Hand reanimation

Abstract

Brachial plexus disruption, major traumatic amputations, and Volkmann’s contracture are all devastating injuries that present difficult reconstructive challenges. Advances in our understanding of nerve injury, regeneration, and refinement of microsurgical techniques have given rise to a number of therapeutic avenues over the last 4 decades. Hand reanimation aims to provide strength, stability, and mobility to a sensate hand. How this is achieved depends on a thorough understanding of the underlying pathophysiology, which in turn dictates what surgical modalities are suitable. Common to all reanimation procedures is the need to ensure full passive range of motion of the target joints prior to definitive surgery. Hand therapy is essential to prevent deleterious sequelae of injury, and to maximize rehabilitation following surgical reconstruction. Options for reanimation include nerve repair, nerve grafting, nerve transfer, tendon transfer, and free functioning muscle transfer.

Introduction

Hand surgeons routinely call on reconstructive techniques that give reliable, predictable outcomes for tendon injuries, fractures, and soft tissue loss. However, there are a group of more devastating injuries for which outcomes are unpredictable, namely following brachial plexus injury, replantation proximal to the wrist, and Volkmann’s contracture. These complex conditions generally involve greater energy transfer during the traumatic event, and are accompanied by loss of specialized tissues that have poor capacity for regeneration. The resulting healing by scar leaves functionless areas that are stiff, unyielding, possibly painful, with progressive contractures.

Historically, such injuries were labelled as ‘unreconstructible’. Patients were left to cope with an unusable upper limb. However, over the last 4 decades huge advances in our understanding of nerve injury and regeneration, coupled with refinement of microsurgical techniques, have presented a number of promising therapeutic avenues for reanimation of the hand. In most cases, return to normal function after severe high energy injuries remains an unreachable goal. However, recalling the adage “when you have nothing, a little is a lot,” seemingly modest functional gains can make a significant difference to patients. In general terms, reanimation surgery aims to provide strength, stability, and mobility to a sensate hand. To what extent these goals are achievable depends upon the severity of injury and tissue loss, coupled with the patient’s biology and motivation. A range of reanimation techniques is available but their appropriate application requires knowledge of the underlying pathology in each case. The ultimate reconstructive plan is drawn up based on a full clinical assessment, that may include specialized imaging and electrophysiological testing.

Volkmann’s contracture

This results from an ischemic insult or direct crush of muscle, leading to necrosis and subsequent fibrosis [1]. Although both intrinsic and extrinsic hand muscles may be involved, the hand typically adopts an intrinsic minus posture because the extrinsics are stronger. The deep volar forearm compartment is most often involved, with contracture of flexor digitorum profundus, flexor pollicis longus and restricted supination because of contracture of the pronator muscles. More severe cases involve the entire volar muscle mass, with or without extensor involvement. Mild cases with some residual function can be adequately managed with tendon lengthening or muscle slide procedures. Severe cases require more sophisticated procedures detailed below. In addition to muscle fibrosis and contracture, nerve lesions are prevalent in Volkmann’s contracture. Both median and ulnar nerves are often affected as a result of either the ischemic insult or mechanical trauma. Mechanical compression may develop gradually as evolving muscle fibrosis encases the nerves. For this reason, more aggressive early excision of infarcted muscle has been advocated [2]. Furthermore, early excision of infarcted muscle is easier while it is soft and tissue planes are relatively preserved.

Brachial plexus injury

Approximately 70 % are closed traction injuries that produce irreparable nerve avulsions from the spinal cord [3]. The hand is affected in the majority of cases. In contrast to nerve avulsions, closed nerve rupture or open nerve division has capacity for regeneration because proximal nerve stumps remain in continuity with the central nervous system. Anatomical nerve reconstruction by primary nerve repair or nerve grafting is relatively effective for shoulder reinnervation, but long regrowth distances render these techniques ineffective for the hand. Some recovery of long flexor function is achievable, but the consistent failure of intrinsic muscle reinnervation means few will attempt proximal C8/T1 repair except in the pediatric population [4]. Alternative techniques of reanimation are considered below.

