Disability and Rehabilitation: Assistive Technology

ISSN: 1748-3107 (Print) 1748-3115 (Online) Journal homepage: [Link]

A narrative review: current upper limb prosthetic

options and design

Lauren Trent, Michelle Intintoli, Pat Prigge, Chris Bollinger, Lisa Smurr
Walters, Dan Conyers, John Miguelez & Tiffany Ryan

To cite this article: Lauren Trent, Michelle Intintoli, Pat Prigge, Chris Bollinger, Lisa Smurr
Walters, Dan Conyers, John Miguelez & Tiffany Ryan (2020) A narrative review: current upper limb
prosthetic options and design, Disability and Rehabilitation: Assistive Technology, 15:6, 604-613,
DOI: 10.1080/17483107.2019.1594403

To link to this article: [Link]

Published online: 11 Apr 2019.

Submit your article to this journal

Article views: 907

View related articles

View Crossmark data

Citing articles: 8 View citing articles

Full Terms & Conditions of access and use can be found at

[Link]
DISABILITY AND REHABILITATION: ASSISTIVE TECHNOLOGY
2020, VOL. 15, NO. 6, 604–613
[Link]

REVIEW ARTICLE

A narrative review: current upper limb prosthetic options and design

Lauren Trent, Michelle Intintoli, Pat Prigge, Chris Bollinger, Lisa Smurr Walters, Dan Conyers, John Miguelez and
Tiffany Ryan
Arm Dynamics, Redondo Beach, CA, USA

ABSTRACT ARTICLE HISTORY

Purpose: This review was conducted to provide an overview of current literature as it relates to upper Received 31 July 2018
limb difference, available componentry, and prosthetic options and design. Emerging technologies com- Accepted 8 March 2019
bined with an increased awareness of the limb difference community have contributed to recent
KEYWORDS
advancements in upper extremity prosthetics.
Upper limb; amputation;
Methods: A search of five major clinical databases utilizing keywords relating to upper limb prostheses, prosthesis; prosthetic
componenty and limb difference levels resulted in over 1200 articles. These articles were subjected to options; design;
inclusion and exclusion criteria in order to identify current peer reviewed research relevant to this topic. rehabilitation
Results: Fifty-five applicable articles and sources of standards were reviewed based on the inclusion and
exclusion criteria, presenting five general options for prosthetic intervention. This information was assimi-
lated and categorized in this article, which provides an overview of the aforementioned options.
Conclusion: While a noteworthy amount of research focuses on technological advancements, the five
options for prosthetic intervention are inherently represented in the current literature. For individuals
with upper limb difference, as well as their care team, successful rehabilitation hinges on awareness of
new components, the functional efficacy of these components, and the evolved techniques used in pros-
thetic design and fabrication. It is noted that the rapid evolution of upper limb prosthetics consistently
outpaces research and publication of information.

ä IMPLICATIONS FOR REHABILITATION

 To provide an overview of prosthetic design considerations and options to help create a more
informed rehabilitation team, leading to improved outcomes in prescription and management of
upper limb prosthetics.
 To bring awareness of current research in the field of upper limb prosthetics in order to provoke fur-
ther exploration of the efficacy of prosthetic options and design considerations.

Introduction from the current literature. Our intent is to provide a comprehen-

sive analysis of the current state of upper limb prosthetic science
Losing a limb, often due to a catastrophic event, can be a devas-
and its impact on the provision of prosthetic rehabilitation for
tating injury. Many people find that this loss is equivalent to los-
those pursuing functional upper limb (UL) prosthetic solutions.
ing a loved one and the grieving process may be similar
This review is relevant to manufacturers, engineers, research and
throughout recovery. In addition, physical rehabilitation is
development communities, device regulation and review entities,
required to restore useful and meaningful function. In the case of
prescribing physicians, prosthetists, occupational and physical
the congenital limb difference, while there is no grieving process,
therapists, and other allied health professionals involved in the
the lifelong effects of limb absence can be similar to that of an
rehabilitative care for this unique population.
amputation. The upper limb prosthetic rehabilitation team under-
stands these complexities and has the task of helping the individ-
ual achieve their goals through the appropriate recommendation
Methods
of prosthetic options along with other rehabilitative strategies.
Technology has advanced significantly, which has made prosthetic A systematic search of literature was conducted in May 2018 using
intervention viable for more levels of limb difference than were the following databases: PubMed, Medline Complete, CINAHL,
possible before. With this improvement of technology and fitting Academic Search Complete and Google Scholar. Keywords relating
methodologies, it is important to stay up to date with current to UL prostheses, componentry and limb difference levels were
practice in order to provide meaningful assistance to those in organized into search statements (Table 1). Search terms were
need of prosthetics rehabilitation. adapted primarily from those used within the VA/DoD Evidence-
This paper will provide an overview of upper limb differences Based Clinical Practice Guideline for the Management of Upper
and subsequently define the five prosthetic options available, Extremity Amputation Rehabilitation Search Terms, Appendix A [1].
their associated componentry, and the design principles reviewed Collectively, these results yielded over 1200 articles.

CONTACT Lauren Trent


ltrent@[Link] Arm Dynamics, 7000 W 121st Street, Overland Park, KS 66209, USA
All authors are members of the clinical team at Arm Dynamics, headquartered in Redondo Beach, California, USA.
ß 2019 Informa UK Limited, trading as Taylor & Francis Group
 CURRENT UL PROSTHETIC OPTIONS AND DESIGN 605

Table 1. Search Statements: PubMed, CINAHL, Medline, Academic Search Complete, Google Scholar.
Search terms 1 Search terms 2 Total results > unique reviewed articles
Amputation and (prosthesis or prosthetic Arm or finger or hand or thumb or wrist or transradial 484 > 13
or prostheses) or transcarpal or elbow or transhumeral or shoulder
or forequarter or upper limb or upper extremity
Prosthesis or prosthetic or prostheses Body-powered 78 > 16
Prosthesis or prosthetic or prostheses Myoelectric or electrically powered or exter- 484 > 15
nally powered
Prosthesis or prosthetic or prostheses Passive or cosmetic or restoration 158 > 3
Prosthesis or prosthetic or prostheses Activity specific 3>0
Prosthesis or prosthetic or prostheses Protocol or prescription or upper extremity or upper 37 > 0
limb or arm
Prosthesis or prosthetic or prostheses Component or terminal device or harness or socket 211 > 2
or frame

