clinical article

J Neurosurg 124:647656, 2016

Extensive traumatic anterior skull base fractures

with cerebrospinal fluid leak: classification and repair
techniques using combined vascularized tissue flaps
Jacob B. Archer, MD,1 Hai Sun, MD, PhD,2 Phillip A. Bonney, BS,1 Yan Daniel Zhao, PhD,3
Jared C. Hiebert, MD,4,5 Jose A. Sanclement, MD,4 Andrew S. Little, MD,2 Michael E. Sughrue, MD,1
Nicholas Theodore, MD,2 Jeffrey James, MD, DDS,3 and Sam Safavi-Abbasi, MD, PhD1,2
Departments of 1Neurosurgery, 3Oral and Maxillofacial Surgery, 4Otorhinolaryngology, and 5Biostatistics and Epidemiology,
University of Oklahoma Health Science Center, Oklahoma City, Oklahoma; and 2Division of Neurological Surgery, Barrow
Neurological Institute, St. Josephs Hospital and Medical Center, Phoenix, Arizona

Objective This article introduces a classification scheme for extensive traumatic anterior skull base fracture to help
stratify surgical treatment options. The authors describe their multilayer repair technique for cerebrospinal fluid (CSF)
leak resulting from extensive anterior skull base fracture using a combination of laterally pediculated temporalis fascial
pericranial, nasoseptal-pericranial, and anterior pericranial flaps.
Methods Retrospective chart review identified patients treated surgically between January 2004 and May 2014 for
anterior skull base fractures with CSF fistulas. All patients were treated with bifrontal craniotomy and received pedicled
tissue flaps. Cases were classified according to the extent of fracture: Class I (frontal bone/sinus involvement only);
Class II (extent of involvement to ethmoid cribriform plate); and Class III (extent of involvement to sphenoid bone/sinus).
Surgical repair techniques were tailored to the types of fractures. Patients were assessed for CSF leak at follow-up. The
Fisher exact test was applied to investigate whether the repair techniques were associated with persistent postoperative
CSF leak.
Results Forty-three patients were identified in this series. Thirty-seven (86%) were male. The patients mean age was
33 years (range 1179 years). The mean overall length of follow-up was 14 months (range 545 months). Six fractures
were classified as Class I, 8 as Class II, and 29 as Class III. The anterior pericranial flap alone was used in 33 patients
(77%). Multiple flaps were used in 10 patients (3 salvage) (28%)1 with Class II and 9 with Class III fractures. Five (17%)
of the 30 patients with Class II or III fractures who received only a single anterior pericranial flap had persistent CSF leak
(p < 0.31). No CSF leak was found in patients who received multiple flaps. Although postoperative CSF leak occurred
only in high-grade fractures with single anterior flap repair, this finding was not significant.
Conclusions Extensive anterior skull base fractures often require aggressive treatment to provide the greatest long-
term functional and cosmetic benefits. Several vascularized tissue flaps can be used, either alone or in combination.
Vascularized flaps are an ideal substrate for cranial base repair. Dual and triple flap techniques that combine the use of
various anterior, lateral, and nasoseptal flaps allow for a comprehensive arsenal in multilayered skull base repair and
salvage therapy for extensive and severe fractures.
http://thejns.org/doi/abs/10.3171/2015.4.JNS1528
Key Words anterior skull base repair; cerebrospinal fluid leak; frontal sinus; nasoseptal flap pericranial flap;
temporoparietal fascial flap; traumatic meningitis

T
anterior cranial fossa and paranasal sinuses are
he
complex structures formed by the frontal, ethmoid,
and sphenoid bones. The ethmoid bone is inter-
posed between the frontal bones and joins the sphenoid
bone and sinus posteriorly.45 Traumatic force to the frontal
and facial bones transmits through the frontal, ethmoid,
and sphenoid sinuses lying below the medial part of the
anterior cranial base. The resulting complex comminuted
fractures involving intimately associated paranasal, cra-
nial, and intracranial structures pose unique management

submitted January 9, 2015. accepted April 7, 2015.

include when citing Published online October 16, 2015; DOI: 10.3171/2015.4.JNS1528.

AANS, 2016 J Neurosurg Volume 124 March 2016 647

J. B. Archer et al.
challenges. Access and visualization of the area around
the attachment of the crista galli and ethmoid bone may be
especially difficult.4,5,11,38 Multilayered repair of large skull
base defects is necessary to prevent delayed brain sagging
and CSF leak.17,18,33 The main goal of treatment in surgical
candidates is reconstitution of the cranial base anatomy
to prevent meningitis and mucocele formation while op-
timizing cosmetic outcome. There are various reports on
and methods for repair of skull base defects.33,35,39,44,50,52,54,55
For decades, the use of the pericranial flap has been the
means to repair anterior cranial floor defects due congeni-
tal disease, tumors, and traumatic injuries.4,5,11,38,45 This flap
is easy to harvest, seldom requires additional incisions,
and quickly incorporates into the surrounding tissue. One
drawback is that, in traumatic repairs, 10% to 17% of
patients have persistent CSF leak after the initial repair,
suggesting possible vascular compromise and subsequent
flap failure.1,10,21 However, the pericranial flap cannot only
be pedicled anteriorly using supratrochlear and supraor-
bital arteries but also laterally using superficial and deep
temporal arteries.42 An additional option for a lateral flap
is the temporalis muscle flap, which consists of the tempo-
ralis muscle (with or without the overlying fascial tissue).
This flap can also be elevated along with the pericranium
to form a longer and more robust flap.25,55,58 Another wide-
ly used flap is the temporoparietal fascia flap, which has
been well described in plastic and reconstructive surgery
for the head and neck; however, reports of its use in the
skull base are sparse.2,7,9,48,57
To evaluate the evolution of our neurosurgical tech-
niques and treatment outcomes for comminuted anterior
skull base fractures, we collected and analyzed data on
patients who underwent surgical repair of extensive an-
terior skull base fractures. In this series, we describe the
use of various combined vascularized tissue flaps for
multilayered repairs of the anterior skull base. We also
introduce a useful novel classification scheme to aid in
planning the required extent of surgical exposure and the
appropriate flap choice for reconstruction in anterior skull
base trauma.
