Clinical Radiology 70 (2015) 25e36

Contents lists available at ScienceDirect

Clinical Radiology
journal homepage: www.clinicalradiologyonline.net

Maxillofacial fibro-osseous lesions

D.S. MacDonald*
Division of Oral & Maxillofacial Radiology, Faculty of Dentistry, UBC, 2199 Wesbrook Mall, Vancouver V6T 1Z3, BC,
Canada

art icl e i nformat ion

During the last decade much has changed in our understanding of fibro-osseous lesions (FOLs)
Article history: of the jaws with regards to their imaging, their nomenclature and classification, and their
Received 5 June 2014 potential impact on the overall health of the patient. The changes in nomenclature,
Accepted 20 June 2014 classification, and the FOLs’ potential association with important syndromes are discussed with
the assistance of a flowchart. The lesions, fibrous dysplasia (FD), ossifying fibroma (OF), and
osseous dysplasia (OD), though with similar histopathology, have very different clinical and
radiological presentations, behaviour, and treatment outcomes. FD of the jaw, which though
becoming inactive, does not involute and therefore requires life-long review to monitor for
occasional reactivation. OF is completely removed surgically as it has a propensity to recur. No
treatment is generally required for an OD unless it is infected or displays expansion. Lesions
outside the jaws associated with FOLs of the jaws are identified. Radiology remains central to
the diagnosis of FOLs of the jaw, because the pathologist is still not able to distinguish between
them. The advent of cone-beam computed tomography (CBCT) with its better radiation dose
reduction, accessibility, and spatial resolution has transformed imaging of FOLs. Their pre-
sentations on CBCT and the clinical indications for its use are discussed.
Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Introduction lesions are now understood, imaged, and treated. The

In the decade since “Fibro-osseous lesions of the face
and jaws”1 (FOLs) was published this important group of
lesions has experienced many significant changes, which
the medical radiologist would want to know, not only to
better diagnose them, but also to understand the man-
agement implications of their diagnoses. In addition to
changes in nomenclature and classification (Fig 1),
detailed case series and systematic reviews (SRs) and
technological developments have all affected how these
* Guarantor and correspondent: D. MacDonald, Division of Oral & Maxil-
lofacial Radiology, Faculty of Dentistry, UBC, 2199 Wesbrook Mall, Vancou-
ver V6T 1Z3, BC, Canada. Tel.: þ1 604 822 9762; fax: þ1 604 822 3562.
E-mail address: dmacdon@dentistry.ubc.ca
2005 edition of the WHO classification of odontogenic
neoplasms2,3 adopted Brannon and Fowler’s histopatho-
logical redefinition of the “cementoid” elements found in
cemento-osseous dysplasia and cement-ossifying fi-
bromas.4 They reported that these cementoid elements
are truly manifestations of abnormal bone and not a form
of cementum. As a result cement-ossifying fibroma and
cement-osseous dysplasia were renamed ossifying fi-
broma (OF; Figs 2 and 3) and osseous dysplasia (OD;
Figs 4e7). OF is the jaw’s equivalent of the extragnathic
“osteofibroma”, which was also called “ossifying fibroma”
until renamed in the second edition of the WHO classifi-
cations of neoplasms of the bone.5 Although the name of
the third FOL, “fibrous dysplasia of the jaws” (FD),
remained unchanged (Figs 8e10) we now understand its
aetiology, presentation, and behaviour better.

http://dx.doi.org/10.1016/j.crad.2014.06.022
0009-9260/Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
26 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36

Figure 1 Changes in the classification, nomenclature and association with syndromes of FOLs affecting the face and jaws. This flowchart
continues on from MacDonald-Jankowski’s Fig 1.1

Figure 2 This medical computed tomograph is of a juvenile OF affecting the maxilla of a child. Its margins are well-defined. The coronal and axial
reconstructions exhibit a complete obturation of the maxillary sinus and a ball-shaped expansion of its bony walls, including a substantial buccal
expansion. The subjacent teeth are displaced downwards and the floor of the orbit is displaced upwards. It has obturated the ipsilateral nasal
cavity and expanded into the contralateral nasal cavity.

