SECONDARY END-TO-END REPAIR OF EXTENSIVE FACIAL

NERVE DEFECTS: SURGICAL TECHNIQUE AND

POSTOPERATIVE FUNCTIONAL RESULTS
Hildegunde Piza-Katzer, MD,1 Brigitta Balogh, MD,1 Emilia Muzika-Herczeg, MD,2
Alexander Gardetto, MD1
1

Department of Plastic and Reconstructive Surgery, Innsbruck University Hospital, Anichstrasse 35,
A-6020 Innsbruck, Austria and Ludwig Boltzmann-Institute for Quality Control in Plastic Surgery, Austria.
E-mail: alexander.gardetto@uibk.ac.at
2
Department of Plastic and Reconstructive Surgery, Hospital of Lainz, Vienna, Austria
Accepted 13 February 2004
Published online 6 August 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/hed.20047

Abstract: Background. Repair of the transected facial nerve

is imperative for restoration of muscle function, including the ability
to produce appropriate facial expressions. Injury might involve the
main trunk and its several branches. Restoration of function presupposes meticulous repair of all injured nerve branches.
Methods. Here we report three cases of secondary tensionfree end-to-end coaptation of a transected trunk and branches
of the facial nerve by removal of the superficial part of the parotid gland.
Results. Facial tone and symmetry at rest and motion were
achieved. In two patients, a slight residual synkinesis is observed
under stress.
Conclusions. Direct end-to-end coaptation of the facial nerve
and its branches by the technique described should be considered
before deciding on grafts or rerouting procedures to deal with gaps
of up to 15 mm. This technique is not recommended in the presence of infection and nerve defects. Intensive postoperative
physiotherapy is required for optimal results. A 2004 Wiley Periodicals, Inc. Head Neck 26: 770 777, 2004
Keywords: facial nerve; nerve lengthening; parotid gland; direct
nerve coaptation; surgical repair

Correspondence to: A. Gardetto

B 2004 Wiley Periodicals, Inc.

770

Secondary Facial Nerve End-to-End Repair

Facial nerve injury resulting in facial paralysis

and loss of mimetic movement is a social handicap
that often leads to severe psychological and economic hardships.1 An understanding of the anatomy and physiology of the facial nerve and the
impact of injury to this nerve on the production of
appropriate facial expressions and a knowledge of
surgical repair techniques are prerequisites for
both the proper evaluation and the treatment of
this problem.2 Peripheral facial nerve defects are
mainly caused by trauma or occur after tumor
resection. Iatrogenic nerve injuries, although not
very frequent, also occur. In contrast to idiopathic
facial palsy, if paralysis is the result of injury to
the peripheral facial nerve, surgery can often
restore at least partial function.
Although the prognosis for facial nerve regeneration after direct repair is better than that for
other motor nerves,3 there are several primary
factors that directly affect outcome: proximity of
the injury to the cell body4,5; type of neural injury5; timing; type of facial nerve repair; and a
host of secondary factors including age,6 nutrition,
HEAD & NECK

September 2004

and related illness. Of these prognosticators, the

surgeon influences only timing and type of repair.
Recent evidence7 suggests that end-to-end repair of the facial nerve with or without rerouting
is functionally and cosmetically superior to grafting, but the repair should be performed immediately after injury. The single most important
prognostic factor, apart from the neurorrhaphy
technique, is tension-free coaptation of the
stumps.8 10 Most authors are of the opinion that
nerve defects of 5 to 10 mm can be overcome by
mastoid-extratemporal rerouting but that greater
defects should be repaired by interpositional grafting.11 14 Yarbrough et al,15 however, described in
their anatomic study that decompression and
mobilization of the facial nerve allows tension-free
closure of 8-mm defects. Furthermore, dissection of
the tympanic portion of bone in addition to decompression and mobilization increases the repairable
defect size to 14 mm. Retroposition of the parotid
gland adds roughly 3.5 mm to the length of facial
nerve gained with either of the preceding procedures. The most extensive mastoid-extratemporal
rerouting procedure with retroposition of the parotid gland thus allows closure of neural defects
of 17.5 mm.
We believe that these extensive rerouting procedures are not necessary. In our anatomic study,
we were able to demonstrate that gaps up to 15 mm
can be bridged by just removing the superficial
part of the parotid gland and mobilizing the
branches of the parotid plexus.16 In this article,
we present the clinical application and postoperative results of this technique in three cases of facial palsyone iatrogenic and two traumaticin
which a secondary tension-free end-to-end coaptation of the trunk of the facial nerve and branches of
the parotid plexus could be achieved by removing
the superficial part of the parotid gland.

