University of Southern California Dissertations and Theses

Upper extremity control and dynamics during manual wheelchair propulsion at different speeds and wheelchair configurations

(USC Thesis Other)

Upper extremity control and dynamics during manual wheelchair propulsion at different speeds and wheelchair configurations

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1
Upper Extremity Control and Dynamics
during Manual Wheelchair Propulsion at
Different Speeds and Wheelchair
Configurations

Ian M. Russell
Dissertation Chair: Jill L. McNitt-Gray
Dissertation Committee Members: David D’Argenio, Rand R. Wilcox
Degree Conferred: Doctor of Philosophy (Biomedical Engineering)
Conferring Body: Faculty of the USC Graduate School
University of Southern California
Defense Date: June 13, 2018
Degree Conferral Date: August 2018

2
Dedicated to my family,
Mom, Miles, Clay, and Geraldine
You have always encouraged me to pursue my dreams and I could not have accomplished this goal
without your support.

ACKNOWLEDGMENTS
I wish to express my sincerest thanks to:
 My advisor Dr. Jill McNitt-Gray for her wisdom and guidance
 The USC Biomechanics Lab (Ed, Travis, Sneha, Antonia, Korkut, Nate, Marisa, Casey, Sam, Justin,
and Jennifer) for their support and friendship
 The ARCS Foundation for their financial support
3
Table of Contents
Chapter 1: Introduction ................................................................................................................................ 6
Mechanics of Wheelchair Propulsion ....................................................................................................... 7
Proposed Set of Experiments .................................................................................................................... 8
References .............................................................................................................................................. 10
Chapter 2: Specific Aims and Hypotheses................................................................................................... 12
Specific Aim 1: Modifications in wheelchair propulsion technique with speed (Chapter 4) .................. 12
Specific Aim 2: Parameterization of the shoulder net joint moment to characterize the mechanical
demand on an individual’s shoulder during manual wheelchair propulsion (Chapter 5) ...................... 13
Specific Aim 3: Determine how wheelchair reconfiguration to address posture, pressure, and stability
affects joint kinetics and torso posture during manual wheelchair propulsion (Chapter 6) .................. 14
Specific Aim 4: Determine how different WC collection scenarios with different stability affect
wheelchair propulsion technique (Chapter 7) ........................................................................................ 15
References .............................................................................................................................................. 17
Chapter 3: Experimental Design ................................................................................................................. 20
Experimental Procedures ........................................................................................................................ 20
Inclusion Criteria ..................................................................................................................................... 20
Exclusion Criteria..................................................................................................................................... 20
Experimental Tasks ................................................................................................................................. 20
Indoor Ergometer ................................................................................................................................ 20
Outdoor Sidewalk................................................................................................................................ 20
Data Collection ........................................................................................................................................ 21
Kinematics ........................................................................................................................................... 21
Lab Setup and Coordinate System .......................................................................................................... 22
Ergometer Indoor Lab Testing ............................................................................................................ 22
Sidewalk Outdoor Testing ................................................................................................................... 22
Data Processing and Analysis .................................................................................................................. 24
Kinematics ........................................................................................................................................... 24
Kinetics ................................................................................................................................................ 27
References .............................................................................................................................................. 28
Chapter 4: Modifications in wheelchair propulsion technique with speed ................................................ 29
Abstract ................................................................................................................................................... 29
4
Introduction ............................................................................................................................................ 30
Methods .................................................................................................................................................. 31
Participants ......................................................................................................................................... 31
Instrumentation .................................................................................................................................. 31
Data Collection .................................................................................................................................... 32
Experimental Protocol ........................................................................................................................ 32
Data Processing and Analysis .............................................................................................................. 32
Statistics .............................................................................................................................................. 34
Results ..................................................................................................................................................... 34
Discussion................................................................................................................................................ 43
Acknowledgements ................................................................................................................................. 45
Suppliers .................................................................................................................................................. 45
References .............................................................................................................................................. 46
Chapter 5: Parameterization of the shoulder net joint moment ................................................................ 49
using four functional axes ........................................................................................................................... 49
Introduction ............................................................................................................................................ 49
Methods .................................................................................................................................................. 50
Results ..................................................................................................................................................... 52
Discussion................................................................................................................................................ 59
References .............................................................................................................................................. 61
Chapter 6: Modifications in wheelchair propulsion technique following clinical wheelchair adjustment 63
Introduction ............................................................................................................................................ 63
Methods .................................................................................................................................................. 64
Participants ......................................................................................................................................... 64
Instrumentation .................................................................................................................................. 64
Experimental Protocol ........................................................................................................................ 65
Data Processing and Analysis .............................................................................................................. 65
Statistics .............................................................................................................................................. 66
Results ..................................................................................................................................................... 67
Discussion................................................................................................................................................ 79
References .............................................................................................................................................. 82
5
Chapter 7: Modifications in wheelchair propulsion technique between ergometer and overground
wheelchair propulsion ................................................................................................................................ 84
Introduction ............................................................................................................................................ 84
Methods .................................................................................................................................................. 85
Participant ........................................................................................................................................... 85
Ergometer Instrumentation ................................................................................................................ 85
Ergometer Protocol ............................................................................................................................. 85
Overground Instrumentation .............................................................................................................. 86
Overground Protocol .......................................................................................................................... 86
Data Processing and Analysis .............................................................................................................. 86
Statistics .............................................................................................................................................. 87
Results ..................................................................................................................................................... 88
Discussion................................................................................................................................................ 93
References .............................................................................................................................................. 95
Chapter 8: Summary of Studies .................................................................................................................. 97
Chapter 9: Conclusions ............................................................................................................................. 100

6
Chapter 1: Introduction

Manual self-propelled wheelchairs (WC) have existed for centuries allowing those who are unable to
walk due to disability or injury, to be mobile without the use of their legs. Today, manual WC provide an
effective form of low-cost wheeled mobility by preserving upper body strength, cardiovascular
conditioning, independence, and participation in the community especially among individuals with
spinal cord injury (SCI). However, repetitive mechanical loading of the upper extremity is often
associated with pain, dysfunction, and poor health-rated quality of life. Shoulder pain, in individuals with
paraplegia, can occur within the first year and the incidence increases with time post- injury (35% at 5
years, 70% at 20 years).
1

While a direct causal link has not been established between mechanical loading and shoulder pain and
pathology, researchers and clinicians have attributed the prevalence of shoulder pain and pathology in
the SCI population (~300,000)
2
to the repetitive mechanical load imposed on the upper limb due to
lower extremity paralysis.
3,4
In contrast to the able-bodied population, individuals with SCI who develop
shoulder pain are unable to rest their arms because they are fully dependent on their upper extremities
for both locomotion and daily activities. Repetitive loading of shoulder structures without adequate
opportunity for recovery can lead to injury, pain, and a devastating loss of function and independence.
Due to the detrimental impact on functional mobility and the difficulty in treatment of shoulder pain
once it occurs, understanding the control and dynamics of the upper extremity during manual WC
propulsion under different conditions is important for improving future patient outcomes.

Structural stability of the shoulder joint is provided by a shallow humeral head socket (glenoid cavity)
and a fibrous labrum.
5
During WC propulsion, the elbow is positioned below the shoulder. In this
segment configuration, the joint capsule tends to loosen and the reinforcing ligaments can become
slack, thereby creating the need for shoulder muscles to maintain stability of the glenohumeral joint.
6

Simultaneously, activation of the upper extremity muscles must be coordinated to control the scapula
and produce the shoulder and elbow net joint moments (NJM) needed to generate propulsive forces on
the pushrim.
7–9
Imposing both joint stability and moment generation requirements on muscles of the
upper extremity during WC propulsion increases the susceptibility to neuromuscular fatigue.
8,9
A
weakened muscle within the shoulder-girdle-complex can result in inadequate dynamic stability of the
system particularly during intervals when large forces applied to the pushrim are required during WC
propulsion.
10
Loss of dynamic stability causes stress on the shoulder structures and other joints of the
upper limb and can lead to the development of shoulder pain.
3,11–13

Current experimental research and clinical guidelines promote improved interaction between the
individual and the WC as a means to mitigate detrimental mechanical loading of the shoulders.
14
Manual
WC propulsion requires generation of a force on the pushrim with a component tangential to the wheel.
The mechanical demand (Net Joint Force (NJF), Net Joint Moment (NJM)) imposed across the upper
extremities, however, is affected by orientation of the resultant reaction force (RF) relative to the body
segments.
15,16
Model simulation results indicate a 2-fold increase in RF magnitude can result in minimal
7
changes in shoulder NJM magnitude without decrements in performance if the orientation of the RF
relative to the upper extremity segments is modified.
15

Mechanics of Wheelchair Propulsion

Manual WC propulsion is a cyclic task that requires repetitive generation of propulsive forces on the
pushrim of the WC. Generation of these forces on the pushrim involves coordinated activation of
muscles responsible for simultaneously maintaining shoulder joint stability and controlling shoulder
rotation. During WC propulsion, RF at the hand/pushrim interface varies in magnitude, direction and
duration between individuals, between tasks and across different performance conditions (e.g. free, fast
steady state velocities, acceleration phases, sustained propulsion up inclines etc.).
During push, the propulsion pattern used by an individual likely reflects a balance between the
mechanical requirements of the task (e.g. generation of a tangential force on the handrim to rotate the
wheel) and the biomechanical possibilities afforded by the individual’s musculoskeletal system
(distribution of the mechanical load resulting from the orientation of the RF generated relative to the
upper extremity segments).
17
In theory, the most effective way to rotate a wheel is to orient the applied
force to be tangential to the wheel. This force applied to the pushrim, however, produces an equal and
opposite RF back onto the upper extremity (Figure 1). If the orientation of the RF passes anterior to the
center of mass (CM) of the forearm, the NJM at the elbow will be an extensor NJM. If the RF passes
posterior to the CM of the forearm, a flexor NJM will be required. The direction of the NJM, along with
the proximal and distal moments imposed on the upper arm by the NJFs at the elbow and shoulder will
also simultaneously affect the magnitude of the shoulder NJM (Figure 1) .

Figure 1: Free body diagrams showing how differences in magnitude and orientation of the reaction force (RF) relative to the
upper extremity segments and the adjacent net joint moment affects the mechanical load distribution characterized by upper
extremity joint kinetics (Net Joint Force (NJF), Net Joint Moment (NJM)).
As found with able-bodied athletes and skilled workers, prevention of shoulder injury depends on
balanced muscle control.
18–20
The rotator cuff muscles contribute to stability of the glenohumeral joint
8
by creating a neuromuscular controlled sleeve around the joint. Control of the trunk and scapula relative
to the humerus without reduction in the subacromial space also requires coordinated muscle activation
of muscles attached to the trunk including the pectoralis major, latissimus dorsi, and serratus anterior.
When the mechanical demand imposed on the shoulder (NJF and NJM) exceeds the muscle force
generating capacity of the involved muscle groups, a loss of control may lead to prolonged impingement
associated with degenerative changes in the rotator cuff tendons.
21
Recent results also indicate that
scapular kinematics patterns during weight relief and manual WC propulsion tends to be specific to the
individual WC user.
22
WC users with paraplegia move toward anterior tilt, downward rotation, and
protraction positions when RFs increase in weight relief and the push phase of WC propulsion.
22

Corresponding reductions in the subacromial space during RF generation are considered a risk factor for
impingement (supraspinatus) and a common source of shoulder pain in individuals with SCI.
22
Recent
work by Westerhoff et al.
23
, in which shoulder joint contact loads were measured during manual WC
propulsion, using instrumented shoulder prostheses, also indicates that shoulder joint contact loading
varies with between subject differences in RF direction, upper extremity segment configuration, and
muscle activation during push phase of manual WC propulsion.

Proposed Set of Experiments

This body of work aims to determine how upper extremity control and dynamics during manual WC
propulsion is affected by changes in speed and seating configuration. Due to the detrimental impact on
mobility and the difficulty in treatment of shoulder pain, exploring upper extremity loading
consequences of propulsion strategies are important to eventually improving patient outcomes. By
determining how individual WC users accommodate expected increases in mechanical demands we can
begin to identify effective multijoint control strategies for achieving desired performance outcomes
while mitigating detrimental mechanical loading of the shoulder. A detailed understanding of upper
extremity demand during propulsion is necessary for describing an individual’s interaction with their WC
and identifying how to maintain shoulder health. However, the nature of the shoulder joint often lends
to complex multi-planar movements that are difficult to fully describe anatomically. By using a novel
method of parsing shoulder net joint moment, we aim to get a clearer understanding of the mechanical
demand imposed on the shoulder during WC propulsion. Clinical fitting visits for WC users aim to
reconfigure the WC to address issues with posture, pressure, and stability. By comparing how individual
WC users modify technique during outdoor WC propulsion following reconfiguration, we can explore the
effect of WC configuration on control and dynamics in a real-world setting. Finally, by analyzing
propulsion mechanics on a stationary fixed ergometer apparatus and outside in a realistic setting we can
begin to investigate the potential shifts in postural control as well as joint kinetics for differing WC
stability scenarios.
9

10
References

1. Sie IH, Waters RL, Adkins RH, Gellman H. Upper extremity pain in the postrehabilitation spinal
cord injured patient. Arch Phys Med Rehabil. 2017;73(1):44-48.
doi:10.5555/uri:pii:000399939290225L.
2. SCIMS & NIDRR. Spinal Cord Injury (SCI) Facts and Figures at a Glance.; 2014.
https://www.nscisc.uab.edu/PublicDocuments/fact_figures_docs/Facts 2014.pdf.
3. Curtis K a, Drysdale G a, Lanza RD, Kolber M, Vitolo RS, West R. Shoulder pain in wheelchair users
with tetraplegia and paraplegia. Arch Phys Med Rehabil. 1999;80(4):453-457. doi:10.1016/S0003-
9993(99)90285-X.
4. Sie IH, Waters RL, Adkins RH, Gellman H. Upper extremity pain in the postrehabilitation spinal
cord injured patient. Arch Phys Med Rehabil. 1992;73(1):44-48.
doi:10.5555/URI:PII:000399939290225L.
5. Inman VT, deC. M. Saunders JB, Abbott LC. Observations On The Function Of The Shoulder Joint. J
Bone Jt Surg. 1944;26(1).
6. Mulroy SJ, Gronley JK, Newsam CJ, Perry J. Electromyographic activity of shoulder muscles during
wheelchair propulsion by paraplegic persons. Arch Phys Med Rehabil. 1996;77(2):187-193.
7. Robertson RN, Boninger ML, Cooper RA, Shimada SD. Pushrim forces and joint kinetics during
wheelchair propulsion. Arch Phys Med Rehabil. 1996;77(9):856-864. doi:10.1016/S0003-
9993(96)90270-1.
8. Kulig K, Rao SS, Mulroy SJ, et al. Shoulder Joint Kinetics During the Push Phase of Wheelchair
Propulsion. Clin Orthop Relat Res. 1998;354:132-143.
9. Koontz AM, Cooper R a, Boninger ML, Souza AL, Fay BT. Shoulder kinematics and kinetics during
two speeds of wheelchair propulsion. J Rehabil Res Dev. 2002;39(6):635-649.
10. McCully SP, Suprak DN, Kosek P, Karduna AR. Suprascapular nerve block results in a
compensatory increase in deltoid muscle activity. J Biomech. 2007;40(8):1839-1846.
doi:https://doi.org/10.1016/j.jbiomech.2006.07.010.
11. Gironda RJ, Clark M, Neugaard B, Nelson A. Upper Limb Pain in Anational Sample of Veterans
With Paraplegia. J Spinal Cord Med. 2004;27(2):120-127. doi:10.1080/10790268.2004.11753742.
12. Samuelsson KAM, Tropp H, Gerdle B. Shoulder pain and its consequences in paraplegic spinal
cord-injured, wheelchair users. Spinal Cord. 2004;42(1):41-46. doi:10.1038/sj.sc.3101490.
13. Alm M, Saraste H, Norrbrink C. Shoulder pain in persons with thoracic spinal cord injury:
Prevalence and characteristics. J Rehabil Med. 2008;40(4):277-283. doi:10.2340/16501977-0173.
14. Dalyan M, Cardenas DD, Gerard B. Upper extremity pain after spinal cord injury. Spinal Cord.
1999;37(3):191-195.
15. Munaretto JM, McNitt-Gray JL, Flashner H, Requejo PS. Simulated effect of reaction force
11
redirection on the upper extremity mechanical demand imposed during manual wheelchair
propulsion. Clin Biomech. 2012;27(3):255-262. doi:10.1016/j.clinbiomech.2011.10.001.
16. Raina S, McNitt-Gray JL, Mulroy S, Requejo PS. Effect of increased load on scapular kinematics
during manual wheelchair propulsion in individuals with paraplegia and tetraplegia. Hum Mov
Sci. 2012;31(2):397-407. doi:10.1016/j.humov.2011.05.006.
17. Veeger HEJ, Rozendaal LA, Van der Helm FCT. Load on the shoulder in low intensity wheelchair
propulsion. Clin Biomech. 2002;17(3):211-218. doi:10.1016/S0268-0033(02)00008-6.
18. Raina S, McNitt-Gray J, Mulroy S, Requejo P. Effect of choice of recovery patterns on handrim
kinetics in manual wheelchair users with paraplegia and tetraplegia. J Spinal Cord Med.
2012;35(3):148-155. doi:10.1179/2045772312Y.0000000013.
19. Kotajarvi BR, Basford JR, An K-N, Morrow DA, Kaufman KR. The Effect of Visual Biofeedback on
the Propulsion Effectiveness of Experienced Wheelchair Users. Arch Phys Med Rehabil.
2006;87(4):510-515. doi:https://doi.org/10.1016/j.apmr.2005.12.033.
20. Rice I, Gagnon D, Gallagher J, Boninger M. Hand Rim Wheelchair Propulsion Training Using
Biomechanical Real-Time Visual Feedback Based on Motor Learning Theory Principles. J Spinal
Cord Med. 2010;33(1):33-42. doi:10.1080/10790268.2010.11689672.
21. Sharkey NA, Marder RA. The Rotator Cuff Opposes Superior Translation of the Humeral Head. Am
J Sports Med. 1995;23(3):270-275. doi:10.1177/036354659502300303.
22. Raina S. Biomechanics of the upper extremity during wheelchair propulsion and weight relief
raise in wheelchair users with spinal cord injury. 2011;(December).
23. Westerhoff P, Graichen F, Bender A, et al. Measurement of shoulder joint loads during
wheelchair propulsion measured in vivo. Clin Biomech. 2011;26(10):982-989.
doi:https://doi.org/10.1016/j.clinbiomech.2011.05.017.
24. Morrow MMB, Hurd WJ, Kaufman KR, An KN. Shoulder demands in manual wheelchair users
across a spectrum of activities. J Electromyogr Kinesiol. 2010;20(1):61-67.
doi:10.1016/j.jelekin.2009.02.001.

12
Chapter 2: Specific Aims and Hypotheses

Background: As part of daily living, manual wheelchair (WC) users need to regulate WC propulsion
speed and can often encounter scenarios where increased speed is necessary, such as when in a rush.
Investigating the mechanical demands during different propulsion scenarios is important to
understanding load on the upper extremity experienced by manual WC users
1
. On average, increases in
propulsion speed has been reported to significantly increase reaction force (RF) magnitudes, decrease
hand contact duration, affect wrist angular position on pushrim
2–4
and influence the mechanical demand
imposed on muscles controlling shoulder stabilization and rotation during WC propulsion
2,3
. Increases in
WC propulsion speed can also lead to disproportionate increases in shoulder net joint moments (NJM)
during hand contact
4
. Understanding the different techniques individual’s use during tasks with
increased upper extremity demands is important for identifying manual WC propulsion strategies that
can help preserve shoulder function.
Specific Aim 1: Modifications in wheelchair propulsion technique with speed (Chapter
4)

Goal: Determine how individual manual WC users with paraplegia modify WC propulsion mechanics to
accommodate expected increases in RF generated at the pushrim with self-selected increases in
propulsion speed.
 Hypothesis 1: Reaction force (RF) magnitude, shoulder net joint force (NJF), and shoulder net
joint moment (NJM) during wheelchair propulsion will increase whereas contact duration will
decrease with increases in self-selected wheelchair propulsion speed
 Hypothesis 2: Orientation of RF relative to the forearm and upper arm will affect the mechanical
demand imposed on the upper extremity with increases in self-selected wheelchair propulsion
speed
 Hypothesis 3: Individuals with paraplegia will use different propulsion techniques to
accommodate the need to increase wheelchair propulsion speed

These hypotheses will be tested by comparing upper extremity joint kinetics and pushrim reaction
forces for 40 experienced manual wheelchair users with paraplegia while propelling on a stationary
ergometer at self-selected free and fast propulsion speeds. Comparison across propulsion conditions
will be calculated using a Sign Test (α = .05). Within-subject differences will also be tested using Cliff’s
analog of the Wilcoxon-Mann-Whitney test (Cliff 1996) to calculate a p-value, then apply a modified
step-down Fisher-type method to control the familywise error rate over multiple comparisons.
5–8

13
Background: Functionally relevant representations of 3D joint kinetics can assist in making
comparisons across tasks and studies. Currently, choosing the coordinate system for reporting joint
kinetics is a laboratory and study specific process which makes interpretation across studies difficult
9
.
The shoulder moment during wheelchair propulsion has similarly been represented with a variety of
coordinate systems
2–4,10–16
. These systems have involved the humeral coordinate system only, the trunk
coordinate system only, or a combination of the two
9
. Morrow et. al 2009 suggested extending Schache
and Baker’s kinetic representation for lower limb kinetics
17
to those of the upper extremity,
representing upper body efforts in the “Joint Coordinate System” (JCS). However, Morrow suggests that
in the presence of axial torque about the distal segment, moments may be improperly attributed to
proximal segment axes due to crosstalk of functional moment descriptions. An extension of the JCS
method proposed by Wagner et al. describes how motion of the humerus (relative to the torso) can be
parsed into two types of rotation: those that change where the elbow is pointing (i.e 3- torso fixed axes)
and those that change the orientation of the humerus about its longitudinal axis (axial rotation),
providing a solution to the functional crosstalk problem
18
. Thus, at the shoulder these same four-axes
can be used for kinetic descriptors: adduction-abduction, horizontal adduction-abduction, and flexion-
extension defined by the torso, and a fourth external-internal rotation axis defined by the longitudinal
axis of the upper arm pointing distally. Including an isolated external-internal NJM component when
describing shoulder kinetics illustrates how multiplanar loading of the upper extremity can be
understood through a four-axis anatomical component parameterization.

Specific Aim 2: Parameterization of the shoulder net joint moment to characterize the
mechanical demand on an individual’s shoulder during manual wheelchair propulsion
(Chapter 5)

Goal: Use a novel method of parsing the shoulder net joint moment into four axes to better understand
the mechanical demand imposed on the shoulder during manual wheelchair propulsion
 Hypothesis 1: Individuals will distribute the mechanical demand imposed on the shoulder about
different axes during wheelchair propulsion
 Hypothesis 2: Distribution of NJM will vary based on elbow position away from the torso and RF
orientation relative to the arm plane

These hypotheses will be tested by comparing upper extremity kinematics, joint kinetics, and pushrim
reaction forces for 3 example participants who are experienced manual wheelchair users with
paraplegia while propelling on a stationary ergometer at self-selected fast propulsion speed.

