CURRENT CONCEPTS
The Management of Digital Nerve Injuries David J. Slutsky, MD
A tension-free coaptation is a key factor for the successful outcome of any nerve repair. A variety of host factors influence the outcome of digital nerve repair more than the type of repair per se. Although autologous graft remains the reference standard for reconstruction of any critical digital nerve defect, allografts and conduits have assumed an increasing role. (J Hand Surg Am. 2014;39(6):1208e1215. Copyright Ó 2014 by the American Society for Surgery of the Hand. All rights reserved.) Key words Nerve, digital, repair, graft, conduits.
N
are composed of 3 terminal branches from either the radial, median, or ulnar nerve. Distal to the transverse carpal ligament, the median nerve splits into lateral and medial branches. The lateral branch supplies a motor branch to the thenar muscles and then divides into 3 proper palmar digital nerves. The common digital nerve to the thumb splits into a radial and ulnar digital branch. The second common digital nerve supplies a motor branch to the index lumbrical and supplies sensation to the radial side of the index. In the distal palm, the medial branch of the median nerve divides into 2 common palmar digital nerves. The first innervates the second lumbrical and runs toward the index and middle fingers, where it divides into 2 proper digital nerves. The second common palmar digital nerve runs toward the middle and ring fingers and splits into 2 proper digital nerves. Each proper digital nerve in the finger gives off a dorsal branch at or proximal to the A1 pulley in the fingers that joins the dorsal digital nerve, a terminal segment of the superficial branch of the radial nerve. The dorsal branch from the proper digital nerve perforates the connective tissue septum (Cleland’s ligament) and anastomoses with the dorsal digital nerve. This ERVES TO THE DIGITS
Current Concepts
From the Hand and Wrist Institute, Torrance, CA. Received for publication November 1, 2013; accepted in revised form December 6, 2013. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: David J. Slutsky, MD, Hand and Wrist Institute, 2808 Columbia Street, Torrance, CA 90503; e-mail: [email protected]. 0363-5023/14/3906-0035$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.12.012
1208
r
Ó 2014 ASSH
r
Published by Elsevier, Inc. All rights reserved.
nerve innervates the skin on the dorsal aspect of the last phalanx. At the terminal aspect of each digit, the proper digital nerve divides again into 2 branches; 1 supplies the pulp and the other splits around and underneath the nail. As they run along the lateral aspect of each finger, the proper digital nerves are superficial to the corresponding arteries. There are no connections between the dorsal sensory nerve and the dorsal branch of the proper digital nerve in the thumb. There is great variation in the course that the proper digital nerve takes to the tip of the finger. Commonly, it passes between the superficial and deep transverse metacarpal ligaments. As it moves distally, it lies palmar to the adjacent artery at the side of the fibrous flexor sheath. At the distal digital crease, it goes on to divide into multiple branches that terminate at the pulp and nailbed. There is no crossover innervation of the pulp. Anatomical studies have shown that there is little change in diameter of the nerve as it makes its way distally. The ulnar nerve bifurcates approximately 5 cm proximal to the wrist into the dorsal branch and volar branch. The dorsal branch of the ulnar nerve passes beneath the flexor carpi ulnaris, perforates the deep fascia, and runs along the ulnar side of the back of the hand. As it proceeds through the midhand, it divides into 2 dorsal digital branches. One branch supplies sensation to the dorso-ulnar side of the little finger, and the other to the adjacent side of the little and ring fingers. It also sends a small branch to join with the superficial branch of the radial nerve for the adjoining sides of the middle and ring fingers. The dorsal digital branch extends to the base of the distal phalanx in the little fingers and to the base of the second phalanx in
DIGITAL NERVE INJURIES
the ring fingers. The distal aspects are supplied by the dorsal branches from the proper volar digital branch of the ulnar nerve. The volar branch of the ulnar nerve crosses the transverse carpal ligament on the lateral side of the pisiformis and ends by dividing into a superficial and a deep branch. The superficial branch supplies the skin on the ulnar side of the hand by dividing into a proper volar digital branch for innervation of the ulnar side of the little finger. In addition, a common volar digital branch is given off that divides into 2 proper digital nerve innervating the adjoining side of the little and ring fingers volarly. Nerve regeneration does not involve mitosis and multiplication of nerve cells. Instead, the cell body restores nerve continuity by growing a new axon. One axon sends out multiple unmyelinated axon sprouts from the tip of the remaining axon or collateral sprouts from a nearby proximal node of Ranvier. The distal sprout contains the growth cone. This sends out filopodia, which adhere to sticky glycoprotein molecules in the basal lamina of Schwann cells, such as laminin and fibronectin (neurite-promoting factors). The filopodia contain actin, which aids in pulling the growth cone distally. The basal lamina of 2 abutting Schwann cells form a potential endoneurial tube into which the regenerating axon grows. These axons will deteriorate if a connection with a target organ is not reached. Unlike the motor end plate, sensory end organs remain viable because there is no end plate and they retain the potential for reinnervation. Reconstruction of a sensory nerve defect, by comparison, may provide protective sensation even after many years.1 A normal nerve has longitudinal excursion, which subjects it to a certain amount of stress and strain in situ. For example, the median nerve moves as much as 12.4 mm with wrist motion at the carpal canal.2 A peripheral nerve is initially easily extensible, but this rapidly diminishes with further elongation owing to stretching of the connective tissue within the nerve. The perineurium is a mechanically strong membrane and is a major load-carrying connective tissue component that can sustain intrafascicular pressure elevations of up to 750 mm before rupturing.3 The epineurium is a loose connective tissue layer that allows a certain amount of nerve gliding. Chronically injured nerves become even stiffer. Elasticity decreases by as much as 50% in the delayed repair of nerves in which Wallerian degeneration has occurred. Experimentally, blood flow is reduced by 50% when the nerve is stretched 8% beyond its in vivo length. Complete ischemia occurs at 15% of nerve elongation. Suture pullout from the neurorrhaphy site does not occur until a 17% increase in length.4 This suggests J Hand Surg Am.
1209
CLINICAL PICTURE The patient typically presents with a history of a laceration to the palm or digit by glass or a knife, although puncture wounds caused by a drill bit can cause more extensive damage. A common digital nerve injury to the thumb will result in a sensory defect along the radial and ulnar aspects of the distal thumb. The patient with a common digital nerve injury in the palm will present with a sensory deficit along the adjacent sides of the second or third webspace (median nerve) or the fourth webspace (ulnar nerve). A proper digital nerve injury will result in an isolated sensory deficit along the radial or ulnar aspect of the thumb or digit. Static 2-point discrimination (S2PD) and moving 2PD (M2PD) testing will be greater than 25 mm on the affected side(s, which is equivalent to a complete sensory loss. Normal S2PD is 6 mm or less, whereas 15 mm is equivalent to a loss of protective sensation. Semmes Weinstein monofilament (SWM) testing will be greater than 6.65. This can easily be tested in the emergency room or office setting using the West Enhanced Sensory Test, which consists of a handle with 5 monofilaments. Measuring individual digital nerve action potentials can aid in the diagnosis of isolated digital nerve injuries (Fig. 1 A, B).5 Errors owing to volume conduction from an intact digital nerve on the opposite side must be watched for. Occasionally, it is necessary to perform a digital nerve block of the unaffected side to prevent contamination of the response in the nerve under consideration. Lacerations proximal to the takeoff of the dorsal sensory branch in the digits will also result in a sensory deficit r
Vol. 39, June 2014
Current Concepts
that ischemia, and not disruption of the neurorrhaphy, is the limiting factor in acute nerve repairs.4 This observation is also applicable to nerve grafting. There is a difference between a nerve gap and a nerve defect. A nerve gap refers to the distance between the nerve ends resulting from the elastic retraction that occurs immediately after nerve severance, whereas a nerve defect refers to the actual amount of nerve tissue that is lost. With simple nerve retraction after division, the fascicular arrangement is similar. As the nerve defect between the proximal and distal stumps increases, there is a greater fascicular mismatch between the stumps. This is of minimal consequence with digital nerve repairs because all of the fascicles contain only sensory fibers, but fascicular mismatch can still lead to an alteration of the sensory distribution to the fingertips in common digital nerve repairs.
