Introduction
Prosthetic limb technology have been evolving for thousands of years, from its primitive beginnings [26][69][58] to its sophisticated present, to the exciting visions of the future.
Today, there are several millions of amputees worldwide, and more than one million new amputations are being carried out every year [3]. It is projected that the amputee population will more than double by the year 2050 [1]. Amongst those living with limb loss, the main causes are:
- vascular disease – including diabetes [46] and peripheral arterial disease [56],
- trauma – such as vehicular accidents and combat injuries, and
- cancer – which includes children with sarcomas [64].
Throughout history, humans have used state-of-the-art technology to help amputees to regain limb function and reintegrate into society. An ideal prosthetic limb would have the natural control, strength, sensation, weight, comfort and appearance of the respective native human limb, mimicking its functionality. However, engineering an artificial limb to provide the same form and function as the natural limb is an enormously challenging task. Unfortunately, the familiar images of restored limbs from Star Wars or iRobot do not exist in reality.
Current prosthetic solutions restore some degree of functional ability, but this ability falls below the standard set by the natural limbs. Ultimately, the measure of success for the rehabilitation of amputees is not how state-of-the-art their prosthetic legs or arms are, but how well they can live their lives.
In this essay, I intend to provide a critical review of the current state-of-the-art in the prosthetics field and discuss future possibilities.
Osseointegration
Quality of life, following amputation, is highly dependent to the ability to effectively use a prosthetic limb. The method of connecting an artificial limb to the amputated stump is paramount to the effectiveness of the prostheses and, therefore, to its adoption.
The current standard approach to attach a prosthetic limb to the body is by using a socket [32]. This approach, albeit practical, has the inherent disadvantage of trying to attach a rigid device to a soft, delicate and variable body. Many amputees experience serious discomfort wearing a prosthesis, because of painful pressure points, skin irritation, instability during walking, sitting discomfort, or sweating and bad odour [22][14]. Ensuring a satisfactory socket fit is a challenging task, since the shape of the socket remains in a fixed state even though residual limbs shrink and swell over the course of a day and change shape over time. Advances in socket suspension systems and new materials for socket liners have improved the security and comfort of the connection between the socket and the residual limb [58]. Nevertheless, even with well fitted sockets, amputees consistently report a range of complications and difficulties [22].
Osseointegration provides an alternative approach to attaching a prosthesis to the residual limb. It involves fitting a titanium-based implant in the medullary canal [53] of the bone of the residual limb. The implanted fixture has a percutaneous component (known as abutment) that is protruding from the skin to connect to the external prosthetic limb [58]. These recently-developed endo-exoprostheses do not have many of the disadvantages of prosthetic sockets and can be a promising option for amputees.
Osseointegration was accidentally discovered in the 1950’s, when it was first observed that titanium-based implants have the property of fusing permanently with living bone [60]. A similar effect can be obtained by coating alternative metals with biomaterials [41], such as hydroxyapatite [50], that mimic bone. Attaching prostheses to the skeleton by osseointegration is now relatively well established in dentistry [45]. It is also being successfully used in total hip arthroplasty [49] and for knee and shoulder replacement [52] [65], which similarly utilise intramedullary implants that are, however, subcutaneous.
Osseointegration-based prosthetic attachments do not apply pressure to the soft tissue of the residual limb, do not loosen with perspiration, and can be performed on residual limbs too short for sockets. Research has also shown that amputees with an osseointegrated prosthesis enjoy increased walking ability, stability, better fixation, larger joint range of motion, diminished phantom limb pain, maximum sitting comfort, quick donning and doffing, improved functional capacity, and an overall increase in quality of life [38].
Furthermore, direct skeletal attachment offers significantly improved levels of proprioception, or more specifically ‘osseoperception’ [13], which translates to increased confidence and efficiency in controlling the prosthesis.
Achieving osteointegration of a titanium implant is now a very well understood and reliable process. However, there are various risks associated with such direct skeletal attachments. Since the stiffness of the intramedullary implant is much higher than that of the bone, high impact loads – e.g. during falling accidents – can cause bone fractures. The abutment connectors of the implants are designed as safety devices, protecting the bone from damage [18].