Methods
After review and approval from the University of Okla-
homa Health Science Center institutional review board
was obtained, we reviewed the clinical and radiological
records of all patients who underwent surgical repair of
anterior skull base fractures between 2004 and 2014 at this
institution. We reviewed the mechanism of injury, clini-
cal findings on admission, and initial imaging findings,
including assembled fracture location, fracture character-
istics, and associated intracranial lesions and outcomes.
Diagnostic Procedures
All patients underwent high-resolution CT examination
upon admission. Serial axial CT images of the head, fa-
cial bones, and cervical spine were obtained (without ad-
ministration of a contrast agent) to assess intracranial pa-
thology and fractures of the cranium, cervical spine, and
face. Patients with fractures involving the sphenoid wing
and carotid canal underwent CT angiography of the head
648 J Neurosurg Volume 124 March 2016
and neck to evaluate possible vascular injury. One patient
with a delayed CSF leak who presented with meningitis
required the use of radionuclide cisternography to confirm
the presence of the CSF leak and identify its location.
Classification
Fractures were classified into 3 groups (Classes IIII),
based on the posterior extent of the fracture along the an-
terior skull base (Fig. 1). Fractures were classified to aid
neurosurgical planning and to accurately anticipate the
extent of exposure needed to adequately prepare the ap-
propriate vascularized flap for repair. Class I fractures
were defined as involving only the frontal sinus or frontal
bone. Patients with involvement of only the anterior table
were excluded from the study because they did not meet
criteria for the surgical approach. All patients with Class
I fractures had comminuted fractures. Class II fractures
were defined as those extending to the ethmoid cribriform
plate. Class III fractures were defined as those extending
to the sphenoid bone or sinus.
Surgical Anatomy and Exposure
Surgery was performed as soon as possible to lessen
the risk that patients would develop meningitis. Until
2010, a lumbar drain was placed routinely prior to surgery.
However, in 2010 the management philosophy of our insti-
tution changed, and this practice was abandoned. Instead,
we adopted a more aggressive intraoperative approach
that used endoscopy and microscopy for identification and
repair of the skull base. If a patient required intracranial
pressure monitoring and drainage, an external ventricu-
lar drain was placed before the operation, and intracranial
pressure was managed prior to the procedure.
All patients underwent a standard bifrontal craniotomy
with both intra- and extradural exposure. Following a bi-
coronal incision and subgaleal dissection, the pericranial
flap can be designed and created.
A standard anteriorly based pericranial flap may be
developed; however, this flap has several disadvantages
for some repairs of anterior skull base fractures. Often
the source of traumatic injury damages the pericranium
in this area (Fig. 2). This compromises the vascular sup-
ply from the supraorbital and supratrochlear arteries, po-
tentially obviating this flap design in some cases (Fig. 3).
Also, when an anteriorly based flap is inset, a gap must be
left in the bone flap and superior orbital rims to maintain
perfusion to the pericranial flap. Initially, glabellar full-
ness will result, secondary to tissue bulk of the pericranial
flap. After tissue edema subsides, a gap may be evident,
which often becomes a source of poor cosmesis. There-
fore, the vascularized flap design evolved recently into a
laterally based flap.
The temporalis muscle flap comprises the temporalis
muscle, overlying temporalis fascia, and attached pericra-
nium, with vascularization from the anterior and posterior
deep temporal arteries. For this flap, the parietal pericra-
nium is marked and designed with a lateral pedicle (Fig.
4). The pericranium is incised contralateral to the pedicle
of the flap, and a blunt subperiosteal dissection ensues to
the level of the zygoma and the vascular pedicle contain-

FIG. 1. Illustration demonstrating the fracture classification scheme. A:
Class I exocranial surfaceextensive comminuted fracture of the an-
terior and posterior tables of the frontal sinus.B: Class I endocranial
surfaceposterior to anterior view showing a lateral comminuted frac-
ture limited to the anterior and posterior walls of the frontal sinus without
extension into the ethmoid bone or the sphenoid bone.C: Class II
exocranial surfacefracture extending down the frontal sinus and back
through the ethmoid bone.D: Class II endocranial surfacefracture
down the lateral posterior wall of the frontal sinus into the ethmoidcrib-
riform plate complex without extension into the sphenoid bone. E: Class
III exocranial surfacefracture along the lateral and calvarial portions of
the frontal bone, with involvement of the orbits and orbital roof, ethmoid
bone, and sphenoid bone. F: Class III endocranial surfaceextensive
comminuted fractures involving bilateral frontal bone and ethmoid bone,
with extension into the sphenoid bone and sinuses. Copyright Barrow
Neurological Institute, Phoenix, Arizona. Published with permission.
Figure is available in color online only.
ing the deep temporal arterial supply. The anteriorly based
flap, if available, can then be dissected in the subperiosteal
plane to the orbital rims as in the standard exposure, and
a dual flap is available for repair (Fig. 4). The length
and degree of dissection of the laterally pedicled tempo-
ralis muscle and fascia-pericranial flap can be tailored ac-
cording to the extent and class of the fractures. For Class
Combined vascularized tissue flaps for skull base fracture repair
FIG. 2. Comminuted frontal bone fracture obviating anteriorly based
pericranial flap design due to severe vascular compromise of the supra-
orbital and supratrochlear arteries. The temporalis fasciapericranial
flap intended to be used for anterior skull base repair is based on the left
superficial and deep temporal perforators. Figure is available in color
online only.
III fractures, a vascularized flap of 1520 cm can be dis-
sected, and the entire anterior skull base can be covered.
