Chapter 18A - Atlas of Limb Prosthetics: Surgical, Prosthetic, and Rehabilitation Principles
Transtibial Amputation: Surgical Procedures and Immediate Postsurgical Management
John H. Bowker, M.D.
Bertram Goldberg, M.D.
Pradip D. Poonekar, M.B.B.S., M.S. (Gen Surg)
Among major amputations in the lower limb, the transtibial (below-knee) amputation is the most common. Many series report a ratio of at least two transtibial amputations to every transfemoral (above-knee) one. It is important to note that it is the most proximal level in the lower limb at which near-normal function is available to a wide spectrum of lower-limb amputees. This is because energy consumption for the transtibial amputee, due to preservation of the knee joint, is far less than for amputees with a transfemoral level. The relative ease of transtibial vs. transfemoral gait is borne out by several studies of prosthesis usage. Combined data from 13 studies from 1943 through 1983 showed an average transtibial prosthesis usage rate of 73.5%. In contrast, analysis of four studies covering the same period disclosed that transfemoral prosthesis usage averaged only 26.5%. Most of the patients in these studies had peripheral vascular disease. Another detailed study was made of 25 unilateral transtibial amputees who were all under the age of 45 years at the time of amputation for trauma. They were reviewed 2½ years following surgery as regards their function and life-style. Eighty-four percent wore their prostheses more than 13 hours a day, 72% could walk a mile if necessary, and 84% drove automobiles. Sports were played by 72%. The most notable finding was that 84% of these unilateral transtibial amputees regarded themselves as minimally or nondisabled.
Another singular advantage of transtibial over transfemoral amputation is markedly reduced perioperative mortality. The combined mortality of three studies for transtibial amputation was 9.5% as compared with 29.7% for transfemoral amputation. Virtually the same findings were reported by Sarmiento and Warren, who noted a fall in mortality rate from 24% to 10% that was directly related to the reversal of their transtibial-to-transfemoral ratio from 1:2 to 2:1.
For many years, transfemoral amputations were preferred to transtibial ones because it was felt that primary healing is easier to obtain at the thigh level. Healing at that level, however, is far from certain. Boontje, in a series of 171 amputations, noted a 28% failure of transfemoral healing as compared with 35% for transtibial cases. By pooling the data from eight reported series totaling 942 cases, each with at least 50 patients, it was found that 70% of transtibial amputations healed primarily and 16% secondarily for a total healing rate of 86%. These series did not separate diabetics with or without ischemia from those with ischemic disease alone.
It was long taught that diabetics should have a primary amputation at the transfemoral level because of their supposed inability to heal at more distal levels. Data were combined from four series that compared the healing rate of transtibial amputations in diabetics with that in patients with purely ischemic disease. Of 194 diabetic patients, 92% healed their wounds. In contrast, only 75% of 188 patients with purely ischemic disease healed. Two additional series of 100 diabetics each reported transtibial healing rates of 99% and 90%, respectively. These studies strongly suggest that the notion that diabetics do best with a primary transfemoral amputation for foot lesions should be discarded.
In ischemic conditions, unilateral transtibial amputation may be followed by loss of the opposite limb with progression of vascular disease. One study of 80 patients noted an interval of 23 months, on average, between transtibial amputations. The risk of contralateral limb loss is 10% per year. With sufficient longevity, therefore, transtibial amputees often face the prospect of opposite lower-limb loss. The chances of ambulation as a bilateral transtibial amputee therefore become a major concern. Pooled data on 137 patients showed that 77% of bilateral transtibial amputees were able to attain functional ambulation.
In summary, the importance of preserving the knee joint cannot be overemphasized. It allows younger patients to continue a vigorous life-style and elderly patients the opportunity to walk as opposed to wheelchair confinement. In view of the high risk of later contralateral amputation, every effort should be made to preserve at least a transtibial level at the first amputation.
With aging of the general population, trauma has been replaced by peripheral vascular disease as a leading cause of lower-limb amputation. Smoking appears to be related to this increase. In a review of 51 male lower-limb amputees in the United Kingdom, Stewart found a significantly higher incidence of smokers as compared with the general population (82.4% vs. 55%). Another series reported that 58% of 110 transtibial amputees were smokers. The precipitating cause of amputation may be gangrene, infection, or intractable claudication. In diabetes mellitus, the vast majority of amputations are related to various types of foot injury secondary to peripheral sensory neuropathy, with often minor foot damage providing a portal for infection. Infection in diabetics may be difficult to combat at the tissue level due to decreased leukocyte activity in the hyperglycemic state. Patients often continue to walk on infected feet due to a loss of deep pain sense, thereby rapidly spreading the infection along tissue planes. Neuropathic arthropathy, which can be initiated by presumably minor trauma, may also lead to amputation if the foot and ankle skeletal structure becomes severely damaged (Fig 18A-1.,A and B). Although diabetics often develop atheromatous disease at an earlier age than the general population does, it may be difficult to distinguish the relative importance, in causation of gangrenous changes, of atheromatous changes seen in larger vessels and more peripheral small-vessel disease.
Although the population of diabetics appears to be growing, it is conjectural whether this is due to earlier detection, increased longevity related to better treatment, or other factors. It is certain, however, that an increasing percentage of lower-limb amputations is being done in diabetics. For example, a 1956 study showed diabetes as a factor in only 16% of cases. In contrast, combined data from 17 studies published between 1961 and 1988 showed that an average of 52% of patients (range, 30% to 75%) had diabetes mellitus as the primary or secondary causal factor in amputation.
In Hansen's disease, infection of peripheral nerves with Mycobacterium leprae will cause foot insensitivity. Progressive loss of bone and soft tissues, aggravated by intractable deep infection following skin ulceration, may require transtibial amputation (Fig 18A-2.). Severe tissue destruction from fungal infection may occur in the presence of normal sensation, as in mycetoma or "Madura foot" (Fig 18A-3.).
In general, transtibial amputation is indicated when the process requiring ablation cannot be effectively eliminated by lesser procedures. In cases of severe foot infection, usually related to diabetes mellitus, a long transtibial level can almost always be saved even if the proximal spread of infection precludes a partial-foot amputation or Syme ankle disarticulation (see Chapter 16A and Chapter 17A). In peripheral vascular disease with distal gangrene, this level is suitable if there is sufficient vascularity present at the level selected. In cases of trauma to the foot and leg, transtibial amputation should be done initially if there is such severe destruction of soft tissue and bone that reconstruction or a more distal amputation is not feasible (see Chapter 2C). In addition, if there has been warm ischemia of the leg and foot for more than 6 hours following severe vascular injury to the lower limb, a primary amputation should be consid-ered.