1210
DIGITAL NERVE INJURIES
FIGURE 1: Digital nerve conduction studies. A Intact sensory nerve action potential to the radial side of the index finger (top tracing) but absent potential to the radial side of the middle finger. B Partial common digital nerve laceration to the third webspace. (Ó 2013 David J. Slutsky, MD.)
fibers, matching is not necessary. Usually 3 to 4 sutures with 8-0 or 9-0 nylon are sufficient. Alternatively, fibrin glue can be used, which allows the placement of fewer sutures. After nerve repair, rehabilitation focuses on 3 areas: initial immobilization to protect the repair, joint mobilization to promote longitudinal excursion of the nerve, and sensory reeducation. Lundborg and Rosen7 demonstrated that training protocols specifically addressing the relearning process substantially increase the possibilities for improved functional outcome after nerve repair. Before wound closure, the adjacent joints are placed in various degrees of flexion and extension, to determine the optimum limb position that unloads the repair site. This position is maintained with a blocking splint for 3 weeks, but a protected short arc of motion may be instituted to provide some nerve gliding. Yu et al8 challenged this notion and demonstrated no significant differences in 16 digits after an isolated digital nerve repair treated with 21 days of immobilization versus 14 digits with combined nerve and flexor tendon repair were treated with early protected mobilization.
over the dorsal aspect of the middle phalanx. There may be an associated flexor tendon or digital artery injury, which may affect the rehabilitation and outcome. Although uncommon, a neurapraxia rather than a nerve laceration may occur after a sharp injury, and may recover spontaneously if treated initially with observation. It is my preference to delay surgery no longer than 3 weeks, however, because the likelihood of the need for a nerve graft increases after 1 month.
Current Concepts
SURGICAL MANAGEMENT Primary nerve repair A tension-free repair is the goal for any nerve neurorrhaphy. When there is a clean transection of the nerve and the gap is caused by elastic retraction, an acute primary repair is indicated. The digital nerves are approached though a midlateral or a volar Brunner incision. If the extent of injury cannot be determined, nerve repair should be delayed. There are often 2 large groups of fascicles macroscopically in the typical digital nerve. In a cadaver dissection of 20 hands, Bonnel et al6 found that histologically, the number of fascicles ranged from 16 to 18 proximally, which increased to 21 to 28 fascicles in the distal pulp. Similarly, the amount of connective tissue relative to the fascicles increases from 70% of the cross-sectional area proximally to 90% distally. An external epineurial repair gives the same results as a fascicular repair. Because they contain only sensory J Hand Surg Am.
Nerve graft When the treatment of a nerve laceration is delayed, fibrosis of the nerve ends prevents approximation of the severed nerve ends; hence, nerve grafting is required even though there is no loss of nerve tissue. One should always be prepared to proceed with a r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
nerve graft or conduit if a nerve repair would lead to undue tension at the coaptation site. Many authors recommend nerve grafting when the gap exceeds 1 cm, with the wrist and all 3 finger joints extended. Because the graft is vascularized from the tissue bed, nerve grafting should not be performed in burned or irradiated tissue. The nerve graft acts to provide a source of empty endoneurial tubes through which the regenerating axons can be directed. Any tissue that contains a basal lamina such as freeze-dried muscle or tendon can be substituted but only the autologous nerve graft also provides a source of viable Schwann cells. A normal nerve can compensate for the change in length with limb flexion and extension because it is surrounded by gliding tissue that permits longitudinal movement. A nerve graft becomes welded to its recipient bed by the adhesions through which it becomes vascularized. As a consequence, the nerve graft is susceptible to tension because it has no longitudinal excursion. The harvested graft must be long enough to span the nerve gap without tension while the adjacent joints are extended. This should also be the position of temporary postoperative immobilization. If the digit is immobilized in flexion, the graft will become fixed in this position. When the limb is then mobilized at 8 days, the proximal and distal stumps will be subject to tension even though the graft was initially long enough. Early attempts at lengthening the graft will lead to disruption of the neurorrhaphy nerve repair.
Outcomes after repair or graft Most series report the results of nerve repair using the British Medical Research Council grading system, which was modified by Dellon.10 In this classification, a score of S0 represents absent sensory function, whereas a score of S1 or S2 represents detectable but less useful function. S3 equates to recovery of pain and touch sensibility, with 2PD of greater than 15 mm, whereas S3þ equates to 2PD of 7 to 15 mm, and S4 is normal sensation. Wang et al11 evaluated 67 adults with 74 primary digital nerve repairs and reported a recovery of 2SPD of 8 mm or less in 80% of the patients. Of these, 26 of 29 patients with a clean-cut primary repair had M2PD less than 7 mm (median, 3 mm) and 18 had S2PD less than 7 mm. Of 42 nerves with a crush or saw-type injury, 37 achieved a median of 6 mm M2PD and a median S2PD of 8 mm. Digits with mild crush or saw injuries had significantly reduced sensibility as measured by an increased 2 PD (P ¼ .02) as compared to digits with simple nerve lacerations. Patients under 40 years of age had better 2PD than those over 40 years. Kallio12 studied 95 patients with 254 completely divided digital nerves at an average of 12.4 years (range, 5e20 y) after repair. A delayed epineurial suture was used in 53, fascicular grafting in 37, and fascicular repair in 5. Useful sensory function was recovered in 79% of the nerves operated on with epineurial or perineurial suture and in 56% with fascicular grafting. Weinzweig et al13 examined the prognostic factors in a multicenter retrospective review of 172 epineural digital nerve repairs in 96 patients, with a mean age of 33.3 years (range, 5e64 y). They found the mechanism of injury and age to be predictors of sensory recovery after digital nerve injury and repair. The follow-up averaged 22.2 months (range, 6e77 mo). The authors found a statistically significant correlation between age and recovery of S2PD, with r
Vol. 39, June 2014
Current Concepts
FIGURE 2: Digital nerve graft after saw injury. Terminal posterior interosseous nerve graft interposed between the 2 fascicles of the terminal digital nerve. (Ó 2013 David J. Slutsky, MD.)
Options for nerve graft donors Higgins et al9 investigated the fascicular crosssectional area and number of fascicles of 5 nerve graft sites to specific digital nerve segments. In the fingertip distal to the distal interphalangeal joint, the anterior interosseous nerve, posterior interosseous nerve, and medial antebrachial cutaneous nerve were all appropriate choices for caliber-matched grafts (Fig. 2). The lateral antebrachial cutaneous nerve, however, was the only similar donor nerve when the number of fascicles was assessed. The lateral antebrachial cutaneous nerve is also the best match in caliber and fascicle number for digital nerve deficits between the metacarpophalangeal joint and the distal interphalangeal joint, and from the common digital nerve bifurcation to the metacarpophalangeal joint (Fig. 3 A, B). The sural nerve is the most appropriate choice when grafting defects between the wrist and the common digital nerve bifurcation, even though there are considerably fewer fascicles and a smaller cross-sectional area than the common digital nerve. J Hand Surg Am.
1211
1212
DIGITAL NERVE INJURIES
FIGURE 3: Lateral antebrachial cutaneous nerve graft. A Bifurcation of lateral antebrachial cutaneous nerve in the proximal radial forearm (arrow). B Bifurcated graft used for reconstruction of the third common digital nerve (red arrow). Lateral antebrachial cutaneous nerve graft to the small finger (white arrow) and primary nerve repair to the index finger (black arrow). (Ó 2013 David J. Slutsky, MD.)