Specific inclusion and exclusion criteria were established to

select appropriate articles relevant to the purpose of this paper.
Given that this review intends to identify current literature, articles
published prior to 2008 were excluded. Furthermore, non-English
articles, experimental or investigational research, and surgical and
animal studies were excluded. The articles were required to be
published within the last 10 years, to be peer-reviewed, and to
pertain to the purpose of this paper by specifically referencing
advantages or disadvantages, individual requirements, available
devices, relevance to current practice, and/or fabrication processes
of upper limb (UL) prosthetic devices. A resulting 55 sources,
including the VA/DoD Evidence-Based Clinical Practice Guideline
mentioned previously, were deemed appropriate for this review
and all were utilized in this paper. All materials identified in the
initial search were then reviewed by certified prosthetists and
subject matter experts, for relevance to the purpose of the paper.
Digital copies of applicable reference material were obtained.

Overview of upper limb difference levels

While there may be variations in the literature, the generally
accepted nomenclature to delineate upper extremity limb differ-
ence is as follows: Interscapulothoracic, Glenohumeral (Shoulder)
Disarticulation, Transhumeral, Elbow Disarticulation, Transradial, Figure 1. Levels of amputation.
Styloid/Wrist Disarticulation, Transcarpal and Partial Hand
(Figure 1) [2]. Approximately 40% of all amputations are of the
activity/task specific [1]. It is important to note that there are mul-
UL, and the majority of these occur distal to the wrist [3]. In the
tiple manufacturers that develop products in these categories
paediatric population (e.g., ages 0–18), most UL differences have
a congenital aetiology, as opposed to acquired [4]. In the con- (Table 2). Discussing considerations regarding combining different
genital population, deficiencies are categorized as transverse, options into one prosthesis is beyond the scope of this paper.
intersegmental or longitudinal, with further subcategorizations. Knowledge of each manufacturer and their systems must be
The level of limb difference is the most fundamental principle taken into account when recommending a particular device as
when considering prosthetic options. not all components are compatible with each other when com-
bined together. Also, many of the options when combined
together may have components that require manual manipula-
Current prosthetic options tion, requiring activation from the opposite arm or pushing the
Many guidelines are in place for selecting an appropriate pros- prosthesis into or against something in the environment. These
thesis [1,5]. The factors associated with this decision include, but actions would be considered “passive” in nature because they are
are not limited to: level of limb difference, vocational/avocational not powered either by the affected body part, nor by a motor.
and recreational needs, desires and functional goals of the indi- One example of this would be a locking wrist. The wrist can be
vidual, residual limb integrity, work and home demands, level and positioned, but typically requires use of the other arm to manipu-
type of activities, as well as aesthetic priorities [1]. It is beyond late the lock and move the wrist into position and then lock
the scope of this article to discuss how to strategically request it again.
multiple devices in today’s health care setting, but it is important
for readers to be aware that one type of prosthesis is rarely suited
Passive prostheses
for all activities and settings, and more than one prosthetic option
is often necessary to meet the user’s complex functional needs Passive prostheses restore the anthropomorphic limb-length
vocationally and avocationally [5]. necessary to carry or stabilize an object during ipsilateral or bilat-
The available options are categorized by the following descrip- eral use, but do not actively move. A passive device can also pro-
tors: passive, body-powered, electrically powered, hybrid and vide opposition to the remaining digits, as is the case for those
 606 L. TRENT ET AL.

with partial hand presentations. Passive prostheses can be

Controllers,
Electric Powered Body Electrically Locking, Electrically Adaptive Batteries,
Misc.
described as static or positional.
Static devices have no moving parts, while positional designs

x
x

x
x

x
incorporate malleable armatures or ratcheting joints. Static passive
Positional Static Powered Powered Powered Utility PrehensorSDOFMDOF Flexion Rotator Rotator Powered Powered Passive Powered devices devices do not inherently provide active grasp, so the user has a
relatively limited grasping ability compared to positional passive

x
x
x
devices. This option is typically created with a highly customized
task in mind and for the most demanding of environments where
Shoulders

other options would have a higher tendency towards failure.

x
Powered,

When a user with partial hand limb difference requires a more

Body

rugged design, passive positional digits offer an industrially

designed articulation. Once positioned, passive fingers options
x
x

x such as those by manufacturers listed in Table 2, can resist forces

through detent-retained or locking joints. These joint mechanisms
Elbows

allow for restoration of a broad range of activities including the

x
x

potential of heavier duty activities. While there is scarce peer-

reviewed research available for these new systems, published case
x

studies are available by manufacturers that demonstrate the

Body

necessity of these devices.

Static and positional passive prostheses may also be designed
x
Wrists

to appear natural; however, their appearance is ancillary to their

function. Sculpted re-creations of the absent limb (Figure 2) pri-
x

x
x

marily serve to protect sensitive areas on the residual limb and to

restore limb length, which together, improves the functional use
x

x
x

of the affected side. The restoration gives the remaining or intact

digits a surface area to oppose for the light grasp of objects and
x

x
x

the material properties of flexible latex, rigid PVC or silicone can

Electrically
Powered
Terminal devices

enhance friction [6–15]. The natural appearance also helps to

restore the individual’s body image, reduce unwanted attention,
and aid in his/her psychological well-being [6]. This option can be
fit to any amputation level and, depending on presentation, may
not require harnessing.
x

Passive prosthetic options are light weight, require minimal

Passive Passive Body Electrically Body

componentry maintenance and have been shown to contribute to

x

psychological improvements for the user [16]. As found in the lit-

erature, there are benefits to the use of passive devices in gen-
eral. It is the experience of these authors that new passive
Partial Hand/fingers

positional designs take this already established benefit and

expand on it for additional patients. This extrapolation is not
experimental, but rather an expansion of previous work such that
x

more people can benefit from the established paradigm.

x

x
x
x

Body-powered prostheses
Table 2. Major manufacturers of ULP componentry, non-exhaustive.

