Hand Surgery
1st Edition

55
Reconstruction for Ulnar Nerve Palsy
Kavi Sachar
Ulnar nerve paralysis most commonly results from traumatic injury. This can be by direct laceration, blunt trauma, severe crush injury, or nerve compression. It can be combined with other peripheral nerve injuries or occur in conjunction with brachial plexus level injuries. Additionally, it can occur in systemic conditions such as Charcot-Marie-Tooth disease and leprosy. The prognosis for recovery is dependent on the nature of the injury, the treatment rendered, and, most important, the age of the patient.
Ulnar nerve paralysis results in a combination of motor and sensory deficits. Because forceful and fine motor manipulation activities may be affected, the limitations may significantly hamper function. Each patient, however, has different needs based on age, sex, and occupation. Anatomic variations, combined lesions, and associated injuries also affect function. A detailed understanding of normal ulnar nerve anatomy allows the physician to localize the level of injury. Knowledge of normal and anomalous innervation patterns allows the physician to document motor and sensory deficiencies. This knowledge, combined with an assessment of the patient’s functional limitations, allows the physician and patient to determine the best treatment options. Reconstructive procedures for ulnar nerve paralysis are considered after careful evaluation of the patient’s subjective complaints, functional deficit, and potential for recovery from the initial traumatic incident.
ANATOMY
The C8 to T1 nerve roots combine to form the lower trunk, which contributes to both the posterior and medial cords. The medial cord gives rise to the medial pectoral nerve, the medial brachial cutaneous nerve, and the medial antebrachial cutaneous nerve. The medial cord terminates as the ulnar nerve. Although the ulnar nerve derives the majority of its fibers from the C8 and T1 nerve roots, it receives contributions from the seventh cervical nerve and, occasionally, higher nerve roots at least 50% of the time. This is accomplished through a lateral root derived from the lateral cord or one of its branches (Fig. 1) (1).
At its origin, the ulnar nerve lies medial to the axillary artery, lateral to the axillary vein, and posterior to the medial antebrachial cutaneous nerve. In the upper arm, it lies medial or posterior to the brachial artery. The ulnar nerve pierces the intermuscular septum at approximately the middle of the arm. It then lies on the front of the medial head of the triceps muscle. The medial epicondyle forms the roof of the cubital tunnel, where the nerve is covered by multiple fascial layers. In the cubital tunnel, the nerve lies beneath the fascial arcade formed by the superficial and deep heads of the flexor carpi ulnaris. As it passes between these heads, it comes to lie on the volar surface of the flexor digitorum profundus (FDP) in the middle of the forearm. The nerve lies beneath the flexor carpi ulnaris at the level of the wrist. Both the ulnar nerve and ulnar artery pass lateral to the pisiform and lie volar to the transverse carpal ligament at the wrist. The ulnar nerve ends at the base of the hypothenar eminence by dividing into deep and superficial palmar branches.
The ulnar nerve’s first motor branch is usually to the flexor carpi ulnaris muscle, and it originates distal to the medial epicondyle. The nerve then goes on to innervate the ring and small finger components of the FDP tendon. The next branch is the dorsal sensory branch of the ulnar nerve, which originates 7 cm proximal to the radial styloid, providing sensation to the dorsal ulnar aspect of the hand. In the hand, the superficial branch of the ulnar nerve terminates as a common digital nerve to the ring and small web space and a proper digital nerve to the ulnar border of the small finger. As the deep motor branch courses through the hypothenar muscles, it innervates the abductor digiti minimi, flexor digiti minimi brevis, and the opponens digiti minimi. The deep branch then goes on to innervate the two medial lumbricals, all the interossei, the adductor pollicis, and the deep head of the flexor pollicis brevis. The nerve ends by innervating the first dorsal interosseous (FDI). A schematic of normal ulnar nerve motor and sensory innervation is depicted in Figure 2.
FIGURE 1. The brachial plexus. The ulnar nerve primarily arises from C8 and T1 but occasionally receives branches from higher nerve roots.
FIGURE 2. Schematic representation of ulnar nerve innervation in sequential fashion. ADM, abductor digiti minimi; FCU, flexor carpi ulnaris; FDI, first dorsal interosseous; FDP, flexor digitorum profundus; FPB, flexor pollicis brevis.
P.980

Anomalous innervation patterns occur in the normally innervated ulnar muscles. A forearm ulnar-median communication pattern known as the Martin-Gruber anastomosis occurs in 17% of patients (2). Type I connections (60%) send motor branches from the median nerve to the ulnar nerve to innervate “median” muscles. Type II connections (35%) send motor branches from the median to the ulnar nerve to innervate “ulnar” muscles. Type III connections (3%) send motor branches from the ulnar to the median nerve to innervate “ulnar” muscles, and type IV connections (1%) send motor fibers from the ulnar to the median nerve to innervate “ulnar” muscles. Palmar communications also exist and have been described by Riche and Cannieu (3). Additionally, the FDI, third and fourth lumbricals, and adductor pollicis all have been described as having dual innervation. All of these anatomic variants affect the functional deficit that is encountered in ulnar nerve lesions.
CLINICAL EVALUATION
Ulnar nerve palsy results in characteristic clinical signs that reflect the motor loss, sensory loss, and the level of injury. Additionally, specific clinical tests can be performed to demonstrate the functional loss that occurs. The clinical signs can vary based on variant innervation patterns, combined injuries, and generalized ligamentous laxity.