When reconstruction after trauma has resulted in an unsatisfactory limb due to deformity, pain, nonunion, or persistent infection, transtibial amputation will provide a good solution (Fig 18A-4.). This should be done as soon as it becomes apparent that further attempts at salvage have little likelihood of success. Adherence to this approach will preserve the patients fiscal, physical, and psychological assets, thereby preventing chronic invalidism. Finally, transtibial amputation should be favored over the transfemoral level whenever there is a reasonable possibility of ambulation.
Inadequate vascularity at amputation sites between the knee and ankle, for any reason, is an absolute contraindication to transtibial amputation. Dependent rubor or gangrenous changes about the upper portion of the tibia, whether gradual or sudden in onset, should lead to consideration of a primary transfemoral amputation. Severe rest pain in the proximal portion of the calf may indicate the need for a primary transfemoral amputation as well. A knee flexion contracture severe enough to prevent use of a transtibial prosthesis may be best served by a knee disarticulation, provided that the skin at that level is viable and will heal primarily. A relative contraindication to transtibial amputation is prolonged nonambulatory status. If the patient is bed bound, a knee flexion contracture will very likely develop. A knee disarticulation can be a good choice in this situation because it provides much better sitting balance than a transfemoral residual limb does. Persson, however, maintains that the tibial portion of the limb will still be useful in transfer and wheelchair sitting activities and is reluctant to remove it on the basis of nonambulation alone.
There are several conditions mistakenly thought of as relative contraindications to transtibial amputation. A diabetic or a patient with Hansen's disease (leprosy) need not be denied a transtibial level on the basis of insensate skin. With good prosthetic fitting and regular observation of the skin for areas of pressure, the amputee should do extremely well. Hemiparetic patients can often manage household ambulation with a transtibial prosthesis. Even poor knee control can be managed easily with a hybrid "prosthosis" that combines a transtibial prosthesis with orthotic knee control componentry, provided that flexion or extension patterning is not extreme and that reasonable balance is present. If they are able to comprehend and follow instructions, they can do quite well (Fig 18A-5.). Even if they are not prosthetic candidates, sitting and kneeling activities will be enhanced by leaving as much of the leg as possible.
Children with congenital foot deformities requiring revision for use of a prosthesis are not well served by transtibial amputation. This will interfere with the growth of the residual limb and make its relative length less in adulthood. In these cases, disarticulation at the ankle joint will preserve end weight-bearing capability and allow a moderate increase in length over time (see Chapter 35).
There are several very important aspects to the preoperative management of prospective amputees. These are largely related to the reason for amputation. Patients undergoing amputation for trauma, although usually young and healthy, often have concomitant injuries to other skeletal parts, soft tissues, or viscera. A careful evaluation must be done to rule out injury to areas other than the affected limb.
When dysvascularity related to peripheral vascular disease with or without diabetes mellitus leads to amputation, the presence of associated disease must be assumed. One study found that 76% of 70 patients coming to transtibial amputation had various other degenerative diseases. Special attention must be directed to control of congestive heart failure, arrhythmias, electrolyte imbalance, dehydration, hypertension, bronchitis, and diabetes for optimum results, with the emphasis on rapid preoperative treatment.
In cases of diabetes mellitus, the infection that has led to the need for amputation often totally disrupts diabetic control. Since the control of infection and of hyperglycemia are interdependent, they must be approached simultaneously for optimum effect. Following initial aerobic and anaerobic wound cultures, broad-spectrum antibiotic therapy should be started, pending bacterial sensitivity studies. Care should be taken to avoid nephrotoxic drugs where possible. If they are needed, renal function should be closely monitored.
Icing of a necrotic/infected limb to control local and systemic effects of the infectious process remains controversial. While its use has been suggested in selected cases, Pedersen et al. condemned this practice and stated that following icing, a transfemoral amputation is unavoidable. Instead, they advocated prompt drainage of abscesses, followed by appropriate antibiotics and bed rest.
A wide range of bacteria may be associated with foot infections in diabetics. They include gram-positive, gram-negative, aerobic, and anerobic organisms, occasionally singly but more often in various combinations. Hoar and Torres found Staphylococcus aureus, Streptococcus hemolyticus, and Proteus vulgaris to be most common. Fearon et al. cultured more than 15 different bacteria in a series of diabetic gangrene cases. Systemic infection secondary to wet gangrene or infections independent of the foot must also be controlled preoperatively. Specifically, evidence of genitourinary and pulmonary infections should be sought. Assessments of wound healing potential are also indicated. These include the serum albumin level as an indicator of nutritional status (normal, 3.5 g/dL or more) and the total lymphocyte count as a measure of immu-nocompetence (at least 1,500/mm). If these values are abnormal, one may expect difficulties with primary wound healing.
Reversal of the catabolic state associated with infection should be initiated preoperatively, preferably by oral intake. The addition of nutritional supplements such as ascorbic acid, zinc, and ferrous sulfate should present no additional clinical problem, but significant caloric enhancement will require matching increases in hypoglycemic agents.
If time and the patients' condition allow, they should be introduced to the team members who will be caring for them postoperatively. The physical therapist can initiate a preoperative program to condition the entire body, prevent contracture of the hip and knee on the side of the amputated limb, and teach safe ambulation with a walker or crutches (see Chapter 23).
Because the patient looks to the amputation surgeon for guidance, a unique opportunity exists to influence the surgical outcome insofar as patient compliance is concerned. A reasonably detailed account of the expected course through prosthetic fitting should be given. This is also an opportunity for the surgeon to promote wound healing by strongly discouraging smoking preoperatively and postoperatively. A Danish study showed a 50% increase in both wound infection and reamputation rates in lower-limb amputees who smoked cigarettes.
A psychologist experienced in dealing with amputees can encourage them to express their anxieties regarding both the surgical and prosthetic phases of care. A preoperative visit by a trained amputee peer counselor matched with the patient by age, sex, and level of amputation can be very beneficial.
There are several aspects to correct selection in the individual case. In trauma, the exact length of recon-structible tissue distal to the knee is usually predetermined by the accident and treatment to that point (Fig 18A-6.). Tumor surgery requires that adequate margins free of disease be the surgeon's first concern, with preservation of limb length secondary (Fig 18A-7.). In dys-vascular cases, the surgeon should first determine that the limb is not salvageable by reconstructive vascular surgery, either entirely or with limited loss at the toe, ray, or transmetatarsal level. Second, it should be determined whether a transtibial level has a reasonable chance of healing. Third, a level that will heal and be durable and optimally functional should be chosen. In cases of foot infection, the proximal extent of infection along tissue planes may determine whether a ray or transmetatarsal amputation or a Syme ankle disarticulation is feasible. If purulence has extended proximal to the ankle, an open ankle disarticulation with fascioto-mies and compartmental debridement is indicated to preserve length.