Replantation

Considerable success has been reported with distal replantations of the digits and hand [5]. However, as the level of amputation migrates proximal to the wrist the magnitude of energy transfer and zone of trauma generally increase. This results in more unpredictable outcomes both in terms of initial success of revascularization and long-term function [6]. Functional problems following replantation relate to the degree of muscle debridement required, which may leave insufficient muscle bulk for strength or inadequate muscle length for excursion. In addition, denervation, scarring, tendon adhesion, and joint contractures all interfere with recovery of function. Our practice is to consider secondary reanimation procedures once the patient’s progress with hand therapy has plateaued, which usually occurs a minimum of 6 months postinjury.

Reconstructive tactics

A variety of reanimation techniques can be used to augment the function of a severely disabled hand. Understanding the underlying pathology will determine whether procedures on nerve, tendon, muscle or combinations of these are most appropriate. However, prior to their deployment, a number of conditions should be met to maximize the chances of a favorable outcome. These are detailed in Table 1. Arguably, the most absolute of these is the need for supple joints. It is unreasonable to expect any reanimation technique to move a stiff joint, and the first step in any reanimation effort is to lay the groundwork to ensure total passive motion exceeds total active motion.

Table 1.

Prerequisites for hand reanimation

1. All fractures should be healed or internally fixed with absolute stability.

2. Skin and soft tissues overlying the planned area of reanimation should be unscarred, well-vascularized and pliable.

3. Full passive range of motion of the target joints should be achieved.

4. Potential donor sites of muscles, tendons or nerves should be carefully evaluated to ensure harvest does not cause new functional problems.

Laying the groundwork

Stiffness of the hand has a detrimental effect on outcomes following injury and reconstructive surgery. It is a particularly challenging problem after major injury, when extensive scarring, prolonged immobilization and Volkmann’s contracture all play a role in its etiology. Successful management requires thorough knowledge of the underlying pathophysiology, integrated team-working between hand surgeons and therapists, and a motivated well-informed patient. In general, maximal surgical gains are achieved only after maximal nonsurgical gains have first been attained. To that end, the importance of early hand therapy cannot be over-emphasized.

Hand therapy

The role of therapy is 2-fold: (1) to preserve hand function as much as possible during recovery from injury or surgery, and (2) to maximize hand function once healing has occurred. Early therapy tackles hand stiffness proactively by minimizing deleterious sequelae of the inflammatory response to injury. Following injury, capillary permeability increases in response to cytokines and chemokines released from platelets and neutrophils. This allows egress of protein-rich fluid into the interstitium causing edema, distension of joint capsules and skin envelope, and increased resistance to normal movement. If ignored, this proteinaceous exudate contributes to interstitial fibrosis and adhesion formation. In addition, nerve and muscle injury often result in decreased joint motion that, if ignored, lead to contractures and ligamentous shortening [7].

The most basic of maneuvers involves elevation of the affected hand above the level of the heart to minimize edema. Statically splinting the hand in a ‘POSI’ (Position Of Safe Immobilization) places the wrist in 30° of extension with metacarpophalangeal joints (MCPJ) flexed to 70°–90°, interphalangeal joints (IPJ) in full extension, and thumb abducted and pronated. This is the opposite of the position of comfort that most patients adopt, sometimes known as the ‘intrinsic minus’ hand. The POSI minimizes risk of joint contracture by placing the MCPJ collateral ligaments and IPJ volar plates on stretch and prevents first webspace contracture.

More advanced multi-modal therapies include elastic bandage wraps, massage/external lymphatic drainage, ultrasound, hot/cold therapy, compression garments, and specialized splinting techniques. The aim is to remodel healing tissues into a configuration that favors normal movement, rather than resists it. Low-load prolonged stress has been shown to be more effective than brief high-load stress, and is normally achieved by serial static splints that are gradually molded over the course of weeks to overcome joint stiffness. The vast majority of hand joint contractures can be improved in this way, with operative intervention only considered when progress stalls with poor resultant function [8].

