O&P Library > Atlas of Limb Prosthetics > Chapter 25

Reproduced with permission from Bowker HK, Michael JW (eds): Atlas of Limb Prosthetics: Surgical, Prosthetic, and Rehabilitation Principles. Rosemont, IL, American Academy of Orthopedic Surgeons, edition 2, 1992, reprinted 2002.

Much of the material in this text has been updated and published in Atlas of Amputations and Limb Deficiencies: Surgical, Prosthetic, and Rehabilitation Principles (retitled third edition of Atlas of Limb Deficiencies), ©American Academy or Orthopedic Surgeons. Click for more information about this text.

Funding for digitization of the Atlas of Limb Prosthetics was provided by the Northern Plains Chapter of the American Academy of Orthotists & Prosthetists

You can help expand the
O&P Virtual Library with a
tax-deductible contribution.

Chapter 25 - Atlas of Limb Prosthetics: Surgical, Prosthetic, and Rehabilitation Principles

Musculoskeletal Complications in Amputees: Their Prevention and Management

John H. Bovvker, M.D. 
Robert D. Keagy, M.D. 
Pradip D. Poonekar, M.B., B.S., M.S. (Gen. Surg.) 

One of the primary aims of any surgical procedure, apart from obtaining primary healing at the operative site, is the prevention of complications due to that procedure. Amputee management has the added dimension of complications related to the amputation-prosthesis interface. The complications of amputation surgery can therefore be divided into preprosthetic and post-prosthetic problems.

In general, the frequency and degree of complications decrease remarkably when the surgeon is keenly interested in the challenge presented by properly designed and executed amputation procedures and keeps abreast of innovations. A casual or defeatist attitude toward amputation surgery engenders many of the problems to be discussed.


Delayed Healing

Delayed healing may be related to several factors that can be operative singly or in combination. These include inappropriate amputation-level selection, sub-optimal operative technique, inadequate postoperative management, and infection.

A common cause of delayed healing is inappropriate amputation-level selection. This can, to a large extent, be obviated by proper use of the vascular laboratory as an aid in level selection. Noninvasive techniques such as segmental Doppler studies and transcutaneous oximetry mapping can yield valuable information (see Chapter 2C).

Delayed healing can also be due to suboptimal operative technique. Handling dysvascular skin with forceps, attempting to close the skin under tension, or placing excessive closure tension on muscle of questionable vascularity can result in ischemic changes leading to dehiscence. Even with minimal closure tension, skin edges may be made ischemic by the placing of too many sutures, especially mattress sutures (Fig 25-1.,A). It is better to place a few widely spaced sutures and reinforce the wound with adhesive paper strips (Fig 25-1.,B). There is also little need for subcutaneous sutures in most amputations if good myofascial and fascial closures are done. Removal of skin sutures prior to firm initial healing of the amputation wound may also lead to dehiscence, especially in the immunocompromised or dysvascular patient.

Prior to any definitive treatment of dehiscence other than debridement, the patient should be thoroughly reevaluated to determine the reason for wound failure. The preoperative vascular studies should be reassessed to be sure that the level previously selected was correct. If the patient's wound healing potential was not evaluated preoperatively, it should be done at this point. This would include a determination of serum albumin level to ascertain nutritional status and a total lymphocyte count to assess immunocompetency. If these are deficient, further surgery should be delayed until nutrition is normalized. In chronic renal failure, this may not be possible, and one may be forced to proceed without this assurance. Although it is good practice for patients to permanently discontinue the use of nicotine or at least delay resumption until the wound is well healed, it is common to find them smoking within a day or two of surgery. Rather than have another immediate failure due to smoking, the surgeon has the option to refuse surgical treatment beyond debridement on these patients if they fail to comply with this request, so long as life is not threatened. If vascular studies and nutritional parameters are normal and the patient stops the use of nicotine, treatment should proceed with the most appropriate technique.

Treatment is determined by the length and depth of the dehiscence. If skin separation is minor, the residual limb may be allowed to heal by secondary intention following conservative debridement under adequate antibiotic coverage. Often a temporary fiberglass or plaster of paris cast helps in the healing of such cases. If wound edge separation due to necrosis is confined to the skin, local debridement that avoids trauma to skin of marginal viability is called for. This may be followed by split-thickness skin grafting once adequate granulation tissue forms. Dehiscence with moderate wound separation can be managed by adequate debridement and secondary closure without tension by utilizing a minor wedge excision with minimal bone shortening (Fig 25-2.,A and B).

If infection is the sole cause of dehiscence, the wound should be widely opened for drainage and appropriate antibiotics given. Once the wound is clean and granulating well, the decision can be made to allow healing by secondary intention, with delayed split-skin grafting coverage of granulating areas, or to revise proximally and maintain the same anatomic level if an adequate soft-tissue envelope for the bone can be constructed. In the presence of gross necrosis or failure of the wound to produce adequate granulation tissue, the choice is limited to a revision amputation.