The pedicled fascia/pericranium sheet is usually sufficient
for skull base reconstruction; the temporalis muscle itself
is not used for reconstruction and is sutured back to the
anterior part of the temporal fossa to help decrease post-
operative temporal fossa hollowing and poor cosmesis
(Fig. 4). The temporoparietal fascial flap, or galeal flap, is
another lateral flap that is highly vascularized from the su-
perficial temporal vessels, and it has been well described.
Both flaps are rarely compromised in trauma patients and
provide an excellent source for repair of skull base defects.
After the tissue flaps have been dissected, a standard
bifrontal craniotomy is performed with or without fronto-
orbital osteotomies to improve the working angles. Using
the operative microscope, the endocranial surface of the
skull base is inspected for additional dural lacerations.
The frontal sinus is cranialized in a standard fashion,
making sure to remove the mucosal surfaces to avoid
mucocele formation. The pericranial flap is placed extra-
durally along the anterior floor of the skull and sutured
microsurgically to the dura for a watertight seal. A pos-
teriorly releasing incision must be made in the laterally
based temporalis flap to allow proper rotation into the
cranial cavity (Fig. 5).
Endoscopic endonasal dissection and repair with a na-
soseptal flap are performed for extensive fractures involv-
ing the sphenoid and ethmoid sinuses with large defects or
as a salvage procedure when other flaps have been used
and patients have persistent CSF leaks. The utilization of
this flap has been well described. Following flap place-
J Neurosurg Volume 124 March 2016 649
J. B. Archer et al.
FIG. 3. Full-thickness pericranial laceration associated with the initial
skull base fracture. This laceration extends from the superior temporal
line to approximately midline, transecting the supraorbital and supra-
trochlear vasculature on the right side. Figure is available in color online
only.
ment, concurrent treatment of associated facial fractures
is completed.
Results
Patient Population
Forty-three patients were identified. Their demographic
characteristics and the mechanisms of injury are summa-
rized in Table 1. The most common clinical finding was
periorbital edema, which was present in 21 (49%) of the
patients. Other clinical findings are summarized in Table
650 J Neurosurg Volume 124 March 2016
FIG. 4. Pericranial flap design. Laterally based pericranial flap at left,
and smaller anteriorly based flap elevated. The smaller anterior pericra-
nial segment may still be used for repair of the frontal sinus. Figure is
available in color online only.
2. The patients neurological presentation varied widely;
the mean Glasgow Coma Scale score was 9.8 (range 315).
The most common anterior skull base fracture site was
the frontal sinus, which was fractured in 41 (95%) of the
43 patients (Table 3), followed by involvement of the eth-
moid cribriform plate (in 27 patients [63%]). Associated
facial fractures were present in nearly 39 patients (91%).
Six patients had Class I fractures, 8 had Class II fractures,
and 29 had Class III fractures; thus, most patients in this
series had posteriorly extensive anterior skull base inju-
ries.
Following fractures, the most common intracranial
pathology was pneumocephalus, which was found in 30
patients (70%). The remaining intracranial pathological
findings are described in Table 4.
Surgical Management and Outcome
The average time from admission to surgery was 5 days.
Hematoma evacuation was required in 5 patients (12%),
and another 5 (12%) required external ventricular drain
placement. Placement of a permanent ventriculoperitoneal
shunt was performed in 3 patients (7%). Associated facial
fractures were repaired during the same operation in 18
patients (42%).
Anterior skull base repair results are listed in Table 5.
All 6 patients with Class I fractures underwent repair of
dural tears, frontal sinus obliteration with removal of the
posterior table, and nasal outflow tract obliteration. An
anteriorly based pericranial flap was used exclusively for
repair in all patients with Class I fractures, and no patient
with a Class I fracture had a persistent CSF leak. Thirty-
three patients (77%) received an anterior pericranial flap
alone, and 10 patients (including 3 with salvage flaps)
(23%) had multiple flaps; 1 with Class II fractures and 9
with Class III fractures.
There were 8 patients with Class II fractures. In one
of these patients, the anterior pericranium was severely
compromised, so a dual flap technique was used. A later-
ally based flap was placed and then the uncompromised

FIG. 5. A releasing incision must be made in the posterior temporalis to
allow proper rotation of the flap (arrow). After inset of the pericranial flap,
the rotated temporalis muscle is secured to the superior temporal crest
in its new position to allow for good cosmetic outcome. Figure is avail-
able in color online only.
anterior pericranium proximal to the pericranial lacera-
tion was used for repair of the frontal sinus (Fig. 4). One
patient with a Class II fracture that was repaired with an
exclusive anteriorly based flap experienced persistent leak,
which was discovered one month after discharge and re-
solved with a lumbar drain.
There were 29 patients with Class III fractures. The
dual flap technique was used for 3 of these patients, but the
laterally based flap was designed and extended over the
sphenoid fractures as well as over the comminuted eth-
moid. Two patients with Class III fractures were treated
with a triple flap technique that consisted of the anteri-
orly based pericranial flap for the frontal sinus, a laterally
based temporoparietal fascial flap for repair of ethmoid
and sphenoid bone defects, and an additional nasoseptal
flap for the exocranial surface. None of these patients who
underwent multiflap repair had persistent CSF leakage.
Four patients with Class III fractures who underwent
repair with purely anteriorly based pericranium had per-
sistent CSF leakage. Three of these patients required sal-
vage endonasal endoscopic augmentation and use of a
nasoseptal flap for reconstruction of the skull base. The
salvage repair time frame ranged from 1 to 3 months after
discharge. In the other patient with a Class II fracture and
a persistent leak, the leak was managed successfully with
a lumbar drain.
Combined vascularized tissue flaps for skull base fracture repair
TABLE 1. Demographic characteristics and mechanism of injury
in 43 patients with anterior skull base fractures*
Characteristic Value
Sex
Male 37 (86)
Female 6 (14)
Age in yrs
Mean 33
Range 1179
Injury
Motor vehicle trauma 26 (60)
Work-related accident 8 (19)
Fall 4 (9)
Gunshot 3 (7)
Assault 1 (2)
Other 1 (2)
* Values represent numbers of patients (%) except where otherwise indicated.
