Rhaïa April 2021

Rhaia banner_2020

The Rhaïa eBulletin is published for the Australasian Faculty of Rehabilitation Medicine (AFRM). The newsletter is primarily a communication for AFRM Fellows and trainees to provide up-to-date information for readers in the areas of education, practice, advocacy, research and AFRM affairs.

Editor's Introduction

Swinging into 2021 – with the year probably already well underway by the time of publication – I hope you have all remained safe and well this year and carrying on with new, exciting activities/projects as best possible. I would like to welcome all of you to our first Rhaïa eBulletin for 2021, which features our Fellows sharing their stimulating opinions, perspectives and experiences within the areas of upper limb prosthetics, spasticity management, and the use of technology in rehabilitation. 

Both Dr Greg Bowring and renowned Alfred Health (Melbourne) plastics surgeon, Mr Frank Bruscino-Raiola highlight more creative technological advances in the realm of upper limb prosthetics and the use of targeted muscle reinnervation (TMR). Our foremost spasticity experts: Associate Professor Barry Rawicki, Associate Professor Ian Baguley, Dr Yuriko Watanabe and Dr Warren Jennings-Bell draw our attention to key and current aspects of spasticity management. We have interesting and excellent articles by Associate Professor Ray Russo and Dr Timothy Scott discussing the increasing use of robots and adaptive technology to help facilitate recovery in rehabilitation patients. 

Once again, I would like to express my sincere thanks and appreciation towards our contributors for providing their time and efforts in writing these articles. I hope you will thoroughly enjoy reading this edition as much as I did.

Dr Krystal Song

AFRM President’s Report 

I would like to share observations about technology and rehabilitation, particularly in the field of upper limb prosthetics. Upper limb prosthetics is being revolutionised by technological advancement, offering a window into how technology might to be used in parallel fields in the future (e.g. spinal cord injury, brachial plexus injury etc.). We need to incorporate these technological innovations into the comprehensive rehabilitation we offer to patients with disabilities.

Most of you will remember the cable and harness body-powered prostheses that you learnt about in outpatient clinics or during upper limb amputee training courses. These designs have changed little since the late 19th century and, although passionate advocates argue that their function has barely been surpassed by the modern battery-powered myoelectric alternatives, the improved performance and cosmesis of myoelectric hands is winning the hearts of the amputee community; soon, they will exceed the function of the previous devices by a wide margin. Cost is the major drawback.
Myoelectric upper limb prostheses have been in use since the 1950s (invented in Germany in the 1940s). These two-site control, battery operated, single axis ‘hands’ improved gradually over the next 50 years. Powered wrist rotators were added and although powered elbows were also available, they were rarely used because control systems made the combined prosthesis too cognitively demanding for most amputees to tolerate.

The USA’s large number of injured veterans from 2002 onwards included thousands of amputees because body armour and rapid resuscitation techniques could often save a life despite massive injuries. The rehabilitation needs of these veterans drove the political will to fund major research and development in upper limb prosthetic technology to improve the outcomes. This was funded by the US Defence Department’s Defence Advanced Research Project Agency (DARPA) via its ’revolutionising prosthetics’ project – a multi-billion-dollar research effort directly targeted at improving upper limb prosthetics.

Multifunction hands, with a myriad of grip patterns, began appearing from 2007 first in the UK, then subsequently in the US, Germany, and most recently Aotearoa New Zealand. The oddly named “i-Limb” hand from Touch Bionics (not Apple!) in the UK was the first of the multifunction hands and its latest generation continues to lead the market. There are now at least half a dozen multifunction hands in the world market, though in Australia – only four have been used. Of these, the i-Limb (now owned by Ossur), the Bebionic (now owned by OttoBock) and the Aotearoa New Zealand designed Taska (the newest entrant) are the three most popular models, with the Michaelangelo (OttoBock) being less popular. 

In the USA, one of the early DARPA funded products was the so-called ‘Luke’ (as in Luke Skywalker) hand by the Deka Corporation, which is seen in US research and promotional videos. Another is called the MPL (modular prosthetic limb). Neither are available in Australia yet.

Ultimately, the improved hands needed better control systems to fully utilise the potential of their multiple group patterns. A major breakthrough came with Targeted Muscle Reinnervation (TMR) designed for trans-humeral amputees. TMR involved the identification of the now disconnected median, ulnar, and distal radial nerve branches and re-attaching them to slips of muscle (one head of biceps, one head of triceps, and brachialis if present). This enables the previous two-site control systems to be increased to four or even five separate sites providing intuitive control of hand function in a transhumeral amputee rather than relying on the previous two-site control which utilised musculocutaneous innervated biceps and radially innervated triceps as the only control signals to control a powered elbow, wrist, and multifunction hand. This was indeed a great breakthrough and made the use of multi-powered limbs far more realistic and practical. However, two-site control has other problems – the prosthetic electrodes must be sited precisely over the strongest surface electromyography (sEMG) signal and the patient must learn to activate agonist and antagonist separately and accurately. When a person needs to learn four-site control with similar ’purity‘ of signal, it was quite a difficult thing to achieve reliably.

Pattern recognition was a conceptual shift in myoelectric control. This concept dated back to at least the 1970s, but no one had tried incorporating it into control systems for commercially available prostheses until the TMR work at The Rehabilitation Institute of Chicago (RIC) in the early 2000s, in collaboration with the University of New Brunswick School of Biomedical Engineering. Pattern recognition sampled all the muscles around the limb simultaneously when movements were attempted, and the computer ‘learned’ what the pattern of activation represented; in terms of what movement the patient desired in the elbow, wrist, or hand. Pattern recognition is much easier to use in a multi-powered limb and in a transradial level myoelectric hand. It also offered movement control beyond what even our multifunction hands could deliver.