SURGICAL TECHNIQUE

Depending on the side of interest, with the patient lying supine, the head was turned to the
lateral position. A skin incision was made as for
parotidectomy, was connected with a retroauricular incision, and was carried inferiorly onto
the mandibular angle. The advantages of this
Y-shaped skin incision are that the extratemporal
segments of the facial nerve up to the stylomastoid
foramen can be identified, and a probable skin
excess after parotidectomy can be managed like a
facelift procedure with a favorable influence on the
postoperative esthetic result (Figure 1). The skin

Secondary Facial Nerve End-to-End Repair

FIGURE 1. Schematic drawing showing the Y-shaped skin

incision.

and subcutaneous tissue were folded medially, and

after cutting the superficial musculo-aponeurotic
system (SMAS) preauricularly, the fascia and the
border of the parotid gland were identified. The
next step of the operation was done with loupe
magnification to expose the injured nerve
branchesdetermined preoperatively by electrophysiologic and the clinical examinationsin
their central and extraparotid course. The entire
superficial part of the parotid gland was then
removed, and the parotid plexus was mobilized up
to the stylomastoid foramen. It was necessary to
resect the entire superficial part of the parotid
gland to obtain an adequate length of nerve and
avoid formation of fistulas. Great care was taken
not to injure the deep part of the gland and the
parotid duct, as well as the secretory fibers to the
deep part. After the injured nerve stumps were
identified, the distal and proximal stumps were
resected until healthy nerve fibers were reached.
Epineural coaptation of the nerve stumps was
performed with 10-0 nonabsorbable sutures under
a surgical microscope. After hemostasis was controlled and drains were placed, the subcutaneous
layers were closed with absorbable sutures. The
skin was closed intracutaneously with a 6-0 nonabsorbable suture.

HEAD & NECK

September

2004

771

CASE REPORTS

Three patients with peripheral facial palsy were

admitted to our department for surgical revision (Table 1). In all patients, a direct tensionfree coaptation of the transected nerve and its
branches was performed after the superficial
part of the parotid gland was removed and
the branches of the parotid plexus were mobilized. The follow-up period was 3 to 7 years.
Good functional results were achieved in all
three patients, with some residual synkinesis in
two patients.

In 1993, a 2-year-old girl was admitted

to our department 5 days after a surgical attempt
to remove a preauricular cyst that resulted in
transaction of the main trunk of the left facial
nerve and complete left facial palsy (Figure 2a).
During surgical revision, the main trunk of the
facial nerve was exposed and found to be transected 2.5 cm distal to the stylomastoid foramen.
The distal stump of the facial nerve could not be
found directly, so we followed the zygomatic and
temporal branches from distal to proximal. By
removing the superficial part of the parotid gland,
we exposed the buccal and mandibular branches
and followed them centrally. This enabled us at
the same time to overcome the gap of about
14 mm. The proximal and distal stumps were
refreshed for nerve suture, and coaptation was
possible without any tension. Six months later,
some movements of the left mimic muscles were
perceptible, and 1 year after nerve suture, en bloc
response was noted. The electroneurography after
24 months confirmed the partial neurologic
recovery. Electrophysiologic tests were conducted
in 2000. We stimulated the facial nerve on both
sides at the stylomastoid foramen and measured
the response with use of surface electrodes. The
muscle action potential elicited from the orbicularis oculi muscle was 63% of the amplitude

Case 1.

height on the healthy side. Clinically, 7 years

after surgery, the patient has facial symmetry at
rest and in action (Figures 2b to 2d).

