Digital Replantation and Thumb Reconstruction

For videos accompanying this chapter see ExpertConsult.com . See inside cover for access details.

Introduction

Vascular surgeons Jacobson and Suarez introduced the microscope to perfect microvascular small vessel anastomosis in 1960, which rapidly led to its application in revascularizing digits and replantation. ^, It was promptly adopted in Taiwan, the USA, Australia and, later, Europe. China independently and unbeknown to the West had been achieving remarkable successes despite primitive instrumentation and facilities. The need for small animal models for transplantation research led to laboratories being equipped for microvascular surgery and they doubled as training centers for skilling budding young surgeons enthused by the potential of this new technique. Establishing “replant services” became an achievable and desirable goal worldwide. Unsurprisingly, the enthusiasm of reconstructive teams was not met with equal passion from anesthetists, hospital administrators, and other surgical disciplines whose routine was often disturbed by seemingly unending theater occupation and repeat operations for what many saw in the overall scheme of things as a trivial pursuit. Microvascular surgery was here to stay, and there quickly evolved a divide between the older generation of surgeons who were skeptical or intimidated by the technical expertise demanded and those of a younger set who relished the challenge, the new anatomy, and the vistas that were on offer.

Indications for Replantation

Like all surgical procedures, the indication for replantation is when the estimated benefit outweighs the possible risks. The more extensive the injury, the greater the likely benefits of replantation, because in the severely damaged hand a little is a lot.

Intuitively, replanting everything seems the best solution to restoring lost function and this will be the desire of most patients, but it presupposes that the results from replantation will be close to the original. Sadly, experience shows that this is often not the case. The experienced surgeon needs to estimate the outcome taking into account the general fitness of the patient, the extent of injury of the parts, their capacity for repair and functional potential, and the suitability of the patient to commit to the arduous postoperative therapy required to maximize this potential. Having balanced these expectations and limitations, the surgeon must present the patient with an honest as well as a compassionate opinion regarding the wisdom of replantation including the potential need for multiple procedures, flap coverage, ultimate poor result, and later amputation. Mostly the patient in a state of shock will plead that the parts are reattached, and mostly this will be the correct decision. With major limb or multiple digit amputations, rational decisions cannot be made until a thorough assessment in theater is performed. What initially looks to be hopeless can sometimes prove quite feasible; alternatively, the opportunity to prioritize salvageable digits to essential sites, use segments of nonreplantable digits for spare parts, and even banking in ectopic sites, presents itself only at the time of initial surgery. Resuscitation and management of concurrent life-threatening injuries take priority and guide decision-making regarding amputation or replantation and its timing. More severe crush and avulsion mechanisms have less likelihood of overall success compared to guillotine-type injuries. Age, general medical health, occupation, handedness, and expectations are also essential considerations guiding decision-making.

Proper handling of the amputated part is imperative to a successful outcome. The severed part is stored in moist saline gauze inside a watertight bag, placed in a container of ice. Care must be taken to avoid direct contact with ice, which can irreversibly damage the microvascular anatomy. Although the expedient transfer to a microsurgical center is critical, multiple reports have shown successful replantation after more than 24 hours of cold ischemia.

Functional Loss Assessment

If the guiding principle of replantation is to restore lost function then a well-grounded understanding of the basic functions of the hand and the expected outcomes of hand surgery are essential. It follows that a hand surgeon with microvascular expertise should be the director of the expedition and not an enthusiastic microsurgeon whose pinpoint focus may be technical vascular gymnastics in the face of big picture stupidity. Because of the daytime injury leading often to nighttime surgery queues, too often the task of replantation is given to junior trainees who have to reinvent the wheel in the trenches at night without supervision. Anesthetists and nursing staff have to suffer interminable time wasting as the resident’s indecision lapses into mesmerizing Brownian movement. It should be remembered that secondary surgery on replanted digits is the most challenging of all hand surgery so that meticulous primary repair of all structures is the goal.

Normal Hand Function

The functional role of the digits can be looked at from the individual finger point of view and from their role as a partner in group hand functions, pinch grip, palmar grip, global pinch etc. Rarely do they function in isolation except perhaps in keyboard practice ( Fig. 50.1 ).

