Head Hisao Hajime
Tonami, Yokota,
MD MD
#{149} Itaru #{149} Tetsuya
Twenty-six patients with orbital fractures diagnosed with plain radiography and computed tomography were examined with surface coil magnetic resonance (MR) imaging. Fifteen patients had blow-out fractures, and 11 had maxillofacial complex fractures. In all patients with blow-out fractures, the location of the fracture was precisely indicated by the presence of prolapsed orbital fat. Incarceration of the extraocular muscle or orbital fat was correctly diagnosed with MR imaging, which was less sensitive in depicting maxillofacial fractures but was useful in assessment of soft-tissue involvement. Postoperative follow-up MR studies provided valuable information about the cause of motility impairment. While Ti-weighted images are useful for the detection of the fracture site, both Ti- and T2weighted images are usually necessary for evaluating soft-tissue lesions. The results of this study mdicate that surface coil MR imaging is an important adjunct procedure in the diagnosis and treatment of orbital fractures.
Surface
P
Index
terms: Face, fractures, 24.1214, 24.418. Magnetic resonance (MR), surface coils #{149} Orbit, fractures, 22.414, 22.418 #{149} Orbit, injuries, 22.414, 22.49 #{149} Orbit, MR studies, 22.1214. Trauma, 22.414, 22.418, 22.49 179:789-794
MR
radiography, tomography, and computed tomography (CT) are the established modes of evaluating orbital fractures (1-15). Recently, magnetic resonance (MR) imaging with surface coils has come to play an important role in orbital imaging (16-19). However, orbital fractures have received relatively little attention in the literature and have been described in only a few cases of blow-out fractures (20, 21). In this artide, we discuss our experience with surface coil MR imaging of patients with orbital fractures and attempt to clarify the current role and limitations of this technique. PATIENTS
AND
Twenty-six
patients,
METHODS aged
9-58
years,
with orbital fractures diagnosed with plain radiography and CT underwent examination with surface coil MR imaging.
Polytomography
was performed
All of the patients had obvious trauma to the orbit. Twenty patients were male and six were female; 15 had blowout fractures (medial wail, n = 6; floor,
n
=
and
6; and medial 1 1 had
tures.
wall and floor,
maxilbofacial
n
complex
CT and MR examinations
3)
frac-
were
per-
within a 3-day period in each and the time of MR examination
case, alter trauma varied from 2 days to 3 years. Sixteen patients underwent surgery. Of these, 10 had undergone postoperative follow-up MR examinations for the evaluof motility
impairment.
CT was performed
with
a third-genera-
tion scanner DR3; Siemens
(Somatom 2 or Somatom Medical Systems, Erlangen, Germany). A 2-mm collimator was used, and sections were obtained at 2-mm intervals. Both axial and direct coronal scanning was performed in 14 patients, and axial scanning only was performed in 12
patients. From the Department of Radiology, Kanazawa Medical University, Daigaku 1-1, Uchinada, Kahoku, Ishikawa 920-02, Japan. From the 1989 RSNA scientific assembly. Received September 24, 1990; revision requested November 16; revision received January 10, 1991; accepted January 16. Address reprint requests to H. Tonami. c RSNA, 1991 I
In all of these
12 patients
in
whom direct coronal scanning was not performed, computer-reformatted images in the coronal or oblique sagittal plane parallel to the axis of the inferior rectus muscle were obtained from the axial sections.
MR examinations a 0.5-T
whole-body
were imaging
performed
with
system
(Mag-
Radiology
Tamamura, Okimura,
MD MD
Imaging’ netom;
Siemens
Medical
Systems).
The
head coil was used for transmission only, and a surface coil 10 cm in diameter was placed over the orbit to act as a receiver. Ti-weighted spin-echo (SE) images with a repetition time (TR) of 400-600 msec and an echo time (TE) of 30-40 msec were obin all cases,
tamed
and
T2-weighted
im-
ages (TRITE = 1,600-2,000/70-90) were added in 20 cases. Each section was 5 mm thick with a 0-2.5-mm intersection gap. The
matrix
four signals
size
was
were
256
X 256,
averaged.
tiple images in the ing the entire orbit
and
coronal plane were obtained.
to
two
Initially,
mul-
coverThere-
after, additional images in other planes, either the axial plane along the course of the medial rectus muscle or the oblique sagittal
plane
parallel
to the
axis
of the
in-
terior rectus muscle, were obtained. All patients were asked to keep their eyes closed during the examination and to refrain from eye movement. RESULTS
in 10 pa-
tients.
ation 1991;
Coil
LAIN
formed
Radiology
Neck
Yamamoto, MD #{149} Masao Matsuda, MD #{149} Hiroyasu Nakagawa, MD #{149} Akira Takarada, MD #{149} Tetsuro
Fractures:
Orbital
and
Blow-out
Fractures
Seventeen fractured surfaces in 14 patients were evaluated with MR imaging. One other patient who underwent examination after surgery only was not included. In all cases, the location of fracture was precisely mdicated on Ti-weighted images by the presence of orbital fat that had prolapsed into the adjacent air-containing paranasal sinuses (Figs 1-3). The displaced bone segments could be seen indirectly as a signal void between prolapsed orbital fat and intrasinus
hemorrhage
or mucosal
thick-
ening in six cases (Fig 1). MR imaging depicted enlargement of extraocular muscle in seven cases (Figs 1, 3), kmnking of muscle toward the bone defect in 10 cases (Figs 1-3), and discontinuity of muscle at the bone defect in two cases (Fig 2). Sur-
Abbreviations: SE time, TR = repetition
spin time.
echo,
TE
echo
789
gery
revealed
these
two
of muscle Orbital shape
muscle
incarceration
with
a discontinuity
cases
in
at MR imaging. fat prolapsed in a saccubar
was
present
in three
patients
at
.-lI-.--
,_4
MR imaging (Fig 2). Surgery disclosed incarceration of probapsed orbital fat at the bone defect in these three patients, all of whom had poor prognoses with restriction of ocular motility even after surgery. Intrasinus
hemorrhage
strated tients.
was
at MR
Complex
Twenty-seven patients
pa-
only
surfaces
evaluated
with
Five other examination
were
not
patients after
included.
27 fractured
b.
Fractures
fractured were
imaging. derwent the
demon-
in five
a.
Maxillofacial
six
well
imaging
surfaces
in MR
who surgery
un-
,
Sixteen
of
(nine
of the
‘
blow-out type and seven of the separated linear type) were well shown, but the 1 1 others were not detected
with MR imaging (Figs 4, 5). Associated soft-tissue abnormalities, including two cerebral contusions (Fig 5), five intrasinus hemorrhages (Fig 6), one intraorbital hematoma (Fig 6), and
one
well
orbital
shown
emphysema
with
Postoperative Studies
imaging. gery and
was
with
The
intervals
disclosed
1.
Blow-out
fracture
of medial
orbital
wall.
