Isometric hip and knee torque measurements as an outcome measure in robot assisted gait training

TLDR: The peak torque recorded during isometric contractions provided an objective outcome measure to record changes in muscle strength following RAGT.

Abstract: BACKGROUND: Strength changes in lower limb muscles following robot assisted gait training (RAGT) in subjects with incomplete spinal cord injury (ISCI) has not been quantified using objective outcome measures. OBJECTIVE: To record changes in the force generating capacity of lower limb muscles (recorded as peak voluntary isometric torque at the knee and hip), before, during and after RAGT in both acute and subacute/chronic ISCI subjects using a repeated measures study design. METHODS: Eighteen subjects with ISCI participated in this study (Age range: 26–63 years mean age = 49.3 ± 11 years). Each subject participated in the study for a total period of eight weeks, including 6 weeks of RAGT using the Lokomat system (Hocoma AG, Switzerland). Peak torques were recorded in hip flexors, extensors, knee flexors and extensors using torque sensors that are incorporated within the Lokomat. RESULTS: All the tested lower limb muscle groups showed statistically significant (p < 0.001) increases in peak torques in the acute subjects. Comparison between the change in peak torque generated by a muscle and its motor score over time showed a non-linear relationship. CONCLUSIONS: The peak torque recorded during isometric contractions provided an objective outcome measure to record changes in muscle strength following RAGT.

Figures

Figure 6a and 6 b: Peak Torque generatd as a result of isometric contraction of the hip flexor and knee extensor muscles, compared against the Lower limb motor score (ASIA Score), for each of the respective muscle in subacute/chronic subjects. Intial measures are denoted by open circles and final measures are denoted by filled circles.

Figure 3a and 3b: Normalised peak torques presented as pooled data for chronic and acute subjects.

Table 2: Change in normalised peak torque compared to week 1. All statistically significant differences (p<0.05) are highlighted in grey cells.

Figure 4a and 4b: Normalised peak torques presented as pooled data for chronic and acute subjects.

Figure 3a and 3b: Normalised peak torques presented as pooled data for chronic and acute subjects.

Table 1 : Details of Subjects who had participated in the study. Subjects who were <6months post injury at the time of recruitment were classified as acute and the rest were classified as Chronic. M=Male and F=Female.

Full text

1
Isometric Hip and Knee Torque measurements as an outcome
measure in Robot Assisted Gait Training
Sujay S Galen
1,2,3
Celia J Clarke
1,2
, Alan N Mclean
2
, David B Allan
2
, Bernard A Conway
1,2
1 Bioengineering Unit, University of Strathclyde, Glasgow, UK.
2 Scottish Centre for Innovation in Spinal Cord Injury (SCISCI), Queen Elizabeth National Spinal
Injuries Unit, Southern General Hospital, Glasgow, UK.
3 Physical Therapy Program, Eugene Applebaum College of Pharmacy and Health Care Sciences,
Wayne State University, Detroit, MI, USA.
*Correspondence to first author at
Physical Therapy Program
Wayne State University
259 Mack Avenue
Detroit, MI 48201.
USA.
Tel: +1 (313) 577 5531
Fax: +1 (313) 577 8685
e-mail: sujay.galen@wayne.edu
Keywords: Incomplete spinal cord injury, Lokomat, Peak torque, Dynamometry, Muscle
strength, Body weight supported treadmill training.

2
Abstract
Background: Strength changes in lower limb muscles following robot assisted gait training
(RAGT) in subjects with incomplete spinal cord injury (ISCI) has not been quantified using
objective outcome measures.
Objective: To record changes in the force generating capacity of lower limb muscles
(recorded as peak voluntary isometric torque at the knee and hip), before, during and after
RAGT in both acute and subacute/chronic ISCI subjects using a repeated measures study
design.
Methods: Eighteen subjects with ISCI participated in this study (Age range: 26-63 years
mean age = 49.3 + 11 years). Each subject participated in the study for a total period of
eight weeks, including 6 weeks of RAGT using the Lokomat system (Hocoma AG,
Switzerland). Peak torques were recorded in hip flexors, extensors, knee flexors and
extensors using torque sensors that are incorporated within the Lokomat.
Results: All the tested lower limb muscle groups showed statistically significant (p<0.001)
increases in peak torques in the acute subjects. Comparison between the change in peak
torque generated by a muscle and its motor score over time showed a non-linear relationship.
Conclusions: The peak torque recorded during isometric contractions provided an objective
outcome measure to record changes in muscle strength following RAGT.