Additionally, given that the abutment protrudes permanently through the skin, there is a potential chance of infection at the stoma of penetration [27]. The possibility of mimicking the biological model of deer antlers has been investigated [30] as a potential solution to this problem.
Neural Interfaces for Control of Limb Prostheses
After its structural interface with the amputated limb, the second most important aspect of any prosthesis is the degree by which it restores functional ability. Current conventional prostheses often fail to meet the daily needs of amputees, to the extent that are frequently abandoned. Recent advancements in science and technology have led to the development of advanced robotic prostheses [37][28][31] and to promising new methods for controlling them using neural information [33].
Although lower-limb prostheses present significant functional challenges, high-fidelity control of upper-limb prostheses is exceedingly dependent on neural information from the amputee. The complexity of the human hands can be figuratively illustrated by Penfield’s cortical homunculus [43], which shows how disproportionally bigger are the motor and sensory projections of the human hands on the cerebral cortex.
Current functional prostheses can be broadly categorised into body-powered or externally-powered prostheses. The design of body-powered arm prostheses has remained largely unchanged for more than a hundred years. They utilize cables and harnesses strapped to the amputee, to mechanically manoeuvre the artificial limb through movements of the shoulder and residual arm [58]. Most amputees today continue to use body-powered prostheses, most likely because of their relatively low cost, light weight, high reliability and functionality, and the sensory feedback they provide through cable forces [69].
Externally-powered prostheses represent an attempt to overcome the limitations of body-powered prostheses, by utilising electric power and an electronic system to control movement. Neural interfaces attempt to control the prostheses, utilising neural motor commands that originate from the human brain. Such neural signals can be intercepted either from muscles or from nerves or even the brain.
Myoelectric prostheses are the most widely available type of upper-limb externally-powered prostheses. Myoelectric prostheses utilise electromyography (EMG) [47] signals from voluntarily contracted muscles within the amputee’s residual limb to control its functions [29].
Most available myoelectric prostheses use two surface EMG electrodes to pick up antagonistic muscle groups to control a single function at a time, such as open/close, flex/extend or pronate/supinate. Muscle contractions basically act as on/off switches. However, some prostheses offer proportional control, which allows for controlling the speed and force of movements based on the length or intensity of contraction [57].
Compared to traditional body-powered prosthesis, a myoelectric arm provides greater comfort, more range of motion, a larger functional area and a more natural appearance [29].
Myoelectric prostheses, however, have several important drawbacks, including [69][39][36]:
- It is unintuitive, error-prone and tedious to learn to isolate muscle signals. The muscles used for control are not related to the intended arm function.
- It is not possible to execute complex movements requiring simultaneous articulation of multiple joints. These must be executed sequentially, resulting slow and unnatural control.
- The surface mounted electrodes can only detect the gross contractions from a group of superficial muscles. This limits the number of specific sites that can serve as unique control inputs.
- Due to the lack of sensory feedback, visual input must be constant, resulting to high cognitive burden.
- There is a delay between initiation of movement command and actual mechanical response.
- Minute fluctuations in the positioning of the electrodes relatively to the underlying muscles, can result in false EMG signals or diminished EMG output.
- Changing skin conditions (e.g. moisture from perspiration) or even external electromagnetic fields may also interfere with EMG signals, which could elicit unwanted actions of the prosthetic limb.
- Despite their widespread availability, they remain very expensive.
These limitations result in less functional control, which in turn can lead to frustration and ultimately to abandonment of myoelectric prostheses [6].
Many of the problems of current myoelectric prosthetic systems are related to using surface-mounted EMG electrodes. An alternative approach is to use implanted myoelectric sensors to transmit EMG signals directly from within the patient’s muscles [39]. Implanted sensors offer improved reliability and signal fidelity and increase the potential number of control input sites. In the case of osseointegrated prostheses, the implanted fixture and percutaneous abutment can potentially act as a conduit for interfacing to such implanted sensors, eliminating the need for dedicated percutaneous or wireless connectors [19].
In the past decade, the technique of targeted muscle reinnervation (TMR) has been developed, which, is arguably the most promising of the immediately applicable neural control strategies for robotic prosthetic limbs [21][66]. Unlike conventional myoelectric control, TMR is intuitive and enables amputees to simultaneously move multiple joints, significantly increasing the ease and speed of task performance.
