BASIC AND PATIENT-ORIENTED RESEARCH

J Oral Maxillofac Surg

67:2559-2570, 2009

Computer-Assisted Planning,
Stereolithographic Modeling, and
Intraoperative Navigation for Complex
Orbital Reconstruction: A Descriptive
Study in a Preliminary Cohort
R. Bryan Bell, DDS, MD,* and
Michael R. Markiewicz, DDS, MPH
Purpose: Post-traumatic or postablative enophthalmos and diplopia and/or facial asymmetry resulting

from inaccurate restoration of orbital anatomy remain relatively frequent sequellae of complex orbital
reconstruction. Recently, preoperative computer-assisted planning with virtual correction and construction of stereolithographic models have been combined with intraoperative navigation in an attempt to
more accurately reconstruct the bony orbit and optimize treatment outcomes. The purpose of the
present study is to review the authors early experience with computer planning, stereolithographic
modeling, and intraoperative navigation in a series of patients who underwent surgical treatment for a
variety of complex post-traumatic and postablative orbital deformities.
Patients and Methods: The investigators initiated a retrospective chart review, and a sample of
patients was derived from the population of patients at Legacy Emanuel Hospital, Portland, OR, between
2007 and 2008. Each patients anatomy was assessed in multiplanar (axial, coronal, sagittal) and
3-dimensional computed tomography (CT) hard-tissue views; virtual correction was made using the
uninjured or anatomically correct side by creating a mirror image that was superimposed on the
traumatized side. The internal orbit was reconstructed with the previously contoured titanium mesh. The
external orbital frame was reduced or repositioned and stabilized using 1.3-mm and/or 1.5-mm titanium
plates and screws. The patients position was identified with a digital reference frame that was fixed to
an adhesive mask. Intraoperative navigation was then used to assess the accuracy of the restored internal
and external orbital anatomy by assessing various points on the virtual image at the workstation. All
patients received a postoperative CT scan, and the preoperative and postoperative images were compared and subjectively analyzed. To be included in the sample, patients must have undergone reconstruction for complex primary or secondary unilateral orbital deformities secondary to traumatic injury
or ablative procedure using computer-assisted treatment during the study enrollment period. Criteria for
using computer-assisted navigation were unilateral, clinically significant disruption of the internal and/or
external orbit, that involved more than one orbital wall and that resulted in or had the potential to result
in enophthalmos, diplopia, ocular dysmotility, or facial asymmetry. Patients excluded from the review
were those who underwent orbital reconstructing using traditional (non computer-assisted) techniques.
Demographic, etiological, treatment, and outcome variables were recorded and analyzed. Outcome
measures included globe position, ocular motility, facial symmetry, and complications. Poor outcome
*Attending Head and Neck Surgeon and Director of Resident
Education, Oral and Maxillofacial Surgery Service, Legacy Emanuel
Hospital and Health Center; and Clinical Associate Professor of Oral
and Maxillofacial Surgery, Oregon Health & Science University,
Portland, OR.
Resident, Department of Oral and Maxillofacial Surgery, Oregon
Health & Science University, Portland, OR.

Address correspondence and reprint requests to Dr Bell:

1849 NW Kearney, Suite 300, Portland, OR 97209; e-mail: bellb@
hnsa1.com
2009 American Association of Oral and Maxillofacial Surgeons

