American Academy of Orthotists & Prosthetists

Biomechanical Discussion of Current and Emergent Upper-Limb Prosthetic Interface Designs

Randall D. Alley, BSc, CP, LP, FAAOP, CFT

Abstract

Traditional functional assessment of upper-limb prosthetic patients as it relates to some measure of operational and functional performance pays scant attention to isolating and assessing interface function. Instead, a predominant focus is brought to bear on the prosthetic system as a whole, or to associated components extrinsic to the interface. As we gain a better understanding of biomechanics and how its proper application is imperative for optimizing functional interface design, we begin to more fully appreciate the interface's primary role in determining prosthetic acceptance and outcomes. With this knowledge comes an ever-increasing realization that optimizing the interface boundary—and the resulting impact on underlying anatomy and physiology—in order to maximize interface function is a core component of any well-designed prosthetic or orthotic system.

The Interface

The primary roles of lower-limb interface design are considered to be load transmission, stability, and efficient control for mobility.¹ When talking about upper-limb prosthetic design, it may serve better to describe mobility control in terms of positional, operational, and functional control.

Load Distribution and Transmission

Distribution and transmission of an applied load is extremely important in both lower- and upper-limb interface design. The basic principles of current load-distribution models revolve around two simple concepts: uniform distribution of load about the entire limb, and concentration of load on load-tolerant areas of the limb, with concomitant relief for those areas deemed "load-sensitive." An alternative load-distribution model, known as the "High-Fidelity" or "Compression-Stabilized" interface, which was developed by the author, is a significant departure from current interface design protocol. It involves achieving additional skeletal control through targeted soft-tissue relief. A brief discussion of the principles behind it will be presented at the end of this paper.

Other than simple containment of the limb, early lower-limb interface models exhibited little biomechanical basis for their particular designs. This holds true for many early and current upper-limb interface designs as well. These initial encapsulatory "sockets" focus on soft-tissue containment without a complete understanding of the associated anatomy or the biomechanical principles involved. Current designs, such as the Muenster², introduced in the early 1950s, and the Northwestern socket³, first written about in the early 1970s, take an arguably simplistic approach when invoking biomechanical principles into their composition. This has been discussed previously in greater detail.^4,5 Current interfaces at all levels of limb loss, or interfaces that focus on enclosing a limb segment for control and function, rely mainly on hydrostatic pressure and anatomical contouring to achieve their goals. The Muenster and Northwestern designs are self-suspending and use different techniques proximal to the cubital fold that distinguish them from one another and from their predecessors. Whereas the Muenster design reduces the anteroposterior (AP) dimension in order to apply an increased load to the cubital region as well as to the area just proximal to the triceps tendon insertion, the Northwestern design applies a concentrated medial-lateral (ML) pressure just proximal to the humeral epicondyles.

It is the reliance on proximal anatomical control—inherent in these and nearly all upper- and lower-limb prosthetic interface designs to date—in addition to a limited understanding of global hydrostatic pressure and its role as a dampener, that renders them insufficient in maximizing overall system performance.

Intrinsic load distribution for improved comfort is but one important characteristic of an upper- or lower-limb interface. Another related principle is the use of load transmission to supply feedback, or kinesthesia. The ability to sense the orientation and direction of an external force is crucial to prosthetic control. If properly applied, the more intimate the interface, the more information that can be transferred to the wearer. Long are the discussions about the value of the control cable in providing feedback in cable-driven upper-limb prostheses, but conspicuously absent in the literature are discussions surrounding the critical nature of upper- and lower-limb interface-derived feedback or even the existence of such a concept.

Figure 1: Intrinsic bone motion in traditional interface rendering.