This innovative surgical procedure reroutes the peripheral nerves in the residual arm and reattaches them to healthy, target muscles elsewhere in the body, such as a chest muscle [21]. TMR is essentially a technique for amplifying the information transmitted through the residual nerves, by contracting the reinnervated muscles. The amputee simply has to think of performing a movement with his missing hand, triggering a motor command, which results in the contraction of the respective reinnervated muscle. EMG sensors pick the resultant neural signal, which is in turn translated to an action performed by the prosthetic limb.
An additional benefit of TMR is that it creates the potential of providing cutaneous sensory feedback to amputees. It has been shown that, in parallel to reinnervating the target muscles, it is possible to reinnervate the skin overlaying them with the respective afferent sensory nerves [66]. The result of this targeted sensory reinnervation (TSR) is that when stimuli – like touch, pressure and heat – are applied to the reinnervated skin, they are perceived as if they were applied on the missing limb. In this way, tactile information from sensors on the prosthetic hand can be indirectly transferred to the peripheral nervous system of the amputee. Tactile sensory feedback is essential to intuitive control of a prosthetic limb.
The holy grail of neural interfaces for the control of limb prostheses is the interception of nerve signals directly from the motor and sensory nerves of the peripheral nervous system. The transmitted signals to be recorded by such direct neural interfaces are approximately a thousand times weaker than EMG signals generated by muscle contractions. Nevertheless, it is possible to detect them using implanted electrodes that are placed inside or around the nerves. This is an active topic of research and some successful attempts have already been demonstrated [9].
It may sound like a slice of science fiction, but technology that could enable us to directly interface to the human brain is also becoming a reality. Various brain-computer interfaces are being developed that allow acquisition of neuroelectric signals, mainly for helping quadriplegics [67] or paraplegics [55]. In one case, researchers surgically attached cortical microelectrode arrays [54] to the specific regions of the patient’s primary motor cortex [62] and of the primary somatosensory cortex [61], which are responsible for the control of voluntary movements and for the sense of touch of the patient’s arms and hands [10]. In this way, two-way communication between brain and machine was enabled, which allowed the patient to control a robotic arm just with the mind and to receive accurate sensory feedback.
Both direct neural interfaces and brain-machine interfaces require significant additional research before they can become available to amputees.
Cosmetic Considerations
Many people who have an amputation, often suffer from significant body image issues. This invariably impacts considerably their psychological well-being, emotional health, and social lives. Coming to terms with the psychological impact of an amputation is often as important as coping with the physical demands [25]. Conventional prosthetic limbs do not help amputees overcome any of these issues because the majority of them do not look or feel like natural limbs.
An ideal prosthetic limb would arguably have a natural appearance, with a size and weight comparable to those of the native limb. In addition, regardless of the underlying mechanical structure, it would move and react to resistance in a lifelike manner.
The present state-of-the-art cosmetic solutions for prosthetic limbs are extremely realistic, high definition silicon coverings that are carefully crafted to precisely match the natural limb in terms of size, skin tone, hair distribution, nail characteristics, and even tattoos [58] [8] [2]. It this way, the prosthetic limbs can become very inconspicuous.
Unfortunately, such custom-made ‘artificial skins’ can be very expensive, particularly given that they need replacement after a few years.
Although most amputees would prefer a limb that is lifelike, when deprived of the option some would rather opt for a prosthesis with futuristic, robotic appearance [34]. This new trend incentivized the use of CAD and 3D printing fabrication methods to successfully produce fashionable designs of cosmetic covers that are customised for individual amputees [35][16].
Other Issues
There is a sharp contrast between the cost of a simple, body-powered prosthesis and that of a state-of-the-art neurally-controlled robotic limb. As reimbursement from either public or private insurers is generally capped for prosthetics, there is limited access to the most innovative devices, with the majority of patients only receiving a simple prosthesis [11]. Advanced prosthetic devices typically need to be paid privately. By providing the possibility of repairing disabilities at great expense, there is increased threat of deepening the gap between the rich and poor in terms of quality of life.