3
Introduction
The primary sensory and motor consequences of spinal cord injury (SCI) are well known, and
vary according to the level and extent of lesion. .However, with paralysis comes a secondary
complication of muscle atrophy. Following injury, the onset of muscle atrophy can be rapid
and is evident soon after injury(Dudley-Javoroski & Shields, 2008). The loss of muscle cross
sectional area in SCI subjects has been reported to be between 18 and 46% on average when
compared with neurologically intact individuals (Giangregorio & McCartney, 2006). This
degree of muscle atrophy and the related loss of force generating capacity can contribute to
functional impairment and can compromise progression of rehabilitation in SCI subjects.
Accordingly, methods for early mobilization may mitigate against atrophy by either reversing
or reducing the rate of muscle loss. In ISCI subjects receiving intensive gait rehabilitation,
this is likely to be a factor that facilitates the functional recovery of gait (Dietz, Colombo,
Jensen, & Baumgartner, 1995).
Bodyweight supported treadmill training (BWSTT) has been shown to improve walking
ability (Nymark et al., 1998; Postans, Hasler, Granat, & Maxwell, 2004) in ISCI subjects and
is increasingly being adopted within physical therapy programs (Hicks & Martin Ginis,
2008). For acute subjects BWSTT also serves as a method for early lower limb mobilization
and exercise.
In recent years the use of robotic assistance during BWSTT has emerged as an alternative to
therapist assisted(Adams et al., 2006; Hicks et al., 2005) or FES assisted BWSTT(Postans et
al., 2004) and may provide a more resource efficient method of delivering intensive and
progressive gait training (Hornby, Zemon, & Campbell, 2005). The Lokomat® (Hocoma
AG, Switzerland) is a robotic driven gait orthosis (DGO) that provides robot assisted gait
training (RAGT) using BWSTT. Although preliminary studies have shown that Lokomat

4
training improves gait in both acute (Hornby et al., 2005), and chronic ISCI subjects (Wirz et
al., 2005) there have been no studies documenting the associated changes in the force
generating capacity of the lower limb muscles, in subjects as they progress through RAGT
using BWSTT
Strength changes in lower limb muscles following locomotor training have been previously
documented in ISCI subjects using objective outcome measures such as peak torque recorded
using a Biodex dynamometer. (Jayaraman et al., 2008) Given that the Lokomat system is
equipped with integrated torque sensors within the DGO exoskeleton, the device has the
capability to be used as a dynamometer in its own right. Recently it has been shown that the
peak torques, recorded using the torque sensors integrated within the DGO exoskeleton,
provides data with good inter and intra-rater reliability in both healthy subjects and subjects
with neuromuscular disorder.(Bolliger, Banz, Dietz, & Lunenburger, 2008) This inbuilt
capability provides the opportunity for monitoring the peak torques generated around hip and
knee during voluntary isometric exertions, and has a potential use as a clinical measure to
monitor the subject’s progress through the rehabilitation programme (Bolliger et al., 2008;
Lunenburger, Colombo, Riener, & Dietz, 2004). In this study we set out to record the change
in voluntary force generating capacity in the hip and knee flexors and extensor muscles,
before, during and at the end of RAGT in both acute and subacute/chronic ISCI subjects. The
quantitative force data is then compared with the equivalent manual muscle scores acquired
from subjects as part of the standard neurological classification of spinal cord injury. (Marino
et al., 2003)
Methodology
Eighteen subjects with ISCI participated in this study (Age range: 26-63 years mean age =
49.3 + 11 years) of which five were classified as sub acute/chronic subjects (> 6 months from

5
time of injury) and thirteen were classified as acute subjects (<6 months from the time of
injury). Subjects were recruited if they were
 Acute or Chronic ISCI subjects graded as AIS classification C/D.(Marino et al., 2003)
 Between the ages of 18-65 years.
 Assessed as medically stable by an (independent) physician.
Subjects were not recruited if they had
 Complications requiring immobilization such as fractures and pressure sores.
 Osteoporosis or contractures limiting range of motion.
 A bodyweight of over 130 kg or thigh length over 47 cms (Lokomat restrictions).
Ethics approval was obtained from the National Health Service (NHS) regional ethics
committee and subjects were recruited from the Queen Elizabeth National Spinal Injuries
Unit, Glasgow. The details of the subjects who participated in the study are provided in
Table1.
Insert Table 1 here.
Informed and signed consent was obtained from all subjects prior to the start of the study.
Each subject participated in the study for a total period of 8 weeks. A repeated measures
study design was used. A battery of neuro-physiological and functional outcome measures
were used to record the subject’s baseline and final measures during week1 and week8 of the
study respectively. RAGT was provided using the Lokomat for an hour each week day
(Monday to Friday) from Week 2 to Week 7 (a total of 6 weeks) for each subject. For more
details on the outcome measures used in this study , refer to Ellaway et al 2010. (Ellaway et