Yet another related principle involves the transmission of forces originating from the user to the interface and to the prosthetic system as a whole. Volitional control of a prosthetic system is greatly affected by the ability of the interface to efficiently, effectively, and accurately transmit these forces. By its nature, a simple encapsulatory design does not possess the capability to perform this function efficiently. The delay between volitional human movement and prosthetic movement caused by motion of the interface about the limb prior to system engagement is an inherent feature of all existing interface designs. Much of this delay is due to the time it takes for the soft tissue between the intrinsic bone and the prosthetic interface to compress to the point of realizing interface response of sufficient magnitude to effect movement (figure 1). And while efficient transmission of forces relies upon intimacy of fit, proper anatomical contouring, and biomechanical control, they in turn influence another important aspect of interface design: stability.

Stability

A properly designed upper-limb interface should optimize stability. This means creating a balance between exerting maximum stability and allowing for the appropriate level of intended interface and prosthetic-system mobility.

Figure 2: Translation axial rotation slip.

Stability has several facets. The most important considerations in upper-limb interface design pertain to axial rotation, slip, and translation. Axial rotation can pertain to rotation of either the soft tissues or the interface itself about the long axis of the primary bone in radial and humeral designs or about the thorax in the transverse plane in thoracic designs. Slip of the soft tissues occurs intrinsically (within the volume of the interface) when applied forces are of sufficient magnitude to overcome the frictional force at the interfacial or human-interface boundary. Finally, translation will be defined herein as any gross movement—excluding axial rotation as defined above—of the interface about the limb segment in radial and humeral designs and in relation to the thorax in thoracic designs (figure 2).

Axial Rotation

A properly designed upper-limb interface should limit unwanted axial rotation of both the soft tissues and the interface itself about the intrinsic limb's long axis. The biomechanical aspects of the interface involved in controlling both types of rotation are intrinsic anatomical and biomechanical contouring and perimeter design. The simple shape of early and traditional sockets allows for significant axial rotation of the soft tissues, especially in response to externally applied torque perpendicular or tangential to the long axis of the enclosed bone segment. In addition, these sockets' perimeter shape is simple and rarely extends up to and beyond the cubital fold. The Muenster and Northwestern designs attempt to limit this rotation at the radial level by bringing the proximal edge of the interface over the humeral epicondyles. The Anatomically Contoured and Controlled Interface (ACCI), introduced by the author in the early 1990s, controls axial rotation by extending the medial and lateral stabilizers more proximally than in earlier designs and adding supracubital fossae anterior and proximal to the humeral epicondyles, avoiding the often-reported discomfort in earlier designs that suspended directly proximal to the epicondyles.⁶ Where the traditional interfaces and the ACCI differ is in how well the interfaces perform under heavy loading, particularly in the initial stages of flexion. The Muenster and Northwestern designs possess a critical flaw in that they perform poorly under heavy loading, as the focus is on the proximal portion of the socket and does not adequately spread loads throughout the mid and distal portions of the interface due to an overreliance on hydrostatic stabilization.

This places an undue burden on the distal end of the radius, increasing discomfort and reducing function. The ACCI adds radial channels on either side of the antecubital area and along the length of the radius, which allows for greater surface-area contact during the crucial stages of early and mid-flexion. This, in turn, allows the user to lift heavier loads more comfortably without restricting flexion and increases rotational stability due to its "capture" of the radius.

At the humeral level, early and traditional designs do little to limit axial rotation of soft tissue and interface rotation about the axis of the humerus, relying instead on auxiliary straps to provide this function. Patients are subjected to excessive harness pressures in the axilla and elsewhere due to prosthetists' attempts to adequately control rotation by increasing the forces exerted by auxiliary strapping. The "Dynamic Socket," introduced by Tom Andrews, CP, in the late 1980s, finally achieved interface control of axial rotation by relying on the interface rather than on ancillary straps to control unwanted axial rotation of the soft tissues about the humerus.⁷

Figure 3: XFrame medial perspective.

