Wrist and Forearm Amputations

Amputations of the upper extremity largely follow the same basic principles as those of any amputated limb, some of which are covered in this article. However, the primary purpose of this article is to highlight the special considerations involved in acquired amputations at the wrist and forearm (below the elbow).[1, 2, 3, 4, 5] Amputations at or above the elbow, as well as the various hand and digital amputations, are discussed more fully elsewhere (see Elbow and Above-Elbow Amputations, Hand Amputations and Replantation, and Digital Amputations of the Upper Extremity).

The true frequency of acquired amputation of the wrist and forearm is unknown. Published estimates of amputated limbs, including those of the upper extremities, vary significantly. Prevalences of 350,000-1,000,000 persons with amputations and annual incidences of 20,000-30,000 new amputations have been cited.

Although acquired amputations in children are discussed because this group of patients deserves special consideration,[6] congenital limb amputations and deficiencies are beyond the scope of this article.

Amputation remains one of the oldest surgical procedures. Archeologists have uncovered evidence of prehistoric people with amputations, both congenital and those acquired from surgery or trauma. Whereas the procedure has evolved significantly since the days of quickly severing a limb from an unanesthetized patient and dipping the stump in boiling oil to achieve hemostasis, it was not until World Wars I and II that modern ideas of amputation and prosthetics developed.

Particularly within the last three decades, prosthetic research and rehabilitation engineering centers supported by federal funding have disseminated new information regarding biomechanics and prosthetic design. With the advent of physical and rehabilitative medicine, surgeons now realize that the care of a person with an amputation does not end with removal of sutures.

As medical technology and surgical techniques are improved in the areas of peripheral vascular disease, diabetes, microsurgery, and limb salvage, the number of amputations will consequently decrease. Ethical questions of "technology over reason" will come to the forefront.

An example of this debate can be observed with the advent of upper-extremity transplantation.[7] Although transplantation offers an attractive alternative to amputation, further discussion is needed to evaluate the risks and benefits of such procedures. Does the functional gain of transplantation justify the use of resources involved and commitment of the patient to lifelong immunosuppressive therapy?

Although the surgical technique of amputation has stabilized and is not likely to undergo radical advances in the near future, prosthetic advances may well lead to improvements in function and quality of life for individuals with an amputation. Likewise, much research has focused on gaining a better understanding of the problem of phantom sensations as it relates to the reorganizational changes in the somatosensory system. Unfortunately, many questions remain unanswered.

Irreparable loss of the blood supply of a diseased or injured upper extremity is the only absolute indication for amputation regardless of all other circumstances (see the image below). Severe peripheral vascular disease, traumatic injury, thermal and electrical injury, and frostbite commonly necessitate amputation.[8, 9, 10, 11] Not only has the part been rendered useless, but it is also a threat to the life of the individual because the toxic products of tissue destruction are disseminated systemically. It is important to remember that no injury severity score exists as a guide for severe upper-extremity trauma. Much of the decision-making is left to the judgment of the surgeon.

Likewise, in individuals with systemic sepsis, amputations are necessary to control an otherwise rampant infection. Occasionally, an injury or condition that does not directly affect the vasculature has disabled the upper extremity to the extent that a prosthesis would be functionally superior to sparing the limb. An indication for amputation after nerve injury is the development of uncontrolled trophic ulcers in an anesthetic upper extremity. Amputation is indicated rarely in persons with quadriplegia, even if the upper extremities have no residual function. Often the upper extremities help maintain balance while sitting and serve to distribute the forces of weightbearing over a larger area, thus minimizing pressure sores.

In general, amputations in the upper extremity are also indicated for persons with malignant tumors without evidence of metastases.[12] Even after metastases, amputation may be necessary to relieve pain when a neoplasm has become ulcerated and infected or has caused a pathologic fracture. In these oncologic cases, the indications for amputations versus limb salvage procedures are evolving constantly and require individual consideration beyond the scope of this article.