(a) Axial
and
(b) direct
coronal
CT scans
show relatively extensive blow-out fracture of medial orbital wall on left (arrows). Medial rectus muscle (arrowheads) is medially displaced. MR imaging was performed 25 days after trauma. (c) Ti-weighted MR image (SE 600/30) clearly shows the fracture (arrows) and the medial rectus muscle (arrowheads) because of bright signal from orbital fat. Although the medial rectus muscle is enlarged and medially displaced, the entire course of the muscle can be seen. (d) Coronal Ti-weighted MR image (SE 600/30) shows the displaced bone segment (arrows) as a signal void between the prolapsed orbital fat and the mucosal thickening of ethmoid sinus. Surgery was not performed because this patient’s clinical symptoms were not serious.
assessed
postoperative
MR
between
MR imaging In six of these
imaging
MR
impairment
10 patients
days.
were
imaging.
Follow-up
Motility in
MR
d.
C.
Figure
sur-
were 17-112 10 patients, MR
postoperative
complications: two scar tissue tions (Fig 7), three inappropriate plants (Fig 3), and one orbital
formaimabscess
(Fig 8). A second operation was performed in four patients after MR imaging, and remarkable improvement of motility was noted in all of these patients. In two other patients with intraorbitab pair was patients,
was
scar tissue, performed. diagnosis
not the
made
on the
basis
surgical reIn these two of scar tissue
of low
signal
at follow-up
MR
2.
Blow-out
imaging was performed 2 days after traushows blow-out fracture of orbital floor on right. Orbital fat is prolapsed in a saccular shape, and inferior rectus muscle (arrowheads) is herniated together with orbital fat. (b) Oblique sagittal Ti-weighted MR image (SE 600/ 40) shows a discontinuity of muscle at the bone defect (arrow). The orbit was explored 3 weeks after trauma, and incarceration of both muscle and orbital fat was confirmed. (Reprinted, with permission, from reference 21.)
ma. (a) Coronal
intensity on T2-weighted images and more than 5 months of stable appear-
ance
Figure
examinations.
fracture
Ti-weighted
of orbital
MR image
floor.
MR
(SE 600/35)
DISCUSSION Orbital
fractures
can
usually
be
vi-
suabized with plain radiography with a 28#{176} Caldwebl view and a Waters view (9). Fractures that are not seen on plain radiographs can almost a!ways
be
localized
or CT scans 790
#{149} Radiology
(10).
with
Of these
tomographic
two,
direct
oblique
sagittal
or direct
coronal
CT
scans have proved more useful (14,6-8,10-15). Direct oblique sagittal scans, however, have several limitations (14,15). The positioning of the patient is sometimes difficult, requir-
ing better technical assistance. It is also difficult to maintain the position of the patient’s head in a scanner with a small gantry aperture. As with direct coronal scans, artifacts from dental fillings can be a problem (1,4).
June
1991
nificant
:)
‘
A ‘-:
improvement
of the
signal-
to-noise ratio, thereby allowing thinner sections and higher spatial resolution, comparable with those of CT, without the expense of an increase in imaging time (16-18). Until now, however, orbital fractures have received little attention in the litera-
:t
ture. It has been believed that MR imaging is not suited to the evalua-
tion of bony lesions because of the inability to demonstrate cortical bone b.
a.
itself (17,18).
Although small bone fragments were not well visualized on MR images in our series of blow-out fractures,
the
4 ,
tion
,/ ‘
;
:
about
A
‘
rare,
b. Le Fort II and III fractures. (a) Computed radiograph shows extensive fractures of all orbital walls (arrows). MR imaging was performed 3 days after trauma. (b) Coronal Tiweighted MR image (SE 500/30) demonstrates five of eight fractures, including diastatic fractures of both frontozygomatic sutures (arrows), and blow-out fractures (arrowheads) of the left floor and roof of both orbits. MR imaging failed to demonstrate fractures of the right and
medial
wall
of both
orbits.
but
when
in which direct oblique sagittal or direct coronal scans are not obtained, computer-reformatted images in the coronal or oblique sagittal plane are sometimes useful, but there is an obvious loss of spatial resolution with computer-reformatted images
Volume
(5).
Furthermore,
the
179
3
#{149} Number
risk
of
harmful ionizing radiation to the lens is not negligible with CT (22). Clinical application of MR imaging to the orbit has several advantages, including absence of ionizing radiation, multiplanar capability, and excellent tissue contrast. In addition, the use of surface coils enables sig-
MR images, at the bone
suggestive Two patients
of musin our
it is substantiated,
We speculate
may be entrapped
that
this
find-
derived from the fact that muscle is usually ebon-
gated and its axis is perpendicular to the fractured orbital wall (4). Furthermore, small defects, which tend to be associated with muscle incarceration, may accentuate this finding (24). On the other hand, it has been sugthat
dipbopia
ocular motility the entrapment
and
limitation
of
usually result from of orbital fat through
the bone defect (3,25). Orbital fat contains numerous fibrous bands that connect the muscle sheath to the periosteum. Increased tension of these fibrous bands after trauma has been suggested as a cause of these symptoms (3,25). On MR images, or-
bital In cases
seen on of muscle
early repair is necessary to prevent tissue necrosis (14,23). One might postulate that discontinuity suggests not an incarceration but a laceration or tear of the muscle. Laceration or tear, however, seldom occurs in orbital fractures, especially in blow-out
gested
4.
between
study had this MR finding, and the presence of muscle incarceration was confirmed at surgery. Actual incarceration of the extraocular muscle is
ing the
floor
relationship
abnormalities discontinuity
fractures.
Figure
of prolapsed
and the extraocular musthe findings of muscle
defect was highly cle incarceration.
Figure 3. Inappropriate implant used for the reconstruction of blow-out fracture. (a) Axial and (b) coronal Tl-weighted MR images (SE 600/35) obtained 5 days after trauma show blow-out fracture of medial orbital wall at left (arrows). Medial rectus muscle (arrowheads) is enlarged and medially displaced. The defect was reconstructed with autogenous bone from the iliac crest 15 days after trauma. Postoperative MR examination was performed because of the exacerbation of motility impairment. (c) Coronal Ti-weighted MR image (SE 600/35) obtained 17 days after surgery shows that the autogenous bone (arrows) is inclined medially. The autogenous bone was removed during the second operation, and a silicone plate was inserted. (d) Coronal Ti-weighted MR image (600/35) obtained 70 days after the second operation shows the adequately reconstructed medial orbital wall. Note the silicone plate (arrowheads), seen as a region of hypointensity. (Reprinted, with permission, from reference 21.)
.4
the
the fracture cle. Among
d.
