Stuart Andrew Crampton, BPO(Hons)

Supervisors:
Dr. T. M. Bach, National Centre for Prosthetics and Orthotics, La Trobe University, and
Mr. A. Ward, School of Human Biosciences, La Trobe University

The aim of this study was to examine the effects of different knee joint control conditions on energy expenditure during walking. An experimental Knee-Ankle-Foot orthosis (KAFO) was developed which incorporated a pneumatic brake and an electronic control system. Knee locking was controlled by Force Sensing Resistors (FSRs) located in the footwear. Electronic controls allowed four test conditions: (1) locking at ipsilateral heel contact and unlocking at contralateral heel contact (HCUL), (2) locking at ipsilateral heel contact and unlocking at ipsilateral toe off (TOUL), (3) always locked in full extension (Fixed) and (4) unlocked at all times (Free). By allowing free flexion during swing, it was assumed that substantial decreases in energy expenditure would be achieved compared to the fixed condition. It was hypothesised that the HCUL condition would further improve walking economy by allowing pre-swing flexion of the knee.

One able-bodied subject was tested during walking on a treadmill. Gait analysis was performed using a VICON 3-D motion analysis system (Oxford Metrics) with four infra-red cameras. Energy expenditure was measured using a Morgan metabolic cart. Four five minute trials at 67 m/min for each brace condition (Free, Fixed, HCUL, TOUL) were conducted. Trials were presented in a random order to minimise series effects. Trials without the brace were conducted prior to and following the braced conditions. A one-way ANOVA for repeated measures was used to test for effects on energy expenditure. Tukey's HSD test was used for post hoc comparisons.

The ANOVA indicated a significant effect for brace condition (F=16.99, df=3, p<0.001). Post hoc tests showed that the TOUL and fixed conditions were not significantly different from one another and that both were significantly higher than both HCUL and Free conditions.

No significant difference was found between the free knee condition and the HCUL condition. This suggests that stability during single limb support could be achieved without significant changes in energy expenditure during walking.

Examination of kinematic data provided some insights into the differences in energy expenditure. The TOUL condition allowed free knee flexion during swing but, unlike the HCUL condition, no phase of pre-swing knee flexion was present. Joint moments were not computed, however, it appeared that greater effort was required to achieve knee flexion over a shorter period in the gait cycle in the TOUL compared to the HCUL condition. Vaulting during swing phase appeared to be another compensatory mechanism which accounted for differences in effort between conditions.

This study supported the hypothesis that maintaining a free knee during swing phase can significantly reduce energy expenditure compared to a fixed knee configuration. The study also demonstrated that pre-swing knee flexion may be vital in achieving reduced energy expenditure.

The Effect of AFO Stiffness on Knee Stability During the Loading Phase of Gait

Priya Davis, BPO(Hons)

Supervisors:
Ms. M. Hodge and Dr. T. M. Bach,
National Centre for Prosthetics and Orthotics, La Trobe University

Previous investigations have shown relationships between knee stability, the amount of toe clearance gained during swing and the flexibility of ankle-foot-orthoses (AFOs). Other investigators have shown that AFOs set in dorsiflexion increase knee flexion moments during loading. AFOs incorporating a plantarflexion stop have produced increases in knee moments when compared to spring assisted AFOs. The purpose of this study was to quantify the relationship between AFO stiffness and knee extensor muscle moments during loading to better understand this aspect of AFO prescription.

Four male subjects with dorsiflexion deficits participated in the study. Two of these were later eliminated either because of data collection problems or inappropriate gait patterns. Each subject's affected lower limb was fitted with a test plastic/metal AFO. The stiffness of the AFO was systematically altered by varying the number and type of springs used in the mechanical ankle joints. Five AFO conditions were used. The testing procedure involved a two hour session during which kinematic and kinetic data were collected while subjects walked with each AFO condition. Springs used within each joint were measured before and after testing, to indicate changes in their characteristics.

A VICON 3D motion analysis system was used to record kinematic data. A Kistler 6-component force platform was used to measure ground reaction forces. Muscle moments were computed using a standard inverse dynamics approach. Three separate trials for each AFO condition were recorded. Data was averaged between trials and across subjects. The relationship between knee extension moment and stiffness was determined using linear and non-linear regression analysis.