0278-2391/09/6712-0002$36.00/0
doi:10.1016/j.joms.2009.07.098

2559

2560

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

was defined as clinically perceptible enophthalmos, persistent dipolopia, facial asymmetry/malar flattening, or ocular dysmotility.
Results: Fifteen consecutive patients with complex primary or secondary unilateral post-traumatic and
postablative orbital deformities received computer-assisted treatment. Anatomic restoration of internal
and external orbital contours was obtained in all but 1 patient based on a comparison of preoperative and
postoperative CT scans. Further evaluation of the postoperative CT images compared favorably to the
virtually planned reconstructions. Despite favorable restoration of internal and external bony anatomy,
the soft-tissue limitations were not completely overcome in some patients with secondary deformities.
Suboptimal correction of globe projection occurred in three patients undergoing secondary enophthalmos repair because of severe, intraconal, soft-tissue scarring posterior to the equator of the globe.
Complications occurred in 4 patients.
Conclusions: Preoperative computer modeling and intraoperative navigation provides a useful guide
for and presumably more accurate reconstruction of complex orbital injuries and postablative orbital
defects. Although probably not necessary for routine use in small orbital blowout fractures, its use in a
shattered orbit or high-velocity injury resulting in severe disruption of the internal and external orbit
shows promise.
2009 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 67:2559-2570, 2009
Fractures involving the orbit are present in as many as
50% of patients with craniomaxillofacial injuries who
present to trauma centers in the United States.1 The
ultimate goal in the treatment of complex orbital
injuries is to restore form and function by reconstructing the preinjury 3-dimensional architecture of the
orbital frame and internal orbit and re-establishing
orbital volume. This is accomplished using craniofacial techniques that have been developed and refined
during the last 50 years.2-4 These techniques use a
variety of autogenous5-8 and alloplastic9-15 materials
that result in acceptable outcomes for most patients.
Despite notable successes, post-traumatic enophthalmos16-19 and diplopia20,21 resulting from inaccurate
restoration of orbital anatomy remain relatively frequent sequellae of complex orbital reconstruction.
The most common cause of enophthalmos and
long-term ocular dysmotility is enlargement of the
bony orbit with herniation of the orbital contents into
a neighboring cavity.22-26 It has been estimated that
increased orbital volume after traditional reconstructive treatment occurs in approximately 8.5% of patients with orbital trauma.19 High-velocity injuries are
especially challenging to the reconstructive surgeon because of limited access, poor visualization, complex
anatomy, and post-traumatic edema or scarring. These
factors often lead to difficulty in identifying and reconstructing orbital bony landmarks, particularly the postero-medial orbital bulge and orbital apex (Fig 1).27
Recently, preoperative computer-assisted planning
with virtual correction and construction of stereolithographic models have been combined with intraoperative navigation in an attempt to more accurately
reconstruct the bony orbit and optimize treatment
outcomes.28-30 Gellrich et al31 described the use of a

navigation system for computer-assisted preoperative

planning with virtual reconstruction to obtain symmetry of the orbits and intraoperative control of
virtual contours in unilateral post-traumatic orbital
deformities. Surgical procedures were preplanned
with virtual correction by mirroring an individually
defined, 3-dimensional segment of the unaffected
side into the deformed side, creating an ideal unilateral reconstruction. These computer models
were then used intraoperatively as a virtual template to navigate the preplanned bony contours and
globe projection (Fig 2). A number of authors have
subsequently refined these techniques by including
customized preforming of titanium meshes and
point-to-point computer-assisted navigation based
upon stereolithographic models with navigation
markers.32-39 The purpose of the present study is to
review the authors early experience with preoperative computer stereolithographic modeling and intraoperative navigation in a series of patients who
underwent surgical treatment for a variety of posttraumatic and postablative orbital deformities.

Patients and Methods

STUDY DESIGN AND SAMPLE

The investigators initiated a retrospective chart review. Because of the small sample size, statistical
analyses would be unreliable and misleading; therefore we will describe our early experience in this
patient population. A sample of patients was derived
from the population of patients at Legacy Emanuel
Hospital, Portland, OR, between 2007 and 2008. The
senior author (R.B.B.) was the primary surgeon for all
patients. This study was approved by the Institutional

2561

BELL AND MARKIEWICZ

EXCLUSION CRITERIA

Patients excluded from the review were those who

underwent orbital reconstructing using traditional
(non computer-assisted) techniques.
STUDY VARIABLES

FIGURE 1. Factors leading to difficulty identifying and accurately reconstructing orbital bony landmarks. A, Sagittal CT scan
demonstrating the normal ascending slope of the posterior orbit
(left) and the common surgical error (right) of inadequate restoration of the height of the posterior orbit. B, Axial CT scan
demonstrating the normal postero-medial orbital bulge (left,
red), and the common surgical error (right, red) of inadequate
restoration of the postero-medial bulge. Green line represents
optimal orbital contour.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

Review Board at Legacy Emanuel Hospital and Health

Center.
INCLUSION CRITERIA

To be included in the sample, patients must have

undergone reconstruction for complex primary or
secondary unilateral orbital deformities secondary to
traumatic injury or ablative procedure using computer-assisted treatment during the study enrollment period. Criteria for using computer-assisted navigation
were unilateral, clinically significant disruption of the
internal and/or external orbit, involving more than
one orbital wall or an isolated orbital roof, and that
either resulted in or had the potential to result in
enophthalmos, diplopia, ocular dysmotility, or facial
asymmetry.