The injury percentages add up to less than 100% due to rounding.
Postoperative Complications and Outcomes
The mean follow-up time was 14 months (range 545
months). For the 31 patients with available follow-up data,
the mean Glasgow Outcome Scale score was 4.4 (range
15, with 5 corresponding to minor neurological deficits).
Most of these patients (n = 21) had a score of 5, with 3 be-
ing the next most common score (n = 6). One patient had a
score of 1, and 3 had a score of 4.
The postoperative course of 2 patients was complicated
by meningitis (5%). In both cases, an Enterococcus spe-
cies grew in the patients drains, and the infections were
treated successfully with intravenous antibiotic therapy.
No patients had postoperative neurological deteriora-
tion attributable to increased cerebral edema or hemato-
ma. Only 1 patient died; this was on postoperative Day 7
after surgical repair due to a middle cerebral artery infarc-
tion that was present preoperatively. Local wound infec-
tion occurred in 2 patients. One patient developed chronic
wound dehiscence, and another developed nasal cellulitis
a year after surgery. At follow-up, no patients had devel-
oped mucoceles. Three patients required a second opera-
tion for closure of skull scalp defect.
Five (12%) of the 43 patients developed CSF leak af-
ter surgical repair. All 5 patients had Class II or Class III
fractures and underwent repair with only an anterior peri-
cranial flap (Table 5). No patient treated with a primarily
combined flap (dual flap or triple flap technique) or with
endoscopic repair required reoperation, had persistent
postoperative CSF leak, or developed meningitis. Based
on the Fisher exact test, there was no statistically signifi-
cant association between the postoperative CSF leak and
the repair strategy (p < 0.31).
Discussion
Indications for Surgical Treatment
Dural tears and CSF leaks are commonly encoun-
J Neurosurg Volume 124 March 2016 651
J. B. Archer et al.

TABLE 2. Clinical findings in 43 patients with anterior skull base TABLE 4. Intracranial pathology associated with anterior skull
fractures base fractures in 43 patients
Finding No. (%) Pathological Finding No. (%)
Periorbital edema 21 (49) Pneumocephalus 30 (70)
Periorbital ecchymosis 19 (44) Subarachnoid hemorrhage 25 (58)
Epistaxis 17 (40) Subdural hemorrhage 23 (53)
Sluggish or nonreactive pupils 12 (28) Intraparenchymal hemorrhage 17 (40)
Hemotympanum 8 (19) Cerebral edema 12 (28)
Hemorrhagic contusions 11 (26)
Ophthalmoplegia 3 (7)
Intraventricular hemorrhage 10 (23)
CN VII palsy 3 (7)
Epidural hematoma 9 (21)
Meningitis 3 (7)
Midline shift 9 (21)
Otorrhea 2 (5) Carotid dissection 2 (5)
Visible brain matter 2 (5) Subfalcine herniation 2 (5)
Seizure 2 (5) Transtentorial herniation 1 (2)
Proptosis 2 (5) Uncal herniation 1 (2)
Rhinorrhea 1 (2)
Decorticate posturing 1 (2)
CN = cranial nerve.
8 days. In light of the high incidence of meningitis and
mortality rates of 18.8% to 25%,5,32 the rapid identifica-
tion and treatment of CSF fistulas are critical to reducing
tered in skull base fractures and are 5 to 6 times more long-term morbidity and mortality. When a dural leak can
frequent in the anterior than in the middle or posterior be sealed successfully, the risk for meningitis decreases
skull base.1,6,2124,28,33,46,48,52,53,56 Ethmoidcribriform plate from 85% to 7% over 10 years.16 Although the spontaneous
fractures are the most common.3,31,35 The preponderance resolution of CSF fistula is possible,40 the risk for menin-
of ethmoidcribriform plate fractures can be explained by gitis must be weighed against potential surgical morbidity
the thin fragile dura over the cribriform plate and its firm and mortality. However, recurrent CSF leakage after initial
adherence.38,49 Additionally, the ethmoid is porous in rela- spontaneous cessation may occur and is associated with
tion to the other skeletal bones; therefore, even low-impact a 30% rate of recurrent meningitis and increased mortal-
trauma can result in ethmoid fractures.49 CSF fistulas oc- ity.15 Spontaneous resolution is also less likely in cases of
cur in 10% to 30% of patients with skull base fractures,61 extensive, comminuted anterior skull base fractures, espe-
and from 7% to 30% of patients with posttraumatic CSF cially those involving the ethmoid and sphenoid sinuses. In
leakage develop meningitis.19 our series of 43 patients, several patients (7%) with chronic
The first successful CSF repair was reported in 1926 by and persistent CSF leaks requiring subsequent conclusive
Dandy,10 who used a fascia lata autograft to seal the pos- reoperation and repair were referred to us after a delay to
terior wall of the frontal sinus. Posttraumatic meningitis treatment. Compared with immediate surgical repair, ob-
is the major concern in patients with a skull base fracture servation has been proven to lead to considerably higher
and a CSF leak. The overall rate of infection ranges from risk for development of meningitis.51
1.3% per day for the 1st week after a skull base fracture High-resolution CT is the primary diagnostic modality
to 8.4% per year after the first 6 months.15,20 Sakas and used to identify cranial and facial fractures. It has been
colleagues49 reported a significantly higher infection rate shown to be highly sensitive in detecting skull base frac-
in patients who had CSF rhinorrhea lasting longer than tures.30,34 CT scans are also useful in the diagnosis of CSF
fistula, with accurate pinpointing of the position and size of
the leak.30,36 Pneumocephalus, which is associated with the
TABLE 3. Locations of anterior skull base fractures in 43 patients presence of dural tears, should be considered the equiva-
lent of a CSF fistula. An air volume greater than 10 ml is
Location No. (%) thought to be a risk factor for development of meningitis.13
Frontal sinus 41 (95) Pneumocephalus was present in 70% of our patients, and
most had extensive, multiple, and comminuted fracture
Orbital plate of frontal bone 28 (65)
sites. In fact, size and location of the anterior skull base
Ethmoid cribriform plate 27 (63) fracture have been found to be predisposing factors for in-
Sphenoid sinus/sella turcica 21 (49) fection and meningitis, independent of the presence of rhi-
Temporal bone 13 (30) norrhea.49 We therefore recommend prompt surgical repair
Sphenoid wing 12 (28) in patients with extensive fractures to avoid infection and
Carotid canal 11 (26) related complications. Radionuclide cisternography was
used in only 1 patient, who had a chronic CSF leak. We
Foramen 1 (2)
believe that this imaging modality is usually not indicated
652 J Neurosurg Volume 124 March 2016

TABLE 5. Classification and flap use in 43 patients for repair of anterior skull base fractures
Type of Flap
Anterior pericranial flap only
Dual flap (anterior pericranial and lateral flaps)
Triple flap (anterior pericranial, lateral pericranial, and nasoseptal flaps)
Anterior pericranial and nasoseptal flap
* One patient had persistent leak that resolved with a lumbar drain.