An exponential function appeared to provide the best fit between peak knee extensor moments and ankle joint stiffness. The correlation for this regression was 0.81 (p<0.5, one-tailed).

Kinematic data was examined to determine any effects of AFO stiffness. Flexion angles during loading appeared to increase with stiffness for one of the subjects but not for the other. Dorsiflexion during swing did not appear to be affected by AFO stiffness.

Changes in spring length occurred with two hours of wear for 92% of the springs used. This indicates a tendency for springs used within mechanical ankle joints to breakdown over short periods of time.

Results from the investigation indicate that AFO stiffness has an effect on knee stability during the loading phase of gait. Changes AFO stiffness have the most pronounced effect at low stiffnesses. Further research with an increased sample size is warranted. Research into deformation and material fatigue in springs used in mechanical joints for AFOs should also be undertaken.

The Effect of Ankle Taping and Bracing on Ankle Stiffness and Energy Consumption During Running

Linda Dodds, BPO(Hons)

Supervisor:
Dr. S. A. Malcolm, School of Human Biosciences, La Trobe University, and
Dr. T. M. Bach, National Centre for Prosthetics and Orthotics, La Trobe University

Traditionally, taping has been the most commonly used method of ankle protection in sports. However, more recently, reusable prophylactic ankle orthoses (braces) have increased in popularity. The rationale underlying the use of ankle stabilisation devices is that by restricting motion at the extremes of range, the ligamentous structures of the ankle will be protected.

The purpose of this study was to examine the effect of ankle stabilisation using taping, a commonly used brace (Swede-O), and a newly developed brace (Grenace) on ankle stiffness at both mid and end range and to observe changes in heart rate, oxygen consumption and energy expenditure during running wearing ankle stabilisers. A shoe only condition served as control.

Eight healthy subjects participated in the study. The procedure involved four testing sessions which included shoes only (control), taping, Granace brace and Swede-O brace.

Ankle stiffness was measured during inversion/eversion and plantar/dorsiflexion after application of each device and following 20 minutes of exercise. Stiffness was measured using the KinCom isokinetic dynamometer by moving the ankle passively throughout range while recording force. Following initial stiffness measurements, subjects ran a figure-of-eight course for ten minutes. This was followed by a self-paced ten minute treadmill run during which heart rate and VO2 were recorded. Ankle stiffness was then remeasured. Subjects were asked to rate comfort of each device on a scale from one to five.

Ankle stiffness was found to be significantly increased at end of range in inversion/eversion following application of both braces when compared to both tape and control conditions. This increase in stiffness continued after 20 minutes of exercise. Following application of braces, ankle stiffness increased significantly at end of range in plantar/dorsiflexion in comparison to tape and control, however this difference was not apparent following exercise. No difference was observed between taping and control conditions in either movement at any time.

Following application, the Swede-O brace was found to significantly increase ankle stiffness during inversion/eversion and plantar/dorsiflexion in mid-range when compared to all other conditions. Although not significant (p=0.08), the trend for increased stiffness in mid-range was apparent post-exercise. Subjective rating of comfort from subjects indicated that the Granace brace was the preferred ankle stabiliser, followed by taping and finally the Swede-O It is possible that the increased stiffness in mid-range while wearing the Swede-O brace contributed to the low user acceptance.

VO2 and energy expenditure during running were not affected by either taping or bracing. Correlations between VO2 and ankle stiffness during plantar/dorsiflexion were not significant suggesting that increased stiffness does not result in increased energy expenditure during running.

Previous studies evaluating the effectiveness of ankle immobilisation have reported that bracing and tape restricted movement following application but lost their restriction following exercise. In the present study ankle stiffness near end of range was found to be greater following bracing than taping and this was not affected by exercise.

The Grenace brace was found to be superior to the other devices tested in that it increased at the end of range with no effect mid-range and was perceived as comfortable by users.