At the thoracic level, there is an inherent resistance of the soft tissues to axial rotation of significant magnitude; however, a measurable amount can and does occur. Soft-tissue rotation about the thorax is largely absent in the encapsulatory designs still in use today, as these designs seldom have the degree of overall intimacy to generate the interfacial friction necessary to impart global soft-tissue rotation about the thorax of any appreciable magnitude. Commonly, slip occurs, and the interface exhibits significant instability, rotating about the thorax until edge pressure at the perimeter or contact pressure elsewhere builds to sufficient resistance levels to counteract it. In early and traditional designs of thoracic interfaces, axial rotation is only marginally controlled by these anteromedial and posteromedial edges. In response to this discomfort, to the inherent instability of encapsulatory designs, and to excessive heat retention, the XFrame was introduced in the early 1990s, also by the author.⁴ This thoracic interface utilizes a technique similar to those in advanced humeral designs, namely the use of an outrigger principle to aggressively control axial rotation. Stabilizers or paddles are placed at each of the four corners comprising the anterosuperior, posterosuperior, anteroinferior, and posteroinferior boundaries, and because of this, unwanted interface rotation in the transverse plane is mitigated (figure 3).

Slip

In all upper-limb interface designs, some amount of slip occurs. The challenge, biomechanically and physiologically, is in designing an interface that allows the proper amount of slip to occur. Excessive slip can result in significant stability and suspension difficulties, requiring additional loading in other areas of the interface to counteract this. Excessive localized control of slip can result in interfacial issues arising from physiological responses to friction and shear.

In the early and traditional radial interface designs—as is the case in nearly every interface design conceived to this point—slip was largely controlled by either volumetric reduction (increasing global skin-interface friction), or by increasing localized pressures in critical areas (often in sensitive, minimal-load-bearing regions).

In the early and traditional humeral interface designs, slip was controlled in similar ways. Reduced volume of the interface in order to increase surface friction was common, as was increased pressure over sensitive regions, such as the acromion, the acromioclavicular joint, the clavicle, and spine of the scapula of the shoulder complex, as a result of encompassing the shoulder. In more advanced humeral designs, volume reduction is also used, but more focus is placed on distributing pressure over the pectoralis area, which improves both suspension and slip while avoiding encapsulation of the shoulder complex.

Early and traditional thoracic designs exhibit an inordinate amount of slip. Slip occurrs in response to a static encapsulatory interface being unable to conform to a dynamic anatomical landscape caused by posturing, joint motion, volume fluctuations, etc. With the advent of the XFrame, the encapsulation method was discarded in favor of a reduced-profile design that covers a smaller surface area and avoids contact with the shoulder joint as well as the lateral aspects of the scapular spine and trapezius muscle, which allows the XFrame to remain stable on the thorax throughout. By compressing the pectoralis area in a method similar to that described above, slip is further prevented.

Interface material also plays a major part in slip control. The balance between allowing slip and controlling slip can be skewed one way or another by the friction properties of materials commonly used, or more advanced materials not often considered. In current radial and humeral levels, a compromise must be reached between the donning and doffing effort as well as interface suspension. Due to the encapsulatory nature of traditional interfaces designed for these levels, friction must be carefully considered due to the effort required when donning or doffing. In the thoracic levels utilizing the XFrame design, higher levels of friction may be employed in order to achieve greater stability and control with minimal regard to application and removal.

Interface Translation

Interface translation often involves friction and shear but can also be the result of improper contouring and inaccurate volume control. Interface translation as defined previously is as any motion of the interface—other than axial rotation—relative to the skeletal structure of the limb (in radial and humeral designs) or, at the thoracic level, relative to the skeletal structure of the thorax.

Translation of the interface is responsible for many of the inefficiencies observed in transfer of the user's input to prosthetic output. Much of this translation occurs through soft-tissue compression. The firmness of the underlying soft tissue will determine the rate and magnitude of compression that occurs in response to an applied normal or tangential force. At some point, resistance to compression will build to a level that begins to exert an equal and opposite force to that which is being applied, and the interface, if unrestricted, will respond by moving in the direction of the applied force.

An applied external or internal force can cause significant translation of the interface until adequate resistance or counterforce is met. This translation can have profound effects on user comfort and prosthetic control as efficiency plummets and function suffers. In traditional encapsulatory designs, both volumetric and perimeter intimacy, together with correct anatomical and biomechanical contouring, are necessary to minimize this type of interface translation.