P.981

FIGURE 3. Wartenberg’s sign. This results from unopposed pull of the small finger extensor tendons.
Flattening of the metacarpal arch and hypothenar muscles (Masse’s sign) (4,5) results from atrophy of the interossei and opponens digiti quinti. Abductor digiti minimi paralysis allows for unopposed pull of the extensor tendon, which results in small finger abduction (Wartenberg’s sign) (Fig. 3) (6,7).
Clawing of the fourth and fifth fingers (Duchenne’s sign) occurs due to loss of intrinsic function (8). Without the intrinsics, there is a loss of metacarpophalangeal (MCP) flexion power and proximal interphalangeal (PIP) extension power. This results in an intrinsic minus position of MCP extension and PIP flexion. The median-innervated lumbricals partially protect the index and middle fingers. The clawing is less severe in high ulnar nerve palsy because the loss of the FDP to the ring and small fingers produces less of a flexion force on the distal digits. Patients with more ligamentous laxity may have worse clawing. The MCP volar plate eventually stretches out, accentuating the MCP hyperextension. Early on, the PIP capsule and extensor tendons remain normal. Eventually, however, patients may lose the ability to extend the PIP joint. This is what distinguishes simple from complex clawing and is the basis for Bouvier’s test (9). In simple clawing, when MCP hyperextension is corrected, the extensor tendons are able to extend the PIP joints, demonstrating integrity of the extensor apparatus. This is considered to be a positive Bouvier’s test. In a negative Bouvier’s test, the patient cannot actively extend the PIP joints when MCP hyperextension is corrected, indicating insufficiency of the extensor apparatus. When tendon transfers to correct clawing are planned in a patient with simple clawing (a positive Bouvier’s test), the tendon transfer only has to correct the clawing. In complex clawing (a negative Bouvier’s test), the tendon transfer must correct the clawing and provide PIP extension.
FIGURE 4. Froment’s sign. This results from loss of adductor pollicis function with compensatory thumb interphalangeal flexion.
As clawing becomes more severe, finger extension becomes more difficult. Patients sometimes flex their wrist to extend their digits, taking advantage of extrinsic tenodesis. This is known as the André Thomas sign (10) and is a poor prognostic indicator for successful tendon transfers because the pattern may become ingrained and difficult to change (11).
The clinical tests in ulnar nerve palsy demonstrate both muscle paralysis and loss of functional hand activities. Key pinch relies on adductor pollicis function. In ulnar nerve palsy, key pinch may be diminished as much as 77% to 80% (7,11,12). The flexor pollicis brevis also contributes to key pinch by causing flexion at the MCP joint and preventing MCP hyperextension. In ulnar nerve palsy, adductor pollicis paralysis is compensated for by using the FPL tendon. This results in hyperflexion of the thumb IP joint when attempting key pinch (Froment’s sign) (Fig. 4) (13,14). The thumb MCP joint may hyperextend during pinch (Jeanne’s sign) (15) if flexor pollicis brevis function is absent (Fig. 5).
Ulnar nerve palsy results in the loss of integration of PIP and MCP flexion (11). This dyssynchronous finger flexion
P.982

results from intrinsic paralysis. In normal gripping, the intrinsics flex the MCP joints before distal interphalangeal flexion, allowing for broad objects to be grasped. In ulnar nerve palsy, ring and small finger MCP joint flexion can occur only after distal joint flexion. Patients therefore tend to push items out of their hands because they cannot secure them in their palms before distal joint flexion (Fig. 6).
FIGURE 5. Jeanne’s sign: hyperextension of the thumb metacarpophalangeal joint.
FIGURE 6. Dyssynchronous finger flexion results from loss of intrinsic function and the normal flexion cascade.
Omer described the “cross-your-finger test” to demonstrate paralysis of the first volar interosseous and second dorsal interosseous (16). This test can reveal an inability to cross the flexed long finger dorsally over the index finger or the index finger over the long finger with the hand on a flat surface.
The Pitres-Testut sign is demonstrated by an inability to abduct the middle finger when on a flat surface (7). This tests the second and third dorsal interossei muscles.
In high ulnar nerve palsy, there is an inability to flex the distal interphalangeal joints of the ring and small fingers due to paralysis of the ulnar innervated FDP tendons (Pollock’s sign) (4). These clinical signs and findings are summarized in Table 1. The sensory loss that occurs in ulnar nerve paralysis depends on the level of nerve injury. If the nerve injury is above the mid-forearm, the dorsal sensory branch of the ulnar nerve is affected, resulting in dorsal ulnar hand anesthesia. There is considerable variability in dorsal hand innervation (17). The radial sensory contribution may be large, so a detailed examination is necessary. Palmar sensation is absent over the volar small finger and ulnar one-half of the ring finger (1).