14
Background: Clinical guidelines, based on clinical and epidemiologic evidence, ergonomics, and expert
opinions, recommend customized WC seating as a promising means to mitigate loading of the shoulder
in manual WC users.
19
These recommendations align with the National Research Council and the
Institute of Medicine report on Musculoskeletal Disorders and the Workplace (2001) which indicates
that modifications to task performance can reduce the incidence of pain and cumulative trauma
disorders in the workplace.
20
Ongoing clinical research using experimental designs investigating specific
changes in WC seating confirms that modifications to WC seating alter multiple factors known to affect
upper extremity joint kinetics during manual WC propulsion in individuals with paraplegia.
21–26
These
multiple interacting factors include modifications to segment kinematics, reaction forces generated at
the pushrim, and neuromuscular control. These are the same factors affecting individual differences in
shoulder joint contact loads as measured during manual WC propulsion in older adults with
instrumented shoulder prostheses.
27
Implementation of customized WC seating into clinical practice
tends to be iterative with an emphasis on the individual’s posture, pressure distribution, and stability
requirements.
19
Personalizing the WC fit to the specific mobility needs of the individual WC user remains
highly dependent on the expertise of the clinician and their ability to account for these multiple
interacting factors known to affect upper extremity mechanical loading. By investigating WC propulsion
technique modifications in response to changes in WC configuration we hope to gain knowledge of WC
adjustment strategies that may prove fruitful in guiding clinical decisions that aim to identify approaches
to preserving shoulder function in individuals with spinal cord injury.
Specific Aim 3: Determine how wheelchair reconfiguration to address posture,
pressure, and stability affects joint kinetics and torso posture during manual
wheelchair propulsion (Chapter 6)

Goal: Determine if individual manual wheelchair users with paraplegia modify control and dynamics
during wheelchair propulsion following wheelchair reconfiguration intended to address posture,
pressure, and stability.
 Hypothesis 1: Reconfiguration of a WC to address posture and balance will result in differences
in shoulder height relative to wheel axle and torso angle during manual WC propulsion.
 Hypothesis 2: Shifts in orientation of the RF relative to the upper extremity will affect upper
extremity joint kinetics during manual WC propulsion.

These hypotheses will be tested by comparing upper extremity kinematics, joint kinetics, and pushrim
reaction forces for 22 manual wheelchair users with paraplegia while propelling outside the seating
center in the courtyard at Rancho Los Amigos National Rehabilitation Center at self-selected fast
propulsion speed. Comparison between baseline WC configuration and 30 days later following fitting
adjustments by a clinician will be calculated using a Wilcoxon-Mann-Whitney percentile bootstrap test
with 2000 bootstrap samples. Within-subject differences will also be tested using Cliff’s analog of the
Wilcoxon-Mann-Whitney test (Cliff 1996) to calculate a p-value, then apply a modified step-down Fisher-
type method to control the familywise error rate over multiple comparisons. Within-subject analysis will
15
be calculated an additional time using a Wilcoxon-Mann-Whitney percentile bootstrap test with 2000
bootstrap samples if the conservative step-down technique failed to reject, however still showed a p-
value less than 0.05 for a particular within-subject comparison.
5–8

Background: Propulsion collection systems like ergometers and treadmills have provided a convenient
way of collecting multiple wheelchair propulsion cycles in a controlled and stationary environment. Due
to the stationary nature of these systems, the WC is secured in place and therefore cannot tip over.
However, during real-world propulsion, this is not the case. Wheelchairs are designed to have their front
wheels free to lift off the ground in order to more easily traverse curbs. However, the WC user must be
aware and account for this degree of instability as to not inadvertently tip backwards, especially during
mechanically demanding tasks like fast or graded propulsion. Current WC seating center clinicians are
acutely aware of the importance of modifying WC stability to the user, which is why it is one of the 3
emphases of individualized WC fitting which include posture, pressure, and balance. After analyzing
modifications in propulsion technique pre and post WC fitting, Russell et al., (Chapter 6) hypothesized a
potential connection between torso posture and WC stability. From the results of their study, they
observed shifts towards more upright posture in wheelchair users who had adjustments specifically
targeting complaints of an unstable wheelchair. Russell et al. further proposed that the hypothesis of
torso posture relating to stability could be further investigated by comparing overground WC propulsion
with propulsion on an ergometer, a context where stability concerns are removed from the mechanical
demands of the task. Previous studies have compared characteristics of wheelchair propulsion in
controlled stationary systems such as ergometers, dynamometer, and treadmills with overground
propulsion. However, these studies analyzed features of the force applied to wheel, physiological
measures, or hand trajectory patterns
28–31
. Other studies have investigated trunk motion as it relates to
fatigue
32
, spinal cord injury level
33
, propulsion speed
34
, and upper-limb impairment
35
. However, no
investigation has compared torso posture or mechanical loading of the upper extremity between the
two collection methods. By investigating differences in WC propulsion technique between the collection
modalities of ergometer and overground propulsion we hope to gain insights into the effect of WC
stability on propulsion mechanics.

Specific Aim 4: Determine how different WC collection scenarios with different
stability affect wheelchair propulsion technique (Chapter 7)

Goal: Determine differences in WC propulsion mechanics and posture between a stationary fixed
ergometer apparatus and outside in a realistic setting to determine the effect of varying WC stability
scenarios.

16
 Hypothesis 1: A higher shoulder height as well as more upright torso angle during WC propulsion
on an ergometer compared to shoulder height and torso angle during overground propulsion
 Hypothesis 2: Decreased push time and reduced RF impulse during WC propulsion on an
ergometer compared to overground propulsion
 Hypothesis 3: This individual with paraplegia will use different propulsion techniques to
accommodate the different stability requirements between the two collection modalities

This hypothesis will be tested by comparing upper extremity joint kinetics, pushrim reaction forces, and
posture for one manual wheelchair users with paraplegia while propelling outside the seating center in
the courtyard at Rancho Los Amigos National Rehabilitation Center at self-selected fast propulsion
speed and on a stationary ergometer at self-selected fast propulsion speed. Comparison between
overground propulsion and ergometer characteristics will be calculated using a Wilcoxon-Mann-Whitney
percentile bootstrap test with 2000 bootstrap samples.

17
References

1. Morrow MMB, Hurd WJ, Kaufman KR, An KN. Shoulder demands in manual wheelchair users
across a spectrum of activities. J Electromyogr Kinesiol. 2010;20(1):61-67.
doi:10.1016/j.jelekin.2009.02.001.
2. Kulig K, Rao SS, Mulroy SJ, et al. Shoulder Joint Kinetics During the Push Phase of Wheelchair
Propulsion. Clin Orthop Relat Res. 1998;354:132-143.
3. Koontz AM, Cooper R a, Boninger ML, Souza AL, Fay BT. Shoulder kinematics and kinetics during
two speeds of wheelchair propulsion. J Rehabil Res Dev. 2002;39(6):635-649.
4. Veeger HEJ, Rozendaal LA, Van der Helm FCT. Load on the shoulder in low intensity wheelchair
propulsion. Clin Biomech. 2002;17(3):211-218. doi:10.1016/S0268-0033(02)00008-6.
5. Cliff N. Answering Ordinal Questions with Ordinal Data Using Ordinal Statistics. Multivariate
Behav Res. 1996;31(3):331-350. doi:10.1207/s15327906mbr3103_4.
6. Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika.
1988;75(4):800-802. doi:10.1093/biomet/75.4.800.
7. Hochberg Y, Tamhane AC. Multiple Comparison Procedures. John Wiley & Sons, Inc.
doi:10.1002/9780470316672.ch1.
8. Wilcox RR, Clark F. Robust Multiple Comparisons Based On Combined Probabilities From
Independent Tests. J Data Sci. 2015;13(1):1-11.
9. Morrow MMB, Hurd WJ, Kaufman KR, An K-N. UPPER EXTREMITY JOINT KINETICS EXPRESSION
DURING WHEELCHAIR PROPULSION. J Rehabil Res Dev. 2009;46(7):939-944.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2857327/.
10. Collinger JL, Boninger ML, Koontz AM, et al. Shoulder Biomechanics During the Push Phase of
Wheelchair Propulsion: A Multisite Study of Persons With Paraplegia. Arch Phys Med Rehabil.
2008;89(4):667-676. doi:10.1016/j.apmr.2007.09.052.
11. Desroches G, Aissaoui R, Bourbonnais D. Relationship Between Resultant Force at the Pushrim
and the Net Shoulder Joint Moments During Manual Wheelchair Propulsion in Elderly Persons.
Arch Phys Med Rehabil. 2008;89(6):1155-1161. doi:10.1016/j.apmr.2007.10.040.
12. Mercer JL, Boninger M, Koontz A, Ren D, Dyson-Hudson T, Cooper R. Shoulder joint kinetics and
pathology in manual wheelchair users. Clin Biomech. 2006;21(8):781-789.
doi:https://doi.org/10.1016/j.clinbiomech.2006.04.010.
13. van Drongelen S, van der Woude LH, Janssen TW, Angenot EL, Chadwick EK, Veeger DH.
Mechanical Load on the Upper Extremity During Wheelchair Activities. Arch Phys Med Rehabil.
2005;86(6):1214-1220. doi:10.1016/j.apmr.2004.09.023.
14. Sabick MB, Kotajarvi BR, An K-N. A new method to quantify demand on the upper extremity
during manual wheelchair propulsion. Arch Phys Med Rehabil. 2004;85(7):1151-1159.
http://www.ncbi.nlm.nih.gov/pubmed/15241767. Accessed June 3, 2018.
18
15. Boninger ML, Cooper RA, Robertson RN, Rudy TE. Wrist biomechanics during two speeds of
wheelchair propulsion: an analysis using a local coordinate system. Arch Phys Med Rehabil.
1997;78(4):364-372. http://www.ncbi.nlm.nih.gov/pubmed/9111455. Accessed June 3, 2018.
16. Cooper RA, Boninger ML, Shimada SD, Lawrence BM. Glenohumeral joint kinematics and kinetics
for three coordinate system representations during wheelchair propulsion. Am J Phys Med
Rehabil. 78(5):435-446. http://www.ncbi.nlm.nih.gov/pubmed/10493454. Accessed June 3, 2018.
17. Schache AG, Baker R. On the expression of joint moments during gait. Gait Posture.
2007;25(3):440-452. doi:10.1016/j.gaitpost.2006.05.018.
18. Wagner E V, Russell IM, Muller-karger C, et al. An approach for characterizing complex
multiplanar upper extremity motion through parsed angular velocity vector components. In:
American Society of Biomechanics.; 2016:1-2.
19. Medicine PV of AC for SC. Preservation of Upper Limb Function Following Spinal Cord Injury: A
Clinical Practice Guideline for Health-Care Professionals. J Spinal Cord Med. 2005;28(5):434-470.
20. Council I of M and NR. Musculoskeletal Disorders and the Workplace. Washington, D.C.: National
Academies Press; 2001. doi:10.17226/10032.
21. Desroches G, Aissaoui R, Bourbonnais D. Effect of system tilt and seat-to-backrest angles on load
sustained by shoulder during wheelchair propulsion. J Rehabil Res Dev. 2006;43(7):871.
doi:10.1682/JRRD.2005.12.0178.
22. Mulroy SJ, Newsam CJ, Gutierrez D, et al. Effect of Fore-Aft Seat Position on Shoulder Demands
During Wheelchair Propulsion: Part 1. A Kinetic Analysis. J Spinal Cord Med. 2005;28(3):214-221.
doi:10.1080/10790268.2005.11753815.
23. Hughes CJ, Weimar WH, Sheth PN, Brubaker CE. Biomechanics of wheelchair propulsion as a
function of seat position and user-to-chair interface. Arch Phys Med Rehabil. 2017;73(3):263-269.
doi:10.5555/uri:pii:0003999392900769.
24. Mâsse LC, Lamontagne M, O’Riain MD. Biomechanical analysis of wheelchair propulsion for
various seating positions. J Rehabil Res Dev. 1992;29(3):12—28.
25. Majaess GG, Lee Kirby R, Ackroyd-Stolarz SA, Charlebois PB. Influence of seat position on the
static and dynamic forward and rear stability of occupied wheelchairs. Arch Phys Med Rehabil.
2017;74(9):977-982. doi:10.5555/uri:pii:000399939390278I.
26. van der Woude L, Bouw A, van Wegen J, van As H, Veeger D, de Groot S. Seat height: Effects on
submaximal hand rim wheelchair performance during spinal cord injury rehabilitation. J Rehabil
Med. 2009;41(3):143-149. doi:10.2340/16501977-0296.
27. Westerhoff P, Graichen F, Bender A, et al. Measurement of shoulder joint loads during
wheelchair propulsion measured in vivo. Clin Biomech. 2011;26(10):982-989.
doi:https://doi.org/10.1016/j.clinbiomech.2011.05.017.
28. Kwarciak AM, Turner JT, Guo L, Richter WM. Comparing handrim biomechanics for treadmill and
overground wheelchair propulsion. Spinal Cord. 2010;49:457.
http://dx.doi.org/10.1038/sc.2010.149.
19
29. Koontz AM, Worobey LA, Rice IM, Collinger JL, Boninger ML. Comparison Between Overground
and Dynamometer Manual Wheelchair Propulsion. J Appl Biomech. 2012;28(4):412-419.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5450157/.
30. Mason B, Lenton J, Leicht C, Goosey-Tolfrey V. A physiological and biomechanical comparison of
over-ground, treadmill and ergometer wheelchair propulsion. J Sports Sci. 2014;32(1):78-91.
doi:10.1080/02640414.2013.807350.
31. Stephens CL, Engsberg JR. Comparison of overground and treadmill propulsion patterns of
manual wheelchair users with tetraplegia. Disabil Rehabil Assist Technol. 2010;5(6):420-427.
doi:10.3109/17483101003793420.
32. Rodgers MM, Keyser RE, Gardner ER, Russell PJ, Gorman PH. Influence of trunk flexion on
biomechanics of wheelchair propulsion - ProQuest. J Rehabil Res Dev. 2000;37(3):283-295.
33. Newsam CJ, Rao SS, Mulroy SJ, Gronley JK, Bontrager EL, Perry J. Three dimensional upper
extremity motion during manual wheelchair propulsion in men with different levels of spinal cord
injury. Gait Posture. 1999;10(3):223-232. doi:10.1016/S0966-6362(99)00034-X.
34. Julien MC, Morgan K, Stephens CL, Standeven J, Engsberg J. Trunk and neck kinematics during
overground manual wheelchair propulsion in persons with tetraplegia. Disabil Rehabil Assist
Technol. 2014;9(3):213-218. doi:10.3109/17483107.2013.775362.
35. Finley M a, Rasch EK, Keyser RE, Rodgers MM. The biomechanics of wheelchair propulsion in
individuals with and without upper-limb impairment. J Rehabil Res Dev. 2004;41(3B):385-395.
doi:10.1682/JRRD.2004.03.0385.

20
Chapter 3: Experimental Design

Experimental Procedures
Participants with complete SCI and varying experience in manual WC were recruited from the outpatient
clinics of the Rancho Los Amigos National Rehabilitation Center volunteered to participate. Each
participant was provided informed consent in accordance with the Institutional Review Board.
Individuals were excluded from participation if they reported a history of shoulder pain that altered
performance of daily activities or required medical treatment.
Inclusion Criteria
Participants in these studies must have paraplegia from complete SCI (SCI level thoracic or lower,
American Spinal Cord Injury Association A or B with no motor function below the level of SCI), be older
than or equal to 18 years of age at time of study entry, and time since spinal cord injury will differ
depending on criteria for that particular study. At entry into the study participants must be free of
shoulder pain that interferes with daily activities or requires medical intervention with a total score on
the WUSPI of 10 or less for new users study and 12 or less for ergometer study.
Exclusion Criteria
Participants in these studies must not have positive impingement signs (positive Hawkins-Kennedy test
and painful arc in shoulder abduction or flexion), biceps tendonitis (positive Speed’s test), adhesive
capsulitis, or cervical radiculopathy at the initial evaluation a history of shoulder injury or surgery or
orthopedic or neurologic disorders (other than SCI) that would impact arm function.
Experimental Tasks
Indoor Ergometer
Each participant will perform WCP at their self-selected free speed, as they do normally when traversing
a tiled floor, and at a self-selected fast speed, as if they are in a hurry to not miss an important
appointment. Preceding the start of data collection, participants will propel for 30-seconds to avoid the
propulsion initiation period. Force and kinematic data will be collected for 10 seconds (6-10 push cycles)
at each speed condition with no additional load applied to the ergometer rollers (i.e. level ground over a
tiled surface). Prior to data collection, participants will be given adequate time to become accustomed
to the wheelchair and experimental conditions.
Outdoor Sidewalk
During the initial biomechanical assessment, each participant will perform three or more trials of
approximately 6-10 WC propulsion cycles at self-selected free speed, fast speed, and on an incline in the
courtyard outside the seating center at RLANRC. Each trial will be initiated from a stationary position.
WC propulsion velocity during the initial acceleration (first 3 cycles) and at steady state velocity (middle
3 cycles) will be kept within 10% of that performed during the initial trials in the field assessment. Each
participant performed WCP at their self-selected free speed, as they do normally when traversing a tiled
floor, and at a self-selected fast speed, as if they are in a hurry to not miss an important appointment.

21
Data Collection
Kinematics
Video
Two-dimensional video will be captured simultaneously in the frontal and sagittal planes (60 Hz JVC, 120
Hz Panasonic). Sagittal video will be used to determine body segment kinematics for outdoor propulsion
collections.

Motion Capture
Three dimensional kinematics of the participants were captured using active infra-red markers and a 6
camera CODA motion analysis system (CODA Motion Analysis system, 100 Hz). Markers will be placed on
the trunk at the manubrium, the xiphoid process, the spinous process of T3 and T10 vertebrae, greater
tubercle of the humerus, lateral epicondyle, medial epicondyle, deltoid tuberosity, middle of the
forearm, radial styloid, ulnar styloid, head of the third metacarpal and head of the fifth metacarpal.
Three reflective markers were also placed on the right wheel.

Kinetics
Reaction force applied by the right arm and hand to the pushrim was measured using three strain gauge
force transducers in an instrumented wheelchair wheel at 200Hz (SmartWheel, Three Rivers Holdings,
Mesa, AZ, USA) mounted on the right side of the wheelchair.

Figure 2: Reference frame of forces applied on SmartWheel instrumented wheel.

22
Lab Setup and Coordinate System

Ergometer Indoor Lab Testing
The WC frame was positioned on a stationary ergometer with the rear wheels resting on the 2
independent rollers of the ergometer. The WC had additional support to prevent movement during
testing by hydraulic jacks in the rear and clamping down the front footrest to a support bar.
1,2

Translational inertia was simulated with flywheels acting proportional to the weight of the participant.
The rollers were coupled by means of a differential to an alternator and a modified Velodyne® bicycle-
ergometer that computer-controlled the resistance.

Figure 3: Indoor lab reference system and stationary ergometer setup.
Forces recorded from the Smartwheel were rotated into the Lab Reference system to match the
recorded kinematic reference system established with the CODA motion analysis cameras.
Sidewalk Outdoor Testing
Each WC propulsion trial was done in the participants own WC and was initiated from a stationary
position outside the seating center in the courtyard of Rancho Los Amigos National Rehabilitation Center
for a distance of about 10m. Subject upper extremity was wrapped in colorful Coban and marked with
tape to facilitate the identification of wrist, elbow, and shoulder joint centers in the sagittal video.
23
Reference system of the SmartWheel was used to establish moving WC reference system by rotating
average wheel camber of 3 degrees about the Fx direction to create a vertical, progressional (forward
direction of wheelchair translation), and perpendicular to sagittal camera right had reference system.
Manually digitized segment endpoints were used to create 3 dimensional reconstructed upper extremity
kinematics in this WC reference system.

Figure 4: Courtyard walkway where outdoor propulsion trials were collected.

Figure 5: Colorful and high contrast marking created with tape and Coban on participant upper extremity for easier segment
digitizing.
24
Data Processing and Analysis

Kinematics

Ergometer Indoor Lab Testing

Kinematic data were filtered in Visual3D using a 6th order low-pass filter with a cutoff frequency of 8Hz.
3

Four segments were constructed based on the ISB standard definitions.
4
The thorax segment was
defined using markers placed at the xiphoid, manubrium, T3 and T10 vertebrae. The upper arm
segment was constructed with the marker at the humeral head, a non-collinear marker on the upper
arm and the lateral humeral epicondyle marker. The forearm segment was created using the lateral
humeral epicondyle marker, a non-collinear marker on the forearm and the marker on the ulnar styloid
process. The hand segment was created using the markers of the radial styloid, ulnar styloid, the head of
the third metacarpal. Segment inertia parameters were based on body segment parameters in de Leva
1996.
5
For comparison between ergometer and overground propulsion, kinematic protocol below was
used for both testing scenarios.

Sidewalk Outdoor Testing

Sagittal video (60 Hz) was used to digitize upper extremity segment endpoints and wheel center using
custom MatLab code. The 2-dimensional pixel location of these endpoints were then scaled to meters
according to the physically measured length from wrist joint center to wheel center when the subject’s
hand is grabbing the pushrim and located at the 12 o’clock position. Next these 2 dimensional points
are extrapolated into 3 dimensions using the assumptions that during hand contact with the pushrim
and force generation above 5 N the elbow lies vertical to wrist. The second assumption was that the
shoulder is offset from both the wrist and elbow by a constant amount throughout the trial and
therefore it’s offset in the Z could be measured from front or rear static photos and similarly set
constant throughout the 3-dimensional reconstructed upper extremity kinematics. These two
assumptions were tested using the 3-dimensional wheelchair propulsion motion capture data described
under the “Indoor Ergometer” headings discussed previously. Position of the right shoulder joint center,
elbow joint center, and wrist joint center were calculated from tracking markers for the 10 second
propulsion trials. Lateral movement of the shoulder joint center was quantified as the difference
between the minimum and maximum positions in the Z-axis (Figure 5). Lateral range of elbow position
relative wrist was quantified as the difference between the minimum and maximum positions in the Z-
axis between elbow joint center and wrist joint center (Figure 6).
25

Figure 6: Distance range of shoulder along the Z-axis during propulsion for free (blue) and fast (red) propulsion trials.

Figure 7: Distance range between elbow joint center and wrist joint center along the Z-axis during propulsion for free (blue) and
fast (red) propulsion trials.
For the 107 subjects collected, shoulder lateral range along the Z-axis tended to be below 6 centimeters
for most subjects. For the same subject group lateral distance along the Z-axis between elbow joint
center and wrist joint center tended to be below 8 centimeters for most subjects. To test these
26
assumptions further, by looking at their effect on joint kinetics, resultant shoulder NJM was calculated
for a constant fixed shoulder Z position and elbow fixed directly above wrist (Figure 7).

Figure 8: Resultant shoulder NJM for free and fast propulsion cycles for two example subjects, showing difference between
calculated NJM with 3-dimensional segment data versus assumptions of elbow in wrist plane and shoulder fixed at constant
offset.
27
Calculation of resultant shoulder NJM magnitude with assumptions was on average less than 5 Nm for
free and fast propulsion and at most showed a maximum difference less than 8 Nm.

Kinetics

Shoulder and elbow net joint moments (NJM) and net joint forces (NJF) were calculated using inverse
dynamics. Force applied to the pushrim was collected with the instrumented wheel (SmartWheel).
Center of pressure for the force was assumed to be at the wrist joint center. Segment inertia parameters
were based on body segment parameters (de Leva 1996).
5
The acceleration of the center of mass (COM)
of the segment was collected using either a multicamera motion capture system or digitized segment
kinematics from sagittal video. According to Newton’s second law of motion
𝐹𝑜𝑟𝑐𝑒𝑠 = 𝑀𝑎𝑠𝑠 ∗ 𝐴𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛
The sum of the forces on the forearm along with the acceleration of the center of mass was can be used
to calculate the proximal and distal NJF. The next step is to calculate the NJM
𝑀𝑜𝑚𝑒𝑛𝑡𝑠 = 𝑀𝑜𝑚𝑒𝑛𝑡 𝑜𝑓 𝐼𝑛𝑒𝑟𝑡𝑖𝑎 ∗ 𝐴𝑛𝑔𝑢𝑙𝑎𝑟 𝐴𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛
Once the process has been used to calculate NJM and NJF on the elbow, it can be repeated moving up
the segment to the upper arm in order to calculate the NJM and NJF on the shoulder. The resultant NJM
acting on the shoulder is due to the NJM at the elbow and the proximal and distal moments imposed the
NJFs.

