Componentry for Lower Extremity Prostheses Abstract Prosthetic components for both transtibial and transfemoral amputations are available for patients of every level of ambulation. Most current suspension systems, knees, foot/ankle assemblies, and shock absorbers use endoskeletal construction that emphasizes total contact and weight distribution between bony structures and soft tissues. Different components offer varying benefits to energy expenditure, activity level, balance, and proprioception. Less dynamic ambulators may use fixed-cadence knees and non–dynamic response feet; higher functioning walkers benefit from dynamic response feet and variable-cadence knees. In addition, specific considerations must be kept in mind when fitting a patient with peripheral vascular disease or diabetes. W ith the advent of new materi- als, designs, and technologic advances, the field of lower extrem- ity prostheses has expanded dramat- ically. Prosthetic components have a significant impact on functional per- formance. The choice of compo- nents varies depending on a patient’s functional level; this is especially true regarding the specific needs of patients with amputation secondary to peripheral vascular disease or dia- betes. These critical needs include protecting the sound limb, consider- ing abnormal and excessive forces on the residual limb, and factoring in the metabolic costs of ambulation. Understanding lower extremity prosthetic componentry and how ap- plication varies is important. Appli- cation is based on the level of ampu- tation in the context of the expected functional level of the user. A classi- fication scale can assist in determin- ing appropriate components corre- sponding to each functional level. Etiology and Incidence of Amputation In the United States, lower extrem- ity amputation is not uncommon; approximately 110,000 people un- dergo some level of lower limb am- putation surgery each year. 1 Of those amputations, most are a result of disease (70%), followed by trauma (22%) and congenital etiology and tumor (4% each). 1 Approximately 54,000 amputations secondary to di- abetes are performed annually in the United States. 2 Further, more than half of all lower limb amputations occur in individuals with diabetes; below-knee or distal amputations are more common in this population than transfemoral amputations. Be- tween 9% and 20% of patients with diabetes who have had an amputa- tion undergo a second amputation ipsilaterally or a new amputation contralaterally within 12 months of the first amputation. 2 Thirty percent Karen Friel, PT, DHS Dr. Friel is Associate Professor and Chair, Department of Physical Therapy, New York Institute of Technology, Old Westbury, NY. Neither Dr. Friel nor the department with which she is affiliated has received any- thing of value from or owns stock in a commercial company or institution re- lated directly or indirectly to the subject of this article. Reprint requests: Dr. Friel, New York Institute of Technology, Room 501, Northern Boulevard, Old Westbury, NY 11568. J Am Acad Orthop Surg 2005;13:326- 335 Copyright 2005 by the American Academy of Orthopaedic Surgeons. 326 Journal of the American Academy of Orthopaedic Surgeons to 50% of patients with amputations performed as a result of diabetes will lose the contralateral limb within 3 to 5 years after the first amputa- tion. 1,2 Therefore, preserving the in- tact limb is of paramount impor- tance and is a significant factor in the prosthetic management of the amputated limb. These data indicate that despite advances in new pros- thetic components, health care pro- viders still face challenges in fitting patients with optimal prosthetic components and in rehabilitating them to a level of functional inde- pendence. Although 85% of persons treated with amputation for a poorly vascu- larized lower limb are fitted with a prosthesis, only 5% use the limb for more than half of their waking hours; 3 furthermore, within 5 years, only 31% are still using the prosthe- sis. 4 In addition, only 26% of patients are walking outdoors 2 years after amputation for an insufficently vas- cularized or compromised limb. 4 Fi- nally, the 5-year death rate for pa- tients with amputation who are fitted with a prosthesis is 48%, whereas the rate is 90% for patients not fitted with a prosthesis. 4 It is not known whether these patients are ill initially or whether a more sedentary lifestyle leads to their decline. Fitting a patient with prosthetic compo- nents that enhance ambulation and increase functional independence is therefore extremely important. Functional Classification Scale A guideline useful in the selection of prosthetic components is the K-rating scale of the US Department of Health and Human Services’ Cen- ter for Medicare and Medicaid Ser- vices. The K-rating scale classifies individuals with amputation into five functional categories. Although primarily used for reimbursement considerations, the scale can provide a context for the prescription of pros- thetic components, particularly prosthetic knees and feet. For in- stance, a knee with a swing rate con- trol mechanism is