Temporal hollowing following coronal incision: a prospective, randomized, controlled trial.
ABSTRACT Coronal incisions are used in traumatic, reconstructive, and cosmetic procedures to access the lateral facial skeleton. Temporal hollowing is a common complication following coronal incision that affects the patient both physically and psychologically. Several dissections have been recommended through this area to avoid injury to the frontal branch of the facial nerve and the temporal fat pad, which is thought to be the cause of hollowing. The purpose of this study was to identify the cause of postoperative temporal hollowing.
Patients requiring a coronal incision were recruited prospectively. Each side of the head in all patients was randomized to suprafascial, subfascial, or deep dissection. An unmarked envelope containing the type of dissection to be performed for each side was used. All envelopes contained equal distributions of all groups. The incidence and severity of temporal hollowing 6 months postoperatively were measured clinically and by computed tomographic volume analysis.
Twenty-seven patients with 54 sides (18 suprafascial, 15 subfascial, and 21 deep) completed the study. There were no demographic differences among the three groups. The incidence and severity of temporal hollowing were lowest with suprafascial dissection. Other factors associated with the presence of temporal hollowing included a reduction in body mass index. There were no injuries to the frontal nerve in any of the dissections. Postoperative temporal hollowing was associated with surgical approach and postoperative weight loss.
Elevation of a coronal flap in the suprafascial plane and minimization of patient weight loss may decrease the incidence of postoperative temporal hollowing.
- [show abstract] [hide abstract]
ABSTRACT: The coronal approach is a versatile surgical technique to expose the craniofacial skeleton. A retrospective clinical study on the use of this approach in treatment of craniomaxillofacial trauma in 28 patients was carried out. The study showed that this technique provides optimum exposure of the fracture sites, allowing for accurate anatomic reduction and fixation of the fractured segments and good cosmetic results in the incision site. The surgical technique, indications, and management and prevention of potential complications of the coronal approach are discussed.Journal of Oral and Maxillofacial Surgery 07/1990; 48(6):579-86. · 1.33 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The primary repair of facial fractures requires sufficient subperiosteal dissection and mobilization of soft tissues to permit accurate fracture reduction. Improper repositioning of soft tissues predisposes the site to deformities with subsequent adverse effects on the aesthetics of the final result. The purpose of this paper is to describe these deformities, the modification of surgical approaches to avoid them, and the various techniques for secondary soft tissue reconstruction. The paper reviews the assessment of soft tissue deformity and the principles of soft tissue reconstruction in addressing temporal contour deformity, cheek ptosis, eyelid deformities, and medial and lateral canthal dystopia. The morbidity of remote incisions and soft tissue degloving used in primary facial fracture repair can be minimized by using a meticulous technique and precise soft tissue repositioning at closing.The Journal of cranio-maxillofacial trauma 02/1999; 5(3):19-29; discussion 30-1.
- [show abstract] [hide abstract]
ABSTRACT: Depression of profile in the temporal region is commonly seen after orbital rim advancement procedures. A newly developed temporalis musculoosseous flap has been designed with the intent to prevent this postoperative occurrence.Journal of Craniofacial Surgery 10/1994; 5(4):271-4. · 0.69 Impact Factor
Temporal Hollowing following Coronal Incision:
A Prospective, Randomized, Controlled Trial
Damir B. Matic, M.D., M.Sc.
Sharon Kim, M.D.
London, Ontario, Canada
Background: Coronal incisions are used in traumatic, reconstructive, and cos-
metic procedures to access the lateral facial skeleton. Temporal hollowing is a
common complication following coronal incision that affects the patient both
physically and psychologically. Several dissections have been recommended
through this area to avoid injury to the frontal branch of the facial nerve and
of this study was to identify the cause of postoperative temporal hollowing.
Methods: Patients requiring a coronal incision were recruited prospectively.