28
References

1. Mulroy SJ, Newsam CJ, Gutierrez D, et al. Effect of Fore-Aft Seat Position on Shoulder Demands
During Wheelchair Propulsion: Part 1. A Kinetic Analysis. J Spinal Cord Med. 2005;28(3):214-221.
doi:10.1080/10790268.2005.11753815.
2. Mulroy SJ, Farrokhi S, Newsam CJ, Perry J. Effects of spinal cord injury level on the activity of
shoulder muscles during wheelchair propulsion: an electromyographic study. Arch Phys Med
Rehabil. 2004;85(6):925-934. doi:https://doi.org/10.1016/j.apmr.2003.08.090.
3. Cooper RA, Digiovine CP, Boninger ML, Shimada SD, Koontz AM, Baldwin MA. Filter frequency
selection for manual wheelchair biomechanics. J Rehabil Res Dev. 2002;39(3):323-336.
4. Wu G, van der Helm FCT, (DirkJan) Veeger HEJ, et al. ISB recommendation on definitions of joint
coordinate systems of various joints for the reporting of human joint motion—Part II: shoulder,
elbow, wrist and hand. J Biomech. 2005;38(5):981-992.
doi:https://doi.org/10.1016/j.jbiomech.2004.05.042.
5. De Leva P. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J Biomech.
1996;29(9):1223-1230. doi:10.1016/0021-9290(95)00178-6.

29
Chapter 4: Modifications in wheelchair propulsion technique
with speed

Ian M. Russell
1,*
, Shashank Raina
1
, Philip S. Requejo
3
, Rand R. Wilcox
4
, Sara Mulroy
3
, Jill L. McNitt-
Gray
1,2

1
Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, USA
2
Department of Biological Sciences, University of Southern California, Los Angeles, CA, USA
3
Pathokinesiology Laboratory, Rancho Los Amigos National Rehabilitation Center, Downey, CA, USA
4
Department of Psychology, University of Southern California, Los Angeles, CA, USA

Abstract
Objective: Repetitive loading of the upper limb joints during manual wheelchair propulsion has been
identified a factor that contributes to shoulder pain, leading to loss of independence and decreased
quality of life. The purpose of this study was to determine how individual manual wheelchair users with
paraplegia modify propulsion mechanics to accommodate expected increases in reaction forces
generated at the pushrim with self-selected increases in wheelchair propulsion (WCP) speed.
Methods: Upper extremity kinematics and pushrim reaction forces were measured for 40 experienced
manual wheelchair users with paraplegia while propelling on a stationary ergometer at self-selected free
and fast propulsion speeds. Upper extremity kinematics and kinetics were compared within-subject
between propulsion speeds. Between group and within subject differences were determined (α =0.05).
Results: Increased propulsion speed was accompanied by increases in Reaction Force (RF) magnitude (22
of 40, >10N) and shoulder Net Joint Moment (NJM, 15 of 40, >10Nm) and decreases in pushrim contact
duration. Within-subject comparison indicated that 27% of participants modified their WCP mechanics
with increases in speed by regulating RF orientation relative to the upper extremity segments.
Conclusions: Reorientation of the RF relative to the upper extremity segments can be used as an
effective strategy for mitigating rotational demands (NJM) imposed on the shoulder at increased
propulsion speeds. Identification of propulsion strategies that individuals can use to effectively
accommodate for increases in RFs is an important step towards preserving musculoskeletal health of the
shoulder and improving health-related quality of life.

Key Words: biomechanics, spinal cord injury, shoulder pain, wheelchair, rehabilitation, propulsion, joint
kinetics, upper extremity

30
Introduction
Preserving shoulder function in individuals with spinal cord injury (SCI) continues to be a significant
problem (Gutierrez et al. 2007; Alm et al. 2008). Effective interaction between an individual and their
manual wheelchair (WC) is essential to preserving quality of life, specifically shoulder function and
overall health (Curtis et al. 1999; Gutierrez et al. 2007). Although the clinical problem of shoulder pain
in individuals with SCI was identified more than three decades ago, the prevalence remains high
(Silfverskiold et al. 1991; Pentland et al. 1994; Jain et al. 2010). Researchers and clinicians have
attributed shoulder pain in the SCI population to the repetitive mechanical loading of the upper limb as
a consequence of lower extremity paralysis (Bayley et al. 1987; Dalyan et al. 1999). In individuals with
paraplegia, shoulder pain can occur within the first year and the incidence increases with time post-
injury (35% at 5 years, 70% at 20 years, Sie et al. 1992). Due to the detrimental impact on functional
mobility and the difficulty in treatment of shoulder pain, effective preventative strategies must be
determined for each wheelchair user. The activities that provoke the highest pain responses for full-time
manual WC users tend to be those that are repetitive and generate high shoulder forces such as manual
wheelchair propulsion (WCP) (Curtis et al. 1999).

Manual WCP is a cyclic task that requires repetitive generation of propulsive forces on the pushrim of
the WC. Generation of these reaction forces (RF) applied at the pushrim involves coordinated activation
of muscles responsible for simultaneously maintaining shoulder joint stability and controlling shoulder
rotation. Structural stability of the shoulder joint is provided by a shallow humeral head socket (glenoid
cavity) and a fibrous labrum (Inman et al. 1944). During WCP, the elbow is positioned below the
shoulder. In this segment configuration, the joint capsule tends to be loose and the reinforcing
ligaments are slack in absence of a RF, thereby creating the need for shoulder muscles to maintain joint
stability (Mulroy et al. 1996). Simultaneously, activation of the upper extremity muscles must be
coordinated to produce the shoulder and elbow net joint moments (NJM) needed to generate
propulsive RFs on the pushrim (Robertson et al. 1996; Kulig et al. 1998; Koontz et al. 2002). Imposing
both joint stability and moment generation requirements on muscles in the shoulder region during WCP
increases the susceptibility to neuromuscular fatigue (Kulig et al. 1998; Koontz et al. 2002). A weakened
muscle within the shoulder-girdle-complex can result in inadequate dynamic stability of the shoulder
particularly during intervals when large RFs are required during WCP (McCully et al. 2007). Loss of
dynamic stability causes stress on the shoulder structures and other joints of the upper limb and can
lead to the development of shoulder pain (Curtis et al. 1999; Gironda et al. 2004; Samuelsson et al.
2004; Alm et al. 2008).

As part of daily living, manual WC users need to regulate WCP speed. On average, increases in WCP
speed has been reported to significantly increase RF magnitudes, decrease hand contact duration, affect
wrist angular position on pushrim (Kulig et al. 1998; Koontz et al. 2002; Veeger et al. 2002) and influence
the mechanical demand imposed on muscles controlling shoulder stabilization and rotation during WCP
(Kulig et al. 1998; Koontz et al. 2002). Increases in WCP speed can also lead to disproportionate
increases in shoulder NJMs during hand contact (Veeger et al. 2002). Understanding how an individual
can effectively interact with the pushrim to achieve required increases in WCP speed provides insights
into how modifications in multijoint control of the upper limb can accommodate for increased
31
mechanical demand imposed on the shoulder. Model simulation results indicate that modifications in RF
orientation relative to the upper extremity segments can effectively redistribute load away from the
shoulder while maintaining WCP speed (Munaretto et al. 2012, 2013). To date, the techniques used by
individuals with SCI to accomplish the changes in propulsion speeds have been difficult to discern from
group mean data of peak NJMs reported during WCP (Kulig et al. 1998; Kulig et al. 2001; Koontz et al.
2002; Mercer et al. 2006).

In this study, we used a within-subject experimental design to determine how individual manual WC
users with paraplegia modify WCP mechanics to accommodate expected increases in RF generated at
the pushrim with self-selected increases in propulsion speed. As found previously, we expect that
reaction force (RF) magnitude, shoulder net joint force (NJF), and shoulder net joint moment (NJM)
during WCP would increase whereas contact duration would decrease with increases in speed (Kulig et
al. 1998; Koontz et al. 2002; Veeger et al. 2002). Consistent with that found in other impulse generating
tasks (McNitt-Gray et al. 2001; Mathiyakom et al. 2005) and experimental-based model simulations of
WCP (Munaretto et al. 2012), we hypothesized that the orientation of RF relative to the forearm and
upper arm would affect the mechanical demand imposed on the upper extremity with increases in WCP
speed. We anticipated that individuals with paraplegia would use different WCP techniques to
accommodate the need to increase WCP speed. Modifications in WCP technique between free and self-
selected fast WCP speeds were characterized by identifying within-subject differences in upper
extremity joint kinetics at peak push during hand contact with the pushrim. Identification of effective
load distribution strategies an individual can use during manual WCP at different speeds provides
evidence to support clinical decisions as to how and when to explore modifications in WCP technique as
a means of preserving shoulder function in individuals with SCI.

Methods
Participants
Forty participants (32 Male, 8 Female) with complete SCI who were experienced manual WC users with
paraplegia (T2-L3) from the outpatient clinics of the Rancho Los Amigos National Rehabilitation Center
volunteered to participate. Each participant was provided informed consent in accordance with the
Institutional Review Board. Individuals were excluded from participation if they reported a history of
shoulder pain that altered performance of daily activities or required medical treatment. Average
(standard deviation) weight of participants was 74.5(18) kg, average height was 1.73 (0.1) m and
average age was 35 years (range: 18 to 62 years). The mean time since occurrence of the injury was 8.25
years (range: 2 to 20 years).

Instrumentation
For this study, the majority of the participants propelled their own wheelchair using an ergometer (27 of
40). In cases when the individual’s WC did not fit the ergometer set-up (13 of 40), the individual used a
rigid frame, lightweight Quickie GPV wheelchair with either a 16” or 18” seat, depending on the size of
the participant. Horizontal and vertical axle positions were matched to that of the individual’s
wheelchair. The height of the footrest, seat back, and inertial parameter of the test wheelchair were
also adjusted to match the participant’s own wheelchair. Each participant used their own seat cushion.
32
The wheelchair was positioned on a stationary ergometer, consisting of a support frame and split rollers,
allowing separate rotation of each wheel. The rollers were coupled by means of a differential to an
alternator and a modified Velodyne® bicycle-ergometer that computer-controlled the resistance. To
quantify the friction force between the tire and ergometer rollers, a coast down test (from 182
m/minute to 35 m/minute) with the participant sitting in the test wheelchair on top of the ergometer
was used. Removable flywheels proportional to the weight of both the person and the wheelchair were
used to simulate the translational inertia of “over ground” propulsion. Further details about the
ergometer instrumentation and calibration steps are described in previous papers (Mulroy et al. 2004;
Mulroy et al. 2005; Requejo et al. 2008; Lighthall-Haubert et al. 2009). Reaction force applied by the
hand to the pushrim was measured using three strain gauge force transducers at 200Hz (SmartWheel,
Three Rivers Holdings, Mesa, AZ, USA).

Data Collection
Three-dimensional trunk, right-side upper extremity and wheel kinematics were collected with active
infra-red markers using a CODA motion analysis system (6-camera, CODA Motion Analysis system, 100
Hz) for 10 seconds of WCP at two speed conditions. Markers were placed on the trunk at the
manubrium, the xiphoid process, the spinous process of T3 and T10 vertebrae, greater tubercle of the
humerus, lateral epicondyle, medial epicondyle, deltoid tuberosity, middle of the forearm, radial styloid,
ulnar styloid, head of the third metacarpal and head of the fifth metacarpal. Three reflective markers
were also placed on the right wheel.

Experimental Protocol
Prior to data collection, participants were given adequate time to become accustomed to the wheelchair
and experimental conditions. Each participant performed WCP at their self-selected free speed, as they
do normally when traversing a tiled floor, and at a self-selected fast speed, as if they are in a hurry to
not miss an important appointment. Preceding the start of data collection, participants propelled for 30-
seconds to avoid the propulsion initiation period. Force and kinematic data were then collected for 10
seconds (6-10 push cycles) at each speed condition with no additional load applied to the ergometer
rollers (i.e. level ground over a tiled surface).

Data Processing and Analysis
The kinematic and force data of consecutive propulsion cycles during the data collection interval (10-
seconds) were analyzed using Visual3D
d
and Matlab
f
. The number of propulsion cycles analyzed for each
subject was the maximum number of propulsion cycles captured in the 10 second window common to
all subjects for that condition (5 for free and 6 for fast). Kinematic data were filtered in Visual3D using a
6
th
order low-pass filter with a cutoff frequency of 8Hz (Cooper et al. 2002). Four segments were
constructed based on the ISB standard definitions (Wu et al. 2005). The thorax segment was defined
using markers placed at the xiphoid, manubrium, T3 and T10 vertebrae. The upper arm segment was
constructed with the marker at the humeral head, a non-collinear marker on the upper arm and the
lateral humeral epicondyle marker. The forearm segment was created using the lateral humeral
epicondyle marker, a non-collinear marker on the forearm and the marker on the ulnar styloid process.
33
The hand segment was created using the markers of the radial styloid, ulnar styloid, the head of the
third metacarpal. Segment inertia parameters were based on body segment parameters (de Leva 1996).

Cycle duration, defined as the elapsed time between successive hand-pushrim contacts, was determined
using measured pushrim RF data. Contact phase of the cycle was defined from the point in time when
the vertical component of the RF exceeded 3N (Newtons) to the time of rim release, when the RF
reduced to below 3N. To characterize differences in initiation of hand contact with the pushrim and
propulsion generation strategies between individuals, the number of peaks in RF observed during the
contact phase were noted (Figure 1). The contact phase was further divided into sub-phases: the impact
(IP) phase when present and a propulsion-generating phase(s) (PGP). The IP was defined as the interval
immediately after pushrim contact (from initial hand contact to time of next local minimum) and was
typically not associated with substantial torque acting to rotate the wheel. Time of peak push was
identified as the time of the maximum peak in the vertical RF measured during PGP.

Figure 9: Vertical reaction force and moment on the wheel for three example propulsion cycles illustrating the
three different propulsion strategies seen in the data. The shaded regions show the duration around peak
averaged to define peak push.

Kinematic and RF at the pushrim were synchronized at time of initial contact with the pushrim and used
to calculate 3D NJM and NJF at the elbow and shoulder (100Hz) using inverse dynamics in Visual3D. The
magnitudes of the RF, NJF, and NJM at the elbow and shoulder are reported for peak push as the
average of the 6 points around the peak in vertical RF during the PGP phase(s). The relative contribution
of the elbow and shoulder to the mechanical demand imposed on the upper extremity was determined
for peak push by the NJM at each joint divided by the sum of the NJMs at both joints (shoulder and
34
elbow). The orientation of the RF relative to the forearm and upper arm was expressed by the angle of
the resultant RF projected into the arm plane (created by the wrist, elbow and shoulder).

Statistics
The probabilities for each variable being less during the free condition than the fast condition when
comparing across propulsion conditions was calculated using a Sign Test. Assuming local independence
for trials and that the free and fast conditions were independent for each subject, these comparisons
were repeated for each variable for within-subject statistical significance as well (R, open-source). A p-
value was then calculated for each subject using Cliff’s analog of the Wilcoxon-Mann-Whitney test (Cliff
1996). A modified, step-down Fisher-type method was then applied to control the familywise error rate
of α = 0.05 over multiple comparisons (Wilcox & Clark 2015). This within-subject analysis was used to
determine which subjects had statistically significant changes when comparing their self-selected free
propulsion cycles to their self-selected fast propulsion cycles.

Results
Consistent with the experimental design, all of the 40 participants significantly increased their WCP
speed between free and fast conditions across all participants (p=0.0001, Figure 2). Mean velocity
across all participants during free condition was 1.02 m/s (0.3) and mean velocity across all participants
during fast condition was 1.72 m/s (0.3). The velocity increase between free and fast conditions was on
average 0.70 (0.2) m/s across participants.

35

Figure 10: Within-subject comparison of self-selected wheelchair propulsion velocity. Black dots are velocity at
free speed condition and blue squares are velocity at fast speed condition. Dotted vertical lines connect each
subject’s free and fast velocities and show velocity increase. All subjects successfully increased propulsion velocity.

As expected, hand-rim contact duration significantly decreased with increases in WCP speed across all
participants (p=0.0001, Figure 3). Within-subject comparisons indicated that 39 of the 40 participants
reduced contact duration with increases in WCP speed. Of those 39 participants, 18 reduced contact
duration by 0.20 seconds or more.

36

Figure 11: Within-subject comparison of average contact time for each subject for both self-selected free and fast
speed conditions. Black dots are contact time at free speed condition and blue squares are contact time at fast
speed condition. SE bars are shown for both conditions. Dotted vertical lines connect each subject’s free and fast
contact times and show magnitude of the change in contact time. Within-subject comparison found 32 of the 40
participants significantly reduced contact duration.

The resultant RF magnitude at peak push significantly increased for the fast as compared to the free
WCP condition across all participants (p=0.0001) (Figure 4). Within-subject comparisons indicated that
26 of the 40 participants increased resultant RF at peak push between the free and fast conditions. Of
those 26 participants, 22 increased the resultant RF by 10 N or more.

37

Figure 12: Within-subject comparison of average resultant RF magnitude at peak push for each subject for both
self-selected free and fast speed conditions. Black dots are average RF magnitude at free speed condition and blue
squares are average RF magnitude at fast speed condition. SE bars are shown for both conditions. Dotted vertical
lines connect each subject’s free and fast RF magnitudes and show magnitude change in RF. Within-subject
comparison found 26 of the 40 participants increased resultant RF at peak push.

The resultant shoulder NJM at peak push significantly increased in the fast as compared to free WCP
conditions across all participants (p=0.0001)(Figure 5). Within-subject comparison revealed that 30 of 40
participants increased resultant shoulder NJM with increases in WCP speed, with 15 participants
increasing shoulder NJM by 10 Nm or more.

38

Figure 13: Within-subject comparison of average resultant NJM magnitude on the shoulder at peak push for each
subject for both self-selected free and fast speed conditions. Black dots are average shoulder NJM magnitude at
free speed condition and blue squares are average shoulder NJM magnitude at fast speed condition. SE bars are
shown for both conditions. Dotted vertical lines connect each subject’s free and fast NJM magnitudes and show
magnitude change in NJM. Within-subject comparison revealed that 30 of 40 participants showed a significant
increase in resultant NJM on the shoulder with increases in WCP speed.

The resultant shoulder NJF at the time of peak push significantly increased in the fast as compared to
free WCP conditions across all participants (p=0.0001)( Figure 6). On average, resultant shoulder NJF
increased by 23 N when propelling under the fast as compared to the free WCP condition.

39

Figure 14: Within-subject comparison of average resultant shoulder NJF magnitude at peak push for each subject
for both self-selected free and fast speed conditions. Black dots are average shoulder NJF magnitude at free speed
condition and blue squares are average shoulder NJF magnitude at fast speed condition. SE bars are shown for
both conditions. Dotted vertical lines connect each subject’s free and fast NJF magnitudes and show magnitude
change in NJF.

As hypothesized, orientation of RF relative to the forearm and upper arm affected the mechanical
demand imposed on the upper extremity with increases in WCP speed. Increases in RF magnitude did
not necessarily result in proportionate increases in shoulder NJM within-subject (Figure 7). For example,
in the fast WCP condition, subjects A and B both generated relatively large RFs (130N and 92N,
respectively) but different techniques led to different magnitudes in shoulder NJMs (Figure 7). Subject A
oriented the RF anterior to forearm resulting in an elbow extensor NJM and a shoulder flexor NJM of 28
Nm. In contrast, Subject B RF was more aligned with the forearm resulting in an elbow flexor NJM and a
shoulder flexor NJM of 41Nm.

40

Figure 15: Average resultant NJM magnitude on the shoulder at peak push for each subject for both self-selected
free and fast speed conditions plotted against average resultant RF magnitude at peak push for each subject.
Diagonal line represents a 1:1 relationship, meaning that a twofold increase in RF would lead to a twofold increase
in NJM on the shoulder. At the higher RF magnitudes, a few subjects deviate further from this relationship.
Subjects A and B illustrate how RF orientation relative to upper extremity can affect shoulder NJMs relationship to
RF magnitude.

Increase is WCP speed, from free to self-selected fast, was accomplished using different techniques
within-subject. In some cases, increases in WCP speed were associated with significant increase in RF
magnitude without modifications in upper extremity kinematics (12 of 40). In other cases, individuals
significantly modified RF orientation, forearm orientation, or both, resulting in modifications in
41
mechanical demand imposed on the shoulder. More vertical orientations of the forearm at peak push
was associated with hand positions more posterior on the pushrim whereas more horizontal orientation
of the forearm at peak push was associated with hand positions that were more anterior on the
pushrim. No significant within-subject differences in elbow angle at peak push were noted between
WCP speeds, suggesting muscle lengths were maintained across WCP conditions.

In some cases, individuals were able to mitigate increases in the rotational demand imposed on the
shoulder with increases in WCP speed whereas others were not. For example, the three exemplar
participants achieved comparable fast WCP velocities with comparable RF magnitudes at peak push
(Figure 8a). However, the magnitude of the shoulder NJM depended on the proximal distal moments
created by the NJFs about the center of mass (CM) of the forearm and upper arm segments as well as
the adjacent joint NJM at the elbow. When the RF is oriented anterior to the forearm CM, an elbow
extensor NJM is needed to achieve the observed motion. The elbow extensor NJM applied to the upper
arm contributes to the reduction in magnitude of the shoulder NJM. In contrast, when the RF is oriented
posterior to the forearm CM, an elbow flexor NJM is needed to achieve the observed motion. The elbow
flexor NJM applied to the upper arm contributes to the increase in magnitude of the shoulder NJM.

In the free WCP condition, the RF orientation relative to the forearm CM at peak push varied across all
participants (anterior (17), aligned (10 within 5 degrees), posterior (13), Figure 8b). Likewise, in the fast
propulsion condition, the RF orientation relative to the forearm CM at peak push tended to be evenly
distributed across all participants (anterior (15), aligned (9 within 5 degrees), posterior (16)).

Within subject comparison in RF orientation relative to the forearm CM at peak push indicated that
shifts in orientation varied with WCP speed. Within-subject analysis indicated 11 of 40 participants made
a significant shift in RF orientation relative to the forearm at peak push when increasing WCP speed. Six
of 11 shifted RF in a direction consistent with increasing the shoulder NJM (Figure 8a), while 5 of 11
participants shifted RF in a direction consistent with decreasing the shoulder NJM. Nine of 11
participants modified the RF orientation relative to the forearm by more than 10 degrees.

42

Figure 16: (A) Effect of RF orientation relative to the upper extremity segments for three example subjects with
comparable propulsion velocities and RF magnitudes. Free body diagrams are drawn for fast speed condition at
the time of peak push. Note elbow NJMs are in opposite directions for the anterior and posterior examples and
43
how that affects shoulder NJM. (B) Population grouping of RF component orientation in the armplane (plane that
connects shoulder, elbow, and wrist) relative to the upper extremity at the time of peak push. Orientation is
grouped into posterior (more than 5° behind the forearm), anterior (more than 5° in front of the forearm), and in
line (within 5° posterior or anterior).

On average there were no consistent shifts across all participants in distribution of the total arm
moment across the upper extremity when increasing WCP speed. Within-subject comparisons indicated
that 10 of 40 participants showed a significant increase in the relative contribution of resultant shoulder
NJM to the total arm moment. The largest shift in load distribution (reduction in shoulder NJM
contribution to total arm moment by 30%) was accomplished by orienting RF more anterior to forearm
(13 deg to 27 deg) and more aligned with the upper arm (28 degrees).

No significant shifts in RF alignment with the arm plane at peak push were observed between WCP
conditions across all participants. Within-subject analysis revealed that 5 of the 40 participants showed a
statistically significant shift in RF alignment (re-alignment of RF relative to arm plane > 5%) with
increases in WCP speed. The RF was less aligned with the arm plane for 4 of 5 of those participants
thereby contributing to out of plane shoulder NJMs.