Study variables included the following: patient age,

gender, etiology of injury (motor vehicle accident
[MVA], tumor, assault, equestrian accident), primary
diagnosis (secondary enophthalmos with or without
malunion of orbital bones, fibrous dysplasia, orbital
blowout fracture, orbito-zygomaticomaxillary complex
fracture, initial injury complex (isolated, panfacial, zygomaticomaxillary complex [ZMC], naso-orbital-ethmoid
[NOE], Le Fort III), history of previous surgery (yes/no),
internal orbital injury (yes/no), external orbital injury
(yes/no), history of preoperative diplopia (yes/no), presence of postoperative diplopia (yes/no), history of preoperative ocular dysmotility (yes/no), postoperative ocular dysmotility (yes/no), preoperative globe projection
(perceptible enophthalmos, normal, proptosis, both),
postoperative globe projection (perceptible enophthalmos, normal, proptosis, both), preoperative facial
symmetry (normal, projected malar prominence, flattened malar prominence), postoperative facial symmetry (normal, projected malar prominence, flattened malar prominence), complication (dysmotility,
decreased visual acuity, inaccurate plate placement,
and entropion). Poor outcome was defined as clinically perceptible enophthalmos, persistent dipolopia,
facial asymmetry/malar flattening, or ocular dysmotility.
PROCEDURE

Spiral computed tomography datasets (Summatome

4; Siemens, Irvine, CA), 1-mm collimation/slice
thickness, 2-mm table feed, 1-mm increment, were
acquired for computer-assisted preoperative planning
and construction of a stereolithographic model before
surgery. Each patients anatomy was assessed in multiplanar (axial, coronal, sagittal) and 3-dimensional
computed tomography (CT) hard-tissue views, and
virtual correction was made using the uninjured or
anatomically correct side by creating a mirror image
that was superimposed on the traumatized side (Intellect Cranial Navigation System, Stryker Navigation,
Kalamazoo, MI) (Fig 3). Stock titanium mesh (Stryker,
Kalamazoo, MI; Synthes, Paoli, PA) was then preformed to the stereolithographic model (Medical
Modeling, Denver, CO; Synthes) to reproduce optimal
orbital contour. Cases that involved the external orbital frame and necessitated bone graft reconstruction
were outlined appropriately on the models and normal anatomy used to guide recontouring of the harvested calvarial bone grafts.

2562

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

FIGURE 2. Components of a surgical navigation system. Intraoperative navigation is comparable to GPS systems commonly used in
automobiles and is composed of 3 primary components: a localizer, which is analogous to a satellite in space; an instrument or surgical
probe, which represents the track waves emitted by the GPS unit in the vehicle; and a CT scan dataset that is analogous to a road map.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

At the time of surgery, all patients were approached

via a transconjunctival incision alone or combined
with an upper blepharoplasty or coronal approach,
depending upon the clinical scenario. The internal
orbit was reconstructed with the previously contoured titanium mesh. The external orbital frame was
reduced or repositioned and stabilized using 1.3-mm
and/or 1.5-mm titanium plates and screws (Stryker;
Synthes). Intraoperative navigation was then used to
assess the accuracy of the restored internal and
external orbital anatomy (Surgical Tool Navigation
System, Stryker Navigation, Kalamazoo, MI). Intraoperative navigation was carried out by means of a
frameless stereotaxy with 3 infrared cameras controlling the pointer via integrated light-emitting diodes
(LEDs). The patients position was identified with a
digital reference frame that was fixed to an adhesive
mask (Fig 4). The mask had a total of 31 LEDs that it
used for registration. A minimum of 21 LEDs were
used to achieve optimal registration accuracy. Various
points on the virtual image at the workstation and
the patient were matched and compared with anatomic landmarks. An acceptable margin of error was
defined as less than 1 mm. If a less than 1-mm margin
of error could not be obtained or a coronal flap
needed to be raised, then the registration was made
using a fixed skeletal reference tool. Proper position
of the bony segments and internal orbit were confirmed in sequential fashion according to the following systematic protocol: malar eminence, infraorbital

rim, lateral orbital rim, orbital floor, medial internal

orbit/postero-medial orbital bulge, lateral internal orbit, posterior orbit/orbital apex, globe projection.
All patients received a postoperative CT scan,
and the preoperative and postoperative images
were compared and subjectively analyzed by the
primary author (R.R.B.). Differences in internal orbital contour or malar projection between the virtually planned and the surgically achieved results
were noted. Each patient was evaluated for the
presence or absence of diplopia, ocular motility,
globe projection, and facial symmetry. Complications were defined as a return to the operating
room or suboptimal results resulting from the operation itself, such as inaccurate plate position or
visual disturbance. Suboptimal restoration of enophthalmos was, for purposes of this study, not
considered a complication. Descriptive statistics
were derived using Microsoft Excel (Microsoft, Redmond, WA).