Body-powered prostheses use a harness to capture proximal body

x
x

x
Alternative Prosthetic Services [Link]

motion, which produces excursion of a cable. The cable termi-

nates on a hook or hand and affects movement of the compo-
nents [10,12,17–22]. Body-powered prostheses (Figure 3) have
Partial Hand Solutions [Link]
Infinite Biomedical Technologies [Link]

been prescribed for those with UL differences for centuries, and a

considerable number of users continue to value the benefits this
College Park Industries [Link]

prosthesis type [23].

Midwest ProCAD [Link]

Body-powered prostheses offer several advantages over the

Mobius Bionics [Link]

Point Designs [Link]

TRS Prosthetics [Link]

Naked Prosthetics [Link]

Vincent Systems [Link]/en

MDOF: Multiple Degrees of Freedom.

alternative options. They typically are lighter weight, more dur-

COAPT [Link]

Texas Assistive Devices [Link]

able, more tolerant of environmental conditions (e.g., wet, dusty,

SDOF: Single Degree of Freedom.

etc.), provide secondary proprioceptive feedback to the user, have

Ottobock [Link]

RSL Steeper [Link]

a lower initial price point, and incurred maintenance costs are less
as compared to electrically powered options [10,12,17–21,23–27].
Ossur [Link]
Fillauer [Link]

Even though these devices are considered more durable and rug-
ged, it is expected with heavy use that repairs to the cable sys-
Manufacturer

tems, terminal devices, and other components are frequent and

necessary to maintain the individual’s functionality.
Despite the advantages of body-powered prostheses, there are
several disadvantages. Harnesses are restrictive in that they: limit
 CURRENT UL PROSTHETIC OPTIONS AND DESIGN 607

Figure 2. Passive Prosthesis, #Arm Dynamics. Photo permission obtained.

Figure 4. Body Powered Figure of 8 Harness, #Arm Dynamics. Photo permission

obtained.

components; these are intended as a brief overview and do not

divulge all options or details.

Harness/control
Successful control through a harness relies on four critical con-
cepts: a secure suspension of the prosthesis, a cabling system to
connect the harness and TD, adequate range of motion in the
body segment affecting the harness, and sufficient force to create
the excursion needed to actuate the components. The most com-
monly utilized harnessing system for transradial prostheses is the
figure-of-8, Northwestern style harness (Figure 4) [17]. It is com-
prised of an axilla loop around the contralateral shoulder; a sus-
pension strap over the ipsilateral shoulder that connects to an
inverted “Y” strap and triceps pad; and a control strap linking the
harness to the cable, which is then connected to the TD.
Variations of this design, including custom options, are applied
when conventional harnessing is insufficient to meet the comfort
or functional requirements of the user.

Terminal devices
Body-powered TDs come in either a hook or hand configuration.
Table 2 lists commonly utilized manufacturers of these terminal
Figure 3. Body Powered Prosthesis, #Arm Dynamics. Photo permis- devices. Body-powered hooks are generally preferred TD because,
sion obtained.
although body-powered hands appear more anatomical, they typ-
ically weigh more, have a preset and limited pinch force [23], are
available motion for function with the device, can exert significant less versatile for handling objects, and their shape can visually
forces on the residual limb, and can compress the contralateral obscure objects being grasped [29]. Hook configurations are fur-
axilla [28]. Users with more proximal levels of limb difference ther identified as either voluntary opening (VO) or voluntary clos-
have greater difficulty generating the necessary excursion to oper- ing (VC). VO TDs remain closed at rest from the tension of rubber
ate these devices, and some individuals dislike the appearance of bands or springs, and they only open with cable excursion. Pinch
the hooks, cable and harness [10,22–24,26,27]. The parameter sur- force is limited to the tension from the rubber bands or springs
rounding a user, where the prosthesis functions consistently, is that hold the TD closed. VC TDs remain open at rest and close
referred to as the “functional work envelope”. As a body-powered with cable excursion, therefore pinch force directly correlates to
prosthesis is moved further away from the body, it becomes more the amount of force exerted on the cable system, so some users
difficult for the user to produce excursion for device operation. can achieve enough force that the resulting grip exceeds that of
Grip force may be limited when compared to electrically powered electrically powered TDs [23]. Cable operation has been docu-
options, depending on the terminal device (TD) configuration. mented to provide proprioceptive feedback relative to the
While body-powered options have been historically viable for amount of pinch force being applied and, for an experienced
higher level presentations, there has been a recent introduction user, can indicate how wide the opening is on the TD [23]. Newly
of body-powered options for the partial hand which will be dis- available designs can interchange between VO and VC via a
cussed later. The following sections describe body-powered switch or change in cable excursion [19].
608 L. TRENT ET AL.

Wrist, elbow and shoulder components

A positive aspect of body-powered prostheses is the available
joint mechanisms for wrists, elbows and shoulders. Commonly uti-
lized manufacturers of these components are listed in Table 2.
Wrist components create additional degrees of freedom such
as flexion/extension and rotation for TD operation in various
planes of space [12,30,31]. Some options on the market for wrist
attachments include friction, quick disconnect, rotational, flexion
and multifunction units.
Available elbow components can be endoskeletal or exoskel-
etal. Depending on the design, humeral rotation of the forearm is
possible. Most versions lock, and some use friction to maintain
their position. Each of these options has its own indication and
function [12].
Shoulder components for body-powered prostheses currently
rely on passive positioning assisted by the sound limb or gravita-
tional prepositioning. Shoulders also have friction and locking
options. Shoulder locking is typically achieved by activating a
lever or nudge control with the contralateral limb or chin. These
are classified by the degrees of motion at the joint, either single-
axis or double-axis, which refer to shoulder flexion/extension and/
or abduction/adduction, respectively [12].