Current Concepts
patients older than 40 years demonstrating significantly poorer recovery of sensibility than patients younger than 40 years (P < .001). There was also a significant correlation between severity of injury and recovery of sensibility (P < .001). Sharp injuries (8.2 mm) demonstrated significantly improved recovery of sensibility compared with mild crush (10.8 mm). Fractures (11.1 mm), fractures and tendon involvement (11.4 mm), and replantations (11.8 mm) demonstrated significantly diminished recovery of sensibility compared with isolated nerve injuries (7.9 mm), tendon involvement (8.1 mm), and revascularizations (9.3 mm). There appeared to be no significant difference in recovery of sensibility according to gender, digit involved, radial or ulnar side of digit, median or ulnar nerve distribution, level of injury, and time interval from injury to repair. Chen et al14 examined 17 patients with an average age of 32 years, with 21 nerve defects of 2.3 cm (range, 1.4e3.5 cm) who were treated with a nerve transfer from the dorsal branch of the proper digital nerve (PDN). A neurorrhaphy was performed between the uninjured dorsal branch of the same digit or the dorsal branch of the adjacent digit and the distal end of the PDN. They compared the results with those of a control group of series of 31 patients with PDN defects treated with conventional sural nerve grafting. At a mean follow-up of 25 months (range, 20e26 mo), the mean S2PD was 6.4 mm and the mean SWM score was 3.6. In the comparison group at a mean follow-up of 23 months (range, 19e27 mo), the mean SPD was 9.2 mm and the mean SWM score was 4.10.
regeneration. Neurotropism describes the directional accuracy of that regeneration. The growth cone is attracted to neurotropic proteins that are derived from the distal degenerating nerve segment after nerve transection. These phenomena have been exploited through the use of synthetic or natural conduits to bridge the nerve gap. Histologically, fibrin clot develops inside the tube within hours. Within the first week, longitudinally oriented fibrin matrix bridges span the nerve gap. In the second week, fibroblasts, Schwann cells, macrophages, and endothelial cells permeate the matrix. Axon sprouts from proximal nerve reach the distal stump and become myelinated by the 4th week. The axons elongate down the distal endoneurial tubes and reinnervate the target organs. Silicone nondegradable tubes were initially used experimentally. More recently, natural conduits have included freezeethawed muscle (which contains a basal lamina) and autologous vein grafts. Biodegradable synthetic material has been used, such as polyglycolic acid (PGA), caprolactone, and collagen. Acellular allografts have also been used. The postoperative rehabilitation includes early mobilization of the part, because tension on the repair site is not a consideration, followed by appropriate sensory retraining. Conduits are indicated for reconstruction of smalldiameter, noncritical sensory nerves with a gap of less than 3 cm. They can be used with digital replantation with nerve gaps of less than 3 cm and in the mangled hand if a primary, tension-free nerve repair cannot be achieved in the acute setting and if the gap is less than 1 to 2 cm. They are indicated in association with flexor tendon repair in zone II to create a tension-free repair and allow early mobilization. Relative indications
NERVE CONDUITS Neurotrophism is the ability of chemotactic hormonal or growth factors to enhance the rate of nerve J Hand Surg Am.
r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
include nerve reconstruction after resection of a traumatic neuroma or a neoplasm. They can be used as a wrap around a partially injured nerve or after neurolysis of a scarred nerve. Contraindications include uncertainty about the viability of the nerve ends, especially with avulsion injuries, blast injuries and gunshot wounds, local or systemic infection, and inadequate soft tissue coverage. When placed in superficial locations, the nerve conduit may be palpable for up to 1 year. Polyglycolic acid conduits are more likely to extrude than collagen conduits owing to their more rigid nature, and require thicker soft tissue coverage. Relative contraindications include nerve gaps greater than 3 cm.
FIGURE 4: Collagen conduit (forceps) bridging a gap in a radial digital nerve deficit of the index finger. (Ó 2013 David J. Slutsky, MD.)
Synthetic conduits The procedure is done with the patient supine and the extremity on an arm board under tourniquet control. Microsurgical instruments are needed; high-power loupe magnification is usually sufficient, although a fascicular dissection may be performed under the operating microscope. The proximal and distal nerve ends are identified and meticulous hemostasis is obtained. The tension-free gap is measured with the adjacent joints in full extension. The nerve stumps are trimmed under magnification until healthy fascicles are identified. The minimum distance between the ends is 5 mm, with a maximum of 3 cm. The diameter of the nerve is measured with a ruler to determine the appropriate size conduit. The inner diameter of the tube should be equal to or slightly larger than the diameter of the nerve to be implanted. Some synthetic conduits require saline immersion before nerve implantation. Conduit sizes range from 2 to 10 mm and 4 cm in length. After the appropriate size is chosen, 1 nerve end is pulled into the conduit so that 2 to 5 mm of the nerve lies within the conduit, or a distance equal to or greater than the nerve diameter. An 8e0 or 9e0 nylon suture is used to draw the nerve ends into the conduit using a horizontal mattress stitch (Fig. 4). Either the proximal or distal end is sutured first. The stitch is passed from the outside to inside of the conduit wall at least 1 mm from its end. The suture is next passed transversely through the epineurium 2 to 3 mm from the stump, and then back through the conduit from inside to out, and tied after the stump is drawn into the conduit. The conduit lumen is filled with sterile saline for collagen tubes and heparinized saline for PGA tubes, as per the manufacturers’ recommendation, to prevent air bubbles or blood clot formation, which may impede axonal ingrowth. J Hand Surg Am.
1213
POSTOPERATIVE REHABILITATION Repair of associated tendon or muscle injuries will influence the rehabilitation. With an isolated nerve reconstruction, the affected part is immobilized with the adjacent joints in neutral position. Composite extension or flexion in the opposite direction of the repair is avoided to prevent tension that could dislocate a nerve end from the conduit. Range of motion exercises are started at 10 days, followed r
Vol. 39, June 2014
Current Concepts
Autologous vein grafts The use of autologous veins as a biologic tube to reconstruct a nerve gap has been well described over the past century. Chiu and Strauch15 championed the use of autologous venous nerve conduits (AVNC) for the reconstruction of digital sensory nerve defects of 3 cm or less. They demonstrated clinical results that were more favorable compared with autologous nerve grafts. Advantages of the AVNC include the avoidance of a cutaneous donor site sensory deficit and minimal cost compared with the commercially manufactured synthetic conduits. The superficial veins on the dorsum of the hand are of an appropriate caliber for digital nerve reconstruction (Fig. 5 A, B). The authors noted that there is a fairly constant dorsal vein that runs longitudinally in line with the fourth webspace and is usually adjacent to the dorsal cutaneous branch of the ulnar nerve. The AVNC is reversed because of the possible presence of venous valves. The cut proximal end of the nerve is placed in the end of the vein that was distal (in situ) using a telescoping suture. The same is done to the other end of the nerve after the vein graft is cut to the appropriate length. Some authors prefer to use only lower-extremity veins because they are thicker and have less of a tendency to collapse.
1214
DIGITAL NERVE INJURIES
FIGURE 5: Vein conduit. A Harvest of autologous dorsal vein conduit. B Reversed vein graft prior reconstruction of 1-cm nerve defect (arrow). (Ó 2013 David J. Slutsky, MD.)
by strengthening exercises at 6 weeks and sensory reeducation for the next year.
results of using a caprolactone tube versus primary repair in 34 digital nerve lesions with gaps less than 2 cm. The results with the conduits were equal to those with repair, as evaluated by the pressure-specified sensory testing device and SP2D. Haug et al19 evaluated 35 patients with a mean age of 47 years (range, 11e83 y), with 45 digital nerve defects. The nerve defects ranged from 5 to 26 mm (mean, 12 mm) and were reconstructed with 2- to 3-mm-diameter collagen tubes with a mean length of 19 mm (range, 11e30 mm). They used a sum score that included a variety of modalities including mechanical stimuli, warm/cold sensation vibration, SWM, 2SPD, number recognition, and electroneurography. Basic sensory functions such as sharp/dull and warm/cold sensation as well as vibration showed faster recovery than more complex functions such as SWM sensation, S2PD, and number recognition. Recovery of sensory functions 12 months after nerve tube implantation, as measured by the sum score, was very good in 4 cases, good in 21, mediocre in 14, and poor in 3.