The radial three digits function as a unit to achieve the manipulative, dextrous function of pulp pinch or radial tripod or chuck grip, typically a dominant hand function. Sensation is preeminent. The pulps convey multiple sensory modalities that in concert allow the brain to interpret what the fingers are feeling, they are the eyes of the hand. This is tactile gnosis and is considerably more sophisticated than what is assessed by simple sensory two-point discrimination. Pinch between index and thumb uses a dominant radial side of the index whereas holding a pen or using fine instruments in the tripod fashion demands sensory acuity of the ulnar side of the index pulp and radial side of the middle. The basal joint is the key to allowing the mobile thumb ray to range variably across the digits to perform the range of pinch grip options from lateral (key) pinch against the stable proximal side of the index where the thumb is in full adduction and retropulsion, to pulp pinch and tripod pinch with index and long finger pulps where the thumb moves into maximum abduction and opposition. The pill rolling motion of the thumb pulp against the finger members is performed by circumduction (abduction, adduction, flexion, and extension) of the metacarpophalangeal joint. At the moment of final application of forced tripod pinch the proximal joints of the three rays must be stabilized by their intrinsic muscles to allow transmission of the flexor tendon force to flex the distal joints. Fortunately, intrinsics are protected in most digital amputations. By understanding the essential elements of this radial pinch grip according to each digit’s contribution we can better make decisions about the residual potential of the amputated part and the value or otherwise of replanting it.

It is estimated that the thumb, which has no substitute, contributes 40% of hand function and it should take priority for replantation. Provided the amputation does not involve the basal joint, and this is very rare, then even a relatively badly damaged thumb that is shortened and stiff can still perform most functions provided adequate sensation returns. In contrast, the index finger is readily substituted for by the middle if it has inferior function, especially sensory, and replantation can hinder residual hand function. Most of the unique function of the index resides beyond the distal joint (glabrous skin with finger printed sweat traction grip, padded stable pulp, multiple modality sensory nerve endings, and a finger nail acting as an exoskeleton to stabilize the pulp against its bony platform and to magnify sensation) so that with the exception of guillotine injuries proximal amputations of the index are likely to be numb and stiff because of tendon adhesions and joint damage and are arguably contraindicated for replantation.

The middle finger functions as a partner in the tripod pinch team but is also the most powerful palmar flexor because of its central position and the size of its tendons. Thus in combination with the ring and little fingers it also participates in the ulnar-sided palmar grip function. Here movement and length of digit rather than sensation are critical to function. All the joints must fold into full flexion for power grip and to prevent objects falling out of the open palm. The little finger is more valuable than the ring in this regard because it exploits full width of the palm and facilitates the stable and secure gripping of long-handled brooms, axes, etc. The torque power is augmented by the hypothenar muscles and the mobile carpometacarpal joints of these two digits.

Replantation of the middle finger proximal to the proximal interphalangeal (PIP) joint in isolation has considerable merit because its sensory need is relative, and even with limited mobility it will contribute to tripod pinch while its mere presence prevents small objects, coins, etc. falling through the cupped hand. A truly successful replant will serve both functions. The isolated ring finger replantation can also contribute usefully for the same reasons even if the result is mediocre although it is possible that flexor tendon adhesions may also limit adjacent finger tendon gliding and produce the quadriga effect. Little finger replants have theoretical logic and limits for the reasons mentioned but for some reason regain less function.

Single digit replantations distal to superficialis insertion were previously thought to be too difficult and not cost-efficient, but with greater experience they have been shown to be technically straightforward, and if nerves are repairable, afford a good outcome even if the distal joint is stiff. Likewise, fingertip amputations through the distal phalanx, although technically more challenging, can often be replanted by joining an artery and vein if possible and give a meaningful sensory return. They are the ultimate local flap, unlike some local flap options which add insult to injury. Terminalization often necessitates further bone shortening which may convert a usable into a nonusable digit. Here replantation may be the better option.