Partial finger/hand
Body-powered finger and hand prostheses have seen improve-
ments over the last decade. Three of the most prominent partial
finger prostheses utilize flexion of the residual finger to cause
flexion of the prosthesis. There may be a significant mechanical Figure 5. Myoelectric Prosthesis, Complex Hand, #Arm Dynamics. Photo permis-
sion obtained.
disadvantage because excursion is reduced over the surface of
the prosthesis as the finger flexes. Partial hand body-powered
prostheses utilize motions of the wrist to create excursion. The
most commonly utilized manufacturers of partial hand options
are listed in Table 2.

Electrically powered prostheses

Electrically powered prostheses utilize motors to effect movement,
which are powered by a rechargeable battery system. The most
common control method is through electromyography (EMG) sig-
nals, although there are other inputs available if these signals are
too weak. Other inputs include servos, linear potentiometers or
transducers, force-sensing resistors, rocker switches, push-button
switches and harness pull switches, as well as Inertial
Measurement Units (IMUs). There is also the option of utilizing
multiple control schemes in the same prosthesis. Electrically pow-
ered prosthetic component options range from a single digit
through hand, wrist elbow and shoulder componentry (Figure 5).
To successfully operate an electrically powered prosthesis, an
individual must be physically capable of operating an electrically
powered device, as well as have sufficient cognitive awareness to
understand and control the device. More proximal levels of limb
difference require more advanced and multifunctional prostheses,
which can impose a greater cognitive burden on the user [32]. Figure 6. Simple Myoelectric Hand, #Arm Dynamics. Photo permis-
sion obtained.
Advantages of electrically powered prostheses include, speed
and grip force that is both VO and VC as well as proportionally
controlled depending on the strength of the muscle contraction; hand and hook designs negate some limitations with operating
more anthropomorphic appearance; ability to control two compo- these devices in contraindicated environments.
nents simultaneously; higher grip force potential; and increased
functional work envelope.
Disadvantages of this prosthesis include, higher initial costs, Terminal devices
required battery maintenance, heavier weight, more complex TDs for electrically powered prostheses are broken down into sin-
repairs, and intolerance to wet, dirty or corrosive environments gle or multiple Degrees of Freedom (DOF) and anthropomorphic
[17,20]. Thankfully, recent advances in waterproofing and rugged or nonanthropomorphic prehensors [33].
 CURRENT UL PROSTHETIC OPTIONS AND DESIGN 609

Single DOF, anthropomorphic prehensors (Figure 6) operate Prosthesis control

using a single motor that provides a powerful and consistent tri- Electrically powered prostheses typically utilize noninvasive elec-
pod grasp. The motor articulates the metacarpophalangeal (MCP) trodes that read surface EMG signals from intact muscles in the
joints of digits one, two and three; the MCP joints of digits two residual limb. A small electrical signal is produced when a muscle
and three move in unison as the first digit abducts/adducts to contracts, and the external electrodes detect the electrical
oppose them. The fourth and fifth digits passively follow, and impulses from these muscle contractions. Determining the specific
there are no motion distal to the MCP and carpometacarpal muscle to use depends on the level of limb difference and the
(CMC) joints. Manufacturers of single DOF hands include those residual anatomy. Most electrodes filter and amplify the electrical
listed in Table 2. Most of these hands can be fit with locking or impulse and produce an output message to the prosthesis
friction-controlled flexion/extension wrist units. [17,18,33,35,38–41].
The single DOF, nonanthropomorphic prehensors are tradition- With multi-articulating prehensors, control strategies have
ally in the shape of a hook or gripper. Referred to as “utility pre- expanded to include specific myoelectric input triggers, Radio-
hensors”, these options were developed to be more durable and Frequency Identification (RFID) tags or chips, smart phone applica-
robust, to increase visual connection, and to provide a stronger tions, motion capture from gyroscopes or the previously men-
pinch; up to 36 pounds of force depending on the type of pre- tioned IMUs. New control methods include Pattern Recognition
hensor. Commonly used manufacturers of utility prehensors are and Targeted Muscle Reinnervation (TMR).
listed in Table 2. These utility prehensors can also be fit with lock- Pattern Recognition is a commercially available control system
ing or friction-controlled flexion/extension wrist units [12]. that uses an array of electrodes covering the entire residual limb
Multiple DOF, anthropomorphic prehensors (Figure 5) have to capture muscle contractions. The muscle signals are analyzed
digits that articulate over multiple joints within the fingers, and and assigned to a “pattern”, which is designated to a specific
some models have multiple motors to move fingers individually, movement and allows for more intuitive myoelectric control. This
thereby increasing the number of available functional grasp pat- promises a more intuitive control method akin to writing a docu-
terns to as many as thirty-six pre-programmed grasps. Having ment using voice recognition instead of manually typing on a
multiple grasp patterns, rather than a single tripod grip, can pre- keyboard. Pattern recognition software like COAPT# (Coapt, LLC.
sent advantages and challenges that depend on the user and his/ Chicago, IL) aims to ease the cognitive burden of switching
her adeptness with this technology. A user must be properly between components, and potentially can provide a newer, faster
trained to access these grasp patterns and gestures by triggering control scheme for controlling the advancing multi-articulating
the device, typically using a muscle trigger (specified myoelectric hands [42].
impulses). The most commonly utilized manufacturers of multiple Targeted Muscle Reinnervation (TMR) is a nerve-transfer sur-
DOF hands are listed in Table 2 [12,34–36]. gery that reassigns the remaining nerves after amputation and
replants them into an intact muscle. For those with transhumeral
Wrist, elbow, and shoulder components amputations, TMR transfers the remaining large brachial plexus
Utilization of prosthetic wrists, including rotation and flexion/ nerves and their motor fibres to viable muscle tissue in the
extension components, reduce compensatory movements that residual limb [43,44]. Like pattern recognition, TMR can allow for
may occur in the elbow, shoulder or torso [37]. Most prehensors a more intuitive control strategy with electric devices. A surface
connect using a quick-disconnect wrist that enables the exchange EMG signal will correspond to the previous nerve function once
of TDs. the transferred nerves and host muscle fibres are reinnervated.
Elbow components can be either electrically powered or body- Multiple nerves can be transferred so that more than two EMG
powered. The body-powered elbows use the same control mecha- sites can control the device. With more sites available, there is the
nisms as any other body-powered system, i.e., harness or gravita- potential for simultaneous control of multiple DOFs (e.g., flexing
tional control. If a body-powered elbow is configured for use with the elbow and closing the hand simultaneously) [43,44]. Without
a electrically powered TD, the prosthesis is then deemed a hybrid TMR, the nerve pathway would be incomplete, and the severed
prosthesis. Electrically powered elbows do not include any body- nerves would not generate useful muscle signals.
powered controls. Electrically powered and body-powered elbows To a similar end, the Starfish procedure has been developed to
have characteristics reflective of their general prosthetic classifica- salvage muscles from partial hand amputations and move them
tion; body-powered elbows are lighter, more durable and have a more proximally. Isolating each muscle not only facilitates more
lower initial cost as opposed to electrically powered elbows. intuitive myoelectric prosthetic control, but it also provides the
Notable advantages of electrically powered elbows are that they potential for controlling individual prosthetic digits [45].
can function in a larger work envelope, and they can offer live lift
functionality where body-powered elbows have virtually no live
Hybrid
lift capability. This significantly changes the function of the elbow
from being a preposition device (in the case of the body-powered Hybrid prostheses (Figure 7) combine two prosthetic options into
elbow) to a functional device (in the case of an electrically pow- one device. The most commonly used hybrid configuration is a
ered elbow). The most commonly utilized manufacturers of elec- body-powered elbow and electric powered hand/wrist for those
trically powered elbow components are listed in Table 2. who present with an above the elbow amputation.
There are a limited number of shoulder components available The advantages and disadvantages of a hybrid prosthesis are
across the industry. The most commonly utilized manufacturers of dependent on which components are powered and in what way.
shoulder components are listed in Table 2. Most shoulder compo- Generally, hybrid devices provide simultaneous control of multiple
nents are passive and function for electrically powered prostheses components, are lighter in weight than a fully electrically pow-
in the same capacity as they do for body-powered prostheses. ered prosthesis and offer increased grip force as compared to a
Prepositioning of the shoulder joint in the sagittal and coronal fully body-powered prosthesis. The disadvantages, as stated
planes can only occur passively, and the shoulders can only lock above, are that the body-powered elbow must be moved by grav-
with a manual locking mechanism in the sagittal plane. ity or a harness and cannot provide live lift. The necessity of a
610 L. TRENT ET AL.