Surgery

Procedures to optimize the hand prior to subsequent reanimation surgery are undertaken with extreme caution. The surgeon is placed in a ‘Catch 22’ situation where surgery may be required to release scar tissue, but the surgery itself will inflict further soft tissue injury, scarring, and edema. Co-ordination with a hand therapist is essential to ensure the initial benefits of surgery are maintained throughout the recovery period. A thorough clinical assessment should determine what structures or tissues require release such that surgery is appropriately targeted and unnecessary dissection avoided. A plan for any reconstruction required should be discussed with the patient. For example, chronic joint contractures may be associated with skin and soft tissue deficits that only become apparent after the joint is released.

There is no clear evidence-based guidance on whether such surgery should be staged or performed at the same time as reanimation procedures. Advantages of single stage surgery include shorter ‘downtime’ for the patient and the possibility of earlier rehabilitation. Extensive dissection is often required to lay open the whole zone of trauma. Intuitively, less scarring is induced if this is only done once. However, single stage surgery may require particular expertise with composite free tissue transfer to simultaneously import bone, muscle, nerve, tendon and skin. Single stage surgery may also be inappropriate when both flexor and extensor deficiencies require reconstruction because optimal rehabilitation and splinting would be compromised in favor of one to the detriment of the other [9]. The major disadvantage of staged surgery is the delay required between surgeries. At least 3 months should separate each stage to allow tissues to reach equilibrium before inflicting further surgical trauma.

Soft tissue reconstruction

Skin contractures can significantly limit excursion of the underlying joints. Reconstructive options range from simple scar release with lengthening Z-plasty or V-Y plasty, to skin grafts, local or regional flaps, and free tissue transfer.

Reconstruction should be accomplished by the simplest method that gives optimal function. Although skin grafting is a relatively simple technique, we often find that it falls short on the requirement for function. Skin grafts are prone to fibrosis and scar contracture, and placement onto paratenon inevitably results in tendon adhesion. It is also near-impossible to perform any secondary surgery, such as tendon transfer or nerve grafting, in areas previously skin grafted with little supple subcutaneous tissue.

Furthermore, vascularized soft tissue flaps of fascia (eg, temporoparietal fascial flap) or muscle (eg, latissimus dorsi) are prone to contracture and tendon adhesion. In our experience, the best approximation of replacing ‘like tissue with like tissue’ is using a supple skin flap. Small defects may be readily covered with local flaps adjacent to the defect, or regional flaps such as the posterior interosseous flap or reverse radial forearm flap. Our preference is free tissue transfer of the anterolateral thigh flap due to its versatility of inset, composition, and favorable donor site morbidity [10]. Using tissue from a distant donor site limits further trauma to the injured limb.

First, webspace contractures are common following denervation seen in brachial plexus, Volkmann’s contracture, and postreplantation cases. Most can be adequately treated with local soft tissue rearrangement in the form of a double-opposed Z-plasty, 4-flap Z-plasty, or ‘jumping man’ Z-plasty. However, some cases of Volkmann’s contracture where the intrinsic muscles may be fibrosed require more extensive release with importation of vascularized tissue (eg, lateral arm free flap, free groin flap, reverse radial forearm flap) to fill the resulting defect.

Tenolysis

Tendon adhesion is a common impediment to rehabilitation following replantation. Tendons may be adherent to each other, to bone or fracture site, to joint capsule, or to overlying skin and soft tissues. Paratenon should be preserved during tenolysis to maintain the tendon’s blood supply. If both flexor and extensor tenolysis is indicated, the extensor surface is addressed first because extensor mobility can be adequately maintained with passive motion whereas flexors tend to do better with active mobilization.

Joint procedures

If tenolysis fails to correct limitation of motion, surgical joint release is considered. This usually involves dorsal capsulotomy or capsulectomy of the MCPJ and check-rein ligament/volar plate release of the PIPJ. If further release of the PIPJ is required, the accessory collateral ligaments may be divided. Total detachment of the true collateral ligaments is avoided to prevent joint instability.