While it may sound far-fetched or at least far off, a future where prosthetics technology can seamlessly integrate with our bodies, may be closer than many people realise. In that future, we may indeed “have the technology” to rebuild the human body “better, stronger, faster”, as in the fictional case of Colonel Steve Austin from Caidin’s Cyborg. It may be possible to augment average human beings into superior beings, who are capable to function beyond a normal range. This possibility can create a multitude of ethical issues. There are existing cases of elective amputation [7][4][20], where people voluntarily amputate their paralysed natural limbs in order to replace them with robotic ones. In a future where artificial limbs surpass their natural counterparts in terms of functionality, will there be able people who will be willing to tinker with human nature for the purpose of enhancement [17]? Furthermore, there is the question of who will be entitled to such inevitably expensive enhancement technologies. Will there be a new elite of enhanced individuals and what will be the social implications? Will our efforts to eradicate disability discrimination result to new forms of discrimination and social inequality?
Extensive man-machine integration may also have security implications. Could someone hack into the control systems of the artificial body parts and manipulate them against the owner’s will? Would, in this way, whole new areas of cybercrime become possible?
Finally, the legal implications of the next generation of prosthetic limbs is an intriguing subject [12]. According to the current law, a person is a separate entity to her prosthesis, which is simply considered as her property [24]. However, as the integrations between man and machine become more intimate, such traditional views can be challenged. For example, a person could have an osseointegrated implant and/or neural interface implants and, therefore, be deeply integrated with her prosthesis. She may have ‘trained’ the control system of the prosthesis to best respond to her own neural patterns. The cosmetic covering of the prosthesis may be an integral part of her appearance and identity. In such cases, shouldn’t the prosthesis be considered as a part of the living person, even though is not biologically human?
Conclusions
It is possible that, in the far future, approaches like limb transplantation [48][5] or even limb regeneration [68][63] advance to a degree that they render non-biological artificial limbs unnecessary. For the near future, however, neuroprosthetic limbs appear to offer the best solution.
The single most critical aspect of any prosthesis is the quality of the interface between the residual limb and the prosthetic device. Sockets can cause excessive pain and discomfort, leading up to 80% of amputees to abandon their prosthesis within two years [27]. In the future, direct skeletal attachment of prostheses, using osseointegration, may be a routine option for a large proportion of amputees.
There has been a relatively recent surge in technological development and scientific understanding that has paved the way for the realization of advanced neural interfaces for prosthetic limb control. Targeted muscle reinnervation has demonstrated efficacy and limited risks and is now a clinical reality for amputees. Direct neural interfaces and brain-machine interfaces require much additional research and development before they become safe and reliable [33]. Nevertheless, at the rate at which the technology is advancing, we may see their realisation sooner than it seems.
The prosthetic limbs are becoming increasingly integral to the persons for whom they remedy lost functions and capacities. As the distinction between the living and the artificial body is eroded, new ethical, social and legal issues are emerging and will need to be addressed.
It seems quite likely that in the future, amputee demand for more versatile, lifelike, higher performance prostheses will fuel further innovation. The use of advanced prosthetic limbs is liable to be controlled by funding constraints. It is really these financial constraints that limit the rate of advancement in prosthetic rehabilitation, and one of the greatest challenges for the future will be to find ways to fund widespread application of prosthetic innovations.
References
- Advanced Amputee Solutions, “Amputee statistics you ought to know”, http://www.advancedamputees.com/amputee-statistics-you-ought-know
- Alatheia Prosthetics, Portfolio, http://alatheia.com/portfolio/
- Amputee Coalition, “Limb Loss Statistics”, http://www.amputee-coalition.org/limb-loss-resource-center/resources-by-topic/limb-loss-statistics/limb-loss-statistics/
- Avramov E., “Experience: I chose to amputate my leg”, January 2015, The Guardian
- BBC News, “First double leg-transplant patient has legs amputated”, June 2013, http://www.bbc.co.uk/news/health-22855670
- Biddiss E., Chau T. “Upper-limb prosthetics: critical factors in device abandonment”, December 2007,
- Bowdler, N., “Bionic hand for 'elective amputation' patient”, May 2011, BBC News
- Bowers R., “The Wonderful World of Cosmesis”, March 2002, inMotion 12-2, http://www.amputee-coalition.org/inmotion/mar_apr_02/cosmesis.pdf
- DARPA, “Nerve Interface for Direct Sensory Feedback”, https://www.youtube.com/watch?v=rpLzLduBGhU
- DARPA, “Providing a Sense of Touch through a Brain-Machine Interface”, https://www.youtube.com/watch?v=A4BR4Iqfy7w
- GlobalData, “Despite Impressive Advances, Prosthetics Will Struggle to Clear Reimbursement Hurdles”, August 2015, https://healthcare.globaldata.com
- Goold I., Herring J., “Great Debates in Medical Law and Ethics”, August 2014, American Journal of Physical Medicine and Rehabilitation, 86-12.