eliminate the need of a harness for ancillary suspension. Clinically

relevant design styles include the Muenster, TRAC (Transradial
Anatomically Contoured), and Northwestern sockets for transradial
level presentations, as well as the Dynamic Socket for transhum-
eral level presentations.
Alley et al. [49] investigated the fabrication of a socket utilizing
alternating areas of compression and release. Alley reported that
despite being self-suspending, sockets may inhibit range of
motion of the residual limb, lose transmission of motion between
residual limb and prosthesis, and load the bone in localized areas.
This work found that the compression release socket (CRS)
achieved greater control of the underlying bone, which “offers
enhanced performance regarding stability, comfort, energy effi-
ciency, ROM and the perceived weight of the prosthesis” [49].
Razak et al. [47], explored an anthropomorphic socket design
that utilized pressure sensors to determine the required socket
size and fitting, and an oscillometric pump which varies the air
volume within the socket. The study revealed that although the
system requires improvement, it may pave the way for new
socket techniques in the near future.
Creative socket designs continue to evolve, and the most suc-
cessful ideas incorporate a strong clinical focus on comfort, sus-
pension, range of motion, and appropriate structural integrity to
house components and control elements that may be integral to
the socket itself, such as electrodes, access fenestrations and
anchor points for frameworks and components.
Figure 7. Hybrid Prosthesis, #Arm Dynamics. Photo permission obtained.

Future and developing technology

Options for prosthetic rehabilitation will continue to evolve which
may require the list of options represented herein to expand.
harness for body-powered component control can be difficult for
Several of the future and developing technologies are described
those with more proximal levels of limb difference.
for the purpose of awareness. Osseointegration is a surgical pro-
cedure developed in Europe circa 1950s as an alternative pros-
Activity/task specific prostheses thetic suspension strategy. The procedure includes either a one or
two-step surgical approach with an initial “fixture” implanted
Activity-specific prostheses are designed to facilitate tasks that
directly into the bone, proceeded by the attachment of a percu-
entail a higher level of performance, such as performing a push-
taneous abutment to the fixture. The primary benefit of osseointe-
up, holding a golf club, or throwing a ball [27]. Often these types
gration is the elimination of a conventional socket that can cause
of activities are contraindicated to perform with the primary pros-
skin breakdown, pain, discomfort and sweating. Other benefits
thesis because of safety or reliability concerns. The TD of an activ-
include the potential for unrestricted movement through minimal
ity-specific prosthesis can replicate the form of the human hand
to no harnessing, ease of donning/doffing the prosthesis, proprio-
during an activity or the tool used for an activity. There are mul- ceptive feedback and increased muscle mass. The potential disad-
tiple TD options, so this type of prosthesis is often designed with vantages include the risk of skin infection and the obligatory daily
a wrist unit that allows TDs to be quickly disconnected and inter- skin care around the abutment [50].
changed so that multiple activities can be accomplished. The use of implantable electrodes for prosthetic control is
becoming a popular area of research due to the advantages these
Upper limb socket design pose over surface EMG systems. With implantable systems, the
control of the prosthesis is not susceptible to body sweat, nor will
The socket interface is imperative for the function and success of it be affected by electrode displacement or poor socket fit. This
the prosthesis [46,47]. It serves to contain the residual limb, con- leads to a reduction of unintended movements and therefore
nect the person to the prosthesis and must be fitted securely and improved control when performing tasks above or below body
comfortably [48]. A variety of materials can be used, such as midline. One such current research system is the Implantable
thermoplastic or high consistency rubber (HCR) silicone [46]. The MyoElectric Sensor (IMES) project [51].
frame surrounds the socket and is typically matched to the length Haptic feedback has been an important topic for many years.
and circumference of the contralateral limb. The frame is typically The ability to replicate sensation has been trialled with a variety
rigid and sturdy and is used as a protective housing mechanism of approaches that include mechanotactile, electrotactile, vibrotac-
for the prosthetic components. tile and modality matched feedback [20,24,33,52]. A more recent
As electrically powered prostheses become more prevalent, and invasive approach allows for direct nerve stimulation by
socket designs have been adjusted to stabilize electrodes against wrapping electrode wires around the nerve or longitudinally plac-
the residual limb [49]. Electrically powered prostheses do not ing electrode wires on the nerves directly. The information from
require a harness for operation of the device, so these designs the tactile sensors placed on the prosthetic hand will then com-
have become self-suspending on the residual anatomy to municate directly to the nerve, allowing for somatosensory
 CURRENT UL PROSTHETIC OPTIONS AND DESIGN 611

