New Prosthetic Alignment Challenges

Gerald Stark, MSEM, CPO/L, FAAOP

Compared to older exoskeletal construction, modernendoskeletal componentry offers both benefits andconstraints to proper alignment. In many instances current suspension and componentry dictate the alignment of the prosthesis rather than the patient presentation with adverse affects to alignment. Although materials and component design have greatly changed, alignment challenges persist and are more pronounced with more active users who demand not only stability, but optimized movement. Alignment parameters established historically by Radcliffe, Foort, Inman, Hampton, McClaurin and others should be revisited to avoid common gait deviations and provide the basis for changing componentry properties.

Unfortunately many of these principles, proved valuable with empirical clinical observation, have been largely forgotten and gait deviations, unacceptable in earlier times such as uneven step length, hyperstability, lateral trunk bending, abducted gait, and rotational whips, have reemerged. Alignment for exoskeletal systems had to be established with strict adherence to established principles since changing the alignment later was difficult and costly. Currently bench alignment is not as stringently observed since it is perceived that endoskeletal componentry will allow for correction later in the fitting. This freedom of adjustment has not translated into better alignment, but primarily into speedier fittings. Unfortunately endoskeletal systems do not have the range of adjustment or easy linear capability to "dial" in the alignment. This explains the popularity of slide and attachment devices which promise to compensate for poor bench alignment and achieve acceptable rather than optimal placement. Endoskeletal componentry, which also promised to lighten the prosthesis, has gradually increased in weight with extra componentry. Compared to the 2 ½-3 lb. transtibial proposed by Een-Holmgren and Fillauer, the modern lower limb prostheses easily weigh much more. Combined with poor alignment, this would indicate that endoskeletal use is not fully optimized.

Although many patients prefer dynamic response feet, this has not been corroborated by laboratory by numerous studies. Roll over shape, a principle first discussed by Hansen et al at Northwestern University Prosthetic Research Laboratory, may help explain the advantage as a certain roll over late in stance 1 . The center of pressure can be easily plotted on a force plate then placed in relation to the ankle-foot or the knee-ankle-foot. An arc of motion can be defined which describes motion late in stance 1 . Different foot designs were shown to have different rollover shapes. In a sense the SACH foot represented the first rollover shape with subsequent designs influencing that shape with keels of various stiffnesses due material and geometry design. What was most interesting is that the various prosthetist equalized the rollover shape to have them be very similar 1 . Rollover shape also showed climbing a ramp to be primarily a function of the ankle-foot and going down an incline to be a function of the knee-ankle-foot. The relative heel height had little effect on the rollover shape except with extreme high heels of 50-60mm 2 . In a related study AFO's had the effect of elongating the length and radius of the rollover shape.

Components that combine the suspension pin lock with the distal attachment have, in effect changed alignment. The suggested posterior placement of the transtibial foot is 18-65mm with a bench alignment for a SACH foot at 37mm 9 . Since the distal attachment is in line with the midpoint this has shifted the alignment more anterior. Although this is more acceptable with dynamic response feet that flex more late in stance, this may demand that the foot be dorsiflexed for the more anterior position resulting in an excessive toe clearance at heel contact. This can be compensated with a slight anterior lean as a result of a linear A-P adjustment. While this adjustment is not often done it would help to adjust for the relative keel stiffness of the foot. This would also have the effect of shifting the roll-over shape anterior or posterior. Another biomechanical principle is to dorsiflex the transtibial prosthesis to unweight the heel. This is especially relevant for heavy heel walkers who have a tendency to crush the heel. Dorsiflexing the foot helps initiate roll over sooner before excessive pressure is observed 9 .

Shock absorbers as they are termed in prosthetics are primarily shock dampers 11 . A spring mechanism is the primary shock absorber and the damper works to slow the response of the spring. These devices are especially relevant to replace the absence of the ankle and knee joint along with transverse rotation. In earlier times a SACH foot with thicker foam rubber in the heel presented with substantial shock absorption. Newer dynamic response feet have less since the heel lever is longer or the heel material has been minimized. Linear shock absorption is relatively non-physiologic in that the limb telescopes rather than using rotary joint motion. Gard has shown that true shock absorption really only takes place with faster walkers at higher speeds generating greater force 11 .