There are significant differences between ethnic groups
for each FOL. The world literature of FOLs, set out in recent
SRs,6e8 was grouped into four global groups, Western
[predominantly Caucasian (European, Middle Eastern and
Indian)], East Asian, sub-Saharan African, and Latin-
American. Further details regarding their composition has
been set out in the SRs,6e8 whereas their global extent and
the tabulation of their clinical and radiological features at
first presentation have been set out in a recent textbook.9
These global groups are based on Cavalli-Sforza et al.,10
supported more recently by Tang et al.,11 which confirmed
that global grouping based on ethnicity is appropriate. The
advantage of this global grouping is that the clinician can
determine those features of a lesion which best reflect the
community served.
During the last decade, the advent of cone-beam
computed tomography (CBCT) has transformed the
imaging of FOLs affecting the jaws. So far there does not
appear to be any published reports on CBCT of FOLs. A
reason for this hitherto dearth is that those FOLs, be they
incidental or symptomatic, if large in size were most
likely to be investigated by conventional CT in a medical
hospital (Figs 2 and 10). Smaller FOLs are more likely to be
investigated and managed in general and specialist dental
practices. Generally, this has been achieved by conven-
tional dental radiography. Now with the growing avail-
ability of CBCT units, with a higher spatial resolution, a
smaller field of view (FOV), and a lower radiation dose,
FOLs can now being optimally investigated by CBCT
(Figs 3e9). There are three main advantages CBCT has
over CT. The spatial resolution of CBCT, which can be as
low as 0.078 mm voxel size, is generally better to that of
CTs. CBCT imparts a much lower radiation dose. CBCT is
more readily available and can be sited in almost any
 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
Figure 3 The periapical radiograph (a) displays a well-defined “ground-glass” radiopacity displacing the adjacent teeth. It is separated from the
normal adjacent bone by a thin radiolucent space (arrows). The CBCT (b and c) exhibits a rounded buccolingual expansion. Both the buccal and
lingual cortices are also eroded (thinned). This is an early OF.
clinical space, due to its lower weight, smaller footprint,
and no need for a high-tension electricity supply. The
main disadvantage of CBCT are poorer contrast resolution
(15 bit-depth or less; therefore, it essentially displays only
a “bone window”).12
Leading authorities on FOL pathology13,14 (see reference
1 for their quotations in full) stated the histopathology can
only determine that the lesion is a FOL. To refine this further
requires clinical and radiographic evaluation. This was the
only safe path that could be taken as the various separate
FOLs differ so much in their nature and behaviour. Con-
ventional radiography is still sufficient for the diagnosis of
almost all cases of FOLs. FD can be distinguished from OF
based on their margins, which are poorly defined for FD and
well defined for OF. Although this radiological feature re-
mains the reference standard for distinguishing between FD
and OF, over the last decade, a genetic test has been able to
distinguish between them with increasing confidence. This
test is based on guanine nucleotide-binding protein (GNAS)
mutations. Jundt et al.15 stated that in FD “Mutations in the
gene (GNAS I) encoding for the G-protein (Gs-a) lead to
increased c-AMP production affecting proliferation and
differentiation of preosteoblasts”. Shi et al.16 reported that
although GNAS mutations were found in most cases of FD,
they were not found in any case of OF. Patel and co-
workers17 reported GNAS not only absent in cases of OF, but
also OD. Therefore, a positive GNAS response is strong in-
dicator of FD and may be useful in the few cases in which
27
the clinical and radiological presentations are too unclear to
allow distinction between FD and OF.
Fibrous dysplasia
FD is a benign, non-inheritable disease and affects about
1:30,000 individuals.18 It presents either as a local lesion,
the monostotic form, or as a systemic lesion, the polyostotic
form. When this last form is combined with hormonal
changes, it is now McCuneeAlbright’s Syndrome (MAS), of
which precocious puberty and cafe -au-lait cutaneous spots
are the most striking features. MAS is also now associated
with other endocrinopathies19 and other lesions such as
ovarian follicular cysts.20 Although Mazabraud syndrome is
associated with FD with or without MAS,21 so far no intra-
muscular myxomas typical of this syndrome have been re-
ported in the face, or jaws.
If it is supposed that all FDs arise during childhood or
puberty, most monostotic cases remain undetected, only to
become active or be reactivated later and thus be detected
for the first time later in life. This conclusion is feasible if the
classical division of FD into monostotic, polyostotic, and
MAS forms is considered to reflect the timing of the mu-
tation and thereby, the initial size of the mass of FD pre-
cursor cells.18,22 Therefore, the earlier the mutation, the
more widespread and the more severe will be the
manifestations.18
28 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36

Figure 4 CBCT of FOD of the left side of 60-year-old Chinese woman. The CBCT panoramic reconstruction of the left side displays multiple well-
defined radio-opacities within the alveolus of the mandible. Note they are all positioned above the mandibular canal (arrows). The axial and
transaxial reconstructions exhibit these lesions within the alveolus without buccolingual expansion. The lesions are separated from their
adjacent normal bone by a radiolucent space and a cortex (in places).