.‘
identification
hyperintense orbital fat enabled bocalization of the fracture site. MR imaging also provided useful informa-
fat probapsed
is clinically indicates
defect. had tion
fat
Three
patients
this MR finding, of the orbital fat
herniated
and
shape
strongly
incarceration
surgery in these been postulated tents
in a saccular
important
at the
in our and was
incarcerafound at
three patients. that the orbital through
bone
series
the
defect
Radiology
It has conat #{149} 791
the time of impact. As the force of the blow dissipated, the bony fragments tended to return toward their
original
position,
thus
the herniated three patients
orbital
incarcerating fat
(26).
In
all
who had incarceration of orbital fat, satisfactory improvement of ocular motility could not be achieved even after surgical repair. The interval between trauma and surgery in these patients was 16-21 days. As with the muscle incarceration, therefore, early surgical repair in patients orbital fat
with the probapsed
MR finding in a saccular
shape is recommended. The need for and timing cal treatment of blow-out still
controversial.
Most
of
of surgifractures
is
surgeons
now agree that a blow-out fracture not a surgical emergency and delay the decision to operate for 10-14 days,
after
which
repair
is
is performed
selectively in cases of persistent diplopia or enophthalmos (27,28). In our experience, however, one exception is an incarceration of extraocular muscle
or orbital
fat.
Early
repair
of
the defect should be performed in such cases, and MR imaging is extremely useful in the diagnosis of these soft-tissue conditions. Detection of the fracture sites in maxillofacial complex fractures at MR imaging remains limited. Fractures at which orbital fat was not prolapsed were sometimes missed at MR imaging. In our series, 1 1 of 27 fractured surfaces (41%) were not detected at MR imaging. However, even in these cases, MR imaging in assessing soft-tissue just as it was for blow-out
was useful involvement, fractures.
Cerebral contusion, epidural ma, intraorbital hematoma,
hematointra-
d.
C.
5. Fractures of orbital roof and floor. (a) Direct coronal fractures of orbital roof (arrows) and floor (arrowheads)
Figure
shows
bital canal soft-tissue
at right. (b) abnormalities.
Direct coronal CT scan (c) MR imaging was (SE 500/30) fails to demonstrate
weighted image MR image (SE 2,000/90) roof fracture.
Figure
6.
Le Fort
clearly
II and
shows
III fractures.
cerebral
CT scan
settings of infraorwith soft-tissue settings does not demonstrate performed 54 days after trauma. Coronal Tieither fracture. (d) Coronal T2-weighted contusion (arrows) associated with orbital
with
with
bone
involvement
MR
imaging was performed 27 days after trauma. Oblique sagittal Ti-weighted MR image (SE 500/40) shows hyperintense intraorbital hematoma (solid arrows) extending from subfrontal epidural hematoma (arrowheads) through the bone defect. Note hyperintense intrasinus hemorrhage (open arrows) in the maxillary sinus.
p
sinus hemorrhage, and orbital emphysema, all of which are commonly
associated with orbital fractures, were well visualized at MR imaging. In evaluating the deep portion of the orbit,
however,
gion
of interest
must
be
kept
theoretical tio
the
because
small-diameter
distance when
ings
lesions
indicate
especially
cations,
it may
of the
surface
re-
coil the
in signal-to-noise
duced as this (18). Therefore, tures,
the
in mind,
gain
from
distance
from
coils
rais re-
is increased clinical findin deep
intracranial be better
struc-
complito use
a stan-
dard head coil in place of or in addition to the surface coil (17). MR imaging also provided valuable information about the condition of the orbit in patients with postoperative motility impairment. In our series
of
tive
motility
792
10 patients
#{149} Radiology
impairment,
with
causes of motility impairment were seen at MR imaging in six cases. In our
MR
imaging
seems
short T2 and the low density of water protons in collagen (29-32). Relative hypointensity of scar tissue on T2weighted images allows it to be distinguished from soft-tissue edema,
hemorrhage, thermore, radiation
follow-up
or muscle tissue. Furlack of harmful ionizing to the
studies
eye
enables
repeated
to be performed
to
In summary, ing,
with
and
its
surface multiplanar
coil
MR
capability
imag-
excellent
olution,
soft-tissue
is
useful
extremely
contrast resin evalu-
ation of orbital fractures. In blow-out fractures, MR imaging may be performed soon after adequate plain radiography. It is our impression that CT may not be necessary
safely.
postopera-
obvious
experience,
be extremely valuable in the evaluation of scar tissue formation. It is known that on T2-weighted images mature fibrosis usually has a low signal intensity, caused by the very
for
fractures equal
the
evaluation
because to CT
in the
of blow-out
MR
imaging
detection
is of the
June
1991
Figure
7. Scar autogenous coronal CT
with rect
tissue formation after surgery for blow-out fracture. In this case, bone 8 days after trauma. CT and MR imaging were performed scan shows soft tissue with abnormally high attenuation (arrows)
sagittal
Ti-weighted
MR image
(SE 500/40),
muscle
is obliterated
(arrowheads).
(c) On
area
(arrows).
The
diagnosis
soft-tissue
structure
coronal T2-weighted was made on the
of scar tissue
is present MR
basis
image
the defect in the orbital floor at right was 51 days after surgery because of persistent between the globe and the orbital floor.
part of the orbital
in the anterior (SE, 2,000/90),
of persistent
the
low signal
lesion
intensity
reconstructed diplopia. (a) (b) On oblique
floor (arrows).
is demonstrated
on T2-weighted
The inferior
as a very images for
Di-
rectus
hypointense 5 months.
8. Orbital abscess after the reconstruction of Le Fort II and III fractures. In this case, the reconstruction of the orbital floor at right was performed 52 days after trauma, and a silicone sheet was inserted. Follow-up CT and MR imaging were performed 87 days after surgery because the patient complained of external strabismus and diplopia. (a) Direct coronal CT scan shows a well-circumscribed extraconal mass (arrows) displacing the inferior rectus muscle (arrowheads) superomedially. Note a large defect in the orbital floor. (b) On oblique sagittal Ti-weighted MR image (SE 500/40), the mass (arrows) is slightly hyperintense relative to muscle and hypointense relative to fat. The globe and the optic nerve are displaced upward. In b and c, arrowheads point to hypointense rim. (c) On coronal T2-weighted MR image (SE 2,000/90), the mass (arrows) is extremely hyperintense to fat. Surgical findings confirmed the mature abscess with thick capsule. Figure
fracture site the assessment malities.
On plays
the
the
and
other
is superior of soft-tissue
hand,
in
MR imaging
a complementary
evaluation
to CT abnor-
role
to CT
of maxilbofacial
in
corn-
plex fractures. CT is superior to MR imaging in the detection of fracture sites in maxiblofacial fractures. MR imaging, however, provides valuable information about soft-tissue lesions. Situations in which MR imaging is preferable in detection of maxilbofacial fractures include unexplained neurobogic deficits, visual or extraocular muscle impairment, and fractures with a high probability of intracranial extension.
preferable
in
cases of postoperative pairment. MR imaging
MR
imaging
motility provides
imuse-
ful
postoperative
information
Volume
179
is also
about #{149} Number
3
complications vided by
that
may
not
be
pro-
CT.