If peripheral vascular parameters are poor, before proceeding to a higher anatomic level, for instance, from transtibial to transfemoral, transcutaneous oximetry can be utilized to determine the potential for skin healing slightly more proximally in the same limb segment. Evaluation should begin with baseline transcutaneous oxygen pressure (Tcp02) determinations on room air at the site of proposed revision. If less than 40 mm Hg, the measurements may be repeated after the patient has been breathing 100% 02 at 1 atm for 20 minutes. If Tcp02 values then meet or exceed 40 mm Hg, postoperative hyperbaric oxygen (HBO) therapy may be considered. In selected cases, readings can be taken in the hyperbaric chamber while the patient is breathing 100% 02 at 2.4 atm. If Tcp02 levels are still borderline, consideration should be given to amputation at the next higher anatomic level, followed by HBO therapy if Tcp02 readings are borderline at that level.

On occasion, repeated attempts at healing a trans-femoral amputation in cases of severe dysvascularity result in wound dehiscence due to necrosis at a higher level. The next proximal level is a hip disarticulation, with no assurance that this level will heal. If necrosis then recurs, the patient is at great risk of death due to the difficulty of controlling a wound at that level without involving the pelvic structures. It is sometimes best to merely debride high transfemoral wounds in a manner so as to avoid trauma to wound edges of marginal viability. This is done by leaving a residual rim of necrotic tissue approximately 2 to 3 mm in width. In this way, the marginally viable skin beyond the necrosis is not traumatized by the scalpel. This will often stop the inexorable spread of necrosis attendant upon repeated aggressive debridement. This small rim of necrotic tissue should separate spontaneously. Gauze dressings applied moist and removed dry three times daily will encourage the formation of granulation tissue, which will lead to either healing by secondary intention or the production of a suitable bed for a split-skin graft. This is, of course, combined with improvement of nutrition and cessation of smoking. HBO treatments are again a useful adjunct in management if pretreatment test results are favorable.

In older dysvascular patients, falls in the early postoperative period are common due to problems with balance, coordination, and weakness during crutch or walker ambulation. Direct falls on a partially healed amputation wound can result in massive dehiscence and leave the bone exposed. Cleansing, debridement, and closure should be done on an emergency basis to prevent infection, flap shrinkage, and prolonged delay in prosthetic fitting (Fig 25-3.,A and B). This complication is usually prevented by application of a cast to the midthigh with the knee in full extension each week for 3 weeks. This will also allow wound inspection at weekly intervals and afford an opportunity for a full range of motion of the knee prior to application of each cast.

Skin Adherence to Bone of the Residual Limb

The ideal closure of an amputation includes the construction of an adequate soft-tissue envelope for the enclosed bone or bones. Myodesis or myoplasty are the two best techniques available to provide both distal padding and to prevent adherence of the incisional scar to the underlying bone. If the skin cannot slide over the underlying bone, it will not be able to comfortably tolerate shear forces applied by the prosthesis at the interface with the residual limb.

If wound closure is to involve split-thickness skin grafting, this should be applied only over deeper soft tissues such as muscle and not directly on bone because the graft is very likely to ulcerate as soon as use of a prosthesis is begun. Exceptions occur in upper-limb amputations, which are not weight bearing, and in children, who do surprisingly well with split-skin grafts once they have matured.

Problems in Shaping of the Residual Limb

Much can be done at the time of surgery to create an amputation stump that, by its configuration, will assist in early prosthetic fitting. On the other hand, many problems that lead to delayed application of a definitive prosthesis are a direct result of suboptimal operative technique. While it is axiomatic that wound closure tension should be minimized by designing ample myofasciocutaneous flaps, it is equally true that redundant skin and muscle can lead to slow stump shrinkage with persistent distal edema despite adequate attempts at shrinkage (Fig 25-4.,A-C).

In order to ensure optimal configuration, proper shaping of the bone end and adequate soft-tissue coverage are important. Here the role of myodesis/myoplasty cannot be overemphasized. The most structurally stable residual limbs are achieved with myodesis in which the surrounding muscles and their fasciae are sutured directly to the bone through drill holes (Fig 25-5.). In the case of transfemoral amputation, the additional advantages of myodesis are stabilization of the femur in adduction by the adductor magnus, enhanced hip flexion by the rectus femoris, and enhanced hip extension by the biceps femoris, all three being muscles that cross the hip joint.

In myoplasty, on the other hand, opposing muscle groups are simply joined to each other by sutures through the myofascia and investing fascia over the end of the bone. In a severely dysvascular residual limb with marginal muscle viability, myoplasty is probably the preferable method but should be done with little closure tension. In either case, tapering of the muscle mass avoids excessive distal bulk (Fig 25-6.).