The Effect of Liner Type on Knee Extension Torque and Antero-Tibial Socket Pressures in Transtibial Amputees

Mark Graf, BPO(Hons)

Supervisor:
Dr. T. M. Bach, National Centre for Prosthetics and Orthotics, La Trobe University

In a repeated measures study involving three male amputees aged 26-58 years, the effect of liner material and resistance pad placement on peak knee extension torque and distal anterior tibial pressure were investigated. A KinCom dynamometer and pressure transducer were used to simultaneously measure torque and pressure respectively, during maximal effort knee extension. Subjects gave a perceived discomfort rating for each condition. Mean torque and mean pressure were correlated. Peak torque and average pressure for the different liner/placement conditions were compared in a two-way repeated measured ANOVA and perceived discomfort was correlated to peak pressure and liner/pad placement conditions.

The results of this study showed a strong linear relationship between knee extension torque and distal tibial pressure (silicon liner r=0.78 and "Pelite" liner r=0.84, p<0.05) Average peak torque was found to be 109.7 Nm for silicon and 94.4 Nm for "Pelite" liners with average pressures of 735 kPa and 576 kPa respectively.

A main effect on torque was detected for liner type, with silicon liners showing higher torques (p<0.05). No effects of liner type were detected for pressure. Pad placement showed no main effects or interactions for either torque or pressure. A high correlation (Spearman's rho = 0.73) between perceived discomfort and peak pressure was obtained.

The results were interpreted as indicating that distal tibia discomfort due to socket pressure was a limiting factor for knee extension torque and that higher torques were obtained for the silicon liner as it more effectively redistributed distal anterior tibial pressure making the socket more comfortable.

It was concluded that distal anterior tibial discomfort, as experienced by transtibial amputees during activities that require large knee extension torques, can inhibit torque generation by the prosthetic side knee and may be a factor in observed prosthetic side quadriceps disuse atrophy. The use of silicon liners allows greater torque generation and may therefore help reduce the extent of quadriceps atrophy and improve amputees' functional capacity.

The Effects of Foot Orthosis Modification on Pain and Plantar Pressure in People with Rheumatoid Arthritis

Margaret Cathrine Hodge, BPO(Hons)

Supervisors:
Dr. T. M. Bach, National Centre for Prosthetics and Orthotics, La Trobe University, and
Mr. G. M. Carter, Murdoch Medical Centre, Perth

The purpose of this study was to investigate the effect of foot orthosis design on foot pain, peak plantar pressures and gait characteristics in people with rheumatoid arthritis. The effects of four different foot orthosis designs were investigated. A semi-rigid, prefabricated foot orthosis and three variations of semi-rigid, custom made foot orthosis were compared to a shoe only condition.

Pain in the second metatarsal head region, in both standing and walking, was measured using visual analogue scales. Peak pressure was measured in five foot regions: the medial midfoot, lateral midfoot, first metatarsal head, second metatarsal head and 3rd-5th metatarsal heads. Cadence and stance phase duration were measured to indicate differences in temporal gait characteristics. An EMED Pedar system was used to measure pressure, cadence and stance phase duration.

Twelve subjects with a history of rheumatoid arthritis with foot involvement were recruited for the study. Feet were excluded from the main analyses if shod weight bearing produced no pain. Results from 20 feet (nine bilateral subjects and two subjects with unilateral involvement) were analysed in the main sample.

Significant reductions in walking and standing pain levels were associated with the custom made foot orthosis with a metatarsal dome (CFOD) and a reduction in walking pain was associated with the standard custom made foot orthosis (CFOS). A highly significant positive correlation was found between walking pain and standing pain and no differences were found between walking pain levels and standing pain levels in the different conditions. No significant correlation was found between either walking pain or standing pain and second metatarsal head peak pressure.

Significant reductions in peak pressures in at least one metatarsal head region were found in all orthosis conditions. The custom made foot orthosis with a metatarsal bar (CFOB), CFOD and the prefabricated foot orthosis (PFO) reduced peak pressure significantly in all metatarsal head regions. No differences in peak pressure between these orthoses were identified. The CFOS reduced peak pressure in the region of the first metatarsal head only.

Neither cadence nor stance phase duration in the four orthoses was found to differ significantly from the control.

A further analysis was made to compare symptomatic and asymptomatic feet. Peak pressure effects in the two sub-groups were not different from the effects in the main sample. There were differences in the orthosis preferred by subjects in the two sub-groups. The CFOD was the orthosis of preference for the main sample and also for the symptomatic sub-group. Asymptomatic subjects preferred the CFOB.