Unfavorable wrist posturing due to contracture or paralysis can be a difficult problem to reanimate, partly because efforts are concentrated on reanimating the hand joints and partly because wrist fusion is considered an excellent option. Long term wrist malposition leads to carpal malalignment and instability, making it difficult for patients to properly position their hand in space. Wrist fusion in a position of function —usually around 20° of extension—gives a stable platform for reanimation of a functional hand [11•].

Primary nerve repair

Acute nerve transection, seen after upper limb amputation or open penetrating brachial plexus injury, may be amenable to primary nerve repair. In general, this is only possible in the acute postinjury period prior to retraction and scarring of the nerve stumps. The aim is to coapt the nerve ends anatomically in a tension-free manner, usually with the aid of an operating microscope. This allows sprouting axons from the proximal stump to cross the repair site and proceed toward their targets at a growth rate of approximately 2.5 cm per month. Excessive tension on a repair reduces microvascular flow, impairs nerve regeneration, and compromises outcome [12]. Alternative methods of nerve reconstruction should be employed if tension-free repair cannot be achieved. In our experience, median and ulnar nerve repairs at forearm level result in return of protective sensation more reliably than return of motor function and alternative tactics of reanimation may still be required. This is particularly the case for proximal nerve lesions, where regenerating motor axons may not reach their target muscles within the critical period of 12–18 months. After this time the motor end-plates degenerate, fibrose, and the muscle becomes unresponsive to neurotization. In contrast, sensory end-organs do not exhibit such a critical period and may be neurotized successfully many years following denervation.

Nerve grafting

If divided nerve stumps cannot be re-approximated without tension, either for reasons of delayed surgery or segmental nerve damage, a nerve gap exists. Such gaps are common after closed brachial plexus ruptures where the longitudinal zone of injury may be extensive. Debridement of chronically constricted nerves in the forearm following neglected Volkmann’s contracture also produces nerve gaps. Segmental nerve loss is a feature of high-energy avulsion amputations of the forearm. It should be borne in mind that motor recovery is unlikely to occur if regenerating axons cannot reach their target muscles within 12–18 months.

Nerve autografting is the current gold standard for bridging gaps in critical major nerves. They provide a route for axonal regrowth and viable Schwann cells, which are known to be required for nerve regeneration [13]. Donor nerves are typically sensory, and include the sural, medial antebrachial cutaneous, lateral antebrachial cutaneous; rarely the superficial radial nerve and saphenous nerve may be used. Large-diameter nerve gaps are bridged with multiple segments or ‘cables’ of smaller diameter nerve graft. The disadvantage of autologous nerve harvest is the requirement for a second surgical site and resultant donor site sensory loss.

Larger nerves may be used as vascularized interpositional grafts, either pedicled or free, if they are non-functional [14]. Large nerves must be vascularized to avoid necrosis of the central portion of the nerve. The ulnar nerve is often used in this manner after nonrepairable lower plexus avulsion injury. If autologous nerve tissue is not available or its harvest undesirable, cadaveric nerve allograft may be used. Systemic immunosuppression is required for up to 18 months while the patient’s axons traverse the segment of allograft [15]. Processed allograft, chemically stripped of cellular material, has been used successfully without need for immunosuppression [16].

Alternatives to nerve tissue include synthetic or biological nerve conduits. Non-neural biological materials, such as segments of vein, artery, small bowel, or muscle, are not widely used nowadays. Synthetic nerve conduits composed of polyglycolic acid and bovine collagen have shown promise but their use is limited to small-diameter nerves, such as digital nerves [13, 16]. They may be used for large-diameter nerves, but the failure rate is higher and their efficacy wanes as the size of the nerve gap exceeds 3–4 cm.

In practice, all but the most noncritical of nerve gaps are managed with autologous nerve graft whenever possible.