Each side of the head in all patients was randomized to suprafascial, subfascial,
or deep dissection. An unmarked envelope containing the type of dissection to
of all groups. The incidence and severity of temporal hollowing 6 months
postoperatively were measured clinically and by computed tomographic volume
Results: Twenty-seven patients with 54 sides (18 suprafascial, 15 subfascial, and
21 deep) completed the study. There were no demographic differences among
the three groups. The incidence and severity of temporal hollowing were lowest
with suprafascial dissection. Other factors associated with the presence of tem-
to the frontal nerve in any of the dissections. Postoperative temporal hollowing
was associated with surgical approach and postoperative weight loss.
Conclusion: Elevation of a coronal flap in the suprafascial plane and minimi-
zation of patient weight loss may decrease the incidence of postoperative tem-
poral hollowing.(Plast. Reconstr. Surg. 121: 379e, 2008.)
exposure to the anterior and lateral craniofacial
skeleton.1A common complication of the coronal
a concavity or hollowing of the temporal region.2,3
Hollowing occurs within 6 months postoperatively
nal exposure is unknown.
he coronal incision is commonly used by mul-
tiple surgical specialties for traumatic, recon-
structive, and cosmetic procedures to provide
During coronal flap elevation, dissection in the
frontal and temporal regions places the fron-
nerve travels in a plane under the temporoparietal
fascia (or superficial temporal fascia) just above the
superficial temporal fat pad.4–6As a result, it has
been suggested that dissection should proceed
through this fat pad to protect the frontal nerve.6
However, this dissection may cause fat pad injury
and/or displacement, which can result in temporal
also been noted as potential causes of hollowing.7
Lacey et al. used magnetic resonance imag-
ing to show that temporal hollowing was caused
From the Division of Plastic and Reconstructive Surgery,
University of Western Ontario.
Received for publication July 10, 2007; accepted October 2,
Presented at the XIth International Congress of the Inter-
national Society of Craniofacial Surgery, in Queensland,
Australia, September 11 through 14, 2005, and at the 60th
in Quebec City, Quebec, Canada, June 2006.
Disclosure: Neither of the authors has any commer-
cial associations or financial disclosures that might
pose or create a conflict of interest with information
presented in this article.
by either a decrease in temporal fat pad volume
or prolapse. The authors therefore recom-
mended dissection immediately under the su-
perficial layer of the deep temporal fascia to
minimize dissection through the fat pad and
prevent temporal hollowing.7To date, no stud-
ies have been performed to test this assumption.
Previous work has identified the neurovascular
the vasculature of the fat pad includes perforators
from the deep and middle temporal arteries; the
zygomaticotemporal nerve travels through the sub-
divide the fat pad into several lobes.8Differing sur-
gical approaches may interrupt these structures to
varying degrees, which may influence the incidence
of temporal hollowing.
Several surgical approaches through the tem-
poral fossa have been described.1,2,4,5,9The three
commonest approaches that allow access to the
zygomaticofrontal suture and zygomatic arch in-
clude (1) dissection over the superficial layer of
the deep temporal fascia (suprafascial); (2) dis-
section immediately deep to the superficial layer
of the deep temporal fascia without violation of
the superficial temporal fat pad (subfascial); and
(3) dissection within the substance of the super-
ficial temporal fat pad (deep) (Fig. 1).
No studies have documented the incidence
of temporal hollowing following any of these
approaches. Therefore, the purpose of this
study was, first, to identify the route of dissection
through the temporal fossa that results in the
lowest incidence and severity of temporal hol-
lowing; and, second, to identify the cause of
postoperative temporal hollowing.
PATIENTS AND METHODS
A prospective, randomized, controlled trial of
all patients requiring coronal flap elevation was
performed. All patients presented to Victoria Hos-
pital, London Health Sciences Centre, London,
Ontario, Canada, between January 1, 2003, and
December 31, 2005. Full approval was received
from the Research Ethics Board at the London
Health Sciences Centre and at the University of
Western Ontario, London, Canada, before the
start of the study. Written and verbal informed
consent was obtained from all patients.