Collectively, the results of this project suggest that the CFOD was most effective in treating people with rheumatoid arthritis who exhibited metatarsalgia. The CFOD produced reductions in metatarsal head peak pressures, relieved walking and standing pain and was the orthosis preferred by subjects. Although the CFOB and the PFO were associated with reduced metatarsal peak pressures, neither reduced levels of pain. The CFOS reduced walking pain but had minimal effects on pressure reduction.

The pressure redistributions and pain reductions achieved in this project suggest that orthotic management of the foot in RA is a useful treatment option. The project demonstrated that a custom made, semi-rigid foot orthosis with an appropriately sized metatarsal dome was effective in reducing metatarsal head peak pressure and pain in the presence of rheumatoid arthritis.

The Functional Ability of Simulated Upper Extremity Dominant Limb Amputees in Fine Motor Tasks

Yvette Marchant, BPO(Hons)

Supervisor:
Dr. D. K. Rogers, School of Human Biosciences, La Trobe University

There has been controversy surrounding the origins of dominance patterns and handedness for many years. Theorists differ on the extent to which heredity and environment play a role. What is generally accepted is that it exists and is fixed early in life. So what happens when the dominant limb is amputated? Literature advocates the practice of transferring dominance patterns to the intact limb however there have been no studies to evaluate the benefits or otherwise of this practice.

This investigation hypothesised that retaining dominance patterns on the prosthetic dominant side would result in improved handwriting skills and may reduce amputee frustration levels and hence training times. An aim of the investigation was to compare handwriting characteristics of subjects using a prosthesis on their dominant limb with those of the intact non-dominant limb.

Eight subjects recruited from the student population of the Prosthetics and Orthotics course at La Trobe University (mean age 24.5 years; 5 females, 3 males) completed a hand preference questionnaire to determine which limb would utilise a transradial training prosthesis (dominant limb). Selection criteria excluded subjects if they had any upper limb pathology, or if they had been forced to use their non-dominant hand (NDH) for writing for any length of time.

Eight one-hour training sessions conducted in pairs were completed by each subject over a four week period (two hours per week). Training began with an orientation to the prosthesis and continued with handwriting practice completed by both the prosthesis and NDH. Subjects were tested using a WACOM digitiser pad connected to a pentium computer. OASIS software was used to record and analyse subjects' writing skills (a familiar word - signature, an unfamiliar word -'gob' and a geometric shape - square within a circle template). Data was analysed using two-way ANOVAs for repeated measures and appropriate post hoc comparisons were made.

Each writing skill tested was analysed for three parameters - duration, control index and spatial characteristic. Results for duration showed similar patterns for all skills tested. NDH was significantly slower than prosthesis and normal was significantly faster than both other hand conditions. This result probably reflects the fact that speed comes with practice, of which 8 hours is insufficient. Prosthetic control of writing was found to be no different from normal control, whereas NDH control was poor for all test elements. Use of the prosthesis was felt subjectively to be automatic and this was reflected in the objective testing. The 'push' vs 'pull' of the pen when using the NDH resulted in spatial characteristics that were significantly condensed horizontally when compared with prosthetic and normal dominant hand conditions.

There was a lack of a generalised learning curve for either prosthetic or NDH conditions which leads to the conclusion that improvement in writing quality takes more time than was available for this study.

Handwriting of a very similar quality to normal can be achieved using a transradial prosthesis in a very short period of time. Training the dominant limb amputee to retain dominance patterns on the prosthetic side reduces frustration levels and may reduce training time.

A Biomechanical Analysis of Able-Bodied and Transtibial Amputees During Cycling

Anna T. Murphy, BHS(Hons)

Supervisors:
Dr. T. M. Bach, National Centre for Prosthetics and Orthotics, La Trobe University, and
Dr. B. McLean, Australian Institute of Sport, Canberra

This study investigated the differences in kinematics, joint moments and powers developed in the lower limbs of able-bodied (AB) and transtibial (TT) amputee cyclists to identify differences in technique and compensations for lost power generating joints in the amputee group. Four male AB non-elite athletes and three male and one female TT amputees (one recreational cyclist, three sub-elite athletes) participated. Normal and shear forces on the bicycle pedals and kinematic data were collected for five seconds at the 10th, 11th and 12 minutes as each subject pedalled at 200W and 90 RPM on their own bicycle or a bicycle setup similar to their own, which was fixed to a road emulator.

