|3475 Torrance Blvd., Ste F
Torrance, CA 90503
Office hours: M-F, 9am
– 5 pm
Worker’s compensation, Medicare
and most insurance plans accepted.
In September 2008, we will be moving to a beautiful new 4,400 sq ft,
state of the art dedicated hand center located at 2808 Columbia Ave
in Torrance, CA., which will feature onsite nerve conduction studies,
occupational hand therapy and digital x-ray.
NERVE TRANSFERS TO RESTORE
WRIST AND FINGER EXTENSION
A loss of wrist and finger extension can result from myriad causes including
brachial plexus injury, radial nerve injury due to trauma, nerve tumors,
compression or idiopathic neuritis as well as posterior interosseous nerve
(PIN) compression. In the latter case some radial wrist extension may
be preserved. Until recently, tendon transfers have been the mainstay
of treatment. Nerve transfers have long been used for brachial plexus
reconstruction. The principals of nerve to nerve transfer, or neurotization
have recently been applied to peripheral nerve injuries with encouraging
results. The following discussion will center on neurotization to restore
wrist and finger extension.
The radial nerve arises from the posterior cord of the brachial plexus.
It receives contributions from C5 - C8 spinal roots. The nerve contains
approximately 16,000 myelinated fibers. It runs medial to the axillary
artery then at the level of the coracobrachialis it courses posteriorly
to lie in the spiral groove of the humerus. In the lower arm it pierces
the lateral intermuscular septum to run between the brachialis and brachioradialis.
Opposite the head of the radius there are some fibrous bands from the
joint’s capsule and immediately distal to this, the nerve is regularly
crossed by several prominent veins, the “leash of Henry.”
It divides 2 cm distal to the elbow into a superficial radial sensory
branch (SRN) and a deep motor branch, the posterior interosseous nerve
(PIN). It gives off branches to the extensor carpi radialis longus (ECRL)
and brevis (ECRB), brachioradialis (BR) and anconeus before giving off
the PIN branch. The PIN continues on between the superficial and deep
head of the supinator muscle, to exit on the dorsal forearm. After it
emerges from the distal border of the supinator, the PIN sends branches
to the extensor digitorum communis (EDC), extensor carpi ulnaris (ECU)
, extensor digiti quinti, extensor pollicus longus(EPL) and brevis and
the extensor indicis proprius (EIP) in descending order, although there
may be considerable variation.
Classification of Nerve Injury
In the Seddon classification, there are 3 types of nerve injury: neuropraxia,
axonotmesis and neurotmesis.1 Sunderland subclassified these injuries
into 6 types.2 Axonotmesis was divided into axonal injuries with or without
an intact basil lamina (2nd and 3rd degree) or with complete scar block
(4th degree). Neurotmesis was divided into those with complete transection
(5th degree) and a combination of conduction block and transection (6th
For the purposes of nerve transfer, Mackinnon grouped these injuries into
three categories: (1) injuries that will recover spontaneously (1st and
2nd degree), (2) Injuries that must be repaired (5th degree or neurotmesis)
and (3) Injuries that have partial recovery and will likely need surgery
(3rd, 4th and 6th degree).
3 Types of Radial neuropathies
In the arm region, the radial nerve is often injured in association with
some form of unconsciousness. In a Saturday night palsy, an obtunded patient
sits with their arm over a chair back or rests his/her head on the lateral
surface of their arm. Alternatively the radial nerve can be compressed
in the groove between the brachialis and forearm muscles when one person
rests their head on the middle third of the arm of another i.e. Honeymooner’s
palsy. The patient will typically present with a wrist drop and an inability
to extend the fingers, thumb or wrist. In addition, the brachioradialis
will be affected along with variable involvement of the triceps. They
will also have diminished sensation over the dorsum of the 1st web space.
The NCS typically demonstrates the absence of the superficial radial sensory
nerve action potential (SNAP). Motor recordings are more difficult since
no muscle is sufficiently isolated from other radially innervated muscles.
A surface electrode over the extensor indicus proprius (EIP) results in
a volume conducted response from the adjacent radial innervated muscles,
which makes side to side amplitude comparisons difficult. Radial nerve
recordings using needle electrodes in the EIP are more common as a result,
which makes it difficult to approximate the degree of axonal loss by assessing
the amplitudes. The EMG however is quite useful, and permits a relatively
accurate localization of the lesion. In a spiral groove lesion for example,
all 3 heads of the triceps should be normal, with denervation of the brachioradialis
and all muscles distal to it.