Results
Fifteen consecutive patients with complex primary
or secondary unilateral post-traumatic and postablative orbital deformities received computer-assisted
treatment. Patient demographics, injury characteristics, and outcome measures are presented in Table 1.
The age of the 15 patients included in the study
ranged from 17 to 67 years (mean 37.3 yrs); 11 were

BELL AND MARKIEWICZ

2563
(n 5). Three patients had large but isolated orbital
blowout fractures, and 2 patients had orbital defects
as part of an orbito-zygomaticomaxillary complex
fracture. Anatomic restoration of internal and external
orbital contours was obtained in all but one patient
based on a comparison of preoperative and postoperative CT scans. Further evaluation of the postoperative CT images compared favorably to the virtually
planned reconstructions.
Despite favorable restoration of internal and external bony anatomy, the soft-tissue limitations were not
completely overcome in some patients with secondary deformities. Suboptimal correction of globe projection occurred in three patients undergoing secondary enophthalmos repair due to severe intraconal softtissue scarring posterior to the equator of the globe.
In each of these patients, the globe could not be
projected without significant inferior rotation, and
the patients also had persistent diplopia.
Complications occurred in 4 patients, 3 of whom
required reoperation: one patient because of intraconal scarring, limited ocular motility and persistent
diplopia; another because of visual changes secondary to plate encroachment on the optic nerve; and
another because of inaccurate mesh placement, despite apparently appropriate positioning on navigated
images. The fourth patient, who had undergone sec-

FIGURE 3. Anatomic assessment in multiplanar (axial, coronal,

sagittal) and 3-dimensional computed tomography (CT) hard-tissue
views. A, uncorrected image. B, Virtual correction made using the
uninjured or anatomically correct side by creating a mirror image
that was superimposed on the traumatized side (Intellect Cranial
Navigation System; Stryker Navigation, Kalamazoo, MI).
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

male and 3 were female; and 6 had undergone previous orbital reconstructive surgery. Most patients were
the victims of high-speed motor vehicle accidents
(n 8), and 6 patients had previously sustained
severe orbital injuries involving the internal orbit and
external buttresses as part of more extensive facial
fractures. At the time of enrollment into this study,
these 6 patients had undergone previous surgery using conventional techniques, and all of these had
undergone navigation-assisted enophthalmos repair
between 6 months and 3 years after the original injury. The remaining 8 patients had undergone primary
reconstruction, either at the time of orbito-zygomatic
ablation (n 2) or within 1 week of being injured

FIGURE 4. Digital reference frame fixed to an adhesive mask. The

mask had a total of 31 light-emitting diodes (LEDs) that it used for
registration. A minimum of 21 LEDs were used to achieve optimal
registration accuracy.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

2564

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

Table 1. PATIENT CHARACTERISTICS

Patient

Age
(yrs)

Gender

Previous
Surgery

Internal
Orbit

External
Orbit

Preoperative
Diplopia

Preoperative
Diplopia

1
2

48
24

M
M

MVA
Tumor

Secondary enophthalmos
Fibrous dysplasia

Panfacial fxs
NA

Yes
No

Yes
Yes

No
Yes

Yes
No

Yes
No

3
4
5

54
28
40

M
M
M

MVA
Assault
Tumor

Orbital blowout fx
Orbital blowout fx
Fibrous dysplasia

Isolated
Isolated
NA

No
No
No

Yes
Yes
No

No
No
Yes

Yes
Yes
No

No
No
No

30

MVA

Secondary enophthalmos

Yes

Yes

No

Yes

No

M
F
M
M
F

MVA
MVA
Assault
Assault
MVA

Yes
No
No
Yes
No

Yes
Yes
Yes
Yes
Yes

Yes
No
Yes
No
Yes

Yes
Yes
Yes
Yes
NA

NA
No
No
No
NA

20

MVA

Panfacial fxs

Yes

Yes

Yes

Yes

Yes

13

34

MVA

Le Fort III

Yes

Yes

Yes

Yes

Yes

14
15

31
46

M
F

Assault
Equestrian
accident

Secondary enophthalmos
Orbital blowout fx
ZMC fx
Secondary enophthalmos
Secondary enophthalmos and
unrepaired ZMC
Secondary enophthalmos and
malunion
Secondary enophthalmos and
malunion
ZMC fx
Orbital roof blowout fx

Panfacial fxs, globe

rupture
Panfacial fxs
Isolated
Isolated
ZMC
NOE w/globe disruption

7
8
9
10
11

46
17
67
35
46

12

Isolated
Isolated

No
No

Yes
Yes

Yes
Yes

Yes
Yes

No
No

Etiology

Primary Diagnosis

Concomitant Facial
Injuries

Abbreviations: F, female; fx, fracture; M, male; MVA, motor vehicle accident; NA, not available; NOE, naso-orbital-ethmoid;
ZMC, zygomatic-maxillary complex.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

ondary enophthalmos correction, developed lower

lid retraction with increased scleral show, despite
significant improvement in orbital volume and globe
projection. Diplopia was improved after revisional
surgery in the first patient. The second and third
patients recovered completely after revision and had
favorable functional and esthetic outcomes.