Discussion
During daily activities, manual wheelchair users often encounter situations that result in increases in the
mechanical demand imposed on the upper extremity, such as speeding up, going up ramps, or
traversing carpets and grass. Understanding the different techniques individual’s use during tasks with
increased upper extremity demands is important for identifying manual WCP strategies that can help
preserve shoulder function, maintain independence, and improve quality of life. The results of this
study indicate that increases in RF magnitudes associated with increases in WCP speed do not
necessarily translate into comparable increases in shoulder NJMs. The magnitude of the shoulder NJM
depends on the proximal distal moments created by the NJFs about the CM of the forearm and upper
arm segments as well as the adjacent joint NJM at the elbow. Within-subject analysis indicated more
than 25% of the participants made a significant shift in RF orientation relative to the forearm at peak
push when increasing WCP speed. In approximately half of these cases, reorienting the RF relative to the
upper extremity segments was used as an effective strategy for mitigating rotational mechanical
demand imposed on the shoulder at increased WCP speeds. In the other cases, the shift in RF
orientation relative to the forearm at peak push at increased WCP speeds contributed to increases in
the shoulder NJM and reductions in the vertical component of the shoulder NJF. By investigating WCP
technique modifications in response to increases in WCP speed using a within-subject design,
preferential shifts in mechanical loading imposed on the shoulder can be determined. This knowledge
of self-selected load mitigation strategies may prove fruitful in guiding clinical decisions that aim to
identify strategies for preserving shoulder function in individuals with SCI.

The self-selected free and fast propulsion velocities attained in our sample population are comparable
to those found in Kulig et al. 1998. Joint kinetics in this study were also found to be in line with
44
magnitudes previously reported in the literature (Kulig et al. 1998; Koontz et al. 2002; Veeger et al.
2002; Collinger et al. 2008). In this study, an ergometer was used to achieve self-selected steady-state
WCP speeds for multiple cycles. To minimize limitations associated with this experimental set-up, a
within-subject design was used as a means for each individual to serve as their own control. Consistent
with previously reported group mean data (Kulig et al. 1998; Koontz et al. 2002; Veeger et al. 2002;
Collinger et al. 2008), the resultant RF as well as the resultant shoulder NJM and NJF at peak push
significantly increased in the fast WCP speed condition when compared to free WCP across subjects.

In order to increase WCP speed, the tangential component of the RF being applied to the pushrim must
increase in magnitude, particularly if the pushrim contact duration decreases with WCP speed. The
participants in this study increased WCP speed using a variety of different techniques. Some participants
increased WCP speed by amplifying RF magnitude without modifications in upper extremity kinematics.
Whereas other individuals significantly modified RF orientation, forearm orientation, or both, resulting
in modifications in mechanical demand imposed on the shoulder. Minimal changes in elbow angle at
peak push were observed across speeds, suggesting individuals may have chosen to maintain a
preferred muscle length when generating RF at peak push. Results of this study illustrated how choice of
orientation of RF relative to the upper extremity affected mechanical demand on the shoulder.
Orientation of RF anterior to the forearm CM created an elbow extensor NJM which contributed to a
reduction in shoulder NJM magnitude. Conversely, when RF was oriented posterior to the forearm CM
the resulting elbow flexor NJM contributed to an increase in shoulder NJM magnitude. These results
suggest that individuals choosing to modify WCP technique by shifting the RF more anterior to the
forearm CM may favor reductions in shoulder NJM over increases in the vertical component of the RF,
and vice versa. Identification of preferences toward a particular load mitigation strategy may prove
fruitful in guiding clinical decisions that aim to identify strategies for preserving shoulder function in
individuals with SCI.

The experimental results of this study are consistent with the model simulation results (Munaretto et
al., 2012, 2013) that demonstrate at a particular WCP speed, increases in resultant pushrim RF can occur
without comparable increase in shoulder NJM. The magnitude of the shoulder NJM is dependent on the
proximal distal moments created by the NJFs at the elbow and shoulder and the elbow NJM (Figure 8).
The magnitude of the proximal and distal moments is dependent on the magnitude of the NJFs and their
orientation relative the upper arm. Redirection of the RF relative to the upper extremity, as shown in
both the experimental and model simulation results, can serve as a potential strategy to redistribute
load imposed on the upper extremity. Simulation results indicate that WCP speed can be maintained
with minimal changes in shoulder NJM even if the corresponding RF doubles in magnitude, provided the
RF is reoriented relative to the forearm and upper arm. These results indicate that alignment of the RF
anterior to the forearm can mitigate the effect of higher pushrim forces on shoulder NJM magnitude.
This strategy may prove to be an effective means of redistributing the mechanical loads imposed on the
upper extremity joints during WCP.

Maintaining shoulder health requires more than reducing mechanical demand. Certain scapular and
glenohumeral orientations have been associated with reducing subacromial space which increases the
45
potential risk of shoulder impingement syndrome. Previous research by Morrow et al. 2011 and Raina et
al. 2012 found that WCP placed the scapula in some of these potentially dangerous orientations that
could contribute to the development shoulder impingement. More specifically, Raina’s study showed
that with increases in propulsion force, wheelchair user’s scapula tended to move into anterior tilt,
downward rotation and protraction. All of these positions are associated with reduced subacromial
space. If this scapular movement occurs in conjunction with upward motion of the humerus in the
glenoid cavity, there is potential for impingement of the supraspinatus. The superiorly-directed forces
transmitted along the axis of the humerus could have a negative long-term consequence if not
adequately controlled by muscles crossing the shoulder complex (Mulroy et al. 2005). However, further
research must be done with more accurate methods of subacromial space estimation to see if the
scapular movement found in wheelchair propulsion is clinically relevant (Raina et al. 2012). Any
recommendation in technique modification must consider the ability of the individual to control RF and
segment motion during task performance to avoid detrimental loading (McNitt-Gray 2000).

By examining how individual wheelchair users organized their upper limb coordination to accommodate
increases in mechanical demands, effective multijoint control strategies for increasing WCP speed
without substantial increases in the shoulder NJM was identified. Future studies will examine how this
WCP technique may benefit those with different upper extremity control capabilities and will explore
the relative contribution of these factors in regulating shoulder loads during WCP. For individuals unable
to modify WCP technique by reorienting pushrim RF relative to the upper extremity segments,
customized modifications in wheelchair configuration, including optimal seat height (Kotajarvi et al.
2004), axle horizontal positions (Mulroy et al. 2005), and seat angle (Desroches et al. 2006), are
alternative ways of re-distributing the mechanical demands within the upper extremity joints.

Acknowledgements
This study was supported by a grant from the National Institute of Health (R01 HD049774). This project
was supported by SC CTSI (NIH/NCATS) through Grant UL1TR000130. Its contents are solely the
responsibility of the authors and do not necessarily represent the official views of the NIH.

Suppliers
a. Digital Equipment Corporation, Cambridge, MA
b. Quickie GPV; Sunrise Medical, Quickie Designs Inc., 2842 Business Park Avenue. Fresno, CA
93727.
c. Velodyne bicycle ergometer; Schwinn Bicycle Company 217 N. Jefferson Street, Chicago. IL
60661.
d. VICON (Oxford Metrics LTD, Oxford, England).
e. C-Motion, Inc.15821-A Crabbs Branch Way, Rockville, MD 20855
f. The MathWorks, Inc. 3 Apple Hill Drive Natick, MA 01760-2098
g. R Programming Language. Link www.r-project.org
46
References

Alm, M., H. Saraste and C. Norrbrink (2008). Shoulder pain in persons with thoracic spinal cord injury:
prevalence and characteristics. J Rehabil Med 40(4): 277-283.
Asato, K. T., R. A. Cooper, R. N. Robertson and J. F. Ster (1993). SMARTWheels: development and testing
of a system for measuring manual wheelchair propulsion dynamics. IEEE Trans Biomed Eng
40(12): 1320-1324.
Bayley, J. C., T. P. Cochran and C. B. Sledge (1987). The weight-bearing shoulder. The impingement
syndrome in paraplegics. J Bone Joint Surg Am 69(5): 676-678.
Boninger, M. L., M. Baldwin, R. A. Cooper, A. Koontz and L. Chan (2000). Manual wheelchair pushrim
biomechanics and axle position. Archives of Physical Medicine & Rehabilitation 81(5): 608-613.
Brose, S. W., M. L. Boninger, B. Fullerton, T. McCann, J. L. Collinger, B. G. Impink and T. A. Dyson-Hudson
(2008). Shoulder ultrasound abnormalities, physical examination findings, and pain in manual
wheelchair users with spinal cord injury. Arch Phys Med Rehabil 89(11): 2086-2093.
Cliff, N. (1996). Answering Ordinal Questions with Ordinal Data Using Ordinal Statistics. Multivariate
Behav Res. 31(3):331-350.
Collinger, J.L., M.L. Boninger, A.M. Koontz, et al. (2008). Shoulder biomechanics during the push phase of
wheelchair propulsion: a multisite study of persons with paraplegia. Arch Phys Med Rehabil.
89(4):667-676.
Cooper, R., C. P. DiGiovine, M. Boninger, S. D. Shimada, A. Koontz and M. A. Baldwin (2002). Filter
frequency selection for manual wheelchair biomechanics. Journal of Rehabilitation Research
and Development 39(323-336).
Curtis, K. A., G. A. Drysdale, R. D. Lanza, M. Kolber, R. S. Vitolo and R. West (1999). Shoulder pain in
wheelchair users with tetraplegia and paraplegia. Arch Phys Med Rehabil 80(4): 453-457.
Dalyan, M., D. D. Cardenas and B. Gerard (1999). Upper extremity pain after spinal cord injury. Spinal
Cord 37(3): 191-195.
de Leva, P. (1996). Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. J Biomech 29(9):
1223-1230.
Desroches, G., R. Aissaoui and D. Bourbonnais (2006). Effect of system tilt and seat-to-backrest angles
on load sustained by shoulder during wheelchair propulsion. J Rehabil Res Dev 43(7): 871-882.
Escobedo, E. M., J. C. Hunter, M. C. Hollister, R. M. Patten and B. Goldstein (1997). MR imaging of
rotator cuff tears in individuals with paraplegia. AJR Am J Roentgenol 168(4): 919-923.
Gironda, R. J., M. E. Clark, B. Neugaard and A. Nelson (2004). Upper limb pain in a national sample of
veterans with paraplegia. J Spinal Cord Med 27(2): 120-127.
Gutierrez, D. D., L. Thompson, B. Kemp and S. J. Mulroy (2007). The relationship of shoulder pain
intensity to quality of life, physical activity, and community participation in persons with
paraplegia. J Spinal Cord Med 30(3): 251-255.
Inman, V. T., J. B. d. M. Saunders and L. C. Abbott (1944). Observations on the function of the shoulder
joint. Journal of Bone and Joint Surgery 26A: 3-30.
Jain N. B., Higgins L. D., Katz J. N., Garshick E. (2010). Association of shoulder pain with the use of
mobility devices in persons with chronic spinal cord injury. Platin. Met. Rev2: 896–900.
Koontz, A. M., R. A. Cooper, M. L. Boninger, A. L. Souza and B. T. Fay (2002). Shoulder kinematics and
kinetics during two speeds of wheelchair propulsion. J Rehabil Res Dev 39(6): 635-649.
Kotajarvi, B. R., M. B. Sabick, K. N. An, K. D. Zhao, K. R. Kaufman and J. R. Basford (2004). The effect of
seat position on wheelchair propulsion biomechanics. J Rehabil Res Dev 41(3B): 403-414.
Kulig, K., C. J. Newsam, S. J. Mulroy, S. Rao, J. K. Gronley, E. L. Bontrager and J. Perry (2001). The effect of
level of spinal cord injury on shoulder joint kinetics during manual wheelchair propulsion.
47
Clinical Biomechanics 16(9): 744-751.
Kulig, K., S. S. Rao, S. J. Mulroy, C. J. Newsam, J. K. Gronley, E. L. Bontrager and J. Perry (1998). Shoulder
joint kinetics during the push phase of wheelchair propulsion. Clinical Orthopaedics & Related
Research(354): 132-143.
Lighthall-Haubert, L., P. S. Requejo, S. J. Mulroy, C. J. Newsam, E. Bontrager, J. K. Gronley and J. Perry
(2009). Comparison of shoulder muscle electromyographic activity during standard manual
wheelchair and push-rim activated power assisted wheelchair propulsion in persons with
complete tetraplegia. Arch Phys Med Rehabil. 90(11): 1904-1915.
Mathiyakom W, McNitt-Gray JL, Requejo PS, Costa KE. (2005). Modifying center of mass trajectory
during sit-to-stand tasks redistributes the mechanical demand across the lower extremity joints.
J Clin Biomech. 20:105-111.
McCully, S. P., D. N. Suprak, P. Kosek and A. R. Karduna (2007). Suprascapular nerve block results in a
compensatory increase in deltoid muscle activity. Journal of Biomechanics 40: 1839-1846.
McNitt-Gray, J. L. (2000). Subject specific coordination of two- and one-joint muscles during landings
suggests multiple control criteria. Motor Control 4(1): 84-88; discussion 97-116.
McNitt-Gray, J. L., D. M. Hester, W. Mathiyakom and B. A. Munkasy (2001). Mechanical demand and
multijoint control during landing depend on orientation of the body segments relative to the
reaction force. J Biomech 34(11): 1471-1482.
Mercer, J. L., M. Boninger, A. Koontz, D. Ren, T. Dyson-Hudson and R. Cooper (2006). Shoulder joint
kinetics and pathology in manual wheelchair users. Clin Biomech (Bristol, Avon) 21(8): 781-789.
Morrow, M.M.B., M.G. Van Straaten, N.S. Murthy, J.P. Braman, E. Zanella, K.D. Zhao (2014). Detailed
Shoulder MRI Findings in Manual Wheelchair Users with Shoulder Pain. 2014:8-10.
Morrow, M.M.B., K.R. Kaufman, K.N. An (2011). Scapula kinematics and associated impingement risk in
manual wheelchair users during propulsion and a weight relief lift. Clin Biomech. 26(4):352-357.
Morrow, M.M.B., W.J. Hurd, K.R. Kaufman, K.N. An (2010). Shoulder demands in manual wheelchair
users across a spectrum of activities. J Electromyogr Kinesiol. 20(1):61-67.
Mulroy, S. J., S. Farrokhi, C. J. Newsam and J. Perry (2004). Effects of spinal cord injury level on the
activity of shoulder muscles during wheelchair propulsion: an electromyographic study. Arch
Phys Med Rehabil 85(6): 925-934.
Mulroy, S. J., J. K. Gronley, C. J. Newsam and J. Perry (1996). Electromyographic activity of shoulder
muscles during wheelchair propulsion by paraplegic persons. Archives of Physical Medicine &
Rehabilitation 77(2): 187-193.
Mulroy, S. J., C. J. Newsam, D. D. Gutierrez, P. Requejo, J. K. Gronley, L. L. Haubert and J. Perry (2005).
Effect of fore-aft seat position on shoulder demands during wheelchair propulsion: part 1. A
kinetic analysis. J Spinal Cord Med 28(3): 214-221.
Munaretto JM, McNitt-Gray JL, Flashner H, Requejo PS. (2012). Simulated effect of reaction force
redirection on the upper extremity mechanical demand imposed during manual wheelchair
propulsion. Clin Biomech (Bristol, Avon) 27(3): 255-262.
Munaretto JM, McNitt-Gray JL, Flashner H, Requejo PS. (2013). Reconfiguration of the upper extremity
relative to the pushrim affects load distribution during wheelchair propulsion. Med Eng Phys.
35: 1141-1149.
National Spinal Cord Injury Statistical Center. (2011, February 2011). "Spinal Cord Injury Facts and
Figures at a Glance." from https://www.nscisc.uab.edu.
Pentland W. E., Twomey L. T. (1994). Upper limb function in persons with long term paraplegia and
implications for independence: Part II. Paraplegia. 32: 219–224.
Raina S, McNitt-Gray JL, Mulroy S, Requejo PS. (2012). Effect of increased load on scapular kinematics
during manual wheelchair propulsion in individuals with paraplegia and tetraplegia. Hum Mov
Sci. 31(2):397-407.
48
Requejo, P. S., S. E. Lee, S. J. Mulroy, L. L. Haubert, E. L. Bontrager, J. K. Gronley and J. Perry (2008).
Shoulder muscular demand during lever-activated vs pushrim wheelchair propulsion in persons
with spinal cord injury. J Spinal Cord Med 31(5): 568-577.
Robertson, R. N., M. L. Boninger, R. A. Cooper and S. D. Shimada (1996). Pushrim forces and joint
kinetics during wheelchair propulsion. Archives of Physical Medicine & Rehabilitation 77(9):
856-864.
Samuelsson, K. A., H. Tropp and B. Gerdle (2004). Shoulder pain and its consequences in paraplegic
spinal cord-injured, wheelchair users. Spinal Cord 42(1): 41-46.
Sie, I. H., R. L. Waters, R. H. Adkins and H. Gellman (1992). Upper extremity pain in the postrehabilitation
spinal cord injured patient. Arch Phys Med Rehabil 73(1): 44-48.
Silfverskiold J., Waters R. L. (1991).Shoulder pain and functional disability in spinal cord injury patients.
Clin. Orthop. Relat. Res. 272: 141–145.
Soltau, S.L., J.S. Slowik, P.S. Requejo, S.J. Mulroy, R.R. Neptune (2015). An Investigation of Bilateral
Symmetry During Manual Wheelchair Propulsion. Front Bioeng Biotechnol. 3(June):3-8.
Veeger, H. E., L. A. Rozendaal and F. C. van der Helm (2002). Load on the shoulder in low intensity
wheelchair propulsion. Clin Biomech (Bristol, Avon) 17(3): 211-218.
Wu, G., F. C. van der Helm, H. E. Veeger, M. Makhsous, P. Van Roy, C. Anglin, J. Nagels, A. R. Karduna, K.
McQuade, X. Wang, F. W. Werner and B. Buchholz (2005). ISB recommendation on definitions of
joint coordinate systems of various joints for the reporting of human joint motion--Part II:
shoulder, elbow, wrist and hand. J Biomech 38(5): 981-992.
Wilcox RR, Clark F. (2015).Robust multiple comparisons based on combined probabilities from
independent tests. J Data Sci. 13(1):140308.
Wyndaele, M. and J. J. Wyndaele (2006). Incidence, prevalence and epidemiology of spinal cord injury:
what learns a worldwide literature survey? Spinal Cord 44(9): 523-529.

49
Chapter 5: Parameterization of the shoulder net joint moment
using four functional axes

Introduction
Joint kinetic analysis is a common way to quantify the mechanical demand imposed on structures
involved in controlling musculoskeletal joints during different movement tasks. Currently, the
coordinate systems used to report net joint moments (NJM) of the upper and lower extremity during
different tasks can vary considerably between studies which makes comparison of results across studies
difficult
1
. In addition, the reporting of the components of the NJM in relation to the three orthogonal
axes used when computing NJMs often loses functional meaning from an anatomical perspective.
Functionally relevant representations of joint kinetics can facilitate comparisons across tasks and
studies. For example, Schache and Baker proposed guidelines for the lower limb kinetics by adopting the
kinematic coordinate standard put forth by the International Society of Biomechanics for motion
2
. This
Joint Coordinate System (JCS), uses a proximal segment axis and a distal segment axis with a third
floating axis and the distal one being longitudinal to the moving body. Depending on the research
question, axes fixed to the proximal segment, distal segment, or a mixture of both have been
implemented to characterize multi-joint control of the extremities within the biomechanical literature
1
.
All of these existing approaches parse the resultant NJM at a joint into three components
3-14
even
though the joint has four rotary degrees of freedom when expressed relative to the proximal adjacent
segment that arise from the kinematic constraints of linked segments.
Maintaining shoulder health among manual wheelchair (WC) users is especially important because they
are dependent on their upper extremities for mobility as well as activities of daily living. A detailed
understanding of the mechanical demand imposed on the shoulder during physically demanding
activities such as manual WC propulsion is needed to describing an individual’s interaction with their WC
and identifying how to maintain shoulder health.
15
The shoulder moment during manual WC propulsion
has been represented with a variety of coordinate systems
5–14
. These systems have involved a humeral
coordinate system only, a trunk coordinate system only, or a combination of the two
1
. Morrow et. al
2009 suggested extending Schache and Baker’s kinetic representation proposal to those of the upper
extremity, representing upper body efforts in the JCS. However, Morrow et al. correctly reasoned that,
in the presence of axial torque about the distal segment, the current NJM parameterization methods
may introduce crosstalk that misrepresents control required about the proximal segment axes. An
extension of the JCS method proposed by Wagner et al. (in review) describes how the motion of the
humerus (relative to the torso) can be parsed into two types of rotation: those that change where the
elbow is pointing in relation to the torso (i.e. 3-fixed axes torso coordinate system) and those that
change the orientation of the humerus about its longitudinal axis (axial rotation). Similarly, these same
four-axes can be used for joint kinetic descriptors for the shoulder: adduction-abduction (AdAb),
horizontal adduction-abduction (HAdAb), and flexion-extension (FE) defined by the torso, and a fourth
external-internal rotation (EI) axis defined by the longitudinal axis of the upper arm pointing distally.
After calculating the angular velocity and the NJM about these 4 axes, we can then take the product of
50
each corresponding torque and velocity to find the parsed instantaneous NJM power. Finding the parsed
instantaneous power, provides kinematic context under which the NJM is generated. If a NJM
component has the same sign as the coaxial angular velocity component, the NJM axis power is positive
and representative of concentric effort. If NJM and angular velocity are in opposing directions, the NJM
axis power will be negative and represent eccentric effort. We expected that parameterizing the
shoulder NJM vector expressed relative to the trunk using this 4-axes parsing approach and combining it
with the parsed angular velocity vector to calculate the NJM power will provide an effective means to
decouple the mechanical demand imposed on the shoulder and provide insight regarding the
mechanical demand imposed on the shoulder using an anatomically relevant reference system.
3

In this paper, we aim to demonstrate the value of this proposed 4-axes parameterization method in
characterizing the mechanical demand imposed on the joints during functional activities.
Parameterization of the NJM using the 4-axes approach is expected to be advantageous in that it can be
applied to any joint and any task, yet still uses functionally intuitive axes that can be compared across
studies. The extension of this method to calculating power provides additional context to the function
behind the NJM. Mathematically, we expect this approach will be valuable because it provides a means
to compare joint level control across studies because it characterizes control of segment re-alignment by
a rotation as well as segment reorientation by a rotation. As a result, axial components of the NJM can
be separated from positional components. Our working hypothesis is that the proposed 4-axes
parameterization of a shoulder NJM and shoulder NJM power will further clarity on shoulder joint
control during manual WC propulsion by individuals with paraplegia. We test this hypotheses by
comparing 4-axes and 3-axes representations of the shoulder NJM during manual WC propulsion. We
expect individuals with paraplegia will distribute the mechanical demand on the shoulder about axes
different during manual WC propulsion and that the 4-axes parameterization will assist in characterizing
manual WC propulsion strategies. We anticipate that distribution of NJM about the 4-axes will vary
based on elbow position away from the torso and reaction force (RF) orientation relative to the arm
plane.
Methods
Joint kinetic data acquired during manual wheelchair propulsion at self-selected speeds by experienced
manual WC users with paraplegia (n=130, 9 ± 6 yrs post spinal cord injury) from the Rancho Los Amigos
National Rehabilitation Center outpatient clinics were analyzed using the proposed 4-axes
parameterization method.