appropriate for K-1 and K-2 levels, whereas a knee that permits a variable cadence swing rate mechanism would be ap- propriate for K-3 and K-4 levels 5,6 (Table 1). To assist clinicians in proper clas- sification, the Amputee Mobility Predictor has been developed to determine functional ambulation ability following amputation. This simple test, which objectively cate- gorizes patients into an appropriate K-level, 8 has proved to be reliable and valid. It assesses sitting and standing balance, quality of ambulation, and ability to perform limited walking skills. Biomechanics of Gait Related to Amputation and Prosthetic Design Walking is a highly efficient activity, with forces absorbed and dissipated throughout the gait cycle. These forces include gravity, inertia, and muscular action. Muscles transform potential energy into kinetic energy through viscoelastic elements and by contracting both concentrically and eccentrically throughout the gait cycle. After amputation, pa- tients lose many of the muscular forces that function during walking; they must rely instead on a variety of bumpers, springs, and hydraulic/ pneumatic mechanisms in an at- tempt to simulate a normal gait pat- tern and enhance energy efficiency. Many studies have investigated the energy expenditure and metabol- The K-Classification System for Functional Ambulation 5,7 K Level Factor K-0 K-1 K-2 K-3 K-4 Description Nonambulator; requires assist with transfers Household ambulator Limited community ambulation Unlimited community ambulation Exceeds basic use Gait activity Nonambulance Fixed cadence; level surfaces Fixed cadence; negotiates minor community barriers (eg, curbs, ramps, stairs) Variable cadence; negotiates environment freely; has use beyond simple gait Exhibits high-energy activity; high-impact activity Recommended feet Not a pros- thesis candi- date Non–dynamic response foot Non–dynamic response foot Dynamic response foot; energy-storing foot Dynamic response foot; energy-storing foot Recommended knees Not a pros- thesis candi- date Fixed-cadence swing rate Fixed-cadence swing rate Variable-cadence swing rate; computer-assisted Variable-cadence swing rate; computer-assisted Table 1 Karen Friel, PT, DHS Volume 13, Number 5, September 2005 327 ic factors related to gait patterns af- ter amputation. Results of these studies show that the cadence fol- lowing amputation is slower (and the metabolic output higher) com- pared with the cadence of patients without amputation. 1,9 These differ- ences are related to factors such as loss of kinetic energy, changes in muscle symmetry, and loss of coor- dination and balance in amputees, not to mass of the prosthetic compo- nents. 9,10 The weight of most pros- theses is approximately equal to 30% of the weight of a normal low- er limb. 11 Therefore, the weight of various components should not be a concern in prosthetic prescription; rather, matching components to the expected functional level of the user should be paramount. During normal gait, the muscu- loskeletal structures of the lower ex- tremity help to attenuate impact forces. This is accomplished through mechanisms that include knee flex- ion from heel strike to midstance during loading response, the plantar fat pad at initial contact, foot prona- tion during foot flat, and eccentric loading of the muscles themselves. After amputation, however, many of these mechanisms are lost, with the prosthesis able to accommodate only partially by using shock absorb- ers and pylons. One study investigated the effect of pylon material on ground reaction forces during gait with a transtibial prosthesis. Results indicated that, compared with prostheses with py- lons made of aluminum, prostheses with flexible pylons composed of ny- lon had force patterns that more closely mimic those of the nonam- putated limbs. Additionally, with the flexible pylons, a smoother tran- sition occurred between the braking phase of gait at initial contact and the propulsive phases of gait. 12 Postema et al 13 suggested that the degree of dorsiflexion allowed by the prosthetic ankle at the end of stance phase influences balance control dur- ing gait. They proposed that in- creased dorsiflexion causes an in- crease in knee flexion torque, thus decreasing knee stability. Conversely, decreasing the amount of available dorsiflexion decreases the flexor mo- ment to the knee, providing the user with added knee extension stability at late stance. Therefore, patients with balance difficulties may feel more secure with an ankle unit that allows for less dorsiflexion. 11 Others have proposed, however, that mechanical stability (ie, balance control) differs from proprioceptive control. 