Four patients had a persistent CSF leak. Three required salvage nasoseptal flap. One additional patient required a lumbar drain.
in patients with acute, extensive trauma. CT scanning is
a faster and less expensive tool for diagnosis and surgical
planning of acute fractures.
When a patient has severe intracranial traumatic in-
jury with concomitant cerebral edema, the treatment of
increased intracranial pressure should occur prior to re-
pair of the skull base fracture. Repair of the CSF leak and
the anterior skull base can usually be initiated within 5 to
7 days after initial stabilization and management. In our
series, the average time between hospital admission and
surgery was 5 days. Preoperative prophylactic antibiotic
treatment has not been shown to be of benefit in this pa-
tient population20 It has been suggested that prophylactic
antibiotic therapy may increase the risk of meningitis by
altering the normal flora of the sinus mucosa.8,13,14,29 We be-
lieve that prophylactic antibiotic treatment is not indicated
and that the risk of meningitis is lessened to an acceptable
extent by early surgical intervention.
Fracture Classification and Treatment Strategies
Major challenges of anterior skull base repair include
not only reconstruction of the defect but also the decision
making about the extent of the surgical approach (endo-
scopic, open, or a combination thereof) and the appropri-
ate reconstruction strategy (choice of vascularized flap).
Prior classification schemes for categorizing fractures of
the anterior skull base have sought to associate fracture
location with long-term risk for meningitis.37,49 Our clas-
sification is based on the greatest posterior extent of the
fracture and is used for surgical planning, thereby allow-
ing anticipation of required exposure, repair extent, and
flap choice.
Since Wigand59 first reported endoscopic CSF fistula re-
pair in 1981, the use of endoscopic techniques in anterior
skull trauma has grown; endoscopic techniques are now
used solely or in combination with conventional surgical
techniques.26,43 The nasoseptal flap, based off the nasosep-
tal artery, has increasingly become the workhorse of en-
donasal skull base reconstruction.25,26,43 Current evidence
suggests that endoscopic skull base repair with vascular-
ized tissue is associated with a lower rate of CSF leaks
than repair with free tissue graft and results in closure rates
similar to those reported for open surgical repair.26,39,54 En-
doscopic repair with use of a vascularized pedicled naso-
septal flap is now considered the gold standard for most
CSF leak repairs because it is safe and effective.26,43 Sole
endoscopic repair is of the greatest use in the absence of
associated intracranial trauma and for patients who do not
Combined vascularized tissue flaps for skull base fracture repair
Class III
Class I (n = 6) Class II (n = 8) (n = 29)
6 7* 23
0 1 3
0 0 2
0 0 1
require repair of facial fractures.12,31,35 However, for pa-
tients with comminuted frontal sinus fractures and asso-
ciated facial fractures, open surgery may be necessary to
prevent mucocele formation and to achieve good cosmetic
outcomes.
Pericranial flaps are traditionally used in open repair of
the anterior skull base. They are easy to harvest and are
quickly incorporated into surrounding tissue because they
are highly vascularized from the supratrochlear and supra-
orbital arteries.41,42,54 However, we found that the pericra-
nial flap may have several disadvantages in patients with
extensive anterior skull base fractures. Viable uncompro-
mised pericranial flap is often unavailable if initial trauma
or comminuted fracture edges result in pericranial lacera-
tions that compromise the blood supply. Furthermore, the
posterior extent of many fractures precludes the use of the
anteriorly based pericranium, which is not versatile and
may not extend far enough to span large defects across
the frontal, ethmoid, and the sphenoid sinuses, which are
found in Class II and III fractures. Additionally, when an
anteriorly based flap is inset, a gap in the bone flap and
the superior orbital rims must be left to maintain perfu-
sion to the pericranial flap. After tissue edema subsides,
a gap may be evident that can become a source of poor
cosmesis. Therefore, for complex traumatic anterior skull
base defects (Class II and Class III), the pericranial design
evolved into a laterally based tissue flap. Lateral blood sup-
plies are often uncompromised and have 2 sides available
on which to base a lateral vascular pedicle. The superfi-
cial, middle, and deep temporal arteries provide the blood
supply for these flaps. In our series, we mostly used lat-
eral flaps combining the temporalis muscle fascia and the
deeper pericranial tissue layer. This combination forms a
resilient tissue flap with a rich vascular supply.27 This flap
is highly versatile, provides a large surface area with a su-
perior arc of rotation, and may be used with minimal donor
site morbidity and excellent cosmetic outcomes. There is
scant literature evaluating utilization and classification of
combination flaps in traumatic repairs.17
An additional laterally based flap is the temporoparietal
fascial flap, which has been used especially in head and
neck reconstruction.47 Taha et al.55 described the use of the
temporoparietal fascial flap for repair of a large traumatic
middle cranial fossa CSF fistula via a middle cranial fossa
approach. The risks associated with this flap, however, are
its potential to damage hair follicles or skin vascular sup-
ply, resulting in postoperative alopecia or scalp necrosis.54
Therefore, we reserve this flap for cases in which repeated
J Neurosurg Volume 124 March 2016 653
J. B. Archer et al.
repair surgery may be necessary and the pericranium has
already been sacrificed.