In traditional radial- and humeral-level designs, their encapsulatory nature demands a high level of volumetric intimacy. This feature, along with their proximity to—and in most cases their encapsulation of—either the elbow or shoulder joint, exposes them to a wide range of motion-related instabilities. In addition, significant cross-sectional changes occur at the radial level during flexion and extension. The ensuing variation in surface topography and hydrostatic pressure creates opportunities for intrinsic instability.

In current radial interface designs, translation is prevented namely via global compression of the soft tissues and increased localized pressure generated by contact of the soft tissue with the perimeter and what can be thought of as the proximal ring of the socket. When compression resistance of the soft tissue reaches a level that exerts enough force on the anteroproximal brim, intrinsic slip (secondary to the generation of friction and shear forces) and interface translation in the distal direction takes place. This translation often occurs to such a degree that it partially or completely displaces the interface from the limb. This partial displacement is often used as a basis for determining what is considered full and functional flexion.

While it indeed is functional for many users to achieve such levels of flexion despite their limb partially ejecting from the interface, it is important to note that this is an unintended consequence of poor interface design. The most frequent solution to this problem is to reduce the proximal extent of the brim in this area. While this frequently solves the interface-translation issue, it reduces the amount of viable load-bearing surface area, often resulting in user discomfort (figure 4). While it is true that all radial interface designs to date are subject to some intrinsic instability due to forearm dynamics, traditional self-suspending as well as more advanced interfaces also must contend with motion-related and anatomy-related instabilities secondary to joint proximity and/or encapsulation. Proximity to and encapsulation of bony prominences, especially those areas that move or translate relative to the interface during motion or whose surface topography alters, for example, create an additional level of potential instability that must be accounted for during the design phase. One solution to this dilemma is simply to modify the rigid areas of the interface such that they do not encompass the epicondyles or olecranon and instead provide either a flexible encapsulation of this area or eliminate encapsulation of this area altogether, as is seen in Sauter's "three-quarter" socket and in more recent developments that borrow from Sauter's original ideas.

Figure 4: Radial lifting surface.

The problem of instability caused by antecubital tissue bunching and compression is addressed in a unique way in the ACCI design. The anteroproximal brim extends up and into the cubital fold, with a relief allowed for the expanding volume of antecubital tissue during flexion. Rather than leveraging the interface distally as the antecubital tissue amasses against the anteroproximal brim, it simply fills in the relief gradually as elbow flexion angle increases, avoiding the buildup of contact pressure commonly responsible for interface translation.

In early humeral designs, translation was a significant problem, due in part to simplistic anatomical contouring distal to the axilla, and as well to the encapsulation of the glenohumeral joint. If large forces were applied, it was common for the interface to exhibit severe instability. Often the harness—in addition to its suspensory function—was chiefly responsible for providing the necessary stability in these early designs. Harness straps were frequently cinched tightly, resulting in some discomfort in the contralateral axilla if it was used as an anchor point, and in some cases, the cinching resulted in nerve-entrapment syndrome. Because the glenohumeral joint was typically encapsulated, its movements would significantly displace the interface, causing large gaps to be created at the perimeter and resulting in excessive contact pressures on bony prominences and other sensitive areas.

The Dynamic Socket and its successors avoid encapsulation of the glenohumeral joint, and the lateral trim line is often brought as low as the insertion of the middle deltoid. Movements of the glenohumeral joint do not typically displace the interface as they do in earlier designs, but rather allow positional control while maintaining intimate interface contact with the limb. One of the key features limiting interface translation in this design is the addition of anterior and posterior stabilizers to the proximomedial aspect of the interface perimeter. Compression of the pectoralis and scapular areas by these stabilizers generate significant resistance to distal migration of the interface, as well as increased frictional forces at the skin-socket interface, limiting interface translation.