Adding sensory perception to upper limb prosthetics is the major missing component to enable prostheses to reach their potential as a replacement for the lost upper limb. Fortunately, while TMR was performed in Chicago, some sensory reattachment occurred fortuitously and was detected by the patients who described areas of skin overlying the TMR which when touched, produced the sensation of touch on part of their missing hand. Targeted Sensory Reinnervation (TSR) was then deliberately carried out as part of the TMR procedures to create areas of skin that were portals to the cortical areas mapped to the missing hand, particularly the fingers and thumb – sensors on the prosthetic fingers and thumb could enable the person to genuinely ‘feel’ objects with the prosthetic hand. Although this has been done in a limited number of cases in research settings, it has never moved into general use. 

The implication that prehensile function is dramatically improved by the ability to feel an object is not in doubt, nor is the improved ‘embodiment’ of the prosthetic limb. An additional bonus is the prospect of a reliable treatment for phantom limb pain – a cortical deafferentation phenomenon with dysfunctional neuroplasticity.

Subsequent to the TSR experiments, a separate DARPA funded group has demonstrated sensory reattachment via implanted nerve cuffs placed around sensory nerves and direct stimulation of the nerves from touch sensors on the hand. The cuffs remain viable without injury to the nerve for periods of at least five years. While these experiments reported in 2017 held great promise that a commercially available sensory system would soon be available, further technical difficulties have not been overcome.

Implantable motor systems have been utilised in research settings for at least the last five years. Groups in Gothenburg and Utah have implanted leads into muscles replacing the previous surface EMG electrodes and can demonstrate a more stable system. These have not been made available commercially yet.

Osseointegration is the direct skeletal attachment of prostheses. This was pioneered by Dr Rickard Brånemark in Gothenburg, Sweden in 1990, building upon the work of his pioneering orthodontist father who discovered that bone would bond to titanium a generation earlier. Dr Brånemark’s work was designed to provide a method of attachment for amputees who could not wear sockets, and the initial cases were typically short transfemoral and short transhumeral amputees, as well as some cases of thumb amputation. Dr Brånemark’s system, using a threaded barrel-shaped implant, requires two surgical stages and a slow, graduated weight-bearing protocol to achieve the most reliable outcome.  Press-fit implants were developed later by other orthopaedic surgeons utilising hip replacement technology. This enabled much earlier weight bearing by comparison. 

The world is currently seeing a minor proliferation of Osseointegration programs, with at least five or six different approaches available but these can generally be grouped into Brånemark-type threaded barrel implant or press-fit implant. Debate continues about whether single stage or two-stage surgery has better outcomes. All approaches carry the risk of sepsis because of the open stoma which remains in all the designs. Other complications include implant loosening and periprosthetic fracture. Implant removal may require further loss of limb length. Trauma patients are the most commonly implanted, with vascular disease generally contraindicated.

In Australia, the Melbourne group based at the Alfred Hospital has used the Brånemark system since 1995 and the current team led by plastic surgeon, Mr Frank Bruscino-Raiola, prides itself on its meticulous stoma formation, striving to minimise any ooze from the stoma and thus reduce the risk of infection. This requires a two-stage procedure. At the Macquarie University Hospital in Sydney, a group headed by Associate Professor Munjed Al Muderis, currently the world’s most prolific osseointegrator, uses its own proprietary press-fit system and takes a single stage approach. 

Ultimately, osseointegration provides advantages for transhumeral amputees by preserving more of the normal shoulder ROM, especially in presence of a shorter remnant. It also provides a more secure fixation when a heavier, multi-level powered prothesis is utilised. 

The upper limb prosthetic experience demonstrates the great strides that have occurred in connecting the human nervous system to robotic componentry, both at a motor and sensory level. These advances offer tantalising evidence that similar re-connection could be offered to patients with different diagnostic problems. The upper limb prosthetic neural interfaces have generally utilised peripheral nerves and peripheral muscles because it is technically easier to utilise signals that are already large enough to detect and specific enough for purpose. In research settings, direct control of devices from the brain has also been demonstrated, with both directly implanted systems and electroencephalogram (EEG) based surface systems. These are often grouped under the name brain machine interface or brain computer interface. By comparison to peripheral signals, brain signals are smaller and more difficult to separate out from the background noise. The intense research interest in this field would suggest that control systems will continue to evolve and may offer future neural reconnection for spinal cord and brachial plexus injuries.

Dr Greg Bowing
AFRM President 

Faculty Policy & Advocacy Committee update

2021 is now well underway and the policy and advocacy work of the Faculty Policy & Advocacy Committee (FPAC) continues chugging along. As is the pattern of all policy and advocacy work, tasks of FPAC move between urgent requests for consultation with very short response times (those that members won’t generally hear about until after they have already happened) all the way through to the ‘slow burn’ projects, those projects that require considerable ongoing work to reach their goal (those that you keep reading about in these reports). I always think that the ‘slow burn ‘projects are the most challenging as it is hard to keep sight of the long term goal and to maintain enthusiasm, so my thanks go out to those members who continue working on those projects. 

Working Groups

Bariatric Rehabilitation Position Statement: By now you all should have received and hopefully completed a survey for members relating to bariatric services. The results of that survey will assist the working group in development of the Faculty position statement.

Stem Cell Therapy for Children with Cerebral Palsy Position Statement Review: Work continues on this position statement. Now with valuable consumer representation included in the group, we will be looking to complete this document in 2021.

Guiding Principles for the Management of Patients with Multi-resistant Organisms (MROs) in Rehabilitation Units Position Statement: As the members of this group are located in Victoria, COVID-19 put a halt to their work in 2020. With things settled (and hopefully to stay that way) they will be recommencing their project in 2021.

Prehabilitation: This is currently still on hold as we focus on completion of other projects. But watch this space in Rhaïa later this year for updates.