In regard to our suggested aging procedure, multiple
MR coronal
images should
orbit (21).
covering be obtained
the
allows visualization cross sections of all and of all extraocular
after,
entire initially
of tangential four orbital muscles.
we recommend
obtaining
im-
This
walls There-
addi-
tional images in another plane, either the axial plane along the course of the medial rectus muscle or the oblique sagittal plane parallel to the axis of the inferior rectus muscle.
These
additional
images
provide
more precise information about relationship between the muscle the bone defect. While Ti-weighted images are useful for localizing
fracture site, both Tied images are usually
the and a
and T2-weightnecessary for
the evaluation of cerebral contusion, hemorrhagic lesions, and scar tissue formation (21). Although a strict comparison between surface coil MR imaging and direct multipbanar high-resolution
CT should be made believe that surface
in the future, we coil MR imaging
is an important adjunct in the nosis and treatment of orbital tures. U
diagfrac-
Acknowledgments: We thank Kazuyuki Sasaki, MD, Department of Ophthalmology; Sadao Tsukada, MD, Department of Plastic Surgery; and Kohichi Yamashita, MD, Department of Otolaryngology, Kanazawa Medical University, for their valuable contributions. We also thank Akiko Ohta for secretarial assistance; Masao Yonezawa, RT, Tomokazu Oku, RT, and Osamu
and
Yamashita,
James
C. Ehrhardt,
RI,
for
technical
PhD,
assistance;
for reviewing
the
manuscript.
Radiology
#{149} 793
12.
References 1.
2.
3.
Grove AS Jr, Tadmor R, New PFJ, et al. Orbital fracture evaluation by coronal computed tomography. Am J Ophthalmol 1978; 85:679-685. Zilkha A. Computed tomography of blow-out fracture of the medial orbital wall. AJR 1981; 137:963-965. Hammerschlag SB, Hughes 5, O’Reilly GV, Naheedy MH, Rumbaugh CL. Blowout fractures of the orbit: a comparison of computed tomography and conventional radiography with anatomical correlation. Radiology
4. 5.
6.
7.
8.
9.
10.
ii.
794
Zilkha
1982;
13.
14.
15.
A.
#{149} Radiology
SM,
Orbital
blowout
Lagouros PA, et al. the prognostic significance of computed tomography. Ophthalmology 1985; 92:1523-1528. Gentry LR, Smoker WRK. Computed tomography of facial trauma. Semin Ultrasound CT MR 1985; 6:129-144. Koornneef L, Zonneveld FW. The role of direct multiplanar high resolution CT in the assessment and management of orbital trauma.
i43:487-492.
Computed tomography in facial trauma. Radiology 1982; i44:545-548. Brant-Zawadzki MH, Minagi H, Federle MP, et al High resolution CT with image reformation in maxillofacial pathology. AJR 1982; 138:477-483. Cooper PW, Kassel EE, Gruss JS. High resolution CT scanning of facial trauma. AJNR i983; 4:495-498. Gentry LR, Manor WF, Turski PA, et al. High resolution CT analysis of facial struts in trauma. I. Normal anatomy. AJR 1983; 140:523-532. Gentry LR, Manor WF, Turski PA, et al. High resolution CT analysis of facial struts in trauma. II. Osseous and soft tissue complications. AJR 1983; 140:533-541. Dolan KD, Jacoby CC, Smoker W. The radiology of facial fractures. RadioGraphics 1984; 4:576-663. Kreipke DL, Moss JJ, Franco JM, Maves MD, Smith DJ. Computed tomography and thin-section tomography in facial trauma. AJNR 1984; 5:185-189. Johnson DH Jr. CT of maxillofacial trauma. Radiol Clin North Am 1984; 22:131144.
Gilbard
16.
Radiol
Mafee
MF,
24.
fractures:
Clin
North
Am
1987;
25.
26.
25:
753-766. Ball JB Jr. Direct oblique sagittal CT of orbital wall fractures. AJNR 1987; 8:147154. Sullivan JA, Harms SE. Surface-coil MR imaging of orbital neoplasms. AJNR 1986;
27.
28.
7:29-34. 17.
Atlas
SW,
LT, Zimmerman RA, HI, Grossman RI. Orbit: initial experience with surface coil spin-echo MR imaging at 1.5 T. Radiology 1987; 164:501-509. Bilaniuk LI, Atlas SW, Zimmerman RA. Magnetic resonance imaging of the orbit. Radiol Clin North Am 1987; 25:509-528. Hosten N, Sander B, Cordes M, Schubert CJ, Sch#{244}rnerW, Felix R. Graves ophthalmopathy: MR imaging of the orbits. Radiology 1989; i72;759-762. McArdle CB, Amparo EG, Mirfakhraee M. MR imaging of orbital blow-out fractures. J Comput Assist Tomogr 1986; 10:116-119. Tonami H, Nakagawa I, Ohguchi M, et al. Surface coil MR imaging of orbital blowout fractures: a comparison with reformatted CT. AJNR 1987; 8:445-449. Lund E, Halaburt M. Irradiation dose to the lens of the eye during CT of the head. Neuroradiology 1982; 22:181-184. Koornneef L. Current concepts on the management of orbital blow-out fractures. Ann Plast Surg 1982; 9:185-200.
Hackney
18.
19.
20.
21.
22.
23.
Bilaniuk
DB, Goldberg
29.
30.
31.
32.
Smith B, Petrelli R. Fractures of the orbit: blow-out and nasoorbital fractures. In: Freeman HM, ed. Ocular trauma. New York: Appleton-Century-Crofts, 1979; 117-123. Hammerschlag SB, Hughes S, O’Reilly GV, Weber AL. Another look at blow-out fractures of the orbit. AJNR 1982; 3:331335. Greenwald HS Jr. Keeney AH, Shannon GM. A review of 128 patients with orbital fractures. Am J Ophthalmol 1974; 78:655-664. Converse
JM,
Smith
B.
On
the
treatment
of blow-out fractures of the orbit. Plast Reconstr Surg i978; 62:100-104. Crikelair GF, Rein JM, Potter GD, et al. A critical look at the “blow-out” fracture. Plast Reconstr Surg 1972; 49:374-379. Glazer HS, Lee JKT, Levitt RG, et al. Radiation fibrosis: differentiation from recurrent tumor by MR imaging. Radiology 1985; 156:721-726. Sundaram M, McGuire MH, Schajowicz F. Soft tissue masses: histologic basis for decreased signal (short 12) on T2-weighted MR images. AJR 1987; 148:1247-1250. Ebner F, Kressel HY, Mintz MC, et al. Tumor recurrence versus fibrosis in the female pelvis: differentiation with MR imaging at 1.5 T. Radiology 1988; 166:333340. Negendank WG, Al-Katib AM, Karanes C, Smith MR. Lymphomas; MR imaging contrast characteristics with clinicalpathologic correlations. Radiology 1990; 177:209-216.