Although level selection is multifactorial, many studies have tried to oversimplify the problem by basing success or failure solely on one criterion. Although both clinical evaluation and objective laboratory measurements of vascularity are reasonably predictive of success or failure at both the high and low ends of measurement spectra, there remains an intermediate gray zone of unpredictability. The inability to reach consensus on the best test or tests for level selection clearly shows that the best test, which does not yet exist, would be one that predicts failure with 100% accuracy and thus guides the surgeon away from that level.This would avoid imposing higher levels of amputation on patients who could heal at the transtibial level but were eliminated by overly strict application of criteria that include a built-in failure rate for reasons that are not determined by the study method. If failure then occurred, operative factors other than tissue blood flow should be sought, such as poor nutritional status, tissue glycosylation secondary to chronic hyperglycemia, infection, suboptimal surgical technique, or poor postoperative wound management.
The more traditional methods of level selection are considered in this chapter. For a detailed discussion of laboratory tests designed to give more objective measurements of limb and tissue blood flow, the reader is referred to Chapter 2C. In practice, level selection by either approach remains somewhat idiosyncratic and is based on the attitudes and prejudices of the surgeon as well as those of the prosthetist regarding the level under consideration. This is attested to by the varying ratios of transtibial to transfemoral amputations performed in similar institutions in different parts of the world and even in various parts of the same country or city.
Even with the development of more sophisticated tests, most surgeons continue to rely on factors that can be easily evaluated by touch and sight, including peripheral pulses, skin warmth and texture, color of the foot dependent and elevated, hair growth, and the presence of indolent ulcers, tissue necrosis, gross infection, or lymphangitis.
Regarding the evaluation of peripheral pulses, if they can be easily felt, they are usually there. If they cannot be felt, however, they may still be present but obscured by edema, hypotension, or obesity. A significant number of transtibial amputations will heal despite the absence of palpable pulsation at any given level, including the superficial femoral level. In a series of 113 transtibial amputations, 57% healed with only an aortic pulse present; the addition of a femoral pulse increased the success rate to 81%. With palpable popliteal or pedal pulses, all transtibial amputations in this series healed. The data from six papers were combined and analyzed regarding the relationship of healing rate to the presence of a palpable popliteal pulse. Although 65% of these patients had no popliteal pulse felt, 82.5% healed at the transtibial level. These findings point out the difficulty in evaluation of collateral circulation by palpation. The profunda femoris artery, which may be the only major vessel providing collaterals to the calf, is inaccessible to the palpating finger.
Arteriography has been found to bear little correlation to the healing potential of transtibial amputations, on a par with reliance on palpable pulses. Arteriography is now used chiefly to determine the feasibility of vascular reconstruction.
Many surgeons have relied on the trial skin incision. The presumption is that if the skin bleeds within 3 minutes after incision at the proposed level, it should heal at that level; if the skin does not bleed, the surgeon should immediately move proximally. Ken-drick, however, noted no correlation between bleeding of a trial skin incision and healing potential. The basic question of how distally the initial trial skin incision should be made remains unaddressed. A distal trial incision that bleeds, however, should encourage the surgeon to proceed at that level.
Once the decision has been made to amputate at the transtibial level, an equally important choice must be made as to the exact length to be retained (Fig 18A-8.). The shortest useful amputation must include the tibial tubercle to preserve knee extension by the quadriceps.Flexion at this level is provided by the semimembranosus and biceps femoris. Beyond universal agreement as to this shortest possible functional level, the ideal length for optimal prosthetic function has not been determined. The amputation method advocated by Burgess, which results in a cylindrical stump, effectively limits length to approximately 15 cm since the leg begins to taper beyond that point. Marsden recommends limiting the length to 15 cm on the basis that the prosthetist will have less trouble fitting a prosthesis.
There are a number of opinions, however, expressed over several decades that cast doubt on this certitude. Harris, although recommending a short transtibial amputation in his paper of 1944, noted that a long transtibial amputation is stronger than a shorter one. Despite this recognized functional advantage, he recommended a short residual limb due to the skin complications seen in longer amputations from wearing the prostheses with plug-fit sockets and thigh corsets that were available at that time. Moore stated that the greatest length compatible with healing should be retained, while Epps stated that the basic rule was to save all length possible, correlating it to function and the prosthetic components to be used. McCollough et al., while not specifying what they considered optimal length, flatly stated that the longer the residual limb, the better the gait. This position is supported by work showing that transtibial amputees with longer limbs require less energy to ambulate.
In summary, there is no longer an ideal length or site of amputation. In dysvascular cases with an absent popliteal pulse, amputation in the proximal half of the leg would seem reasonable, with a bony level as distal as the junction of the proximal and middle thirds. In cases with good blood flow to the ankle, bone length at the junction of the middle and distal thirds will provide a very functional residual limb. Modern prosthetic components can be easily matched to these more distal levels.
The various types of anesthesia useful in transtibial amputation may be classified as follows:
- Local anesthesia.-An extremely ill or even moribund patient can have a transtibial amputation done without pain under local anesthesia. The agent is injected along the proposed incision line and deeper tissues infiltrated as necessary. Nerves, especially the posterior tibial, are individually injected prior to any manipulation and section. Agents containing epinephrine are avoided.
- Regional anesthesia.-For patients with severe cardiopulmonary compromise, a sciatic-femoral block can be very effective. It can be supplemented, if necessary, with a local anesthetic.
- Low spinal anesthesia.-This technique has little effect on the pulmonary system. Control of blood pressure, however, can be problematic. If hypotension occurs, it is corrected with fluid administration and/or vasopressors.
- General anesthesia.-In the healthy patient undergoing amputation for trauma, it can be quite safe and effective. If the patient has severely compromised cardiopulmonary function, however, it may not be the best choice.
In summary, the choice of anesthesia depends on the patient's condition, the skills and experience of the individual anesthetist, and the patient's choice, if he is fit to receive any type of anesthetic.