We have integrated the use of the surgical microscope
and endoscopic techniques for Class II and Class III skull
base fractures. The microscope allows minimal retraction
of the already injured and edematous frontal lobes, while
also allowing effective inspection and repair. None of the
patients in our series who underwent a combined endo-
scopic and microsurgical approach required reoperation
or had persistent CSF leak, obviating the need for lumbar
drains. Recently, no significant difference was found in the
recurrence of CSF leak, the incidence of meningitis, or the
duration of hospitalization between patients who did and
did not receive lumbar drains.18,60 In our practice, we have
abandoned routine placement of lumbar drains, given their
potential for complications, such as intracranial hypoten-
sion and pneumocephalus,18 especially in posttraumatic
patients with increased intracranial pressure and greater
potential for herniation.
The use of multiple flaps is determined by fracture ex-
tent on preoperative CT. For patients with complex Class
II and Class III fractures, we occasionally use dual flap or
triple flap repair techniques. The dual flap technique may
also be used if microsurgical inspection reveals multiple,
large defects; in such cases, the anteriorly pedicled pericra-
nium is used (if available and uncompromised) for frontal
sinus repair and the laterally based temporoparietal fascial
flap is used for ethmoid and sphenoid fractures. This tech-
nique may be further augmented with an endoscopic ap-
proach of the exocranial surface and use of the triple flap
technique, which includes application of a nasoseptal flap.
However, the nasoseptal flap may be reserved for salvage
therapy when there is a persistent CSF leak after open re-
pair of the endocranial surface. Persistent CSF leaks after
previous attempted repair can be challenging to manage
because of the lack of available healthy autologous tissue
after failure of a pericranial flap. Five of our patients with
more extensive fractures (1 Class II and 4 Class III) and
traditional exclusive anterior pericranial repair had per-
sistent CSF leakage. Three required salvage management
with additional nasoseptal flap repair, while the other two
were managed with a lumbar drain.
Our data, although not statistically significant, suggest
that combining an anterior pericranial flap with an addi-
tional flap (nasoseptal flap or temporoparietal fascial flap)
may lead to prevention of CSF leak. Whether combina-
tion flaps or laterally based flaps are truly superior to the
anteriorly based pericranial flap alone remains to be de-
termined in further larger studies. Nevertheless, our data
indicate that dual flap and triple flap techniques allow for a
comprehensive armamentarium for anterior skull base re-
pair and salvage therapy for extensive and severe anterior
skull base fractures.
Conclusions
Early and aggressive surgical repair of anterior skull
base fractures may be necessary for patients in whom con-
servative management has failed and for those who pres-
ent with meningitis or extensive depressed or comminuted
fractures. The use of vascularized flaps is an ideal substrate
654 J Neurosurg Volume 124 March 2016
for multilayered skull base repair. Dual flap and triple flap
techniques that combine the use of various anterior, lat-
eral, and nasoseptal flaps allow for a comprehensive arse-
nal in skull base repair and salvage therapy for extensive
and severe fractures. When patients meet surgical criteria,
these repair techniques result in a low rate of morbidity
and no mortality, with excellent functional and cosmetic
outcomes devoid of recurrent CSF leaks and meningitis.
Knowledge about the surgical anatomy of various flaps for
complex skull base defects, and expertise in using them,
should be in the armamentarium of comprehensive skull
base surgery centers worldwide.
Acknowledgment
Dr. Sun is the recipient of funding from the Congress of Neuro-
logical Surgeons Christopher C. Getch Flagship Fellowship Award.
References
1. Bell RB, Dierks EJ, Brar P, Potter JK, Potter BE: A protocol
for the management of frontal sinus fractures emphasizing
sinus preservation. J Oral Maxillofac Surg 65:825839,
2007
2. Brent B, Upton J, Acland RD, Shaw WW, Finseth FJ, Rogers
C, et al: Experience with the temporoparietal fascial free flap.
Plast Reconstr Surg 76:177188, 1985
3. Brodie HA, Thompson TC: Management of complications
from 820 temporal bone fractures. Am J Otol 18:188197,
1997
4. Cairns H: Injuries of the frontal and ethmoidal sinuses with
special reference to cerebrospinal rhinorrha and aeroceles.