Posterior interosseous nerve entrapment
In posterior interosseous nerve syndrome the presenting symptoms are weakness
and/or paralysis of the extensor muscles, which result in a wrist or finger
drop. There may be a history of a fall onto an extended and pronated arm
although many cases are spontaneous, especially if due to an underlying
lipoma, ganglion or rheumatoid nodule arising from the radiocapitellar
joint. The patient will prevent with variable weakness or paralysis of
the EPL, EIP, EDC and ECU. Motor function of the ECRB/L should be preserved
since they are innervated before the PIN dives between the two heads of
the supinator muscle. The patient will hence extend their wrist in radial
deviation. As it travels distally through the radial tunnel the PIN may
potentially be entrapped by fibrous bands anterior to the radiocapitellar
joint, the radial recurrent leash of vessels, the fibrous edge of the
ECRB, the proximal border of the supinator i.e. the arcade of Frohse or
the distal edge of the supinator muscle.
PIN lesions do not affect the superficial radial SNAP, which should be
normal. The compound motor action potential of PIN innervated muscles
may show a drop of conduction velocity or amplitude, but this is difficult
to assess with surface electrodes. Needle EMG is the best technique for
localization, especially with partial lesions. In acute denervation decreased
recruitment, increased insertional activity and fibrillation potentials
" positive sharp waves are present. In chronic lesions seen after
3-6 months, decreased recruitment may still be seen along with giant motor
unit potentials and polyphasia due to peripheral axonal ingrowth.
Indications for Nerve Transfers
The time for reinnervation must take the distance from the injury to the
motor endplate into account. As a general rule, motor endplates degrade
at about 1% per week. Nerve growth is limited to 1 inch/month or 1 to
1.5 mm/day.4 By this reckoning a nerve will have regenerated 12 inches
at 1 year, but 50% of the endplates will be gone. The maximum length that
a nerve can grow to restore motor function is hence approximately 13 -
18 inches due to a critical loss of endplates. Proximal radial nerve repairs
typically demonstrate superior results as compared to those for the median
or ulnar nerves since they are closer to the endplate. Nerve to nerve
transfers should therefore be considered in cases of delayed treatment,
where the time for nerve ingrowth would exceed this window of time for
reinnervation. Proximal radial nerve injuries that are 8-10 months old
would be an ideal indication for neurotization for this reason. Brachial
plexus lesions and traction injuries that show no EMG evidence of recovery
by 3 months may also be appropriate candidates.
Proximal radial nerve lesions that are < 6 months old and PIN lesions
< 1 year old are best treated with conventional methods of nerve repair
or graft. Concomitant median nerve injuries would preclude use of this
Nerve Transfers Options
A number of nerve transfers to restore wrist and finger extension have
been described. Palazzi and coworkers reported on the use of the motor
branch to the brachialis muscle for neurotization of the radial nerve
for a C7-C8-T1 avulsion.5 Tubbs et al performed an anatomic study on the
anterior interosseous nerve and described a transinterosseous membrane
tunnelling technique for transfer to the PIN.6 Ustun et al performed a
cadaver study of the widths and lengths of the median motor nerve branches
from their points of divergence. They found that the motor branches to
the pronator teres, flexor pollicus longus and pronator quadratus were
sufficiently long to permit neurotization of the PIN at different levels
and in various combinations. The motor branches to the flexor digitorum
profundus muscle were too short to use for transfer.7 None of these transfers
have enjoyed widespread clinical use however.
Mackinnon et al have demonstrated that redundant median nerve branches
to the FDS, FCR or PL may be used for transfer.8 They described two successful
cases of transfer of the FDS fascicles to the pronator teres branch to
restore pronation.9 They subsequently reported the use of this transfer
to the ECRB branch and the PIN in a case of a high radial nerve palsy
and a brachial plexopathy with excellent results. 10 In a separate study
they noted that the FCU branch of ulnar nerve can also be used, but this
required a second incision.11, 12
Relevant Median Nerve Anatomy
Just distal to the cubital fossa, the motor branches of the median nerve
consistently collect into 3 fascicular groups.13 There is an anterior
group (to the PT and FCR), a middle group (motor to the FDS and hand intrinsics,
sensory to the thumb, index and middle fingers) and a posterior group
(to the AIN branch). These branch groups can be traced proximally without
harm, within the main trunk of the median nerve for 2.5 to 10 cm. The
nerve and artery pass through the antecubital fossa underneath the lacertus
fibrosis and gives off branches to the palmaris longus (PL), flexor carpi
radialis (FCR), flexor digitorum superficialis (FDS), and rarely the flexor
digitorum profundus (FDP). The nerve then dives between the deep and superficial
heads of the PT to which it supplies 1-4 branches. The fibrous arch of
the PT lies 3 to 7.5 cm below the humeral epicondylar line. The fibrous
arch of the superficialis arch lies 6.5 cm below the humeral epicondylar
line. The median nerve enters the forearm deep to the fibrous arch of
the FDS and emerges beneath the radial side of the muscle belly of the
middle finger superficialis where it is quite superficial and near the
palmaris longus tendon.