Other work

COVID-19 and rehabilitation
The AFRM COVID-19 Rehabilitation Group has recommenced regular meetings to update and discuss current issues relevant to Faculty members. It is a valuable opportunity to discuss current issues at a binational level. Members are welcome to join the group. The Faculty continues to have a representative on the RACP COVID-19 Expert Advisory Group (ERG) and discussions/advice from our COVID-19 Rehab Group can help inform the RACP COVID-19 ERG about COVID-19 issues from a rehabilitation perspective. 

Australian Council on Healthcare Standards (ACHS) Clinical Indicators
The revised recommendations for rehabilitation clinical indicators were submitted to the ACHS for their review. If accepted, these are expected to replace the current indicators used by ACHS by 2022.

Interesting reading

Finally, there are often other issues that come up from time to time that the Faculty may or may not directly work on, but I will note in this article as they are potentially of interest to members.

Review of National Disability Insurance Scheme (NDIS) Assessments
You will be aware of the extensive reviews being conducted of NDIS processes. It is worth members keeping an eye on this review as items do come up that are relevant to rehabilitation physicians. For example, there has been recent review work underway of how NDIS assessment processes are conducted. There are many opportunities for both individuals and organisations to submit views to the review. Link: Access and eligibility policy with independent assessments | NDIS

Royal Commission into Aged Care Quality and Safety Report
By the time of this publication the Commission should have released their final report. In the interim report released in late 2020, several recommendations were made regarding access to rehabilitation services for the aged and for young people in nursing homes. I’m sure the final report will be worth a read. Royal Commission into Aged Care Quality and Safety

Dr Jennifer Mann
AFRM President-elect
Chair, FPAC

Adrian Paul Prize

The Adrian Paul Prize is awarded annually for the best scientific work in the field of rehabilitation medicine submitted as part of the AFRM Advanced Training Program.

2020 Recipient – Dr Petria Carter

Dr Petria CarterDr Petria Carter recently completed her Master of Medicine (Clinical Epidemiology) at the University of Sydney. She completed her postgraduate MBBS at the University of Sydney in 2012 after graduating with first class honours in Science (Psychology). Petria is a final year rehabilitation medicine trainee, most recently positioned at St Vincent’s Hospital in Sydney. She has been a national representative for Australian universities in rowing and developed her keen interest in musculoskeletal medicine because of this. Petria is also interested in neurological rehabilitation, developmental psychology, research and quality improvement.

The 2019 recipient was Dr Tim Butson.

The Prize was established by the AFRM in 1986 through a donation by the late Mrs Nancy Paul, wife of the late Dr Adrian Paul, the Director of the Department of Rehabilitation Medicine of the Royal Alfred Hospital.

Targeted Muscle Reinnervation: An update

Major limb amputations result in significant long term functional disability, pain, psychological and social problems amongst amputee patients. Over the past 30 years, surgical developments in the area of bone anchored prosthesis and residual nerve management have reinvigorated optimism in this field. 

In 2016, The Alfred Hospital in Melbourne established a multidisciplinary Advanced Surgical Amputee Program (ASAP)* to specifically treat major limb amputees. Simply summarised, this service provides a trifactor of: 

  1. Osseointegration (OI)
  2. Residual nerve management
  3. Soft tissue management.

OI was discovered by Swedish Professor Per-Ingvar Brånemark in the 1960s. The Alfred performed the first femur OI in Australia in 2000. Our major indication for OI remains failure of standard socket prosthesis, usually in transhumeral and transfemoral amputees. OI eliminates the socket and its related problems especially to the skin residuum. It improves comfort, simplifies, and expedites prosthetic attachment. It also improves range of motion, gait and decreases energy expenditure. It provides a conduit for sensory feedback (osseoperception). Not surprisingly, prosthetic use improves by four-fold in our patients.

These benefits must be balanced against the risks. These risks will vary depending on three factors:

  1. the system used 
  2. the team using it
  3. patient selection. 

This is reflected in the wide range of reported complications within the literature (Table 1).

Table 1: Risks reported in the literature.

 Risks  Failure rate 
 Componentry failure (abutment) 10-60% 
 OI failure 10-20% 
 Pain 10-40% 
 Deep infection 10-40% 
 Superficial infection 15-100% 
 Fractures 0-10% 

We use the Osseointegrated Prostheses for the Rehabilitation of Amputees (OPRA™) Implant System, a Swedish two-stage procedure three to six months apart. Although The Alfred performed the first femur OI in 2001. It was not until 2016, following the establishment of the multidisciplinary service, that significant changes were made to our protocol (Figure 1).

Figure 1-Alfred Osseointegration protocol
Figure 1: Alfred Osseointegration protocol with OPRA™ Implant System.

Abutment fracture is a mechanical failure. Although not a major complication, this requires a trip to theatre to replace the abutment. During this time, the patient will be off the system. This was a major issue with the OPRA™ Implant System (up to 60%) until modifications were made in 2017; since then we have had no abutment issues (0%). Risks for OI failure (15%) in our cohort are smoking, being overweight (>90kg), and poor compliance with the rehabilitation program. Infection reflects soft tissue instability at the soft tissue metal interface. Due to our extensive soft tissue reconstruction, our risks are relatively small (12%). Pain (40%), as expected, was a significant issue following any stump surgery in amputees, but this problem has been resolved with the addition of TMR in all our patients in stage 1 (0%). Fractures have not been an issue in our patients. 

Although placement of the fixture (intramedullary titanium rod) is important, the vast majority of the surgery is spent on the soft tissue. It is critical to provide a stable soft tissue metal interface to avoid discharging sinus, commonly associated with superficial infections, exposure of bone, and eventually osteomyelitis/deep infection. Also important as part of the soft tissue management is to deal with the residual nerves to reduce or avoid phantom limb pain and residual limb pain. 

Figure 2 - second stage x-ray
Figure 2: Second stage X-ray.