The only absolute contraindication for amputation is a spared limb or part of a limb that would be functionally superior to what could be achieved after amputation. Many factors regarding the functional potential of a spared limb or part of a limb must be considered, including tactile and protective sensation, range of motion, pain, and the intended purpose of the limb or part.

Anatomy

The distal third of the forearm and wrist lack much of the bulky vascular musculature located more proximally. An envelope consisting largely of capsule, ligaments, tendon, and fascia surrounds the distal radius and ulna. Muscle bellies of the flexor pollicis longus (FPL) and pronator quadratus may be encountered; however, they provide very little in the way of stump padding. Radial and ulnar vessels are identified at this level under the flexor carpi radialis (FCR) and flexor carpi ulnaris (FCU), respectively. The median nerve is located between the volar fascia and flexor tendons, and the ulnar nerve lies under the FCU just medial to the ulnar artery.[13]

Special attention should be given to identifying and properly addressing the more superficial palmar cutaneous branch of the median nerve and the dorsal cutaneous branch of the ulnar nerve to prevent painful neuroma formation. Unlike at the level of the median and ulnar nerves and their associated branches, at this level, the radial nerve has fragmented into several branches; some are small filaments, crossing the dorsum of the hand and thumb, and are often difficult to isolate. (See Wrist Joint Anatomy for more information.)

In the distal forearm, the anterior and posterior interosseous nerves and vessels may be identified running along their respective courses anterior and posterior to the interosseous membrane toward the elbow. Proximally, the forearm becomes divided into distinct anterior and posterior muscular compartments. These compartments, combined with thicker skin and subcutaneous tissue, create healthier flaps for closure and stump padding. The radial artery courses down the lateral aspect of the forearm, with the superficial branch of the radial nerve in the interval between the brachioradialis and the FCR.

The median nerve passes deep to the fibrous arch of the origin of the flexor digitorum superficialis (FDS), where it continues distally in the middle of the anterior compartment. Just deep to the median nerve lies the anterior interosseous nerve and artery. The ulnar artery and nerve travel together deep to the FCU and FDS for most of the forearm length. However, near the elbow, the artery leaves the ulnar nerve laterally and travels toward its bifurcation from the brachial artery just distal to the antebrachial fossa.

For more proximal forearm amputations, care must be taken to preserve the pronator teres, which originates from the medial epicondyle of the humerus and the medial aspect of the coronoid and inserts on the middle third of the lateral radius. Not only do a significant number of essential neurovascular structures pass through or in close proximity to the muscle, but also, if a Krukenberg procedure is being considered, the pronator teres serves as the only motor unit for function.

Procedural planning

The surgeon faces many challenges over the course of treatment of the individual with an amputation. As with any potential amputation, the surgeon must decide the salvageability of a limb; this decision is often made quickly in cases of trauma or sepsis. Once the decision to amputate has been made, the level of amputation must be determined.

Knowledge of the functional limitations of amputation levels and prosthetic designs, as well as the patient's emotional, physical, and vocational background, must be carefully considered. This is especially true with the upper extremity. As a result, the surgeon walks a tightrope. Preservation of length in the upper extremity is paramount, but not at the cost of lost stump viability, appropriate bone coverage and padding, and optimal prosthetic fitting.

The surgery itself has its own risks of anesthesia and cardiovascular collapse, as well as early postoperative infections and pulmonary embolism. Later, and perhaps more specific to individuals with amputations, are occurrences of joint contractures, phantom limb pain, neuroma formation, stump breakdown, and, in children, bony overgrowth.

Individuals with amputations should undergo comprehensive physical and emotional rehabilitation. A person with an amputation is a patient for life. Close coordination with a team of specialists in physiatry or rehabilitative medicine, prosthetist, physical therapist, and psychologist is ideal. Unfortunately, with shrinking medical reimbursements and funds to support such endeavors, the surgeon is often left to direct all rehabilitative efforts alone. This is a formidable task for anyone, especially for a surgeon who frequently has no training, experience, or interest in the area.