Frequently asked questions

Q1. What have the authors contributed in "Isometric hip and knee torque measurements as an outcome measure in robot assisted gait training" ?

The Lokomat this paper is a robotic driven gait orthosis ( DGO ) that provides robot assisted gait training ( RAGT ) using BWSTT. 

Q2. What are the future works in "Isometric hip and knee torque measurements as an outcome measure in robot assisted gait training" ?

A further option is that in the chronic subjects the threshold level of activity to induce new muscle hypertrophy is elevated compared with that in acute subjects ( i. e. loss of muscle plasticity ) and further research will be required to explain these findings. RAGT, in addition to having the potential to improve the force generating capacity of the lower limb muscles and gait function in responding ISCI subjects has also been reported to have an exercise impact on cardiovascular fitness and so contributes to improvements in the overall health and wellbeing of subjects irrespective of time post injury. This suggests that when designing new pharmaceutical or regenerative treatments careful consideration should be given to the level and intensity of any concomitant rehabilitation programme. 

Q3. What is the effect of RAGT on the body?

in addition to having the potential to improve the force generating capacity of the lower limb muscles and gait function in responding ISCI subjects has also been reported to have an exercise impact on cardiovascular fitness and so contributes to improvements in the overall health and wellbeing of subjects irrespective of time post injury. 

Q4. What is the significance of the change in peak torques?

a change in peak torques generated by the lower limb muscles that exceeds the baseline measures, has the potential to be used as an indicator that helps the clinician to decide whether to continue training or to alter the rehabilitation strategy. 

Q5. What is the effect of the RAGT programme on the body?

For the acute subjects, the functional gains in gait and muscle power are likely to reflect the combination of the natural recovery process post injury (Basso, 2000) supplemented by the impact of the intensive RAGT programme (Dietz, 2008). 

Q6. What is the significance of manual muscle testing?

Manual muscle testing is an important component of the standard neurological classification of spinal cord injury (ASIA) scoring system. 

Q7. What was the peak torque generated in the hip flexors and extensors?

The peak torque generated as a result of isometric contraction of the muscle was successfully recorded in the hip flexors and extensors and knee flexors and extensors in all subjects. 

Q8. What was the test used to determine the difference in the generated peak torque?

If the Friedman’s test indicated that there were significant differences in the generated peak torques over the six weeks of training, a Wilcoxon signed rank test was used to identify the between week differences in the generated peak torque. 

Q9. What is the advantage of RAGT in a clinical trial?

Use of interventions such as RAGT may offer advantages in such trials as both the dose and intensity of the rehabilitation programme can be easily monitored and standardized and that the integrated dynamometry provides for simple longitudinal collection of a set of important outcome measures. 

Q10. What is the main reason for the loss of muscle?

This degree of muscle atrophy and the related loss of force generating capacity can contribute to functional impairment and can compromise progression of rehabilitation in SCI subjects. 

Q11. What is the significance of the observation that RAGT is used to identify the subjects who are?

The observation that acute subjects who positively responded to RAGT had significant increase in peak torques within the first 3 weeks of training, suggests that this type of quantitative monitoring could also be considered as a useful marker in identifying subjects who are positively responding to RAGT using BWSTT. 

Q12. Why did the subjects who participated in this study have a better ambulatory capacity?

This is likely to be due to the fact that these subacute/chronic subjects were ambulant; therefore a better ambulatory capacity (Figure 2) following ISCI seems to stabilize muscle strength. 

Q13. What is the difference between therapist assisted and BWSTT?

In recent years the use of robotic assistance during BWSTT has emerged as an alternative to therapist assisted(Adams et al., 2006; Hicks et al., 2005) or FES assisted BWSTT(Postans et al., 2004) and may provide a more resource efficient method of delivering intensive and progressive gait training (Hornby, Zemon, & Campbell, 2005).