In a dissection of 31 cadaver arms, Tung and MacKinnon noted that double
innervation of the FDS was found in 94% of the specimens.9 The most common
branching pattern was a proximal branch that also carried the branch to
the PL, and a distal branch a distal branch that arose from the median
nerve distal to the origin of the AIN branch. The proximal branch arose
3.1 cm " 1.3 cm distal to the medial epicondyle, and was 2.1 "
0.7 cm long. The distal branch arose 7.4 cm " 2.7 cm distal to the
medial epicondyle, and was 2.3 " 0.8 cm long.
Surgical technique (link to FDS transfer to PIN video)
The patient is positioned supine with the arm abducted on an arm board.
The procedure is performed under tourniquet control. The tourniquet time
is limited to 1 hour and must be released at least 20 minutes before stimulating
the nerve. If intraoperative nerve conduction is performed Halothane and
muscle relaxants are avoided since they extinguish the nerve response.
The nerve should be irrigated with warm saline, since cold nerves do not
conduct. A short acting agent such as Fentanyl can be used after infiltration
of the wound margins with local anesthetic.
An anteromedial incision is made crossing the antecubital fossa, starting
5 cm above the elbow flexion crease. The medial antebrachial cutaneous
nerve is identified next to the basilic vein and protected. Proximal to
the elbow, the median nerve can be found medial to the brachial artery.
The nerve is followed distally as it passes through the two heads of the
PT. At this level the branches arising from the anterior bundle can be
seen innervating the FCR and PT. Both the superficial head of the PT and
the FDS can be divided and tagged for later repair if needed. The FDS
branches along with the FCR and PL fascicles arise from the middle group.
They are carefully differentiated from the hand intrinsics using the hand
held stimulator. The intrinsic fascicles and sensory fascicles travel
distally in the middle group along with the posterior (AIN branch) group
which pass underneath the sublimus arch. This arch is divided. The 2nd
or distal group of FDS fascicles may arise distal to the sublimus arch.
At this point it is helpful to place vessel loops around the branches
of interest. A disposable 2 mA nerve stimulator is used to sequentially
stimulate the branches. Careful observation of the hand will allow one
to discern the individual motor branches by their corresponding muscle
twitch (see FDS nerve transfer video). The sensory fascicles can be identified
by placed recording ring electrodes on the index and middle fingers, since
stimulation will not elicit a muscle twitch, but this is usually unnecessary.
The radial nerve is isolated through the same incision. It can be found
between the brachioradialis and brachialis as it divides into the superficial
sensory nerve branch and the PIN branch. In PIN palsies with preserved
radial wrist extension, the PIN is divided distal to the motor branch
of the ECRB (Figure 1 A - G). In proximal radial nerve injuries the ECRB
branch is divided and included in the transfer (Figure 2 A - E). These
branches are stimulated to confirm that there is total extensor muscle
denervation prior to nerve division. An intraoperative EMG can be performed
if necessary by placing the needle recording electrode in the ECRB or
extensor muscle mass with the reference electrode in the subcutaneous
tissue. The presence of a compound motor action potential might alter
the prognosis for recovery and require modification of the procedure to
a neurolysis. Alternatively, a nerve action potential can be sought by
stimulating and recording across a 4 cm segment of the PIN. Kline has
estimated that approximately 4,000 myelinated axons are needed to produce
a nerve action potential.14
MacKinnon recommends use of the PL and the FDS or FCR branches to the
PIN and the ECRB branch. She also states that the strongest donor should
be transferred to the ECRB branch to restore strong wrist rather than
strong finger extension for maximum function.3 The redundant median nerve
fascicles are harvested in close proximity to the PIN to avoid undue tension
on the repair site. If need be a short nerve graft can be interposed.
Similar to nerve repair, there should be < 8% strain on the repair
site.15 For practical purposes, if the repair can withstand the tension
during passive elbow motion intraoperatively then a graft is not needed.
The anastomosis is performed with 9-0 nylon since the fascicles are typically
1-3 mm in diameter.
An above elbow splint is applied with the elbow at 90 degrees and the
shoulder, wrist, and fingers free. Gentle elbow flexion is started after
the first week followed by gradual elbow extension. Motor retraining is
akin to tendon transfers. The patient is instructed in active sublimus
contractions, which will ultimately produce wrist and finger extension.