Figure 3 - Metal soft tissue interface (transhumeral)
Figure 3: Metal soft tissue interface (transhumeral).

Figure 4 - Metal soft tissue interface (transfemoral)
Figure 4: Metal soft tissue interface (transfemoral).

Targeted muscle reinnervation (TMR) is a recent surgical technique designed to improve myoelectric prosthetic control in upper limb amputees. The procedure was first performed in 2002 at Northwestern Memorial Hospital, Chicago (Illinois, USA) and reported in the literature in 2004. Simply stated, TMR results in the formation of multiple strong, reliable and independent EMG signals to allow intuitive use of myoelectric prosthesis (Figure 5). Nerves that control a specific movement in the amputated limb will trigger a similar movement in the myoelectric prosthesis.

Figure 5 - TMR and use of myoelectric prosthesis
Figure 5: TMR and use of myoelectric prosthesis.

For example, in a transhumeral amputee, TMR will allow intuitive six-degree myoelectric control, negating the need of switching mechanisms (Graphs 1 and 2). The Alfred performed the first successful TMR in a transhumeral amputee for myoelectric control in 2016.

Graph 1
Graph 1: Shows increase in EMG signals in transhumeral non-acute patients post TMR.

Graph 2
Graph 2: Shows increase in EMG signals in transhumeral acute patients post TMR.

To undergo TMR (Figures 6 and 7), a patient must satisfy the standard criteria for a general anesthetic, avoid smoking and be cleared of proximal nerve injuries (brachial plexus). There needs to be soft tissue and skeletal stability, as well as enough muscle present or able to be harvested to provide targets for the residual nerves. Consideration needs to be given to the age of the patient.

Our experience is that early TMR performed in the acute post-injury stage and prior to fitting of a prosthesis allows for greater preservation of muscle mass, maintenance of neuromuscular junction receptivity, improved pain control and a more intuitive and expedited training process.

Figure 6-Proximal transhumeral amputation
Figure 6: Proximal transhumeral amputation. Divided nerves identified. Each nerve will be given a muscle target.

Figure 7-Tranhumeral amputation
Figure 7: Tranhumeral amputation. Nerves identified with potential muscle targets.

An extensive rehabilitation process including pre-prosthetic virtual reality training and a pattern recognition system are essential to achieve optimal results (Figures 8 and 9). 

Figure 8 - virtual reality training
Figure 8: Virtual reality training.

Figure 9 - COAPT pattern recognition system with TASKA prosthetic hands
Figure 9: COAPT pattern recognition system with TASKA prosthetic hands.

Post amputation pain can be simply categorised into Phantom Limb Pain and Residual Limb Pain. This has been reported in up to 85 per cent of amputees. To date, no single treatment has proven consistently effective for pain control following major limb amputation. Many theories have been postulated as to the cause and an array of surgical and non-surgical treatments advocated over the years. A biproduct of TMR for myoelectric control has been a significant improvement in post-amputation pain. As a result, nerve surgeries like TMR and Regenerative peripheral nerve interface (RPNI) are now becoming more routine in specialised centres.

TMR gives the regenerating nerve fascicles somewhere to go and something to do. It provides a distal nerve receptor for the nerve to innervate and therefore attempt to heal. It is important to understand that these procedures are not a silver bullet, but a big step in a process. A thorough assessment in a multidisciplinary setting to identify the cause and contributing factors to the pain is essential. This includes assessing the distal end of the nerve, the nerve in transit and the central nervous system including patient characteristics that may predispose them to pain syndromes. We use MRI and US guided injections routinely in our work up.

Our experience suggests that TMR in the acute setting (following amputation) would offer better pain control presumably by preventing abnormal neural pathways and cortical re-organisation. In the chronic setting, results are extremely encouraging but ideally, this surgery should be done at time of initial amputation.

Mr Frank Bruscino-Raiola MBBS FRAC
Director of Plastics, Hand & Faciomaxillary Surgery Unit 
The Alfred Hospital

*ASAP incorporates The Alfred Hospital Melbourne Plastics, Hands and Faciomaxillary Surgery Unit, Epworth Hospital Rehabilitation (Melbourne) and Promotion Prosthetics.

Medical management of spasticity: A current approach

In this short article, I will only be able to give an overview of the medical management of spasticity.  It goes without saying, of course, that medical management without appropriate and expert physical management will always lead to suboptimal outcomes.

The two decisions that need to be made are: firstly, should the spasticity be treated and secondly, are you dealing with focal, multifocal, regional, or general spasticity. It should be noted that spasticity of primarily spinal origin is usually regional or general in nature, and that spasticity of primarily cerebral origin is often focal or multifocal – although may be regional and occasionally general. Spasticity of primarily cerebral origin is almost always associated with dystonia and should usually be considered as a spastic dystonia rather than pure spasticity.


There are several medications that can be used for the treatment of regional and generalised spasticity. I will rarely, if ever, use systemic medications for the management of focal or multifocal spasticity. Although we don’t have time or space to discuss the pathophysiology of spasticity and dystonia, there are strong theoretical and experiential reasons to avoid oral medications when treating primarily cerebral origin spastic dystonia.

Baclofen, the most commonly used oral medication, is a GABA-B agonist that inhibits the release of excitatory neurotransmitters. It has a short oral half-life. The most frequent errors in baclofen prescription are inadequate dose and inadequate regime. It should be given at least tds and preferably qid (making compliance a problem) – in therapeutic doses always of course, considering side effects.

Diazepam acts at the GABA-A receptor site increasing the efficacy of endogenous gamma aminobutyric acid (GABA). It can be taken bd.

Clonidine is an α2 agonist, therapeutically very similar to the unavailable in Australia – Tizanidine. Careful consideration should be given to dosing above 300mcg/day as it does have some α1 activity and may cause an increase in spasms.