Finally, the medial-lateral (ML) dimension is compressed along the medial and lateral aspects of the limb in order to increase frictional forces and allow room for biceps and triceps hypertrophy and shape changes during muscle shortening and physiological development. In later designs by the author and others, compression of the pectoralis muscle in a more localized area, as close to the center of rotation of shoulder abduction as possible without impinging on the head of humerus, provides a more stable result. In addition, the increased AP of the Dynamic Socket allows excessive motion of the humerus within the interface in flexion and extension and is considered a detriment to cosmesis. Later designs tend to minimize ML compression and AP expansion for this reason.

In early and traditional thoracic designs, encapsulation of the glenohumeral joint causes problems with interface translation that are similar to those found in the early and traditional humeral designs described above. Large displacements of the interface are common during shoulder movement, and the resulting instability is a significant problem due to the heavier total weight of prosthetic systems typically encountered at this level of amputation or amelia, as well as to a loss of contact with surface electrodes when, for example, a myoelectric control strategy is used.

Finally, encapsulation of the thorax in early and traditional thoracic designs results in interface translation whenever the spine is articulated or the scapulae are ranged. The varying thoracic surface topography during these movements induces intrinsic instabilities within the closed volume, having a negative effect on interface stability.

Because the XFrame does not incorporate the glenohumeral joint within its perimeter and because at its superior aspect the frame's anterior stabilizer applies a concentration of contact pressure at or near the shoulder's center of rotation in the frontal plane, very little interface translation occurs during shoulder ranging. In addition, this interface is not a closed volume and can be described as being discontinuous, so surplus translation is not significant during scapular and/or spinal articulation. Finally, the XFrame adds a pair of stabilizers to the inferior aspect of its anterior and posterior borders, which, in addition to counteracting axial rotation and distal migration, play an important role in reducing interface translation by increasing frictional forces at the interfacial boundary.

Positional Control

Positional control of the prosthesis most often begins with the transfer of human input to the interface. Whether this input is extrinsically derived, as in a positioning task requiring gross body movement, or whether it is intrinsically derived, as in radioulnar supination or pronation, the passing of this "information" and/or energy must be efficiently executed in order to optimize system or component activation. In both cases, an interface must maintain adequate intimacy and biomechanical control in order to ensure maximum functional efficiency or responsiveness.

In early radial designs, their simplistic conical shapes rely on auxiliary suspension and volumetric efficiency to effect positional control, and the designs' functional efficiency is extremely sensitive to limb length. In the Muenster and Northwestern interfaces, their self-suspending design and specific anatomical contouring enables the proximal brim to exert more influence and impart an added measure of positional control. However, lacking significant contouring distal to the proximal third of the interface, these designs still rely chiefly on the intimacy of the interface volume for positional control.

The ACCI adds radial channels to the distal volume, which offers additional stability as well as an extended brim up into the cubital fold. By imparting additional contouring proximal to this region, it achieves improved positional control over that of its predecessors. While traditional interfaces place excessive pressure on the distal aspect of the limb under load, the radial channels of the ACCI not only aid rotational stability, but also ensure sufficient soft-tissue contact along the length of the entire limb throughout all ranges of motion, reducing distal-end discomfort (figure 5).

Figure 5: Fully labeled ACCI showing radial channel.

Positional control in early and traditional humeral designs relies on a combination of auxiliary suspension, glenohumeral encapsulation, and global compression. In extreme ranges of glenohumeral motion, the efficiency of the auxiliary suspension degrades significantly. In addition, any control afforded the user via encapsulation of the shoulder joint diminishes considerably because at extreme ranges of adduction and flexion, it often no longer remains within the confines of the interface. The user generally has to rely on the intimacy of the interface's distal volume to provide any measure of stability, and hence positional control.

Figure 6: AHI versus traditional overlay biomechanical.

In the Dynamic Socket, the anterior and posterior stabilizers add greatly to positional control throughout even extreme ranges of glenohumeral joint range of motion. While soft-tissue compression within the distal volume adds a measure of stability, the stabilizers increase their role in maintaining positional control as limb length, and hence lever arm, are reduced (figure 6).