The following are major goals of amputation surgery in the upper extremity:

• Preservation of functional length
• Durable coverage
• Preservation of useful sensation
• Prevention of symptomatic neuromas
• Prevention of adjacent joint contractures
• Controlled short-term morbidity
• Early prosthetic fitting, when applicable
• Early patient return to work and play

Persons with amputations at the appropriate level with proper surgical technique do very well. Complications outlined previously are generally prevented and managed successfully. A patient's preamputation attitude, motivation, and desire strongly influence the overall outcome. However, if proper follow-up care and rehabilitation are not coordinated with a multispecialty team of surgeons, physical therapists, physiatrists, prosthetists, and psychologists, then a less optimal result is inevitable.[14]

Generally, the longer the residual stump, the greater the residual function with or without a prosthesis. A study examining the outcomes of upper-extremity amputations found that individuals with below-elbow amputations performed two-handed activities more easily than did persons with above-elbow amputations; however, there was no significant difference between the two groups in performing activities of daily living. Although all attempts should be made to preserve elbow function, this finding suggests that perhaps the quality of life of the person with an upper-extremity amputation is not as dependent on the elbow as had previously been believed.[15, 16, 17]

Maldonado et al retrospectively assessed the role of elective amputation after brachial plexus injury in nine patients, all of whom had (1) panplexus injury; (2) nonrecovery (midhumeral amputation) or elbow flexion recovery only (forearm amputation) 1 year after all other surgical options were performed; and (3) at least one chronic complication (chronic infection, nonunion fractures, full-thickness burns, chronic neck pain with arm weight, etc).[18]

Five of the nine patients experienced reduced pain.[18]  There were no statistically significant improvements in subjective patient assessments and visual analogue pain scores after amputation. Nevertheless, four patients described their shoulder pain as feeling better after amputation than it did before, and two stated that the operation completely cured their chronic pain. 

Salminger et al reported normative outcome data of prosthetic hand function in 17 below-elbow amputees, using four different objective measurements closely related to activities of daily living (ADLs): the Action Research Arm Test, the Southampton Hand Assessment Procedure, the Clothespin-Relocation Test, and the Box and Block Test.[19]  The mean Action Research Arm Test score was 35.06 ± 4.42 out of 57; the mean Southampton Hand Assessment Procedure score was 65.12 ± 13.95 points; the mean time for the Clothespin-Relocation Test was 22.57 ± 7.50 s; and the mean score on the Box and Block Test was 20.90 ± 5.74.

Clinical presentation

Patients with vascular compromise or occlusion present very differently, depending on the etiology. For example, patients with vascular occlusion secondary to acute embolic phenomena from a more proximal arterial graft typically present with a cold, pale, initially painful portion of the upper extremity with absent capillary refill. Because of collateral circulation, the location of embolic occlusion is often difficult to determine on the basis of clinical appearance. In such individuals, arteriography or magnetic resonance angiography (MRA) confirms the location of the occlusion and assists in determining the level of intervention.

In persons with acute occlusion, medical or surgical thrombolytic recanalization or vascular bypass efforts should be pursued. However, if these efforts fail or if the devascularized tissue has undergone irreversible injury, then amputation is indicated.[8]

Incidents of chronic ischemia, such as occur in persons with diabetes or peripheral vascular disease, are less common in the upper extremity than in the lower. However, revascularization efforts in this situation are less successful, and frequently the decision is made to proceed to amputation. Chronic ischemic injury begins distally and usually progresses proximally to more viable tissue. For this reason, the extent of ischemic injury may not be appreciated fully soon after the initiation of clinical changes.

In individuals with chronic vascular insufficiency, patients remain quite functional for many years despite intermittent reports of mild pain with activity or cold intolerance. The only skin changes, if any, may be those of atrophy (shiny hairless skin). For example, patients may present with acute onset of pain in the ipsilateral index and long digits with no or minimal skin discoloration acutely. Over the course of 24-72 hours, skin usually turns cyanotic, and pain is replaced with decreased sensation. Provided the patient is not acutely ill from sepsis, amputation at this time is discouraged. Patients with severe chronic vascular insufficiency may experience constant symptoms and may be more susceptible to irreversible ischemia.