Case 2. A 33-year-old man had palsy of the right

side of his face 1 day after mandibular osteosynthesis (Figure 3a) for treatment of a bilateral
mandibular fracture sustained during a brawl in
1993. Ten days after surgery, the patient was
admitted to our unit for surgical revision. The
facial nerve was exposed on both sides, at the
stylomastoid foramen as well as at the border of
the parotid gland. At the area of the osteosynthesis, the zygomatic and the buccal branches
were found to have been transected. The superficial part of the parotid gland was removed, and the
nerve stumps, exhibiting a gap of 14 and 12 mm,
respectively, were coapted without tension. In addition, a ligature constricting the facial nerve at
its bifurcation was released, and epineurotomy
was performed. The patients postoperative
course was uneventful, and the wound healed
primarily. Lid closure was restored 5 months
after surgery, and 1 year later, the muscle action
potential on electromyography showed 67% of the
intensity of the left side. Today, the patient has
facial symmetry at rest and during motion and
complete lid closure (Figures 3b to 3f ). According
to the scale recommended by the British Research
Council for clinically establishing the level of
muscle function, our patient has achieved the
grade M4 with respect to restoration of facial
muscle function. Very infrequently, however,
synkinesis between the patients eye and the
orbicularis oculi muscle occurs, particularly when
he is under stress.

Case 3. An 18-year-old woman had facial palsy

on the left side caused by a traffic accident in 1998
(Figure 4a). In the first surgical procedure, severe

Table 1. Characteristics of patients who have undergone tension-free coaptation by removal of the superficial part of the parotid gland.
Patient
age, y/sex

Affected
hemiface

2/F
33/M

Left
Right

18/F

Left

772

Facial nerve injury

Total no.

Main trunk
Zygomatic branches
Buccal branch
Zygomatic branches
Buccal branches
Cervical branch

1
2
1
2
3
1

Secondary Facial Nerve End-to-End Repair

Time until
reconstruction, d

Time until return

of function, mo

5
21

6
5

43

HEAD & NECK

September 2004

FIGURE 2. (a) Two-year-old girl with left facial paralysis after facial nerve transection. (b) Seven years after repair, symmetric innervation
of the orbicularis oris muscle. (c) Good functional results of the muscles innervated by the zygomatic and buccal branches. (d) Complete
lid closure. (e) Facial symmetry at rest.

bleeding on the left side of the neck was stopped

by coagulation and ligation of the vessels: an
additional 3 weeks of intensive care was necessary because of severe polytrauma. Three months
later, the patient was admitted to our department
for surgical revision of the facial nerve. By removing masses of coagulated blood and the superficial part of the parotid gland, we found that
the cervical branch and then the buccal and the
temporozygomatic branches had been cut by a
glass splinter. All in all, under microscopic control, six branches of the left facial nerve were
coapted after the nerve stumps were refreshed
and intact nerve tissue was found. The removal of
the superficial part of the parotid gland allowed

Secondary Facial Nerve End-to-End Repair

coaptation without tension. No postoperative complications were encountered. Three months later,
the reconstructed facial branches showed a mass
reinnervation of the mimic muscles, and there was
complete lid closure. Motor conduction studies
were performed 14 months after nerve reconstruction, with stimulation being applied on the
central part of the left extracranial facial nerve at
the stylomastoid foramen while the compound
muscle action potential of the orbicularis oculi
and orbicularis oris muscles were recorded with
surface electrodes. The action potential elicited
from the orbicularis oculi muscle was 77% of
the amplitude height on the right side, and the
action potential elicited from the orbicularis oris

HEAD & NECK

September

2004

773

FIGURE 3. (a) A 33-year-old man with palsy of the right side of his face after transection of the zygomatic and buccal branches. (b) Seven
years after repair, facial symmetry at rest. (c) Symmetric innervation of the orbicularis oris muscle. (d) Good functional results of the
muscles innervated by the buccal branch. (e) Symmetry of the periorbital region. (f) Complete lid closure.

muscle was 54% of the amplitude height on the

healthy side. Three years after surgery, the patients facial tone has been restored and symmetry at rest and in motion has been achieved
(Figures 3b to 3d). The patient has some residual synkinesis.
In summary, the tone of facial muscles showed
signs of recovery within 3 to 6 months, with symmetry of the face at rest also restored. In all cases,
movement appeared first about the oral commissure and then progressed to the check, lips,
and orbit.