TABLE 1. CLINICAL SIGNS AND FINDINGS IN ULNAR NERVE PALSY
Signs Findings
Masse’s sign Hypothenar atrophy
Wartenberg’s sign Small finger abduction
Duchenne’s sign Ring and small finger clawing
André Thomas sign Wrist flexion to extend digits
Jeanne’s sign Thumb metacarpophalangeal hyperextension
Froment’s sign Thumb interphalangeal hyperflexion with key pinch
Cross-your-finger test Inability to cross index and middle finger
Pitres-Testut sign Inability to abduct middle finger
Pollock’s sign Loss of flexor digitorum profundus to ring and small finger (high ulnar nerve palsy)
SURGICAL PLANNING
The traumatic or systemic incident that led to the ulnar nerve paralysis should be evaluated in detail. The potential for nerve recovery is variable depending on multiple factors (18,19,20,21 and 22). The most important prognostic factor for recovery is the age of the patient. For example, a child with a low ulnar nerve laceration that has been repaired acutely has a very good prognosis for recovery and should be followed until sufficient time has passed during which recovery should occur. In contrast, an adult with chronic ulnar nerve compression, established atrophy, and deformity has a poor prognosis for recovery and is a candidate for tendon transfers if appropriate.
The location of the nerve injury and mechanism of injury should be noted. High-energy proximal traction injuries, such as from motorcycles or snowmobiles, have a worse prognosis than low-level, sharp lacerations that have been repaired. It should be noted if the patient had an associated fracture or tendon injuries that may affect available tendons for transfer.
Combined nerve injuries present a special problem because of tendon availability and specific functional needs (23). Careful documentation of available tendons and their motor strength should be made.
In all cases, a chart should be made that documents the functional deficits, tendons available for transfer, their strength, tendon transfer options, and proposed transfers. Patients should be well informed of the proposed procedure and what functional gains should result from the transfers.
All joints to receive transfers should be passively supple. Associated fractures should be healed, and soft tissue wounds should be stable.
SURGICAL MANAGEMENT
Tendon Transfers for Pinch
Key pinch involves forcefully applying the tip of the thumb to the radial border of the index finger. This is primarily accomplished by the adductor pollicis and FDI. The extensor pollicis longus and flexor pollicis longus contribute by
P.983

providing a small adductor moment. The flexor pollicis brevis provides MCP stability (24,25).
Several tendon transfers have been designed to restore pinch (11,14,16,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41 and 42). Some restore only adductor pollicis function; others restore FDI function as well. The motors that have been described include wrist extensors, the brachioradialis, digital extensors, and digital flexors. Regardless of the motor, most tendon transfers provide only 25% to 50% of normal pinch strength (11).
The decision to restore pinch is based on a patient’s functional needs. In a series by Hastings and Davidson, many patients with ulnar nerve palsy were satisfied with their pinch despite significant weakness (11).
Transfers to Restore Adductor Pollicis Function
The tendon motors that are available to restore adductor pollicis function are the extensor carpi radialis brevis (ECRB), flexor digitorum superficialis (FDS), brachioradialis, and extensor digitorum communis.
Extensor Carpi Radialis Brevis
Smith described using the ECRB with a tendon graft to restore adductor pollicis function (24,25). The ECRB is released from its insertion and lengthened with a free palmaris tendon graft. This is then passed through the second metacarpal interspace, dorsal to the adductor pollicis, flexor tendons, and neurovascular structures and volar to the interossei. It is sutured to the adductor pollicis insertion with a tendon weave. Tension is set so that when the wrist is neutral, the thumb lies palmar to the index finger. This transfer has the added benefit of thumb abduction with wrist flexion and thumb adduction with wrist extension through a tenodesis effect (Fig. 7). Omer modified this technique slightly (16). He advocates passing the transfer through the third intermetacarpal space and securing it to the fascia over the abductor tubercle of the first metacarpal. He believes this insertion point improves pronation for pinch (36).
Use of the ECRB offers several advantages to other transfers for pinch. It is an extremely strong motor that is not missed because of the numerous other tendons available for wrist extension. The vector of the transfer recreates the transverse head of the adductor pollicis, which is more functional than restoring the oblique head. It also does not sacrifice a finger flexor in a hand that has already weakened grasp.
Flexor Digitorum Superficialis
The FDS tendons are useful motor donors in low ulnar nerve palsy because they provide a strong motor, have excellent excursion, and are less missed if the FDP tendon is innervated. In low ulnar nerve palsy, the FDS to the ring finger can be sacrificed. In high ulnar nerve palsy, the FDS to the middle finger can be used.
FIGURE 7. Extensor carpi radialis brevis tendon transfer for pinch. The tendon is elongated with a tendon graft and sutured to the adductor pollicis insertion.
Littler et al. described using the FDS of the ring or middle finger to restore thumb adduction (Fig. 8) (37,38 and 39). An oblique incision is made at the base of the finger in the distal palm similar to a trigger finger incision. The finger is flexed, and the FDS tendon is divided distal to the decussation, but the insertion is left to prevent hyperextension deformity of the PIP joint. The tendon is then tunneled to a second incision over the adductor pollicis insertion. The tendon is passed deep to the flexor tendons and neurovascular bundles, paralleling the transverse head of the adductor pollicis. The vertical septae of the palmar fascia serve as a pulley. Littler described passing the transfer into bone with a pullout button. Alternatively, the transfer can be sutured to the adductor pollicis insertion. Tension is set so that with 30 degrees of wrist flexion, the thumb lies in moderate flexion adduction. The hand is immobilized for 3 to 4 weeks and then active motion started. Hamlin and Littler have shown that 70% of pinch power can be restored with this procedure (39).