Discussion
Preoperative computer modeling and intraoperative navigation provides a useful guide for accurate
restoration of form and function in individuals with
congenital, developmental, and acquired deformities
of the craniomaxillofacial skeleton. This investigation
reviews the authors experience with computerassisted surgery in a series of patients who underwent
surgical treatment for a variety of complex post-traumatic and postablative orbitozygomatic deformities.
All patients were treated with a protocol that included preoperative stereolithographic models and
intraoperative navigation. All patients underwent reconstruction, with adequate restoration of form and
function. Although 1 patient required revisional surgery for extraocular muscle entrapment, the restoration of orbital volume was achieved with a seemingly
improved degree of predictability. The second patient
who required reoperation did so because the complication overaggressive placement of the mesh plate in
an attempt to completely restore the posteromedial

aspect of the internal orbit, resulting in encroachment

upon the optic nerve and transient loss of visual
acuity. She recovered completely after revision. The
third patient who required revisional surgery presented a more problematic case because the complication appeared to be the result of inaccurate patient
registration. Despite apparently favorable plate placement confirmed by intraoperative navigation, the postreduction CT scan demonstrated suboptimal, lateral
placement of the mesh plate. It is unclear why this
occurred, but it is presumed that error was introduced into the patient registration by either data collection or patient soft-tissue swelling or both. The
most likely explanation is that there was a gap of
several days between the CT scan used during the
navigation procedure (the scan was obtained on admission to the hospital) and the time of surgical correction. It is likely that swelling in this patient contributed to this inaccuracy. This complication has not
occurred in any other patient to date. If a patient
develops significant postinjury edema, a new CT scan
image is obtained the same day as the navigated procedure, according to the above-mentioned protocol.
The goals of treatment for complex orbital deformities include re-establishment of a structural framework; normalization of orbital volume; prevention
of complications, such as enophthalmos, extraocular
muscle restriction, and visual disturbances; and
achievement or improvement of facial esthetics. Most
routine, low-velocity injuries can be successfully man-

2565

BELL AND MARKIEWICZ

Table 1. CONTINUED
Preoperative Postoperative
Ocular
Ocular
Dysmotility Dysmotility
Yes
No

Yes
No

Yes
No
No

No
No
No

No

Preoperative Globe
Projection

Postoperative Globe
Projection

Preoperative
Facial Symmetry

No

Perceptible enophthalmos Perceptible enophthalmos Normal

Normal
Normal
Excessive malar
prominence
Proptosis
Normal
Normal
Perceptible enophthalmos Normal
Normal
Normal
Normal
Excessive malar
prominence
Perceptible enophthalmos Normal
Normal

Yes
Yes
Yes
Yes
No

Yes
No
No
No
No

Perceptible enophthalmos
Proptosis
Proptosis
Perceptible enophthalmos
Globe rupture

No

Postoperative
Facial
Symmetry

Complication

Outcome

Normal
Normal

Ocular dysmotility
None

Poor
Good

Normal
Normal
Normal

None
None
None

Good
Good
Good

Normal

None

Good

Perceptible enophthalmos Malar depression

Normal
Normal
Normal
Malar depression
Normal
Normal
NA
Malar depression

Normal
Normal
Normal
Normal
Normal

None
Decreased visual acuity
None
Entropion
Inaccurate plate placement

Fair
Fair
Good
Fair
Fair

No

Perceptible enophthalmos perceptible enophthalmos Malar depression

Normal

None

Fair

No

No

Perceptible enophthalmos Normal

Malar depression

Normal

None

Fair

No
Yes

No
No

Proptosis
Proptosis

Malar depression
Normal

Normal
Normal

None
None

Good
Good

Normal
Normal

aged by conventional techniques via well-established

periorbital approaches using various alloplastic and
autogenous materials without the use of custom implants or navigation-assisted techniques. Futhermore,
navigation is not a substitute for proper surgical technique. However, patients with complex orbital injuries involving the deep orbit posterior to the equator
of the globe, large internal defects, or external frame
defects may indeed benefit from these technological
advancements.
High-velocity injuries often result in a shattered
orbit with large-volumetric increases internally,
massive herniation of periorbital contents into the
surrounding anatomic spaces, and cranial neuropathies. Although advances in craniomaxillofacial surgical approaches and biological materials have improved our ability to more predictably restore these
patients form and function, a significant number of
these individuals will still require revisional surgery
despite the best efforts of an experienced surgeon.
When the entire orbit is disrupted and there are no
posterior bony landmarks to guide the reconstruction, accurate positioning of bone grafts or mesh
plates becomes problematic. There is difficulty in
establishing proper orbital contour, volume, and
medial bulge projection, as well as risk to encroachment upon the orbital apex and optic nerve. Presurgical planning using stereolithographic models
to establish proper plate contour, as well as intraoperative navigation to ensure accurate and safe