Nerve transfers

Nerve transfers involve division and coaptation of a healthy donor nerve to a nonfunctioning recipient nerve. Nerve transfers have become a popular method of replacing functional nerve loss to the hand as a result of brachial plexus and proximal nerve lesions. Such injuries, if repaired or grafted, require nerve regrowth over long distances through scarred wound beds that hinder muscle reinnervation within the critical 12- to 18-month period. In the case of brachial plexus root avulsions, proximal nerve stumps are not available for either repair or grafting. Nerve transfers circumvent many of these problems by supplying viable donor axons close to the target end-organ (motor or sensory), thus by-passing any proximal zone of trauma and scarring and minimizing the distance required for nerve regrowth. This technique is, therefore, suitable for patients that present relatively late following injury, because neurotization can occur within weeks following nerve transfer [17••].

Nerve transfers are of limited value in hand reanimation following Volkmann’s contracture because the muscle end-organ is usually damaged. Similarly, extensive pan-plexus injuries render most distal nerve transfers to forearm and hand unusable, although more proximal extra-plexal transfers are commonly used for the shoulder, elbow, and wrist, such as intercostal nerve transfers to median nerve. Harvest of the contralateral C7 nerve root, extended by a vascularized ulnar nerve graft to reach the median nerve, is rarely used outside of Asia due to modest functional improvements and theoretical risks of sacrificing function in the remaining uninjured upper limb [18, 19].

Nerve transfers require an expendable motor or sensory nerve close to the intended target. The transferred nerve is traditionally coapted end-to-end to the recipient nerve, however, end-to-side and reverse end-to-side techniques have also been shown to be effective. End-to-side neurorrhaphy joins the distal end of the recipient nerve to the side of an intact donor nerve. The intact donor nerve then sprouts axons into the recipient nerve, albeit with variable results. It would appear that sensory axons sprout readily, but motor axons of the donor nerve require an iatrogenic axonotmesis-type injury to stimulate sprouting [20]. This is typically delivered as a crush of the donor nerve proximal to the neurorrhaphy.

Reverse end-to-side neurorrhaphy involves dividing the donor nerve and coapting its proximal end to the side of the recipient nerve. This is useful when a proximal nerve injury is expected to recover as a result of primary repair or grafting, but regrowth distances are sufficiently long that there is significant risk of permanent changes to motor end-plates. Reverse end-to-side transfers reinnervate the target muscle within weeks, effectively eliminating any concern about the critical 12–18 month period, while still allowing the native proximal nerve to reinnervate its targets in due course. The transferred nerve effectively ‘super-charges’ the proximal nerve repair by supplying additional axons close to the target.

Common nerve transfers for hand reanimation are detailed in Table 2.

Table 2.

Common nerve and tendon transfers

Nerve transfers	Donor	Recipient
Finger extension	FCR branch	PIN
Finger flexion	FDS branch, or  ECRB branch, or  Supinator branch, or  Brachialis branch	AIN
Hand intrinsics	AIN	Deep motor branch of ulnar nerve
Tendon transfers	Donor	Recipient
Finger extension	FCU or FCR  PL	EDC and EPL  APL and EPB
Finger flexion	ECRL  FCR  Brachialis or biceps	FDP  FDP  FDP and FPL
Pinch	Brachioradialis  Brachioradialis	FPL  FDS opponensplasty

AIN anterior interosseous nerve, APL abductor pollicis longus, ECRB extensor carpi radialis brevis, ECRL extensor carpi radialis longus, EDC extensor digitorum communis, EPB extensor pollicis brevis, EPL extensor pollicis longus, FCR flexor carpi radialis, FCU flexor carpi ulnaris, FDP flexor digitorum profundus, FDS flexor digitorum superficialis, FPL flexor pollicis longus, PIN posterior interosseous nerve, PL palmaris longus.

Tendon transfers

Although commonly known as tendon transfer, ‘musculotendinous transfer’ is a more accurate descriptor of this technique because transfer of a tendon alone will not bring about reanimation. Although stating the obvious, it emphasizes the importance of a comprehensive clinical assessment to ascertain what potential tendon transfers are functioning and available for use, and what muscles will remain to compensate for donor site morbidity following transfer. Common tendon transfers for hand reanimation are detailed in Table 2.