Complete pedal cycles over the five second period for each trial were analysed. Joint kinematic patterns were determined for qualitative analysis, peak extensor moments and the percentage contribution of joint power to crank power at the hip, knee and ankle were determined for quantitative analysis and used to statistically compare the groups. The mean over the three trials for the TT group and ±1 SD for the AB group were calculated for both the right and left sides. Subject data in each group were averaged.

The TT group displayed substantially altered and asymmetrical joint moment and power profiles compared to the AB group. Hip contribution to total crank power was significantly different (p<0.05) between sides in the TT group and also significantly different (p<0.05) to the AB group. When compared with the AB group, knee extensor moments and contribution to total crank power on the sound and prosthetic sides were significantly reduced (p<0.05). In the TT group, ankle joint contribution to total power was significantly reduced (p<0.05) on the prosthetic side and for the TT group ankle joint contribution to total power was significantly lower compared to the AB group. The sound side dominated crank power in the TT amputee group with the sound hip making the largest contribution to crank power. These disparities in the contribution of each joint to total crank power resulted in a significant (P<0.05) difference between groups and between the left and right sides. This interaction effect was a direct result of the different power contributions make by the sound and prosthetic sides in the TT group.

The attempt to cycle symmetrically by the subjects in the TT amputee group was evident by the similar changes made in the development of moments and powers in the sound side and the prosthetic side when compared to the AB group. However, the contribution made by each joint to the development of total crank power was asymmetrical. Less power was generated at the knees than was absorbed in the TT amputee group resulting in net power absorption at this joint on both sides. This was compensated by the dominant contribution of the hip extensors.

The importance of the hip in the development of crank power in TT amputee cycling has been highlighted by this study and suggests the need to optimise bicycle setup to utilise the hip efficiently. Optimum seat height, the height of the pedal platform and the seat tube angle which assists in defining pelvic orientation appear important in the design of the bicycle setup for the TT amputee. Componentry designs may also need to be altered to enhance the application of the force to the pedal.

Material Fatigue in the Prosthetic SACH Foot: Effects on Mechanical Characteristics and Gait

Jody van Rooyen, BPO(Hons)

Supervisors: Mr. D. Orr, School of Human Biosciences, La Trobe University, and Mr. B. Contoyannis, Rehab TECH, Monash University

The effects of material fatigue on the prosthetic foot appear to have received little attention. The purpose of this study was to examine fatigue effects in terms of their influence on amputee gait. It was hypothesised that the mechanical characteristics of the prosthetic SACH feet would be compromised after fatigue loading and that the artificial 'aging' of these feet would influence biomechanical characteristics of gait.

Kingsley SACH type K051 and Otto Bock SACH type 1S49, right side, size 26 were tested. Three specimens of each were fatigued by a cyclic tester which approximated forces on the foot during the gait cycle. Static compliance of the heel and forefoot was measured and impact tests were performed to evaluate shock absorption capacity of the prosthetic feet. Four transtibial amputees were recruited and each was fitted with new and fatigued prosthetic feet of both types. Temporal data was collected using a Clinical Stride Analyser and ground reaction force data was obtained using a Kistler force platform. Two-way (foot type x foot age) repeated measured ANOVAs were used to analyse the data.

The Kingsley prosthetic feet were found to undergo premature delamination and failure without the recommended interface plate between the foot and the pyramid adapter. Further tests incorporating an interface plate between solved this problem.

Drop test results indicated that Kingsley feet provided significantly better shock absorption and that shock absorption increased with a minimum amount of fatigue. The Otto Bock feet were significantly stiffer in the forefoot region. Both brands of feet followed a similar trend of an initial increase in stiffness after 5,000 cycles, then a decrease after 10,000 cycles. The mean duration of the heel loading phase of the 'new' Kingsley test sample was significantly longer then the fatigued test sample. It was also demonstrated that as the stiffness of the rearfoot increases, so does the duration of the loading phase (r=0.952).

The Kinglsey feet appeared to be more susceptible to changed mechanical characteristics with ageing. Mechanical characteristics of feet influence gait patterns in amputees.