Prior to data collection, participants were given adequate time to become accustomed to the wheelchair
and experimental conditions. Each participant performed WC propulsion at their self-selected fast
speed, as if they are in a hurry to not miss an important appointment. Three-dimensional trunk, right-
side upper extremity and wheel kinematics were collected with active infra-red markers using a CODA
motion analysis system (6-camera, CODA Motion Analysis system, 100 Hz) for 10 seconds of WC
propulsion at fast propulsion speed. Markers were placed on the trunk at the manubrium, the xiphoid
process, the spinous process of T3 and T10 vertebrae, greater tubercle of the humerus, lateral
epicondyle, medial epicondyle, deltoid tuberosity, middle of the forearm, radial styloid, ulnar styloid,
51
head of the third metacarpal and head of the fifth metacarpal. Three reflective markers were also
placed on the right wheel. Forces applied by the hand to the pushrim was measured using three strain
gauge force transducers at 200Hz (SmartWheel, Three Rivers Holdings, Mesa, AZ, USA). Propulsion was
done on stationary ergometer and further details about the ergometer instrumentation and calibration
steps are described in previous papers
16–19
.
The kinematic and force data of consecutive propulsion cycles during the data collection interval (10-
seconds) were analyzed using Visual3D and Matlab. Kinematic data were filtered in Visual3D using a 6
th

order low-pass filter with a cutoff frequency of 8Hz
20
. Four segments were constructed based on the
ISB standard definitions
21
. The thorax segment was defined using markers placed at the xiphoid,
manubrium, T3 and T10 vertebrae. The upper arm segment was constructed with the marker at the
humeral head, a non-collinear marker on the upper arm and the lateral humeral epicondyle marker. The
forearm segment was created using the lateral humeral epicondyle marker, a non-collinear marker on
the forearm and the marker on the ulnar styloid process. Segment inertia parameters were based on
body segment parameters
22
. Kinematic and RF at the pushrim were synchronized at time of initial
contact with the pushrim and used to calculate 3D NJM shoulder using inverse dynamics in Matlab. Push
phase of the propulsion cycle is defined when resultant force measured by the instrumented wheel
exceeds 5N.

Approach for 4-Axes Parameterization of the Shoulder NJM
The NJM vector (𝜏 ⃗) was calculated in the inertial reference frame and projected into the longitudinal axis
of the upper arm (𝑥 ) to find the degree of External/Internal rotation (EI) moment about the upper
arm. From this point on all vectors including (𝜏 ⃗) and (𝑥 ) are expressed in the torso reference frame:
𝜏 ⃗
= 𝑥 (𝑥 ∙ 𝜏 ⃗)
where the signed scalar form of this measure is given by:
𝜏 = 𝑠𝑖𝑔𝑛 (𝜏 ⃗
)|𝜏 ⃗
|
Next the vector form of the external/internal moment, 𝜏 ⃗
, component is subtracted from 𝜏 ⃗. The
remaining torque describes the NJM vector perpendicular to the upper arm’s longitudinal axis.
𝜏 ⃗
= 𝜏 ⃗ − 𝜏 ⃗

By expressing 𝜏 ⃗
in the torso reference frame, the components in the 𝚤 ̂, 𝚥 ̂, and 𝑘 directions
correspond to the moments about Adduction/Abduction (AdAb), Horizontal Adduction/Abduction
(HAdAb), and Flexion/Extension (FE) axes respectively (reference system shown in Figure 1):
𝜏 ⃗
= (𝜏 𝐴𝑑𝐴𝑏 )𝚤 ̂ + (𝜏 𝐻𝐴𝑑𝐻𝐴𝑏 )𝚥 ̂ + (𝜏 𝐹𝐸 )𝑘
We now have the components of the shoulder NJM in four independent axes
(𝜏 , 𝜏 , 𝜏 , 𝜏 ).
52

Figure 17: In the 4-axes parameterization approach, the shoulder net joint moment vector is parsed into four components with
each axis describing a unique anatomically relevant functional axis of torque thereby providing an effective means for
characterizing mechanical demand on a joint during multidimensional tasks as a function of time. Reference system of four
anatomical axes. Three torso fixed axes FE, HAdAb, AdAb and the longitudinal axis of the upper arm EI.
Approach for 4-Axes Calculation of the Shoulder NJM Power
The elementwise product of the 4-term ω with the 4-term NJM (τ) gives a 4-term power vector (P). The
result is a positive/negative NJM power metric in each of the 4 anatomically-relevant axes.

𝑷 = 𝝉 ∘ 𝝎 = 𝜏 𝜏 𝜏 𝜏 ∘ 𝜔 𝜔 𝜔 𝜔 = 𝜏 𝜔 𝜏 𝜔 𝜏 𝜔 𝜏 𝜔

Results
Parsing of the resultant shoulder NJM about 4-axes revealed consistent coordinated control of the
shoulder about multiple axes within participant. These NJM distribution profiles provide additional
insight about how an individual distributed control of the shoulder about each of the four functional
axes as compared to reporting of resultant NJMs. Parsing of the NJMs about 4-axes provided additional
clarity by avoiding crosstalk observed in NJMs parsed using 3 axes.
between subjects we can easily see the necessity of parsing NJM into these 4 anatomical axes. The first
and third subjects analyzed, Figure 2 and Figure 6, favor a more flexion dominant technique, while the
other subject distributes NJM more evenly between Flexion and Horizontal Adduction.
53

Figure 18: Left side snapshot of torso, arm, and force orientation during push with 4 anatomical axes and NJM vector (T)
overlaid (excluding External-Internal component). Right side plot shows 4-axis parsed NJM magnitude about each anatomical
axis External-Internal Rotation Axis (blue), Flexion-Extension Axis (brown), Adduction-Abduction (yellow), and Horizontal
Abduction-Adduction (purple). NJM is plotted against hand position on pushrim reported in degrees with 90 degrees
representing 12:00 position and 0 degrees at the 3:00 position on the handrim, if subject was facing to the right.
Investigating factors that lead to a particular NJM distribution about the shoulder, we examine RF
orientation relative to the plane connecting shoulder, elbow, and wrist joint centers (arm-plane) as well
as the orientation of the upper arm relative to the torso. We can see in Figure 2 a RF aligned in the arm-
plane, shown here by the dark blue RF vector located at the wrist oriented along the forearm at the
frame of push. This subject keeps the RF mostly in the arm-plane throughout the push resulting in very
little NJM about the external-internal rotation axis, which is confirmed in the corresponding parsed NJM
distribution plot with relatively small magnitude about that axis. The subject’s upper arm is kept
relatively close to his torso resulting an upper arm orientation in the picture similar to the superior-
inferior axis of the torso. When the NJM vector (excluding the external-internal rotation component) is
analyzed against the 3 torso axes (shown in Figure 2 as dotted τ vector perpendicular to the longitudinal
axis of the upper arm), we can see that it mostly falls along the flexion axis leading to a largely flexion
based shoulder NJM in the corresponding parsed NJM distribution plot on the right.
54

Figure 19: Positive power signifies NJM assisting angular velocity, while negative signifies opposing angular velocity. Flexion
dominant angular velocity as well as flexion dominant NJM leads to power about the flexion axis. (External/Internal (E/I),
Adduction/Abduction (Ad/Ab), Horizontal Adduction/Abduction (HAd/HAb), Flexion/Extension (F/E))
The kinematic context of the shoulder NJM for this same subject can be seen in the corresponding
parameterized plots of angular velocity and power in Figure 3. We can see that just as in the NJM
distribution, this subject displays angular velocity largely in the Flexion direction, signifying that the
flexion NJM assists in generating the flexion motion which is confirmed by the positive power about the
flexion axis. Relatively small NJM components about the HAdAb, AdAb, and EI axes as well as minimal
angular velocity about those axes (less than 2 degrees per second), leads to small NJM power in
comparison to the power about the FE axis. Average NJM power impulse about the FE axis was 8.77 Nm,
while the other 3 axes were much less (HAdAb = 0.77 Nm, AdAb = 0.21 Nm, EI = -0.02).
55

Figure 20: Left side snapshot of torso, arm, and force orientation during push with 4 anatomical axes and NJM vector (T)
overlaid. Right side plot shows 4-axis parsed NJM magnitude about each anatomical axis External-Internal Rotation Axis (blue),
Flexion-Extension Axis (brown), Adduction-Abduction (yellow), and Horizontal Abduction-Adduction (purple). NJM is plotted
against hand position on pushrim reported in degrees with 90 degrees representing 12:00 position and 0 degrees at the 3:00
position on the handrim, if subject was facing to the right.
Looking at the same factors of RF orientation relative to the arm-plane and orientation of the upper arm
relative to the torso for another subject in Figure 4 we can see again the dark blue RF vector located at
the wrist mostly aligned with the arm-plane resulting in a small magnitude of NJM about the external-
internal rotation axis in the corresponding parsed NJM distribution plot. However, this subject differs
from the previous example in that his upper arm is out and away from his torso, no longer aligned with
the superior-inferior axis of the torso but between that and the medial-lateral axis pointing to the
subject’s right. When the NJM vector (excluding the external-internal rotation component) is analyzed
against the 3 torso axes (shown in Figure 4 as dotted τ vector perpendicular to the longitudinal axis of
the upper arm), we can see that it falls between the Flexion and Horizontal Adduction vectors. The
resulting parsed NJM distribution due to this abducted elbow position is therefore divided more evenly
between Flexion and Horizontal Adduction, which is confirmed in the parsed NJM distribution plot in the
similar magnitudes of NJM about those axes.
56

Figure 21: Positive power signifies NJM assisting angular velocity, while negative signifies opposing angular velocity. Similar
angular velocity magnitudes about flexion and horizontal adduction align with similar NJM distribution. However, significant
angular velocity in the external rotation direction does not have correspondingly significant NJM relative to the NJM about the
two major axes of FE and HAdHAb. (External/Internal (E/I), Adduction/Abduction (Ad/Ab), Horizontal Adduction/Abduction
(HAd/HAb), Flexion/Extension (F/E))
The kinematic context of the shoulder NJM for this subject can be seen in the corresponding
parameterized plots of angular velocity and power in Figure 5. We can see that just as in the NJM
distribution, this subject displays noticeable and similar magnitude angular velocity in the flexion and
horizontal adduction direction. However, NJM about the EI axis is rather small in comparison to the
angular velocity seen in the external rotation direction. The resulting power plot highlights the active
concentric torques about the FE and HAdHAb axes, and also shows very little EI NJM power due to the
small NJM component about this axis. Average NJM power impulse about the FE axis was 3.30 Nm and
about the HAdHAb was 2.20, which are much closer in magnitude than the previous example subject
who showed a flexion dominant technique. Average NJM power impulse was relatively small about the
remaining two axes (AbAd = 0.14 Nm, EI = 0.61 Nm).
57

Figure 22: Left side snapshot of torso, arm, and force orientation during push for subject with RF oriented lateral to armplane.
Right side plot shows 4-axis parsed NJM magnitude about each anatomical axis Internal-External Rotation Axis (blue), Flexion-
Extension Axis (brown), Abduction-Adduction (yellow), and Horizontal Abduction-Adduction (purple). NJM is plotted against
hand position on pushrim reported in degrees with 90 degrees representing 12:00 position and 0 degrees at the 3:00 position
on the handrim, if subject was facing to the right.
In the third example subject, shown in Figure 6, we can see a propulsion technique that differs from the
two previous examples where the RF is instead oriented laterally to the arm-plane (dark blue vector at
the wrist points off to the side relative to forearm). In this scenario the RF is creating an external
rotation torque that is then countered at the shoulder with an internal rotation NJM component. The
presence of NJM in the internal direction is confirmed in the corresponding parsed NJM distribution plot
by the negative blue curves. While orientation of the upper arm relative to the torso determines the
parsed NJM distribution about the three torso fixed axes of Flexion-Extension, Abduction-Adduction,
and Horizontal Abduction-Adduction, we can see in Figure 6 that orientation of the RF relative to the
arm-plane determines the presence of NJM about the external-internal rotation axis of the upper arm.
58

Figure 23: Positive power signifies NJM assisting angular velocity, while negative signifies opposing angular velocity. Internal
rotation NJM opposes small external rotation angular velocity until about the 75 degree position on the pushrim, where angular
velocity switches direction and accelerates in the internal rotation direction. (External/Internal (E/I), Adduction/Abduction
(Ad/Ab), Horizontal Adduction/Abduction (HAd/HAb), Flexion/Extension (F/E))
The kinematic context of the shoulder NJM for this subject can be seen in the corresponding
parameterized plots of angular velocity and power in Figure 7. We can see that the internal rotation
NJM opposes external rotation angular velocity of upper arm during the first half of push cycle until the
hand is around the 75 degree position on the wheel where motion switches direction and accelerates in
the internal rotation direction. Also, close to zero angular velocity about the AdAb axis signifies little
movement in either direction about that axis. However, the presence of adduction NJM suggests that
this subject is pushing medially into the pushrim which is confirmed in the outward orientation of the
reaction force. This torque component with little resulting motion leads to 0.07 Nm of NJM power
impulse about the AbAd axis. NJM power impulse for this subject was largely about the flexion axis
11.28 Nm, compared to 2.09 Nm about the HAdHAb and -0.26 Nm about the EI axis.
Comparing between parameterization about 3 torso axes and this novel 4-axis parameterization
technique in Figure 8, we can see how each method would lead to a different interpretation of
mechanical demand on the shoulder. Subject (a) displays a horizontal adduction moment above 20 Nm
when parsing among just the 3 torso axes. However, with the 4-axis approach we can see that the
period of high Horizontal Adduction is actually a large moment in the internal rotation direction.
Inclusion of this EI axis decreases the NJM about the HAdHAb axis below 15 Nm for this subject.
Similarly, subject (b) also shows a reduction in NJM about the HAdHAb axis when including 4 axes
particularly at the end of their push. When the upper arm hangs down beside the torso, as is common
during WC propulsion since the object is to grab the pushrims on either side of the WC user’s hips, the
longitudinal axis of the upper arm aligns most closely with the superior-inferior axis of the torso. In
scenarios like this, moments about the EI axis will be attributed to the HAdHAb axis if just the 3 torso
59
axes are used to parse the NJM (Figure 8).

Figure 24: Side-by-side comparison of 4-axis parsing versus 3-axis torso parsing for 2 example subjects (a,b). Note, NJM
component about the external-internal axis on left gets incorrectly attributed to NJM in the horizontal adduction direction.
Anatomical axis External-Internal Rotation Axis (blue), Flexion-Extension Axis (brown), Adduction-Abduction (yellow), and
Horizontal Abduction-Adduction (purple). NJM is plotted against hand position on pushrim reported in degrees with 90 degrees
representing 12:00 position and 0 degrees at the 3:00 position on the handrim, if subject was facing to the right.
Discussion
This proposed method of parsing the NJM is designed to capture axis specific mechanical demand and be
applicable to any joint, any task, and any subject without the need for modification. Including the
longitudinal axis of the distal segment in conjunction with the three fixed axes of the adjacent proximal
60
segment when parsing the NJM, removes the need for investigators to choose a specific 3-axis coordinate
system for their research study. With 4 axes, all directions of joint effort are accounted for so the demand
during any task at any joint can be reported accurately and completely while using clinically intuitive
terms. Applying this new method of parsing the shoulder NJM to the task of manual WC propulsion
revealed multiple strategies in terms of axis distribution. A mixture of single axis dominant techniques
was observed, with NJM about the FE axis being the most common, Figure 2. Other subjects employed an
even split between two or three axes (Figure 3), where the NJM vector aligns between anatomical axes.
In the case of WC propulsion, differences in NJM distribution arise from different orientations of upper
arm relative to the torso. When the upper arm is abducted away from the torso the expected NJM
distribution would include more effort in the Horizontal Adduction direction, Figure 4. When RF
orientation is maintained in line with the arm-plane very little axial torque about the upper arm is created
and therefore little NJM about the EI axis is seen. Conversely, when RF is oriented out of the arm-plane
NJM about the EI will appear, Figure 6.
Including an isolated external-internal NJM component when describing shoulder kinetics illustrates
how multiplanar loading of the upper extremity can be understood through a four-axis anatomical
component parameterization. With more clearly delineated anatomical axes, subject loading technique
characterization becomes more robust and eliminates scenarios where the amount external-internal
NJM may otherwise be obscured and attributed to torque about one of the three body fixed torso axes
(Figure 8). This shows that even during WC propulsion, a task which may not have been thought to have
torque about the EI axis can have significant axial torque for specific propulsion techniques.
Further extending this method of 4-axis parametrization to calculation of the power about each of these
four axes helps provide kinematic context under which the NJM is generated. With the three profiles of
angular velocity, NJM, and power we can determine which torques acted concentrically to assist in
motion of the upper arm as well as which torques acted eccentrically opposing motion of the arm. In the
case of WC propulsion motion in the flexion and horizontal adduction direction was achieved through
torques about their corresponding axis in the same direction. NJM about the EI was seen opposing
rotational motion of the upper arm, like in Figure 7, eventually causing a switch in direction of motion.
Additionally, NJM about the AbAd axis was seen without corresponding movement of the upper arm in
the adduction direction signifying effort beyond the direction of motion perhaps to maintain
grip/friction on the pushrim.
An effective means of characterizing the mechanical demand imposed on the shoulder may potentially
provide clinical value in being able to better prepare patients for the demands of the task. Using this
method of reporting NJM we get a detailed picture of the net mechanical load distribution on any joint
described about muscularly relevant axes. While this paper primarily focused on shoulder NJM during
manual wheelchair propulsion, this method can and should be applied to other joints and tasks.

61
References
1. Morrow MMB, Hurd WJ, Kaufman KR, An K-N. Upper extremity joint kinetics expression during
wheelchair propulsion. J Rehabil Res Dev. 2009;46(7):939-944.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2857327/.
2. Schache AG, Baker R. On the expression of joint moments during gait. Gait Posture.
2007;25(3):440-452. doi:10.1016/j.gaitpost.2006.05.018.
3. Wagner E V, Russell IM, Muller-Karger C, et al. An approach for characterizing complex
multiplanar upper extremity motion through parsed angular velocity vector components. In:
American Society of Biomechanics.; 2016:1-2.
4. Wagner E V. Using nonlinear feedback control to model human landing mechanics. 2018.
http://digitallibrary.usc.edu/cdm/ref/collection/p15799coll40/id/472327/rec/1.
5. Collinger JL, Boninger ML, Koontz AM, et al. Shoulder Biomechanics During the Push Phase of
Wheelchair Propulsion: A Multisite Study of Persons With Paraplegia. Arch Phys Med Rehabil.
2008;89(4):667-676. doi:10.1016/j.apmr.2007.09.052.
6. Desroches G, Aissaoui R, Bourbonnais D. Relationship Between Resultant Force at the Pushrim
and the Net Shoulder Joint Moments During Manual Wheelchair Propulsion in Elderly Persons.
Arch Phys Med Rehabil. 2008;89(6):1155-1161. doi:10.1016/j.apmr.2007.10.040.
7. Mercer JL, Boninger M, Koontz A, Ren D, Dyson-Hudson T, Cooper R. Shoulder joint kinetics and
pathology in manual wheelchair users. Clin Biomech. 2006;21(8):781-789.
doi:https://doi.org/10.1016/j.clinbiomech.2006.04.010.
8. van Drongelen S, van der Woude LH, Janssen TW, Angenot EL, Chadwick EK, Veeger DH.
Mechanical Load on the Upper Extremity During Wheelchair Activities. Arch Phys Med Rehabil.
2005;86(6):1214-1220. doi:10.1016/j.apmr.2004.09.023.
9. Sabick MB, Kotajarvi BR, An K-N. A new method to quantify demand on the upper extremity
during manual wheelchair propulsion. Arch Phys Med Rehabil. 2004;85(7):1151-1159.
http://www.ncbi.nlm.nih.gov/pubmed/15241767. Accessed June 3, 2018.
10. Veeger HEJ, Rozendaal LA, Van der Helm FCT. Load on the shoulder in low intensity wheelchair
propulsion. Clin Biomech. 2002;17(3):211-218. doi:10.1016/S0268-0033(02)00008-6.
11. Koontz AM, Cooper R a, Boninger ML, Souza AL, Fay BT. Shoulder kinematics and kinetics during
two speeds of wheelchair propulsion. J Rehabil Res Dev. 2002;39(6):635-649.
12. Kulig K, Rao SS, Mulroy SJ, et al. Shoulder Joint Kinetics During the Push Phase of Wheelchair
Propulsion. Clin Orthop Relat Res. 1998;354:132-143.
13. Boninger ML, Cooper RA, Robertson RN, Rudy TE. Wrist biomechanics during two speeds of
wheelchair propulsion: an analysis using a local coordinate system. Arch Phys Med Rehabil.
1997;78(4):364-372. http://www.ncbi.nlm.nih.gov/pubmed/9111455. Accessed June 3, 2018.
14. Cooper RA, Boninger ML, Shimada SD, Lawrence BM. Glenohumeral joint kinematics and kinetics
for three coordinate system representations during wheelchair propulsion. Am J Phys Med
Rehabil. 78(5):435-446. http://www.ncbi.nlm.nih.gov/pubmed/10493454. Accessed June 3, 2018.
62
15. Finley M a, Rasch EK, Keyser RE, Rodgers MM. The biomechanics of wheelchair propulsion in
individuals with and without upper-limb impairment. J Rehabil Res Dev. 2004;41(3B):385-395.
doi:10.1682/JRRD.2004.03.0385.
16. Mulroy SJ, Farrokhi S, Newsam CJ, Perry J. Effects of spinal cord injury level on the activity of
shoulder muscles during wheelchair propulsion: an electromyographic study. Arch Phys Med
Rehabil. 2004;85(6):925-934. doi:https://doi.org/10.1016/j.apmr.2003.08.090.
17. Mulroy SJ, Newsam CJ, Gutierrez D, et al. Effect of Fore-Aft Seat Position on Shoulder Demands
During Wheelchair Propulsion: Part 1. A Kinetic Analysis. J Spinal Cord Med. 2005;28(3):214-221.
doi:10.1080/10790268.2005.11753815.
18. Lighthall-Haubert L, Requejo PS, Mulroy SJ, et al. Comparison of Shoulder Muscle
Electromyographic Activity During Standard Manual Wheelchair and Push-Rim Activated Power
Assisted Wheelchair Propulsion in Persons With Complete Tetraplegia. Arch Phys Med Rehabil.
2009;90(11):1904-1915. doi:10.1016/j.apmr.2009.05.023.
19. Santos Requejo P, Lee S, Mulroy S, et al. Shoulder Muscular Demand During Lever-Activated Vs
Pushrim Wheelchair Propulsion In Persons With Spinal Cord Injury. J Spinal Cord Med.
2008;31(5):568-577. doi:10.1080/10790268.2008.11754604.
20. Cooper RA, Digiovine CP, Boninger ML, Shimada SD, Koontz AM, Baldwin MA. Filter frequency
selection for manual wheelchair biomechanics. J Rehabil Res Dev. 2002;39(3):323-336.
21. Wu G, van der Helm FCT, (DirkJan) Veeger HEJ, et al. ISB recommendation on definitions of joint
coordinate systems of various joints for the reporting of human joint motion—Part II: shoulder,
elbow, wrist and hand. J Biomech. 2005;38(5):981-992.
doi:https://doi.org/10.1016/j.jbiomech.2004.05.042.
22. De Leva P. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J Biomech.
1996;29(9):1223-1230. doi:10.1016/0021-9290(95)00178-6.