Tendon transfers have limited applicability following pan-plexus injury, although a successfully reinnervated biceps may be transferred to provide finger and thumb flexion. As previously stated, Volkmann’s contracture has a predilection for the deep volar forearm compartment, meaning more superficial muscles such as the wrist flexors may be available for transfer. A spared flexor digitorum superficialis may be transferred to the deep finger flexors. Alternatively, spared dorsal muscles such as extensor carpi radialis longus can be transferred to the deep finger flexors, with either extensor indicis, extensor digiti minimi or brachioradialis transferred to flexor pollicis longus.

In the context of major forearm replantation, tendons may be transferred not only for secondary reconstruction, but as acute transfers at the time of initial surgery. If a musculotendinous unit has been disrupted distally, such that the proximal muscle with its neurovascular supply is intact, rather than reattach it to its original location it may be transferred to fulfil a function that may otherwise be lost, such as transfer of a wrist extensor for thumb or finger extension.

Acute tendon transfers may also be employed following nerve repair, so that some function is restored immediately to allow rehabilitation while waiting for nerve recovery to occur. Should nerve recovery be poor, the transferred tendon can act as permanent substitute for the nerve’s function.

Free functioning muscle transfer (FFMT)

This is a complex procedure where a healthy donor muscle is transferred from one part of the body to another, where it is revascularized using microvascular techniques and coapted to a local recipient motor nerve. It is ideally suited to situations where attempting nerve reconstruction would be inappropriate due to lack of suitable target muscle. Cases of muscle denervation that exceed the critical 12–18 month period or muscle loss due to trauma or infarction with subsequent Volkmann’s contracture should be considered for FFMT, particularly if other tendon transfer options are not available. Gracilis is the most commonly-used muscle, although latissimus dorsi, rectus femoris and tensor fascia lata are also popular. Selection of a suitable muscle for transfer is based on its strength, excursion, size relative to the defect to be filled, suitable neurovascular anatomy, and resultant donor defect [21••]. In order to ensure the muscle is inset with the correct resting tension, the muscle should be marked either with ink or temporary sutures at predetermined points, eg, every 1 cm or 5 cm, before dividing its origin and insertion. Selection of an appropriate recipient motor nerve is also important. Pure motor nerves that have synergistic function to that replaced by the FFMT are preferred. When functioning peripheral nerves are available in the forearm, FFMT for finger flexion is best motored by the anterior interosseous nerve, or a motor fascicle to the flexor digitorum superficialis (FDS). FFMT for finger extension is best motored by the posterior interosseous nerve, or the motor branch to FDS may be used. When peripheral nerves are not available, as is often the case after a severe brachial plexus injury, extraplexal nerves are used to motor the FFMT. Common donors include 3 contiguous intercostal nerves, or the spinal accessory nerve. To avoid using interposition nerve grafts, the FFMT may be inset in the upper arm to allow direct coaptation of the nerve to either spinal accessory nerve or intercostals. The proximal gracilis muscle is secured to the coracoid process of the scapula, and its long tendon is passed around a pulley of, eg, pronator teres, into the forearm to be woven into finger flexors. Restoring appropriate mechanical balance between wrist and finger flexors and extensors may require transfer of two FFMTs. This was pioneered by Doi, who reported that two-thirds of his patients achieved more than 30° total active finger motion [22]. FFMT can restore 10 %–40 % of grip strength compared with the normal side.

Conclusions

Patients presenting with devastating loss of function following brachial plexus injury, Volkmann’s contracture, or major upper limb amputation present huge reconstructive challenges. Previously unreconstructible problems can now be addressed using combinations of nerve repair, nerve grafting, nerve transfer, tendon transfer, and free functioning muscle transfer. These techniques are constantly being refined as ongoing basic science research sheds light on mechanisms of nerve regeneration. Although current practice focusses on using autologous tissue for reconstruction, advances in our understanding of immunology and induction of immunological tolerance may allow composite tissue allotransplantation of more complete upper limbs than hand transplantation alone. Parallel technological advances are producing increasingly sophisticated myoelectric prostheses that convincingly reproduce hand function without need for further donor site morbidity.