In persons with chronic ischemia, allowing the extent of the ischemia to declare itself clinically is far preferable. Cyanotic fingers may turn black with time, and more proximal tissue that had been assumed to be viable may follow, with cyanosis observed in the fingers. Once the progression of ischemia has stabilized, plans for definitive amputation may commence. If the level of amputation is in question, specific tests may be performed to assess the viability and healing potential of the tissue.

Thermal burns and frostbite rarely result in amputation more proximal to the hand. However, with extensive injury, amputation may be required. In general, both thermal burns and frostbite injuries should be managed nonoperatively until the extent of the damage can be assessed accurately and the amputation can be performed at the most distal level consistent with good healing. Pyrophosphate nuclear scanning has been demonstrated to be useful in predicting the need for amputation in these situations.

Even in the setting of trauma, the level of amputation may be difficult to determine. Most cases of trauma involve significant avulsion and crush components that leave obvious devitalized tissue exposed.[20] The complete extent of the injury zone may not be apparent on initial presentation. When in doubt, especially for grossly contaminated wounds, it is wise to proceed with open amputation to allow the wound to declare itself prior to closure over a definitive stump length.

Neglected compartment syndromes in the upper extremities with resulting global ischemia often necessitate amputation. Provided that the patient is not in a septic state or otherwise medically compromised, it is preferable to avoid acute amputation for the status of the tissues to be better assessed. Initially, fasciotomies are performed, and if the patient remains systemically stable, the role of initial debridement should be to remove tissue that is obviously dead. Tissue that is neither contractile nor bleeding should be removed at this time. Tissue that is noncontractile but bleeding and otherwise healthy appearing should be left intact, and fasciotomies should be left open with a sterile dressing that prevents desiccation.

The patient should then return to the operating room within 24-48 hours for a second observation of the tissues, and the tissues should be debrided as described above. This conservative process continues until the tissues have stabilized and the surgeon is convinced that all remaining tissue is viable.

This step-by-step conservative debridement, though labor-intensive, ensures that tissue removal is minimal and that the patient is left with maximal function. Even in severe cases where amputation is indicated, this stepwise process dictates the level of amputation and ensures maximum length rather than arbitrarily guessing at the level, which may be inappropriately long (resulting in failure of healing) or inappropriately short (resulting in decreased functional potential). However, the step-by-step process is contraindicated in patients with systemic sepsis, renal compromise secondary to disseminated myoglobin, or another critical illness in which the patient cannot sustain multiple surgeries.

Laboratory studies

Hematocrit and hemoglobin levels should be monitored. In trauma situations, acute blood loss is a concern. Even with elective amputations, postoperative bleeding and hematoma formation require careful assessment. Acceptable levels are individualized according to age, associated medical problems and injuries, and baseline values. In general, a young, otherwise healthy patient should maintain a hematocrit/hemoglobin level greater than 20/6. Elderly patients or patients with underlying cardiovascular disease should be maintained at higher levels (30/10).

Creatinine levels should be monitored. In individuals with muscle injury and necrosis, myoglobin enters the systemic circulation and can lead to renal insufficiency and failure. This is especially true in individuals with thermal and electrical burns. If creatinine levels continue to rise more than 0.4 over baseline, preoperative, or preinjury levels, more aggressive surgical intervention and fluid hydration should be considered.

Potassium and calcium levels should be monitored. As dead tissue is metabolized, destroyed cells release intracellular stores of potassium and calcium into the extracellular space. Elevated levels of these electrolytes may lead to cardiac arrhythmias and seizures.

White blood cell count, C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) values should be monitored in persons with infection. Observe that these values normalize following amputation, thus suggesting resolution of the infection. Expect CRP to be the first laboratory value to respond to treatment; the two others may take several days to weeks to normalize despite eradication of the infection. If these values remain elevated or rise further, treatment should be reassessed by considering changing to a more appropriate antibiotic, searching for an unrelated occult infection or hidden abscess, and possibly revising the amputation at a more proximal level.