Where multiple digits are injured and/or amputated then the disruption to both radial and ulnar hand functions can be dire and every attempt to replant as many digits as is feasible is called for on the basis that something, even if compromised, is better than nothing. The best replantable digits should be strategically targeted to the most important sites so that elements of both radial pinch and ulnar grip can be restored. In severe mutilations, digits may be transposed for maximal function but when the index or little fingers are not replanted their metacarpal head stumps should not be resected for cosmetic reasons because torque (rotational) grip is reduced by more than 50%. Spare parts from nonreplantable digits can be used to salvage other damaged digits and in extreme cases parts can be banked in anticipation of later use, for example large pieces of innervated glabrous volar digit or palmar flap skin. An isolated index amputation in association with a middle finger pulp loss or divided radial digital nerve now becomes a strong indication for replantation. A compound PIP joint injury to a viable ring finger that will result in stiffness strengthens the indication for replanting the otherwise debatable adjacent little finger. The ulnar digits institute the power grip by fully flexing their terminal phalanges into the palm. Whereas a single proximal ulnar digit replantation risks restricting movement of the other, distal ulnar digit, replantation, which will not interfere with proximal tendon gliding, would restore full length of the digit to accomplish palmar closure even if the distal interphalangeal joint is stiff. Similarly, replantation of a single distal ring or little finger may be thought to be a trivial indication but if both are amputated then the critical locking of objects into the palm is lost and replantation of at least one becomes strongly indicated.

A rudimentary listing of indications to justify the resources to patient, surgeon, and health system would look something like: the thumb, single digits distal to superficialis insertion, multiple digits, and any pediatric digit. With the only reservations being single digits proximal to the PIP joint, which because of the circumferential injury to extensor and flexor tendon systems inevitably risk unyielding stiffness and reduced sensitivity requiring prolonged costly therapy, secondary surgery, and extended time from work. In reality, when confronted with a patient bringing with them their amputated part and immense hope, an honest conversation regarding the realistic likelihood of a functioning, noninterfering digit is paramount but not likely heard.

Contraindications

Contraindications apply to the digit and the patient.

Technique and Principles

Replantation requires disciplined leadership and continued momentum, which quickly wanes in the early hours of the morning. There needs to be communication with anesthetics regarding thromboembolism prophylaxis, a warmed theatre to prevent vasospasm, adequate warming and hydration to maximize peripheral vasodilatation, appropriate antibiotics (usually cephazolin with the occasional addition of Gram-negative and aerobic cover for agricultural mechanisms), pressure area prevention, and an indwelling catheter.

The techniques of digital replantation have been described in innumerable texts, and personal preferences determine the surgeon’s specific routine. Several principles, however, should guide the procedure involving identification and meticulous primary repair of each component of the severed digit.

• 1.

  Supervision by a person with hand experience. Upper limb replantation is hand surgery, not microsurgery. The indication for surgery should not just be technical feasibility but the estimated functional potential, which requires experienced intraoperative assessment.

• 2.

  Primary repair of every structure. Revisional surgery in replants is exceptionally difficult, hazardous, and often unrewarding. Take time and care to perfect each step. Early postoperative mobilization is the key to achieving maximum functional recovery, and this demands skeletal stability and primary wound healing. Where tendon damage demands a staged reconstruction, it is probably wiser despite infection risk to insert a silicone rod at the primary operation to avoid secondary reopening of the digit at the replantation level.

• 3.

  Predetermined routine. Random repair of one structure and then another and back to the first wastes time and confuses assistants. Two teams are ideal, especially if multiple digits are being repaired. Replantation is a time and motion study.

• 4.

  Debridement rarely needs to be radical, especially of the skin. Infection is rare and primary skin closure is the ideal without grafts or flaps.

• 5.

  Skin access incisions should ideally be Bruner type. Avoid midlateral, so that if the wound is too tight to close the neurovascular repairs will be flap covered. Bruner incisions can be extended safely, as is often required for avulsion injuries where vascular resection may be extensive. Precious time is wasted because of inadequate exposure to the proximal vascular bundle, so avoid hesitation in gaining greater access proximally.

• 6.

  Tag vessels, nerves, and tendons in the amputated segment and at the proximal stump. The proximal exploration is best under tourniquet control until arterial assessment, when it should be released to assess for pulsatile flow.

• 7.