14 Although the more rigid foot may provide increased mechan- ical stability, active users interpret good balance as having a wider range of balance options on uneven ter- rain, as can be accomplished with the more flexible design. Suspension All of the various types of suspen- sion mechanisms are designed to hold the prosthesis securely onto the residual limb, prevent pistoning, and minimize breakdown. Traditional Suspension Systems The supracondylar cuff was an ex- tremely popular means of suspen- sion for the transtibial prosthesis in the 1970s and 1980s. However, this type of suspension should not be used for the individual with vascular compromise 15 because, to hold the prosthesis on the patient, the cuff mechanism relies on constriction proximal to the knee. 16 Although still used today, the supracondylar cuff is being replaced by more cos- metic, secure means of suspension. It is now most appropriate for the less active user and limited ambula- tor (K-1 level). The suspension sleeve is another option for suspension of the transtib- ial prosthesis. A sleeve made of neo- prene, latex, or elastomer materials is fitted onto the upper aspect of the prosthesis. The other end is rolled above the prosthesis onto the pa- tient’s skin, adhering to the skin through negative pressure. Sleeves are simple to use, inexpensive, fair- ly cosmetic, and appropriate for any level of user. The sleeve may be dif- ficult to don, however, for patients with hand weakness or poor dexter- ity, as is commonly seen in individ- uals with diabetes. 15 The suprapatellar/supracondylar suspension system uses the bony structures of the knee to suspend the transtibial prosthesis. The medial condyle of the femur and the supra- patellar aspect of the knee form bony locks against slippage of the prosthesis during the swing phase of gait and other activities when pis- toning may occur. This suspension system may be used when there is an exceedingly short residual limb or when additional knee stability is re- quired. In some circumstances, aux- iliary suspension, such as a sleeve, may also be used with this design. 16 Suspension around the waist can be used both as the primary and the auxiliary means of suspension. The Silesian belt and elastic suspension are composed of a strap or sleeve that attaches to the proximal end of the prosthesis and ascends to encircle the patient’s waist. These straps may be composed of neoprene (called a total elastic suspension system) or of cotton. 15 Neither of these methods helps to control the hip in the presence of instability. Contemporary Suspension Systems The shuttle lock system, also known as the pin-and-lock system, continues to gain in popularity for both the transtibial and transfem- oral prostheses. This system pro- vides cushioning, torque control, and shock absorption because the outer surface of the liner acts as an interface between the skin and the socket. 17 This interface dissipates forces that would affect the skin in a total suction situation , which does not use a liner or interface between the residual limb and the socket. Componentry for Lower Extremity Prostheses 328 Journal of the American Academy of Orthopaedic Surgeons The shuttle lock system uses a gel or silicone liner with a locking pin on the bottom, which is rolled onto the skin. The pin is then inserted into a shuttle lock inside the socket (Fig. 1). This system helps to provide for a total-contact fit, which minimizes distal edema, distributes pressure over the entire limb, and prevents movement of the limb against the socket. The coefficient of friction be- tween the stump-liner interface and the liner-socket interface needs to be high to minimize any movement be- tween the surfaces. The soft, flexible gel liners can accomplish this. 16,18 In fact, indications for these systems in- clude patients whose skin is sensitive to shear forces and uncontrolled pis- toning in the socket. 15 Because of the potential for skin breakdown and sub- sequent infection, pistoning is a threat to further loss of limb to an amputee with vascular disease or di- abetes. Prosthetic socks can be added to the shuttle lock system in the event of limb girth fluctuations. The shuttle lock system is appro- priate for all levels of users because of the security afforded by this sus- pension method as well as the im- proved cosmesis and ease of don- ning. When the transfemoral residual limb is long, there may be a difference between the involved limb and the sound limb in the knee centers of rotation when the locking hardware is placed inside the pros- thesis. 15 The shuttle lock system is an excellent alternative for users who have difficulty donning the full suction socket. 19 Suction is a popular means of sus- pension, particularly for the patient with a transfemoral amputation. It provides for an intimate fit between the limb and the socket, which en- hances proprioception and muscular control of the prosthesis. 20 Comfort level is also enhanced because auxil- iary suspension, such as a belt, waist strap, or thigh corset, is not needed, although an additional means of sus- pension may be used when a higher activity level requires it. The socket is held on through negative pressure and surface tension. Because the pa- tient must stand to ensure that the limb is fully entered into the socket, patients with poor balance or prob- lems with manual dexterity may have difficulty donning this type of socket. 15,19 Total suction is not often used with the transtibial prosthesis because the bony characteristics of the lower limb make it difficult to obtain a tight seal. Prosthesis Construction Traditional Construction In the past, prostheses were fabri- cated in an exoskeletal fashion: the strength of the prosthesis derived from the solid outer walls. Exoskel- etal prostheses were composed of a solid piece of wood or rigid polyure- thane covered with plastic laminate and fashioned into the shape of a leg. The components were embedded or built-in and thus were not inter- changeable. 21 Unless an external frame was used, the entire prosthesis needed to be refabricated to change componentry. In addition, these prostheses were heavy and bulky. Exoskeletal prostheses are not usual- ly fabricated today unless a user spe- cifically requests such construction; some long-term prosthesis users have become accustomed to the exoskeletal design and opt not to change. Contemporary Construction Today, most prostheses are of an endoskeletal design: components are located inside the prosthesis. The strength of the prosthesis comes from the pylon—usually made of lightweight nylon, aluminum, or carbon/graphite—which is enclosed in a cosmetic foam covering. Bene- fits of the endoskeletal design are that components of a standardized design are completely interchange- able, the prosthesis is easily repaired, and the design is lighter and more cosmetic than the exoskeletal de- sign. 21 However, these prostheses are subject to external moisture and de- bris. Figure 1 A, Liner with attached pin for shuttle lock mechanism. B, Shuttle lock mechanism in clear check socket. Karen Friel, PT, DHS Volume 13, Number 5, September 2005 329 Transtibial Prostheses The prosthetic socket has several important functions. It is designed to accommodate the residual limb, allow for weight bearing, distribute forces, and provide total contact to prevent distal pooling of fluid with- in the residual limb. The sockets are custom-fitted and have specific areas of weight bearing incorporated into their design. Traditional Socket Design Since the 1950s, the most com- mon socket design has been patellar tendon–bearing (PTB), still consid- ered the standard today. 16 The design is based on increasing weight- bearing pressures in areas that are pressure tolerant. These areas in- clude, but are not restricted to, the patellar tendon, medial and lateral tibial flares, and gastrocnemius- soleus complex. Conversely, the socket is designed to decrease pres- sures in areas that are pressure- sensitive, such as the proximal and distal fibula and the tibial crest. Contemporary Socket Design With the advent of new materials and fabrication principles, an in- creasingly common adjunct to the PTB design is the use of hydrostatic loading. Hydrostatic loading stabiliz- es the bony anatomy within the soft tissues through the use of compres- sion and elongation of the tissues during casting for the socket. The forces of weight bearing are distrib- uted through a greater surface area, thus decreasing pressures to any one area. This technique is also known as total-surface bearing; the force is evenly distributed throughout the entire limb. 16 This distribution may help to prevent breakdown of the skin and enhance comfort for the user. Foot/Ankle Assembly Advances in the design of pros- thetic feet are occurring at a dramat- ic rate, and new feet are introduced to the market regularly. Numerous factors must be considered when fit- ting a prosthetic foot (Table 2). The most notable factor related to the be- havior of the prosthetic foot is the presence or absence of a joint that al- lows for plantar flexion. This factor is significant because the ability to have both plantar flexion and dorsi- flexion range of motion forms the basis for the classification system of articulated and nonarticulated ankle designs. 24 Many of the newer designs have an integrated pylon/ankle/foot mechanism, which allows for both dorsiflexion and energy return to the user. It should be noted that there is no difference between the prosthetic feet used for transtibial prostheses and those used for transfemoral pros- theses. The choice of foot depends on the patient’s mobility, stability, and functional use and control of the prosthesis. Non–Dynamic Response Feet The solid ankle cushioned heel (SACH) foot (Sheck and Siress, Chi- cago, IL) (Fig. 2) has been extremely popular since its inception in the 1950s and is very economical com- pared with other prosthetic feet. The SACH foot uses compressible mate- rial in the heel to simulate plantar flexion at heelstrike. It incorporates a rigid, wooden keel that is unable to dorsiflex through the midstance phase of gait. Because of this, during midstance, the center of mass on the prosthetic side is comparatively higher than on the nonamputated side. This inequity leads to increased loads placed on the sound side during the weight acceptance phase of gait; 25 instead of the normally smooth tran- sition provided by adequate dorsiflex- ion, the user tends to “fall onto” the sound side during weight transfer. Studies have shown that ambulating with the SACH foot produces the greatest ground reaction forces on the sound side compared with both dy- namic response feet and other non– dynamic response feet. 26,27 This means that the SACH foot is not op- timal at protecting the sound limb from excessive forces, which is a con- cern because of the high rate of con- tralateral amputation in the popula- tion with diabetes. 2 However, the Key Concepts for Foot Prescription Ability to adequately absorb impact forces Ability to accommodate to uneven terrain Avoidance of the prosthesis being too heavy distally 22 Dynamic response of the foot (ie, ability to return energy to the user during push-off 23 ) Maintenance of proper balance Table 2 Figure 2 Solid ankle cushioned heel (SACH) foot. Componentry for Lower Extremity Prostheses 330 Journal of the American Academy of Orthopaedic Surgeons SACH foot is still appropriate for the limited ambulator, the K-1 level user, and the individual in the beginning stages of rehabilitation. One major ad- vantage of using this type of pros- thetic foot is that the rigid keel may provide more balance than would a dynamic response foot. 6 Feet specifically designed for the geriatric patient have keels com- posed of flexible polypropylene. This design replicates a more pronated position of the foot, with more of the foot in contact with the ground. This factor provides for added stability and a softer rollover, thus minimiz- ing forces to the residual limb. 5 The Dycor ADL uniaxial design (Dycor, Missouri City, TX) is currently cat- egorized for the K-2 level user. Dynamic Response Feet Currently, the more responsive prosthetic feet are generally reserved for the more active ambulators. These feet are available in both artic- ulated and nonarticulated designs. The dynamic response foot uses a keel that deforms under pressure but returns to its original shape when the load is removed. The keel acts as a spring that on return to its original shape returns energy to the user, thereby assisting push-off. The flex- ibility of the keel allows for dorsi- flexion. 6 The increased dorsiflexion afforded by the dynamic response foot allows for a longer midstance time in the gait cycle. Hafner et al 28 noted that increased time spent in midstance may increase the percep- tion of stability, compared with the rapid heel rise and toe-only support in the non–dynamic response foot. Hafner et al 28 compared patient perception of energy-storing feet ver- sus their perception of conventional prosthetic feet using biomechanical gait analysis. Results indicated that, despite advantages perceived by us- ers when ambulating with a dynam- ic response foot, supportive biome- chanical data were inconsistent. The advantages that users reported when ambulating at higher velocities with a dynamic response foot were in- creased gait velocity, increased sta- bility, increased ankle motion, de- creased shock at the hip and knee, and enhanced performance in “high activity” gait (ie, activities requiring increased ankle power and propul- sion). 28 The impact that foot selection has on forces taken through the sound limb also has been investigated. Spe- cifically, the Flex-Foot (Össur, Aliso Viejo, CA) (Fig. 3) was compared to SACH, Carbon Copy II (Ohio Willow Wood, Mt. Sterling, OH), Seattle (Model and Instument W orks, Seattle, WA), and Quantum (Hosmer Dor- rance Corp, Campbell, CA) feet. The Flex-Foot notably reduced peak ver- tical ground reaction forces to the sound limb compared with the other feet. In fact, the other feet on average increased peak forces to the sound limb 17% over normal values. The authors therefore hypothesized that the increased dorsiflexion achieved with the Flex-Foot design allows for less of a fall onto the sound limb dur- ing the weight-acceptance phase of gait. 26 All of the dynamic response feet are usually prescribed for the K-3 or K-4 level ambulator. Several shock absorbers are avail- able, many of them built into the an- kle mechanism of the foot/ankle as- sembly. The Reflex Vertical Shock Pylon (VSP) (Össur), is a variation of the Flex-Foot, with the vertical shock absorber built into the ankle mech- anism. 5 Results of a study by Hsu et al 29 indicated that the Reflex VSP al- lowed for improved energy cost and gait efficiency compared with the SACH foot or Flex-Foot. Specifically addressing gait parameters, Miller and Childress 30 found that vertical compliance of the pylon caused little change in gait parameters during nor- mal speeds of walking. With the Re- flex VSP system, greater changes were noted in ground reaction forces, vertical trunk displacement, and py- lon compression at faster walking and jogging speeds compared with normal walking speeds. The most re- cent version of this foot is called the Ceterus (Össur) (Fig. 4). Transfemoral Prostheses The design principles for the trans- femoral socket are similar to those for the transtibial socket. Currently, there are three primary designs. The plugfit original sockets for transfem- oral prostheses were cylindrical and used the soft tissues of the thigh for weight bearing. Today’s sockets, whether the traditional quadrilateral socket or more contemporary ischial containment or flexible sockets, all feature some level of shared weight bearing between the skeleton of the pelvis and the soft tissues of the thigh. Traditional Socket Design Quadrilateral design sockets first appeared in the 1950s. They are so named because each of the four walls of the socket has distinct features to apply forces and distrib- ute pressures. Weight bearing is achieved primarily through the is- chial tuberosity and gluteal muscu- lature sitting atop a posterior shelf. This socket provides for lateral sta- bilization of the femur to assist with pelvic stability. 19 Figure 3 Flex-Foot, an integrated pylon/ankle/ foot. Karen Friel, PT, DHS Volume 13, Number 5, September 2005 331 Critics have suggested that use of this socket results in skin irritation in the ischium and pubis, tender- ness over the anterior distal femur, and discomfort from the anterior wall when sitting, as well as poor cosmesis and a tendency toward a Trendelenburg-type gait. 31 This de- sign is rarely used today. Contemporary Socket Design The ischial containment socket design, the current standard (Fig. 5), resulted from addressing some of the criticisms of the quadrilateral sock- et. Specifically, certain parameters regarding transfemoral socket fit in- corporate the design principles of the ischial containment socket devel- oped in 1987 by the International Society for Prosthetics and Orthot- ics 19 (Table 3). This design emphasiz- es maintaining adequate femoral ad- duction for enhanced pelvic stability and improved gait. Improved force distribution and stability are empha- sized by having more of the pelvis housed within the socket rather than sitting on top of the socket, as in the quadrilateral design. 20,31 The flexible above-knee socket (also known as the Icelandic, Scandi- navian, or New York socket), while still employing ischial containment principles, incorporates a flexible in- ner socket supported by a rigid out- er frame with cut-out sections 31 (Fig. 6). This design minimizes pressures within the socket of contracting muscles and soft tissues. All of these socket designs can be used with any type of suspension. Knees The variable that determines which knee is appropriate for each functional K-level is whether the knee allows for a fixed pendu- lum swing or a variable cadence of Figure 5 Ischial containment socket. Overhead view. Figure 4 Left, Reflex VSP with integrated shock absorption. Right, Ceterus with integrated shock absorption. Design Principles of the Transfemoral Socket 19 Maintain normal femoral adduction and narrow-based gait Enclose the ischial tuberosity and ramus within the socket to create a skeletal lock Distribute forces along the shaft of the femur Decrease emphasis on a narrow anterior-posterior diameter Provide total contact Use suction suspension when possible Table 3 Componentry for Lower Extremity Prostheses 332 Journal of the American Academy of Orthopaedic Surgeons swing. A fixed-swing rate control knee is appropriate for K-1 and K-2 level functional ambulators. The unlimited community ambulators, K-3 and K-4 users, are capable of us- ing a variable-cadence swing mech- anism (Table 1). This category of prosthesis uses both hydraulic and pneumatic mechanisms to control the rate of swing. 7 Fixed-Cadence Knee Mechanisms Conventionally damped pros- thetic limbs use fixed resistance in the knee unit to control the pendu- lum action of the prosthesis. This rate of swing is set by the prosthet- ist. When the cadence of gait changes, the user must compensate for the fixed pendulum speed by us- ing gait deviations to change the rate of extension or by forcefully throw- ing the limb forward to ensure that the foot will be in the correct loca- tion at heel strike. 7 Many of these knees have a stance lock control so that the knee will not buckle during stance. This is useful for the patient who has poor prosthetic control and balance or for the K-1 and K-2 level ambulator. Variable-Cadence Knee Mechanisms Variable-cadence knees use pneu- matics or hydraulics to accommo- date to the user’s walking speed. The range of velocities of swing rate set into the unit is dependent on the us- er’s typical level of functioning. The ambulator is free to change walking speed within that range and still avoid gait deviations. One option available to the user is the addition of a stance flexion com- ponent. In normal gait, the stance knee will flex approximately 15° to 18° as load is transferred onto the weight-bearing leg. This lowering of the center of mass allows for a de- creased load on the limb 7 as well as a cushioned support with a gradual weight transfer onto the sound limb. 32 Given the propensity for contralateral limb loss in patients with vascular disease or diabetes, decreasing the loads placed on the sound limb may help to prolong and protect the health of that limb. Stance flexion devices incorporate some degree of flexion during stance. They have been devel- oped to decrease load as well as to add stability during gait by lowering the center of mass. For the patient with potential proprioceptive difficulties, this could be advantageous for the safety and efficiency of gait. In addi- tion, some degree of stance control is favorable for the more active person with a lower limb amputation when put into compromising situations for which additional stability may be necessary. 7 A computer-assisted knee mech- anism uses a computer chip im- planted into the hydraulic knee unit to accommodate to the walking speed of the user. This allows for correction and control of the knee continuously throughout the gait cycle—up to 50 times per second with little or no thought required by the prosthesis user—to ensure proper swing rate and stance con- trol. 7 Hence, there is no need to compensate with gait deviations. 33 Computer-assisted knees (eg, the C-leg [Otto Bock, Minneapolis, MN] and the Intelligent Knee [Endolite, Centerville, OH]) can assume part of the energy-absorbing functions of the quadriceps and hamstrings nor- mally seen during early and late swing phases of gait 11 (Fig. 7). Be- cause these knees allow for variable cadence, they would be appropriate only for the high activity−level user—K-3 or K-4 on the K-rating scale. The expense of these knees is not warranted for the more limited ambulator who is unable to benefit from its advantages. Datta and Howitt 34 compared user satisfaction and overall use when ambulating with a pneumatic swing phase–control knee versus a microprocessor-controlled intelli- gent knee. Using a questionnaire for- mat, they found that most users pre- ferred the microprocessor-controlled knee unit. In fact, 95% reported walking at different speeds to be “a lot easier” or “easier.” More than 81% said they could walk farther, and 59% found walking on slopes and hills “a lot easier.” An over- whelming 95% felt that walking was more nearly “normal.″ 34 Figure 6 Left, Flexible above-knee socket. Right, Outer socket for flexible system. Karen Friel, PT, DHS Volume 13, Number 5, September 2005 333 Studies addressing energy expen- diture show that at gait velocities >3.2 km/h, a decrease in energy ex- penditure of approximately 10% oc- curred when ambulating with a mi- croprocessor knee compared with ambulation using a conventional knee prosthesis. 33,35 A common re- port of the elderly prosthesis user is that the leg feels “heavy” or that the prosthesis is too fatiguing to use. A knee that can markedly decrease en- ergy expenditure may have consider- able implications for the overall ac- tivity level, health, and well-being of the patient. Summary Rapid advances in prosthesis tech- nology have led to an expansion of prosthetic options for individuals with transtibial and transfemoral amputations, regardless of cause of the amputation. These options may be grouped into classes of compo- nents, which can then be viewed in the context of the needs of users with different functional levels. Re- gardless of the functional level of the user, contemporary prostheses generally use endoskeletal con- struction, sockets that emphasize total contact, and weight distribu- tion between bony structures and soft tissues. Such prostheses also use suspensions that minimize the use of constrictive belts and cuffs proximal to the level of amputation. For individuals expected to be household ambulators or limited community ambulators, tradi- tional, non–dynamic response pros- thetic feet and fixed-cadence knees may be appropriate. For individuals who are expected to be unlimited community ambulators, or for those who will place high work or recre- ational demands on their prosthe- ses, contemporary dynamic re- sponse feet and variable-cadence knees should be prescribed. Specific considerations exist for persons with peripheral vascular dis- ease. One primary concern is preser- vation of the intact limb, which can be improved by components that help to lower the center of mass as well as ease weight transfer onto the sound limb. Second, skin integrity is equally important and can be aided by liners composed of gel, silicone, or similar materials that serve to de- crease shear and dissipate friction forces. The physician, therapist, prosthetist, and patient should all be actively engaged in the decision- making process. Acknowledgment The author wishes to thank Eliza- beth Domholdt, PT, EdD, for her as- sistance with significant revisions of this manuscript. References 1. Gailey RS: One Step Ahead: An Inte- grated Approach to Lower Extremity Prosthetics and Amputee Rehabilita- tion. Miami, FL: Advanced Rehabili- tation Therapy, 1994. 2. Reiber GE, Boyko EJ, Smith DG: Lower extremity foot ulcers and amputations in diabetes, in Diabetes in America, ed 2. Washington, DC:National Insti- tutes of Health, 1995, pp 409-427. Available at www.niddk.nih.gov/ health/diabetes/dia/contents.htm Ac- cessed July 20, 2005. 3. Collin C, Collin J: Mobility after lower-limb amputation. Br J Surg 1995;82:1010-1011. 4. McWhinnie DL, Gordon AC, Collin J, Gray DW, Morrison JD: Rehabilita- tion outcome 5 years after 100 lower- limb amputations. Br J Surg 1994;81: 1596-1599. 5. Nassan S: The latest designs in pros- thetic feet. Phys Med Rehabil Clin N Am 2000;11:609-625. 6. Romo HD: Specialized prostheses for activities: An update. Clin Orthop 1999;361:63-70. 7. Romo HD: Prosthetic knees. Phys Med Rehabil Clin N Am 2000;11:595-607. 8. Gailey RS, RoachKE, ApplegateEB, et al: The amputee mobility predictor: An instrument to assess determi- nants of the lower-limb amputee’s ability to ambulate. Arch Phys Med Rehabil 2002;83:613-627. 9. Gailey RS, Nash MS, Atchley TA, et al: The effects of prosthesis mass on metabolic cost of ambulation in non- vascular trans-tibial amputees. Pros- thet Orthot Int 1997;21:9-16. 10. Selles RW, Bussmann JB, Wagenaar RC, Stam HJ: Effects of prosthetic mass and mass distribution on kine- matics and energetics of prosthetic gait: A systematic review. Arch Phys Med Rehabil 1999;80:1593-1599. 11. Rietman JS, Postema K, Geertzen JH: Gait analysis in prosthetics: Opin- ions, ideas and conclusions. Prosthet Orthot Int 2002;26:50-57. 12. Coleman KL, Boone DA, Smith DG, Czerniecki JM: Effect of trans-tibial prosthesis pylon flexibility on ground reaction forces during gait. Prosthet Orthot Int 2001;25:195-201. 13. Postema K, Hermens HJ, de Vries J, Koopman HF,Eisma WH:Energy stor- age and release of prosthetic feet: I. Biomechanical analysis related to user benefits. Prosthet Orthot Int 1997;21:17-27. 14. Nielsen DH, Shurr DG, Golden JC, Meier K: Comparison of energy cost and gait efficiency during ambulation in below-knee amputees using differ- ent prosthetic feet—a preliminary report. Journal of Prosthetics and Orthotics 1989;1:24-31. 15. Kapp S: Suspension systems for pros- theses. Clin Orthop 1999;361:55-62. 16. Edwards ML: Below knee prosthetic Figure 7 Otto Bock computerized leg C-leg. (Courtesy of Otto Bock Health Care, Minneapolis, MN.) Componentry for Lower Extremity Prostheses 334 Journal of the American Academy of Orthopaedic Surgeons socket designs and suspension sys- tems. Phys Med Rehabil Clin N Am 2000;11:585-593. 17. Geertzen JH, Martina JD, Rietman HS: Lower limb amputation: II. Reha- bilitation—a 10 year literature re- view. Prosthet Orthot Int 2001;25:14- 20. 18. Emrich R,Slater K:Comparative anal- ysis of below-knee prosthetic socket liner materials. J Med Eng Technol 1998;22:94-98. 19. Kapp SL: Transfemoral socket design and suspensionoptions. PhysMed Re- habil Clin N Am 2000;11:569-582. 20. Dingwell JB, Davis BL, Frazier DM: Use of an instrumented treadmill for real-time gait symmetry evaluation and feedback in normal and trans- tibial amputee subjects. Prosthet Orthot Int 1996;20:101-110. 21. May BJ: Prosthetic components, in Amputations and Prosthetics: A Case Study Approach. Philadelphia, PA: F.A. Davis Co, 1996, pp 118-159. 22. Lehmann JF, Price R, Okumura R, Questad K, de Lateur BJ, Négretot A: Mass and mass distribution of below- knee prostheses: Effect on gait effica- cy and self-selected walking speed. Arch Phys Med Rehabil 1998;79:162- 168. 23. Fergason JR, Boone DA: Custom de- sign in lower limb prosthetics for ath- letic activity. Phys Med Rehabil Clin NAm2000;11:681-699. 24. Cortés A, Viosca E, Hoyos JV, Prat J, Sánchez-Lacuesta J: Optimisation of the prescription for trans-tibial (TT) amputees. Prosthet Orthot Int 1997; 21:168-174. 25. Lehmann JF, Price R, Boswell- Bessette S, Dralle A, Questad K: Com- prehensive analysis of dynamic elas- tic response feet: Seattle Ankle/Lite Foot versus SACH foot. Arch Phys Med Rehabil 1993;74:853-861. 26. Powers CM, Torburn L, Perry J, Ayyappa E: Influence of prosthetic foot design on sound limb loading in adults with unilateral below-knee amputations. Arch Phys Med Rehabil 1994;75:825-829. 27. Hayden S, Evans R, McPoil TG, Corn- wall MW, Pipinich L: The effect of four prosthetic feet on reducing plan- tar pressures in diabetic amputees. Journal of Prosthetics and Orthotics 2000;12:92-96. 28. Hafner BJ, Sanders JE, Czerniecki J, Fergason J: Energy storage and return prostheses: Does patient perception correlate with biomechanical analy- sis? Clin Biomech (Bristol, Avon) 2002;17:325-344. 29. Hsu MJ, Nielsen DH, Yack HJ, Shurr DG: Physiological measurements of walking and running in people with transtibial amputations with 3 differ- ent prostheses. J Orthop Sports Phys Ther 1999;29:526-533. 30. Miller LA, Childress DS: Analysis of a vertical compliance prosthetic foot. J Rehabil Res Dev 1997;34:52-57. 31. Esquenazi A, Leonard JA Jr, Meier RH III, Hicks JE, Fisher SV, Nelson VS: Prosthetics, orthotics, and assistive devices: III. Prosthetics. Arch Phys Med Rehabil 1989;70:S206-S209. 32. Cochrane H, Orsi K, Reilly P: Lower limb amputation: III. Prosthetics a 10 year literature review. Prosthet Orthot Int 2001;25:21-28. 33. Taylor MB, Clark E, Offord EA, Baxter C: A comparison of energy expendi- ture by a high level trans-femoral am- putee using the Intelligent Prosthesis and conventionally damped prosthet- ic limbs. Prosthet Orthot Int 1996;20: 116-121. 34. Datta D, Howitt J: Conventional ver- sus microchip controlled pneumatic swing phase control for trans-femoral amputees: User’s verdict. Prosthet Orthot Int 1998;22:129-135. 35. Buckley JG, Spence WD, Solomonidis SE: Energy cost of walking: Compari- son of “Intelligent Prosthesis” with conventional mechanism. Arch Phys Med Rehabil 1997;78:330-333. Karen Friel, PT, DHS Volume 13, Number 5, September 2005 335