Amputation is no longer to be considered as purely the ablation of a useless or debilitating part, but rather as a reconstructive procedure to restore ambulatory function. As more functional goals for the transtibial amputee have been appreciated, new techniques have been developed in an attempt to enhance function at that level. To achieve optimal function, the surgeon must be willing at times to do staged procedures. The ultimate goal is a residual limb that will interface well with a prosthesis. To achieve this end, transtibial amputation should be performed or directly supervised by an experienced surgeon and not delegated to the least experienced surgeon-in-training to do unsupervised. Use of a thigh tourniquet is recommended in cases of amputation for trauma. In dysvascular cases, a tourniquet may be in place, but only inflated if bleeding is problematic during surgery.
There are two criteria for the primary healing of transtibial residual limbs. First, as discussed above, is proper selection of level. A second and equally important criterion is the proper technical management of tissues during the procedure. The placement and measurement of flaps must be accurately related to the cross-sectional area of the leg at the bony level selected. Otherwise, either the bone will need to be shortened to avoid closure under tension, or redundant soft tissue will have to be excised. Successful use of a variety of flap configurations has shown that incision placement is not crucial so long as the incisional scar is not adherent to the underlying bone.
Transtibial amputations may be classified as follows:
- Closed amputations
- End weight-bearing amputations
- Open amputations
Long Posterior Myofasciocutaneous Flap
In 1943, Bickel reported on the use of a long posterior myofasciocutaneous flap in 110 transtibial amputations. The major impetus for the acceptance of this concept, however, may be attributed to the educational efforts of Burgess. The senior author's technique, adapted from Wagner, follows: reference points are marked medially and laterally on the leg at the junction of the anterior two thirds and posterior third of the leg diameter at the level selected for bone section. The two points are joined to form an anterior flap that is convex distally and no more than 0.5 cm in length. The posterior flap is then drawn with its length equal to the distance from the original reference point to the anterior third of the tibia. This line passes slightly anterior and then gradually posterior to pass around the posterior aspect of the leg and is completed in a similar fashion on the opposite side (Fig 18A-9.). The anterior and posterior flaps meet in a small half circle that will eliminate the "dog ear."
The subcutaneous tissue and investing fascia are cut in line with the skin incision. At this point, the greater saphenous vein is ligated and the superficial peroneal and saphenous nerves transected under slight tension to cause them to retract into the soft tissues. The anterior compartment muscles are carefully divided to expose the neurovascular structures. The artery should be doubly ligated, the veins ligated, and the nerve cut under slight tension. The tibia is stripped of periosteum only to the level of transection to reduce the chance of bone spur formation. It is then cut transversely with a saline-cooled power saw. The fibula is cleared with an elevator and cut obliquely to form a facet facing posterolateraly. It should be made equal to or slightly shorter then the tibia to prevent undue distal tibial prominence as seen in a conical rather than cylindrical residual limb (Fig 18A-10.). A long amputation knife is passed behind the tibia and fibula and drawn distally to create a tapered myofasciocutaneous flap (Fig 18A-11.). The peroneal and posterior tibial arteries are clamped, divided, and doubly ligated, and the veins are singly ligated. The posterior tibial nerve may be ligated to secure its intrinsic vessels, or it may be dissected out and lightly cauterized. The nerve is then cut and allowed to retract proximal to the bone end. The lesser saphenous vein is found in the posterior flap and ligated while the adjacent sural nerve is cut under tension.
The deep calf musculature is excised to reduce the bulk of the posterior flap (Fig 18A-12.). A bulky soleus may also be tapered further to contour the distal tibia padding. If necessary, the flap edges can be trimmed further to obtain a good fit. The cooled power saw is now used to cut a bevel in the anterior end of the tibia. Both bone cuts are now carefully smoothed and contoured with a bone file (Fig 18A-13.).
The wound may be closed by using either a myodesis or myoplasty technique. Myodesis provides firm fixation of the posterior muscle padding to the tibia, thus preventing later retraction. It is contraindicated, however, in cases of severe dysvascularity in which the blood supply to the muscle appears compromised. In these cases, myoplasty will be sufficient. To effect myodesis, drill holes are placed each side of the tibial crest bevel. Other holes may be placed medially and laterally as well. All bone detritus must be carefully washed from the wound after drilling. Following placement of a suction drain, the myodesis sutures are inserted. The tissues joined to the bone by these sutures include the anterior investing fascia, the gastrocnemius (or tapered Achilles tendon in the case of a long transtibial amputation), and the posterior investing fascia (Fig 18A-14.). A heavy absorbable suture works well for this. The medial and lateral portions of investing fascia and muscle flap are sutured with further interrupted absorbable suture. No subcutaneous sutures are necessary, and the skin is closed with interrupted nylon sutures widely spaced. The intervals are reinforced with adhesive paper strips (Fig 18A-15.).
A posterior myofasciocutaneous flap can be formed down to the distal extent of the soleus muscle, with the technique becoming progressively easier in more distal amputations (Fig 18A-16.). There are several anatomic reasons for this. With distal tapering of the calf, the cross-sectional area of the leg decreases, and this results in a much shorter, widely based posterior flap. Distal muscle bulk is much less, thus resulting in minimal muscle excision to allow closure. Less tissue mass also results in less tendency for "dog ear" formation. Finally, control of venous bleeding is simplified because of fewer venous plexuses distally.
A well-padded plaster or fiberglass cast is applied with the knee in full extension. The drain tube is run between the layers of cast padding out the top of the cast so that it can be removed after 24 to 48 hours without disturbing the cast (Fig 18A-17.). The cast is made as light as possible to allow the patient greater mobility in bed and on crutches.
Equal Anterior and Posterior Flaps
In this technique, the length of each flap is equal to half the diameter of the leg at the level of bone transection (Fig 18A-18.). Starting from a midlateral apex on either side, the skin is cut to form equal anterior and posterior flaps. The anterior investing fascia and the muscle of the anterolateral compartment are then cut down to the anterior tibial neurovascular bundle. This and the superficial peroneal nerve and the neurovascular bundle are managed as described in the previous section. The tibia and fibula are cut as noted before. The long amputation knife is used to create a posterior myofasciocutaneous flap. The vessels and nerves are dealt with as described. The tibia is beveled and contoured with a bone file. The wound is irrigated, and myoplasty is carried out by suturing the investing fascia and myofascia of the posterior and anterior flaps together over the end of the bone. The absolute indication for equal anterior and posterior flaps is to conserve bone length when relatively little bone remains (Fig 18A-19.,A-C).
