to propel on the dynamometer at a steady state speed of 0.9 m/s (2mph) and 1.8 m/s (4mph) for 20 seconds (4). The kinetic data was checked for errors and was filtered at 30Hz with an eighth order zero phase digital Butterworth type (5). A local coordinate system was used to determine the tangential force (Ft), radial force (Fr) and resultant force (F) (7).

Statistical analysis: Pushrim biomechanical data peak points were collected for the first five strokes and averaged and combined. Biomechanical data was normalized for body weight. The left and right sided biomechanical and NCS data were averaged together because they were highly correlated. We used a backward linear regression model with the dependent being NSC at the second visit. NCS at the first visit were forced into the model as an independent variable (IV). Other IVs studied included subject weight, height, year post injury, and wheelchair biomechanics data. All statistical calculations were performed by SPSS v10.1 statistical package (SPSS, Inc., Chicago, IL). Alpha level set at (p<0.05) for all tests performed.


Pushrim biomechanical results are found in Table 1. The results of the linear regression models can be found in Table 2. Four linear regression models were found to be significant and used for the analysis. In summary MMA at the second visit was significantly related to MMA at the first visit and the non-planar moment. This refers to the moment at the pushrim not causing forward motion. In addition, MML was significantly related to the magnitude and rate of rise of the result force as well as to subject weight.

Table 1: Pushrim biomechanics (weight normalized). (n = 23)

Biomech. Variables



0.9 m/s

1.8 m/s

Stroke frequency (Hz)

1.01 ± 0.14

1.38 ± 0.24

Velocity m/s

0.96 ± 0.14

1.70 ± 0.18

Average contact angle (θ)

97.68 15.40

110.60 ± 12.49

Resultant force

0.09 ± 0.03

0.14 ± 0.04

Non-planar moment peak (m)

0.01 ± 0.01

0.02 ± 0.01

Non-planar rate of rise (m/s)

0.16 ± 0.08

0.32 ± 0.18

Rate of rise of F (s)

1.44 ± 0.58

2.60 ± 1.05

m/s = meters per second


Table 2. Backward linear regression model.

Dependent Variable (DV)

Independent Variable (IV)



Significant p<. 05

MMA (2)

Non-planar moment peak




MMA (1)


<. 0001

MML (2)

Rate of rise of F




MML (1)


< .0001


Weight (N)



MML (2)

F (N)




MML (1)


< .0001

(1)- taken from the first NCS
(2)- taken from the last NCS DISCUSSION

Previous studies have indicated a relationship between pushrim biomechanics and carpal tunnel syndrome (CTS) and have suggested that longitudinal data should be analyzed to prove a direct cause and effect relationship. This study is the first to find such a link. The way an individual propelled their wheelchair at time 1 predicted worsening of median nerve function over a course of 2 to 6 years. It is interesting to note that significant relationships were only found at the 0.9m/s speed. This might be because manual wheelchair users (MWU) typically propel closer to 0.9 m/s (near walking speed) then 1.8m/s. One of the limitations of this study is a small n of twenty-three. A greater number of participants would allow us to develop complex models. Having stated this, the fact that we found a significant relationship with this small an n is impressive.


The greater the forces and how fast a participant reaches those forces can lead to median nerve damage and may explain the high incidence of CTS in MWU and why CTS progresses over time. Based on the results of this study, clinicians can advise patients on appropriate propulsion techniques. Wheelchair users should take long, smooth strokes that minimize the peak forces and don't rapidly load the wrist. In addition, the twisting at the wrist should be minimized. Further intervention studies are needed to see if this type of wheelchair propulsion technique can really prevent injury.


  1. Boninger, ML, Cooper, RA, Baldwin ,MA, Shimada, SD, & Koontz, AK (1999). Wheelchair Pushrim kinetics: Body weight and median nerve function. Arch Phys Med Rehabil, 80(8): 910-915.
  2. Veeger, HEJ, Meershoek, LS, van der Woude, LHV & Langenhoff, JM (1998). Wrist motion in handrim wheelchair propulsion. J Rehabil Res Dev, 35: 305-313.
  3. Asato, KT, Cooper, RA, Robertson, RN, Ster, JF (1993). SMARTWheels: development and testing of a system for measuring manual wheelchair propulsion dynamics. IEEE Trans Biomed Eng, 40: 1320-1324.
  4. DiGiovine, CP, Cooper, RA & Dvorznak, M (1997). Modeling and analysis of a manual wheelchair coast down protocol. In: Proceedings of the 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL. The Netherlands: Soetehoun A/V Productions; 1888-1891.
  5. Cooper, RA, DiGiovine, CP, Boninger, ML, Shimada, SD & Robertson, RN (1998). Frequency analysis of 3-dimensional pushrim forces and moments for manual wheelchair propulsion. Automedica, 16: 355-365.
  6. Vosse AJ, Cooper RA, & Dhaliwal B (1990). Computer control of a wheelchair dynamometer. In: Proceedings of the 13th Annual RESNA Conference; 1990 Jun. 15-20; Washington, DC. Washington (DC): Resna Press; p.59-60.
  7. Souza, A.L., Boninger, M.L., Koontz, A.M., Fay, B.T., & Cooper, R.A. (2000). A Kinetic Analysis of Propulsion Patterns in Manual Wheelchair Users. Proceedings of the 23rd Annual RESNA Conference, Orlando, FL. 405-407.


The study was supported partly by the Department of Veteran Affairs, National Institutes of Health (NIH), National Institute for Disability and Rehabilitation Research, Paralyzed Veterans of America (PVA), and Eastern Paralyzed Veterans of America (EPVA).

RESNA Conference Logo