stimulation of different peripheral receptors [33,52]. The Defense suggested clinical guideline, and some of their results were con-
Advanced Research Projects Agency (DARPA) Hand Proprioception flicting with other recently published studies [17]. Ayub also men-
and Touch Interfaces (HAPTIX) programme is advancing this tech- tions that “unfortunately, few studies have attempted to explore
nology, which allows for constant, permanent communication the effectiveness or optimize the operation of a Bowden cable-
between an upper limb prosthesis and the peripheral nervous sys- operated gripper” [17]. The body-powered prosthesis is one of the
tem of the user [53]. Further proprietary information regarding oldest and least variable mainstays of prosthetic care, and with
the DARPA and HAPTIX projects are available through their indi- minimal research attention in the last ten years, users could still
vidual programme sites. benefit from enhancing the control capabilities of these sys-
tems [27].
As innovation evolves, the issues of durability and complexity
Discussion
of electrically powered prostheses are being addressed. The com-
Function of the UL is far more complex than that of the lower ponentry and technology are comparatively newer than the body-
limb and often involves an open kinetic chain of movement to powered prostheses, and therefore more current research is avail-
perform an array of activities such as self-care, interaction with able; however, upon review of the available publications, most
the environment, interaction with others, self-expression, and focus on future and developing technologies, and few address
other fine and gross motor activities. Contributing factors to func- the current options for electrically powered prostheses. Those
tion are the ability to perform tasks correctly, quickly, rationally that do address the purpose of this paper are limited in
and resourcefully [54]. There is little consistency across the litera- their scope.
ture when defining the function as it specifically pertains to pros- Functional requirements of the patient and the process that
theses, and currently utilized measures of UL prosthetic function the rehabilitation team goes through to align the individual’s
are insufficient at doing so effectively and objectively [55]. requirements with the appropriate prosthetic option is beyond
Additionally, technological advances in electrically powered the scope of this review, however; that exploratory process is
components are rapid. By the time research is completed and equally important to understand as are the prosthetic options
results are published on the benefits of a specific powered com- available. Utilizing outcome measures to assess function specific
ponent, findings may be considered “outdated” as new genera- to this patient population will greatly improve understanding of
tions of a component are commercially available. The small effectiveness and aid in a proper prescription of prosthetic devi-
population size and heterogeneous variables across patients with ces [55]. Further, keeping pace with the changes in available tech-
upper limb difference create an added challenge to produce evi- nologies, fitting techniques and therapy protocols is a worthy
dence that is considered adequate to demonstrate the efficacy of effort for the rehabilitation team. Wang et al. [55] specifically
a device. Belter et al. [29] provided a systematic review on address outcome measures directly applicable to ULP
anthropomorphic prosthetic hands in 2013. Four of the hands rehabilitation.
reviewed (Vincent [Vincent Systems, Karlsruhe, Germany], i-limb
and i-limb Pulse [Touch Bionics by Ossur, € Foothill Ranch, CA],
Limitations
Bebionic and Bebionic v2 [Ottobock, Austin, TX]) have already
been updated with newer versions (i-limb Quantum [Touch Given the purpose of this review, specific inclusion and exclusion
€
Bionics by Ossur, Foothill Ranch, CA] and Bebionic v3[Ottobock, criteria were developed. This eliminated some studies (e.g., single
Austin, TX]) and new hands such as the TASKA (Fillauer, case studies, magazine articles) that, despite their value, could not
Chattanooga, TN) are now available with similar functions such be included as it would decrease the confidence in the informa-
that the review would apply to these updates. Even though the tion delivered herein. These outside publications and presenta-
technology has been updated, the category of prosthetic option tions demonstrate that a consensus of the effectiveness of these
that the change applies to has not and the value of those technologies and techniques is present in the field.
options remain. The literature search for this publication was limited in scope
Passive prostheses have developed over the years in terms of to more current publications. It was necessary to set this limita-
detail and durability. Passive devices designed for the purpose of tion since our purpose is to analyze current practice. There are
restoring limb length and surface area have been studied for the older references in literature, thus excluded because of their age,
psychosocial implications on the individual and have been found which show pertinent fittings and techniques that are still in prac-
to be beneficial [6]; however, the literature did not reveal any tice today but have been modernized because of materials and
studies regarding current updates to fabrication, fit and function. available componentry. These papers are therefore necessary for
Newly developed passive positional options are now available the body of knowledge but again were excluded because of
and case studies have shown the efficacy of these devices. The their age.
continual evolution to this category of prosthetic options with the
addition of new technologies does not make the category experi-
Conclusion
mental; on the contrary, the new options make the category
more valuable. A noteworthy amount of research focuses on the technological
The available published works for body-powered prostheses advancements of modern UL prosthetic componentry and control
vary in age and application. Due to the longevity of use of body- systems. The five general options for prosthetic intervention are
powered systems, most commonly cited research on these devi- inherently represented in the current literature. While some of
ces was conducted more than ten years ago, with some publica- these options are better developed than others, there is enough
tions dating back to the 1950s and 1960s [28]. One recent information to provide an overview of each prosthetic option.
publication attempted to define a global optimal setup with the Updated publications are outpaced by the rapid evolution of UL
current standard configuration for fitting a transradial body-pow- prosthetic rehabilitation; however, the value of this review is to
ered prosthesis, determined through robotic testing of prostheses equip the rehabilitation team with the fundamentals of limb dif-
[17]. Despite this group’s efforts, it was not feasible to define a ference and their respective prosthetic options. Advancements in
612 L. TRENT ET AL.