Current Concepts
OUTCOMES Chiu and Strauch15 conducted a clinical series of animal experiments bridging gaps of 1.0 to 6.0 cm using autologous vein grafts, and reported results comparable to those obtained with nerve repair and grafting for sensory nerve defects of less than 3 cm. Tang et al16 examined patients after 22 digital nerve repairs using autologous vein grafts. The S2PD averaged 4.6 mm for 11 acute digital nerve repairs using vein conduits 1 to 3 cm in length. Delayed digital nerve repair with vein conduits yielded poor results. Weber et al17 performed a randomized, prospective, multicenter evaluation of 98 patients with 136 nerve transections in the hand, comparing a PGA conduit with either end-to-end repair or a nerve graft. A total of 56 nerves repaired in the control group and 46 nerves repaired with a conduit were available for follow-up. The overall results showed no significant difference between the 2 groups as a whole. Nerves with gaps of 4 mm or less had better sensation when repaired with a conduit; the mean M2PD was 3.7 1.4 mm for the PGA tube repair and 6.1 3.3 mm for end-to-end repairs (P ¼ .03). All injured nerves with deficits of 8 mm or greater were reconstructed with either a nerve graft or a conduit. This subgroup also demonstrated a significant difference in favor of the PGA tube. The mean MP2D for the conduit was 6.8 3.8 mm, with excellent results obtained in 7 of 17 nerves, whereas the mean MP2D for the graft repair was 12.9 2.4 mm, with excellent results obtained in none of the 8 nerves (P < .001 and P ¼ .06, respectively). Bertleff et al18 examined the J Hand Surg Am.
Vein conduit with skeletal muscle Marcoccio and Vigasio20 treated 17 patients with 21 digital nerve defects using a vein conduit filled with fresh skeletal muscle. The donor vein and fresh muscle segments were obtained through a 5-cm skin incision in the proximal forearm. A suitable vein graft was harvested and a 22-gauge cannula was introduced into the lumen of the vein. A longitudinal segment of muscle fibers was harvested slightly longer than the vein graft. A 7e0 nylon suture was used to draw the muscle into the vein graft, and then removed. A tension-free repair was performed as described above. The average nerve gap was 2.2 cm (range, 1e3.5 cm). r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
1. Sunderland S. Capacity of reinnervated muscles to function efficiently after prolonged denervation. AMA Arch Neurol Psychiatry. 1950;64(6):755e771. 2. Coppieters MW, Alshami AM. Longitudinal excursion and strain in the median nerve during novel nerve gliding exercises for carpal tunnel syndrome. J Orthop Res. 2007;25(7):972e980. 3. Selander D, Sjostrand J. Longitudinal spread of intraneurally injected local anesthetics: an experimental study of the initial neural distribution following intraneural injections. Acta Anaesthesiol Scand. 1978;22(6):622e634. 4. Clark WL, Trumble TE, Swiontkowski MF, Tencer AF. Nerve tension and blood flow in a rat model of immediate and delayed repairs. J Hand Surg Am. 1992;17(4):677e687. 5. Terai Y, Senda M, Hashizume H, Nagashima H, Inoue H. Selective measurement of digital nerve conduction velocity. J Orthop Sci. 2001;6(2):123e127. 6. Bonnel F, Foucher G, Saint-Andre JM. Histologic structure of the palmar digital nerves of the hand and its application to nerve grafting. J Hand Surg Am. 1989;14(5):874e881. 7. Lundborg G, Rosen B. Hand function after nerve repair. Acta Physiol. 2007;189(2):207e217. 8. Yu RS, Catalano LW III, Barron OA, Johnson C, Glickel SZ. Limited, protected postsurgical motion does not affect the results of digital nerve repair. J Hand Surg Am. 2004;29(2):302e306. 9. Higgins JP, Fisher S, Serletti JM, Orlando GS. Assessment of nerve graft donor sites used for reconstruction of traumatic digital nerve defects. J Hand Surg Am. 2002;27(2):286e292. 10. Dellon AL. A numerical grading scale for peripheral nerve function. J Hand Ther. 1993;6(2):152e160. 11. Wang WZ, Crain GM, Baylis W, Tsai TM. Outcome of digital nerve injuries in adults. J Hand Surg Am. 1996;21(1):138e143. 12. Kallio PK. The results of secondary repair of 254 digital nerves. J Hand Surg Br. 1993;18(3):327e330. 13. Weinzweig N, Chin G, Mead M, et al. Recovery of sensibility after digital neurorrhaphy: a clinical investigation of prognostic factors. Ann Plast Surg. 2000;44(6):610e617. 14. Chen C, Tang P, Zhang X. Finger sensory reconstruction with transfer of the proper digital nerve dorsal branch. J Hand Surg Am. 2013;38(1):82e89. 15. Chiu DT, Strauch B. A prospective clinical evaluation of autogenous vein grafts used as a nerve conduit for distal sensory nerve defects of 3 cm or less. Plast Reconstr Surg. 1990;86(5):928e934. 16. Tang JB, Gu YQ, Song YS. Repair of digital nerve defect with autogenous vein graft during flexor tendon surgery in zone 2. J Hand Surg Br. 1993;18(4):449e453. 17. Weber RA, Breidenbach WC, Brown RE, Jabaley ME, Mass DP. A randomized prospective study of polyglycolic acid conduits for digital nerve reconstruction in humans. Plast Reconstr Surg. 2000;106(5):1036e1045; discussion 1046e1038. 18. Bertleff MJ, Meek MF, Nicolai JP. A prospective clinical evaluation of biodegradable neurolac nerve guides for sensory nerve repair in the hand. J Hand Surg Am. 2005;30(3):513e518. 19. Haug A, Bartels A, Kotas J, Kunesch E. Sensory recovery 1 year after bridging digital nerve defects with collagen tubes. J Hand Surg Am. 2013;38(1):90e97. 20. Marcoccio I, Vigasio A. Muscle-in-vein nerve guide for secondary reconstruction in digital nerve lesions. J Hand Surg Am. 2010;35(9): 1418e1426. 21. Taras JS, Amin N, Patel N, McCabe LA. Allograft reconstruction for digital nerve loss. J Hand Surg Am. 2013;38(10): 1965e1971. 22. Cho MS, Rinker BD, Weber RV, et al. Functional outcome following nerve repair in the upper extremity using processed nerve allograft. J Hand Surg Am. 2012;37(11):2340e2349.
Processed nerve allografts Processed nerve allografts (Avance Nerve Graft; AxoGen, Inc., Alachua, FL) consist of decellularized and predegenerated nerve tissue that is rapidly revascularized and repopulated with host Schwann cells. The immunogenicity of the graft is decreased by removing residual nerve debris and Schwann cells while preserving the normal internal neural architecture and important guidance molecules such as laminin. Taras et al21 reported the outcomes of 14 patients with an average age of 39 years (range, 18e76 y) with 18 digital nerve reconstruction using processed nerve allograft at an average of 15 months (range, 12e20 mo). The average nerve gap was 11 mm (range, 5e30 mm). Seven of 18 achieved S2PD of 6 mm or less, 8 achieved 7 mm, and 3 achieved 8 mm. Cho et al22 analyzed the data from a registry database of 35 subjects with a mean age of 46 14 years (range, 23e68 y). The time to surgery was 190 349 days and the follow-up was 306 184 days. The gap was 19 9 mm. There were 24 lacerations, 6 neuromas, and 5 complex lacerations. There was meaningful recovery of S3eS4 in 31 of 35 digital nerve repairs. The take-home messages are as follows: a tensionfree coaptation is a key factor in any nerve repair. Ischemia and not disruption of the neurorrhaphy is the limiting factor in acute nerve repairs. Grouped fascicular repairs do not appear to have any advantages over an epineurial repair. It is apparent that a variety of host factors influence the outcomes of digital nerve repairs more than the type of repair per se. Younger patients do better than older patients, sharp injuries do better than crush injuries, and isolated nerve injuries do better than those with associated fractures, tendon involvement. and revascularizations. Although autologous graft remains the reference standard for reconstruction of any critical digital nerve defect, the use of allografts and conduits is increasing in popularity.
r
Vol. 39, June 2014
Current Concepts
REFERENCES
At a mean follow-up of 43 months (range, 18e96 mo), 5 patients achieved S4 with complete recovery of S2PD of 2 to 6 mm and MP2D of 2 to 3 mm; 9 patients achieved S3þ with SPD of 7 to 15 mm and MP2D of 4 to 7 mm with but good localization of stimulus; and 3 patients achieved S3 with S2PD of greater than 15 mm, MP2D greater than 7 mm, and recovery of pain and touch sensibility. One patient had S2 with recovery of superficial pain and some touch sensibility, and 4 patients had S1 with recovery of superficial pain and only some touch sensibility.
J Hand Surg Am.