63
Chapter 6: Modifications in wheelchair propulsion technique
following clinical wheelchair adjustment

Introduction
Dependence on the upper extremity for mobility and daily activities exposes individuals with spinal cord
injury (SCI) to repetitive mechanical loading that often exceeds current ergonomic recommendations.
1,2

Clinical guidelines, based on clinical and epidemiologic evidence, ergonomics, and expert opinions,
recommend customized wheelchair (WC) seating as a promising means to mitigate loading of the
shoulder in manual WC users.
3
These recommendations align with the National Research Council and
the Institute of Medicine report on Musculoskeletal Disorders and the Workplace (2001) which indicates
that modifications to task performance can reduce the incidence of pain and cumulative trauma
disorders in the workplace.
2
Ongoing clinical research using experimental designs investigating specific
changes in WC seating confirms that modifications to WC seating alter multiple factors known to affect
upper extremity joint kinetics during manual WC propulsion in individuals with paraplegia.
4–9
These
multiple interacting factors affecting upper extremity joint kinetics include modifications to segment
kinematics, reaction forces (RF) generated at the pushrim, and neuromuscular control. These are the
same factors affecting individual differences in shoulder joint contact loads as measured during manual
WC propulsion in older adults with instrumented shoulder prostheses.
10
In addition, modification in task
performance conditions (e.g. fast propulsion, up an incline) are known to affect push angle, RF
magnitude, direction and time of peak RF during the push phase of WC propulsion.
11

Implementation of customized WC seating into clinical practice tends to be iterative with an emphasis
on the individual’s posture, pressure distribution, and stability requirements.
3
Personalizing the WC fit
to the specific mobility needs of the individual WC user remains highly dependent on the expertise of
the clinician and their ability to account for multiple interacting factors known to affect upper extremity
mechanical loading. In this study, we used a within-subject experimental design to determine how WC
reconfiguration to address posture, alleviating pressure areas, and improving balance affects posture
and joint kinetics during manual WC propulsion. The clinician’s goal when configuring the WC is to aid
the subject in sitting upright, but also avoid being reclined too far back. The balance of the wheelchair
should be such that the patient does not easily tip over when sitting stationary or during propulsion. In
this study we analyzed the task of fast WC propulsion because it is of the more mechanically demanding
yet common skills a WC user must have and also provides a more difficult dynamic context to evaluate
the effect of WC configuration.
In this study, we used a within-subject experimental design to determine how individual manual WC
users with paraplegia modify WC propulsion mechanics following a clinical WC fitting visit to adjust WC
configuration for the purpose of addressing posture, balance, and pressure. As found in previous
investigations that looked at specific modifications in WC seating
4–9
, we expected that individuals would
use different WC propulsion techniques as well as posture to accommodate the different WC
configurations. Modifications in WC propulsion technique between baseline and following one month in
the new configuration were characterized by identifying within-subject differences in posture and upper
64
extremity joint kinetics. We hypothesized that reconfiguration of the WC to address posture and
balance will result in differences in shoulder height relative to wheel axle and torso angle relative to
right horizontal during self-selected fast propulsion at the start of elbow extension during the push. We
expect an adjustment made to improve a hunched posture would increase the torso angle bring it closer
to 90 degrees reflecting a straight-up torso position over the main wheel axle as well as increase in
shoulder height relative to the wheel axle. We also expect differences in mechanical demand on the
upper extremity to be affected by shifts in orientation of the RF relative to the forearm and upper arm,
as found in other impulse generating tasks
12,13
, WC propulsion model simulations
14
, and WC ergometer
studies
11
. By investigating WC propulsion technique modifications in response to changes in WC
configuration we hope to gain knowledge of WC adjustment strategies that may prove fruitful in guiding
clinical decisions that aim to identify approaches to preserving shoulder function in individuals with SCI.

Methods

Participants
Twenty two participants (21 Male, 1 Female) with complete SCI who were experienced manual WC users
with paraplegia (T2-L3) from the outpatient clinics of the Rancho Los Amigos National Rehabilitation
Center volunteered to participate. Each participant was provided informed consent in accordance with
the Institutional Review Board. Individuals were excluded from participation if they reported a history of
shoulder pain that altered performance of daily activities or required medical treatment. Average
(standard deviation) weight of participants was 79.96 (20) kg, average height was 1.76 (0.1) m and
average age was 31 years (range: 18 to 47 years). The mean time since occurrence of the injury to the
baseline data collection was 2 years (range: 2 months to 25 years).

Inclusion criteria: paraplegia from complete SCI (SCI level thoracic or lower, American Spinal Cord Injury
Association A or B with no motor function below the level of SCI), and in their own WC for at least one
month, and at entry into the study be free of shoulder pain that interferes with daily activities or
requires medical intervention with a total score on the Wheelchair Users Shoulder Pain Index of 10 or
less.
Exclusion criteria: positive impingement signs (positive Hawkins-Kennedy test and painful arc in
shoulder abduction or flexion), biceps tendonitis (positive Speed’s test), adhesive capsulitis, or cervical
radiculopathy at the initial evaluation a history of shoulder injury or surgery or orthopedic or neurologic
disorders (other than SCI) that would impact arm function.

Instrumentation
For this study, all participants propelled their own wheelchairs in a courtyard outside of the Seating
Center at RLANRC. Reaction force applied by the hand to the pushrim was measured using three strain
gauge force transducers at 240Hz (SmartWheel, Three Rivers Holdings, Mesa, AZ, USA). Two-
dimensional video was captured in the frontal and sagittal planes (60 Hz JVC). Sagittal video was used to
determine body segment kinematics during propulsion cycles. Inertial measurement units (APDM) were
65
also secured to the upper extremity and collected during propulsion to test feasibility of collecting
during a clinical fitting visit.

Experimental Protocol
Prior to data collection, subject upper extremity was wrapped in colorful Coban and marked with tape
to facilitate the identification of wrist, elbow, and shoulder joint centers in the sagittal video. Next, each
participant’s anthropometric and wheelchair measurements were taken, as well as pictures sitting in
their wheelchair with their hand at the top of the pushrim as well as arm hanging down towards the
axle. Each of these pictures was taken from the front, side and rear views. Next, the participants
performed three or more trials of approximately 6-10 WC propulsion cycles at self-selected fast speed in
the courtyard outside the seating center at Rancho Los Amigos National Rehabilitation Center. Each trial
was initiated from a stationary position covered a distance of about 10m. Each participant performed
WC propulsion at their self-selected fast speed, as if they are in a hurry to not miss an important
appointment.

Data Processing and Analysis
The kinematic and force data were synchronized using a tap on the pushrim which could be identified as
a spike in force data as well as seen in the video. The reference system of the SmartWheel was used to
establish a moving WC reference system by rotating average wheel camber of 3 degrees about the Fx
direction to create a vertical, progressional (forward direction of wheelchair translation), and
perpendicular to sagittal camera right hand reference system. Manually digitized segment endpoints
were used to create 3 dimensional reconstructed upper extremity kinematics in this WC reference
system. Kinematic data were digitized from the 2D digitized joint segment endpoints from the sagittal
video (60 Hz). Digitized data was filtered in Matlab using a 4
th
order butterworth filter with a cutoff
frequency of 6Hz and up-sampled to 240 Hz to match the force data.
15
The 2D pixel location of these
endpoints were then scaled to meters according to the physically measured length from wrist joint
center to wheel center when the subject’s hand is grabbing the pushrim and located at the 12 o’clock
position. Next, these 2D points were extrapolated into 3 dimensions using the assumptions that during
hand contact with the pushrim and force generation above 5 N the elbow lies in plane with the wrist.
The second assumption was that the shoulder is offset from both the wrist and elbow by a constant
amount throughout the trial and therefore it’s offset in the Z could be measured from front or rear static
photos and similarly set constant throughout the 3D reconstructed upper extremity kinematics.

To account for the difficulty and therefore variation in digitizing of the shoulder joint center in sagittal
video and to maintain the accuracy of the longitudinal vector orientation of the upper arm, the (X,Y)
shoulder position was adjusted to maintain constant 3D length and constant measured offset along the
Z-axis. 3D reconstructed kinematics of the upper extremity were then combined with the 3D collected
force data to calculate 3D NJM at the elbow and shoulder using inverse dynamics in Matlab. Segment
mass, moment of inertia, and center of mass locations, were based on body segment parameters.
16

Push duration, was determined using measured pushrim RF data and was defined as the phase of torque
application on pushrim when moment about the wheel axle was greater than 5 Nm. The magnitudes of
66
the resultant NJM impulses at the elbow and shoulder were calculated by integrating (area under the
NJM time curves) over the push duration when torque on the wheel was greater than 5 Nm. The
orientation of the RF relative to the forearm was expressed by the angle of the resultant RF and the
forearm projected into sagittal plane with positive angles representing anterior to the forearm and
negative angles being posterior. Torso angle is defined as the angle between the sagittal vector
connecting WC axle and shoulder joint center relative to the right horizontal (Figure 1). Shoulder height
is the vertical distance between wheel axle and projected shoulder joint (Figure 1).

Figure 25: Shoulder height is the vertical distance between wheel axle and projected shoulder joint. Torso angle is defined as
the angle between the sagittal vector connecting wheel axle and shoulder joint relative to the right horizontal.
Statistics
The probabilities of each variable being less during the 30 days follow-up than the baseline when
comparing across propulsion days was calculated using a Wilcoxon-Mann-Whitney percentile bootstrap
test with 2000 bootstrap samples. Assuming local independence for trials and that the baseline and
follow-up propulsion cycles were independent for each subject (R, open-source). For a within subject
analysis, a p-value was calculated for each subject using Cliff’s analog of the Wilcoxon-Mann-Whitney
test.
17
A modified step-down Fisher-type method was then applied to control the familywise error rate
of α = 0.05 over multiple comparisons.
18
This within-subject analysis was used to determine which
subjects had statistically significant changes when comparing their propulsion cycles from the baseline
collection day to their propulsion cycles on the follow-up collection day. Within-subject analysis was
calculated an additional time using a Wilcoxon-Mann-Whitney percentile bootstrap test with 2000
67
bootstrap samples if the conservative step-down technique failed to reject, however still showed a p-
value less than 0.05 for a particular within-subject comparison.

Results
Changes to WC configuration varied across subjects depending on individual patient needs, baseline
configuration, and that WC’s adjustability. Twenty out of 22 participants had some adjustment made to
the seatback with the most common adjustment being made to the seatback angle, which was often
repositioned to be as close to perpendicular to the ground as possible while still allowing for the
participant feel comfortable (Figure 2). This means that the seatback in the baseline configuration was
commonly reclined away from perpendicular to the ground and was adjusted to be more vertical. Other
adjustments to the seatback included shifting it higher or lower, forward or backward according to
clinician’s goal of properly supporting the torso as well as avoiding any discomfort from rubbing or
leaning against it. In addition to the translational adjustments to the seatback, two subjects had more
foam added, one participant had the suspension tightened (to make the upholstery less convex), and
one participant switched from a tension adjustable upholstery seatback to a solid seatback. Adjustments
to the seatback had to carefully account for the patient’s spinal cord injury level and muscular control
over their trunk. Six participants had adjustments to their main axle position, five of which were shifted
forward relative to the seatback. Seven adjustments were made to the seat dump (difference between
rear and front seat height). In addition to documenting the WC adjustments, it was also noted whether
the participant complained or the clinician made particular note of a posture or balance issue in the
baseline WC configuration (Figure 2). These issues included whether the WC felt “too tippy” where the
participant felt their WC would tip backwards and over too easily, whether their posture was too
hunched over or slouched, or whether the WC was too reclined. These issues were made of particular
note since they make up two out of the three factors (posture, balance, pressure) clinicians specifically
adjust WC to account for.
68

Figure 26: Summary table of adjustments made to participant wheelchairs. Changes to seatback angle are labeled as vertical if
the angle was adjusted to be closer to perpendicular. Direction of axle shift was designated as forward when it increased the
distance in front of the seatback post and back when it decreased that distance. Dump is the difference in front and rear seat
height in inches. If a modification was made to seat dump, both the baseline and new configuration dumps are shown (baseline
to new config.).
Mean velocity across all participant propulsion cycle 3 and cycle 4 pushes during period when moment
on the wheel was greater than 5 Nm for baseline collection was 1.66 m/s with a standard deviation of
0.30 m/s. Mean velocity across all participant propulsion cycle 3 and cycle 4 pushes for follow up was a
similar 1.62 m/s with a standard deviation of 0.26 m/s. Mean velocity did not significantly change across
69
all participants between baseline and follow up propulsion cycles (p = 0.271). Within-subject analysis
indicated that 7 participants out of 22 made shifts in velocity between baseline and follow up propulsion
velocity with 2 increasing their speed and 5 decreasing (Figure 3). The largest shifts in mean velocity
were seen in participants 22 and 17, who reduced velocity by 0.42 m/s and 0.45 m/s respectively.
However, the other six shifts in velocity although statistically significant, were less than 0.2 m/s. From
previous work by Russell et al, 2015 a shift from self-selected free propulsion speed to self-selected fast
propulsion speed was found to be on average an increase 0.7 m/s.

Figure 27: Within-subject comparison of mean propulsion velocity for each cycle for Baseline (blue dots) and Follow-up (red
dots) data collections. Subjects ordered by mean baseline velocity. Subjects with statistically significant changes in velocity are
signified with a curly bracket and an asterisk (black Cliff’s step-down, gold Wilcoxon-Mann-Whitney) above their column of
data. Calculated over push duration, when moment about wheel axle was greater than 5 Nm.
The shoulder height relative to wheel axle at the start of elbow extension did significantly change across
all participants (p=0.004) between baseline and follow up propulsion cycles. Within-subject
comparisons indicated that 8 of 22 participants shifted shoulder height at the start of elbow extension
between baseline and follow-up propulsion cycles (Figure 4). All but one of the significant shifts were in
the direction of increasing distance away from the wheel axle which was hypothesized based on the goal
of encouraging more upright posture during propulsion. When analyzing shift in shoulder height it is also
important to note any adjustments in rear seat height as the change in shoulder height may be a result
of the increased rear seat height and not a more upright posture. For example, participant 22 shifted
higher by 0.055 meters and had their rear seat height raised by 0.05 meters. Although, participant 4 who
shifted shoulder height lower at the start of elbow extension did so by 0.066 meters and had their WC
70
rear seat dropped by only 0.038 meters during adjustments, which suggests an additional shift due to
change in torso posture. Additionally, participant number 3 had adjustments to drop rear seat height by
0.006 meters but raised shoulder position by 0.078 meters. Three participants had shifts in shoulder
height without any adjustment in rear seat height for example participant 6 raised shoulder height by
0.062 meters.

Figure 28: Within-subject comparison of shoulder to axle height at the start of elbow extension for each push cycle during
Baseline (blue dots) and Follow-up (red dots) data collections. Subjects ordered by mean baseline velocity. Subjects with
statistically significant changes in shoulder height velocity are signified with a curly bracket and an asterisk (black Cliff’s step-
down, gold Wilcoxon-Mann-Whitney) above their column of data.
The torso angle relative to the right horizontal at the start of elbow extension did not significantly
change across all participants (p=0.952) between baseline and follow-up propulsion cycles. Almost all
torso angles across participants at the start of elbow extension remain below 90 degrees, which would
represent the angle where shoulder is directly above wheel axle. Only participant 1 showed a torso
angle consistently greater than 90 degrees during his follow up propulsion cycles. Within-subject
comparison of torso angle relative to the right horizontal at the start of elbow extension revealed that 9
out 22 participants made significant shifts between baseline and follow-up propulsion cycles (Figure 5).
Four subjects increased torso angle in a more upright or reclined direction and 5 subjects decreased
torso angle. The largest shift in torso angle was by participant 1 who shifted in the upright direction by
10 degrees. All other significant shifts were by less than 5 degrees. All four of the participants that
shifted towards more upright torso angle were participants who also had complaints about posture or
balance. Interestingly, posture improved (more upright) for two subjects that had adjustments for
71
feeling unstable, which may suggest that a hunched posture is potentially a side effect of an unstable
WC configuration. None of the five participants who shifted towards decreasing torso angle at the start
of elbow extension had complaints about posture or balance (Figure 2). Additional participants
displayed bent over torso angles beyond the participants who had posture complaints, however it was
not addressed as a complaint at the time of fitting.

Figure 29: Within-subject comparison of torso angle relative to right horizontal at the start of elbow extension for each push
cycle during Baseline (blue dots) and Follow-up (red dots) data collections. Subjects ordered by mean baseline velocity. Subjects
with statistically significant changes in torso angle are signified with a curly bracket and an asterisk (black Cliff’s step-down, gold
Wilcoxon-Mann-Whitney) above their column of data.
Shoulder NJM impulse did not significantly change across all participants (p=0.866) between baseline
and follow-up propulsion cycles, calculated over period when torque on the wheel was above 5 Nm.
Within-subject analysis of shoulder NJM impulse between baseline and follow-up propulsion cycles
showed 4 out of 22 participants had significant shifts (Figure 6). Three participants decreased NJM
impulse on the shoulder with participants 6 and 1 decreasing by 43% and 47% respectively from
baseline shoulder NJM impulse. Participant 12 increased shoulder NJM impulse by 29% compared to
baseline push cycles.
72

Figure 30: Within-subject comparison of NJM impulse on the shoulder for each push cycle during Baseline (blue dots) and
Follow-up (red dots) data collections. Subjects ordered by mean baseline velocity. Subjects with statistically significant changes
in shoulder NJM impulse are signified with a curly bracket and an asterisk (black Cliff’s step-down, gold Wilcoxon-Mann-
Whitney) above their column of data. Calculated over push duration, when moment about wheel axle was greater than 5 Nm.
Elbow NJM impulse did not significantly change across all participants (p=0.774) between baseline and
follow up propulsion cycles, calculated over period when torque on the wheel was above 5 Nm. Within-
subject analysis of elbow NJM impulse between baseline and follow-up propulsion cycles revealed 7 out
of 22 participants had significant shifts (Figure 7). Four increased elbow NJM impulse with participant 6
showing the largest increase of 57%. Participants 9, 10, and 11 also increased elbow NJM impulse by
31%, 41%, and 48% respectively. Three participants decreased elbow NJM impulse comparing baseline
to one month follow up with participant 22 showing the largest decrease of 42%. The other two
participants (7 and 19) decreased by 32% and 33%.
73

Figure 31: Within-subject comparison of NJM impulse on the elbow for each push cycle during Baseline (blue dots) and Follow-
up (red dots) data collections. Subjects ordered by mean baseline velocity. Subjects with statistically significant changes in
elbow NJM impulse are signified with a curly bracket and an asterisk (black Cliff’s step-down, gold Wilcoxon-Mann-Whitney)
above their column of data. Calculated over push duration, when moment about wheel axle was greater than 5 Nm.
Resultant RF impulse did not significantly change across all participants (p=0.821) between baseline and
follow up propulsion cycles, calculated over period when torque on the wheel was above 5 Nm. Within-
subject testing of resultant RF impulse between baseline and follow-up propulsion cycles indicates 7 out
22 subjects showed a significant shift (Figure 8). Four participants increased RF impulse and three
decreased RF impulse. The largest increase in RF impulse was seen in participant 10 who increased by
42%. The largest decrease in RF impulse was by participant 22 who dropped by 34%. Only participant 4
had a significant change in push duration that contributed to the RF imp shift, which means all other RF
impulse shifts were due to changes in RF magnitude. The other participants with shifts in RF impulse
ranged from 13%-30%.
74

Figure 32: Within-subject comparison of resultant RF impulse on the elbow for each push cycle during Baseline (blue dots) and
Follow-up (red dots) data collections. Subjects ordered by mean baseline velocity. Subjects with statistically significant changes
in RF impulse are signified with a curly bracket and an asterisk (black Cliff’s step-down, gold Wilcoxon-Mann-Whitney) above
their column of data. Calculated over push duration, when moment about wheel axle was greater than 5 Nm.
Mean RF angle relative to the forearm from the start of elbow extension to the end of the push cycle did
not significantly change across all participants (p=0.210) between baseline and follow up propulsion
cycles. Within-subject analysis of mean RF angle relative to forearm between baseline and follow-up
propulsion cycles revealed that 7 out of 22 participants had significant shifts in RF orientation (Figure 9).
From Figure 10, we can see that participants move RF orientation relative to the forearm throughout
the push duration. Five of those participants who were identified as having significant shifts between
mean orientations over the elbow extension phase also appear to have shifted RF orientation
throughout the push duration, not just the mean value over the phase of elbow extension (Figure 10).
These five participants also all shifted in a direction towards mitigating the potential magnitude of the
shoulder NJM.
11
These shifts include orienting the RF less posteriorly or from posterior to inline with the
forearm thus reducing the magnitude of the elbow flexor NJM, or shifting towards a more anteriorly
oriented RF which creates an elbow extensor NJM that contributes to a reduction in shoulder NJM.
75

Figure 33: Within-subject comparison of mean RF angle relative to forearm from start of elbow extension to end of each push
cycle during Baseline (blue dots) and Follow-up (red dots) data collections (negative angles posterior to forearm, 0 degrees
representing inline with forearm, and positive angles anterior to forearm). Subjects ordered by mean baseline velocity. Subjects
with statistically significant changes in orientation are signified with a curly bracket and an asterisk (black Cliff’s step-down, gold
Wilcoxon-Mann-Whitney) above their column of data.
76

Figure 34: Five participants who showed significant shifts in mean reaction force (RF) orientation relative to forearm from start
of elbow extension to end as well as throughout the push cycle (negative angles posterior to forearm, 0 degrees representing
inline with forearm, and positive angles anterior to forearm). These five participants all shift RF orientation in a direction
consistent with mitigating the potential magnitude of the shoulder net joint moment. Shown during push duration when
moment on wheel is greater than 5 Nm.
Four out of the 6 participants with posture or balance complaints displayed shifts towards a more
upright torso angle at the start of elbow extension as well as higher shoulder position relative to the
wheel axle at the same instant (Figure 11). The adjustment for each of these six individuals with posture
or balance complaints was to angle the seatback to be more vertical moving it from a reclined position
to be more perpendicular to the ground. Among the participants with shifts in shoulder NJM impulse,
we can see that those who shifted in the direction of reducing demand on the shoulder all had a more
upright torso angle, higher shoulder position as well as either a re-orientation of RF or a reduction in RF
impulse. The largest reductions in shoulder NJM impulse were of 47% and 43% by participants 1 and 6
and given no significant shift in RF impulse, the decrease in shoulder demand can be tied to re-
orientation of the RF relative to the upper extremity. If the orientation of the RF is anterior to the
center of mass (CM) of the forearm, the NJM at the elbow will be an extensor NJM. If the RF is posterior
to the CM of the forearm, a flexor NJM will be required. The direction of the NJM, along with the
proximal and distal moments imposed on the upper arm by the net joint forces at the elbow and
shoulder will affect the magnitude of the shoulder NJM. For the same RF magnitude, the lowest
77
corresponding shoulder NJM will occur in the region where RF is oriented anterior to the forearm and
close to but not anterior to the shoulder joint center. The one participant with an increase in shoulder
NJM impulse also had an increase in RF impulse.

Figure 35: Summary table of within-subject significant changes in propulsion technique. Also included in the final right column
are the participants with specific posture or balance complaints. Percent shift from baseline magnitude to follow up magnitude
is included in parenthesis next to the direction of shift for Shoulder NJM impulse, Elbow NJM impulse, and RF impulse. Shift in
RF orientation angle to forearm is described based on the value of mean RF angle over elbow extension period.
If we look at the coordination between elbow angle and torso angle over the entire push cycle for the
subgroup of participants with balance and postural complaints we can see noticeable shifts towards
more upright torso angles in 5 of the six individuals (red lines above the blue lines Figure 12). This shift
towards a more upright torso angle throughout push cycle is present among this subgroup for both
participants with posture complaints as well as WC balance complaints, again highlighting that there
78
may be a link between addressing stability of the WC and an upright posture. Even though participant
11 did not have a statistically significant shift in torso angle at the start of elbow extension, their angle-
angle plot shows during extension the baseline torso angle for the majority of the push cycles is more
bent over than the follow up propulsion cycles. Additionally, in participant 1’s plot in Figure 12 we can
see a reduction in variability with more repeatable torso angles during propulsion in the one month
follow up cycles compared to baseline propulsion.