J Laryngol Otol 52:589623, 1937
5. Calvert CA: Discussion on injuries of the frontal and eth-
moidal sinuses. Proc R Soc Med 35:805810, 1942
6. Chen KT, Chen CT, Mardini S, Tsay PK, Chen YR: Frontal
sinus fractures: a treatment algorithm and assessment of
outcomes based on 78 clinical cases. Plast Reconstr Surg
118:457468, 2006
7. Cheney ML, Varvares MA, Nadol JB Jr: The temporoparietal
fascial flap in head and neck reconstruction. Arch Otolaryn-
gol Head Neck Surg 119:618623, 1993
8. Choi D, Spann R: Traumatic cerebrospinal fluid leakage: risk
factors and the use of prophylactic antibiotics. Br J Neuro-
surg 10:571575, 1996
9. Collar RM, Zopf D, Brown D, Fung K, Kim J: The versatility
of the temporoparietal fascia flap in head and neck recon-
struction. J Plast Reconstr Aesthet Surg 65:141148, 2012
10. Dandy WE: Pneumocephalus (intracranial pneumatocele or
aerocele). Arch Surg 12:949982, 1926
11. Dandy WE: Treatment of rhinorrhea and otorrhea. Arch
Surg 49:7585, 1944
12. Dodson EE, Gross CW, Swerdloff JL, Gustafson LM: Trans-
nasal endoscopic repair of cerebrospinal fluid rhinorrhea and
skull base defects: a review of twenty-nine cases. Otolaryn-
gol Head Neck Surg 111:600605, 1994
13. Eftekhar B, Ghodsi M, Nejat F, Ketabchi E, Esmaeeli B: Pro-
phylactic administration of ceftriaxone for the prevention of
meningitis after traumatic pneumocephalus: results of a clini-
cal trial. J Neurosurg 101:757761, 2004
14. Eljamel MS: Antibiotic prophylaxis in the management of
CSF fistula. Surg Neurol 50:387, 1998 (Letter)
15. Eljamel MS, Foy PM: Acute traumatic CSF fistulae: the risk
of intracranial infection. Br J Neurosurg 4:381385, 1990
16. Eljamel MS, Foy PM: Post-traumatic CSF fistulae, the case
for surgical repair. Br J Neurosurg 4:479483, 1990

17. Eloy JA, Choudhry OJ, Christiano LD, Ajibade DV, Liu JK:
Double flap technique for reconstruction of anterior skull
base defects after craniofacial tumor resection: technical
note. Int Forum Allergy Rhinol 3:425430, 2013
18. Eloy JA, Kuperan AB, Choudhry OJ, Harirchian S, Liu JK:
Efficacy of the pedicled nasoseptal flap without cerebrospinal
fluid (CSF) diversion for repair of skull base defects: inci-
dence of postoperative CSF leaks. Int Forum Allergy Rhi-
nol 2:397401, 2012
19. Friedman JA, Ebersold MJ, Quast LM: Persistent posttrau-
matic cerebrospinal fluid leakage. Neurosurg Focus 9(1):e1,
2000
20. Friedman JA, Ebersold MJ, Quast LM: Post-traumatic cere-
brospinal fluid leakage. World J Surg 25:10621066, 2001
21. Gabrielli MFR, Gabrielli MAC, Hochuli-Vieira E, Pereira-
Fillho VA: Immediate reconstruction of frontal sinus frac-
tures: review of 26 cases. J Oral Maxillofac Surg 62:582
586, 2004
22. Gerbino G, Roccia F, Benech A, Caldarelli C: Analysis of
158 frontal sinus fractures: current surgical management and
complications. J Craniomaxillofac Surg 28:133139, 2000
23. Gonty AA, Marciani RD, Adornato DC: Management of
frontal sinus fractures: a review of 33 cases. J Oral Maxil-
lofac Surg 57:372381, 1999
24. Gossman DG, Archer SM, Arosarena O: Management of
frontal sinus fractures: a review of 96 cases. Laryngoscope
116:13571362, 2006
25. Hadad G, Bassagasteguy L, Carrau RL, Mataza JC, Kassam
A, Snyderman CH, et al: A novel reconstructive technique
after endoscopic expanded endonasal approaches: vascular
pedicle nasoseptal flap. Laryngoscope 116:18821886, 2006
26. Harvey RJ, Parmar P, Sacks R, Zanation AM: Endoscopic
skull base reconstruction of large dural defects: a systematic
review of published evidence. Laryngoscope 122:452459,
2012
27. Horowitz JH, Persing JA, Nichter LS, Morgan RF, Edgerton
MT: Galeal-pericranial flaps in head and neck reconstruction.
Anatomy and application. Am J Surg 148:489497, 1984
28. Ioannides C, Freihofer HP, Vrieus J, Friens J: Fractures of
the frontal sinus: a rationale of treatment. Br J Plast Surg
46:208214, 1993 (Erratum in Br J Plast Surg 46:718, 1993)
29. Klastersky J, Sadeghi M, Brihaye J: Antimicrobial prophy-
laxis in patients with rhinorrhea or otorrhea: a double-blind
study. Surg Neurol 6:111114, 1976
30. Kral T, Zentner J, Vieweg U, Solymosi L, Schramm J: Diag-
nosis and treatment of frontobasal skull fractures. Neurosurg
Rev 20:1923, 1997
31. Lanza DC, OBrien DA, Kennedy DW: Endoscopic repair of
cerebrospinal fluid fistulae and encephaloceles. Laryngo-
scope 106:11191125, 1996
32. Levin S, Nelson KE, Spies HW, Lepper MH: Pneumococcal
meningitis: the problem of the unseen cerebrospinal fluid
leak. Am J Med Sci 264:319327, 1972
33. Liu P, Wu S, Li Z, Wang B: Surgical strategy for cerebrospi-
nal fluid rhinorrhea repair. Neurosurgery 66 (6 Suppl Op-
erative):281286, 2010
34. Lloyd MN, Kimber PM, Burrows EH: Post-traumatic cere-
brospinal fluid rhinorrhoea: modern high-definition comput-
ed tomography is all that is required for the effective demon-
stration of the site of leakage. Clin Radiol 49:100103, 1994
35. Locatelli D, Rampa F, Acchiardi I, Bignami M, De Bernardi
F, Castelnuovo P: Endoscopic endonasal approaches for re-
pair of cerebrospinal fluid leaks: nine-year experience. Neu-
rosurgery 58 (4 Suppl 2):ONS-246ONS-256, 2006
36. Lund VJ, Savy L, Lloyd G, Howard D: Optimum imaging
and diagnosis of cerebrospinal fluid rhinorrhoea. J Laryngol
Otol 114:988992, 2000
37. Madhusudan G, Sharma RK, Khandelwal N, Tewari MK:
Nomenclature of frontobasal trauma: a new clinicoradio-
Combined vascularized tissue flaps for skull base fracture repair
graphic classification. Plast Reconstr Surg 117:23822388,
2006
38. McHugh HE: Treatment of fractures of the frontal and eth-
moid sinuses. Laryngoscope 68:16161640, 1958
39. McCormack B, Cooper PR, Persky M, Rothstein S: Extra-
cranial repair of cerebrospinal fluid fistulas: technique and
results in 37 patients. Neurosurgery 27:412417, 1990
40. Mincy JE: Posttraumatic cerebrospinal fluid fistula of the
frontal fossa. J Trauma 6:618622, 1966
41. Neligan PC, Mulholland S, Irish J, Gullane PJ, Boyd JB,
Gentili F, et al: Flap selection in cranial base reconstruction.