1215
The Management of Digital Nerve Injuries David J. Slutsky, MD
A tension-free coaptation is a key factor for the successful outcome of any nerve repair. A variety of host factors influence the outcome of digital nerve repair more than the type of repair per se. Although autologous graft remains the reference standard for reconstruction of any critical digital nerve defect, allografts and conduits have assumed an increasing role. (J Hand Surg Am. 2014;39(6):1208e1215. Copyright Ó 2014 by the American Society for Surgery of the Hand. All rights reserved.) Key words Nerve, digital, repair, graft, conduits.
N
are composed of 3 terminal branches from either the radial, median, or ulnar nerve. Distal to the transverse carpal ligament, the median nerve splits into lateral and medial branches. The lateral branch supplies a motor branch to the thenar muscles and then divides into 3 proper palmar digital nerves. The common digital nerve to the thumb splits into a radial and ulnar digital branch. The second common digital nerve supplies a motor branch to the index lumbrical and supplies sensation to the radial side of the index. In the distal palm, the medial branch of the median nerve divides into 2 common palmar digital nerves. The first innervates the second lumbrical and runs toward the index and middle fingers, where it divides into 2 proper digital nerves. The second common palmar digital nerve runs toward the middle and ring fingers and splits into 2 proper digital nerves. Each proper digital nerve in the finger gives off a dorsal branch at or proximal to the A1 pulley in the fingers that joins the dorsal digital nerve, a terminal segment of the superficial branch of the radial nerve. The dorsal branch from the proper digital nerve perforates the connective tissue septum (Cleland’s ligament) and anastomoses with the dorsal digital nerve. This ERVES TO THE DIGITS
Current Concepts
From the Hand and Wrist Institute, Torrance, CA. Received for publication November 1, 2013; accepted in revised form December 6, 2013. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: David J. Slutsky, MD, Hand and Wrist Institute, 2808 Columbia Street, Torrance, CA 90503; e-mail: [email protected]. 0363-5023/14/3906-0035$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.12.012
1208
r
Ó 2014 ASSH
r
Published by Elsevier, Inc. All rights reserved.
nerve innervates the skin on the dorsal aspect of the last phalanx. At the terminal aspect of each digit, the proper digital nerve divides again into 2 branches; 1 supplies the pulp and the other splits around and underneath the nail. As they run along the lateral aspect of each finger, the proper digital nerves are superficial to the corresponding arteries. There are no connections between the dorsal sensory nerve and the dorsal branch of the proper digital nerve in the thumb. There is great variation in the course that the proper digital nerve takes to the tip of the finger. Commonly, it passes between the superficial and deep transverse metacarpal ligaments. As it moves distally, it lies palmar to the adjacent artery at the side of the fibrous flexor sheath. At the distal digital crease, it goes on to divide into multiple branches that terminate at the pulp and nailbed. There is no crossover innervation of the pulp. Anatomical studies have shown that there is little change in diameter of the nerve as it makes its way distally. The ulnar nerve bifurcates approximately 5 cm proximal to the wrist into the dorsal branch and volar branch. The dorsal branch of the ulnar nerve passes beneath the flexor carpi ulnaris, perforates the deep fascia, and runs along the ulnar side of the back of the hand. As it proceeds through the midhand, it divides into 2 dorsal digital branches. One branch supplies sensation to the dorso-ulnar side of the little finger, and the other to the adjacent side of the little and ring fingers. It also sends a small branch to join with the superficial branch of the radial nerve for the adjoining sides of the middle and ring fingers. The dorsal digital branch extends to the base of the distal phalanx in the little fingers and to the base of the second phalanx in
DIGITAL NERVE INJURIES
the ring fingers. The distal aspects are supplied by the dorsal branches from the proper volar digital branch of the ulnar nerve. The volar branch of the ulnar nerve crosses the transverse carpal ligament on the lateral side of the pisiformis and ends by dividing into a superficial and a deep branch. The superficial branch supplies the skin on the ulnar side of the hand by dividing into a proper volar digital branch for innervation of the ulnar side of the little finger. In addition, a common volar digital branch is given off that divides into 2 proper digital nerve innervating the adjoining side of the little and ring fingers volarly. Nerve regeneration does not involve mitosis and multiplication of nerve cells. Instead, the cell body restores nerve continuity by growing a new axon. One axon sends out multiple unmyelinated axon sprouts from the tip of the remaining axon or collateral sprouts from a nearby proximal node of Ranvier. The distal sprout contains the growth cone. This sends out filopodia, which adhere to sticky glycoprotein molecules in the basal lamina of Schwann cells, such as laminin and fibronectin (neurite-promoting factors). The filopodia contain actin, which aids in pulling the growth cone distally. The basal lamina of 2 abutting Schwann cells form a potential endoneurial tube into which the regenerating axon grows. These axons will deteriorate if a connection with a target organ is not reached. Unlike the motor end plate, sensory end organs remain viable because there is no end plate and they retain the potential for reinnervation. Reconstruction of a sensory nerve defect, by comparison, may provide protective sensation even after many years.1 A normal nerve has longitudinal excursion, which subjects it to a certain amount of stress and strain in situ. For example, the median nerve moves as much as 12.4 mm with wrist motion at the carpal canal.2 A peripheral nerve is initially easily extensible, but this rapidly diminishes with further elongation owing to stretching of the connective tissue within the nerve. The perineurium is a mechanically strong membrane and is a major load-carrying connective tissue component that can sustain intrafascicular pressure elevations of up to 750 mm before rupturing.3 The epineurium is a loose connective tissue layer that allows a certain amount of nerve gliding. Chronically injured nerves become even stiffer. Elasticity decreases by as much as 50% in the delayed repair of nerves in which Wallerian degeneration has occurred. Experimentally, blood flow is reduced by 50% when the nerve is stretched 8% beyond its in vivo length. Complete ischemia occurs at 15% of nerve elongation. Suture pullout from the neurorrhaphy site does not occur until a 17% increase in length.4 This suggests J Hand Surg Am.
1209
CLINICAL PICTURE The patient typically presents with a history of a laceration to the palm or digit by glass or a knife, although puncture wounds caused by a drill bit can cause more extensive damage. A common digital nerve injury to the thumb will result in a sensory defect along the radial and ulnar aspects of the distal thumb. The patient with a common digital nerve injury in the palm will present with a sensory deficit along the adjacent sides of the second or third webspace (median nerve) or the fourth webspace (ulnar nerve). A proper digital nerve injury will result in an isolated sensory deficit along the radial or ulnar aspect of the thumb or digit. Static 2-point discrimination (S2PD) and moving 2PD (M2PD) testing will be greater than 25 mm on the affected side(s, which is equivalent to a complete sensory loss. Normal S2PD is 6 mm or less, whereas 15 mm is equivalent to a loss of protective sensation. Semmes Weinstein monofilament (SWM) testing will be greater than 6.65. This can easily be tested in the emergency room or office setting using the West Enhanced Sensory Test, which consists of a handle with 5 monofilaments. Measuring individual digital nerve action potentials can aid in the diagnosis of isolated digital nerve injuries (Fig. 1 A, B).5 Errors owing to volume conduction from an intact digital nerve on the opposite side must be watched for. Occasionally, it is necessary to perform a digital nerve block of the unaffected side to prevent contamination of the response in the nerve under consideration. Lacerations proximal to the takeoff of the dorsal sensory branch in the digits will also result in a sensory deficit r
Vol. 39, June 2014
Current Concepts
that ischemia, and not disruption of the neurorrhaphy, is the limiting factor in acute nerve repairs.4 This observation is also applicable to nerve grafting. There is a difference between a nerve gap and a nerve defect. A nerve gap refers to the distance between the nerve ends resulting from the elastic retraction that occurs immediately after nerve severance, whereas a nerve defect refers to the actual amount of nerve tissue that is lost. With simple nerve retraction after division, the fascicular arrangement is similar. As the nerve defect between the proximal and distal stumps increases, there is a greater fascicular mismatch between the stumps. This is of minimal consequence with digital nerve repairs because all of the fascicles contain only sensory fibers, but fascicular mismatch can still lead to an alteration of the sensory distribution to the fingertips in common digital nerve repairs.