Platelets should be monitored periodically if subcutaneous heparin is administered postoperatively. Heparin-induced thrombocytopenia may occur even in small subcutaneous doses.

Imaging studies

With few exceptions, plain radiography should be included in preoperative planning. The presence of hardware or occult pathology in an extremity to be amputated is an embarrassing intraoperative discovery.

In oncology cases, preoperative computed tomography (CT) and magnetic resonance imaging (MRI) have proved invaluable in assessing the extent of tumor involvement and level of amputation. Occasionally, preoperative CT or MRI is helpful to evaluate the extent of infection and abscess in these particular settings.

Arteriography remains the criterion standard for the definitive analysis of vascular status. However, because this is an invasive procedure, arteriography carries increased risks for pseudoaneurysm, hematoma, and vascular embolism.

MRA remains a noninvasive alternative to arteriography. The complications of arterial puncture are avoided, the risk of contrast-related renal failure is eliminated, and sensitivity is higher than in contrast angiography in the identification of severe peripheral arterial occlusive disease. Because of the expense and expertise involved, many medical centers cannot offer this alternative.

Pyrophosphate nuclear scanning has been introduced as another noninvasive method of evaluating tissue viability.[21] It has been demonstrated to be a useful adjunct in predicting the need for amputation in persons with extremities damaged by electrical injury, frostbite, or invasive infection. Pyrophosphate nuclear scanning has a sensitivity of 94%, a specificity of 100%, and an accuracy rate of 96%.

Doppler ultrasonography (US) detects blood flow, and, when used in conjunction with blood pressure cuffs, can measure arterial pressure at different levels in the upper extremity.

Other studies

Transcutaneous oxygen tensions reflect tissue perfusion. Significant occlusive disease causes these measurements to fall below 35 mm Hg. When considering the level of amputation, it is imperative that transcutaneous oxygen tensions at the level of incision are at least 35 mm Hg because measurements below this are associated with decreased healing and wound problems. Measurement of tissue oxygen tension is not affected by incompressible calcified vessels and appears to be very sensitive in evaluating arterial occlusive disease during exercise.

The time for prosthetic fitting in the person with upper-extremity amputation has been debated. Traditionally, fitting commences once stump shrinkage has subsided, usually at 8-12 weeks. However, many have advocated immediate or early prosthetic fitting to improve outcome, particularly improvement in early bimanual activities and increased user rates.

Malone et al[22] describe a "golden period" of fitting within the first month following amputation of the upper extremity in order to maximize acceptance rates and user patterns, regardless of the type of prosthesis initially provided. They also observed that if the fitting occurred within this time, patients demonstrated decreased edema, decreased postoperative pain, decreased phantom pain, accelerated wound healing, improved rehabilitation, and decreased hospital stays compared to those with later fittings. These benefits were less pronounced at amputation levels above the elbow.

Unlike persons with lower-extremity amputations, most persons with upper-extremity amputations have excellent vascularity in their stumps and are much less prone to wound-related problems. It would appear that in the upper extremity, particularly for below-elbow amputations, the advantages of immediate or early prosthetic fitting far outweigh the disadvantages.

A patient's preamputation attitude, motivation, and desire strongly influence the overall outcome. Grunert et al[23] strongly recommended a psychological consultation within the first 3 days of injury. Psychological testing has proved useful not only for predicting outcome but also for assisting in the patient's return to gainful employment.

Transcarpal amputation

At this level, supination and pronation of the forearm, as well as flexion and extension of the wrist, are preserved and can improve the patient's overall function. Furthermore, in comparison with more proximal amputations, the long lever arm provided by transcarpal amputation increases the ease and power with which a prosthesis can be used. However, prosthetic fitting is more difficult and requires a skilled prosthetist.

Ideally, a long full-thickness palmar and shorter dorsal flap should be created in a ratio of 2:1. Finger flexor and extensor tendons should be drawn, divided, and allowed to retract deep into the proximal wound. Conversely, wrist flexor and extensor tendons are identified and released from their distal insertions and reflected proximally out of the way.