P.984

FIGURE 8. Flexor digitorum superficialis (FDS) tendon transfer to restore pinch. The palmar fascia serves as a pulley. Alternatively, the FDS to the middle can be used.
Brachioradialis
Boyes described using the brachioradialis tendon as a motor, citing its benefit of functioning in both wrist flexion and extension (40). A lengthy incision from the first dorsal compartment to the mid-forearm allows for mobilization of the tendon and its muscle belly. The muscle is released at the base of the first dorsal compartment. Excursion of up to 30 mm can be obtained by releasing the muscle to the proximal third of the radius. The brachioradialis tendon is elongated with a palmaris longus graft and passed in the same manner as the ECRB transfer as described by Smith.
Extensor Digitorum
When a finger flexor is not available as in combined nerve palsies, the extensor indicis proprius (EIP) can be used as a motor. The tendon is released just proximal to the extensor hood and then passed between the third and fourth meta-carpals, dorsal the adductor pollicis. The tendon is brought through the web space and secured to the adductor pollicis insertion (24). Edgerton and Brand have recommended inserting the tendon into the abductor pollicis tendon as a more functional insertion (36).
Bunnel described using the extensor indicis communis tendon because its muscle is stronger than the EIP (40,42). The tendon is elongated with a tendon graft and passed subcutaneous around the ulnar border of the hand and deep to the flexor tendons. It is sutured to the adductor pollicis insertion as previously described.
Restoration of First Dorsal Interosseous
Numerous transfers have been described to restore FDI function (32,40,41,42,43 and 44), which, when combined with restoration of adductor pollicis function, would truly restore pinch. Most patients, however, can stabilize the index finger against the middle finger, and the true vector of the FDI is difficult to reproduce. These transfers have been reported to improve pinch only 10% to 15%. A strong motor for FDI may produce a “nose-picker finger” with persistent radial deviation of the index finger (11).
Abductor Pollicis Longus Transfer
This transfer is performed by isolating an abductor pollicis longus slip that does not contribute to abduction of the thumb (32). This is usually a more radial slip. The tendon is lengthened with a tendon graft. A dorsal radial incision is made along the FDI, and a dorsal subcutaneous tunnel is created. The tendon graft is woven into the FDI insertion. This transfer does not make the FDI much stronger but does stabilize it during pinch (Fig. 9).
FIGURE 9. Abductor pollicis longus (APL) tendon transfer with graft to restore first dorsal interosseous (FDI) function.
Extensor Digitorum
Bunnel described transfer of the EIP to restore FDI function, and this was later modified by Omer (16,40,42). Omer describes taking the EIP proximal to the extensor hood, withdrawing it at the wrist, and passing it around the radial border of the second metacarpal. The tendon is split so that one slip inserts at the FDI tendon and the other at the adductor pollicis. This transfer does not restore strong power pinch but does stabilize the index finger (16).
The extensor digiti quinti can be used in a manner similar to the EIP (34). The two slips of the extensor digiti quinti are divided and brought distal to the fourth dorsal compartment retinaculum. They are then passed subcutaneously to the insertion of the FDI and adductor pollicis with one slip sewn to each. Although the extensor digiti quinti is a weak motor, recovery of adequate pinch has been reported (34).
Correction of Claw Deformity
The claw deformity in ulnar nerve paralysis is due to loss of intrinsic function. This results in MCP hyperextension and PIP flexion. MCP hyperextension can be corrected with either a static tissue tightening procedure or a dynamic tendon transfer.
P.985

In patients with a positive Bouvier’s test, there is no need to restore PIP extension. If the Bouvier’s test is negative, a tendon transfer must restore both MCP flexion and PIP extension.
Procedures to improve clawing include bony procedures, soft tissue tightening procedures, and tendon transfers. Tendon transfers use the FDS, extensor carpi radialis longus (ECRL), ECRB, and flexor carpi radialis (FCR) tendons.
Bone and Soft Tissue Procedures for Correction of Clawing
Early attempts at correcting MCP hyperextension involved use of a bone block dorsally over the MCP joint (42,45). These techniques have largely been abandoned as soft tissue techniques have improved.
Bunnel described a procedure that creates bowstringing to increase the flexor moment on the MCP joint (46). He called this a “flexor pulley advancement.” The A1 and A2 pulleys are released on both sides but not in the midline to the level of the midproximal phalanx. This allows the flexor tendon to partly bowstring, moving its axis more volar to the MCP joints.
Tightening the volar plate may be useful in patients with mild clawing and good PIP function. Zancolli described proximal advancement of the volar plate (47). Through a longitudinal incision for each digit, the A1 pulley is released, and a distally based rectangular flap is created from the volar plate. The flap is advanced proximally and sutured with the digit in 20 degrees of flexion. Omer modified the procedure by securing the flap to the deep transverse metacarpal ligament (Fig. 10) (16). Leddy et al. recommend securing the proximally advanced volar plate to bone (48).
Brown combined the Bunnel flexor pulley advancement and the Zancolli capsulorrhaphy secured to bone and added the excision of 1.5 cm of skin. He critically evaluated this technique and found a high complication rate. Complications included PIP and distal interphalangeal contractures and a gradual stretching of the capsulorrhaphy. All recurrences occurred within the first year (49).