Following amputation surgery, there is a tendency for terminal swelling to occur. Control of this edema will greatly reduce the tendency for wound complications and thus facilitate early healing. This is a major benefit of rigid circumferential dressings of plaster of paris or fiberglass. As postoperative swelling decreases, the rigid dressing may slip distally and produce a circumferential constriction with distal choking and aggravation of distal swelling. If the cast is changed promptly on loosening, this problem will not occur.

Whenever removal of a limb for malignancy is to be followed by chemotherapy, it is recommended that nonabsorbable sutures be used to oppose the fascia and other deep structures. Fascia and other soft tissues tend to heal poorly in the presence of chemotherapeu-tic agents, and soft-tissue retraction occurs if absorbable sutures are used. The overall conformation of the residual limb will then be altered, possibly leading to difficulties in fitting.

The Syme ankle disarticulation presents a special case. If the heel pad is secure and well centered, the patient will be able to tolerate a great deal of end bearing. If it is not properly anchored to the end of the tibia, it may migrate posteriorly or to one side in the socket (Fig 25-7.). If passively correctable, however, it can be held in the proper weight-bearing position by a carefully fabricated prosthesis. On the other hand, if the heel pad becomes fixed off center by contracture of the triceps surae, other muscle groups, or a scar, it should be surgically repositioned by division of the contracted tendons or scar and removal of an ellipse of excess skin, including the incisional scar, opposite the original contracture. It may also be necessary to remove a wafer of distal tibia and fibula to allow reduction of the heel pad. The plantar fascia of the heel pad may then be firmly sutured to the distal-anterior portions of the tibia and fibula through drill holes.

Prior vascular surgery may have resulted in a longitudinal scar beginning on the lower portion of the abdomen and crossing the inguinal crease into the thigh (Fig 25-8.). This presents two possible problems for the transfemoral amputee. One is scar irritation by the socket brim. The other is maintaining suction suspension, especially if the scar is depressed. Another incisional approach that results in a scar that will not cross the socket brim begins in the proximomedial portion of the thigh and extends laterally, parallel and just distal to the inguinal crease.

Repeated application of an elastic bandage has been the time-honored method of shrinking and shaping a residual limb. Unfortunately, less-than-expert application of the bandage will produce a poorly shaped residual limb, generally with distal edema (Fig 25-9.). The bandages are meant to be applied on the bias with gradually decreasing pressure as the wrapping proceeds proximally. Since layers of bandage tend to shift with movement, frequent rewrapping is necessary to avoid circumferential constriction and distal edema formation. As a result, many surgeons and prosthetists now recommend the use of an elastic shrinker sock. This sock not only is easy to don and doff but also results in the proper pressure gradient. When first fitted, it should be snug. A tuck may be sewn in the sides of the sock every 7 to 14 days to keep it snug as the residual limb decreases in volume (Fig 25-10.). Depending on limb configuration and activity level, the sock may need to be fitted with a waist belt to keep it in place. The sock is briefly removed daily for skin care. Two socks should be supplied so that a clean one can be worn each day to assist in hygienic care of the residual limb. Very muscular or obese amputees may show virtually no shrinkage of their residual limbs by wrapping or shrinker sock usage. In these cases, a temporary prosthesis or weight-bearing rigid dressing fitted as soon as the wound is sound will cause both the bulky calf and the obese thigh to shrink most rapidly. This technique may also be applied to the standard residual limb as well. This shrinkage includes the removal of edema as well as atrophy of all the soft tissues of the residual limb. Shrinkage is the greatest during the first 6 weeks of compression by prosthesis use but continues for approximately 1 year after amputation, at which time a new socket will probably be required. Shrinkage may be further enhanced by applying a snug elastic shrinker sock when the preparatory socket is removed each night. When a definite plateau in shrinkage has been reached, as determined by no further need for shrinker sock tightening or by stable weekly circumferential measurements of the residual limb, a definitive prosthesis may be fitted.


The joint immediately proximal to an amputation site tends to develop contractures if full range of motion is not initiated early in the postoperative phase. Contractures most often occur as a result of the patient keeping the residual limb in a comfortable flexed position. In lower-limb amputees, a variety of contractures may occur. These are serious complications that will interfere with proper prosthetic gait and increase the energy requirements of ambulation.