Median and ulnar nerves are identified and sectioned well proximal to the amputation site so that the inevitable neuroma formation will occur in a more proximal padded site. At this level, the radial nerve has divided into fine filaments, and each should be sectioned as described above.

After the radial and ulnar arteries are ligated just proximal to the intended level of bone section, the carpus is transected, and all rough edges are rasped to form a smooth rounded contour. The wrist flexors and extensors should be anchored to the remaining carpus in line with their insertions so as to preserve active wrist motion.

Wrist disarticulation

Wrist disarticulation has many of the same advantages as transcarpal amputation with regard to providing a long lever arm and preserved supination and pronation. However, wrist flexion and extension are sacrificed. Because of their length, conventional wrist units are not used, and myoelectric fitting is problematic, as in persons with transcarpal amputations. However, current wrist disarticulation prostheses can be fashioned with thin wrist units that minimize the length discrepancy between upper extremities.

A long full-thickness palmar flap and short dorsal flap at a 2:1 ratio are elevated, with the incision beginning and ending 1.5 cm distal to the radial and ulnar styloid processes, respectively. Identify and ligate arteries and nerves as described above.

Alternatively, Louis et al[24] described a technique for minimizing postoperative pain from neuroma formation, which involves extending the incisions proximally between the pronator teres and brachioradialis just distal to the elbow flexion crease and doubly ligating the median, ulnar, and superficial radial nerves at this level. This allows a neuroma-in-continuity to form at this site and away from the prosthetic wall or scar, where it may cause symptoms.

Martini et al[25] reported excellent results achieved by freeing the epineural sleeve of the nerve stump approximately 5 mm from the nerve fascicles. The epineurium tube is then filled and sealed with butyl-cyanoacrylate. However, this technique has not been as commonly used in the United States. It is not necessary to ligate the antebrachial cutaneous nerves in the proximal forearm. Electrocautery is used to achieve hemostasis of the anterior and posterior interosseous vessels.

Preserving the triangular fibrocartilage and avoiding damage to the distal radioulnar joint (DRUJ) are imperative. Otherwise, pronation and supination will be reduced, painful, or both. The prominent radial and ulnar styloid processes should be rounded off with a rasp; however, shortening of the radial styloid process should be avoided. Preserving the radial styloid flare improves prosthetic suspension.

Wrist disarticulation is the procedure of choice in children. In general, disarticulations are preferable to transections through bone at a more proximal level because the distal physis is spared and, consequently, the growth of the distal stump continues at a normal rate. In addition, disarticulation prevents terminal overgrowth of the bone.

Distal forearm amputation

Whereas maintaining length remains an important consideration in the upper extremity, the underlying soft tissues in the distal forearm consist of relatively avascular structures, such as fascia and tendon, and may not always offer adequate padding for the bony stump. Furthermore, the skin and subcutaneous tissue in this area are thin and may be predisposed to wound problems.

A good compromise between adequate functional length and adequate wound healing appears to be at the junction of the distal and middle third of the forearm. Despite resection of the DRUJ, some degree of pronation and supination is preserved in persons with forearm amputations. The extent of motion is dependent on the length of residual forearm stump; the longer the stump, the greater the arc of motion.

As with more distal amputations, flaps are fashioned distal to the intended level of bone amputation. However, at this level, anterior and posterior flaps of equal dimension are created and reflected proximally. Vessels, nerves, and tendons or muscle bellies (depending on the level of amputation) are transected similarly to the methods already described.

Proximal forearm amputation

The technique for proximal forearm amputation is similar to that already described for more distal amputations. A short stump having as little as 4 cm of ulna is preferable to an above-elbow amputation. To facilitate prosthetic fitting in these extreme cases, detaching the biceps tendon and reattaching it proximally to the ulna at a position approximating its resting length is advisable. Distal reattachment should be avoided because it may cause a flexion contracture at the elbow.