Flexor Digitorum Superficialis Transfer for Correction of Clawing
There are several variations of the FDS transfer for correction of clawing (50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66 and 67). The basic principle is to keep the
P.986

transferred tendon volar to the deep intermetacarpal ligament, reproducing the vector of the lumbrical. In the Zancolli lasso, the transfer is sewn on itself as a loop (50). In the Stiles-type transfers, the tendon is sewn to the flexor tendon sheath, to the proximal phalanx, or into the lateral band if the Bouvier’s test is positive (55).
FIGURE 10. Volar plate advancement for correction of clawing. After releasing the A1 pulley (left), a distally based flap of volar plate is developed (middle). The flap is advanced proximally (arrow) and secured with a suture anchor to the metacarpal (right).
In general, FDS transfers do not improve power grip because they sacrifice at least one finger flexor. Hastings and McCollam have shown a decrease in grip strength after FDS transfers (62). There is also a high incidence of PIP swan-neck deformities after these transfers (49). The FDS of the ring and small fingers should not be used in high ulnar nerve palsies because these are the only finger motors to the ring and small fingers. Transfers should be performed on all fingers because dynamic clawing may be present in the index and middle fingers as well (16). These transfers do well only in young, ligamentously lax individuals. They are unpredictable in people with stiff hands. Patients with a positive André Thomas sign do not do well with dynamic tendon transfers because of ingrained behavior that makes it difficult to train the transfer (11).
Zancolli Lasso
Zancolli described a tendon transfer that involves using the FDS tendon as a lasso around the A1 pulley (Fig. 11) (50). This provides both static flexion and dynamic flexion strength to the MCP joint. The flexor tendon sheath is exposed through a Bruner incision. The interval between the A1 and A2 pulley is incised, and the FDS tendon is divided, leaving its insertion to prevent hyperextension. The FDS is then brought volar to the A1 pulley and sutured to itself. The transfer is sutured with the MCP joint in neutral position and the FDS pulled fairly tight. Tension is set so that the MCP joint sits in 40 to 60 degrees of flexion with the wrist in neutral. The small finger tension is set tighter than the ring finger. The MCP joint should be able to be passively extended to 0 degrees with some difficulty with the wrist in neutral but no difficulty with the wrist flexed. A dorsal blocking splint should be used for 6 weeks after surgery. Typically, this procedure is done with the sublimis tendon being used for its own finger. However, if the ring and small finger do not have a strong sublimis tendon or have no active profundus as in high ulnar nerve palsy, the middle finger sublimis can be split and used for both fingers (11). Omer describes looping the FDS around the A2 pulley and performs the procedure on all four fingers (16).
Stiles, Bunnel, Riordan Techniques
Bunnel modified the original Stiles FDS transfer by using all of the FDS tendons and transferring them to both sides of the clawed fingers (46,55). This is considered too strong of a transfer and results in PIP swan-neck deformity (46). The Stiles-Bunnel transfer now involves using one FDS tendon to motor two digits (16).
Midaxial incisions are made on the radial sides of each digit. A midpalmar incision is made to retrieve the FDS donor tendons. The ring finger FDS is harvested through a window between the A1 and A2 pulleys. It is split so that it
P.987

can be used for two fingers. If all four fingers are to be corrected, a second FDS tendon is harvested. Each slip is then passed through the lumbrical canal volar to the deep intermetacarpal ligament. This can be accomplished by passing a Carroll tendon passer from the digital wound proximally into the palmar wound along the path of the lumbrical. With a positive Bouvier’s test, the transfer is sutured to the flexor tendon sheath. If the Bouvier’s test is negative, the transfer is sewn to the lateral band. Tension is set with the wrist in neutral, the MCP joints in 45 to 50 degrees of flexion, and the IP joints in full extension. One must be careful not to set the tension too tightly and risk PIP hyperextension.
FIGURE 11. Flexor digitorum superficialis (FDS) tendon transfers for correction of clawing. The flexor digitorum superficialis can be sewn to the lateral band (A), to bone (B), or on itself in the Zancolli lasso (C).
FIGURE 12. A: Riordan flexor digitorum superficialis (FDS) transfer for correction of clawing. The middle finger flexor digitorum superficialis is split into four slips. The slips are passed volar to the deep transverse metacarpal ligament. If the Bouvier’s test is negative, the slips are sewn to a lateral band (B). If the Bouvier’s test is positive, the slips are sewn to the flexor tendon sheath (C).
The Riordan modification involves using only the FDS to the middle finger and splitting it into four slips. Omer considers this the standard technique (Fig. 12) (16). The surgical procedure is similar to that described previously.
Wrist Extensor Transfer for Correction of Clawing
When restoration of a strong power grip is as important as correction of clawing, a wrist motor should be used as a tendon transfer. The ECRL, ECRB, FCR, and brachioradialis are all available in an isolated ulnar nerve palsy and can be extremely powerful motors. In addition, the FDS tendons, which serve as weak flexors, do not have to be sacrificed.
All of the wrist motor transfers require a free tendon graft. Donors include the palmaris longus, plantaris, or toe extensors.