1210
DIGITAL NERVE INJURIES
FIGURE 1: Digital nerve conduction studies. A Intact sensory nerve action potential to the radial side of the index finger (top tracing) but absent potential to the radial side of the middle finger. B Partial common digital nerve laceration to the third webspace. (Ó 2013 David J. Slutsky, MD.)
fibers, matching is not necessary. Usually 3 to 4 sutures with 8-0 or 9-0 nylon are sufficient. Alternatively, fibrin glue can be used, which allows the placement of fewer sutures. After nerve repair, rehabilitation focuses on 3 areas: initial immobilization to protect the repair, joint mobilization to promote longitudinal excursion of the nerve, and sensory reeducation. Lundborg and Rosen7 demonstrated that training protocols specifically addressing the relearning process substantially increase the possibilities for improved functional outcome after nerve repair. Before wound closure, the adjacent joints are placed in various degrees of flexion and extension, to determine the optimum limb position that unloads the repair site. This position is maintained with a blocking splint for 3 weeks, but a protected short arc of motion may be instituted to provide some nerve gliding. Yu et al8 challenged this notion and demonstrated no significant differences in 16 digits after an isolated digital nerve repair treated with 21 days of immobilization versus 14 digits with combined nerve and flexor tendon repair were treated with early protected mobilization.
over the dorsal aspect of the middle phalanx. There may be an associated flexor tendon or digital artery injury, which may affect the rehabilitation and outcome. Although uncommon, a neurapraxia rather than a nerve laceration may occur after a sharp injury, and may recover spontaneously if treated initially with observation. It is my preference to delay surgery no longer than 3 weeks, however, because the likelihood of the need for a nerve graft increases after 1 month.
Current Concepts
SURGICAL MANAGEMENT Primary nerve repair A tension-free repair is the goal for any nerve neurorrhaphy. When there is a clean transection of the nerve and the gap is caused by elastic retraction, an acute primary repair is indicated. The digital nerves are approached though a midlateral or a volar Brunner incision. If the extent of injury cannot be determined, nerve repair should be delayed. There are often 2 large groups of fascicles macroscopically in the typical digital nerve. In a cadaver dissection of 20 hands, Bonnel et al6 found that histologically, the number of fascicles ranged from 16 to 18 proximally, which increased to 21 to 28 fascicles in the distal pulp. Similarly, the amount of connective tissue relative to the fascicles increases from 70% of the cross-sectional area proximally to 90% distally. An external epineurial repair gives the same results as a fascicular repair. Because they contain only sensory J Hand Surg Am.
Nerve graft When the treatment of a nerve laceration is delayed, fibrosis of the nerve ends prevents approximation of the severed nerve ends; hence, nerve grafting is required even though there is no loss of nerve tissue. One should always be prepared to proceed with a r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
nerve graft or conduit if a nerve repair would lead to undue tension at the coaptation site. Many authors recommend nerve grafting when the gap exceeds 1 cm, with the wrist and all 3 finger joints extended. Because the graft is vascularized from the tissue bed, nerve grafting should not be performed in burned or irradiated tissue. The nerve graft acts to provide a source of empty endoneurial tubes through which the regenerating axons can be directed. Any tissue that contains a basal lamina such as freeze-dried muscle or tendon can be substituted but only the autologous nerve graft also provides a source of viable Schwann cells. A normal nerve can compensate for the change in length with limb flexion and extension because it is surrounded by gliding tissue that permits longitudinal movement. A nerve graft becomes welded to its recipient bed by the adhesions through which it becomes vascularized. As a consequence, the nerve graft is susceptible to tension because it has no longitudinal excursion. The harvested graft must be long enough to span the nerve gap without tension while the adjacent joints are extended. This should also be the position of temporary postoperative immobilization. If the digit is immobilized in flexion, the graft will become fixed in this position. When the limb is then mobilized at 8 days, the proximal and distal stumps will be subject to tension even though the graft was initially long enough. Early attempts at lengthening the graft will lead to disruption of the neurorrhaphy nerve repair.
Outcomes after repair or graft Most series report the results of nerve repair using the British Medical Research Council grading system, which was modified by Dellon.10 In this classification, a score of S0 represents absent sensory function, whereas a score of S1 or S2 represents detectable but less useful function. S3 equates to recovery of pain and touch sensibility, with 2PD of greater than 15 mm, whereas S3þ equates to 2PD of 7 to 15 mm, and S4 is normal sensation. Wang et al11 evaluated 67 adults with 74 primary digital nerve repairs and reported a recovery of 2SPD of 8 mm or less in 80% of the patients. Of these, 26 of 29 patients with a clean-cut primary repair had M2PD less than 7 mm (median, 3 mm) and 18 had S2PD less than 7 mm. Of 42 nerves with a crush or saw-type injury, 37 achieved a median of 6 mm M2PD and a median S2PD of 8 mm. Digits with mild crush or saw injuries had significantly reduced sensibility as measured by an increased 2 PD (P ¼ .02) as compared to digits with simple nerve lacerations. Patients under 40 years of age had better 2PD than those over 40 years. Kallio12 studied 95 patients with 254 completely divided digital nerves at an average of 12.4 years (range, 5e20 y) after repair. A delayed epineurial suture was used in 53, fascicular grafting in 37, and fascicular repair in 5. Useful sensory function was recovered in 79% of the nerves operated on with epineurial or perineurial suture and in 56% with fascicular grafting. Weinzweig et al13 examined the prognostic factors in a multicenter retrospective review of 172 epineural digital nerve repairs in 96 patients, with a mean age of 33.3 years (range, 5e64 y). They found the mechanism of injury and age to be predictors of sensory recovery after digital nerve injury and repair. The follow-up averaged 22.2 months (range, 6e77 mo). The authors found a statistically significant correlation between age and recovery of S2PD, with r
Vol. 39, June 2014
Current Concepts
FIGURE 2: Digital nerve graft after saw injury. Terminal posterior interosseous nerve graft interposed between the 2 fascicles of the terminal digital nerve. (Ó 2013 David J. Slutsky, MD.)
Options for nerve graft donors Higgins et al9 investigated the fascicular crosssectional area and number of fascicles of 5 nerve graft sites to specific digital nerve segments. In the fingertip distal to the distal interphalangeal joint, the anterior interosseous nerve, posterior interosseous nerve, and medial antebrachial cutaneous nerve were all appropriate choices for caliber-matched grafts (Fig. 2). The lateral antebrachial cutaneous nerve, however, was the only similar donor nerve when the number of fascicles was assessed. The lateral antebrachial cutaneous nerve is also the best match in caliber and fascicle number for digital nerve deficits between the metacarpophalangeal joint and the distal interphalangeal joint, and from the common digital nerve bifurcation to the metacarpophalangeal joint (Fig. 3 A, B). The sural nerve is the most appropriate choice when grafting defects between the wrist and the common digital nerve bifurcation, even though there are considerably fewer fascicles and a smaller cross-sectional area than the common digital nerve. J Hand Surg Am.
1211
1212
DIGITAL NERVE INJURIES
FIGURE 3: Lateral antebrachial cutaneous nerve graft. A Bifurcation of lateral antebrachial cutaneous nerve in the proximal radial forearm (arrow). B Bifurcated graft used for reconstruction of the third common digital nerve (red arrow). Lateral antebrachial cutaneous nerve graft to the small finger (white arrow) and primary nerve repair to the index finger (black arrow). (Ó 2013 David J. Slutsky, MD.)