June
1991
Tonami, Yokota,
MD MD
#{149} Itaru #{149} Tetsuya
Twenty-six patients with orbital fractures diagnosed with plain radiography and computed tomography were examined with surface coil magnetic resonance (MR) imaging. Fifteen patients had blow-out fractures, and 11 had maxillofacial complex fractures. In all patients with blow-out fractures, the location of the fracture was precisely indicated by the presence of prolapsed orbital fat. Incarceration of the extraocular muscle or orbital fat was correctly diagnosed with MR imaging, which was less sensitive in depicting maxillofacial fractures but was useful in assessment of soft-tissue involvement. Postoperative follow-up MR studies provided valuable information about the cause of motility impairment. While Ti-weighted images are useful for the detection of the fracture site, both Ti- and T2weighted images are usually necessary for evaluating soft-tissue lesions. The results of this study mdicate that surface coil MR imaging is an important adjunct procedure in the diagnosis and treatment of orbital fractures.
Surface
P
Index
terms: Face, fractures, 24.1214, 24.418. Magnetic resonance (MR), surface coils #{149} Orbit, fractures, 22.414, 22.418 #{149} Orbit, injuries, 22.414, 22.49 #{149} Orbit, MR studies, 22.1214. Trauma, 22.414, 22.418, 22.49 179:789-794
MR
radiography, tomography, and computed tomography (CT) are the established modes of evaluating orbital fractures (1-15). Recently, magnetic resonance (MR) imaging with surface coils has come to play an important role in orbital imaging (16-19). However, orbital fractures have received relatively little attention in the literature and have been described in only a few cases of blow-out fractures (20, 21). In this artide, we discuss our experience with surface coil MR imaging of patients with orbital fractures and attempt to clarify the current role and limitations of this technique. PATIENTS
AND
Twenty-six
patients,
METHODS aged
9-58
years,
with orbital fractures diagnosed with plain radiography and CT underwent examination with surface coil MR imaging.
Polytomography
was performed
All of the patients had obvious trauma to the orbit. Twenty patients were male and six were female; 15 had blowout fractures (medial wail, n = 6; floor,
n
=
and
6; and medial 1 1 had
tures.
wall and floor,
maxilbofacial
n
complex
CT and MR examinations
3)
frac-
were
per-
within a 3-day period in each and the time of MR examination
case, alter trauma varied from 2 days to 3 years. Sixteen patients underwent surgery. Of these, 10 had undergone postoperative follow-up MR examinations for the evaluof motility
impairment.
CT was performed
with
a third-genera-
tion scanner DR3; Siemens
(Somatom 2 or Somatom Medical Systems, Erlangen, Germany). A 2-mm collimator was used, and sections were obtained at 2-mm intervals. Both axial and direct coronal scanning was performed in 14 patients, and axial scanning only was performed in 12
patients. From the Department of Radiology, Kanazawa Medical University, Daigaku 1-1, Uchinada, Kahoku, Ishikawa 920-02, Japan. From the 1989 RSNA scientific assembly. Received September 24, 1990; revision requested November 16; revision received January 10, 1991; accepted January 16. Address reprint requests to H. Tonami. c RSNA, 1991 I
In all of these
12 patients
in
whom direct coronal scanning was not performed, computer-reformatted images in the coronal or oblique sagittal plane parallel to the axis of the inferior rectus muscle were obtained from the axial sections.
MR examinations a 0.5-T
whole-body
were imaging
performed
with
system
(Mag-
Radiology
Tamamura, Okimura,
MD MD
Imaging’ netom;
Siemens
Medical
Systems).
The
head coil was used for transmission only, and a surface coil 10 cm in diameter was placed over the orbit to act as a receiver. Ti-weighted spin-echo (SE) images with a repetition time (TR) of 400-600 msec and an echo time (TE) of 30-40 msec were obin all cases,
tamed
and
T2-weighted
im-
ages (TRITE = 1,600-2,000/70-90) were added in 20 cases. Each section was 5 mm thick with a 0-2.5-mm intersection gap. The
matrix
four signals
size
was
were
256
X 256,
averaged.
tiple images in the ing the entire orbit
and
coronal plane were obtained.
to
two
Initially,
mul-
coverThere-
after, additional images in other planes, either the axial plane along the course of the medial rectus muscle or the oblique sagittal
plane
parallel
to the
axis
of the
in-
terior rectus muscle, were obtained. All patients were asked to keep their eyes closed during the examination and to refrain from eye movement. RESULTS
in 10 pa-
tients.
ation 1991;
Coil
LAIN
formed
Radiology
Neck
Yamamoto, MD #{149} Masao Matsuda, MD #{149} Hiroyasu Nakagawa, MD #{149} Akira Takarada, MD #{149} Tetsuro
Fractures:
Orbital
and
Blow-out
Fractures
Seventeen fractured surfaces in 14 patients were evaluated with MR imaging. One other patient who underwent examination after surgery only was not included. In all cases, the location of fracture was precisely mdicated on Ti-weighted images by the presence of orbital fat that had prolapsed into the adjacent air-containing paranasal sinuses (Figs 1-3). The displaced bone segments could be seen indirectly as a signal void between prolapsed orbital fat and intrasinus
hemorrhage
or mucosal
thick-
ening in six cases (Fig 1). MR imaging depicted enlargement of extraocular muscle in seven cases (Figs 1, 3), kmnking of muscle toward the bone defect in 10 cases (Figs 1-3), and discontinuity of muscle at the bone defect in two cases (Fig 2). Sur-
Abbreviations: SE time, TR = repetition
spin time.
echo,
TE
echo
789
gery
revealed
these
two
of muscle Orbital shape
muscle
incarceration
with
a discontinuity
cases
in
at MR imaging. fat prolapsed in a saccubar
was
present
in three
patients
at
.-lI-.--
,_4
MR imaging (Fig 2). Surgery disclosed incarceration of probapsed orbital fat at the bone defect in these three patients, all of whom had poor prognoses with restriction of ocular motility even after surgery. Intrasinus
hemorrhage
strated tients.
was
at MR
Complex
Twenty-seven patients
pa-
only
surfaces
evaluated
with
Five other examination
were
not
patients after
included.
27 fractured
b.
Fractures
fractured were
imaging. derwent the
demon-
in five
a.
Maxillofacial
six
well
imaging
surfaces
in MR
who surgery
un-
,
Sixteen
of
(nine
of the
‘
blow-out type and seven of the separated linear type) were well shown, but the 1 1 others were not detected
with MR imaging (Figs 4, 5). Associated soft-tissue abnormalities, including two cerebral contusions (Fig 5), five intrasinus hemorrhages (Fig 6), one intraorbital hematoma (Fig 6), and
one
well
orbital
shown
emphysema
with
Postoperative Studies
imaging. gery and
was
with
The
intervals
disclosed
1.
Blow-out
fracture
of medial
orbital
wall.