Patients with partial-foot amputations between the transmetatarsal and Syme ankle disarticulation levels are likely to develop an equinus deformity due to the relatively unopposed action of the triceps surae. This may be prevented during tarsometatarsal (Lisfranc) and midtarsal (Chopart) amputations by reattaching the extrinsic muscle-tendon units of the foot to more proximal bony structures in a balanced fashion and by lengthening the Achilles tendon percutaneously. A postoperative cast applied with the partial foot in a plantigrade position will prevent contractures until a definitive prosthesis is made. A plastic ankle-foot orthosis fitted with an anterior ankle strap can be similarly used. If, despite these precautions, a contracture later develops, a second percutaneous Achilles tendon lengthening or revision to the Syme ankle disarticulation level may be required.

Transtibial amputees, especially those with a short tibial segment, are prone to develop knee flexion contractures in the first or second week postoperatively (Fig 25-11.). For this reason, a circumferential rigid dressing of plaster of paris or fiberglass with the knee in full extension is advised until the wound heals sufficiently to allow the removal of sutures. This is replaced weekly for 3 weeks with a full range of knee motion at each change. The patella should be well padded to prevent pressure necrosis of the prepatellar skin. Even with a cast in place, pillows should not be placed under the residual limb, or a hip flexion contracture may be encouraged.

Severe knee flexion contractures are virtually impossible to reduce by exercise once they become fixed. In amputations not done for vascular insufficiency, hamstring lengthening and release of the posterior knee joint capsule should be considered. The dysvascular amputee with a short contracted residual limb may be fitted with a bent-knee prosthesis, which is functionally no better and cosmetically inferior to that for a knee disarticulation (Fig 25-12.). Occasionally, moderate knee contracture in a proximal-third amputation may be improved by fitting a prosthesis with the foot in slight equinus to provide a knee extension moment on foot contact.

At the transfemoral level of amputation, a hip flexion-abduction contracture can be devastating because the already high energy requirement for ambulation at this level is further increased by contracture. Again, prevention is the key. During transfemoral amputations, flexion-abduction contracture can be discouraged by a balanced myodesis, including reattachment of the adductor magnus tendon to the lateral aspect of the femur as it is held in adduction and extension (see Chapter 20A). Postoperatively, pillows under the thigh are forbidden. Within a few days of surgery, the patient should be taught to lie prone for 15 minutes three times a day to stretch out any early flexion contracture and to actively adduct the residual limb to prevent abduction contracture. Active extension of the residual limb while flexing the opposite thigh to the chest is also effective. Excessive wheelchair use, which encourages contractures, is discouraged by early walking with crutches or a walker.

An alternative approach may be used in anticipation of prosthetic use by vigorous transfemoral amputees. A unilateral hip spica is applied in the operating room immediately following wound closure. Application is easily done by lifting the patient by the opposite leg. The benefits are analogous to those achieved by postoperative casting of transtibial amputees. The hip is aligned to discourage contractures, distal constriction edema from bandaging is avoided, and the wound is protected from shear and direct pressure, thereby reducing pain. At the first cast change, a pylon and foot may be added to convert it to a preparatory prosthesis.

At the short transfemoral level, flexion contracture of up to 25 degrees may be accommodated by prosthetic alignment, but hip extensor power, needed for good prosthetic knee stability, is compromised. As one progresses distally to the midthigh level, it is increasingly difficult to compensate prosthetically for a hip flexion contracture. Even then, the resulting cosmesis of the prosthesis will leave something to be desired. More than 15 degrees of hip flexion contracture will require a marked compensatory increase in lumbar lordosis that, even if available, may lead to low back pain.

When prescribing a prosthesis in cases with significant flexion contracture of the hip or knee, the patient and family must be forewarned of the relatively grotesque appearance of the prosthesis. Otherwise, if the patient and family do not understand the rationale for this initial fitting in the hope that prosthesis usage will tend to decrease the contracture, they may be very dissatisfied with the prosthesis and reject it.

In children, knee and hip flexion contractures can be stretched out by ignoring their presence and fitting the patient with conventional alignment techniques. Spontaneous use will usually stretch the contractures without other special treatment.

Contractures also occur in upper-limb amputations. Limitation of glenohumeral abduction and forward flexion is common in short transhumeral amputations. Elbow flexion contracture occurs readily in a short transradial case. Either can be easily prevented by instituting range-of-motion exercises as soon as postoperative pain has subsided at 5 to 7 days. Gentle muscle-strengthening exercises begun at 2 to 3 weeks postoperatively are also helpful. If contractures become fixed, even an extensive program of stretching may be ineffective and require selective release of contracted muscles to allow fitting of a prosthesis.

Chronic Wound Sinus

The patient who appears with a small adhesive bandage on his wound with minimal drainage has a problem that may or may not be related to one of the many conditions mentioned above. This small opening may only lead to a superficial suture abscess, or it may be a sinus related to a bone spur or low-grade, localized osteomyelitis. Since a sinus is likely to become inflamed or infected, it is useful to know that it is present. A good way to determine this is to probe the opening with a malleable metal probe or a flexible polyethylene intravenous catheter after antiseptic skin preparation. Radiographs will help to determine bone involvement. A sinus is best managed surgically during a benign, nonacute interval. A sinogram followed by excisional surgery can be curative.