the field of prosthetics broaden the implementation of these [12] Kistenberg RS. Prosthetic choices for people with leg and
options so that more individuals can benefit from their arm amputations. Phys Med Rehabil Clin N Am. 2014;25:
application. 93–115.
[13] Kuret Z, Burger H, Vidmar G, et al. Impact of silicone pros-
thesis on hand function, grip power and grip-force tracking
Acknowledgements ability after finger amputation. Prosthet Orthot Int. 2016;40:
We would like to thank Kerstin Baun, Kimberly Konston, Nathan 744–750. Accessed May 22, 2018.
Kearns and Julian Wells for their assistance in writing and tech- [14] Maat B, Smit G, Plettenburg D, et al. Passive prosthetic
nical editing of the manuscript. hands and tools: a literature review. Prosthet Orthot Int.
2018;42:66–74.
[15] Reddy K, Bandela V, Bharati M, et al. Acrylic finger pros-
Disclosure statement thesis: a case report. J Clin Diagn Res. 2014;8:7–8.
The mention of commercial products, their sources, or their use in [16] Ahmad M, Balakrishnan D, Narayan A, et al. Comprehensive
connection with material reported herein is not to be construed rehabilitation of partially amputated index finger with sili-
as an actual or implied endorsement of such products by cone prosthesis: a case report with 3 years of follow up. J
Department of Health and Human Services. Indian Prosthodont Soc. 2014;14:222–226.
[17] Ayub R, Villarreal D, Gregg RD, et al. Evaluation of transra-
dial body-powered prostheses using a robotic simulator.
Funding Prosthet Orthot Int. 2017;41:194–200.
This work was sponsored by the Defence Advanced Research [18] Behrend C, Reizner W, Marchessault JA, et al. Update on
Projects Agency (DARPA) BTO under the auspices of Dr. Al advances in upper extremity prosthetics. J Hand Surg Am.
Emondi through the [Space and Naval Warfare Systems Centre, 2011;36:1711–1717.
Pacific OR DARPA Contracts Management Office] Grant/Contract [19] Berning K, Cohick S, Johnson R, et al. Comparison of body-
No. N66001-17C-4060. powered voluntary opening and voluntary closing prehen-
sor for activities of daily life. J Rehabil Res Dev. 2014;51:
253–261.
References [20] Brown JD, Kunz TS, Gardner D, et al. An empirical evalu-
ation of force feedback in body-powered prostheses. IEEE
[1] Department of Veteran Affairs, Department of Defense. VA/
Trans Neural Syst Rehabil Eng. 2017;25:215–226.
DoD Evidence-Based Clinical Practice Guideline for the [21] Gemmell KD, Leddy MT, Belter JT, et al. Investigation of a
Management of Upper Extremity Amputation passive capstan based grasp enhancement feature in a vol-
Rehabilitation. Washington D.C.; 2014. untary-closing prosthetic terminal device. Conf Proc IEEE
[2] ISO/TC 8549-4. Prosthetics and orthotics – vocabulary – part Eng Med Biol Soc. 2016;2016:5019–5025.
4: terms relating to limb amputation. International [22] Huinink L, Bouwsema H, Plettenburg D, et al. Learning to
Organization for Standardization. 2014. use a body-powered prosthesis: changes in functionality
[3] Ziegler-Graham K, MacKenzie EJ, Ephraim PL, et al. and kinematics. J Neuroeng Rehabil. 2016;13:1–12.
Estimating the prevalence of limb loss in the United States: [23] Smit G, Bongers R, Van der Sluis C, et al. Efficiency of vol-
2005 to 2050. Arch Phys Med Rehabil. 2008;89:422–429. untary opening hand and hook prosthetic devices: 24 years
[4] Egermann M, Kasten P, Thomsen M. Myoelectric hand pros- of development? Jrrd. 2012;49:523–534.
theses in very young children. Int Orthop. 2009;33: [24] Hichert M, Plettenburg D. Ipsilateral scapular cutaneous
1101–1105. anchor system: an alternative for the harness in body-pow-
[5] Melton D. Physiatrist perspective on upper-limb prosthetic ered upper-limb prostheses. Prosthet Orthot Int. 2018;
options: using practice guidelines to promote patient edu- 42(1):101-106.
cation in the selection and the prescription process. J [25] Sensinger J, Weir R. Modeling and preliminary testing
Prosthet Orthot. 2018;12:40–44. socket-residual limb interface stiffness of above-elbow
[6] Arazpour M, Mardani MA, Ahmadi Bani M, et al. Design prostheses. IEEE Trans Neural Syst Rehabil Eng. 2008;16:
and fabrication of a finger prosthesis based on a new 184–190.
method of suspension. Prosthet Orthot Int. 2013;37: [26] Hashim NA, Abd Razak NAB, Gholizadeh H, et al. Analysis
332–335. of voluntary opening Ottobock Hook and Hosmer Hook for
[7] Arazpour M, Mardani M, Bahramizadeh M, et al. The effect upper limb prosthetics: a preliminary study. Biomed Tech
of new method of suspension on quality of life, satisfac- (Berl). 2017;62:447–454.
tion, and suspension in patients with finger prostheses. [27] Carey SL, Lura DJ, Highsmith MJ, CP: FAAOP. Differences in
Prosthet Orthot Int. 2015;39:197–203. myoelectric and body-powered upper-limb prostheses: sys-
[8] Cabibihan J. Patient-specific prosthetic fingers by remote tematic literature review. J Rehabil Res Dev. 2015;52:
collaboration-a case study. Plos ONE. 2011;6:1–6. 247–262.
[9] Goyal A, Goel H. Prosthetic rehabilitation of a patient with [28] Smit G, Plettenburg D. Efficiency of voluntary closing hand
finger amputation using silicone material. Prosthet Orthot and hook prostheses. Prosthet Orthot Int. 2010;34:411–427.
Int. 2015;39:333–337. [29] Belter J, Segil J, Dollar A, et al. Mechanical design and per-
[10] Imbinto I, Peccia C, Controzzi M, et al. Treatment of the formance specifications of anthropomorphic prosthetic
partial hand amputation: an engineering perspective. IEEE hands: a review. Jrrd. 2013;50:599–618.
Rev Biomed Eng. 2016;9:32–48. [30] Abd Razak N, Abu Osman N, Wan Abas W. Kinematic com-
[11] Jacob PC, Shetty KH, Garg A, et al. Silicone finger pros- parison of the wrist movements that are possible with a
thesis. A clinical report. J Prosthodont. 2012;21:631–633. biomechatronics wrist prosthesis and a body-powered
 CURRENT UL PROSTHETIC OPTIONS AND DESIGN 613