- Hagberg K. et al., “Osseoperception and Osseointegrated Prosthetic Limbs”, 2008, Psychoprosthetics.
- Hagberg K. et al., “Socket versus bone-anchored trans-femoral prostheses: hip range of motion and sitting comfort.”, 2005, Prosthetics & Orthotics International
- Hanson W., “Advances in Cosmetic Covers for Upper-Limb Prostheses”, July 2001, inMotion 11-4, http://www.amputee-coalition.org/inmotion/jul_aug_01/cosmetic.pdf
- HealthManagement.org, “Fashionable Fairings: 3D-Printed Limb Covers”, https://healthmanagement.org/c/it/news/fashionable-fairings-3d-printed-limb-covers
- Honigsbaum M., “The future of robotics: in a transhuman world, the disabled will be the ones without prosthetic limbs…”, June 2013, The Guardiane
- Integrum, “OPRA Implant System”, http://integrum.se/our-solutions/opra-implant-systems
- Integrum, “Neural Control of Prostheses”, http://integrum.se/our-solutions/mind-controlled-prostheses/
- Kiss J., “What if a bionic leg is so good that someone chooses to amputate?”, April 2015, The Guardian
- Kuiken T.A. et at., “The use of targeted muscle reinnervation for improved myoelectric prosthesis control in a bilateral shoulder disarticulation amputee.”, December 2004, Prosthetics and Orthotics International
- Levy S. W., “Skin problems of the leg amputee”, 1980, Prosthetics & Orthotics International, 4
- Marks L.J., Michael J. W., “Artificial Limbs”, September 2001, BJM Vol. 323
- Mason L. R., “Is Harm to a Prosthetic Limb Property Damage or Personal Injury?”, January 2016, http://motherboard.vice.com
- NHS Choices, “Amputation”, July 2016, http://www.nhs.uk/conditions/amputation
- Norton K.M., “A Brief History of Prosthetics”, November 2007, inMotion 17-7 http://www.amputee-coalition.org/resources/a-brief-history-of-prosthetics
- Kang N., “Prosthetic Implants”, Royal Free Hospital, https://www.royalfree.nhs.uk/services
- Ottobock, “Michelangelo prosthetic hand”, http://www.ottobockus.com/prosthetics/upper-limb-prosthetics/solution-overview/michelangelo-prosthetic-hand
- Ottobock, “Myoelectric prosthetics 101”, http://www.ottobockus.com/prosthetics/info-for-new-amputees/prosthetics-101
- Pendegrass C.J. et al., “Nature's answer to breaching the skin barrier: an innovative development for amputees”, July 2016, Journal of Anatomy
- RSLSteeper, “beBionic Hand”, http://bebionic.com/the_hand
- Sabolich S., “Prosthetic Sockets”, inMotion 16-5, http://www.amputee-coalition.org/resources/prosthetic-sockets
- Schultz A.E., et al., “Neural Interfaces for Control of Upper Limb Prostheses: The State of the Art and Future Possibilities”, January 2011, PM&R 3-1
- Sansoni S. et al., “The Aesthetic Appeal of Prosthetic Limbs and the Uncanny Valley: The Role of Personal Characteristics in Attraction”, 2015, International Journal of Design, Vol.9 No.1
- 35. The Ortho Remedy, “The New Trend in Prosthetics Covers”, http://www.orthoremedy.com/blog/article/new-trend-prosthetics-covers
- The War Amps, “Myoelectric Arms”, 2010, http://www.waramps.ca/pdf/english-site/ways-we-help/artificial-limbs-and-devices/upper-limb/myoelectric-arms.pdf