Painful Residual Limb

In this section we will be dealing with causes of residual-limb pain other than phantom pain. Pain in a residual limb can have as many causes as pain in an intact limb. Taking an adequate history and performing a physical examination continue to be appropriate, with inspection, palpation, performance tests, radiography, and other imaging studies used as necessary. Since all persons with acquired amputations have phantom-limb sensation, there is a tendency for patients, families, and physicians not familiar with amputees to consider all pain in the residual limb to be phantom pain. In fact, chronic phantom pain is very rare and presents a specific syndrome of a totally preoccupying pain in a stocking-glove distribution. Another equally poor presumption is that all pain problems are related to a poorly fitting prosthesis, and this results in the making of many unnecessary new limbs.

Preamputation traumatic disruption of the interos-seus membrane and proximal tibiofibular joint, with resultant hypermobility of the fibula, can be a cause of pain in the transtibial amputee. Ertl has recommended distal tibiofibular synostosis to prevent or treat this hypermobility. Fusion of the proximal tibiofibular joint is a much simpler procedure that does not require shortening of the residual limb or disturbance of the otherwise satisfactory distal tissues.

A common site of discomfort and skin breakdown in a transtibial amputation is over the distal-anterior part of the tibia. This is commonly due to inadequate contouring (beveling) of the tibia at the time of amputation. Although this can often be corrected by local socket relief, with or without injection of the local bursa with steroids, surgical revision may be required.

Symptomatic spurs from ectopic bone formation of the transected bone may be generated in one of two ways. One is by surgical extension of periosteal stripping onto bone that is to be retained (Fig 25-13.). The periosteum should therefore be disturbed as little as possible on any bone that one anticipates keeping. In addition, the approach to the fibula should be extra-periosteal to avoid fibular regeneration. Widespread ectopic bone formation also occurs when the trauma leading to amputation has resulted in wide stripping of periosteum from the bone that is to be retained. This can be quite massive and may lead to considerable discomfort in the residual limb. Before proceeding to revision, the surgeon should determine exactly those portions that are contributing to discomfort and not disturb all ectopic bone just because it is present. Following any bone transection, the soft tissues should be copiously washed to remove minute bone fragments.

If the fibula is inadvertently left longer than the tibia, the resulting distal bony prominence may be weight bearing and tender. Socket modification should be attempted, but surgical revision is often necessary. In very short transtibial residual limbs, if the fibular head and shaft are unduly prominent or hypermobile, the fibular remnant may require complete secondary resection (Fig 25-14.). This complication is prevented by routine primary excision of the fibular remnant if amputation is close to the tibial tubercle. In the transtibial amputee, torn knee ligaments may result in painful instability while wearing the standard patellar tendon-bearing (PTB) prosthesis. Depending on the degree of instability, it may be controlled by using a supracondylar-suprapatellar or supracondylar prosthesis or a PTB prosthesis with a thigh corset.

In transfemoral amputations, if a dynamically balanced myodesis has not been performed, the femur may drift anterolaterally through the soft tissue to present its distal end subcutaneously (Fig 25-15.). This will produce local tenderness and even ulceration with or without use of a prosthesis. In the absence of ulceration, prosthetic modifications including socket relief over the bony prominence or anterior filling-in of the socket just above the prominence may be effective. If simple socket adjustments do not produce relief, surgical revision, including myodesis, may be necessary. Pain-producing bone spurs may develop at the cut end of the femur and require similar socket relief or excision (Fig 25-16.). Adventitious bursae develop over bony prominences and occasionally need treatment beyond socket relief.

Transfemoral amputees may complain of a burning sensation in the ischial weight-bearing area, particularly in the early phases of using a quadrilateral socket. With the increasing utilization of ischial containment sockets and the advent of flexible socket materials, pressure discomfort over the ischium is less common. Nonetheless, the residual limb should be carefully examined at each visit for areas of local inflammation secondary to excessive pressure.