At the thoracic level, positional control of the interface is not typically considered a crucial factor in interface design. Users typically rely on either active or passive movement of prosthetic components rather than on glenohumeral joint excursion, with the exception of some fully cable-driven and hybrid systems. While much of this control is effectively borne by contact of the glenohumeral complex (typically the humeral head and surrounding soft tissue) against the inner wall of the socket in encapsulatory designs, this increased mobility has a concomitantly negative effect on overall stability.

The XFrame must be modified if it is intended to be used in partially or fully cable-driven form. The anterior stabilizer is brought more superior to engage the humeral head, if present, and its associated soft tissues in order to capture maximum motion derived from scapular protraction. The XFrame was designed to be extremely stable under all conditions; hence, its unmodified version avoided contact with the mobile glenohumeral joint. The frame remains relatively motionless throughout glenohumeral joint ranging without the humeral-head modification although scapular protraction does allow some voluntary axial rotation of the interface about the thorax for improved positioning of the prehensors, which is especially important in bilateral applications as the user attempts to reach his or her midline.

Discussion

In summarizing the general biomechanical principles of proper upper-limb interface design, it is important to note that an efficient interface will optimize load transmission and stability, while maximizing positional control. In other words, anatomical contouring and biomechanical control must regard optimal locations for the application of contact pressures as well as the overall surface area upon which contact pressure is applied. It is imperative that pressure distribution is not necessarily maximized but optimized, as slip—often the result of poorly managed friction and shear forces—must be carefully modulated to achieve the desired result.8 In addition, load transmission, axial rotation, and translation must all be modulated carefully to ensure superior function and comfort while imparting maximal stability and capture of volitional movement.

The enclosure of an individual's limb for the express intent of capturing movement, load transmission, providing suspension, limb protection, and stability, as well as providing an attachment for prosthetic components, is traditionally achieved through the use of an encapsulatory or closed-volume socket designed to contain the soft tissue. Improvements to these earlier solid-body designs have resulted in the incorporation of a framestyle interface consisting of a rigid frame and a semiflexible thermoplastic inner liner. Though this inner liner flexes to some degree where it is not backed by a rigid frame element, it still encapsulates the intrinsic limb and significantly restricts soft-tissue flow, thus maintaining hydrostatic pressure and providing a global compressive force to the limb's associated soft tissue.

High-Fidelity Interface

Over the last 50 years, despite significant advances in interface design, there is a common thread that links them all: with the exception of the ACCI, very little attention has been paid to the soft tissues and skeletal anatomy distal to the proximal region of the interface. While it is not possible to detail all the biomechanical and physiological elements involved within the scope of this article, the author will briefly discuss the fundamental principle inherent in an emergent and highly successful design that incorporates not only anatomical contouring (which by its very definition is present in all interface designs that encapsulate the limb), but optimizes biomechanical control through the use of much more significant compression forces strategically placed upon the intrinsic bone of the target limb. This highly advanced design concept forms the basis for the interface utilized in key military research-and-development programs centering around upper-limb prosthetics, with which the author is involved.

Because hydrostatic pressure in a traditional encapsulatory interface restricts the amount of localized and targeted compression on the intrinsic bone that can be attained within the interface, there exists a significant amount of compressible soft tissue between the skeletal structure and the interfacial boundary. Because soft-tissue compression in this area between the interface wall and the skeletal structure is sub-optimal, significant skeletal movement occurs prior to attaining sufficient soft-tissue compression at the interfacial boundary to initiate interface response. This redundant motion decreases prosthesis stability, the wearer's positional precision, functional range of motion, and overall efficiency of movement, thus increasing energy expenditure while concurrently increasing the perceived weight of the prosthesis.

Figure 7: High Fidelity Radial Interfaces.