Published clinical series on this transfer are still lacking. In Mackinnon’s
series a 51 y.o. male with a proxiimal left radial nerve palsy underwent
transfer of the nerve branch to the PL and the FDS to the PIN and ECRB
branch respectively. He achieved 4/5 power for wrist extension but lacked
simultaneous finger extension at 14 months postoperatively. The second
patient was a 24 y.o. female following an iatrogenic radial nerve injury
at the plexus level. The nerve branch from the PL and FDS were transferred
to the PIN. A 2nd FDS branch was transferred to the ECRB. The patient
achieved 4/5 power of wrist and finger extension.10
My own experience is limited to two cases. The 43 y.o. male with a high
radial nerve palsy featured in the accompanying video underwent transfer
of an FDS branch to the PIN proximal to the ECRB branch. By 9 months he
had achieved 4/5 power of wrist extension but still lacked independent
finger extension. The second case of the 26 y.o. male with a PIN palsy
did not achieve any appreciable wrist extension at 6 months followup.
Because of the history of a 10 year old nonunion of the medial epicondyle,
and the intraoperative findings of complete denervation of the supinator
muscle it was uncertain as to whether this represented a much longer standing
injury than reported by the patient, and may have contributed to the failure.
It is evident that nerve transfers to restore wrist and finger extension
are a viable alternative to tendon transfers when the time for standard
nerve repair or grafting has passed. The critical number of axons that
need to be transferred are still unknown, but from the extensive work
by Dr. MacKinnon, it seems reasonable to perform separate transfers to
the PIN and ECRB branch in cases where both finger and wrist extension
Figure. 1 Posterior Interosseous Nerve Palsy
A. 26 y.o. male with an 8 month old idiopathic PIN palsy. Note the radial
wrist extension due to preservation of the ECRB branch.
B. View of the median nerve with exposure of the FCR branch, the PT branch
and the middle fascicular group (*) passing under the fibrous sublimus
C. Labeling of the median motor nerve branches after nerve stimulation.
D. Close up of the median nerve with the middle fascicular group (*) and
the AIN branch, demonstrating its proximity to the PIN.
E. Harvesting of the proximal FDS branch arising from the median nerve.
F. View of the PIN as it runs beneath the brachioradialis with the overlying
radial leash of vessels (*).
G. After the completed transfer (*) which has been passed underneath the
brachial artery. MN = median nerve Figure 2. High Radial Nerve Palsy
A. 10 month old malunion of the distal humerus with a secondary high radial
B. Dissection of the median nerve with exposure of the AIN and the FDS
C. Relative positions of the median nerve (MN) and PIN (*)
D. View of the FDS transfer just prior to anastomosis
E. Completed transfer
1. Seddon H. Nerve Injuries. Med Bull (Ann Arbor) 1965: 31: 4-10.
2. Sunderland S. The anatomy and physiology of nerve injury. Muscle Nerve
1990: 13: 771-84.
3. Mackinnon SE WR. Upper Extremity Nerve Transfers. In Slutsky DJ HV,
ed. Peripheral Nerve Surgery: Practical Applications in the Upper Extremity.
Philadelphia: Elsevier, Inc., 2006: 89-108.
4. Seddon HJ, Medawar PB, Smith H. Rate of regeneration of peripheral
nerves in man. J Physiol 1943: 102: 191-215.
5. Palazzi S, Palazzi JL, Caceres JP. Neurotization with the brachialis
muscle motor nerve. Microsurgery 2006: 26: 330-3.
6. Tubbs RS, Custis JW, Salter EG, et al. Quantitation of and superficial
surgical landmarks for the anterior interosseous nerve. J Neurosurg 2006:
7. Ustun ME, Ogun TC, Buyukmumcu M. Neurotization as an alternative for
restoring finger and wrist extension. J Neurosurg 2001: 94: 795-8.
8. Nath RK, Mackinnon SE. Nerve transfers in the upper extremity. Hand
Clin 2000: 16: 131-9, ix.
9. Tung TH, Mackinnon SE. Flexor digitorum superficialis nerve transfer
to restore pronation: two case reports and anatomic study. J Hand Surg
[Am] 2001: 26: 1065-72.
10. Lowe JB, 3rd, Tung TR, Mackinnon SE. New surgical option for radial
nerve paralysis. Plast Reconstr Surg 2002: 110: 836-43.
11. Mackinnon SE, Novak CB. Nerve transfers. New options for reconstruction
following nerve injury. Hand Clin 1999: 15: 643-66, ix.
12. Boutros S, Nath RK, Yuksel E, Weinfeld AB, Mackinnon SE. Transfer
of flexor carpi ulnaris branch of the ulnar nerve to the pronator teres
nerve: histomorphometric analysis. J Reconstr Microsurg 1999: 15: 119-22.
13. Zhao X, Lao J, Hung LK, et al. Selective neurotization of the median
nerve in the arm to treat brachial plexus palsy. An anatomic study and
case report. J Bone Joint Surg Am 2004: 86-A: 736-42.
14. Kline DG. Surgical repair of peripheral nerve injury. Muscle Nerve
1990: 13: 843-52.
15. 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: 677-87.