Figure 36: Angle-Angle plots of elbow and torso over the push cycles for the subgroup of participants with postural or balance
complaints. Blue curves are baseline wheelchair configuration propulsion cycles and red curves are from one month follow up
propulsion cycles
Shifts in torso posture at the start of elbow extension can further be seen in the stills of the sagittal
video (Figure 13). Large differences in posture at this instant are especially apparent for participants 1
and 2 when comparing between their baseline and follow-up photos. You can see their torsos are bent
over almost to where their heads are over their knees in the baseline photos but on follow up their head
is closer to being over their hips. Participant 6 shows rounding of the upper back in the baseline picture
but a taller less slouched shoulders and neck posture in the follow-up.
79

Figure 37: Still pictures from sagittal video of the subgroup of participants with postural or balance complaints shown at the
start of elbow extension. Resultant reaction force vector on wrist overlayed onto image with length of vector corresponding to
force magnitude (green anterior to forearm, red posterior to forearm, yellow inline with forearm). Shoulder and elbow net joint
moments shown as yellow circles, radius corresponding to magnitude of moment.
Discussion
Manual wheelchairs are an effective form of low-cost wheeled mobility especially for those with a spinal
cord injury. In addition to providing mobility, manual WC help preserve upper body strength, maintain
cardiovascular health, and independence in the community. However, the high incidence of shoulder
pain among this population has been attributed to the repetitive mechanical loading of the upper limb
as a consequence of lower extremity paralysis.
19,20
Individualized fitting of the manual WC to the user
has been highlighted as a promising means to improve the interaction and potentially mitigate loading
on the shoulder.
3
Current clinical WC adjustment focuses on addressing posture, alleviating pressure
areas, and improving balance. However, the biomechanical outcomes of addressing these factors is
largely unknown. In this study, we used a within-subject experimental design to determine how WC
reconfiguration to address posture, pressure and stability affects both joint kinetics and posture during
WC propulsion. By investigating WC propulsion technique modifications in response to adjustments in
WC configuration we hope to gain insights into potential approaches for preserving shoulder function in
individuals with SCI.
We used a within-subject analysis approach for this study to account for the differences in WC
adjustments, which varied depending on the specific needs of that individual. Mean velocity for each
participant was kept within 0.5 m/s between their baseline configuration propulsion and one month
following new WC configuration, even though fast propulsion speed was self-selected. When analyzing
shifts in shoulder height, we also made sure to note any adjustments in rear seat height as the change in
80
shoulder height may be a result of the increased rear seat height and not the result of a more upright
posture. There are additional participants that displayed torso angles in the lower range that were not
part of the group with noted posture complaints. It is possible that more important adjustments to that
individual were focused during the session and posture could have been overlooked. Given the many
interrelated factors that a clinician must keep track of while adjusting the WC future studies should
investigate whether the fitting process could be aided by providing clinicians with more biomechanical
information. It is also important to note that among these other participants with low torso angles, were
some with large amounts of adipose tissue around their hips making the boundary surface of their
bottom a few inches behind the surface of their torso and shoulder. In these scenarios the wheel axle is
commonly positioned further back relative to the shoulder due to the extra body mass. This would likely
appear as bent over torso angle when in reality it may not be because the torso angle was measured by
the line from wheel axle to shoulder joint.
Changes in propulsion technique varied across the population and included shifts in postural kinematics
as well as joint kinetics. The WC adjustments for 20 out of 22 subjects involved modifications to the
seatback with the goal of supporting the lumbar region of the spine, encouraging neutral position of the
pelvis, and allowing for the torso to sit more comfortably in an upright posture. The specific
adjustments to the seatback varied depending on the baseline setup of the wheelchair, the subject’s
ability to use their core muscles (level of spinal cord injury), and the limitations of the wheelchair. Four
out of the 6 participants with posture or balance complaints displayed shifts towards more upright torso
angle at the start of elbow extension as well as higher shoulder position relative to the wheel axle at the
same instant. The shift in torso angle for these 4 participants was additionally seen to appear
throughout the push (Figure 12) and not just at the start of elbow extension. As noted previously this
subgroup includes two participants that had adjustments for feeling “too tippy”, which may suggest a
link between hunched posture and an unstable WC configuration.
The adjustment for each of these six individuals with posture or balance complaints was to angle the
seatback to be more vertical. Adjusting the angle of the seatback from reclined to be more
perpendicular to the ground changes the resting position of the torso relative to the wheel axle and the
boundary surface from which subject can push their back against. If the seatback positions the torso
and therefore the subject’s center of mass too close to the edge of the base of support, the subject
could feel too unbalanced to sit up straight or lean back onto the seatback while propelling. A setup like
this would cause the subject to slouch or hunch over while propelling in order to position their center of
mass safely within the boundaries of the base of support. This theory explains why the WC adjustment
and corresponding posture shift is the same for participants with balance complaint as it is for those
with a posture complaint. If torso posture adjusts to maintain WC balance during overground
propulsion, additional research should investigate the torso posture in scenarios where WC balance is
removed such as on an ergometer. When the risk of tipping over is no longer present because the WC is
secured to a platform, a hunched torso posture might be less common.
The largest reductions in shoulder NJM impulse were of 47% and 43% by participants 1 and 6 and given
no significant shift in RF impulse, the decrease in shoulder demand can be tied to re-orientation of the
RF relative to the upper extremity. Both participants shifted from having RF oriented posteriorly during
81
the first half of their push cycles to orienting the RF inline or anterior to their forearm during this same
phase for their follow up push cycles (Figure 10). When the orientation of the RF is anterior to the CM
of the forearm, the NJM at the elbow will be an extensor NJM, which contributes to a reduction in
shoulder NJM magnitude. Conversely, when RF is oriented posterior to the forearm CM the resulting
elbow flexor NJM contributes to an increase in shoulder NJM magnitude. The results of this study align
with previous experimental work by Russell et al. 2015 and model simulation work by Munaretto et al.
2011, which looked at the effect of RF orientation on upper extremity mechanical demand during WC
propulsion.
At present, clinical guidelines recommend that clinicians simultaneously consider multiple interacting
factors influencing an individual’s posture, pressure, and stability needs when making WC configuration
decisions. Exploring the corresponding effect of addressing these specific areas on a subject’s propulsion
technique helps to reveal the dynamic and joint kinetic outcomes for WC adjustment decisions. Future
studies will examine how a WC users torso posture relates to the balance and stability of the wheelchair
by comparing WC propulsion done on an ergometer with overground WC propulsion.

82
References
1. NIOSH. Musculoskeletal Disorders and Workplace Factors - A Critical Review of Epidemiologic
Evidence for Work-Related Musculoskeletal Disorders of the Neck, Upper Extremity, and Low
Back.; 1997.
2. Council I of M and NR. Musculoskeletal Disorders and the Workplace. Washington, D.C.: National
Academies Press; 2001. doi:10.17226/10032.
3. Medicine PV of AC for SC. Preservation of Upper Limb Function Following Spinal Cord Injury: A
Clinical Practice Guideline for Health-Care Professionals. J Spinal Cord Med. 2005;28(5):434-470.
4. Desroches G, Aissaoui R, Bourbonnais D. Effect of system tilt and seat-to-backrest angles on load
sustained by shoulder during wheelchair propulsion. J Rehabil Res Dev. 2006;43(7):871.
doi:10.1682/JRRD.2005.12.0178.
5. Mulroy SJ, Newsam CJ, Gutierrez D, et al. Effect of Fore-Aft Seat Position on Shoulder Demands
During Wheelchair Propulsion: Part 1. A Kinetic Analysis. J Spinal Cord Med. 2005;28(3):214-221.
doi:10.1080/10790268.2005.11753815.
6. Hughes CJ, Weimar WH, Sheth PN, Brubaker CE. Biomechanics of wheelchair propulsion as a
function of seat position and user-to-chair interface. Arch Phys Med Rehabil. 2017;73(3):263-269.
doi:10.5555/uri:pii:0003999392900769.
7. Mâsse LC, Lamontagne M, O’Riain MD. Biomechanical analysis of wheelchair propulsion for
various seating positions. J Rehabil Res Dev. 1992;29(3):12—28.
8. Majaess GG, Lee Kirby R, Ackroyd-Stolarz SA, Charlebois PB. Influence of seat position on the
static and dynamic forward and rear stability of occupied wheelchairs. Arch Phys Med Rehabil.
2017;74(9):977-982. doi:10.5555/uri:pii:000399939390278I.
9. van der Woude L, Bouw A, van Wegen J, van As H, Veeger D, de Groot S. Seat height: Effects on
submaximal hand rim wheelchair performance during spinal cord injury rehabilitation. J Rehabil
Med. 2009;41(3):143-149. doi:10.2340/16501977-0296.
10. Westerhoff P, Graichen F, Bender A, et al. Measurement of shoulder joint loads during
wheelchair propulsion measured in vivo. Clin Biomech. 2011;26(10):982-989.
doi:https://doi.org/10.1016/j.clinbiomech.2011.05.017.
11. Russell IM, Raina S, Requejo PS, Wilcox RR, Mulroy S, McNitt-Gray JL. Modifications in Wheelchair
Propulsion Technique with Speed . Front Bioeng Biotechnol . 2015;3:171.
https://www.frontiersin.org/article/10.3389/fbioe.2015.00171.
12. McNitt-Gray JL, Hester DM, Mathiyakom W, Munkasy BA. Mechanical demand and multijoint
control during landing depend on orientation of the body segments relative to the reaction force.
J Biomech. 2001;34(11):1471-1482. http://www.ncbi.nlm.nih.gov/pubmed/11672722. Accessed
May 25, 2018.
13. Mathiyakom W, McNitt-Gray JL, Requejo P, Costa K. Modifying center of mass trajectory during
sit-to-stand tasks redistributes the mechanical demand across the lower extremity joints. Clin
Biomech. 2005;20(1):105-111. doi:10.1016/j.clinbiomech.2004.08.005.
83
14. Munaretto JM, McNitt-Gray JL, Flashner H, Requejo PS. Simulated effect of reaction force
redirection on the upper extremity mechanical demand imposed during manual wheelchair
propulsion. Clin Biomech. 2012;27(3):255-262. doi:10.1016/j.clinbiomech.2011.10.001.
15. Winter D. Biomechanics and Motor Control of Human Movement 2nd Edition.; 1999.
16. De Leva P. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J Biomech.
1996;29(9):1223-1230. doi:10.1016/0021-9290(95)00178-6.
17. Cliff N. Answering Ordinal Questions with Ordinal Data Using Ordinal Statistics. Multivariate
Behav Res. 1996;31(3):331-350. doi:10.1207/s15327906mbr3103_4.
18. Wilcox RR, Clark F. Robust Multiple Comparisons Based On Combined Probabilities From
Independent Tests. J Data Sci. 2015;13(1):1-11.
19. Bayley JC, Cochran TP, Sledge CB. The weight-bearing shoulder. The impingement syndrome in
paraplegics. JBJS. 1987;69(5).
20. Dalyan M, Cardenas D, Gerard B. Upper extremity pain after spinal cord injury. Spinal Cord.
1999;37(3):191-195. doi:10.1038/sj.sc.3100802.

84
Chapter 7: Modifications in wheelchair propulsion technique
between ergometer and overground wheelchair propulsion

Introduction
Chronic pain among individuals with a spinal cord injury (SCI) often leads to a decline in health and
independence.
1,2
Current experimental research has looked at investigating wheelchair (WC) propulsion
technique and WC configuration with the hope of improving the interaction between the individual and
the WC.
3
Propulsion collection systems like ergometers and treadmills have provided a convenient way
of collecting multiple wheelchair propulsion cycles in a controlled and stationary environment. These
systems are especially useful when using motion capture cameras, electromyography, or
cardiopulmonary monitoring systems where the equipment does not easily allow for the subject to
move linearly over large distances. Due to the stationary nature of these systems, the WC is secured in
place and therefore cannot tip over. However, during real-world propulsion, this is not the case.
Wheelchairs are designed to have their front wheels free to lift off the ground and able to “pop-a
wheelie” in order to more easily traverse curbs. However, the WC user must be aware and account for
this degree of instability as to not inadvertently tip backwards, especially during mechanically
demanding tasks like fast or graded propulsion. Current WC seating center clinicians are acutely aware
of the importance of modifying WC stability to the user, which is why it is one of the 3 emphases of
individualized WC fitting which include posture, pressure, and balance. Modifying a WC is a difficult task
since making adjustments is also known to affect both posture and propulsion mechanics (shown in
Chapter 5)
4–9
.

After analyzing modifications in propulsion technique pre and post WC fitting, Russell et al., (Chapter 6)
hypothesized a potential connection between torso posture and WC stability. From the results of their
study, they observed shifts towards more upright posture in wheelchair users who had adjustments
specifically targeting complaints of an unstable wheelchair. They theorized that in WC configurations
where the user felt more stable (center of mass (CM) further within the base of support) the user may
be more inclined to sit upright and lean against the seatback without fear of tipping backwards.
Conversely, in scenarios where the CM is positioned too close to the rear edge of the base support the
user might choose to lean forward off of the seatback to shift CM position more safely within the base
of support. Russell et al. further proposed that the hypothesis of torso posture relating to stability could
be further investigated by comparing overground WC propulsion with propulsion on an ergometer, a
context where stability concerns are removed from the mechanical demands of the task.
Previous studies have compared characteristics of wheelchair propulsion in controlled stationary
systems such as ergometers, dynamometer, and treadmills with overground propulsion. However, these
studies analyzed features of the force applied to wheel, physiological measures, or hand trajectory
patterns
10–13
. Other studies have investigated trunk motion as it relates to fatigue
14
, spinal cord injury
level
15
, propulsion speed
16
, and upper-limb impairment
17
. However, no investigation has compared torso
posture or mechanical loading of the upper extremity between the two collection methods.
85
In this pilot study, a manual WC user with paraplegia propelled on an stationary fixed ergometer
apparatus and outside in a realistic setting to determine the differences between WC propulsion
mechanics in varying WC stability scenarios. Based on the theory provided by Russell et al., (Chapter 5)
we hypothesized a more upright torso angle as well as a higher shoulder height during WC propulsion
performed on an ergometer compared to torso angle and shoulder height during overground
propulsion. From data presented in an investigation by Rodgers et al., 2000 which compared propulsion
characteristics between a torso flexion group and a non-torso flexion group of manual WC users during
propulsion, we expected decreased push time and reduced RF impulse during ergometer push cycles.
We also expected differences in upper extremity joint kinetics as a result of shifts in RF magnitude and
orientation. We anticipated that this individual with paraplegia would use different WC propulsion
techniques to accommodate the different stability requirements. By investigating differences in WC
propulsion technique between the collection modalities of ergometer and overground propulsion we
hope to gain insights into the affect of WC stability on propulsion mechanics.
Methods
Participant
One participant with complete SCI who was an experienced manual WC user with paraplegia (T12) from
the clinics of the Rancho Los Amigos National Rehabilitation Center volunteered to participate. The
participant was provided informed consent in accordance with the Institutional Review Board. Weight of
the participant was 68 kg, height was 1.70 m and age was 45 years old. The time since occurrence of the
injury was 25 years.

Ergometer Instrumentation
The participant used a rigid frame, lightweight Quickie GPV wheelchair with front seat height, rear seat
height, horizontal axle position, footrest, seatback angle, and inertial parameter of the test wheelchair
adjusted to match the participant’s own wheelchair. The participant used their own seat cushion. The
wheelchair was positioned on a stationary ergometer, consisting of a support frame and split rollers,
allowing separate rotation of each wheel. The rollers were coupled by means of a differential to an
alternator and a modified Velodyne® bicycle-ergometer that computer-controlled the resistance. To
quantify the friction force between the tire and ergometer rollers, a coast down test (from 182
m/minute to 35 m/minute) with the participant sitting in the test wheelchair on top of the ergometer
was used. Removable flywheels proportional to the weight of both the person and the wheelchair were
used to simulate the translational inertia of “over ground” propulsion. Further details about the
ergometer instrumentation and calibration steps are described in previous papers
5,18–20
. Reaction force
applied by the hand to the pushrim was measured using three strain gauge force transducers at 200Hz
(SmartWheel, Three Rivers Holdings, Mesa, AZ, USA). Two-dimensional video was captured in the frontal
and sagittal planes (60 Hz JVC). Sagittal video was used to determine body segment kinematics during
propulsion cycles.

Ergometer Protocol
Prior to data collection, the participant was given adequate time to become accustomed to the
wheelchair and experimental conditions as well as a few practice trials. The participant performed WC
86
propulsion at their self-selected fast speed, as if in a hurry to not miss an important appointment. Force
and video were then collected for about 15 seconds at the fast speed condition with no additional load
applied to the ergometer rollers (i.e. level ground over a tiled surface).

Overground Instrumentation
For overground propulsion the participant propelled their own wheelchair in a courtyard outside of the
Seating Center at the Rancho Los Amigos National Rehabilitation Center. Reaction force applied by the
hand to the pushrim was again measured using three strain gauge force transducers at 240Hz
(SmartWheel, Three Rivers Holdings, Mesa, AZ, USA). Two-dimensional video was captured in the frontal
and sagittal planes (60 Hz JVC). Sagittal video was used to determine body segment kinematics during
propulsion cycles.

Overground Protocol
The participant performed four trials of approximately 10 WC propulsion cycles at self-selected fast
speed in the courtyard outside the seating center at Rancho Los Amigos National Rehabilitation Center.
Each trial was initiated from a stationary position and covered a distance of about 10m. The participant
performed WC propulsion at their self-selected fast speed, as if they are in a hurry to not miss an
important appointment.

Data Processing and Analysis
The kinematic and force data were synchronized using a tap on the pushrim which could be identified as
a spike in force data as well as seen in the video. The reference system of the SmartWheel was used to
establish a moving WC reference system by rotating average wheel camber of 3 degrees about the Fx
direction to create a vertical, progressional (forward direction of wheelchair translation), and
perpendicular to sagittal camera right hand reference system. Prior to data collection, subject upper
extremity was wrapped in colorful Coban and marked with tape to facilitate the identification of wrist,
elbow, and shoulder joint centers in the sagittal video. Manually digitized segment endpoints were used
to create 3 dimensional reconstructed upper extremity kinematics in this WC reference system.
Kinematic data were digitized from the 2D digitized joint segment endpoints from the sagittal video (60
Hz). Digitized data was filtered in Matlab using a 4
th
order butterworth filter with a cutoff frequency of
6Hz and up-sampled to 240 Hz to match the force data.
21
The 2D pixel location of these endpoints were
then scaled to meters according to the physically measured length from wrist joint center to wheel
center when the subject’s hand is grabbing the pushrim and located at the 12 o’clock position. Next,
these 2D points were extrapolated into 3 dimensions using the assumptions that during hand contact
with the pushrim and force generation above 5 N the elbow lies in plane with the wrist. The second
assumption was that the shoulder is offset from both the wrist and elbow by a constant amount
throughout the trial and therefore it’s offset in the Z could be measured from front or rear static photos
and similarly set constant throughout the 3D reconstructed upper extremity kinematics.

To account for the difficulty and therefore variation in digitizing of the shoulder joint center in sagittal
video and to maintain the accuracy of the longitudinal vector orientation of the upper arm, the (X,Y)
shoulder position was adjusted to maintain constant 3D length and constant measured offset along the
87
Z-axis. 3D reconstructed kinematics of the upper extremity were then combined with the 3D collected
force data to calculate 3D net joint moment (NJM) at the elbow and shoulder using inverse dynamics in
Matlab. Segment mass, moment of inertia, and center of mass locations, were based on body segment
parameters.
22

Push duration, was determined using measured pushrim RF data and was defined as the phase of torque
application on pushrim when moment about the wheel axle was greater than 5 Nm. The magnitudes of
the resultant NJM impulses at the elbow and shoulder were calculated by integrating (area under the
NJM time curves) over this push duration when torque on the wheel was greater 5 Nm. The orientation
of the RF relative to the forearm was expressed by the angle of the resultant RF and the forearm
projected into sagittal plane with positive angles representing anterior to the forearm and negative
angles being posterior. Torso angle is defined as the angle between the sagittal vector connecting WC
axle and shoulder joint center relative to the right horizontal (Figure 1). Shoulder height is the vertical
distance between wheel axle and projected shoulder joint (Figure 1).

Figure 38: Shoulder height is the vertical distance between wheel axle and projected shoulder joint. Torso angle is defined as
the angle between the sagittal vector connecting wheel axle and shoulder joint relative to the right horizontal.
Statistics
The probabilities for each variable being less during the overground propulsion condition than the
ergometer propulsion condition was calculated using a Sign Test between the 8 trials for each condition.

88
Results
Experimental design for this study had the participant propel at self-selected fast speed in both
collection scenarios: ergometer and overground (Figure 2). Propulsion on the ergometer shows
continued push cycles beyond 6 seconds as is common practice to maintain velocity for period of data
collection. Overground propulsion distance was restricted by the boundaries of the courtyard and
therefore shows acceleration of wheelchair velocity and then braking to stop forward motion. The cycles
used to compare across conditions eliminated the first two push cycles (start-up cycles) and for
overground propulsion included cycles 3 and 4 from four seperate push trials. Representative push
cycles for the ergometer included cycles 3 through 10, as is typical of wheelchair ergometer studies to
use repeated sequential pushes from a single bout of propulsion. Mean WC velocity during overground
push cycles was 1.9 m/s and mean WC velocity during ergometer push cycles was 2.6 m/s. Average max
velocity reached over the 4 overground trials was 2.9 m/s and max velocity for the ergometer trial was
3.1 m/s.

Figure 2: Comparison of self-selected fast wheelchair velocities between overground (green) and ergometer (purple) propulsion
trials.
As expected, push duration significantly decreased for ergometer propulsion compared with overground
propulsion (p=0.004, Figure 3). Average push duration for ergometer propulsion was 0.16 seconds while
duration for overground propulsion was 0.24 seconds. Magnitude of resultant reaction force impulse
also decreased for ergometer propulsion compared with overground propulsion (p=0.004). Average RF
impulse was 15 Ns for ergometer propulsion and 25 Ns for overground propulsion. We can see in Figure
89
3 higher resultant RF magnitude towards the end of the push cycle in overground compared to the RF
magnitudes seen in ergometer propulsion. The increase in RF impulse was therefore due to both an
increase in force magnitude as well as push duration.

Figure 3: Comparison of resultant reaction force magnitude measured by the pushrim during overground (green) and
ergometer (purple) propulsion. Reaction forces shown during phase of torque application on pushrim when moment about the
wheel axle was greater than 5 Nm.
As hypothesized, posture during ergometer propulsion was performed with a more upright torso angle
throughout the push cycle (Figure 4a). Further analysis comparing torso angle at the start of elbow
extension between conditions also resulted in a significant shift (p=0.004). On average torso angle was
79 (±2.0) degrees during overground propulsion at the start of elbow extension and 86 (±0.9) degrees
during ergometer propulsion. Figure 4a shows that there was more movement within each trial in torso
angle during overground propulsion as well as more variability between trials. The maximum torso angle
range during a single cycle of overground propulsion was at most 4 degrees, and over all the trials torso
angle reached a minimum of 75 degrees and maximum of 84 degrees. Whereas, during ergometer
propulsion the largest single cycle torso angle range showed movement of just 2 degrees and over all
the trials reached a minimum torso angle of 84 degrees and a maximum torso angle of 88 degrees.
Comparing shoulder height between ergometer and overground propulsion, we see much less of a
visual difference (Figure 4b). Analysis of shoulder height at the start of elbow extension does result in a
significant difference between conditions (p=0.045). Average shoulder height at start of elbow extension
during overground propulsion was 0.67 (±0.01) m and the average height during ergometer propulsion
at the start of extension was 0.68 (±0.01) m. During overground propulsion the shoulder reached a
minimum height of 0.65m and a maximum height of 0.69m. During ergometer propulsion the shoulder
reached a minimum height of 0.65m and a maximum height of 0.70m. Although shoulder height did
90
result in a statistically significant difference and was hypothesized to be lower during overground
propulsion, the magnitude of the difference is small.