In individuals with tenuous soft tissue coverage for the stump, rather than resorting to an above-elbow amputation, Jones et al[26]  described a salvage technique in which a free latissimus dorsi flap is used. From a functional standpoint, preserving the patient's own elbow is extremely important. Even in persons with very short stumps, improved fitting prosthetic devices such as the Munster device or a split socket with step-up hinges can be used to achieve excellent function.

In the so-called spare parts method sometimes used for large complex defects, tissue for the flap is sourced from the amputated part itself.[27]

Krukenberg procedure

More than 80 years ago, Krukenberg described a technique that converts a forearm stump into a pincer that is motorized by the pronator teres. Indications for this procedure have been debated; however, they generally include bilateral upper-extremity amputations, especially in those who are also blind. The procedure also has been used successfully in persons in developing countries who lack the means to obtain expensive prostheses.

This procedure preserves proprioception and stereognosis in the functional stump to allow for effective maneuvering in the dark. It is important to note that this procedure is not recommended as a primary procedure at the time of an amputation, and the procedure must be preceded with appropriate counseling because of cosmetic concerns. Conversely, once this procedure is performed, it does not preclude the use of a functional prosthesis. Therefore, the patient is afforded the option to use either functional strategy.

For this surgical option to be considered, the ulna and radius must extend distal to the majority of the pronator teres (the motor for pinching), and an elbow flexion contracture of less than 70° is required. Swanson et al,[28]  Nathan et al,[29]  and Garst[30]  described several modifications of Krukenberg's original surgical technique, focusing on conservative debulking and flap closure without the need for skin grafts.

The success of this procedure depends directly on the strength of the pronator teres, the sensibility of the skin surrounding both ulna and radius, elbow mobility, and mobility of the ulna and radius at the proximal radioulnar joint (PRUJ). Individual patient expectations and motivations, although more difficult to assess, probably play a major role in outcomes as well.

A soft compressive dressing is applied to the stump. The elastic bandage is applied more tightly distally than proximally to prevent stump edema. Rigid dressings and casts, such as those often used on lower-extremity stumps, are unnecessary in the upper extremity. If a drain is used, it is removed within 24-48 hours. If no contraindications exist for anticoagulation, low-dose subcutaneous heparin (5000 U q12hr) may be administered for deep venous thrombosis (DVT) prophylaxis, especially in patients at high risk. Immediate active range of motion of the shoulder and elbow (and wrist, if applicable) is implemented to prevent joint contractures.

Hematoma

Meticulous hemostasis and, if necessary, use of a drain can minimize the occurrence of postoperative hematomas. If allowed to accumulate, hematomas may provide an attractive medium for bacteria and possibly inhibit proper wound healing. Aspiration of the hematoma under sterile conditions is recommended, followed by a compressive dressing to reduce the recurrence rate. If hematoma accumulation persists, surgical exploration, rather than repeat aspirations, may be required to achieve adequate hemostasis.

Infection

Upper-extremity amputation stump infections are more likely to occur in patients who are immunocompromised, those who have previous vascular diseases, and patients who have grossly contaminated or infected wounds. Like any postoperative infection, superficial infections may be treated with proper wound care, antibiotics, and close observation. However, deep infections with or without an abscess may necessitate further surgical debridement and possibly revision of the amputation to a more proximal level.

A study by Ali et al suggested that delaying formal closure of the amputation for at least 5 days after the procedure can lower the postoperative infection rate.[31]

Necrosis

As with infection, superficial areas of skin necrosis may be treated conservatively because healing usually continues under the eschar. However, larger areas of necrosis indicate insufficient vascularization and demand wedge resection or possibly revision of the amputation to a more proximal level.

Contracture

Joint contractures usually are prevented by immediate postoperative active motion. If contractures develop, more aggressive physical therapy may be required, including strengthening the opposing musculature and gentle passive motion. Caution is advised around the elbow because overly forceful passive motion may incite heterotopic ossification.