prosthesis: a preliminary study. Disabil Rehabil Assist [43] Kuiken TA, Dumanian GA, Wilson C, et al. Targeted reinner-
Technol. 2013;8:255–260. vation – surgical outcomes of 30 Patients. Academy Annual
[31] Deijs M, Bongers R, Ringeling-van Leusen N, et al. Flexible Meeting-Journal of Proceedings (American Academy of
and static wrist units in upper limb prosthesis users: func- Orthotists & Prosthetists). 2010.
tionality scores, user satisfaction and compensatory move- [44] Kuiken T. Targeted muscle reinnervation: a promising new
ments. J Neuroeng Rehabil. 2016;13:26. approach to the prevention and treatment of painful
[32] Hancock L, Correia S, Ahern D, et al. Cognitive predictors amputation neuromas. Acad Today. 2016;12:10–12.
of skilled performance with an advanced upper limb multi- [45] Gaston R, Bracey J, Tait M, et al. A novel muscle transfer
function prosthesis: a preliminary analysis. Disabil Rehabil for independent digital control of a myoelectric prosthesis:
Assist Technol. 2017;12:504–511. the starfish procedure. J Hand Surg. 2018;44(2):163.e1-
[33] Saikia A, Mazumdar S, Sahai N, et al. Recent advancements 163.e5.
in prosthetic hand technology. J Med Eng Technol. 2016; [46] Lake C. The evolution of upper limb prosthetic socket
40:255–264. design. JPO. 2008;20:85–92.
[34] Atzori M, Mu €ller H. Control capabilities of myoelectric [47] Razak N, Gholizadeh H, Hasnan N, et al. An anthropo-
robotic prostheses by hand amputees: a scientific research morphic transhumeral prosthesis socket developed based
and market overview. Front Syst Neurosci. 2015;9:162. on an oscillometric pump and controlled by force-sensitive
[35] Kulkarni T, Uddanwadiker R. Overview: mechanism and resistor pressure signals. Biomedizinische Technik [Biomed
control of a prosthetic arm. Mol Cell Biomech. 2015;12: Eng]. 2017;62:49–55.
147–195. [48] Schofield J, Schoepp K, Williams H, et al. Characterization
[36] Luchetti M, Cutti A, Verni G, et al. Impact of Michelangelo of interfacial socket pressure in transhumeral prostheses: a
prosthetic hand: findings from a crossover longitudinal case series. Plos One. 2017;12:e0178517.
study. J Rehabil Res Dev. 2015;52:605–618. [49] Alley R, Williams T, Albuquerque M, et al. Prosthetic sockets
[37] Kyberd P, Lemaire E, Brookeshaw M, et al. Two-degree-of- stabilized by alternating areas of tissue compression and
freedom powered prosthetic wrist. Jrrd. 2011;48:609–617. release. Jrrd. 2011;48:679–696.
[38] Chadwell A, Kenney L, Thies S, et al. Corrigendum: the real- [50] Li Y, Brånemark R. Osseointegrated prostheses for rehabili-
ity of myoelectric prostheses: understanding what makes tation following amputation: the pioneering Swedish
these devices difficult for some users to control. Front model. Unfallchirurg. 2017;120:285–292.
Neurorobot. 2018;12:15. [51] Scheme E, Englehart K. Electromyogram pattern recogni-
[39] Head JS, Howard D, Hutchins SW, et al. The use of an tion for control of powered upper-limb prostheses: state of
adjustable electrode housing unit to compare electrode the art and challenges for clinical use. Jrrd. 2011;48:
alignment and contact variation with myoelectric pros- 643–659.
thesis functionality: a pilot study. Prosthet Orthot Int. 2016; [52] Hsiao SS, Fettiplace M, Darbandi B. Sensory feedback for
40:123–128. upper limb prostheses. Prog Brain Res. 2011;192:69–81.
[40] Segil J, Controzzi M, Weir R, et al. Comparative study of [53] Emondi A. Hand Proprioception and Touch Interfaces
state-of-the-art myoelectric controllers for multigrasp pros- (HAPTIX). Defense Advanced Research Projects Agency
thetic hands. J Rehabil Res Dev. 2014;51:1439–1454. Website; [cited January 2018]. Available from: [Link]
[41] Toledo C, Simon A, Mun ~oz R, et al. A comparison of direct [Link]/program/hand-proprioception-and-touch-interfaces
and pattern recognition control for a two degree-of-free- [54] Bongers RM, Kyberd PJ, Bouwsema H, et al. Bernstein’s lev-
dom above elbow virtual prosthesis. Conf Proc IEEE Eng els of construction of movements applied to upper limb
Med Biol Soc. 2012;2012:4332–4335. prosthetics. J Prosthet Orthot. 2012;24:67.
[42] Simon AM, Lock BA, Stubblefield KA. Patient training for [55] Wang S, Hsu J, Trent L, et al. Evaluation of performance-
functional use of pattern recognition–controlled prosthe- based outcome measures for the upper limb: a compre-
ses. J Prosthet Orthot. 2012;24:56–64. hensive narrative review. Pm&R. 2018;10:951–962.e3.