Extensor Carpi Radialis Brevis and Extensor Carpi Radialis Longus
Brand describes a dorsal technique using the ECRB tendon (16,58). Through a dorsal incision, the ECRB is released from its origin. Brand describes using a plantaris tendon, but the palmaris longus may be used. The tendon grafts are secured and passed superficial to the dorsal carpal ligament, through the intermetacarpal spaces, through the lumbrical canal volar to the deep transverse metacarpal ligament, and attached to the radial lateral bands of the middle ring and small fingers and the ulnar lateral band of the index finger. Brand believed that a stronger pinch could be obtained with the index finger stabilized in adduction.
Brand also describes using the ECRL and bringing it volar to the wrist by passing it under the brachioradialis tendon. It is then elongated with tendon grafts, with the grafts being passed through the carpal canal and sutured similarly to the FDS transfers previously described. Omer notes that this may cause a wrist flexion deformity and potentially crowd the carpal canal (16).
Flexor Carpi Radialis
Riordan believed that the FCR could be used as a motor for correction of claw finger and improvement in grip strength (54). He passed the FCR to the dorsal forearm and then proceeded with a technique similar to Brand’s dorsal ECRB transfer. The FCR can also be used as a volar motor similar to Brand’s volar ECRL transfer (51).
Internal Splint Technique
Omer Superficialis Y Technique
Omer describes an internal splint technique that can improve integration of MCP and IP flexion, key pinch for the thumb, and the flattened metacarpal arch (16). The superficialis Y technique involves using a single superficialis tendon combined with thumb MCP fusion. It combines
P.988

many of the basic concepts in ulnar nerve reconstruction. The superficialis of the ring finger is used in low ulnar nerve palsy and the superficialis of the middle finger in high ulnar nerve palsy. Thumb MCP fusion is done through a dorsal incision using the surgeon’s preferred technique. This fusion improves distal stability for tip pinch. Omer describes harvesting the FDS tendon through a volar Bruner incision. The radial half of the tendon is tenodesed to prevent hyperextension. The ulnar half is released at its insertion. The tendon is split first in half, and then the ulnar half is again divided in half. The radial half is used to restore thumb pinch. It is passed deep to the flexor tendons and neurovascular structures and sutured to the insertion of the abductor pollicis. The pulley for this transfer is the distal edge of the palmar fascia. The ulnar half of the FDS is used as a Zancolli lasso if strength is desired and if Bouvier’s test is positive. Omer loops the tendon around the A2 pulley instead of around the A1 pulley. Otherwise, the FDS is used as a Stiles-Bunnel–type transfer inserting in the dorsal extensor apparatus if the Bouvier’s test is negative. Tension is set with the wrist in neutral and the hand supinated. The MCP joints of the clawed fingers are placed in 45 degrees of flexion with the PIP joints in 0 degrees’ extension. The thumb is adducted so it is parallel to the second metacarpal in an anterior-posterior projection. This position is maintained for 4 weeks until active extension is allowed.
FIGURE 13. Extensor digiti minimi (EDM) transfer for correction of small finger abduction (A). The transfer is sutured to the radial collateral ligaments if no clawing is present (B). FDS, flexor digitorum superficialis.
This is not considered a definitive procedure because it divides the strength of a single tendon into multiple insertions. This technique is used as an internal splint after ulnar nerve repair to allow for function during nerve regeneration.
Small Finger Abduction
Wartenberg’s sign results from paralysis of the third volar interosseous muscle, resulting in unopposed pull of the extensor digit minimi (EDM). Burge believed that the deformity also may result from reinnervation of the abductor digiti minimi but not the third volar interosseous (68). Numerous transfers have been described to correct this deformity, most involving transfer of the EDM to the radial side of the finger (68,69,70,71,72,73,74 and 75). Either the entire EDM or half of the EDM can be used. If the entire EDM is to be used, a functional extensor digitorum communis to the fifth finger must be present (76,77).
Fowler described correction of this deformity and based it on the presence or absence of clawing (77). If clawing is not present, the ulnar slip of the EDM can be released and transferred to the radial collateral ligament of the small finger MCP joint. If clawing is present, it can be transferred more distally to the radial proximal phalanx of the small finger. The tendon is harvested through a dorsal incision and dissected proximally to the dorsal retinaculum. A volar incision is then made, and the tendon is passed through the fourth and fifth metacarpal space into the volar wound. This allows the transfer to pass volar the intermetacarpal ligament. The tendon is tensioned with the wrist in neutral and the MCP joint flexed 20 degrees (Fig. 13).
REFERENCES
1. Hollonshead WH. Anatomy for surgeons, 3rd ed. Philadelphia: Harper & Row, 1982.
2. Leibovic SJ, Hastings H 2nd. Martin-Gruber revisited. J Hand Surg [Am] 1992;17:47–53.
3. Kaplan EB, Spinner M. Normal and anomalous innervation patterns in the upper extremity. In: Omer GE Jr., Spiner M, ed. Management of peripheral nerve problems. Philadelphia: WB Saunders, 1980:75–99.