774

Secondary Facial Nerve End-to-End Repair

DISCUSSION

Patients with facial palsy present one of the most

difficult challenges to reconstructive surgeons.2
The ultimate goal of any nerve repair is maximum sensory and motor recovery of the distal
denervated part.9 Diagnosis of a facial nerve
injury is not always easy, because distortion,
swelling, pain, and lowered levels of consciousness frequently accompany severe facial injury.
However, a careful and complete diagnostic assessment is the precondition for the efficacy of
surgical intervention.

HEAD & NECK

September 2004

FIGURE 4. (a) An 18-year-old woman with left facial palsy after transection of the zygomatic, buccal, and cervical branches. (b) Three
years after surgery, facial symmetry at rest. (c) Incomplete lid closure (preoperative). (d) Complete lid closure (3 years postoperative).

Facial function should be thoroughly evaluated in all cases of laceration in the anatomic
distribution of the seventh nerve. It is important
to document function before surgery. In particular, electrophysiologic examinations such as direct electric stimulation of the main trunk and
the distal branches with minimal amperage
should be performed on both sides for purposes
of comparison and maximal stimulation for establishing the degree of paralysis. However,
these tests are not sensitive until 72 hours after
injury to the facial nerve. Electroneurography
measures evoke summation potentials of the
facial muscles after a nerve stimulus is applied;
responses of less than 10% to 25% of the normal
side imply severe degeneration. The level of lesion
can be precisely determined with nerve conduc-

Secondary Facial Nerve End-to-End Repair

tion velocity tests and by means of electromyography on the basis of the pattern of the injured
nerve branches.
The basic principles of facial nerve repair have
changed little since Bunnell and Martin performed the first successful infratemporal coaptation in the late 1920s.17,18 Traumatic facial
nerve injuries often require use of several technical resources currently available to operating
surgeons. The widespread use of the operating
microscope and fine suture has expanded clinical
applicability, and atraumatic techniques have
optimized surgical procedures. The results of
microsurgical reconstruction of injured nerve
branches of the past 20 to 30 years show that for
achieving functionally good results, appropriate
surgical techniques, the choice of the right point

HEAD & NECK

September

2004

775

in time for the reconstruction, and a concept of

reconstruction tailored to patients age and level
and extension of nerve lesions are the most decisive factors. How to perform surgical repair of
peripheral nerve injuries when a large defect is
present between the severed nerve ends continues to be a controversial issue. In surgical management, the following question needs to be
addressed: should one mobilize extensively and
stretch the nerve to bridge the gap, thereby
achieving end-to-end repair with a single anastomosis, or should one eliminate tension by bridging
with a nerve graft, leaving the regenerating axons
to find their way through two anastomoses?
Spector7 ranked the techniques by functional and
cosmetic results as follows: first, direct end-to-end
coaptation; second, end-to-end coaptation after rerouting the nerve; and third, interpositional nerve
grafting. As far as timing is concerned, the fundamental principle here is that direct coaptation
of the transected nerve or its branches should be
performed as soon as possible after the injury
is identified.
Yarbrough et al,15 in their anatomic study, described an extensive rerouting technique of direct
coaptation that allows tension-free closure of
neural defects of 17.5 mm. In our previous
anatomic study, we described a technique that
allows lengthening of the branches of the facial
nerve without extensive rerouting and grafting.16
This technique is less invasive, and the risk of
bone infection with all the associated complications is negligible. We were able to demonstrate
that removing the superficial part of the parotid
gland enables direct coaptation of cut branches of
the facial nerve with a gap of 15 mm, with even
greater defects managed by mobilizing not only
the proximal stumps but also the distal stumps.
Furthermore, we also showed that gaps of the
main trunk up to 17 mm could be coapted directly
when the branches were mobilized in their intraparotid and extraparotid course and drawn
toward the proximal stump of the trunk.16 After removal of the superficial part of the parotid
gland, which does not affect the function of the
deep part, the parotid plexus can be mobilized. It
is indispensable to resect the entire superficial
part of the parotid gland to avoid fistula formation.
During this procedure, the surgeon should be careful not to injure the deep part of the gland and
the parotid duct, as well as the secretory fibers to
the deep part. The remaining glandular tissue is
sufficient for the secretory function, as evidenced
in postoperative sialograms. The results of our