Figure 4: Comparison of torso angle (a) measured from angle between the line from the wheel axle to the shoulder joint
relative to right horizontal and shoulder height (b) measured as the vertical distance between wheel axle and shoulder joint
during overground (green) and ergometer (purple) propulsion. Torso angle and shoulder height shown during phase of torque
application on pushrim when moment about the wheel axle was greater than 5 Nm.
Analysis of NJM on the shoulder between overground and ergometer propulsion cycles, appear to show
similar magnitudes but different durations (Figure 5a). Further analysis of the NJM impulse on the
shoulder showed a significant increase (p=0.008) from an average of 3.6 (±1.4) Nms during ergometer
propulsion to an average shoulder NJM impulse of 5.2 (±0.8) Nms. This represents an increase of 43%
when comparing mechanical demand on the shoulder between the two conditions. Comparison of
Elbow NJM shows a large rise in elbow NJM magnitude towards the end of the push cycle. Analysis of
the NJM impulse on the elbow also showed a significant increase (p=0.004) from an average of 1.3 (±0.4)
Nms during ergometer propulsion to an average shoulder NJM impulse of 2.4 (±0.4) Nms, which
represents an increase of 76%.
91

Figure 5: Comparison between shoulder net joint moment (a) and elbow net joint moment (b) during overground (green) and
ergometer (purple) wheelchair propulsion cycles. NJMs are shown during phase of torque application on pushrim when
moment about the wheel axle was greater than 5 Nm.
Orientation of the RF relative to the forearm over the push cycles appears to be shifted less posterior
during the first half of the push cycle and more anterior during the second half the push cycle for
overground propulsion compared to ergometer propulsion (Figure 6). Further analysis of the mean RF
angle over the period of elbow extension between overground and ergometer propulsion confirms a
significant shift in orientation (p=0.004). Although, RF magnitude is larger at the end of the push in
overground propulsion, re-orientation of the RF to be even more anterior to the forearm increases the
extensor NJM at the elbow (Figure 5b), which contributes to a reduction in the possible magnitude of
shoulder NJM magnitude (Figure 5a).
92

Figure 6: Comparison of reaction force orientation relative to forearm (negative angles posterior to forearm, 0 degrees
representing inline with forearm, and positive angles anterior to forearm) during overground (green) and ergometer (purple)
wheelchair propulsion, normalized to duration of each push cycle. Reaction force orientation shown phase of torque
application on pushrim when moment about the wheel axle was greater than 5 Nm.
Analysis of coordination pattern between elbow angle and torso angle over the push cycle between
ergometer and overground propulsion, again reveals a noticeable difference in torso angle between
conditions (Figure 7). For overground propulsion, torso angle flexion (leaning forward) appears to
coincide with elbow flexion at the beginning of the push, followed by elbow extension and a constant or
very slight torso extension. However, this pairing of torso flexion with elbow flexion looks to be absent
from the ergometer propulsion where instead torso angle remains largely constant throughout the
push.
93

Figure 7: Comparison of angle-angle plots showing coordination of elbow angle and torso angle over the push cycle during
overground (green) and ergometer (purple) wheelchair propulsion. Angle-angle plot shown during phase of torque application
on pushrim when moment about the wheel axle was greater than 5 Nm.
Discussion
Individualized fitting of manual WC has been identified as a potential method to mitigate the onset of
chronic pain
3
. Currently, WC fitting is iterative and largely depends on the expertise and experience of
the clinician. A significant part of getting the fit of a WC correct for an individual is accounting for that
specific users capacity to maintain balance in their WC. Understanding the implications of WC stability
on propulsion technique and posture is important for achieving a better interaction between WC and
user. This study shows the potential shifts in postural control as well as joint kinetics for differing WC
stability scenarios. These preliminary results highlight the need for further investigation of the influence
of WC stability on WC propulsion mechanics.
We collected data for this pilot study on a single subject and had him propel in matched WC
configurations on a stationary ergometer and outdoors overground in a realistic context. Rolling
resistance of the ergometer was set to match level ground over a tiled surface. Mean WC velocity over
the cycles used for comparison stayed within 0.7 m/s between conditions with overground push cycles
averaging 1.9 m/s and ergometer push cycles averaging 2.6 m/s. Based on results from Rodgers et al.,
2000 study comparing propulsion characteristics of WC users with a torso flexion propulsion technique
with those with a non-torso flexion technique, we correctly hypothesized a decrease in RF impulse and
push duration during ergometer propulsion compared to overground propulsion. Although, this result
differs from some of the previous WC collection comparison studies. Koontz et al., 2012 similarly
compared 24 experienced manual WC users at self-selected speed between a dynamometer (2 roller
system) and overground propulsion on a tile floor and found larger peak force on the dynamometer
94
compared with overground. However, Koontz posits that the observed shift was likely the result of
larger rolling friction in the dynamometer setup, which was not matched to the rolling resistance of the
overground surface or the subject’s inertial characteristics. Mason et al., 2014 also compared manual
WC propulsion characteristics of 15 able-bodied participants between ergometer and overground
propulsion on a wooden sprung floor and found larger mean resultant force, peak resultant force, and
push time during ergometer propulsion. However, similar to Koontz et al., they also reported measuring
higher rolling resistance in the ergometer setup than overground.
As proposed in the experimental design of the study, the front and rear of the WC were secured to the
testing apparatus in the ergometer setup to ensure that the main wheel axle stayed centered over the
roller and eliminate the risk of tipping over. However, in the realistic overground context the WC was
left unconstrained with the front wheels free to lift off the ground. When the WC is free to tip, CM of
the system must stay within the base of support or risk tipping backwards. Since this is not a desired
outcome during propulsion it was hypothesized that the torso would move forward to maintain CM
within the base of support in the overground propulsion scenario. When the WC is fixed on the
ergometer this torso action would not be necessary. As expected, overground propulsion showed
movement of the torso angle throughout the cycle as well as a torso angle range below that of
ergometer propulsion Figure 4a. On average torso angle was 79 (±2.0) degrees during overground
propulsion at the start of elbow extension and 86 (±0.9) degrees during ergometer propulsion. The
maximum angle of excursion during a single cycle of overground propulsion was 4 degrees while
ergometer was just 2 degrees. Across all the overground trials torso angle ranged between 75 and 84
degrees. Whereas, during ergometer propulsion it was between 84 and 88 degrees. During propulsion
on the ergometer, the user could safely lean back into the seatback and even push against it, without
risk of tipping over. Leaning against the seatback could also additionally explain the much smaller range
of motion seen in ergometer propulsion. Torso angle was only seen to move a maximum of 2 degrees
over the push cycle and across all ergometer cycles appears much more consistent, which makes sense
when torso is resting against something.
We anticipated differences in WC propulsion techniques to accommodate the different stability
requirements between the overground and ergometer testing scenarios. Further analysis still needs to
be done, however these findings begin to suggest that stability of the WC may play a role in torso
posture. Future studies should continue this investigation on a larger population and potentially
incorporate a pressure map on the seatback to quantify the different levels to which the torso is leaning
against the surface during propulsion.

95
References
1. Dalyan M, Cardenas DD, Gerard B. Upper extremity pain after spinal cord injury. Spinal Cord.
1999;37(3):191-195.
2. Bayley JC, Cochran TP, Sledge CB. The weight-bearing shoulder. The impingement syndrome in
paraplegics. JBJS. 1987;69(5).
3. Medicine PV of AC for SC. Preservation of Upper Limb Function Following Spinal Cord Injury: A
Clinical Practice Guideline for Health-Care Professionals. J Spinal Cord Med. 2005;28(5):434-470.
4. Desroches G, Aissaoui R, Bourbonnais D. Effect of system tilt and seat-to-backrest angles on load
sustained by shoulder during wheelchair propulsion. J Rehabil Res Dev. 2006;43(7):871.
doi:10.1682/JRRD.2005.12.0178.
5. Mulroy SJ, Newsam CJ, Gutierrez D, et al. Effect of Fore-Aft Seat Position on Shoulder Demands
During Wheelchair Propulsion: Part 1. A Kinetic Analysis. J Spinal Cord Med. 2005;28(3):214-221.
doi:10.1080/10790268.2005.11753815.
6. Hughes CJ, Weimar WH, Sheth PN, Brubaker CE. Biomechanics of wheelchair propulsion as a
function of seat position and user-to-chair interface. Arch Phys Med Rehabil. 2017;73(3):263-269.
doi:10.5555/uri:pii:0003999392900769.
7. Mâsse LC, Lamontagne M, O’Riain MD. Biomechanical analysis of wheelchair propulsion for
various seating positions. J Rehabil Res Dev. 1992;29(3):12—28.
8. Majaess GG, Lee Kirby R, Ackroyd-Stolarz SA, Charlebois PB. Influence of seat position on the
static and dynamic forward and rear stability of occupied wheelchairs. Arch Phys Med Rehabil.
2017;74(9):977-982. doi:10.5555/uri:pii:000399939390278I.
9. van der Woude L, Bouw A, van Wegen J, van As H, Veeger D, de Groot S. Seat height: Effects on
submaximal hand rim wheelchair performance during spinal cord injury rehabilitation. J Rehabil
Med. 2009;41(3):143-149. doi:10.2340/16501977-0296.
10. Kwarciak AM, Turner JT, Guo L, Richter WM. Comparing handrim biomechanics for treadmill and
overground wheelchair propulsion. Spinal Cord. 2010;49:457.
http://dx.doi.org/10.1038/sc.2010.149.
11. Koontz AM, Worobey LA, Rice IM, Collinger JL, Boninger ML. Comparison Between Overground
and Dynamometer Manual Wheelchair Propulsion. J Appl Biomech. 2012;28(4):412-419.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5450157/.
12. Mason B, Lenton J, Leicht C, Goosey-Tolfrey V. A physiological and biomechanical comparison of
over-ground, treadmill and ergometer wheelchair propulsion. J Sports Sci. 2014;32(1):78-91.
doi:10.1080/02640414.2013.807350.
13. Stephens CL, Engsberg JR. Comparison of overground and treadmill propulsion patterns of
manual wheelchair users with tetraplegia. Disabil Rehabil Assist Technol. 2010;5(6):420-427.
doi:10.3109/17483101003793420.
14. Rodgers MM, Keyser RE, Gardner ER, Russell PJ, Gorman PH. Influence of trunk flexion on
biomechanics of wheelchair propulsion - ProQuest. J Rehabil Res Dev. 2000;37(3):283-295.
96
15. Newsam CJ, Rao SS, Mulroy SJ, Gronley JK, Bontrager EL, Perry J. Three dimensional upper
extremity motion during manual wheelchair propulsion in men with different levels of spinal cord
injury. Gait Posture. 1999;10(3):223-232. doi:10.1016/S0966-6362(99)00034-X.
16. Julien MC, Morgan K, Stephens CL, Standeven J, Engsberg J. Trunk and neck kinematics during
overground manual wheelchair propulsion in persons with tetraplegia. Disabil Rehabil Assist
Technol. 2014;9(3):213-218. doi:10.3109/17483107.2013.775362.
17. Finley M a, Rasch EK, Keyser RE, Rodgers MM. The biomechanics of wheelchair propulsion in
individuals with and without upper-limb impairment. J Rehabil Res Dev. 2004;41(3B):385-395.
doi:10.1682/JRRD.2004.03.0385.
18. Mulroy SJ, Farrokhi S, Newsam CJ, Perry J. Effects of spinal cord injury level on the activity of
shoulder muscles during wheelchair propulsion: an electromyographic study. Arch Phys Med
Rehabil. 2004;85(6):925-934. doi:https://doi.org/10.1016/j.apmr.2003.08.090.
19. Santos Requejo P, Lee S, Mulroy S, et al. Shoulder Muscular Demand During Lever-Activated Vs
Pushrim Wheelchair Propulsion In Persons With Spinal Cord Injury. J Spinal Cord Med.
2008;31(5):568-577. doi:10.1080/10790268.2008.11754604.
20. Lighthall-Haubert L, Requejo PS, Mulroy SJ, et al. Comparison of Shoulder Muscle
Electromyographic Activity During Standard Manual Wheelchair and Push-Rim Activated Power
Assisted Wheelchair Propulsion in Persons With Complete Tetraplegia. Arch Phys Med Rehabil.
2009;90(11):1904-1915. doi:10.1016/j.apmr.2009.05.023.
21. Winter D. Biomechanics and Motor Control of Human Movement 2nd Edition.; 1999.
22. De Leva P. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J Biomech.
1996;29(9):1223-1230. doi:10.1016/0021-9290(95)00178-6.

97
Chapter 8: Summary of Studies

Manual wheelchairs (WC) provide an effective form of low-cost wheeled mobility by preserving upper
body strength, cardiovascular conditioning, independence, and participation in the community
especially among individuals with spinal cord injury. However, repetitive mechanical loading of the
upper extremity is often associated with pain, dysfunction, and poor health-rated quality of life. Due to
the detrimental impact on functional mobility and the difficulty in treatment of shoulder pain once it
occurs, understanding the control and dynamics of the upper extremity during manual WC propulsion
under different conditions and configurations is important for improving future patient outcomes.

Chapter 4: Modifications in wheelchair propulsion technique with speed
As part of daily living, manual WC users need to regulate WC propulsion speed. The purpose of this
study was to determine how individual manual wheelchair users with paraplegia modify propulsion
mechanics to accommodate expected increases in reaction forces generated at the pushrim with self-
selected increases in wheelchair propulsion speed. Upper extremity kinematics and pushrim reaction
forces were measured for 40 experienced manual wheelchair users with paraplegia while propelling on a
stationary ergometer at self-selected free and fast propulsion speeds. Upper extremity kinematics and
kinetics were compared within-subject between propulsion speeds. Increased propulsion speed was
accompanied by increases in Reaction Force (RF) magnitude (22 of 40, >10N) and shoulder Net Joint
Moment (NJM, 15 of 40, >10Nm) and decreases in pushrim contact duration. Within-subject comparison
indicated that 27% of participants modified their WC propulsion mechanics with increases in speed by
regulating RF orientation relative to the upper extremity segments. Reorientation of the RF relative to
the upper extremity segments can be used as an effective strategy for mitigating rotational demands
(NJM) imposed on the shoulder at increased propulsion speeds.

Chapter 5: 4-Axis Parameterization of the shoulder net joint moment
Due to the nature of the shoulder joint, upper arm motion is often complex and multi-planar. With no
standard method for reporting shoulder joint moments and a variety of coordinate systems to choose
from, interpretation of results across studies can be difficult. The aim of this study was to use a novel
method of parsing the shoulder NJM into four axes to better understand the mechanical demand
imposed on the shoulder during manual WC propulsion. Upper extremity kinematics and pushrim
reaction forces were analyzed for 3 example participants from a larger data set of (130+) experienced
manual wheelchair users with paraplegia while propelling on a stationary ergometer at self-selected fast
propulsion speed. Applying this new method of parsing the shoulder NJM to the task of manual WC
propulsion revealed multiple strategies in terms of axis distribution. A mixture of single axis dominant
techniques were observed, with NJM about the flexion-extension axis being the most common. Other
subjects employed an even split between two or three axes, where the NJM vector aligns between
anatomical axes. Differences in NJM distribution arise from different orientations of upper arm relative
to the torso. When the upper arm is abducted away from the torso the expected NJM distribution
includes more effort in the Horizontal Adduction direction. When RF orientation is maintained in line
with the arm-plane very little axial torque about the upper arm is created and therefore little NJM about
98
the external-internal axis is seen. Conversely, when RF is oriented out of the arm-plane NJM about the
external-internal axis will appear. Including an isolated external-internal NJM component when
describing shoulder kinetics illustrates how multiplanar loading of the upper extremity can be
understood through a four-axis anatomical component parameterization. With more clearly delineated
anatomical axes, subject loading technique characterization becomes more robust and eliminates
scenarios where the amount external-internal NJM may otherwise be obscured and attributed to torque
about one of the three body fixed torso axes.

Chapter 6: Modifications in wheelchair propulsion technique following clinical wheelchair adjustment
Individualized fitting of the manual WC to the user has been highlighted as a promising means to
improve the interaction and potentially mitigate loading on the shoulder. In this study, we used a within-
subject experimental design to determine how individual manual WC users with paraplegia modify WC
propulsion mechanics following a clinical WC fitting visit focusing on addressing posture, balance, and
pressure. Upper extremity kinematics and pushrim reaction forces were measured for twenty two
participants (21 Male, 1 Female) with paraplegia while propelling in a realistic overground setting
outdoors at self-selected fast propulsion speed. Torso posture and upper extremity kinetics were
compared within-subject between baseline configuration and 30 days following new configuration.
Twenty out of 22 participants had some adjustment made to the seatback with the most common
adjustment being made to the seatback angle, which was often repositioned to be as close to
perpendicular to the ground as possible while still allowing for the participant feel comfortable. Six
participants had specific complaints concerning posture or WC stability in their baseline configuration
and 5 out of those 6 shoed shifts towards a more upright torso angle throughout the push cycle. Since
shifts towards upright posture occurred for participants with complaints of an unstable WC, there may
be a connection between stability of the WC and torso posture during propulsion. Exploring the
corresponding effect of current clinical fitting practice on a subject’s propulsion technique helps to
reveal the dynamic and joint kinetic outcomes for WC adjustment decisions.

Chapter 7: Modifications in wheelchair propulsion technique between ergometer and overground
wheelchair propulsion
Following analysis of modifications in propulsion technique pre and post WC fitting, (Chapter 6) it was
hypothesized that there may be a potential connection between torso posture and WC stability, since
shifts towards upright posture occurred for participants with complaints of an unstable WC. In this pilot
study, a manual WC user with paraplegia propelled on a stationary fixed ergometer apparatus and
outside in a realistic setting to determine the differences between WC propulsion mechanics in varying
WC stability scenarios. Upper extremity kinematics and pushrim reaction forces were measured for one
participant with paraplegia (T12) who was an experienced manual WC user while propelling in a realistic
overground setting as well as on a stationary ergometer at self-selected fast propulsion speed. Torso
posture and upper extremity kinetics were compared between collection scenarios. Torso angle was 79
(±2.0) degrees during overground propulsion at the start of elbow extension and 86 (±0.9) degrees
during ergometer propulsion. Maximum torso angle range over a single propulsion cycle decreased from
4 deg during outdoor propulsion to 2 degrees during ergometer propulsion. A significant part of getting
the fit of a WC correct for an individual is accounting for that specific users capacity to maintain balance
99
in their WC. Understanding the implications of WC stability on propulsion technique and posture is
important for achieving a better interaction between WC and user. This study shows the potential shifts
in postural control as well as joint kinetics for differing WC stability scenarios.

100
Chapter 9: Conclusions

This body of work aimed to determine how upper extremity control and dynamics during manual WC
propulsion is affected by changes in speed and seating configuration. By determining how individual WC
users accommodated expected increases in mechanical demands we can begin to identify effective
multijoint control strategies for achieving desired performance outcomes while mitigating detrimental
mechanical loading of the shoulder. The results of this study indicate that increases in reaction force
magnitudes associated with increases in WC propulsion speed do not necessarily translate into
comparable increases in shoulder net joint moments (NJMs). The magnitude of the shoulder NJM
depends on the proximal distal moments created by the net joint forces about the center of mass of the
forearm and upper arm segments as well as the adjacent joint NJM at the elbow. This knowledge of self-
selected load mitigation strategies may prove fruitful in guiding clinical decisions that aim to identify
strategies for preserving shoulder function.

A detailed understanding of upper extremity demand during propulsion is necessary for describing an
individual’s interaction with their WC and identifying how to maintain shoulder health. However, the
nature of the shoulder joint often lends to complex multi-planar movements that are difficult to fully
describe anatomically. The proposed method of parsing the NJM is designed to capture axis specific
mechanical demand and be applicable to any joint, any task, and any subject without the need for
modification. Including the fourth moving axis pointing distally along the longitudinal axis of the distal
segment in conjunction with the three fixed axes of the adjacent proximal segment when parsing the
NJM, removes the need for investigators to choose a specific 3-axis coordinate system for their research
study. With 4 axes, all directions of joint effort are accounted for so the demand during any task at any
joint can be reported accurately and completely while using clinically intuitive terms. Applying this new
method of parsing the shoulder NJM to the task of manual WC propulsion revealed multiple strategies in
terms of axis distribution. A mixture of single axis dominant techniques were observed, with NJM about
the Flexion-Extension axis being the most common. Other subjects employed an even split between two
or three axes, where the NJM vector aligns between anatomical axes. In the case of WC propulsion,
differences in NJM distribution arise from different orientations of upper arm relative to the torso. Using
this method of reporting NJM we get a detailed picture of the net mechanical load distribution on any
joint described about muscularly relevant axes. While this work primarily focused on shoulder NJM
during manual wheelchair propulsion, this method can and should be applied to other joints and tasks.

Current clinical WC adjustment focuses on addressing posture, alleviating pressure areas, and improving
balance. By comparing how individual WC users modify technique during outdoor WC propulsion
following reconfiguration, we explored the effect of WC configuration on control and dynamics in a real-
world setting. Changes in propulsion technique varied across the population and included shifts in
postural kinematics as well as joint kinetics. Four out of the 6 participants with posture or balance
complaints displayed shifts towards more upright torso angle at the start of elbow extension as well as
higher shoulder position relative to the wheel axle at the same instant. The shift in torso angle for these
4 participants was additionally seen to appear throughout their push cycles. This subgroup included two
101
participants that had adjustments for feeling “too tippy”, which may suggest a link between hunched
posture and an unstable WC configuration.

Finally, by analyzing propulsion mechanics on a stationary fixed ergometer apparatus and outside in a
realistic setting we began to investigate the potential shifts in postural control as well as joint kinetics
for differing WC stability scenarios. A significant part of getting the fit of a WC correct for an individual is
accounting for that specific user’s capacity to maintain balance in their WC. Understanding the
implications of WC stability on propulsion technique and posture is important for achieving a better
interaction between WC and user. The preliminary results of this study highlight the need for further
investigation of the influence of WC stability on WC propulsion mechanics. Further analysis still needs to
be done, however these findings begin to suggest that stability of the WC may play a role in torso
posture.

Future studies should continue the investigation of stability on posture during propulsion through a
comparison of overground and ergometer collection scenarios and potentially incorporate a pressure
map on the seatback to quantify the levels to which the torso is leaning against the surface. Depending
on pelvic posture and flexibility, propulsion techniques that involve a large degree of torso flexion may
be pushing the limits of muscle lengths in the lower back. Therefore, it might be valuable for studies to
look at posture during propulsion and lower backpain among WC users. Additional methods of
modifying technique such as augmented video feedback given to the users would also be an area of for
future research.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Control and dynamics of turning tasks with different rotation and translation requirements

PDF

Using nonlinear feedback control to model human landing mechanics

PDF

Modeling human regulation of momentum while interacting with the environment

PDF

Lower extremity control and dynamics during the golf swing

PDF

Using biomechanics to understand how an individual accomplishes a task in a variety of contexts

PDF

Regulation of linear and angular impulse generation: implications for athletic performance

PDF

Trunk control during dynamic balance: effects of cognitive dual-task interference and a history of recurrent low back pain

Asset Metadata

Creator Russell, Ian Miles (author)

Core Title Upper extremity control and dynamics during manual wheelchair propulsion at different speeds and wheelchair configurations

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Publication Date 08/01/2018

Defense Date 06/13/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biomechanics,ergometer,fitting,joint kinetics,kinematics,manual wheelchair,mechanical loading,net joint moment,OAI-PMH Harvest,posture,propulsion,shoulder,spinal cord injury,upper extremity,wheelchair configuration,wheelchair propulsion

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor McNitt-Gray, Jill (committee chair), D'Argenio, David (committee member), Wilcox, Rand (committee member)

Creator Email ian.m.russell@gmail.com,irussell@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-44014

Unique identifier UC11668653

Identifier etd-RussellIan-6586.pdf (filename),usctheses-c89-44014 (legacy record id)

Legacy Identifier etd-RussellIan-6586.pdf

Dmrecord 44014

Document Type Dissertation

Format application/pdf (imt)

Rights Russell, Ian Miles

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA