It is estimated that patients with a knee disarticulation (through-knee amputation) represent 2% or less of the overall amputee cohort in the United States.^1,2 However, in other countries, the knee disarticulation cohort may be as high as 24%.³ This amputation level is characterized by distinct advantages and disadvantages relative to the more commonly performed, more proximal transfemoral amputation levels.⁴ The residual limb is subjected to fewer bony and muscular disruptions, yielding more balanced muscular control at the hip and a limb that is generally capable of load bearing at the distal end. In addition, the distal bony contours often lend themselves to anatomic suspension of the prosthesis. However, the length of the residual limb precludes the ability to match the knee center of the prosthesis to the anatomic knee center, creating both functional and aesthetic considerations.

As with most amputation levels, the outcomes associated with through-knee amputation vary with the amputation etiology and comorbid health considerations. Polfer et al⁵ reported on a small convenience sample of 10 individuals seen at US military trauma centers with knee disarticulation with an average age of 23 years at the time of amputation. These individuals were matched to a separate cohort of individuals with transfemoral amputations based on time since amputation, preinjury activity levels, and comorbid injury to the sound side extremity. The authors found no significant differences between the two cohorts with respect to outcomes of general health, satisfaction, perceived disability, activity, or participation.

By contrast, Nijmeijer et al⁶ recently reported on 153 knee disarticulation amputations in 138 patients with peripheral arterial disease (98%) and diabetes mellitus (63%) at a mean age of 74 years. Survival rates at 1, 6, and 12 months were reported at 86%, 65%, and 55%, respectively. Only 10% of these primary amputations were ultimately revised to transfemoral amputations. Seventy-one percent of the patients expressed a preoperative intent to ambulate with a prosthesis, and 91% of this contingent were successfully fit with a prosthesis. Of these, 35% walked without the help of others.

A presurgical consultation with the treating prosthetist should be considered whenever possible. Although there are clear functional benefits associated with this amputation level, they come at the expense of cosmetic appearance and functional convenience in certain sitting environments. Patients should anticipate the functional and cosmetic implications of the distally positioned prosthetic knee center that are inherent to this amputation level before
fabricating the prosthesis, and if possible, before the amputation. Given the opportunity to engage in this decision, some patients may elect a transfemoral amputation to afford a more cosmetically acceptable prosthetic appearance.

Knee disarticulation amputation has some distinct advantages over a transfemoral amputation. In the disarticulation, the femoral condyles typically provide the weight-bearing surface within the prosthesis. This provides the ipsilateral hip adductors a stable platform to act upon during single limb stance. As a result, rather than pulling the distal femur laterally, as is generally seen with transfemoral amputation, the hip abductors are able to stabilize the pelvis in the coronal plane, keeping the contralateral hip from dropping during swing. This maintenance of the femoral adduction angle may limit or even eliminate the need for a compensatory Trendelenburg gait during single-limb support. Additional advantages include the bulbous shape of the distal femur, which enables an anatomic suspension of the prosthesis (Figure 1), as well as the ability to bear weight on the limb without a prosthesis in a kneeling position. This latter consideration is particularly important for an individual with a bilateral lower limb amputation who can benefit greatly if either or both residual limbs provide improved sitting balance, assist with transfers, and enable limited knee-walking without prostheses.

In general, the individual using a knee disarticulation prosthesis should have function similar to or better than that of an individual wearing a transfemoral prosthesis.⁷ Walking speed may be faster, oxygen consumption lower, and donning the prosthesis a bit easier.⁸ Patients who are healthy enough to run can often do so with the appropriate prosthesis, training, and practice.

Disadvantages of knee disarticulation prostheses are generally associated with the bulbous nature of the femoral condyles. This can result in a bulky appearance of the thigh segment of the prosthesis. In addition, and often of greater concern, the full length of the ipsilateral femur, coupled with even the lowest profile prosthetic componentry will displace the prosthetic knee center distally (Figure 2). This creates an asymmetry in swing phase as the shortened prosthetic shank creates a smaller knee flexion arc. Further, the shortened tibial length fails to reach the ground on many sitting surfaces, which often creates a fulcrum point at the edge of the chair, placing pressure along the posterior proximal aspect of the socket. The abnormally long femoral segment can prove extremely disruptive in tight seating environments with limited legroom. Other reasons for rejection can be a painful patella, or pain or skin breakdown over bony areas.⁹

Many of the benefits associated with knee disarticulation are lost if the distal end of the limb is not tolerant of load-bearing forces. To ensure a stable, well-padded distal end, surgeons have recommended that, when possible, the gastrocnemius muscle bellies and overlying posterior skin should be used to close a knee disarticulation.¹⁰ Regardless of the surgical technique, it is not possible to predict all complications, and the ability of the residual limb to tolerate the forces of ambulation may be compromised by poor wound healing. An imperfect soft-tissue envelope may be somewhat mitigated in the prosthesis by the use of gel interfaces and end padding. However, thicker liners will adversely affect the overall length of the femoral section of the prosthesis and increase the knee center discrepancy described earlier. In some instances, if the residual limb is not able to provide end weight bearing, revision to a higher level may be indicated.¹¹

Most of the prosthetic benefits of this amputation level depend on the intact bony anatomy of the femur and the presence of a well-healed, scar-free, soft-tissue envelope surrounding the distal femur. When these conditions exist, it is possible for the prosthetic socket brim to be located below the level of the ischial tuberosity (Figure 3). With the ischium free of the socket, sitting comfort and freedom of hip motion can be improved. This characteristic is particularly beneficial to individuals with bilateral amputations.

The caveat of this design is that such sockets need to maintain a tight medio-lateral dimension throughout their length to control the amount of sideto-side shifting that can occur at heel strike and throughout stance. A loose mediolateral dimension will lead to poor prosthetic control, a higher risk of falls, and increased energy expenditure as users attempt to control the coronal position of the prosthesis. In an associated concern, proximal tissues must be monitored and contained as necessary to prevent the progressive development of proximal tissue bunching (Figure 4).

In instances where the residual limb is intolerant to distal load-bearing forces, the prosthesis may require a proximal brim similar to that used in transfemoral socket designs to provide proximal weight bearing and offload the distal end of the femur (Figure 5). As weight bearing is moved to the proximal brim of the socket, the biomechanical benefits of distal load bearing described earlier are reduced.