positioning of the plate in a poorly visualized anatomic region, affords even the experienced surgeon
greater confidence and predictability in the deep
orbit. Furthermore, patients who may require
overcorrection to counteract the effects of fat
atrophy and to ensure adequate long-term globe
projection can undergo treatment planning by assessing both skeletal contour and globe projection
with intraoperative navigation (Figs 5, 6).
In addition to navigating the orbital apex, computerized planning and prebent orbital mesh have
the potential to predictably restore the difficult-toaccess posterior medial bulge region, also known as
the key area. Recently, Metzger et al33 have described a semiautomatic procedure for individual
preforming of titanium meshes for orbital fractures.
By using CT scan data, the topography of the orbital
floor and wall structures can be recalculated. After
mirroring the unaffected side onto the affected
side, the defect can be reconstructed virtually. Data
for the individual virtual model of the orbital cavity
are then sent to a template machine that reproduces the surface of the orbital floor and medial
walls automatically. A titanium mesh can then be
adjusted preoperatively for exact 3-dimensional reconstruction, and an individually milled titanium
implant is used with navigation to guarantee intraoperative placement of the preformed mesh for an
even more precise reconstruction.

2566

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

FIGURE 5. A, Postoperative coronal CT scan after initial orbital repair of a patient with severely disrupted orbital fractures demonstrating
inaccurate plate placement posterior to the equator of the globe. Note increased orbital volume. B, Postoperative axial CT scan after initial
orbital repair, demonstrating inadequate restoration of the postero-medial orbital bulge and significantly increased orbital volume. C,
Calvarial bone graft construct at the time of secondary enophthalmos repair. D, Bone graft try-in using stereolithographic model. E, Bone
graft inset into patient. (Figure 5 continued on next page.)
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

Manipulation of the CT datasets in the current

study was limited to mirror imaging of the uninvolved
side. A major limitation of this technique is that it
assumes orbital/facial symmetry. Studies have shown,
however, that there are measurable differences in
orbital volume for any given individual.40 Even more
apparent, there is tremendous variation in facial
symmetry among individuals. The recent introduction of computer-assisted design and computerassisted manufacturing (CAD/CAM) software pro-

vides the surgeon an opportunity to perform customized virtual manipulations of the CT datasets rather
than relying on mirror imaging. An anatomic structure or region, such as an orbit or orbito-zygomatic
complex, can be virtually repositioned into true
orthogonal planes, segmented, mirrored, osteotomized, or contoured. Once completed, a stereolithographic model can then be milled and a custom
implant constructed. Intraoperative navigation can
then be used to confirm accurate skeletal reposi-

2567

BELL AND MARKIEWICZ

FIGURE 5 (contd). F, Intraoperative navigation demonstrating accurate placement of bone graft construct along the medial orbital wall
based upon a mirror image (red) of the opposite (unaffected) side. G, Intraoperative navigation demonstrating accurate placement of bone
graft construct posterior to the equator of the globe along the antral bulge. H, Postoperative coronal CT scan demonstrating favorable
restoration of orbital volume. I, Postoperative axial CT scan demonstrating favorable restoration of orbital volume posterior to the equator of
the globe.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

tioning of the virtually reconstructed segments, as

described by Pham et al.36
Klein and Glatzer39 described an individual CAD/
CAM-fabricated glass bio-ceramic implant for similar
purposes. A glass bio-ceramic implant was shaped
from CT coordinates with CAD/CAM. In this prefabrication process, the implants were milled out of a
solid block of Bioverit II and used for accurate reconstruction of the orbital floor.
Finally, complex postablative or post-traumatic
conditions that involve the external orbital frame can
be secondarily reconstructed using a point-to-point
computer-assisted navigation for more precise transfer of planned periorbital osteotomies from the stereolithographic model into reality. Klug et al34 described this approach in five patients in whom
osteotomies and repositioning of the zygomatic complex were planned using stereolithographic models.
The desired position of the zygoma in the patient was
determined by fixing individualized osteosynthesis
plates to predefined screw positions. The stereolithographic model and the patient were registered to the

same three-dimensional CT dataset via an occlusal

reference frame on the patient and corresponding
reference markers on the model. Prebent osteosynthesis plates from the surgical simulation on the
model were fixed to corresponding screw positions
on the patient, which were located by computerassisted navigation.
In conclusion, use of preoperative computer modeling and intraoperative navigation provides a useful
guide for, and presumably more accurate reconstruction of, complex orbital injuries and postablative orbital defects. Although probably not necessary for
routine use in small orbital blowout fractures, its use
in treating a shattered orbit or high-velocity injury
resulting in severe disruption of the internal and external orbit shows promise. As computer-assisted surgery continues to evolve and new software programs
are developed, it is thought that even more predictable replication of virtually reconstructed orbital anatomy will be possible in the future. The introduction
of added layers of monetary expense must be justified
by documentation of improved outcomes. Additional