Current Concepts
patients older than 40 years demonstrating significantly poorer recovery of sensibility than patients younger than 40 years (P < .001). There was also a significant correlation between severity of injury and recovery of sensibility (P < .001). Sharp injuries (8.2 mm) demonstrated significantly improved recovery of sensibility compared with mild crush (10.8 mm). Fractures (11.1 mm), fractures and tendon involvement (11.4 mm), and replantations (11.8 mm) demonstrated significantly diminished recovery of sensibility compared with isolated nerve injuries (7.9 mm), tendon involvement (8.1 mm), and revascularizations (9.3 mm). There appeared to be no significant difference in recovery of sensibility according to gender, digit involved, radial or ulnar side of digit, median or ulnar nerve distribution, level of injury, and time interval from injury to repair. Chen et al14 examined 17 patients with an average age of 32 years, with 21 nerve defects of 2.3 cm (range, 1.4e3.5 cm) who were treated with a nerve transfer from the dorsal branch of the proper digital nerve (PDN). A neurorrhaphy was performed between the uninjured dorsal branch of the same digit or the dorsal branch of the adjacent digit and the distal end of the PDN. They compared the results with those of a control group of series of 31 patients with PDN defects treated with conventional sural nerve grafting. At a mean follow-up of 25 months (range, 20e26 mo), the mean S2PD was 6.4 mm and the mean SWM score was 3.6. In the comparison group at a mean follow-up of 23 months (range, 19e27 mo), the mean SPD was 9.2 mm and the mean SWM score was 4.10.
regeneration. Neurotropism describes the directional accuracy of that regeneration. The growth cone is attracted to neurotropic proteins that are derived from the distal degenerating nerve segment after nerve transection. These phenomena have been exploited through the use of synthetic or natural conduits to bridge the nerve gap. Histologically, fibrin clot develops inside the tube within hours. Within the first week, longitudinally oriented fibrin matrix bridges span the nerve gap. In the second week, fibroblasts, Schwann cells, macrophages, and endothelial cells permeate the matrix. Axon sprouts from proximal nerve reach the distal stump and become myelinated by the 4th week. The axons elongate down the distal endoneurial tubes and reinnervate the target organs. Silicone nondegradable tubes were initially used experimentally. More recently, natural conduits have included freezeethawed muscle (which contains a basal lamina) and autologous vein grafts. Biodegradable synthetic material has been used, such as polyglycolic acid (PGA), caprolactone, and collagen. Acellular allografts have also been used. The postoperative rehabilitation includes early mobilization of the part, because tension on the repair site is not a consideration, followed by appropriate sensory retraining. Conduits are indicated for reconstruction of smalldiameter, noncritical sensory nerves with a gap of less than 3 cm. They can be used with digital replantation with nerve gaps of less than 3 cm and in the mangled hand if a primary, tension-free nerve repair cannot be achieved in the acute setting and if the gap is less than 1 to 2 cm. They are indicated in association with flexor tendon repair in zone II to create a tension-free repair and allow early mobilization. Relative indications
NERVE CONDUITS Neurotrophism is the ability of chemotactic hormonal or growth factors to enhance the rate of nerve J Hand Surg Am.
r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
include nerve reconstruction after resection of a traumatic neuroma or a neoplasm. They can be used as a wrap around a partially injured nerve or after neurolysis of a scarred nerve. Contraindications include uncertainty about the viability of the nerve ends, especially with avulsion injuries, blast injuries and gunshot wounds, local or systemic infection, and inadequate soft tissue coverage. When placed in superficial locations, the nerve conduit may be palpable for up to 1 year. Polyglycolic acid conduits are more likely to extrude than collagen conduits owing to their more rigid nature, and require thicker soft tissue coverage. Relative contraindications include nerve gaps greater than 3 cm.
FIGURE 4: Collagen conduit (forceps) bridging a gap in a radial digital nerve deficit of the index finger. (Ó 2013 David J. Slutsky, MD.)
Synthetic conduits The procedure is done with the patient supine and the extremity on an arm board under tourniquet control. Microsurgical instruments are needed; high-power loupe magnification is usually sufficient, although a fascicular dissection may be performed under the operating microscope. The proximal and distal nerve ends are identified and meticulous hemostasis is obtained. The tension-free gap is measured with the adjacent joints in full extension. The nerve stumps are trimmed under magnification until healthy fascicles are identified. The minimum distance between the ends is 5 mm, with a maximum of 3 cm. The diameter of the nerve is measured with a ruler to determine the appropriate size conduit. The inner diameter of the tube should be equal to or slightly larger than the diameter of the nerve to be implanted. Some synthetic conduits require saline immersion before nerve implantation. Conduit sizes range from 2 to 10 mm and 4 cm in length. After the appropriate size is chosen, 1 nerve end is pulled into the conduit so that 2 to 5 mm of the nerve lies within the conduit, or a distance equal to or greater than the nerve diameter. An 8e0 or 9e0 nylon suture is used to draw the nerve ends into the conduit using a horizontal mattress stitch (Fig. 4). Either the proximal or distal end is sutured first. The stitch is passed from the outside to inside of the conduit wall at least 1 mm from its end. The suture is next passed transversely through the epineurium 2 to 3 mm from the stump, and then back through the conduit from inside to out, and tied after the stump is drawn into the conduit. The conduit lumen is filled with sterile saline for collagen tubes and heparinized saline for PGA tubes, as per the manufacturers’ recommendation, to prevent air bubbles or blood clot formation, which may impede axonal ingrowth. J Hand Surg Am.
1213
POSTOPERATIVE REHABILITATION Repair of associated tendon or muscle injuries will influence the rehabilitation. With an isolated nerve reconstruction, the affected part is immobilized with the adjacent joints in neutral position. Composite extension or flexion in the opposite direction of the repair is avoided to prevent tension that could dislocate a nerve end from the conduit. Range of motion exercises are started at 10 days, followed r
Vol. 39, June 2014
Current Concepts
Autologous vein grafts The use of autologous veins as a biologic tube to reconstruct a nerve gap has been well described over the past century. Chiu and Strauch15 championed the use of autologous venous nerve conduits (AVNC) for the reconstruction of digital sensory nerve defects of 3 cm or less. They demonstrated clinical results that were more favorable compared with autologous nerve grafts. Advantages of the AVNC include the avoidance of a cutaneous donor site sensory deficit and minimal cost compared with the commercially manufactured synthetic conduits. The superficial veins on the dorsum of the hand are of an appropriate caliber for digital nerve reconstruction (Fig. 5 A, B). The authors noted that there is a fairly constant dorsal vein that runs longitudinally in line with the fourth webspace and is usually adjacent to the dorsal cutaneous branch of the ulnar nerve. The AVNC is reversed because of the possible presence of venous valves. The cut proximal end of the nerve is placed in the end of the vein that was distal (in situ) using a telescoping suture. The same is done to the other end of the nerve after the vein graft is cut to the appropriate length. Some authors prefer to use only lower-extremity veins because they are thicker and have less of a tendency to collapse.
1214
DIGITAL NERVE INJURIES
FIGURE 5: Vein conduit. A Harvest of autologous dorsal vein conduit. B Reversed vein graft prior reconstruction of 1-cm nerve defect (arrow). (Ó 2013 David J. Slutsky, MD.)
by strengthening exercises at 6 weeks and sensory reeducation for the next year.
results of using a caprolactone tube versus primary repair in 34 digital nerve lesions with gaps less than 2 cm. The results with the conduits were equal to those with repair, as evaluated by the pressure-specified sensory testing device and SP2D. Haug et al19 evaluated 35 patients with a mean age of 47 years (range, 11e83 y), with 45 digital nerve defects. The nerve defects ranged from 5 to 26 mm (mean, 12 mm) and were reconstructed with 2- to 3-mm-diameter collagen tubes with a mean length of 19 mm (range, 11e30 mm). They used a sum score that included a variety of modalities including mechanical stimuli, warm/cold sensation vibration, SWM, 2SPD, number recognition, and electroneurography. Basic sensory functions such as sharp/dull and warm/cold sensation as well as vibration showed faster recovery than more complex functions such as SWM sensation, S2PD, and number recognition. Recovery of sensory functions 12 months after nerve tube implantation, as measured by the sum score, was very good in 4 cases, good in 21, mediocre in 14, and poor in 3.