(a) Axial
and
(b) direct
coronal
CT scans
show relatively extensive blow-out fracture of medial orbital wall on left (arrows). Medial rectus muscle (arrowheads) is medially displaced. MR imaging was performed 25 days after trauma. (c) Ti-weighted MR image (SE 600/30) clearly shows the fracture (arrows) and the medial rectus muscle (arrowheads) because of bright signal from orbital fat. Although the medial rectus muscle is enlarged and medially displaced, the entire course of the muscle can be seen. (d) Coronal Ti-weighted MR image (SE 600/30) shows the displaced bone segment (arrows) as a signal void between the prolapsed orbital fat and the mucosal thickening of ethmoid sinus. Surgery was not performed because this patient’s clinical symptoms were not serious.
assessed
postoperative
MR
between
MR imaging In six of these
imaging
MR
impairment
10 patients
days.
were
imaging.
Follow-up
Motility in
MR
d.
C.
Figure
sur-
were 17-112 10 patients, MR
postoperative
complications: two scar tissue tions (Fig 7), three inappropriate plants (Fig 3), and one orbital
formaimabscess
(Fig 8). A second operation was performed in four patients after MR imaging, and remarkable improvement of motility was noted in all of these patients. In two other patients with intraorbitab pair was patients,
was
scar tissue, performed. diagnosis
not the
made
on the
basis
surgical reIn these two of scar tissue
of low
signal
at follow-up
MR
2.
Blow-out
imaging was performed 2 days after traushows blow-out fracture of orbital floor on right. Orbital fat is prolapsed in a saccular shape, and inferior rectus muscle (arrowheads) is herniated together with orbital fat. (b) Oblique sagittal Ti-weighted MR image (SE 600/ 40) shows a discontinuity of muscle at the bone defect (arrow). The orbit was explored 3 weeks after trauma, and incarceration of both muscle and orbital fat was confirmed. (Reprinted, with permission, from reference 21.)
ma. (a) Coronal
intensity on T2-weighted images and more than 5 months of stable appear-
ance
Figure
examinations.
fracture
Ti-weighted
of orbital
MR image
floor.
MR
(SE 600/35)
DISCUSSION Orbital
fractures
can
usually
be
vi-
suabized with plain radiography with a 28#{176} Caldwebl view and a Waters view (9). Fractures that are not seen on plain radiographs can almost a!ways
be
localized
or CT scans 790
#{149} Radiology
(10).
with
Of these
tomographic
two,
direct
oblique
sagittal
or direct
coronal
CT
scans have proved more useful (14,6-8,10-15). Direct oblique sagittal scans, however, have several limitations (14,15). The positioning of the patient is sometimes difficult, requir-
ing better technical assistance. It is also difficult to maintain the position of the patient’s head in a scanner with a small gantry aperture. As with direct coronal scans, artifacts from dental fillings can be a problem (1,4).
June
1991
nificant
:)
‘
A ‘-:
improvement
of the
signal-
to-noise ratio, thereby allowing thinner sections and higher spatial resolution, comparable with those of CT, without the expense of an increase in imaging time (16-18). Until now, however, orbital fractures have received little attention in the litera-
:t
ture. It has been believed that MR imaging is not suited to the evalua-
tion of bony lesions because of the inability to demonstrate cortical bone b.
a.
itself (17,18).
Although small bone fragments were not well visualized on MR images in our series of blow-out fractures,
the
4 ,
tion
,/ ‘
;
:
about
A
‘
rare,
b. Le Fort II and III fractures. (a) Computed radiograph shows extensive fractures of all orbital walls (arrows). MR imaging was performed 3 days after trauma. (b) Coronal Tiweighted MR image (SE 500/30) demonstrates five of eight fractures, including diastatic fractures of both frontozygomatic sutures (arrows), and blow-out fractures (arrowheads) of the left floor and roof of both orbits. MR imaging failed to demonstrate fractures of the right and
medial
wall
of both
orbits.
but
when
in which direct oblique sagittal or direct coronal scans are not obtained, computer-reformatted images in the coronal or oblique sagittal plane are sometimes useful, but there is an obvious loss of spatial resolution with computer-reformatted images
Volume
(5).
Furthermore,
the
179
3
#{149} Number
risk
of
harmful ionizing radiation to the lens is not negligible with CT (22). Clinical application of MR imaging to the orbit has several advantages, including absence of ionizing radiation, multiplanar capability, and excellent tissue contrast. In addition, the use of surface coils enables sig-
MR images, at the bone
suggestive Two patients
of musin our
it is substantiated,
We speculate
may be entrapped
that
this
find-
derived from the fact that muscle is usually ebon-
gated and its axis is perpendicular to the fractured orbital wall (4). Furthermore, small defects, which tend to be associated with muscle incarceration, may accentuate this finding (24). On the other hand, it has been sugthat
dipbopia
ocular motility the entrapment
and
limitation
of
usually result from of orbital fat through
the bone defect (3,25). Orbital fat contains numerous fibrous bands that connect the muscle sheath to the periosteum. Increased tension of these fibrous bands after trauma has been suggested as a cause of these symptoms (3,25). On MR images, or-
bital In cases
seen on of muscle
early repair is necessary to prevent tissue necrosis (14,23). One might postulate that discontinuity suggests not an incarceration but a laceration or tear of the muscle. Laceration or tear, however, seldom occurs in orbital fractures, especially in blow-out
gested
4.
between
study had this MR finding, and the presence of muscle incarceration was confirmed at surgery. Actual incarceration of the extraocular muscle is
ing the
floor
relationship
abnormalities discontinuity
fractures.
Figure
of prolapsed
and the extraocular musthe findings of muscle
defect was highly cle incarceration.
Figure 3. Inappropriate implant used for the reconstruction of blow-out fracture. (a) Axial and (b) coronal Tl-weighted MR images (SE 600/35) obtained 5 days after trauma show blow-out fracture of medial orbital wall at left (arrows). Medial rectus muscle (arrowheads) is enlarged and medially displaced. The defect was reconstructed with autogenous bone from the iliac crest 15 days after trauma. Postoperative MR examination was performed because of the exacerbation of motility impairment. (c) Coronal Ti-weighted MR image (SE 600/35) obtained 17 days after surgery shows that the autogenous bone (arrows) is inclined medially. The autogenous bone was removed during the second operation, and a silicone plate was inserted. (d) Coronal Ti-weighted MR image (600/35) obtained 70 days after the second operation shows the adequately reconstructed medial orbital wall. Note the silicone plate (arrowheads), seen as a region of hypointensity. (Reprinted, with permission, from reference 21.)
.4
the
the fracture cle. Among
d.
.‘
identification
hyperintense orbital fat enabled bocalization of the fracture site. MR imaging also provided useful informa-
fat probapsed
is clinically indicates
defect. had tion
fat
Three
patients
this MR finding, of the orbital fat
herniated
and
shape
strongly
incarceration
surgery in these been postulated tents
in a saccular
important
at the
in our and was
incarcerafound at
three patients. that the orbital through
bone
series
the
defect
Radiology
It has conat #{149} 791
the time of impact. As the force of the blow dissipated, the bony fragments tended to return toward their
original
position,
thus
the herniated three patients
orbital
incarcerating fat
(26).