  Bone shortening facilitates primary skin closure. It also allows judicious debridement of damaged nerve to enable tension-free end-to-end coaptation. Bone shortening is not to avoid vein grafting the artery. The arterial debridement should not be compromised by concerns with end-to-end repair but determined by pulsatile forward flow from the proximal vessel, and this may require aggressive resection proximal to the next branch point. Aim to shorten bone from the amputated end because if the replant fails, at least the stump length is preserved.

• 8.

  Rigid bone fixation is key to early active movement of the replant. Care should be taken to perfect this step as it lays the foundation both intraoperatively and postoperatively. ^, A dorsal meshed plate is ideal where possible, leaving the adjacent joints free. Unfortunately, abandoning these ideals for the ease and speed of Kirschner wire (K-wire) fixation is standard; K-wire fixation also requires minimal periosteal dissection and allows flexibility in rotational correction. K-wire can risk infection, tethers joints and tendons, and creates movement at the fracture site. In a Korean study of 1247 successful replantations, single axial K-wire was not associated with poorer outcomes compared to rigid fixation.

• 9.

  Periosteal repair where possible facilitates tendon gliding, especially dorsally beneath the broad expanse of the extensor tendon complex at middle and proximal phalangeal levels. Bone shortening often enables periosteal repair.

• 10.

  Dorsal structures repaired first – periosteum, extensor tendon including the intrinsics, veins, and skin.Venous repair before arterial avoids unnecessary blood loss from open veins or the need to clamp the artery post repair or maintain tourniquet. Veins tend to rupture rather than avulse, and for that reason rarely require vein grafts. If bony fixation is correctly aligned, dorsal veins should line up directly. K-wire fixation allows slight rotational correction if required. Subcutaneous bruises and blood clot provide clues to their location, also applying circumferential pressure to the cut edge should encourage them into vision. Lastly, if these methods fail, an arterial repair can be performed first to force them open. At least two veins must be repaired. With very distal replants the veins will be necessarily volar, most likely in the 4 and 8 o’clock regions, but anastomosable vessels may be absent. When no suitable veins exist, temporary venous outflow by leech therapy, subcutaneous heparin, topical heparin to the sterile matrix or subdermal pocketing of the de-epithelialized pulp into palm or chest will help. Venous anastomosis improves the overall success rate in distal replantation by 15%. However, substantial evidence suggests it carries no survival benefit distal to the lunula. Even when venous anastomosis proves impossible, published success rates over 75% justify attempts at replantation in distal amputation. Following venous repair, the skin is loosely approximated avoiding kinking of the veins. Epithelial only sutures will avoid distortion of the tissues.

• 11.

  Volar repair begins at the periosteum, flexor tendons, arteries, nerves and skin. The periosteum is again repaired where possible to reduce adhesions. The tendons should be repaired before the vessels and nerves to restore the preinjury elastic tension-length of the volar digit. This then determines the tension-length of the vessel repair and is especially important in assessing the length of interpositional vein grafts, which will elongate once flow is restored in the artery. Judgment will guide whether one or both flexors are repaired.

  • a.

    Arterial repair under tourniquet control allows the vessels to be identified in their anatomical sites. The ulnar artery of the thumb is dominant and must be selected. Access to its proximal reaches is difficult and requires appropriate assistance and positioning. The radial vessels of the index and little fingers are usually dominant. Relevant to distal amputations, just distal to the level of the DIP joint the digital arteries merge from their lateral location into one central vessel which passes deep along the periosteal surface of the distal phalanx. Once located the vessel ends are inspected for signs of intimal tears and medial splits, bruising or delamination of the wall or intraluminal thrombus. Distal damage is challenging to judge beyond gross retraction, coiling or delamination. Once debridement is done the tourniquet is released to evaluate pulsatile flow from the proximal artery. If reperfusion fails, start investigating proximally for the underlying cause. If the patency test demonstrates no flow up to the anastomosis, then the explanation is spasm, proximal vessel damage or central causes. Aggressive resection of the proximal vessel and grafting may be required. If the flow goes up to but not across the anastomosis, then the diagnosis is thrombus and requires reanastomosis or grafting. In both of these situations, a stat dose of IV heparin is helpful before application of the vessel clamps. If the flow passes across the anastomosis, but the finger does not perfuse then the explanation is distal or intrinsic microvascular damage due to crush, spasm, or prolonged ischemia causing reperfusion injury. This problem may respond to Trinitrin patches, intravenous heparin, dependent positioning of the hand to increase arterial inflow pressure or thrombolytics such as urokinase. Beyond these, fate will determine the outcome.