Coronal plane alignment has also changed as a result of combined suspension/attachment components. The relative inset guideline has changed from 0-12 mm inset 9 depending on the limb length to very little, if any, inset. The inset that is present is usually dependent on the varus presentation of the distal end since most technicians simply position the attachment at the distal end apex. The main challenge is that alignment has become less narrow which for the most part is acceptable for greater stability, but sacrifices a much more narrow energy efficient, cosmetic gait. The relative inset should be measured from the MTP level not the distal end as is common practice. Empirically the distal end is usually more medial and "hangs off" of the foot shell with longer limb lengths or Syme's ankle disarticulations.

In general transfemoral alignment has over emphasized involuntary knee stability with respect to alignment and knee design. Transfemoral alignment has also been compromised with sleeve suspension/distal attachment components. Pre-flexion of the interface, so essential for normal step length, is often not present unless using a series of offset plates. This remains a critical factor when using microprocessor controlled knees that depend on the accurate assessment of the reaction line. If the interface is not preflexed their may be a greater tendency to unlock stance beyond the knee's capability to adjust stance. Preflexion must be 5° in addition to the hip flexion contracture 9 . In normal human locomotion the pelvis lordoses 3°, 5° hip extension, and 15° knee flexion for a total of 23°. The 5° pre-flexion compensates for the patients inability to provide hip extension. Pelvic lordosis must compensate with an increased 10° and 5° preflexion amounting to 15°. This may be one reason active, short, transfemoral amputees and hip disarticulation patients often experience lower back pain.

Although the Berkeley and European plum line alignment methods are utilized the nuances often obviate the original intention of the methods. Radcliffe recommended the Berkeley method originally with the Quadrilateral socket. Never intended to be an emulation of the reaction line, the Alignment Reference Line was a convention to provide stability. The proximal mark can also be approximated with the medial bisection. The knee pivot should fall 3-6 mm posterior to the projection of ankle bisection 4 . Unfortunately the medial bisection is difficult to assess during dynamic alignment. The lateral bisection is slightly different due to the external rotation of the knee. The knee center should then be placed 6-10 mm posterior. The Berkeley method did not employ anterior placement of the knee, but rather anterior placement of the interface. The European Alignment method utilizes a plum line from the bisector with the knee 6-10 mm posterior (depending on knee design) and the midpoint of the foot 10-25mm anterior (also depending on the foot design). This creates an anterior lean of the pylon possible only with a distal pyramid. The Berkeley method sets the knee center in alignment and increases stability with increased preflexion and knee design. The method compensates for the stiffness of the keel and incorporates the slight "safety factor" of plantarflexion also advocated originally by the European method. Radcliffe remarks that the ultimate alignment arrived at should look very similar in both cases.

Radcliffe emphasizes the importance of preserving voluntary control with the transfemoral prosthesis. Through hip extension and limb length the patient influences the placement of the alignment line. At initial contact the patient uses their limb length and hip musculature to influence the position of the reaction line. A patient with strong musculature and longer limb length can shift the reaction line more anteriorly. Conversely the late in stance the amputee can shift the reaction line anterior to flex the knee. Some knee designs, very stable by design, require an extra toe moment to aid in the flexion of the knee.

The Zone of Stability and Control proposed by Radcliffe projects the reaction line in early stance and late in stance. The area between these two lines represents the area in which the knee center may be placed to insure stability early in stance and easy flexion late in stance. The area is greater more proximally making the use of polycentric knees, which place the knee center proximal, advantageous. The patient with good voluntary control can influence this area by shifting the reaction lines to increase stability and control 4 .

Knee stability is frequently increased by shifting the knee center more posterior. This increases the relative involuntary locking of the knee flexion late in stance 5 . At the same time moving the knee posterior is counteracted by also moving the foot posterior increasing knee flexion moment. Plantarflexing the foot has the effect of shifting the Zone of Stability and Control anterior. The foot reaches foot flat faster and then pops up on the metatarsal heads to allow for easy knee flexion late instance to provide better stability without sacrificing voluntary control. This heel airspace may feel strange to the patient during standing at first but provides 3-6mm of posterior knee stability for every 1° of plantarflexion with out sacrificing voluntary control 5 .