Equal Medial and Lateral (Sagittal) Flaps
The advantages of this approach were outlined by Persson. The flaps are less apt to become necrotic in dysvascular cases for two reasons. The first is that the placement of flaps medially and laterally automatically reduces the amount of poorly vascularized anterior skin that is left. The second reason is that the resultant flaps are widely based and very short, thus enhancing their viability (Fig 18A-20.,A-C). Persson also stated that a side-to-side myoplasty covers the bone better and provides good spontaneous drainage. In trauma cases, another advantage is that the sagittal flap configuration allows the skin to be more easily cut proximal to any anterior or posterior damaged skin, thus helping to preserve bony length (Fig 18A-21.,A and B and Fig 18A-22.,A and B).
This approach is designed for the severely dysvascu-lar patient in whom all major vessels are occluded. From thermographic and intradermal radioisotope studies it has been shown that collateral circulation through small arteries accompanying the saphenous and sural nerves will provide blood supply to flaps that incorporate these nerves and their vessels. To take advantage of this fact, the creation of anteromedial and posterolateral flaps is suggested. If the posterolateral flap is seen to have very poor blood supply at the time of skin incision, it can be shortened. This approach combines features of the long posterior flap and sagittal flaps (Fig 18A-23.,A and B).
End Weight-Bearing Transtibial Amputations
Osteomyoplasty (Ertl Procedure)
This procedure was designed for revision of transtibial residual limbs in the war wounded. The resultant amputation has end weight-bearing capability. Two osteoperiosteal flaps are elevated from the anteromedial and lateral aspects of the tibia beginning approximately 10 cm distal to the proposed level of bone transection. The proximal attachment of these osteoperiosteal flaps is preserved as the remainder of the amputation is performed. Once the amputation has been completed, the lateral flap is sewn to the medial aspect of the fibula, and the anteromedial flap is sewn to its lateral aspect. They are then sewn to each other to create an osteoperiosteal tube joining the ends of the bones, which will ossify to form a sturdy weight-bearing bone bridge (Fig 18A-24.). This method has been employed in the American military, but to a much lesser extent in the civilian population. It has been recommended as a useful technique in young traumatic amputees, both initially and in revision surgery. Its chief disadvantage is the sacrifice of 10 cm of bone length, particularly in cases where there is sufficient mobile soft tissue present to cover a greater length of tibia.
This is another approach to end weight bearing in transtibial amputation for trauma. The indications are limited, criteria are strict, and the surgery is precise. Tibial diaphyseal bone loss must be extensive enough to preclude skeletal reconstruction, but the posterior tibial nerve and foot should be intact. The heel pad and sole tissues are dissected from the skeleton of the foot, with the posterior tibial nerve left in continuity. The nerve is folded into the soft tissues of the residual limb, and a posterior tibial-popliteal arterial anastomosis is done. The heel pad is then sutured over the end of the residual limb to provide end weight bearing after healing.
Primary open amputation is indicated whenever primary closure of the wound is likely to result in initial or continuing infection and/or necrosis. This applies equally in traumatic amputations and in cases of infection in which an attempt will be made to preserve maximum limb length below the knee to enhance prosthetic function. The "guillotine" amputation in which all soft tissue and bone is transected at the same level should be reserved for emergency situations and then done only at distal levels to leave enough proximal tissue for a functional transtibial amputation at the time of revision. The open circumferential technique, whereby each successive layer is cut and allowed to retract before cutting deeper layers, has the advantage of less exposure of the deeper soft tissues and bone and perhaps conserves some bone length. It will, however, have to be revised to allow a good soft-tissue envelope reconstruction over the end of the bones. A much better technique utilizes open flaps. In this case, all viable tissue is preserved by forming rough myofasciocutaneous flaps whose length and orientation are dictated by the trauma or infection. While such flaps may appear excessively long initially, considerable shrinkage will occur before closure is feasible. If the flaps are so long that some distal viability is lost, this portion will be removed at the time of closure. Of these three, this last technique preserves the maximum amount of bone length (Fig 18A-25.,A and B).
In cases of irreparable loss of foot vascularity and sensation associated with segmental tibial fracture, there may be segments of tibial shaft that are still well attached to soft tissue which can be closed secondarily to form a good soft-tissue envelope. These segments can be fixed to the proximal part of the shaft by internal or external fixation to provide a longer residual limb (Fig 18A-26.,A and B and Fig 18A-27.,A-C).
In trauma cases, there may be instances in which there is enough muscle to adequately cover the bones but insufficient skin to completely close the wound. It is not necessary in these cases to shorten the bones to the level where full coverage by skin is possible. Available skin can be rotated to cover the anterodistal part of the tibia, the site where the greatest stress occurs during prosthetic walking. The remainder of the muscle is covered with a split-thickness skin graft (Fig 18A-28.,A and B and Fig 18A-29.,A-C).
In cases of severe foot infection, an open ankle disarticulation is useful. If tissue planes proximal to the ankle are involved, they may be easily opened medially and/or laterally to thoroughly debride the infected compartments (Fig 18A-30.,A-D).
SPECIAL CONSIDERATIONS DURING TRANSTIBIAL AMPUTATION
Treatment of Skin
In order to have a successful amputation, the one tissue structure that must heal is the skin. The skin-subcutaneous incision should be made at 90 degrees to the surface to avoid having portions of skin unsupported by subcutaneous tissue and, hence, more difficult to accurately oppose and more prone to necrosis. At no time should the skin be traumatized by grasping with forceps. Instead, skin edges can be everted for closure by the suturing needle, skin hooks, or gloved fingers. The skin must be precisely approximated without closure tension. Simple interrupted sutures are widely spaced and alternated with adhesive paper strips to contain subcutaneous fat. In dysvascular cases, the sutures are kept in place for 3 weeks to allow for the slower healing that is common in these cases.There should be no separation of layers in the creation of myofasciocutaneous flaps because this may interfere with the blood supply to the skin. The residual limb should not be left with inverted scars, redundant skin, or "dog ears" that will not promptly atrophy.
Treatment of Fascia
The crural or investing fascia should be cut at the same level as the skin and subcutaneous tissue. It should never be separated from the surrounding soft tissues in order to prevent damage to any small perforating vessels serving the skin. In closing a myofasciocutaneous flap, care should be taken to ensure that the crural fascia is indeed found and firmly sutured both to ensure maximal wound strength and to take tension off the overlying skin. This, in turn, allows the use of fewer skin sutures, which may contribute to less skin necrosis. Complete closure of the fascia also prevents scarring of skin directly to bone, which would prevent dissipation of shear forces generated at the socket-skin interface.