776

Secondary Facial Nerve End-to-End Repair

clinical study are in agreement with and confirm

the appropriateness of the surgical technique we
used in the three cases reported here.
In conclusion, our results show that secondary
end-to-end, tension-free coaptation of the transected facial nerve and its distal branches with
gaps up to 15 mm can be accomplished with good
functional results, without performing extensive
rerouting procedures, just by removing the superficial part of the parotid gland. The presence of
infection and nerve defects are contraindications
for this procedure. To progress from mass facial
muscle movements, associated with eating, chewing, swallowing, and talking, to volitional and
coordinated facial motion, an orderly and intensive postoperative physiotherapy for more than
2 years must be carried out. This includes intensive training in front of a mirror in accordance
with an exact protocol. A residual defect composed of infrequent occurrences of synkinesis
might persist; a paralyzed face will never return
to complete normalcy.

Acknowledgments. We would like to thank

Rajam Csordas for useful discussions and critical
reading and editing of the manuscript.

REFERENCES
1. Miehlke A. Surgery of the facial nerve. Philadelphia: W.B.
Saunders; 1973. p 1 2.
2. Hoffmann WY. Reanimation of the paralyzed face. Otolaryngol Clin North Am 1992;25:649 667.
3. Millesi H. Nerve suture and grafting to restore the extratemporal facial nerve. Clin Plast Surg 1979;6:333 341.
4. Grafstein B. The nerve cell body response to axotomy. Exp
Neurol 1975;48:32 51.
5. Spector JG, Lee P, Perterein J, Toufa D. Facial nerve
regeneration through autologous nerve grafts: a clinical
and experimental study. Laryngoscope 1991;101:537 554.
6. Spector JG. Mimetic surgery for the paralyzed face.
Laryngoscope 1985;95:1494 1522.
7. Spector JG. Neural repair in facial paralysis: clinical and
experimental studies. Eur Arch Otorhinolaryngol 1997;
254:68 75.
8. Millesi H, Meissl G, Berger A. The interfascicular nerve
grafting of the median and ulnar nerve. J Bone Joint Surg
[Am] 1972;54:727 750.
9. Terzis J, Faibisoff B, Williams HB. The nerve gap: suture under tension vs. graft. Plast Reconstr Surg 1975;56:
166 170.
10. Millesi H. Nerve grafting. Clin Plast Surg 1984;11:105 113.
11. Adkins WY, Osguthorpe JD. Management of trauma of
the facial nerve. Otolaryngol Clin North Am 1991;24:
587 612.
12. Kettel K. The techniques of intratemporal facial nerve
surgery. In: Miehlke A, editor. Surgery of the facial nerve.
Philadelphia: W.B. Saunders; 1973.

HEAD & NECK

September 2004

13. Pillsbury HC, Price HC, Gardiner LJ. Primary tumors of

the facial nerve: diagnosis and management. Laryngoscope 1983;93:1045 1048.
14. Lathrop FD. Management of traumatic lesions of the facial
nerve. Arch Otolaryngol 1952;55:410 416.
15. Yarbrough WG, Brownlee RE, Pillsbury HC. Primary
anastomosis of extensive facial nerve defects: an anatomic
study. Am J Otology 1993;14:238 246.

Secondary Facial Nerve End-to-End Repair

16. Gardetto A, Kovacs P, Piegger J, Rainer C, Meirer R,

Piza-Katzer H. Direct coaptation of extensive facial nerve
defects after removal of the superficial part of the parotid
gland: an anatomic study. Head Neck 2002;24:1047 1053.
17. Bunnell S. Surgical repair of the facial nerve. Arch
Otolaryngol 1937;25:235 259.
18. Martin RC. Late results of facial nerve repair. Ann Otol
Rhinol Laryngol 1955;64:859 869.

HEAD & NECK

September

2004

777