Neuroma

Regardless of the technique employed to divide peripheral nerves, a neuroma always forms. If the neuroma is compressed against a rigid surface (bone or prosthetic wall) or if the neuroma experiences traction as it remains trapped in the healing scar, then pain is inevitably produced. Efforts to prevent compression or entrapment of neuromas have been described previously.

When a neuroma becomes symptomatic, it usually can be treated by altering the prosthetic socket to avoid pressure or traction on the lesion. When all nonoperative efforts to relieve the pain have failed, the neuroma may be excised successfully and the nerve divided at a more proximal level. In children, neuromas seldom call for surgical intervention.

Phantom sensation

This poorly understood phenomenon is defined as the patient's awareness of the amputated portion of the limb.[32, 33, 34, 35] The sensations may be disturbing but are rarely painful. Many modalities have been used in an attempt to prevent and minimize the intensity of these sensations. Amitriptyline and gabapentin can be considered first-line agents in the pharmacologic treatment of phantom sensations. Other, less commonly used agents are capsaicin, calcitonin,[36] mexiletine, carbamazepine, propranolol, metoprolol, and clonazepam. Many of these drugs have only case report usage and require further investigation.

Promising therapies in the upper extremity have included immediate or early prosthetic fitting. Substantial evidence also exists to support perioperative epidural anesthesia and postoperative intraneural anesthesia applied to transected nerves in amputated lower extremities, which may be effective in the upper extremity. However, such data in the upper extremity are not yet available. Phantom sensations are uncommon in children.

Deep venous thrombosis

Patients who have undergone amputation may be at increased risk for DVT. Although persons with upper-extremity amputations are able to mobilize following their surgery more easily than those with lower-extremity amputations and consequently present with fewer occurrences of DVT, their underlying medical conditions often predispose them to this complication.

Multiple risk factors for DVT, such as age, immobility to some degree, and the amputation itself that involves the ligation of vessels, exist in this population. Furthermore, 25% of patients undergoing vascular surgery have an identifiable hypercoagulable state. Unfortunately, most data pertain to individuals with lower-extremity amputations; therefore, it is difficult to make conclusions about the risks for persons with upper-extremity amputations.

Nevertheless, it would be advisable to administer low-dose subcutaneous heparin to patients while they are in the hospital after amputations, provided that they have no contraindications for anticoagulation. Low-molecular-weight heparins (LMWHs), such as enoxaparin and dalteparin, may increase the risk of bleeding and hematoma formation and may induce more anticoagulation than is necessary in the upper extremity. However, more research is required to assess the risks and benefits of this therapy.

Terminal overgrowth

Although many amputation complications, such as neuroma or phantom pain, are less problematic in children, terminal overgrowth occurs to some degree in all children with amputations; up to 12% of children with amputations require one or more stump revisions.

Forearm amputations are less problematic in the upper extremity than above-elbow amputations are, given that the humerus most commonly is associated with this phenomenon. The radius is the second most likely bone in the upper extremity to demonstrate terminal overgrowth. In individuals with forearm amputations, the most common overgrowth occurs as a pincerlike contour of the radius in relation to the ulna. This deformity occasionally causes the proximal radial epiphysis to tilt. Less commonly, the ulna may overgrow with subcutaneous projection.

The cause of overgrowth is still controversial. Contracture of the soft tissue envelope and disproportionate growth of bone from the proximal physis to the soft tissue have been implicated. However, the current hypothesis, according to Speer,[37] is that the established mechanisms of normal fracture healing and local wound healing malfunction and accelerate.

Pediatric patients with disarticulations do not demonstrate terminal overgrowth, because the articular cartilage acts as a natural barrier for this activity. For this reason, as well as to preserve the distal physis and maintain normal stump growth, disarticulations are the treatment of choice in children whenever possible. Most attempts to prevent terminal overgrowth, including capping the bony ends with Silastic or allograft and autograft tissues, have failed. The only treatment for symptomatic terminal overgrowth is revision of the amputation.

Bony spur

Bony spurs occasionally form at the ends of the bones, especially in children. Unlike terminal overgrowth, bony spurs almost never require resection and are well tolerated with prosthetic socket modifications.