Neuroma formation is a natural consequence of nerve section, and all amputees will therefore have several neuromas. If nerves are divided at a level that avoids both inclusion in the wound scar and weight-bearing or other significant pressure from the prosthesis, they will rarely be symptomatic. If a symptomatic mass is suspected of being a neuroma, direct manipulation should produce a Tinel's sign with a tingling discomfort in the peripheral nerve patterns of the missing portion of the limb. If a neuroma is incidentally found by palpation in an asymptomatic residual limb, the amputee should be reassured that it is a normal finding and does not require corrective surgery. Firm fibrous nodules that are only locally sensitive are probably not neuromas. The treatment of neuromas should usually begin with socket accommodation. If this approach is unsuccessful after several attempts, the neuroma should be moved to a deeper site, either by proximal division under moderate tension or by placing the nerve end in bone. This is certainly the case when the neuroma is directly over bony prominences where pressure from use with or without a prosthesis is unavoidable, such as in the hand at the metacarpal heads or at the neck of the fibula. In the case of a very short transtibial limb with a symptomatic peroneal neuroma, the neuroma can be easily deafferented by removal of peroneal nerve proximal to the knee through an incision posterior to the distal portion of the biceps femoris muscle.

In some transtibial amputees who have had traumatic disruption of the interosseus membrane and subluxation of the proximal tibiofibular joint, there may be ill-defined pain related to fibular hypermobility producing pressure on the peroneal, tibial, and sural neuromas. Ertl advocated a distal tibial fibular fusion to correct this problem and provide an "end-bearing" bone. It is easier to simply fuse the proximal tibiofibular joint and not shorten the residual limb or disturb its distal soft tissues.

In dysvascular patients, a healed skin wound may be associated with considerable ischemia of the underlying muscles and result in intermittent claudication during walking. In these cases, there will be no signs of inordinate prosthetic pressures, and the pain will occur regularly when the patient walks a specific distance. Medication may be prescribed, but the amputee should be assured that refabricating the prosthesis will not help.

Whenever late pain occurs in a limb amputated because of tumor, local recurrence is a possibility. The proper course of action depends upon the type of tumor and may range from radiation or chemotherapy to amputation at a higher level. Consultation with an oncologist is essential before proceeding.

Adherence of Skin to Bone

In any residual limb, but especially at the transtibial level, the adherence of skin to bone or the application of split-thickness skin grafts directly to bone may rapidly lead to pain and ulceration when a prosthesis is used. This occurs because skin without underlying soft tissue has little resistance to direct or shear forces imposed by the prosthesis. In an attempt to avoid revision, the amputee can be taught to mobilize tissue by gentle persistent fingertip massage on a daily basis over several weeks as a routine part of self-care. A gel socket may be useful. A nylon sheath should be provided to reduce shear forces. If a transtibial amputation stump continues to show skin breakdown despite competent socket adjustments, other methods may be helpful. The addition of a rotator unit will decrease rotational shear forces. Partial unloading of the transtibial level by the addition of knee joints and a long thigh corset or the use of ischial weight bearing may be necessary to allow continued ambulation.

Surgical revision is done as a last resort and will usually require only a local wedge excision, including a small length of bone. In the case of adherent split-thickness skin grafts, often a relatively small area of adherent split graft can be excised, with primary closure effected by advancement of surrounding skin and subcutaneous tissue once muscle atrophy has resulted in relative skin redundancy at 10 to 12 months (Fig 25-17.,A and B). Amputation to a higher anatomic level is rarely indicated.

Other sites where split-thickness skin grafts may not stand up to prosthetic use is over the adductor longus tendon in the groin, at socket brims over the biceps tendon in the antecubital fossa, and in the anterior axillary fold in relation to transhumeral sockets.

Skin stretching, either by traction or by tissue expanders, is probably the best way to develop enough normal skin to eliminate large areas of split-skin grafts. To cover the adductor tendon region in the groin, however, it is usually sufficient to swing a small flap distally from the abdomen downward or proximally from the thigh.

Insensitive Skin

Amputees with diminished sensation in the residual limb are seen quite commonly. The largest group are diabetics, but other neurologic disorders such as my-elomenigocele, Hansen's disease, and alcoholic neuropathy are also seen. These patients are not deterred by pain from continuing to walk on a locally ischemic or ulcerated residual limb and must be taught to remove their prostheses at regular intervals for skin inspection, especially during the early phases of prosthetic use. Areas of skin blanching and/or erythema should be noted by the prosthetics team for prompt corrective action. Multiple short periods of daily ambulation will usually allow gradual skin adaptation. The presence of insensate but otherwise normal skin on the residual limb should not be considered an indication for more proximal amputation. It does, however, demand precise prosthetic fitting and attentive follow-up care.

Poor Fit

After a variable period of use, most amputees find that prosthetic fit can no longer be effectively adjusted by further socket padding and additional stump socks. With an excessive number of sock plies, usually 10 to 15, the socket/residual-limb interface is disturbed, as manifested by a reduction in rotational control and an increased tendency to piston. In these cases, the prosthesis no longer fits properly, and a new socket is needed promptly if dangerous, costly skin breakdown is to be avoided. On the other hand, many problems are easily corrected with minor sock or socket adjustments. All team members should therefore be aware of the signs of both loose and tight socket fit, especially at the transtibial level. Evaluation of a residual limb for prosthetic pressures is exactly the same as evaluation of a foot for shoe fitting. One looks for areas of prolonged erythema after walking in the prosthesis, erythema in abnormal places, callus or bursa formation, and local tenderness under erythematous areas.