In
all
who had incarceration of orbital fat, satisfactory improvement of ocular motility could not be achieved even after surgical repair. The interval between trauma and surgery in these patients was 16-21 days. As with the muscle incarceration, therefore, early surgical repair in patients orbital fat
with the probapsed
MR finding in a saccular
shape is recommended. The need for and timing cal treatment of blow-out still
controversial.
Most
of
of surgifractures
is
surgeons
now agree that a blow-out fracture not a surgical emergency and delay the decision to operate for 10-14 days,
after
which
repair
is
is performed
selectively in cases of persistent diplopia or enophthalmos (27,28). In our experience, however, one exception is an incarceration of extraocular muscle
or orbital
fat.
Early
repair
of
the defect should be performed in such cases, and MR imaging is extremely useful in the diagnosis of these soft-tissue conditions. Detection of the fracture sites in maxillofacial complex fractures at MR imaging remains limited. Fractures at which orbital fat was not prolapsed were sometimes missed at MR imaging. In our series, 1 1 of 27 fractured surfaces (41%) were not detected at MR imaging. However, even in these cases, MR imaging in assessing soft-tissue just as it was for blow-out
was useful involvement, fractures.
Cerebral contusion, epidural ma, intraorbital hematoma,
hematointra-
d.
C.
5. Fractures of orbital roof and floor. (a) Direct coronal fractures of orbital roof (arrows) and floor (arrowheads)
Figure
shows
bital canal soft-tissue
at right. (b) abnormalities.
Direct coronal CT scan (c) MR imaging was (SE 500/30) fails to demonstrate
weighted image MR image (SE 2,000/90) roof fracture.
Figure
6.
Le Fort
clearly
II and
shows
III fractures.
cerebral
CT scan
settings of infraorwith soft-tissue settings does not demonstrate performed 54 days after trauma. Coronal Tieither fracture. (d) Coronal T2-weighted contusion (arrows) associated with orbital
with
with
bone
involvement
MR
imaging was performed 27 days after trauma. Oblique sagittal Ti-weighted MR image (SE 500/40) shows hyperintense intraorbital hematoma (solid arrows) extending from subfrontal epidural hematoma (arrowheads) through the bone defect. Note hyperintense intrasinus hemorrhage (open arrows) in the maxillary sinus.
p
sinus hemorrhage, and orbital emphysema, all of which are commonly
associated with orbital fractures, were well visualized at MR imaging. In evaluating the deep portion of the orbit,
however,
gion
of interest
must
be
kept
theoretical tio
the
because
small-diameter
distance when
ings
lesions
indicate
especially
cations,
it may
of the
surface
re-
coil the
in signal-to-noise
duced as this (18). Therefore, tures,
the
in mind,
gain
from
distance
from
coils
rais re-
is increased clinical findin deep
intracranial be better
struc-
complito use
a stan-
dard head coil in place of or in addition to the surface coil (17). MR imaging also provided valuable information about the condition of the orbit in patients with postoperative motility impairment. In our series
of
tive
motility
792
10 patients
#{149} Radiology
impairment,
with
causes of motility impairment were seen at MR imaging in six cases. In our
MR
imaging
seems
short T2 and the low density of water protons in collagen (29-32). Relative hypointensity of scar tissue on T2weighted images allows it to be distinguished from soft-tissue edema,
hemorrhage, thermore, radiation
follow-up
or muscle tissue. Furlack of harmful ionizing to the
studies
eye
enables
repeated
to be performed
to
In summary, ing,
with
and
its
surface multiplanar
coil
MR
capability
imag-
excellent
olution,
soft-tissue
is
useful
extremely
contrast resin evalu-
ation of orbital fractures. In blow-out fractures, MR imaging may be performed soon after adequate plain radiography. It is our impression that CT may not be necessary
safely.
postopera-
obvious
experience,
be extremely valuable in the evaluation of scar tissue formation. It is known that on T2-weighted images mature fibrosis usually has a low signal intensity, caused by the very
for
fractures equal
the
evaluation
because to CT
in the
of blow-out
MR
imaging
detection
is of the
June
1991
Figure
7. Scar autogenous coronal CT
with rect
tissue formation after surgery for blow-out fracture. In this case, bone 8 days after trauma. CT and MR imaging were performed scan shows soft tissue with abnormally high attenuation (arrows)
sagittal
Ti-weighted
MR image
(SE 500/40),
muscle
is obliterated
(arrowheads).
(c) On
area
(arrows).
The
diagnosis
soft-tissue
structure
coronal T2-weighted was made on the
of scar tissue
is present MR
basis
image
the defect in the orbital floor at right was 51 days after surgery because of persistent between the globe and the orbital floor.
part of the orbital
in the anterior (SE, 2,000/90),
of persistent
the
low signal
lesion
intensity
reconstructed diplopia. (a) (b) On oblique
floor (arrows).
is demonstrated
on T2-weighted
The inferior
as a very images for
Di-
rectus
hypointense 5 months.
8. Orbital abscess after the reconstruction of Le Fort II and III fractures. In this case, the reconstruction of the orbital floor at right was performed 52 days after trauma, and a silicone sheet was inserted. Follow-up CT and MR imaging were performed 87 days after surgery because the patient complained of external strabismus and diplopia. (a) Direct coronal CT scan shows a well-circumscribed extraconal mass (arrows) displacing the inferior rectus muscle (arrowheads) superomedially. Note a large defect in the orbital floor. (b) On oblique sagittal Ti-weighted MR image (SE 500/40), the mass (arrows) is slightly hyperintense relative to muscle and hypointense relative to fat. The globe and the optic nerve are displaced upward. In b and c, arrowheads point to hypointense rim. (c) On coronal T2-weighted MR image (SE 2,000/90), the mass (arrows) is extremely hyperintense to fat. Surgical findings confirmed the mature abscess with thick capsule. Figure
fracture site the assessment malities.
On plays
the
the
and
other
is superior of soft-tissue
hand,
in
MR imaging
a complementary
evaluation
to CT abnor-
role
to CT
of maxilbofacial
in
corn-
plex fractures. CT is superior to MR imaging in the detection of fracture sites in maxiblofacial fractures. MR imaging, however, provides valuable information about soft-tissue lesions. Situations in which MR imaging is preferable in detection of maxilbofacial fractures include unexplained neurobogic deficits, visual or extraocular muscle impairment, and fractures with a high probability of intracranial extension.
preferable
in
cases of postoperative pairment. MR imaging
MR
imaging
motility provides
imuse-
ful
postoperative
information
Volume
179
is also
about #{149} Number
3
complications vided by
that
may
not
be
pro-
CT.