Treatment of Muscle
Muscle is considered to carry at least some blood from the deep arteries of the leg to the skin. It is therefore generally accepted that muscle should not be dissected from its overlying investing fascia. Muscle may be trimmed to provide sufficient padding for the end of the tibia without unnecessary bulk. Any ischemic or necrotic muscle should be excised. This condition appears most commonly in the anterior compartment. If muscle tissue is merely pale, it may be left because it will probably fibrose in time. Healing can occur following the complete removal of necrotic muscles so long as the skin remains viable.
Treatment of Nerves
The nerves to be specifically found and transected during transtibial amputation include the superficial peroneal, saphenous, deep peroneal, sural, and posterior tibial. The posterior tibial nerve may present sufficient intrinsic vascular supply to warrant ligation or cauterization of its vasa nervorum. A variety of methods have been advocated to inhibit neuroma formation by traumatizing the proximal cut end of the nerve. The best approach appears to be simple sharp division following mild traction on the nerve. The cut proximal end retracts into the soft tissues where the inevitable neuroma will be protected during prosthetic gait. Dellon et al. demonstrated that nerve ends surgically buried in muscle show no tendency toward neuroma formation. Malawer et al. have advocated the control of postoperative pain in the residual limb by intraoperative placement of a small Silastic catheter within or next to the posterior tibial nerve sheath for the continuous administration of local anesthetic (bupivacaine, 0.25% to 0.5%) by a standard infusion pump at the rate of 2 to 3 mL/hr over the first 72 hours. The catheter is easily removed where it exits the postoperative cast.
Treatment of Bone
Beveling of the tibia combined with careful smoothing of the bone edges will prevent damage to the skin in its position between the hard bone surface and the firm prosthetic socket. Various authors have suggested a bevel of 45 to 60 degrees as being optimal. All bone cutting with a power saw should be done with saline cooling to prevent thermal necrosis. If the surgeon wishes to avoid fluid splattering, a Gigli saw may be used instead to cut the tibial shaft from posterior to anterior. As the saw enters the anterior cortex, it is directed proximally to cut the bevel.
The fibula should be no more than 0.5 to 1 cm shorter than the tibia if a conical shape of the distal residual limb with a prominent distal end of the tibia is to be avoided. To prevent complaints of soft-tissue impingement during prosthesis use, the fibula may be cut with a bevel facing posterolaterally. Both bones should be carefully filed to remove all sharp edges and points. Prior to closure, the wound should be generously irrigated to wash away bone detritus.
Total removal of the fibula may be required in cases of fibular osteomyelitis or bony necrosis due to circumferential muscle loss or abscess formation. Removal may also be beneficial in a very short transtibial amputation at the level of the tibial tubercle where, if left in place, the fibular head may produce pain by its ball-like presence in the socket (Fig 18A-31.). Bleeding from the tibia or fibula can be controlled by electrocautery and closure of the wound. Bone wax should not be used because of its tendency to provoke a foreign-body reaction and its interference with firm healing of the muscle flap to the bone.
IMMEDIATE POSTOPERATIVE MANAGEMENT
A rigid dressing from the end of the residual limb to midthigh with the knee in full extension meets a number of worthwhile goals. Knee flexion contracture is prevented during the first few painful postoperative days, the wound is protected from bed trauma, and edema formation is limited. The patient is generally comfortable and, if the rigid dressing is light enough, can move about in bed quite easily. Another advantage of the rigid dressing is that it protects against falls onto the residual limb while the patient is learning to manage a walker or crutches. If necessary, it can be secured with a waist belt. The only disadvantage is that the wound cannot be readily inspected. Careful attention to the patient's general status, however, will inform the surgeon of any indication for removal of the cast such as an otherwise unexplained fever or evidence of excessive drainage. The cast is worn for 3 weeks with weekly changes for wound inspection and full range of motion of the knee.
A soft dressing, on the other hand, allows easy access to the wound for inspection and for motion of the knee with or without the guidance of a therapist. It does not, however, offer any protection to the wound from trauma, nor does it prevent knee flexion contracture if the patient does not move the knee on a regular basis. One randomized study comparing soft and rigid dressings showed that rigid dressings resulted in less pain, improved sense of well-being, and enhanced prosthetic fitting progress. In another series, hospital stay was reduced from 14 to 7 days. A posterior plaster splint will keep the knee straight so long as the splint is not broken and the wrapping is firm. If it is necessary to look at the wound, however, a better plan is to make a strong posterior hemicylinder by removing the anterior half of a full cast.
The issue that most concerns patients in the immediate postoperative period is pain control. They should be given an amount of narcotics sufficient for good pain relief every 3 to 6 hours or by means of an on-demand machine for a maximum of 5 days. After this, therapy can be switched to oral narcotics or nonsteroidal anti-inflammatory drugs. In this way, habituation should not occur. Alternatively, a small Silastic catheter may be inserted at the time of surgery within or next to the posterior tibial nerve sheath for the administration of local anesthetic for the first 72 hours postoperatively. Avoidance of wound dependency will also contribute to prevention of pain.
Infection prevention is an important aspect of postoperative management that is met in most cases with perioperative intravenous antibiotics. If infection was an overriding factor in the amputation, however, one or more antibiotics chosen from organism sensitivities should be continued for 2 to 5 days postoperatively.Further need for antibiotics can be determined by direct evaluation of the wound. Atelectasis may be prevented by positioning and by deep-breathing exercises using various types of incentive respiration devices.
The patient should be made mobile as soon as possible to prevent the deconditioning that may occur within just a few days. On the first postoperative day, the patient should be sitting out of bed with the residual limb elevated to the level of the chair seat. By the next day, the patient should be in the physical therapy department beginning ambulation on the parallel bars. This is followed by the use of crutches or a walker as conditioning and balance improve (see Chapter 23).
Early mobilization has been enhanced in recent years with the introduction of the immediate postoperative prosthesis and its more commonly used component, the rigid postoperative cast. If an immediate postoperative prosthesis has been applied, limited weight bearing on the residual limb can start almost immediately provided that the patient demonstrates sufficient strength, balance, proprioception, and cognition to accurately determine the weight applied.
The cost of hospital stay has become a major issue in recent years. In the past, many patients remained in the hospital or rehabilitation center following surgery until they had healed, been fitted with a prosthesis, and thoroughly trained in its use. In the United States, this is no longer financially feasible. Transtibial amputees are often discharged from the hospital 4 to 5 days after surgery unless they have failed to achieve their maximum level of independence in transfers and one-legged ambulation. In that case, they will stay until these goals have been achieved or abandoned as unrealistic. All further care, including prosthetic fitting and follow-up, is accomplished on an outpatient basis. Hospitalization for prosthetic gait training can be justified in cases of marked deconditioning, advanced age, bilateral concomitant lower-limb amputations, or great distance from the center.