Dantrolene acts at the sarcoplasmic reticulum to prevent Ca++ release. It works primarily by weakening muscle, both spastic and unaffected. It exerts its anti-spasticity effect by also acting on the intrafusal fibres of the muscle spindle. I reserve its use to people with severe general spasticity who have little voluntary function where weakening will not be a great problem.

Gabapentin, despite its name and design does not act like GABA, or on the GABA-A or GABA-B receptor site. Its mechanism of action is unknown, although it possibly acts as an α2 agonist. I personally have been unimpressed with its use for spasticity, although it is arguably and currently the most popular medication. I note that there is limited current literature available.

Medicinal cannabis consists of varying combinations of THC and CBD. The THC component (the bit that makes you stoned) has been shown to be useful in the management of spasticity associated with multiple sclerosis (MS). It is not difficult to access. The jury is certainly still out with respect to benefits and side effects.

Botulinum Toxin (BoNT-A)

There are two commercially available Pharmaceutical Benefit Scheme (PBS) approved preparations of BoNT-A, Botox, and Dysport. Xeomin is available but not PBS-approved for treatment of spasticity. Despite BoNT-A having been used in Australia and world-wide for spasticity management since the early 1990s, there is no consensus on dose or dilution. There is reasonable consensus on the need to localise the specific muscle to be injected. Rehabilitation physicians tend to use either ultrasound or muscle stimulation, whereas neurologists tend to use EMG. In my opinion, ultrasound is far superior to either of the other two methods. The art and pleasure in using BoNT-A is in the clinical assessment and muscle selection and a great deal of experience and expertise is necessary to achieve optimal results. Despite the temptations to be an injector, the worst results and least benefits are those of the occasional injector.


Phenol has been used for spasticity management since 1965 (tibial nerve). Phenol is injected perineurally (unlike BoNT-A, which is intramuscular). It causes patchy demyelination of the nerve slowing down both afferent and efferent traffic in a mixed nerve. Average duration of benefit is around 12 months. It affects all muscles in the distribution of the nerve, so is best suited to nerves that innervate multiple muscles performing the same general function. The main complication is dysaesthesia so is best used for nerves with limited sensory function. The obturator nerve is by far the most commonly injected nerve. Other nerves that are frequently injected are the axillary and musculocutaneous nerves.

Intrathecal Baclofen (ITB)

The principle behind ITB is that baclofen is given directly to the subarachnoid space, bypassing the systemic circulation. The average dose of ITB is around 400mcg/day compared to an oral dose of say, 80mg (80,000mcg)/day, which is usually ineffective compared to the ITB. ITB is indicated for generalised and especially lower limb spasticity and is much more effective for trunk and lower limb than upper limb spasticity, irrespective of the placement of the catheter. A knowledge of cerebrospinal fluid flow and dynamics explains this. ITB is a highly specialised but invaluable tool in the management of difficult to treat generalised or regional spasticity.

Associate Professor Barry Rawicki
Rehabilitation Physician
Associate Professor Medicine, Monash University

Spasticity: When I was a medical student

When I was a medical student 400 or so years ago, ’spasticity’ was easy: easy to recognise and easy to treat. 

Easy to recognise because it was essentially a spot diagnosis – if the person walked with a stiff knee or equinovarus foot and/or had their arm up around their ribcage then they had spasticity. 

I could do a quick knee jerk test with my tendon hammer and detect hyperreflexia – case closed. 

Easy to treat because there were only a few oral medications available that either had a risk of addiction (diazepam), hepatotoxicity (dantrolene), or cognitive impairment (baclofen). Motor point blocks existed but were time consuming and required a high degree of skill to be successful. Consequently, most pharmacological treatment failed, and few patients would persist with splints or AFOs. Thus, spasticity for the clinician was something to document in the notes and for the patient was something to endure. Treatment was largely supportive or targeted at redirecting the patient to think about things that could be achieved. 

As often happens with science and medicine, things have got a whole lot more complicated. This was largely the effect of the development of more powerful interventions such as botulinum toxin and intrathecal baclofen. These medications also allowed us to better explore the nature of the condition, and to rethink what we were trying to achieve.

In my first few years of injecting, I had some wonderful successes (I remember one gentleman who returned to playing the piano six weeks after Botox), but many more cases where treatment was beneficial but not as good as I would have liked. In thinking about it over the following decade or two, it became evident that ’spasticity‘ was not one thing. 

I would argue that Lance’s ground-breaking 1980 definition of spasticity is largely irrelevant for rehabilitation medicine – I don’t want to know what happens on an EMG or a passive stimulus (such as that from a tendon hammer). I want to know how sensorimotor dyscontrol impacts upon the person’s ability to undertake movements during active function. Hyperreflexia is clearly real, but by itself it does not necessarily limit function. This observation has caused confusion in some clinical circles who interpret the absence of a ’spasticity angle‘ during Tardieu testing, for example, to either mean the muscle is irreversibly contracted or that pharmacological intervention is not indicated. It ignores other drivers of positive upper motor neuron (UMN) features such as abnormal postural reflexes or action dystonias, just to name two.

There had also been considerable confusion at a government level, with ’irrational rationing’ of access to BoNT-A creating an accidental funding apartheid of those who qualified versus those unworthy of injection. I was proud to work with Alex Ganora in his role leading the Botulinum Toxin Expert Working Group (run by the Rehabilitation Medicine Society of Australia and New Zealand) for my one foray into medico-political activism. After submissions and meetings with a subgroup of the Therapeutic Goods Administration (TGA) over a couple of years, we were successful in helping the TGA see the need to broaden accessibility for people with focal spasticity from acute neurological disease, rather than just stroke and cerebral palsy. While still not perfect, it was refreshing to see that the system was open to logic and sensible revision.