2568

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

FIGURE 6. Thirty-year-old woman involved in a high-speed motor vehicle accident 1 year before presentation for right enophthalmos repair.
The patient had undergone repair of panfacial fractures at the time of her accident; however, the right orbit was not reconstructed because
of a ruptured globe. The globe was repaired and the patient recovered baseline vision. A, Preoperative appearance demonstrating right
enophthalmos and mild upper lid ptosis. B, Preoperative worms eye view. C, Preoperative coronal CT scan demonstrating increased orbital
volume and large defects of the inferior, medial, and posterior orbital walls. D, Stereolithographic model used to analyze defect and prebend
titanium plate. E, Virtual reconstruction of the patient with mirror image of the opposite orbit and intraoperative navigation confirming
accurate plate placement, favorable contour of the orbit plate, and adequate globe projection. (Figure 6 continued on next page.)
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

2569

BELL AND MARKIEWICZ

FIGURE 6 (contd). F, Postoperative coronal CT scan demonstrating anatomic restoration of orbital volume with plate. G, Postoperative
sagittal CT image demonstrating favorable restoration of the posterior slope of the orbit and safe plate placement at the orbital apex. H,
One-year postoperative appearance of patient. I, Worms eye view 1 year postoperatively.
Bell and Markiewicz. Planning of Complex Orbital Reconstruction. J Oral Maxillofac Surg 2009.

clinical investigations are necessary, therefore, to further examine outcomes and to define the ideal indications for this promising technology.

References
1. Bell RB: The role of oral and maxillofacial surgery in the trauma
care center. J Oral Maxillofac Surg 65:2544, 2007
2. Hammer B: Orbital Fractures: Diagnosis, Operative Treatment,
Secondary Corrections. Seattle, WA, Hogrefe and Huber, 1995
3. Manson PN, Hoopes JE, Su CT: Structural pillars of the facial
skeleton: An approach to the management of Le Fort fractures.
Plast Reconstr Surg 66:54, 1980

4. Gruss JS, Bubak PJ, Egbert MA: Craniofacial fractures. An algorithm to optimize results. Clin Plast Surg 19:195, 1992
5. Converse JM, Smith B: Reconstruction of the floor of the orbit
by bone grafts. Arch Ophthalmol 44:1, 1950
6. Converse JM, Smith B, Obear MF, et al: Orbital blowout fractures: A ten year survey. Plast Reconstr Surg 39:20, 1967
7. Antonyshyn O, Gruss JS, Galvraith DJ, et al: Complex orbital
fractures: A critical analysis of immediate bone graft reconstruction. Plast Reconstr Surg 22:220, 1989
8. Constantian MB: Use of auricular cartilage in orbital floor reconstruction. Plast Reconstr Surg 69:951, 1982
9. Browning C, Walker RV: The use of alloplastics in 45 cases
of orbital floor reconstruction. Am J Ophthalmol 60:684,
1965