- Touch Bionics, “i-Limb Hand”, http://www.touchbionics.com/products/how-i-limb-works
- Van de Meent et al., “Walking Ability and Quality of Life in Subjects with Transfemoral Amputation: A Comparison of Osseointegration with Socket Prosthesis”, 2013, Archives of Physical Medicine and Rehabilitation, 94
- Vandersea J., “Prosthetic Design Considerations for IMES-Controlled Prostheses”, Summer 2015, http://www.oandp.org/AcademyTODAY/2015Jul/4.asp
- Wikipedia, “Amputation”, https://en.wikipedia.org/wiki/Amputation
- Wikipedia, “Biomaterial”, https://en.wikipedia.org/wiki/Biomaterial
- Wikipedia, “Cochlear implant”, https://en.wikipedia.org/wiki/Cochlear_implant
- Wikipedia, “Cortical homunculus”, https://en.wikipedia.org/wiki/Cortical_homunculus
- Wikipedia, “Cosmesis”, https://en.wikipedia.org/wiki/Cosmesis
- Wikipedia, “Dental implant”, https://en.wikipedia.org/wiki/Dental_implant
- Wikipedia, “Diabetes mellitus”, https://en.wikipedia.org/wiki/Diabetes_mellitus
- 47. Wikipedia, “Electromyography”, https://en.wikipedia.org/wiki/Electromyography
- Wikipedia, “Hand transplantation”, https://en.wikipedia.org/wiki/Hand_transplantation
- Wikipedia, “Hip replacement”, https://en.wikipedia.org/wiki/Hip_replacement
- Wikipedia, “Hydroxylapatite”, https://en.wikipedia.org/wiki/Hydroxylapatite
- Wikipedia, “Iron hand (prosthesis)”, https://en.wikipedia.org/wiki/Iron_hand_(prosthesis)
- Wikipedia, “Knee replacement”, https://en.wikipedia.org/wiki/Knee_replacement
- Wikipedia, “Medullary Cavity”, https://en.wikipedia.org/wiki/Medullary_cavity
- Wikipedia, “Multielectrode array”, https://en.wikipedia.org/wiki/Multielectrode_array
- Wikipedia, “Paraplegia”, https://en.wikipedia.org/wiki/Paraplegia
- Wikipedia, “Peripheral artery disease”, https://en.wikipedia.org/wiki/Peripheral_artery_disease
- Wikipedia, “Proportional myoelectric control”, https://en.wikipedia.org/wiki/Proportional_myoelectric_control
- Wikipedia, “Prosthesis”, https://en.wikipedia.org/wiki/Prosthesis
- Wikipedia, “Oral and maxillofacial surgery”, https://en.wikipedia.org/wiki/Oral_and_maxillofacial_surgery
- Wikipedia, “Osseointegration”, https://en.wikipedia.org/wiki/Osseointegration
- Wikipedia, “Postcentral gyrus”, https://en.wikipedia.org/wiki/Postcentral_gyrus
- Wikipedia, “Primary motor cortex”, https://en.wikipedia.org/wiki/Primary_motor_cortex
- Wikipedia, “Regeneration in humans”, https://en.wikipedia.org/wiki/Regeneration_in_humans
- Wikipedia, “Sarcoma”, https://en.wikipedia.org/wiki/Sarcoma
- Wikipedia, “Shoulder replacement”, https://en.wikipedia.org/wiki/Shoulder_replacement
- Wikipedia, “Targeted reinnervation”, https://en.wikipedia.org/wiki/Targeted_reinnervation
- Wikipedia, “Tetraplegia”, https://en.wikipedia.org/wiki/Tetraplegia
- Yong E., “Will we ever… regenerate limbs?”, March 2013, http://www.bbc.com/future/columns/will-we-ever
- Zuo K.J., Olson J.L., “The evolution of functional hand replacement: From iron prostheses to hand transplantation”, Spring 2014, Plastic Surgery Vol.22 No.1