Relative socket looseness will commonly cause excessive direct and shear forces over the tibia and fibula, fibular head, tibial tubercle, and distal end of the patella as the residual limb enters the socket too deeply. This problem is usually related to residual-limb volume decrease by atrophy or weight loss. Relative socket tightness will cause direct tibial tubercle pressure on the patellar tendon bar and verrucous hyperplasia of the limb end due to loss of distal contact. This problem is often related to wearing excessive sock plies or due to weight gain. Pressure and shear forces result in inflamed and/or ulcerated areas of skin in either case.

Another transtibial problem of fit related to distal circumferential shrinkage is usually associated with ill-defined pain in the residual limb. In this case, the amputee has good suspension at the socket inlet but relative freedom of motion distally so that the residual limb moves inside the socket like a clapper in a bell and strikes the anterior socket wall each time the knee is extended during swing phase. There is no sign of inordinate prosthetic pressure, but it may be noted, during donning or doffing of the prosthesis, that while there is a snug fit proximally, there is room distally for an examining finger or that a soft insert feels loose. A weight-bearing radiograph of the residual-limb/socket interface is useful to confirm the presence of a distal void. Often, the same situation leads to choking. This problem may sometimes be corrected by filling in the socket posteriorly.

Lower-limb edema resulting from renal and/or cardiac disease will adversely affect socket fit. If these amputees are unable to use their prosthesis for any reason, such as any sudden change in their health, it may be impossible to get the socket back on. It is extremely important that they have appropriate shrinker socks to wear in bed. If they are admitted for treatment of their underlying condition, compression of the residual limb should be started promptly while in the hospital rather than being neglected for a period of several days. The edema can become relatively chronic, and resumption of ambulation can be very difficult as one struggles to shrink the residual limb again.

Bony Overgrowth in Children

The traumatic transosseous child amputee may experience rapid growth in length of the residual limb to the point where the bone grows through the skin. This is appositional bone growth and is not related to physeal growth, even though bony overgrowth usually ceases when physeal growth ceases. It is often seen in the humerus, fibula, tibia, and femur in that order of relative frequency (Fig 25-18.). This may occur several times during childhood and is easily treated by resection of sufficient bone to allow coverage with an adequate soft-tissue envelope. Caps, plugs, chemical cautery, or electrocautery have not proved to be consistently useful in controlling overgrowth. Proximal epiphysiodesis is con-traindicated since this has no influence on distal appositional bony overgrowth and will lead to unnecessary shortening.

Degenerative Arthritis

Since most people who sustain amputations are middle-aged to elderly, some will have arthritis of the joints proximal to the site of amputation. Superimposed ambulation with a prosthesis may put more strain on the proximal joints, thereby contributing to arthritis pain. Arthritis of the hip joint in the transfemoral amputee may be alleviated to some degree since the trans-femoral socket bypasses the hip joint by utilizing a portion of the pelvis for weight bearing. Also, the lightest-possible prosthesis should be fabricated to require less forceful contractions of the muscles crossing the hip joint, thus reducing joint compression forces. If pain is not relieved, a total-hip arthroplasty should be considered to maintain function in a prosthesis user. Likewise, transtibial amputees with significant symptomatic hip joint arthritis should not be denied the benefits of hip joint arthroplasty if it is otherwise indicated. Weight-bearing pain in the knee secondary to femo-rotibial joint arthritis may be partially relieved by the addition of knee joints and a thigh corset to allow shared weight bearing between the residual limb and the thigh. Patellofemoral arthritis has not proved to be a major concern. In cases of internal derangement of the knee joint, arthroscopic evaluation and surgery should be considered.


Although uncommon, fracture in a residual limb following amputation does occur sufficiently often to warrant a careful design of treatment methods to allow an early, effective return to prosthesis use. By applying current knowledge of the gait cycle and energy expenditure in lower-limb amputees, certain goals in the treatment of late residual-limb fracture become clear.The general principles of fracture management, however, remain the same as in any other individual, but a different approach is allowed due to the reduction in distal limb segment mass and lever arm length.

A combined American and Canadian study produced 90 cases with sufficient information to provide both epidemiologic data and some specific recommendations for management. The average age at injury was 50 years, with a fall while wearing the prosthesis as the usual cause of injury. It was notable that knee joints and a thigh corset did not prevent supracondylar fractures in transtibial amputees, nor did a hip joint with a pelvic belt prevent fractures about the hip in transfem-oral amputees.