Inclusion criteria included adult patients
(?16 years of age) with a clinical condition re-
quiring coronal flap elevation for access to the
anterior and lateral facial skeleton. Patients with
previous coronal flap elevation or facial surgery
were excluded. In addition, patients who were
deemed not capable of giving consent such as
head injury trauma patients or intubated patients
of three groups: (1) suprafascial dissection, above
the superficial layer of the deep temporal fascia; (2)
subfascial dissection, below the superficial layer of
the deep temporal fascia but not into the fat pad;
and (3) deep dissection, into the superficial tempo-
ral fat pad. The randomization technique involved
the selection of an unmarked envelope containing
All envelopes contained equal distributions of all
groups randomized for each side of the face.
Demographic data collection included gender,
age, weight at the time of enrollment, height, and
diagnosis. In patients with traumatic injuries, direct
traumatic injury to the temporal fossa, high- versus
low-velocity injuries, and isolated versus multisystem
injuries were recorded. In addition, data including
operative procedure, extent of lateral dissection (zy-
gomaticofrontal suture versus zygomatic arch), in-
jury of any nerves or arteries within the temporal
dissection, duration of surgery, and postoperative
Fig. 1. Schematic depiction of the anatomy of the temporal
tion. The blue line represents the route of the subfascial dissec-
tion. The dotted black line represents the route of the deep dis-
deep temporal fascia, and the dark yellow compartment repre-
sents the superficial temporal fat pad. TB, temporal bone; TM,
temporalis muscle; Z, zygomatic arch; SC, subcutaneous tissue
(reflected back); blue oval, frontal branch of the facial nerve.
Plastic and Reconstructive Surgery • June 2008
complications (infection, facial nerve weakness, he-
matoma, and seroma) were recorded.
Postoperative data at 2 weeks, 6 weeks, and 6
months included the presence of temporalis mus-
cle contraction, interincisal mouth opening, and
weakness or loss of the frontal branch of the facial
nerve. Body mass index was calculated preopera-
tively and at the 6-month follow-up in all patients.
In addition, clinical temporal hollowing was as-
10 ? severe temporal hollowing) and was per-
formed by the senior surgeon (D.B.M.), who was
blinded to the level of dissection per side, at 6
All coronal exposures were performed by the
extended to the postauricular region, allowing ac-
cess to the zygomatic arch. Dissection was per-
plane to a level 2.5 cm from the palpable supraor-
bital rims. The periosteum was incised with the
scalpel at this point centrally to minimize injury to
the supraorbital sensory nerves. Subperiosteal dis-
section with a periosteal elevator continued to the
supraorbital rims anteriorly.
Laterally at the superior temporal fusion line,
dissection down to the deep temporal fascia with
a no. 15 scalpel blade was performed until the
superior extent of the superficial temporal fat pad
was identified. The dissection was then altered
depending on the randomization of each side
within each patient. Great care was taken to avoid
injury to the neurovascular structures superficial
minimize heat injury to adjacent tissues including
the frontal branch of the facial nerve. The lateral
orbit down to the frontozygomatic suture and zy-
gomatic body was exposed through a subperios-
teal dissection using a periosteal elevator. The lat-
eral periosteum during this dissection was incised
with a scalpel after it was elevated off of the bone,
allowing connection to the temporal dissection.
The temporal muscle itself was not reflected from
and lateral orbital rim and wall. Only a periosteal
dissection was performed in this area, allowing
appropriate visualization and alignment of frac-
tures. This was performed in all trauma patients
whether or not rigid fixation was placed at the
frontozygomatic suture to minimize variability of
the dissection between sides. All elevated perios-
was resuspended with sutures before closure to
ensure that no space between the lateral orbit and
temporal muscle existed. This was performed to
minimize the risk of hollowing secondary to mus-
cle malposition. It should also be noted that re-
traction of the coronal flap was performed at all
times with right-angled retractors and not with
rake retractors to further reduce the risk of a trac-
tion injury to the facial nerve.