Without apology, rehabilitation medicine focuses on the individual, and by corollary, their family and carers. These days, in describing the UMN motor consequences to patients, I refer to muscles that won’t switch on (the negative features) and muscles that won’t switch off (positive features, or what I would have confidently called ’spasticity’ in my younger years) when you want them to. In a spasticity clinic context, my assessment goals now focus on what drives the second part of this dyad. 

Associate Professor Ian Baguley OAM
Brain Injury Rehabilitation Service
Westmead Hospital 

Obturator Nerve Blocks with Phenol for Adductor Spasticity Management

Botulinum toxin type-A (BoNT-A) is the most widely used treatment for focal spasticity. In Australia, it has finally become available on the PBS with the patient co-payment. However, the PBS S100 Botulinum Toxin Program has ongoing restrictions for some conditions such as MS (e.g. maximum treatment periods per year and maximum BoNT-A units per session). 

Obturator Nerve Blocks (OBNs) are an alternative option for hip adductor spasticity/spasms, in order to improve pain, gait, posture, lower body dressing, and/or perineal hygiene. ONB with phenol are significantly less expensive (AUD $75 for 600mg/10mL) than BoNT-A injections and effects can last up to 12 months (or more). If clinically necessary, BoNT-A can be additionally used to other muscles such as hamstrings, gastrocnemius and soleus. There are consensus guidelines, protocols and workshops of BoNT-A injections; however, in contrast there is limited literature to guide how ONB can be used in a spasticity clinic setting. 

So how did I start doing ONBs?      


There was a rotating pain fellow who worked in our rehabilitation ward four to five years ago. He was an anaesthetist in his country and had performed ONB. He taught me the distal approach with ultrasound and electric stimulator guidance, and I have been using this approach since. To place a transducer at the inguinal crease and block the anterior branch (between the adductor longus and brevis) and posterior branch (between the adductor brevis and magnus) of the obturator nerve. However, positioning the patient and placing the transducer at inner groin can be very challenging due to limited abduction and involuntary adductor spasms. Anatomical variations of the obturator nerve and echogenic muscle changes make visual anatomical landmarks more difficult. When I see the muscle twitching with minimum current (1-2 mA), I am sure ONB will work.             


I perform ONB with local anaesthetic (LA) (0.5% bupivacaine, max 5 mL per leg) first to see if the patient achieves the treatment goals. We can see the results within 15 minutes, which I find fascinating. We can see huge differences (before vs after ONB) during one consultation if adductor spasticity was a main contributing factor, unlike BoNT-A that takes a couple of days to work. As the effectiveness of bupivacaine lasts up to four hours, patients are encouraged to do their usual activities (e.g. toileting) at home. If patients are happy with the changes from LA, ONB with phenol is then organised. At our hospital, only one preparation (6% phenol aqueous) is available. I use 240-300mg of phenol per leg based on body weight and age (not exceeding 600mg in total). 

Side effects

From my experiences, one patient developed dysesthesia that continued for six weeks and two patients reported light-headedness/dizziness that improved within two hours.        

Searching for educational opportunities 

I have only six to eight ONB procedures per year, in comparison to 60 to 70 BoNT-A injections. I am keen for more educational opportunities (e.g. workshops) to further develop my anatomical and practical skills (my current teacher = YouTube). I am certainly interested in hearing from the experiences of other rehabilitation colleagues in this area.

Patient example

This patient with MS received ONBs with phenol and BoNT-A to both hamstrings followed by an intensive rehabilitation program. Post intervention, he was able to manage lower body dressing independently (photographs used with consent). 

Image 1

Image 2 
Post treatment

Dr Yuriko Watanabe
St Vincent’s Hospital Sydney

Setting up and running a spasticity service: an experience worth sharing 

From 2008 to 2020, I was involved in setting up and running a spasticity management service at the Royal Hobart Hospital, Tasmania’s only tertiary referral hospital. Spasticity management, in particular botulinum toxin (BTX) injections, is the area of my career that I enjoy most. It is extremely rewarding to see the clinical benefits, but there are aspects that are extremely challenging. My hope in this brief is to share aspects of our system and you can consider for yourself what kind of system to run.

When I started as a registrar, the administrator of our spasticity service was an occupational therapist. She organised clinics and even provided the occasional piece of advice to medical staff as to where to point the needle. It wasn’t long before she moved on. I was then given a handwritten A4 sheet of paper with patient names and medical record numbers and told to organise the clinic. The first thing that needed sorting was a database. Given I’m no expert, Excel was my database of choice. It was sufficient to record names, medical record numbers, diagnosis, source of toxin funding, and how many procedures had previously been performed. The upside of the administrative responsibility as a registrar means one learns quickly the ins and outs of a spasticity service. The downside is the time it takes to record data and organise the correspondence with patients, clinics, pharmacy and allied health. It takes significant time away from traditional registrar time, perhaps to one’s detriment.

Fortunately, an outpatient spasticity management clinic was already in place. This was a once-per-month multidisciplinary pre-assessment and review clinic only with the capacity to see about eight patients. Some centres choose to have an all-in-one clinic model, where the patient turns up, is assessed by the multidisciplinary team and then the medical staff perform the injections. This model can run well, but in my opinion, can be restrictive in public practice. Once everyone assesses, has their say, addresses goals, and the procedure is performed, it becomes a lengthy process. In my opinion, it is difficult to scale up this model to cope with demand due to cutbacks in allied health staffing. Our outpatient pre-assessment and review model was also limited, eight patients assessed/reviewed per month, which was less than the number of patients injected with BTX per month – it’s unsustainable.