Ivan Long in the late 1970's advocated strong femoral adduction as the goal of narrow M-L type interface designs. Although different nuances on this principle have emerged, Mr. Long reminds of the importance of adduction the limb. His use of the mechanical axis line utilized by orthopedic surgeons advocated a 4° line from the mid femoral neck to the bisection of the femur. With shorter limb lengths this can be difficult to achieve. Uellendahl established a simple parameter of having the medial 25mm brim intersect the medial socket distally. All to often the line of femoral adduction is established by the distal attachment which is not easily offset with three prong attachments. If the correct adduction is established the foot may be too far inset. Many times a linear shift is required. Technicians will often "correct" socket adduction by making the socket more vertical.

The transverse alignment is important not only to eliminate whips but also the rotation of the foot late in stance affecting the relative keel loading. The knee and the foot are to be 5° externally rotated to approximate the 15° of external rotation of the trailing limb to allow for the transverse rotation of the pelvis. A common error is to not have the foot and knee in the same plane. Although not a classic whip, which is a result of improper knee rotation, the foot and knee appear to track uncosmetically in different arcs. Coronal alignment should be 0-65 mm outset from the ischium 8 . A knee disarticulation can be placed at 0 mm while a relatively short limb length may be 65 mm laterally displaced. Often the foot is set too far outset because of the distal attachment. In many cases involving geriatric patients to error with stability is satisfactory. When optimizing movement the patient should be challenged gradually to have a less outset presentation. This demands a linear shift or successive alignment goal with each interface.

Hip Disarticulation alignment can be difficult due to the lack of established anatomic landmarks. In the sagittal plane the hip-knee projection should be place 2550 mm posterior to the heel to insure a knee extension moment. Coronally the limb is placed 10 mm lateral to the lateral 1/4 mark 6 . Componentry available makes it difficult to find the proper placement of the ab/ adduction, transverse rotation, and flexion. Forming blocks can be used with the anterior block externally rotated 3° but these often deform the interface. Often the attachment must be repeatedly cut and pasted.

The Fillauer Companies, Inc.; Chattanooga, Tennessee


    1. Hansen, A., Alignment of Trans-tibial Prostheses based on Roll-over Shape Principles, Prosthetics and Orthotics International, Vol. 27., No. 2, 89-99.


    1. Hansen, A., Effects of shoe height on biologic rollover characteristics during walking., Journal of Rehabilitation Research and Development., Vol. 41, No. 4, 547-554.


    1. Radcliffe CW, Foort J. The patellar-tendon-bearing be-low-knee prosthesis. Berkeley (CA): University of California, Biomechanics Laboratory; 1961.


    1. Radcliffe, CW, The Knud Jansen lecture: Above-Knee Prosthetics. Pros and Orth Int 1977; 1:3:146-60.


    1. Radcliffe, CW, Hip-Knee-Foot Interaction with Four-Bar Prosthetic Knees, Annual Scientific Meeting of the International Society for Prosthetics and Orthotics, Australian National Member Society National Centre for O & P, Latrobe University, Melbourne Australia, November 27-29 2003.


    1. Hampton, F., McLaurin, C., Diagonal Hip Disarticulation Method, Northwestern University Prosthetic Research Center, V.A.-V1005 M 1079, Chicago, Illinois, May 1962.


    1. Long, I., Normal Shape ,Normal Alignment (NSNA) Above Knee Prosthesis, Clincial Prosthetics and Orthotics. 9, 1985, pp.9-14.


    1. Northwestern University Transfemoral Fitting Manual, Uellendahl, J, Edwards, M.. Ischial Containment Above-Knee Prosthesis. Northwestern University Prosthetic Certification Program, 1998.


    1. Edwards, M., Northwestern University Prosthetic Program, Transtibial Manual, Chicago, Illinois, 1996.


    1. Lightweight Unitized Endoskeletal Prosthesis-BK (L.U.E.P.-B.K), Durr-Fillauer Medical, Inc, 1991.


  1. Gard, S., The Influence of Prosthetic Shock Absorbing Pylons on Transtibial Amputee Gait. Journal of Proceedings, American Academy of Orthotists and Prosthetists, 2001.