Plast Reconstr Surg 98:11591168, 1996
42. Parhiscar A, Har-El G: Frontal sinus obliteration with the
pericranial flap. Otolaryngol Head Neck Surg 124:304
307, 2001
43. Patel MR, Taylor RJ, Hackman TG, Germanwala AV, Sasaki-
Adams D, Ewend MG, et al: Beyond the nasoseptal flap: out-
comes and pearls with secondary flaps in endoscopic endona-
sal skull base reconstruction. Laryngoscope 124:846852,
2014
44. Ray BS, Bergland RM: Cerebrospinal fluid fistula: clinical
aspects, techniques of localization, and methods of closure. J
Neurosurg 30:399405, 1969
45. Rhoton AL Jr: The anterior and middle cranial base. Neuro-
surgery 51 (4 Suppl):S273S302, 2002
46. Rodriguez ED, Stanwix MG, Nam AJ, St Hilaire H, Simmons
OP, Christy MR, et al: Twenty-six-year experience treating
frontal sinus fractures: a novel algorithm based on anatomi-
cal fracture pattern and failure of conventional techniques.
Plast Reconstr Surg 122:18501866, 2008
47. Rose EH, Norris MS: The versatile temporoparietal fascial
flap: adaptability to a variety of composite defects. Plast Re-
constr Surg 85:224232, 1990
48. Sailer HF, Grtz KW, Kalavrezos ND: Frontal sinus frac-
tures: principles of treatment and long-term results after sinus
obliteration with the use of lyophilized cartilage. J Cranio-
maxillofac Surg 26:235242, 1998
49. Sakas DE, Beale DJ, Ameen AA, Whitwell HL, Whittaker
KW, Krebs AJ, et al: Compound anterior cranial base frac-
tures: classification using computerized tomography scanning
as a basis for selection of patients for dural repair. J Neuro-
surg 88:471477, 1998
50. Scher RL, Cantrell RW: Anterior skull base reconstruction
with the pericranial flap after craniofacial resection. Ear
Nose Throat J 71:210217, 1992
51. Schoentgen C, Henaux PL, Godey B, Jegoux F: Management
of post-traumatic cerebrospinal fluid (CSF) leak of anterior
skull base: 10 years experience. Acta Otolaryngol 133:944
950, 2013
52. Scholsem M, Scholtes F, Collignon F, Robe P, Dubuisson A,
Kaschten B, et al: Surgical management of anterior cranial
base fractures with cerebrospinal fluid fistulae: a single-insti-
tution experience. Neurosurgery 62:463471, 2008
53. Sivori LA II, de Leeuw R, Morgan I, Cunningham LL Jr:
Complications of frontal sinus fractures with emphasis on
chronic craniofacial pain and its treatment: a review of 43
cases. J Oral Maxillofac Surg 68:20412046, 2010
54. Snyderman CH, Janecka IP, Sekhar LN, Sen CN, Eibling DE:
Anterior cranial base reconstruction: role of galeal and peri-
cranial flaps. Laryngoscope 100:607614, 1990
55. Taha M, Carroll T, McMahon J: Vascularized temporopari-
etal fascial flap for the treatment of a traumatic cerebrospinal
fluid fistula in the middle cranial fossa. Technical note. J
Neurosurg 111:393395, 2009
56. Tedaldi M, Ramieri V, Foresta E, Cascone P, Iannetti G:
Experience in the management of frontal sinus fractures. J
Craniofac Surg 21:208210, 2010
57. Telliolu AT, Tekdemir I, Erdemli EA, Tccar E, Ulusoy G:
Temporoparietal fascia: an anatomic and histologic reinvesti-
J Neurosurg Volume 124 March 2016 655
J. B. Archer et al.
gation with new potential clinical applications. Plast Recon-
str Surg 105:4045, 2000
58. Tolhurst DE, Carstens MH, Greco RJ, Hurwitz DJ: The sur-
gical anatomy of the scalp. Plast Reconstr Surg 87:603614,
1991
59. Wigand ME: Transnasal ethmoidectomy under endoscopical
control. Rhinology 19:715, 1981
60. Yeo NK, Cho GS, Kim CJ, Lim GC, Jang YJ, Lee BJ, et al:
The effectiveness of lumbar drainage in the conservative and
surgical treatment of traumatic cerebrospinal fluid rhinor-
rhea. Acta Otolaryngol 133:8290, 2013
61. Yilmazlar S, Arslan E, Kocaeli H, Dogan S, Aksoy K, Kor-
fali E, et al: Cerebrospinal fluid leakage complicating skull
base fractures: analysis of 81 cases. Neurosurg Rev 29:64
71, 2006
Disclosure
The authors report no conflict of interest concerning the materi-
656 J Neurosurg Volume 124 March 2016
als or methods used in this study or the findings specified in this
paper.
Author Contributions
Conception and design: Safavi-Abbasi. Acquisition of data:
Archer, Bonney, Hiebert, James. Analysis and interpretation of
data: Little, Sun, Theodore, Safavi-Abbasi. Drafting the article:
Archer, Sughrue, Safavi-Abbasi. Critically revising the article:
Little, Sun. Reviewed submitted version of manuscript: Little,
Theodore, James, Safavi-Abbasi. Statistical analysis: Zhao.
Administrative/technical/material support: Hiebert, Sanclement,
Sughrue. Study supervision: Safavi-Abbasi.
Correspondence
Andrew S. Little, c/o Neuroscience Publications, Barrow
Neurological Institute, St. Josephs Hospital and Medical Center,
350 W. Thomas Rd., Phoenix, AZ 85013. email: neuropub@
dignityhealth.org.