Figures from the Stroke Foundation illustrate the future demand on our spasticity management services. More than 56,000 strokes experienced by Australians this year, more than 475,000 Australians currently living with the effects of stroke and more than 1,000,000 living with stroke by 2050. Further, the literature states that approximately 30 per cent of stroke survivors develop spasticity. In my mind, those figures suggest that our traditional ways of assessing, treating and reviewing patients with spasticity in our multidisciplinary clinics need reviewing, particularly because those figures are just for stroke. In 2008, I was handed a sheet of paper with about 30 names on it. By 2020, we ran a database with more than 120 patients actively receiving BTX across Tasmania. 

My approach was to increase the number of assessment/review clinics and the number of injection clinics. Over the space of 13 years, we went from performing BTX injections for approximately three patients/week to 10-12 patients/week in three separate injection clinics. Each of our physicians still ran a monthly multidisciplinary pre-assessment/review clinic, but this was usually set aside for more complicated patients. Patients with relatively straight-forward spasticity were seen by the physician alone on a weekly basis, with assessments and management plans developed without allied health present. Referrals were then sent to allied health so that pre-injection assessments and reviews could be arranged in their time, thereby maintaining the multidisciplinary nature of the model. It was by no means a perfect system, but it allowed us to meet demand. If we had stuck to the system of the single monthly multidisciplinary appointment or trialled the model where assessments and procedures are performed on the same day, we would not have treated as many and patients that needed our help would have gone without.

Whichever way our services run changes will be necessary. Demand is outgrowing supply. The turnover of allied health in public hospitals is generally out of our control, but somehow more therapists with experience in spasticity management are required. Administrative support should be available to manage services. Although there are advantages to a registrar-run administration model, I do not think it is fair to leave all those duties with trainees. Last but certainly not least, training for our current and future physicians is vital. This training needs to be of excellent quality and occur frequently. If it is not, then I believe patients will suffer and the demand on our health services will not be met. Training needs to be in assessment and procedural. As physicians, we ought to be able to carry out the more difficult aspects, perform an accurate assessment and implement an effective management plan. The procedural side requires further training with localisation techniques, but in my opinion, this is easier than assessment and management. Unfortunately, a once-per-year workshop is not quite enough.

After 13 years in Hobart, my family and I decided it was time for a change, so we moved to the Sunshine Coast in Queensland. Here we have decided to open a private spasticity management service and after five months of preparation we should be ready to open in one more month. Having experience in setting up a public service has been useful for setting up a private service, but it doesn’t mean it has been simple. Important steps included writing the business plan, securing finance, finding suitable real estate, fitting out the office, brand and business registration, organising a website, setting up practice management software, organising practice and ultrasound accreditation, a Medicare Location Specific Provider Number, ordering consumables, stationery, a BTX ordering and delivery system, a reliable vaccine fridge with a backup power supply, setting up professional connections, writing practice policies and guidelines – the list goes on. If I can offer advice here, it is that though the process seems daunting and overwhelming, aim to tackle one issue at a time and it will come together. It has taken longer than I initially anticipated, so it is worth expecting a lengthy process. Patience is a virtue; I better work on that.

Dr Warren Jennings-Bell FAFRM, MBBS, B Med Sc, B Sc
Rehabilitation Physician
NeuroRehabilitation, Sunshine Coast QLD

Robotics in rehabilitation medicine

As clinicians in neurological rehabilitation, we are often fighting against the natural history of the disorders we are managing. Most commonly, this occurs in stroke and traumatic brain injury in the adult sector, and traumatic brain injury and cerebral palsy in paediatrics. 

In many studies, intensity of therapy is identified as a critical factor for maximising change in functional outcome.1,2 However, achieving the intensity needed to prevent deterioration and promote functional improvement is often elusive, given the amount of therapy input needed to bring about recovery. An exciting area that could help fill this gap is robotic technology. 

Robots are machines capable of carrying out a complex series of actions automatically and are typically programmed by a computer. Robotics is the branch of technology that deals with the design, construction, operation, research and application of robots. There has been a shift in the use of robotic technologies away from helping individuals cope with their environment, to one of interactive robots used to directly help facilitate recovery.3 The critical elements for neuroplasticity and motor learning are not only for intensity of therapy, but for repetition, task specific training and motivation as well as tactile, visual and auditory feedback for self-monitoring and learning.4 Robotic technology has the potential to provide these requisites and contribute to an intensive rehabilitation program. 

We are currently investigating the role that robotic technology can have in children with neurological disability through The Little Heroes Foundation Centre for Robotics and Innovation at the Women’s and Children’s Hospital in Adelaide, South Australia. We have conducted one randomised controlled trial in traumatic brain injury and a feasibility trial of robotics in fragile oncology patients post-chemotherapy, with the aim of improving fitness and functional outcome. Our findings to date indicate that there is feasibility to these technologies both in efficacy and acceptance by our paediatric patients that seems to be primarily related to improved motivation, repetition of tasks and overall dose of therapy, as well as improvements in sensory feedback where motor learning is optimised.

An example of the advantage of sensory feedback provided by the robots is illustrated in a case study. We assisted a 10-year-old boy with cerebral palsy whose goal was to improve his gait efficiency and endurance. Even before this treatment in robotics, his physiotherapist was trying to teach him to isolate his quadriceps muscles on the right, a difficult task for him due to poor selectivity and difficulty understanding the therapist’s instruction. It was not until he got a visual feedback loop from the Lokomat® robotic gait orthosis, having seen a 'peak' on the graph noting knee extension, that he finally understood what “contract your quads on the right” really meant. Once he understood this action, he was able to generalise this through practise in land-based physiotherapy more efficiently and improve the quality of his gait, which was documented objectively in his Gross Motor Function Measure.

There is concern that robotic technology has several deficiencies, such as the cost of providing robotics, the lack of variability of movement within the machines, and a lack of research evidence demonstrating efficacy to the point where the cost of the robotic technology can be fully supported. With ongoing research, this view is changing such as the cost equivalency of robotics with conventional therapy as demonstrated in one study.5 However, there are advantages to the application of robotics, such that this technology is likely to be a valuable adjunct to rehabilitation therapies into the future. 