4. Earle AS, Vlastou C. Crossed fingers and other tests of ulnar nerve motor function. J Hand Surg [Am] 1980;5:560–565.
5. Masse L. Contribution à l’ étude de l’ achon des inter-osseous. J Med (Bordeaux) 1916;46:198.
P.989

6. Wartenberg R. A sign of ulnar palsy. JAMA 1939;112:1688.
7. Mannerfett L. Studies on the hand in ulnar nerve paralysis. Acta Orthop Scand (Suppl) 1966;87:4.
8. Duchenne GB. Physiology of motion. Kaplan EB (trans). Philadelphia: JB Lippincott Co, 1949:141–154.
9. Bouvier. Note sur une paralysie partielle des muscles de la main. Bull Acad Nat Med (Paris) 1851;18:125.
10. Andre-Thomas T. Le tonus du poignet dans la paralysie du nerf cubital. Paris Med 1917;25:473.
11. Hastings H 2nd, Davidson S. Tendon transfers for ulnar nerve palsy. Evaluation of results and practical treatment considerations. Hand Clin 1988;4:167–178.
12. Bowden REM, Napier JR. The assessment of hand functions after peripheral nerve injuries. J Bone Joint Surg Br 1961;43:481–492.
13. Froment J. La paralysie de l’adducteur du pouce et le signe de la prehension. Rev Neurol 1914;28:1236.
14. Brown PW. Reconstruction for pinch in ulnar intrinsic palsy. Orthop Clin North Am 1974;2:289.
15. Jeanne M. La deformation du pouce dans la paralysie cubitale. Bull Med Soc Chir 1915;41:703.
16. Omer GE. Ulnar nerve palsy. In: Green DP, Hotchkiss WC, Pederson WC, eds. Green’s operative hand surgery, 4th ed. New York: Churchill Livingstone, 1999:1526–1541.
17. Rowntree T. Anomalous Innervation of the hand muscles. J Bone Joint Surg Br 1949;31:505.
18. Omer GE. Evaluation and reconstruction of the forearm and hand after acute traumatic peripheral nerve injuries. J Bone Joint Surg Am 1968;50:1454.
19. Brown RE, Zamboni WA, Zook EG, et al. Evaluation and management of upper extremity neuropathies in Charcot-Marie-Tooth disease. J Hand Surg [Am] 1992;17:523–530.
20. Omer GE. Timing of tendon transfers in peripheral nerve injury. Hand Clin 1988;4:317–322.
21. Trevett MC, Tuson C, de Jager LT, et al. The functional results of ulnar nerve repair. Defining the indications for tendon transfer. J Hand Surg [Br] 1995;20:444–446.
22. Smith RJ. Indications for tendon transfers to the hand. Hand Clin 1986;2:235–237.
23. Omer GE. Tendon transfers in combined nerve lesions. Orthop Clin North Am 1974;5:377–387.
24. Smith RJ. Tendon transfers of the hand and forearm. Boston: Little, Brown and Company, 1987:85–133.
25. Smith RJ. Extensor carpi radialis brevis tendon transfer for thumb adduction—a study of power pinch. J Hand Surg [Am] 1983;8:4–15.
26. Zancolli EA. Structural and dynamic basis of hand surgery, 2nd ed. Philadelphia: JB Lippincott Co, 1978:159–206.
27. Thompson TC. A modified operation for opponens paralysis. J Bone Joint Surg 1942;24:632.
28. Bruser P. Motor replacement operations in chronic ulnar nerve paralysis. Orthopade 1997;26:690–695.
29. Alnot JY, Masquelet A. Restoration of thumb-index pinch by a double tendon transfer following ulnar nerve paralysis. Ann Chir Main 1983;2:202–210.
30. Robinson D, Aghasi MK, Halperin N. Restoration of pinch in ulnar nerve palsy by transfer of split extensor digiti minimi and extensor indicis. J Hand Surg [Br] 1992;17:622–624.
31. De Abreu LB. Early restoration of pinch grip after ulnar nerve repair and tendon transfer. J Hand Surg [Br] 1989;14:309–314.
32. Neviaser RJ, Wilson JN, Gardner MM. Abductor pollicis longus transfer for replacement of first dorsal interosseous. J Hand Surg [Am] 1980;5:53–57.
33. Omer GE Jr. Reconstruction of a balanced thumb through tendon transfers. Clin Orthop 1985;195:104–116.
34. Zweig J, Rosenthal S, Burns H. Transfer of the extensor digiti quinti to restore pinch in ulnar palsy of the hand. J Bone Joint Surg Am 1972;54:51–59.
35. Reference deleted.
36. Edgerton MT, Brand PW. Restoration of abduction and adduction to the unstable thumb in median and ulnar paralysis. Plast Reconstr Surg 1965;36:150–164.
37. Littler JW. Tendon transfers and arthrodesis in combined median and ulnar nerve palsies. J Bone Joint Surg Am 1949;31:225–234.
38. North ER, Littler JW. Transferring the flexor superficialis tendon: technical considerations in the prevention of proximal interphalangeal joint instability. J Hand Surg [Am] 1980;5:498–501.
39. Hamlin C, Littler JW. Restoration of power pinch. J Hand Surg [Am] 1980;5:396–401.
40. Boyes JH. Bunnell’s surgery of the hand, 5th ed. Philadelphia: JB Lippincott Co, 1970:366.
41. Goldner JL. Replacement of the function of the paralyzed adductor pollicis with the flexor digitorum sublimis—a ten year review. Proceedings of the American Society for Surgery of the Hand. J Bone Joint Surg Am 1967;49:583–584.
42. Bunnel S. Surgery of the hand. Philadelphia: JB Lippincott Co, 1944.
43. Bruner JM. Tendon transfer to restore abduction of the index finger using the extensor pollicis brevis. Plast Reconstr Surg 1948;3:197–201.