In regard to our suggested aging procedure, multiple
MR coronal
images should
orbit (21).
covering be obtained
the
allows visualization cross sections of all and of all extraocular
after,
entire initially
of tangential four orbital muscles.
we recommend
obtaining
im-
This
walls There-
addi-
tional images in another plane, either the axial plane along the course of the medial rectus muscle or the oblique sagittal plane parallel to the axis of the inferior rectus muscle.
These
additional
images
provide
more precise information about relationship between the muscle the bone defect. While Ti-weighted images are useful for localizing
fracture site, both Tied images are usually
the and a
and T2-weightnecessary for
the evaluation of cerebral contusion, hemorrhagic lesions, and scar tissue formation (21). Although a strict comparison between surface coil MR imaging and direct multipbanar high-resolution
CT should be made believe that surface
in the future, we coil MR imaging
is an important adjunct in the nosis and treatment of orbital tures. U
diagfrac-
Acknowledgments: We thank Kazuyuki Sasaki, MD, Department of Ophthalmology; Sadao Tsukada, MD, Department of Plastic Surgery; and Kohichi Yamashita, MD, Department of Otolaryngology, Kanazawa Medical University, for their valuable contributions. We also thank Akiko Ohta for secretarial assistance; Masao Yonezawa, RT, Tomokazu Oku, RT, and Osamu
and
Yamashita,
James
C. Ehrhardt,
RI,
for
technical
PhD,
assistance;
for reviewing
the
manuscript.
Radiology
#{149} 793
12.
References 1.
2.
3.
Grove AS Jr, Tadmor R, New PFJ, et al. Orbital fracture evaluation by coronal computed tomography. Am J Ophthalmol 1978; 85:679-685. Zilkha A. Computed tomography of blow-out fracture of the medial orbital wall. AJR 1981; 137:963-965. Hammerschlag SB, Hughes 5, O’Reilly GV, Naheedy MH, Rumbaugh CL. Blowout fractures of the orbit: a comparison of computed tomography and conventional radiography with anatomical correlation. Radiology
4. 5.
6.
7.
8.
9.
10.
ii.
794
Zilkha
1982;
13.
14.
15.
A.
#{149} Radiology
SM,
Orbital
blowout
Lagouros PA, et al. the prognostic significance of computed tomography. Ophthalmology 1985; 92:1523-1528. Gentry LR, Smoker WRK. Computed tomography of facial trauma. Semin Ultrasound CT MR 1985; 6:129-144. Koornneef L, Zonneveld FW. The role of direct multiplanar high resolution CT in the assessment and management of orbital trauma.
i43:487-492.
Computed tomography in facial trauma. Radiology 1982; i44:545-548. Brant-Zawadzki MH, Minagi H, Federle MP, et al High resolution CT with image reformation in maxillofacial pathology. AJR 1982; 138:477-483. Cooper PW, Kassel EE, Gruss JS. High resolution CT scanning of facial trauma. AJNR i983; 4:495-498. Gentry LR, Manor WF, Turski PA, et al. High resolution CT analysis of facial struts in trauma. I. Normal anatomy. AJR 1983; 140:523-532. Gentry LR, Manor WF, Turski PA, et al. High resolution CT analysis of facial struts in trauma. II. Osseous and soft tissue complications. AJR 1983; 140:533-541. Dolan KD, Jacoby CC, Smoker W. The radiology of facial fractures. RadioGraphics 1984; 4:576-663. Kreipke DL, Moss JJ, Franco JM, Maves MD, Smith DJ. Computed tomography and thin-section tomography in facial trauma. AJNR 1984; 5:185-189. Johnson DH Jr. CT of maxillofacial trauma. Radiol Clin North Am 1984; 22:131144.
Gilbard
16.
Radiol
Mafee
MF,
24.
fractures:
Clin
North
Am
1987;
25.
26.
25:
753-766. Ball JB Jr. Direct oblique sagittal CT of orbital wall fractures. AJNR 1987; 8:147154. Sullivan JA, Harms SE. Surface-coil MR imaging of orbital neoplasms. AJNR 1986;
27.
28.
7:29-34. 17.
Atlas
SW,
LT, Zimmerman RA, HI, Grossman RI. Orbit: initial experience with surface coil spin-echo MR imaging at 1.5 T. Radiology 1987; 164:501-509. Bilaniuk LI, Atlas SW, Zimmerman RA. Magnetic resonance imaging of the orbit. Radiol Clin North Am 1987; 25:509-528. Hosten N, Sander B, Cordes M, Schubert CJ, Sch#{244}rnerW, Felix R. Graves ophthalmopathy: MR imaging of the orbits. Radiology 1989; i72;759-762. McArdle CB, Amparo EG, Mirfakhraee M. MR imaging of orbital blow-out fractures. J Comput Assist Tomogr 1986; 10:116-119. Tonami H, Nakagawa I, Ohguchi M, et al. Surface coil MR imaging of orbital blowout fractures: a comparison with reformatted CT. AJNR 1987; 8:445-449. Lund E, Halaburt M. Irradiation dose to the lens of the eye during CT of the head. Neuroradiology 1982; 22:181-184. Koornneef L. Current concepts on the management of orbital blow-out fractures. Ann Plast Surg 1982; 9:185-200.
Hackney
18.
19.
20.
21.
22.
23.
Bilaniuk
DB, Goldberg
29.
30.
31.
32.
Smith B, Petrelli R. Fractures of the orbit: blow-out and nasoorbital fractures. In: Freeman HM, ed. Ocular trauma. New York: Appleton-Century-Crofts, 1979; 117-123. Hammerschlag SB, Hughes S, O’Reilly GV, Weber AL. Another look at blow-out fractures of the orbit. AJNR 1982; 3:331335. Greenwald HS Jr. Keeney AH, Shannon GM. A review of 128 patients with orbital fractures. Am J Ophthalmol 1974; 78:655-664. Converse
JM,
Smith
B.
On
the
treatment
of blow-out fractures of the orbit. Plast Reconstr Surg i978; 62:100-104. Crikelair GF, Rein JM, Potter GD, et al. A critical look at the “blow-out” fracture. Plast Reconstr Surg 1972; 49:374-379. Glazer HS, Lee JKT, Levitt RG, et al. Radiation fibrosis: differentiation from recurrent tumor by MR imaging. Radiology 1985; 156:721-726. Sundaram M, McGuire MH, Schajowicz F. Soft tissue masses: histologic basis for decreased signal (short 12) on T2-weighted MR images. AJR 1987; 148:1247-1250. Ebner F, Kressel HY, Mintz MC, et al. Tumor recurrence versus fibrosis in the female pelvis: differentiation with MR imaging at 1.5 T. Radiology 1988; 166:333340. Negendank WG, Al-Katib AM, Karanes C, Smith MR. Lymphomas; MR imaging contrast characteristics with clinicalpathologic correlations. Radiology 1990; 177:209-216.
June
1991