During suprafascial dissection, the superficial
scalpel dissection at the superior extent of the fat
pad. Blunt dissection then proceeded on top of
a periosteal elevator. In this manner, dissection to
the palpable zygomatic arch was performed in a
bloodless fashion without need of sharp dissec-
tion. Once the arch was palpable, the superficial
leaf of the deep temporal fascia and the fat pad
above the palpable upper border of the zygomatic
arch to avoid frontal nerve injury. A periosteal
elevator was then used within the fat pad to find
and elevate the periosteum at the upper border of
the arch itself. The dissection performed in this
manner minimizes injury to the fat pad and also
protects the nerve appropriately (Fig. 1). Both
arterial and venous hemostasis during dissection
within the fat pad was achieved meticulously using
bipolar cautery only. The superficial layer of the
deep temporal fascia, when violated, was resus-
pended with sutures in all patients irrespective of
group before skin closure.
Computed Tomographic Imaging and Volume
All patients underwent preoperative facial
computed tomographic scanning. Similar scans
were obtained 1 day and 6 months postoperatively
to allow for isolation and measurement of the
performed on a GE CT Lightspeed VCT machine
encompassed the skull vertex to the tip of the
mentum. Segmentation of the fat pad was per-
formed for each axial slice by the same individual.
These images were then used to measure the vol-
ume of the superficial temporal fat pad on the
Advantage Workstation (AW 4.2_05) using Vol-
Healthcare). This program counts the number of
pixels in each outline and multiplies the total by
the cubic voxel size (in cubic millimeters) to cal-
Volume 121, Number 6 • Postoperative Temporal Hollowing
culate the volume of each axial slice outline. All
slice volumes are then summed to obtain the total
volume of each fat pad.
Initial sample size calculations indicated that
with an alpha error of 0.05 and a statistical power
of 0.80, 60 patients would be required to detect a
statistically significant difference in postoperative
temporal hollowing among the three dissection
groups. This was determined assuming that clin-
ically detectable postoperative temporal hollow-
ing occurs in 30 percent of patients.
The data were applied to distribution curves.
Data analysis was performed using analysis of vari-
ance, Kruskal-Wallis, and chi-square tests to de-
termine differences between groups for clinical
and radiologic measurements of temporal hollow-
ing. Nonparametric analysis was also performed
using the median test. Correlation testing was per-
formed to determine whether any patient vari-
ables had an influence on temporal hollowing. A
value of p ? 0.05 was considered statistically sig-
nificant throughout the analysis.
Interim statistical analysis was performed at
the midpoint of the study to determine whether a
difference in postoperative temporal hollowing
was present between groups to ensure that no
harm was being done. Thirty consecutive patients
were included in the analysis, resulting in 60 ran-
domized sides. Twenty-eight of the 30 patients
required coronal flap elevation for acute facial
fracture repair. Eleven patients had low-velocity
injuries (altercation or fall from a height ?5 feet)
and 17 had high-velocity injuries (motor vehicle
collision, explosives, fall from a height ?5 feet).
Twelve patients had multisystem injuries and 11
required admission to the intensive care unit. No
patients had penetrating injuries or lacerations
within the temporal regions preoperatively. Two
of the 30 patients had coronal flap elevation for
benign tumor extirpation (Table 1).
There were 25 male patients. Average patient
age was 36 years (range, 16 to 87 years) at the time
of surgery. A total of 27 patients (54 sides) com-
pleted the study. Three patients were lost to fol-
Table 1. Demographic Data for All Patients*
ICU, intensive care unit; N/A, not available.
*Note that, in the randomization columns, 1 represents suprafascial dissection, 2 represents subfascial dissection, and 3 represents deep
dissection. Also note that patients 17, 26, and 29 were lost to follow-up and not included in the analysis of results.
Plastic and Reconstructive Surgery • June 2008
low-up and were not included in the analysis (Table
kept on a soft nonchewing diet for 4 weeks postop-
eratively to rest the muscles of mastication and re-
duce pain during the healing process.