Current Concepts
OUTCOMES Chiu and Strauch15 conducted a clinical series of animal experiments bridging gaps of 1.0 to 6.0 cm using autologous vein grafts, and reported results comparable to those obtained with nerve repair and grafting for sensory nerve defects of less than 3 cm. Tang et al16 examined patients after 22 digital nerve repairs using autologous vein grafts. The S2PD averaged 4.6 mm for 11 acute digital nerve repairs using vein conduits 1 to 3 cm in length. Delayed digital nerve repair with vein conduits yielded poor results. Weber et al17 performed a randomized, prospective, multicenter evaluation of 98 patients with 136 nerve transections in the hand, comparing a PGA conduit with either end-to-end repair or a nerve graft. A total of 56 nerves repaired in the control group and 46 nerves repaired with a conduit were available for follow-up. The overall results showed no significant difference between the 2 groups as a whole. Nerves with gaps of 4 mm or less had better sensation when repaired with a conduit; the mean M2PD was 3.7 1.4 mm for the PGA tube repair and 6.1 3.3 mm for end-to-end repairs (P ¼ .03). All injured nerves with deficits of 8 mm or greater were reconstructed with either a nerve graft or a conduit. This subgroup also demonstrated a significant difference in favor of the PGA tube. The mean MP2D for the conduit was 6.8 3.8 mm, with excellent results obtained in 7 of 17 nerves, whereas the mean MP2D for the graft repair was 12.9 2.4 mm, with excellent results obtained in none of the 8 nerves (P < .001 and P ¼ .06, respectively). Bertleff et al18 examined the J Hand Surg Am.
Vein conduit with skeletal muscle Marcoccio and Vigasio20 treated 17 patients with 21 digital nerve defects using a vein conduit filled with fresh skeletal muscle. The donor vein and fresh muscle segments were obtained through a 5-cm skin incision in the proximal forearm. A suitable vein graft was harvested and a 22-gauge cannula was introduced into the lumen of the vein. A longitudinal segment of muscle fibers was harvested slightly longer than the vein graft. A 7e0 nylon suture was used to draw the muscle into the vein graft, and then removed. A tension-free repair was performed as described above. The average nerve gap was 2.2 cm (range, 1e3.5 cm). r
Vol. 39, June 2014
DIGITAL NERVE INJURIES
1. Sunderland S. Capacity of reinnervated muscles to function efficiently after prolonged denervation. AMA Arch Neurol Psychiatry. 1950;64(6):755e771. 2. Coppieters MW, Alshami AM. Longitudinal excursion and strain in the median nerve during novel nerve gliding exercises for carpal tunnel syndrome. J Orthop Res. 2007;25(7):972e980. 3. Selander D, Sjostrand J. Longitudinal spread of intraneurally injected local anesthetics: an experimental study of the initial neural distribution following intraneural injections. Acta Anaesthesiol Scand. 1978;22(6):622e634. 4. Clark WL, Trumble TE, Swiontkowski MF, Tencer AF. Nerve tension and blood flow in a rat model of immediate and delayed repairs. J Hand Surg Am. 1992;17(4):677e687. 5. Terai Y, Senda M, Hashizume H, Nagashima H, Inoue H. Selective measurement of digital nerve conduction velocity. J Orthop Sci. 2001;6(2):123e127. 6. Bonnel F, Foucher G, Saint-Andre JM. Histologic structure of the palmar digital nerves of the hand and its application to nerve grafting. J Hand Surg Am. 1989;14(5):874e881. 7. Lundborg G, Rosen B. Hand function after nerve repair. Acta Physiol. 2007;189(2):207e217. 8. Yu RS, Catalano LW III, Barron OA, Johnson C, Glickel SZ. Limited, protected postsurgical motion does not affect the results of digital nerve repair. J Hand Surg Am. 2004;29(2):302e306. 9. Higgins JP, Fisher S, Serletti JM, Orlando GS. Assessment of nerve graft donor sites used for reconstruction of traumatic digital nerve defects. J Hand Surg Am. 2002;27(2):286e292. 10. Dellon AL. A numerical grading scale for peripheral nerve function. J Hand Ther. 1993;6(2):152e160. 11. Wang WZ, Crain GM, Baylis W, Tsai TM. Outcome of digital nerve injuries in adults. J Hand Surg Am. 1996;21(1):138e143. 12. Kallio PK. The results of secondary repair of 254 digital nerves. J Hand Surg Br. 1993;18(3):327e330. 13. Weinzweig N, Chin G, Mead M, et al. Recovery of sensibility after digital neurorrhaphy: a clinical investigation of prognostic factors. Ann Plast Surg. 2000;44(6):610e617. 14. Chen C, Tang P, Zhang X. Finger sensory reconstruction with transfer of the proper digital nerve dorsal branch. J Hand Surg Am. 2013;38(1):82e89. 15. Chiu DT, Strauch B. A prospective clinical evaluation of autogenous vein grafts used as a nerve conduit for distal sensory nerve defects of 3 cm or less. Plast Reconstr Surg. 1990;86(5):928e934. 16. Tang JB, Gu YQ, Song YS. Repair of digital nerve defect with autogenous vein graft during flexor tendon surgery in zone 2. J Hand Surg Br. 1993;18(4):449e453. 17. Weber RA, Breidenbach WC, Brown RE, Jabaley ME, Mass DP. A randomized prospective study of polyglycolic acid conduits for digital nerve reconstruction in humans. Plast Reconstr Surg. 2000;106(5):1036e1045; discussion 1046e1038. 18. Bertleff MJ, Meek MF, Nicolai JP. A prospective clinical evaluation of biodegradable neurolac nerve guides for sensory nerve repair in the hand. J Hand Surg Am. 2005;30(3):513e518. 19. Haug A, Bartels A, Kotas J, Kunesch E. Sensory recovery 1 year after bridging digital nerve defects with collagen tubes. J Hand Surg Am. 2013;38(1):90e97. 20. Marcoccio I, Vigasio A. Muscle-in-vein nerve guide for secondary reconstruction in digital nerve lesions. J Hand Surg Am. 2010;35(9): 1418e1426. 21. Taras JS, Amin N, Patel N, McCabe LA. Allograft reconstruction for digital nerve loss. J Hand Surg Am. 2013;38(10): 1965e1971. 22. Cho MS, Rinker BD, Weber RV, et al. Functional outcome following nerve repair in the upper extremity using processed nerve allograft. J Hand Surg Am. 2012;37(11):2340e2349.
Processed nerve allografts Processed nerve allografts (Avance Nerve Graft; AxoGen, Inc., Alachua, FL) consist of decellularized and predegenerated nerve tissue that is rapidly revascularized and repopulated with host Schwann cells. The immunogenicity of the graft is decreased by removing residual nerve debris and Schwann cells while preserving the normal internal neural architecture and important guidance molecules such as laminin. Taras et al21 reported the outcomes of 14 patients with an average age of 39 years (range, 18e76 y) with 18 digital nerve reconstruction using processed nerve allograft at an average of 15 months (range, 12e20 mo). The average nerve gap was 11 mm (range, 5e30 mm). Seven of 18 achieved S2PD of 6 mm or less, 8 achieved 7 mm, and 3 achieved 8 mm. Cho et al22 analyzed the data from a registry database of 35 subjects with a mean age of 46 14 years (range, 23e68 y). The time to surgery was 190 349 days and the follow-up was 306 184 days. The gap was 19 9 mm. There were 24 lacerations, 6 neuromas, and 5 complex lacerations. There was meaningful recovery of S3eS4 in 31 of 35 digital nerve repairs. The take-home messages are as follows: a tensionfree coaptation is a key factor in any nerve repair. Ischemia and not disruption of the neurorrhaphy is the limiting factor in acute nerve repairs. Grouped fascicular repairs do not appear to have any advantages over an epineurial repair. It is apparent that a variety of host factors influence the outcomes of digital nerve repairs more than the type of repair per se. Younger patients do better than older patients, sharp injuries do better than crush injuries, and isolated nerve injuries do better than those with associated fractures, tendon involvement. and revascularizations. Although autologous graft remains the reference standard for reconstruction of any critical digital nerve defect, the use of allografts and conduits is increasing in popularity.
r
Vol. 39, June 2014
Current Concepts
REFERENCES
At a mean follow-up of 43 months (range, 18e96 mo), 5 patients achieved S4 with complete recovery of S2PD of 2 to 6 mm and MP2D of 2 to 3 mm; 9 patients achieved S3þ with SPD of 7 to 15 mm and MP2D of 4 to 7 mm with but good localization of stimulus; and 3 patients achieved S3 with S2PD of greater than 15 mm, MP2D greater than 7 mm, and recovery of pain and touch sensibility. One patient had S2 with recovery of superficial pain and some touch sensibility, and 4 patients had S1 with recovery of superficial pain and only some touch sensibility.
J Hand Surg Am.
1215