2570
10. Romano JJ, Iliff NT, Manson PN: Use of Medpor porous polyethylene implants in 140 patients with facial fractures. J Craniofac Surg 4:142, 1993
11. Polley JW, Ringler SL: The use of Teflon in orbital floor reconstruction following blunt facial trauma: A 20 year experience.
Plast Reconstr Surg 79:39, 1987
12. Sargent LA, Fulks DK: Reconstruction of internal orbital fractures with vitallium mesh. Plast Reconstr Surg 88:31, 1991
13. Rubin PAD, Bilyk FR, Shore JW: Orbital reconstruction using
porous polyethylene sheets. Ophthalmology 101:1697, 1994
14. Bedrossian EJ: Banked fascia lata as an orbital floor implant.
Ophthal Plast Reconstr Surg 9:66, 1993
15. Cordewener FW, Bos RRM, Rozema FR, et al: Poly(L-lactide)
implants for repair of human orbital floor defects: Clinical and
MRI evaluation of long term results. J Oral Maxillofac Surg 54:9,
1996
16. Mathog R, Hillstrom RP, Nesi FA: Surgical correction of enophthalmos and diplopia: A report of 38 cases. Arch Otolaryngol
Head Neck Surg 115:169, 1989
17. Kawamoto HK, Jr: Late posttraumatic enophthalmos: A correctable deformity? Plast Reconstr Surg 69:423, 1982
18. Fries R: Some problems in therapy of traumatic enophthalmos.
Mod Probl Ophthalmol 14:637, 1975
19. Hammer B, Prein J: Correction of post-traumatic orbital deformities: Operative techniques and review of 26 patients. J Craniomaxillofac Surg 23:81, 1995
20. Wojno TH: The incidence of extrocular muscle and cranial
nerve palsy in orbital floor blow-out fractures. Ophthalmology
94:682, 1987
21. Mathog RJ, Archer KF, Nesi F: Post-traumatic enophthalmos
and diplopia. Otolaryngol Head Neck Surg 94:69, 1989
22. Manson PN, Grivas A, Rosenbaum A, et al: Studies on enophthalmos II: The measurement of orbital injuries by quantitative
computed tomotgraphy. Plast Reconstr Surg 77:203, 1986
23. Bite U, Jackson IT, Forbes GS, et al: Orbital volume measurements in enophthalmos using three-REF CT imaging. Plast Reconstr Surg 75:502, 1985
24. Dolynchuk V, Tadjalle H, Manson P: Orbital volumetric analysis: Clinical application in orbitozygomatic injuries. J Cranio
Maxillofac Trauma 2:56, 1996
25. Raskin E, Millman A, Lubkin U: Prediction of late enophthalmos
by volume analysis of orbital fracture. Ophthal Plast Reconstr
Surg 14:19, 1998
26. Kolk A, Pautke C, Schott V, et al: Secondary post-traumatic
enophthalmos: High resolution magnetic resonance imaging
compared with multislice computed tomography in postoperative orbital volume measurement. J Oral Maxillofac Surg 65:
1926, 2007

PLANNING OF COMPLEX ORBITAL RECONSTRUCTION

27. Hammer B, Kunz C, Schramm A, et al: Repair of complex
orbital fractures: Technical problems, state of the art solutions
and future perspectives. Ann Acad Med Singapore 28:687,
1999
28. Watzinger F, Wanschitz F, Wagner A, et al: Computer-aided
navigation in secondary reconstruction of post-traumatic deformities of the zygoma. J Craniomaxillofac Surg 25:198, 1997
29. Ewers R, Schicho K, Undt G, et al: Basic research and 12 years
of clinical experience in computer-assisted navigation technology: A review. Int J Oral Maxillofac Surg 34:1, 2005
30. Schramm A, Schon R, Rucker M, et al: Computer-assisted oral
and maxillofacial reconstruction. J Comput Informations Technol CIT 14:71, 2006
31. Gellrich NC, Schramm A, Hammer B, et al: Computer-assisted
secondary reconstruction of unilateral posttraumatic orbital
deformity. Plast Reconstr Surg 110:1417, 2002
32. Schmelzeisen R, Gellrich NC, Schoen R, et al: Navigation-aided
reconstruction of medial orbital wall and floor contour in
craniomaxillofacial reconstruction. Inj Int J Care Injured 35:
955, 2004
33. Metzger MC, Schon R, Schulze D, et al: Individual preformed
titanium meshes for orbital fractures. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 102:442, 2006
34. Klug C, Schicho K, Ploder O, et al: Point to point computer
assisted navigation for precise transfer of planned zygoma
osteotomies from the stereolithographic model into reality.
J Oral Maxillofac Surg 64:550, 2006
35. Westendorff C, Kaminsky J, Ernemann U, et al: Image-guided
sphenoid wing meningioma resection and simultaneous computer-assisted cranioorbital reconstruction: Technical case report. Operative. Neurosurgery 60(Suppl 1):E174, 2007
36. Pham AM, Rafii AA, Metzger MC, et al: Computer modeling and
intraoperative navigation in maxillofacial surgery. Otolaryngol
Head Neck Surg 137:624, 2007
37. Schicho K, Figi M, Seemann R, et al: Accuracy of treatment
planning based on stereolithography in computer assisted surgery. Med Phys 33:3408, 2006
38. Fan X, Zhou H, Lin M, et al: Late reconstruction of the complex
orbital fractures with computer-aided design and computeraided design and computer-aided manufacturing technique. J
Craniofac Surg 18:665, 2007
39. Klein M, Glatzer C, Cad I: CAM fabricated glass bioceramic
implants in reconstructive surgery of the bony orbital floor.
Plast Reconstr Surg 117:565, 2006
40. Forbes G, Gehring DG, Gorman CA, et al: Volume measurements of normal orbital structures by computed tomographic
analysis. Am J Roentgenol 145:149, 1985