Associate Professor Ray Russo PhD, MBBS, FRACP, FAFRM 
Head of Research, Paediatric Rehabilitation
Women’s and Children’s Hospital

1. Horn SD, DeJong G, Smout RJ, Gassaway J, James R, Conroy B. Stroke rehabilitation patients, practice, and outcomes: is earlier and more aggressive therapy better? Archives of Physical Medicine and Rehabilitation. 2005; 86(12):101-14.
2. Damiano DL. Activity, activity, activity: rethinking our physical therapy approach to cerebral palsy. Physical Therapy. 2006; 86(11):1534-40.
3. Krebs HI, Dipietro L, Levy-Tzedek S, Fasoli SE, Rykman-Berland A, Zipse J, Fawcett JA, Stein J, Poizner H, Lo AC, Volpe BT. A paradigm shift for rehabilitation robotics. IEEE engineering in medicine and biology magazine. 2008; 27(4):61-70.
4. Esquenazi A, Packel A. Robotic-assisted gait training and restoration. American Journal of Physical Medicine & Rehabilitation. 2012; 91(11):S217-31.
5. Lo AC, Guarino PD, Richards LG, Haselkorn JK, Wittenberg GF, Federman DG, Ringer RJ, Wagner TH, Krebs HI, Volpe BT, Bever Jr CT. Robot-assisted therapy for long-term upper-limb impairment after stroke. New England Journal of Medicine. 2010; 362(19):1772-83.

Shameless wearable technology: Impact on assistive technology for people with disability

As a 10-year-old boy, my difficulties in reading the blackboard at school brought the need for corrective glasses. I was a boy who was very self-conscious and I was a non-compliant wearer. Seeing other children in my class getting glasses at the same time did not change my resolve. Sensitivity to my appearance to others limited me and was further fueled by comments from my peers (“Do you have to wear those now?”).

I am a Kids Rehab Doctor with a strong interest in use of technology for innovation in reducing disability. I have always been acutely aware of my own situation as a boy and the things that we ask children to bear (even something as simple as wearing an ankle foot orthotic [AFO]). One of my kindergarten age patients was asked by a peer “Are you a robot?” on sighting his plastic AFO. A term that I heard in the 1990s associated with optimising assistive technology compliance was that it must not be a 'badge of disability'. 

Since the 1990s, there has been a proliferation of wearable technology for mainstream consumer use. This is changing our perception and acceptance for wearable technology. Shamelessly, people in the community are increasingly wearing their devices such as Bluetooth earphones, smartwatch, fitness tracker and safety wearables (including call alarms). Smart garments embracing the convergence of textiles and electronics into e-textiles and smart glasses, to provide an access to data with the potential for augmented reality, have been emerging. Such progress builds a new state of play for children (and adults), with differences embracing assistive technology and its, more recently, increasing 'cool factor' in the broader community. 

A key element in such a factor is appearance (underlined in my own experience above). The term “uncanny valley” was introduced by Mori in Japan in 1970 (Figure 1). It mainly refers to human likenesses in robotic technology. It is also applicable to adaptive technology such as prostheses, exoskeletons, and brain-machine interfaces. Mori wrote “I have noticed that, in climbing toward the goal of making robots appear human, our affinity for them increases until we come to a valley, which I call the uncanny valley”. This is the point at which the likeness is almost but ’not.quite.right’. It produces a feeling of eeriness and becomes uncanny. This progression is shown in the figure below. The valley shown in the curve becomes steeper and deeper when movement of the robot is introduced.

Figure 1 - The uncanny valley graph.tif
Figure 1: The uncanny valley graph created by Masahiro Mori: As a robot’s human likeness [horizontal axis] increases, our affinity towards the robot [vertical axis] increases too, but only up to a certain point. For some lifelike robots, our response to them plunges, and they appear repulsive or creepy. That’s the uncanny valley.

Our response to this as rehabilitation technologists can take one of two divergent approaches with both having merit. Firstly, we can use our innovation to progress our technology to the right of Mori’s curve by more closely emulating natural systems and appearances. The development used in prostheses of incredibly lifelike silicon skin with matching pigment and hair is an example. This is a slowly progressing combination of cosmetic aspects of technology and societal change which is, nonetheless, likely to remain asymptotic to the goal. 

Secondly, we can abandon the goal of being lifelike and embrace the difference by moving to the left of Mori’s curve. Existing prosthetic examples might include the Greifer hand or the running blade. In these cases, the differences are worn confidently as a badge of honour this time. Here, the commitment to function and play supersedes concern for the appearance of differences. This is an outcome of a gradual progression in attitude of the individual and the community where the achievements of people with differences can be paramount and celebrated.

In my own journey, wearing glasses eventually became a part of my life (although I do enjoy contact lenses at times). I accepted the changes it would make to my appearance and, perhaps, I have become a little more resilient. I can laugh with the ophthalmologist when he likens my glasses to the bottom of soft drink bottles (albeit feeling the comment perhaps just a little inappropriate from an eye professional). My father in recent years had a cataract operation that relieved him of his lifetime requirement (to that point) to wear really thick glasses due to an intraocular lens replacement. Recently, on seeing me wearing my glasses, stated animatedly “They’re really thick glasses you’re wearing”. All I could do was shrug, smile and reflect on my own journey. For the individual, growth and change is personal. Others may not understand another’s journey, and this is okay. By appreciating this fact, effective growth and acceptance of change, including emerging assistive technology, without the need for the approval of others, will allow the individual to flourish. 

Dr Timothy Scott
Paediatric Rehabilitation Specialist
Rehab2Kids, Sydney Children’s Hospital 

1. https://spectrum.ieee.org/automaton/robotics/humanoids/the-uncanny-valley 

Close overlay