The psychological needs of the amputee must also be met. Counseling by various team members can be quite helpful in allaying anxiety regarding the prosthetic phase of care. Visits by a trained amputee peer counselor matched with the patient by age, sex, and amputation level can be of inestimable help. Amputee/ consumer peer support groups can be extremely helpful in smoothing the amputee's transition to the community, especially by providing a comfortable social, educational, and recreational outlet.
Transtibial amputation, by saving the knee joint, provides the amputee with the possibility of near-normal function in regard to ambulation and overall life-style. With the availability of new information on the efficacy of transtibial amputation and improved methods of determining potential healing levels in a limb, the majority of major lower-limb amputations are now being done at the transtibial rather than the transfemoral level. Diabetes is now seen to be a primary or secondary cause of amputation in at least 50% of cases. Most patients with dysvascular limbs have one or more significant associated diseases calling for detailed preoperative management and skilled care in the immediate postoperative period.
The aim of amputation surgery is a well-healed, sen-sate, functional end organ that will interface well with a prosthesis. Selection of length is based on etiologic factors and on clinical and laboratory evaluation. As much length as possible should be preserved, compatible with disease eradication and good prosthetic function. Meticulous management of tissues will lead to preservation of the length obtained at surgery. Myodesis is advocated in cases in which local dysvascularity is not a problem. Postoperative rigid dressings are strongly recommended because of local protection of the wound and the prevention of edema and knee flexion contractures. Early mobilization prevents deconditioning, thereby allowing early discharge to an outpatient status. Early prosthetic weight bearing has great value in selected cases if closely monitored. Optimal amputee management is best achieved through a team approach beginning even before surgery.
- Alter AH, Moshein J, Elconin KB, et al: Below-knee amputation using the sagittal technique: A comparison with the coronal amputation. Clin Orthop 1978; 131:195-201.
- Bagdade JD, Nielsen K, Root R, et al: Host defense in diabetes mellitus: The feckless phagocyte during poor control and ketoacidosis. Diabetes 1970; 19:364.
- Baker WH, Barnes RW, Shurr DG: The healing of below-knee amputations: A comparison of soft and plaster dressings. Am J Surg 1977; 133:716-718.
- Barber GG, McPhail NV, Scobie TK, et al: A prospective study of lower limb amputations. Can J Surg 1983; 26:339-341.
- Bard G, Ralston HJ: Measurement of energy expenditure during ambulation, with special reference to evaluation of assistive devices. Arch Phys Med Rehabil 1959; 40:415-420.
- Bickel WH: Amputations below the knee in occlusive arterial diseases. Surg Clin North Am 1943; 23:982-994.
- Block MA, Whitehouse FW: Below-knee amputation in patients with diabetes mellitus. Arch Surg 1963; 87: 682-689.
- Boontje AH: Major amputations of the lower extremity for vascular disease. Prosthet Orthot Int 1980; 4:87-89.
- Bowker JH: Surgical techniques for conserving tissue and function in lower-limb amputation for trauma, infection, and vascular disease. AAOS Instr Course Lect 1990; 39:355-360.
- Brodie IAO: Lower limb amputation. Br J Hosp Med 1970; 4:596-604.
- Burgess EM: The below-knee amputation. Bull Prosthet Res 1968; 10:19-25.
- Castronuovo JJ, Deane LJ, Deterling RA, et al: Below-knee amputation. Is the effort to preserve the knee joint justified? Arch Surg 1980; 115:1184-1187.
- Cheng EY: Lower extremity amputation level: Selection using noninvasive hemodynamic methods of evaluation. Arch Phys Med Rehabil 1982; 63:475-479.
- Chilvers AS, Briggs J, Browse NL, et al: Belowand through-knee amputations in ischaemic disease. Br J Surg 1971; 58:824-826.
- Cranley JJ, Krause RJ, Strasser RS, et al: Below-the-knee amputation for arteriosclerosis obliterans. Arch Surg 1969; 98:77-80.
- Cumming JGR, Jain AS, Walker WF, et al: Fate of the vascular patient after below-knee amputation. Lancet 1987; 2:613-615.
- de Cossart L, Randall P, Turner P, et al: The fate of the below-knee amputee. Ann R Coll Surg Engl 1983; 65:230-232.
- Deffer PA: More on the Ertl osteoplasty. Amputee Clin 1970; 2:7-8.
- Dellon AL, MacKinnon SE, Pestronk A: Implantation of sensory nerve into muscle: Preliminary clinical and experimental observations on neuroma formation. Ann Plast Surg 1984; 12:30-40.
- Dickhaut SC, DeLee JC, Page CP: Nutritional status: Importance in predicting wound-healing after amputation. J Bone Joint Surg [Am] 1984; 66:71-75.
- Dwars BJ, Rauwerda JA, van den Brock TAA, et al: A modified scintigrafic technique for amputation level selection in diabetics. Eur J Nucl Med 1989; 15:38-41.
- Epps CH Jr: Amputation of the lower limb, in Evarts CM (ed): Surgery of the Musculoskeletal System, ed 2. New York, Churchill Livingstone Inc, 1990.
- Ecker ML, Jacobs BS: Lower extremity amputation in diabetic patients. Diabetes 1970; 19:189-195.
- Eraklis A, Wheeler B: Below-knee amputations in patients with severe arterial insufficiency. N Engl J Med 1963; 269:933-943.
- Ertl J: About amputation stumps. Chirurgie 1949; 20:2-12, 218.
- Fearon J, Campbell DR, Hoar CS, et al: Improved results with diabetic below-knee amputees. Arch Surg 1985; 120:777-780.
- Fleurant FW, Alexander J: Below knee amputation and rehabilitation of amputees. Surg Gynecol Obstet 1980; 151:41-44.
- Gonzalez EG, Corcoran PJ, Reyes RL: Energy expenditure in below-knee amputees: Correlation with stump length. Arch Phys Med Rehabil 1974; 55:111-119.
- Harris JP, Page S, Englund R, et al: Is the outlook for the vascular amputee improved by striving to preserve the knee? J Cardiovasc Surg 1988; 29:741-745.
- Harris PD, Schwartz SI, DeWeese JA: Midcalf amputation for peripheral vascular disease. Arch Surg 1961; 82:381-383.
- Harris RI: Amputations. J Bone Joint Surg 1944; 26:626-634.
- Harris WR: Below-knee amputation: A technical note. Can J Surg 1987; 30:392-393.