  • b.

    Nerve repair; if a nerve graft is required it should be done at the primary operation. Use the posterior interosseous or medial antebrachial cutaneous nerve of the forearm or, if another digit is to be discarded, then nerve from this digit is ideal. However, where possible trade skeletal shortening for direct nerve coaptation. The neural recovery is vital for meaningful function of radial-sided digits.

  • c.

    Skin closure is the key to early healing and pos-operative mobilization. It should be as loose as possible with the minimum number of stitches in the knowledge that the tissues will swell in the initial postoperative period. Where the skin cannot be closed thin fenestrated grafts for small defects will take, even over vascular repairs, but this compromises the use of heparin. More substantial defects require flap coverage, especially if tendon or bone is exposed. Thumb or mutilating hand injuries may require regional flaps or distant free flaps including venous flow-through flaps.

Multiple Digital Replantation

Multiple digit loss is an absolute indication for replantation, despite sometimes limited expectations and increased surgical time and resources. In the mutilated hand a little can mean a lot and be the critical difference between employability or functional independence.

Replanting all digits in their intended positions is not always possible or desirable, and decision-making depends on the quality of amputated parts and stumps. When amputated parts are not all replantable, the guiding principle is thumb replantation is prized above all; the dominant hand should take priority and strive to restore primary hand function; either radial pinch or ulnar power grip. Theoretically, the dominant hand should favor radial pinch grip for dextrous work and the nondominant hand ulnar grip for power to stabilize objects while being manipulated by the dominant hand. Transposing ulnar digits to a radial position may mean the dominant digital arteries are reversed so that anastomosis may require vein grafting across the flexor sheath ( Table 50.1 ).

The operative sequence should be cognizant of time and motion. Addressing one component of all the digits, i.e., all bone fixation, then all dorsal repairs and finally volar structures, is time-efficient rather than doing one complete finger at a time. It means the bones and tendons are stabilized before microsurgical repair and has the advantage of reduced risk of accidental injury to delicate adjacent microsurgical repair and multiple applications of the tourniquet. This method is likely to lead to higher survival rates. The distinct disadvantage is prolonged warm ischemia, alleviated by keeping the amputated parts in cold storage until required.

Ring Avulsion Injury

Ring avulsions can be the most challenging injuries for hand surgeons to treat, and the decision to attempt salvage for high-grade injuries remains difficult.

Kay and Adani have expanded Urbaniak’s original classification, shown in Table 50.2 . Although classifications do not seem to add anything to what is obvious, assess avulsions according to the layer(s) involved from superficial to deep.

Skin breach alone with dorsal vein disruption needs venous repair only and has an excellent prognosis.

Deeper more severe degloving without amputation will involve vascular bundle damage, frequently with extensive proximal arterial and nerve injury. The skin is sleeved distally to a variable degree in which case the veins are also divided but not always depending on the angle of incidence of the ring injury; these require arterial vein grafting but have a good prognosis.

The next stage involves skeletal fracture/dislocation at the DIP joint or neck of the middle phalanx distal to the superficialis tendon insertion leaving the digit attached by the intact flexor profundus. Here the extensor tendon is destroyed, the joint damaged, nerves variably avulsed, and the skin sleeve on the degloved segment devascularized over a considerable distance and relies on retrograde fill following arterial reconstitution. Venous flaps may provide a vascularized paddle of skin as well as a segment of a vein to bridge arterial or venous defects. It is evident that the functional potential of these injuries is restricted even if they can be revascularized.

The final stage is complete avulsion/amputation with proximal disruption of the flexor profundus tendon. In the process of disruption, the muscle belly with the tendon is destroyed as it is drawn through the carpal tunnel and presents as a delayed carpal tunnel syndrome. Attempted salvage of these injuries is mainly for cosmetic reasons.

Postoperative Care

Keep the patient warm, hydrated, and pain-free with the limb elevated to avoid vasospasm. Antibiotics are given intravenously for a surgeon-specified period.