One important goal in the treatment of intertrochanteric fractures that applies to both transtibial and trans-femoral amputees is the restoration of a normal neck-shaft angle to restore hip abductor function. Although manipulation and casting often suffice in two-part intertrochanteric fractures, those amputees with unstable fractures are best served by open reduction and internal fixation. Displaced femoral neck fractures in both groups may be managed either by reduction and internal fixation or by endoprosthetic replacement. Excision of the femoral head alone will lead to an unstable gait. Instead, femoral endoprosthetic replacement or total-hip arthroplasty may be undertaken based on the same criteria as in any patient with otherwise intact limbs.

Because of the small residual-limb mass and lever arm length in transfemoral amputees, most nondis-placed peritrochanteric fractures and shaft fractures can be successfully managed by non-weight bearing alone or minispica casts after appropriate manipulation of malaligned fractures.

In transtibial amputees, preservation of knee motion and restoration of limb alignment, especially in more proximal femoral fractures, are paramount. Patients with stable supracondylar femoral fractures can be mobilized rapidly by the use of the cast-brace technique (Fig 25-19.). Unstable supracondylar fractures should be fixed primarily, if possible, to preserve knee motion. Severely comminuted supracondylar fractures unsuitable for fixation may be managed by casting with or without preliminary skeletal traction and/or manipulation (Fig 25-20.,A and B). Moderate malunion or loss of length at the transtibial level is easily compensated by prosthetic adjustment, but an effort should be made to avoid flexion contracture of the knee, which is much less compensable (Fig 25-21.). In displaced intra-articu-lar fractures of the knee, joint congruity should be restored as accurately as possible.

In this study, transtibial amputees were more likely to resume the use of their prosthesis than were trans-femoral amputees due to lesser energy demands. Operative scars did not interfere with the fitting or use of prostheses. Only 25% required a prosthesis modification following fracture, and all of these were transtibial amputees. Proximal revision of amputations through the fracture site was not found to be necessary or desirable.

Fractures of residual upper limbs are very rare. It is recommended that humeral fractures be treated by splinting. If delayed union or nonunion ensues, open reduction, internal fixation, and bone grafting should be considered, especially in transradial amputees. Fractures about the elbow may be managed by open or closed methods so long as treatment is designed to maintain elbow range of motion. In summary, good results in the management of fractures of residual limbs may be expected if they are treated with the same care and expertise accorded fractures occurring in intact limbs.


Knowledge of the common complications of amputation surgery should lead to their prevention in most cases and their speedy resolution when they occur. The primary goal of complication prevention and management is the successful prosthetic restoration of the amputee. This chapter provides numerous examples of preprosthetic and postprosthetic problems and their possible solutions. The amputation surgeon, working with a dedicated prosthetic team, will be able to use these proposed solutions as a creative starting point to upgrade the care of amputees in the local community.


  1. Bowker JH, Kazim M: Biomechanics of ambulation, in Moore WS, Malone JM (eds): Lower Extremity Amputation. Philadelphia, WB Saunders Co, 1989, pp 261-273.
  2. Bowker JH, Rills BM, Ledbetter CA, et al: Fractures in lower limbs with prior amputation. A study of ninety cases. J Bone Joint Surg [Am] 1981; 63:915-920.
  3. Dickhaut SC, DeLee JC, Page CR: Nutritional studies: Importance in predicting wound-healing after amputation. J Bone Joint Surg [Am] 1984; 66:71-75.
  4. Ertl J: Uber amputationsstumpfe. Chirurgie 1949; 20:218.
  5. Hauser CJ: Tissue salvage by mapping of skin surface transcutaneous oxygen tension index. Arch Surg 1987; 111:1128-1130.
  6. Lind J, Kramhoft M, Bodtker S: The influence of smoking on complications after primary amputations of the lower extremity. Clin Orthop 1991; 267:211-217.
  7. Matos LA: Enhancement of healing in selected problem wounds, in Mader JT (chairman): Hyperbaric Oxygen Therapy. A Committee Report. Bethesda, Undersea and Hyperbaric Medical Society, 1989, pp 37-44.
  8. Matos LA: Personal communication.
  9. Wagner FW Jr: Amputations of the foot and ankle, in Moore WS, Malone JM (eds): Lower Extremity Amputation. Philadelphia, WB Saunders Co, 1989, pp 93-117.
  10. Watts HG: Special considerations in amputations for malignancies, in Atlas of Limb Prosthetics. St Louis, Mosby-Year Book, 1981, pp 459-463.

Chapter 25 - Atlas of Limb Prosthetics: Surgical, Prosthetic, and Rehabilitation Principles

O&P Library > Atlas of Limb Prosthetics > Chapter 25

The O&P Virtual Library is a project of the Digital Resource Foundation for the Orthotics & Prosthetics Community. Contact Us | Contribute