Eighteen of 54 sides had a suprafascial dissec-
tion, 15 had a subfascial dissection, and 21 had a
deep dissection. There was no statistical difference
between groups with respect to age, medical/surgi-
cal history, preoperative body mass index, length of
surgery, or mechanism of injury (Table 2). Both the
middle temporal artery (p ? 0.02) and the zygomat-
icotemporal nerve (p ? 0.34) were transected more
the artery reached statistical significance. There was
no injury or weakness to the frontal branch of the
facial nerve in any group. Two patients were re-
debridement of an abscess and the other for evac-
uation of a hematoma (Table 2). All patients had
normal interincisal mouth opening (?40 mm) 6
Definite trends existed in the incidence and
average severity of temporal hollowing among
the three treatment arms. Fifty percent of pa-
tients with a suprafascial dissection had some
amount of temporal hollowing, compared with
76 percent of patients with a subfascial and 71
percent with deep dissections. These differences
did not reach statistical significance with direct
group comparisons (p ? 0.20) (Table 2). How-
ever, a significant difference between groups
was measured with the median test (Fig. 2).
Significantly more patients with a dissection that
was deep to the superficial layer of the deep
temporal fascia (subfascial and deep dissec-
tions) had temporal hollowing that was worse
than the median for the entire study population.
In other words, the suprafascial dissection re-
sulted in patients having significantly less tem-
poral hollowing compared with the median (p ?
0.009). Mean severity of temporal hollowing, as
measured by the surgeon using a visual analogue
scale (0 ? none and 10 ? severe), was 1.8 for
suprafascial dissections, 3.9 for subfascial dis-
sections, and 3.1 for deep dissections (p ? 0.09).
The presence of temporal hollowing corre-
lated significantly with a change in body mass in-
dex over the study period. A decrease in 6-month
an increase in temporal hollowing severity (p ?
0.02). The severity of temporal hollowing also in-
creased as the computed tomographic volume of
the fat pad at 6 months postoperatively decreased,
but this was not significant (p ? 0.27) (Table 3).
The suprafascial approach caused the least
amount of trauma to the superficial temporal
fat pad and its neurovascular supply. This ap-
proach also resulted in the lowest severity of
postoperative temporal hollowing compared
with the other approaches. In addition, the su-
prafascial dissection did not result in facial
nerve injury and was used safely to expose the
lateral facial skeleton.
Numerous fibrous septations divide the fat
pad into discrete sections.8These septa attach to
both the superficial and deep layers of the deep
temporal fascia. Subfascial and deep dissections
therefore by definition disrupt these septa. De-
spite resuspension of the superficial layer of the
deep temporal fascia after these dissections, tem-
poral hollowing still occurred in corresponding
Table 2. Perioperative Findings Divided by the Type of Dissection at the Superficial Temporal Fat Pad
Type of Dissection
4:53 ? 2:05
5:52 ? 2:08
No. of dissections
High-velocity injuries (%)
Mean ? SD anesthesia time (hr:min)
Dissection to the zygomatic arch (%)
Fascia resuspended (%)
ZTN transected (%)
MTA transected (%)
Incidence of clinical temporal hollowing (%)
Mean ? SD severity of clinical temporal following
Mean ? SD preoperative BMI (kg/m2)
Mean ? SD postoperative BMI (kg/m2)
ZTN, zygomaticotemporal nerve; MTA, middle temporal artery; VAS, visual analogue scale; BMI, body mass index; N/A, not applicable.
*A value of p ? 0.05 was considered statistically significant.
4:49 ? 1:52
1.8 ? 2.9
24.6 ? 3.3
25.5 ? 3.4
3.9 ? 3.4
24.9 ? 3.1
24.0 ? 2.7
3.1 ? 3.3
24.8 ? 3.8
25.2 ? 3.3
Volume 121, Number 6 • Postoperative Temporal Hollowing