The preceeding issues are overcome with the application of a new fundamental principle in interface design: the patent-pending "High-Fidelity" or "Compression-Stabilized" Interface. This design principle has been applied to both upper- and lower-limb prosthetic and orthotic applications and functions to provide superior interface response and control by capturing intrinsic bone motion earlier in the volitional motion cycle (response) while reducing avolitional and/or redundant skeletal motion when acted on by an outside force (control). Discontinuous regions of low or no compression and high compression are oriented in such a way as to allow optimum soft-tissue flow away from targeted areas of high compression (figure 7).

The crucial paradigm shift in interface theory is this: by allowing specific soft tissue to flow out of the interface (or into an appropriate relief area in a solid-body configuration) with minimal constraint, it is possible to apply optimal compression at the interfacial boundary far beyond what is possible in encapsulating designs while maintaining superior comfort. This increased localized soft-tissue compression decreases the effective distance the intrinsic skeletal anatomy must travel before opposing forces reach a critical level and interface response, sufficient to capture or control movement, occurs.

An additional benefit of the High-Fidelity design for upperlimb wearers is the significant heat dissipation due to the exposed soft tissue flowing out of the interface. As new technologies are being developed for lower-limb interface materials, heat dissipation without inducing severe window edema is on the horizon. Currently, blood pressure volume monitoring, skin temperature, and other instruments are utilized to ensure long-term viability of tissues under optimum compression and appropriate management of the effects of window edema.

At the time of this writing, a number of patients with varying levels of amputation/amelia as well as intact limb involvement of both the upper and lower limbs have been fit with the High-Fidelity Interface, and clinical trials involving the patent-pending compression-stabilized concept are underway through several government contracts. Results will be published in subsequent papers.

Acknowledgments

The author would like to thank the pioneers in prosthetics—past, present, and yet to come—for their refusal to follow the crowd simply because there is one. The patient and clinician communities are well served by those who question, those who challenge, and those who set about to change the rules in the quest for excellence.

References

1. Mak AF, Zhang M, Boone DA. State-of-the-art research in lower-limb prosthetic biomechanics-socket interface. Journal of Rehabilitation Research and Development. 2001;38(2).

2. Fishman S, Kay HW. The Munster-type below elbow socket, an evaluation. Artificial Limbs. 1964;8:4–14.

3. Billock JN. The Northwestern University supracondylar suspension technique for below elbow amputation. Orthotics and Prosthetics. 1972:16–23.

4. Alley RD, Sears HH. Powered upper limb prosthetics in adults. In: Muzumdar A, ed.: Powered Upper Limb Prostheses: Control, Implementation and Critical Application. Berlin, Germany: Springer; 2004;134–5.

5. Alley RD. Advancement of upper extremity prosthetic interface and frame designs. Paper presented at: American Orthotic & Prosthetic Association National Assembly; October 24–27, 2001; Phoenix, AZ.

6. Alley RD. Advancement of upper extremity prosthetic interface and frame design. Paper presented at: University of New Brunswick's Myoelectric Controls/Powered Prosthetics Symposium; 2002; Fredericton, NB, Canada.

7. Andrew JT. Elbow disarticulation and transhumeral amputation: prosthetic principals. In: Bowker JH, Michael JW, eds. Atlas of Limb Prosthetics, 2nd ed. St. Louis, MO; Mosby Year Book; 1992;255–264.

8. Zhang M, Turner-Smith AR, Tanner A, Roberts VC. Frictional action at residual limb/prosthetic socket interface. Med Eng Phys. 1996;18(3):207–14.

Randall Alley, BSc, CP, LP, FAAOP, CFT, is the CEO and chief prosthetist for biodesigns inc., Santa Monica, California. Alley is the former director of the world's largest upperlimb prosthetic program, is an adjunct professor for the O&P program at California State University Dominguez Hills, and is the chair of the Academy's CAD/CAM society. He is an international lecturer, clinical columnist, and has contributed to numerous prosthetics textbooks. He received a Certificate of Appreciation from the Department of the Army for his upper-limb training of military medical personnel as well as the Chairman's Award and the Excellence Award in Innovation for his contributions to the prosthetic field. He can be contacted by e-mail at

Earn PCE Credits from this article

Take the quiz at the Academy's Online Learning Center.