44. Hirayama T, Atsuta Y, Takemitsu Y. Palmaris longus transfer for replacement of the first dorsal interosseous. J Hand Surg [Br] 1986;11:84–86.
45. Mikhail IK. Bone block operation for clawhand. Surg Gynecol Obstet 1964;118:1077–1079.
46. Bunnell S. Surgery of the intrinsic muscles of the hand other than those producing opposition of the thumb. J Bone Joint Surg 1942;24:1.
47. Zancolli EA. Claw hand caused by paralysis of the intrinsic muscles. A simple surgical procedure for its correction. J Bone Joint Surg Am 1957;39:1076–1080.
48. Leddy JP, Stark HH, Ashworth CR, et al. Capsulodesis and pulley advancement for the correction of claw-finger deformity. J Bone Joint Surg Am 1972;54:1465–1471.
49. Brown PW. Zancolli capsulorrhaphy for ulnar claw hand. J Bone Joint Surg Am 1970;52:868–877.
50. Zancolli EA. Claw hand caused by paralysis of the intrinsic muscles: a simple surgical procedure for its correction. J Bone Joint Surg Am 1957;37:1076.
51. Riordan DC. Intrinsic paralysis of the hand. Bull Hosp Jt Dis Orthop Inst 1984;44:435–441.
52. Riordan DC. Rehabilitation and re-education in tendon transfers. Orthop Clin North Am 1974;5:445–449.
53. Brandsma JW, Brand PW. Claw-finger correction. Considerations in choice of technique. J Hand Surg [Br] 1992;17:615–621.
54. Riordan DC. Tendon transplantation in median nerve and ulnar nerve paralysis. J Bone Joint Surg Am 1953;35:312.
P.990

55. Stiles HJ, Forreseter-Brown MF. Treatment of injuries of the spinal peripheral nerves. London: Fronde & Hodder-Stoughton, 1922:166.
56. Brand PW. Biomechanics of tendon transfers. Hand Clin 1988;4:137–154.
57. Brand PW, Beach RB, Thompson DE. Relative tension and potential excursion of muscles in the forearm and hand. J Hand Surg [Am] 1981;6:209–219.
58. Brand PW. Paralytic claw hand. J Bone Joint Surg Br 1958;40:618.
59. Parkes A. Paralytic claw fingers—a graft tenodesis operation. Hand 1973;5:192.
60. Smith RJ. Metacarpal ligament sling tenodesis. Bull Hosp Jt Dis 1984;44:466.
61. Brooks AL. A new intrinsic tendon transfer for the paralytic hand. J Bone Joint Surg Am 1975;57:730.
62. Hastings H 2nd, McCollam SM. Flexor digitorum superficialis lasso tendon transfer in isolated ulnar nerve palsy: a functional evaluation. J Hand Surg [Am] 199419:275–280.
63. Brandsma JW, Ottenhoff-De Jonge MW. Flexor digitorum superficialis tendon transfer for intrinsic replacement. Long-term results and the effect on donor fingers. J Hand Surg [Br] 1992;17:625–628.
64. Shah A. Correction of ulnar claw hand by a loop of flexor digitorum superficialis motor for lumbrical replacement. J Hand Surg [Br] 1984;9:131–133.
65. Enna CD, Riordan DC. The Fowler procedure for correction of the paralytic claw hand. Plast Reconstr Surg 1973;52:352–360.
66. Burkhalter WE, Carneiro RS. Correction of the attritional boutonniere deformity in high ulnar-nerve paralysis. J Bone Joint Surg Am 1979;61:131–134.
67. Anita NH. The palmaris longus motor for lumbrical replacement. Hand 1969;2:139.
68. Burge P. Abducted little finger in low ulnar nerve palsy. J Hand Surg [Br] 1986;11:234–236.
69. Bellan N, Belkhiria F, Touam C, et al. Extensor digiti minimi tendon “rerouting” transfer in permanent abduction of the little finger. Chir Main 1998;17:325–333.
70. Voche P, Merle M. Wartenberg’s sign. A new method of surgical correction. J Hand Surg [Br] 1995;20:49–52.
71. Hoch J. Correction of post-traumatic adduction insufficiency of the small finger by transposition of the tendon of the extensor indicis muscle. Handchir Mikrochir Plast Chir 1993;25:179–183.
72. Dellon AL. Extensor digiti minimi tendon transfer to correct abducted small finger in ulnar dysfunction. J Hand Surg [Am] 1991;16:819–823.
73. Goloborod—ko SA, Andruson MV, Goridova LD. Method of correcting abduction of the 5th finger after injuries to the ulnar nerve. Ortop Travmatol Protez 1985;11:48–50.
74. Blacker GJ, Lister GD, Kleinert HE. The abducted little finger in low ulnar nerve palsy. J Hand Surg [Am] 1976;1:190–196.
75. Burge P. Abducted little finger in low ulnar nerve palsy. J Hand Surg [Br] 1986;11:234–236.
76. Gonzalez MH, Gray T, Ortinau E, et al. The extensor tendons to the little finger: an anatomic study. J Hand Surg [Am] 1995;20:844–847.
77. Fowler SB. Extensor apparatus of digits. J Bone Joint Surg Br 1949;31:477.