- Heller RF, Hayward D, Hobbs MST: Decline in rate of death from ischaemic heart disease in The United Kingdom. Br Med J 1983; 286:260-262.
- Hoar CS, Torres J: Evaluation of below-the-knee amputation in the treatment of diabetic gangrene. N Engl J Med 1962; 266:440-443.
- Kacy SS, Wolma FJ, Flye MW: Factors affecting the resuits of below knee amputation in patients with and without diabetes. Surg Gynecol Obstet 1982; 155:513-518.
- Keagy BA, Schwartz JA, Kotb M, et al: Lower extremity amputation: The control series. J Vasc Surg 1986; 4:321-326.
- Kendrick RR: Below-knee amputation in arteriosclerotic gangrene. Br J Surg 1956; 44:13-17.
- Lange R, Bach A, Hanse S, et al: Open tibial fractures with associated vascular injuries: Prognosis for limb salvage. J Trauma 1985; 25:203-208.
- Lepantalo M, Isoniemi H, Kyllonen L: Can the failure of a below-knee amputation be predicted? Ann Chir Gynaecol 1987; 76:119-123.
- Lim RC, Blaisdell FW, Hall AD, et al: Below-knee amputation for ischemic gangrene. Surg Gynecol Obstet 1967; 125:493-501.
- Lind J, Kramhoff M, Bodtker S: The influence of smoking on complications after primary amputations of the lower extremity. Clin Orthop 1991; 267:211-217.
- Loon HE: Below-knee amputation surgery. Artif Limbs 1961; 6:86-99.
- Louie TJ, Bartlett JG, Tally FP, et al: Aerobic and anaerobic bacteria in diabetic foot ulcers. Ann Intern Med 1976; 85:461-463.
- McCollough NC, Jennings JJ, Sarmiento A: Bilateral be-low-the-knee amputation in patients over fifty years of age: Results in 31 patients. J Bone Joint Surg [Am] 1972; 54:1217-1223.
- McCollough NC III, Harris AR, Hampton FL: Below-knee amputation, in Atlas of Limb Prosthetics. St Louis, Mosby-Year Book, 1981, pp 341-368.
- McCollum PT, Spence VA, Walker WF, et al: A rationale for skew flaps in below-knee amputation surgery. Pros-thet Orthot Int 1985; 9:95-99.
- Mclntyre KE Jr, Bailey SA, Malone JM, et al: Guillotine amputation in the treatment of nonsalvagable lower-extremity infections. Arch Surg 1984; 119:450-453.
- Malawer MM, Buch R, Khurana JS, et al: Postoperative infusional continuous regional analgesia (PICRA): A technique for relief of postoperative pain following major extremity surgery. Clin Orthop 1991; 266:227-237.
- Marsden FW: Amputation: Surgical technique and postoperative management. Aust N Z J Surg 1977; 47:384-392.
- Moore TJ: Amputations of the lower extremity, in Chapman M (ed): Operative Orthopaedics. Philadelphia, JB Lippincott, 1988.
- Moore WS, Hall AD, Lim RC: Below the knee amputation for ischemic gangrene. Comparative results of conventional operation and immediate postoperative fitting technique. Am J Surg 1972; 124:127-134.
- Murdoch G: Amputation surgery in the lower extremity. Prosthet Orthot Int 1977; 1:72-83.
- Murray DG: Below-knee amputations in the aged: Evaluation and prognosis. Geriatrics 1965; 20:1033-1038.
- Paloschi GB, Lynn RB: Major amputations for oblitera-tive peripheral vascular disease with particular reference to the role of below-knee amputation. Can J Surg 1967; 10:168-171.
- Pedersen HE, LaMont RL, Ramsey RH: Below-knee amputation for gangrene. South Med J 1964; 57:820-825.
- Perry T: Below-knee amputations. Arch Surg 1963; 86: 199-202.
- Persson BM: Sagittal incision for below-knee amputation in ischaemic gangrene. J Bone Joint Surg [Br] 1974; 56:110-114.
- Pohjolainen T, Alaranta H: Lower limb amputations in southern Finland. Prosthet Orthot Int 1988; 12:9-18.
- Purry NA, Hannon MA: How successful is below-knee amputation for injury? Injury 1989; 20:32-36.
- Rizzo RL, Matsumoto R: Above vs. below knee amputations: A retrospective analysis. Int Surg 1980; 65:265-267.
- Robinson K: Long-posterior-flap myoplastic below-knee amputation in ischaemic disease: Review of experience in 1967-1971. Lancet 1972; 2:193-195.
- Robinson K: Skew flap myoplastic below-knee amputation: A preliminary report. Br J Surg 1982; 69:554-557.
- Robinson KP: Skew-flap below-knee amputation. Ann R Coll Surg Engl 1991; 73:155-157.
- Roon AJ, Moore WS, Goldstone J: Below-knee amputation: A modern approach. Am J Surg 1977; 134:153-158.
- Rush DS, Huston CC, Bivins BA, et al: Operative and late mortality rates of above knee and below knee amputations. Am Surg 1981; 47:36-39.
- Sarmiento A, Warren WD: A re-evaluation of lower extremity amputations. Surg Gynecol Obstet 1969; 129:799-802.
- Singer DI, Morrison WA, McCann JJ, et al: The fillet foot for endweight-bearing cover of below knee amputations. Aust NZJ Surg 1988; 58:817-823.
- Smith BC: A twenty year follow-up in fifty below-knee amputations for gangrene in diabetics. Surg Gynecol Obstet 1956; 103:625-630.
- Stewart CPU: The influence of smoking on the level of lower limb amputation. Prosthet Orthot Int 1987; 11:113-116.
- Termansen NB: Below-knee amputation for ischaemic gangrene. Acta Orthop Scand 1977; 48:311-316.
- Thornhill HL, Jones GD, Brodzka W, et al: Bilateral below-knee amputations: Experience with 80 patients. Arch Phys Med Rehabil 1986; 67:159-163.
- Wagner FW Jr: Resident Training Manual. Rancho Los Amigos Medical Center, Calif.
- Waters RL, Perry J, Antonelli D, et al: Energy cost of walking amputees: The influence of level of amputation. J Bone Joint Surg [Am] 1976; 58:42-46.
- Yaramenko D, Andruhova RV: Below-knee amputation in patients with vascular disease and prosthetic fitting problems. Prosthet Orthot Int 1986; 10:125-128.
Chapter 18A - Atlas of Limb Prosthetics: Surgical, Prosthetic, and Rehabilitation Principles