The polyostotic form is the easiest form to diagnosis,
because many bones are affected, and it occurs in childhood
and is generally obvious to the patient’s doctor and/or
parents. Nevertheless, Volkl and Do € rr recently suggested
that non-endocrine afflictions that may accompany poly-
ostotic FD include hepatobiliary dysfunction and cardiac
disease.23 These may be important risk factors for early
death.23
A recent SR of reports of consecutive case series of FD
affecting the jaws reveals that over 93% were of monostotic
form,6 which is higher than the 60% monostotic overall.18 A
third of cases of monostotic FD affect craniofacial bones.24
Monostotic means “one bone”, which is the correct term
when applied to FD affecting the mandible, but is not
strictly true when applied to FD affecting the maxilla. FD
affecting the maxilla may involve one or more contiguous
bones.
The majority of cases of FD affecting the jaws first pre-
sented with a mean of 25 years,6 in contrast to a mean age of
15 years at first presentation overall for all cases of FD.18
Although the distribution between the sexes is almost
equal with females prevailing slightly, males predominated
2-to-1 in the second decade in which 36% percent of all FDs
of the jaws first presented. Males also slightly predomi-
nated in the third decade, but are in the minority in all other
decades.6
Most cases of FD affecting the jaws do not become
“quiescent during puberty” typical of extragnathic cases of
FD.25 Instead, almost all cases first present to the clinician
with symptoms. Only 2% of monostotic FD were discovered
incidently.6 This may be due to the fact that FDs on con-
ventional radiographs have poorly-defined margins and
therefore do not readily draw the dentist’s attention to their
presence unless they are so large, affecting all or most of the
hemimandible or hemimaxilla.26 Although FDs may not be
readily obvious as incidental findings, many patients may
be aware of their lesions on average 5.2 years prior to first
presenting with them.9
Ninety percent of cases of FD affecting the jaws first
present with swelling,6 in contrast to pain in all cases of FD
overall.18 Swelling was significantly more frequent in the
Western than in the East Asian global group.6 Overall, 18%
presented with pain.6 A higher proportion of East Asian
cases present with pain in comparison to other global
communities. In the jaws the maxilla is affected in 58% of FD
cases. FD displays an overwhelming predilection for the
posterior sextants of both jaws.9
FD affecting the jaws differs radiologically and histolog-
ically from that of the rest of the skeleton on conventional
radiography. FD affecting the jaws is poorly-defined, ac-
cording to the Slootweg and Muller’s 1 mm “zone-of-tran-
sition” criterion,27 whereas that of the extragnathic FD is
 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36 29

Figure 5 CBCT of FOD presenting with bilateral lesions in the posterior sextants and one in the anterior sextant of the mandible. The axial
section displays a radiolucency bilaterally in the premolaremolar areas. Although extensive in their mesiodistal extent, there is minimal buc-
colingual expansion. Nevertheless, the cortices are substantially eroded but remain intact. The coronal section reveals radio-opacities at the
apices of the molar teeth. The sagittal section reveals a well-defined round radiopacity at the apex of a central incisor. It is mainly ground glass.

Figure 6 CBCT of an OD arising apically to the second premolar and the mesial root of the first molar. The lesion is a central radiopacity separated
from normal adjacent bone by a radiolucent periphery. There is no buccolingual expansion. The buccal and lingual cortices are nevertheless
eroded (thinned). Although there is no evidence of root resorption, the mesial root of the first molar tooth displays hypercementosis; compare
this with the cementoblastoma of MacDonald-Jankowski’s fig. 15. This is FocOD as it is the sole such lesion in this patient’s jaws. The periodontal
pocket (bone loss), radiolucency associated with the hypercementosis and the hypercementosis itself are not related to the FocOD. Although it is
not uncommon to observe root-treated teeth associated with ODs, especially FocODs, in absence of evidence of pulp-vitality testing at the time,
it is frequently surmised that such treatment was provoked by the FocOD’s early radiolucent stage -mistaken for a periapical-radiolucency-of-
inflammatory-origin (arising from a non-vital tooth).

well-defined.5,25 A possible reason for this difference is that
the jaws are derived from membrane and long bones are
from cartilage.28 Support for this contention comes from
the WHO’s second edition of the classification of neoplasms
affecting the extragnathic skeleton, which included
cartilage as an expected histopathological element of FD
affecting the skeleton outside the jaws.5 Conversely, the
WHO’s 2005 edition of the classification of the odontogenic
neoplasms15 did not include cartilage as an expected
feature of FD affecting the jaws.
30 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36

Figure 7 CBCT of multiple periapical lesions affecting the lower anterior sextant of a female. The lesions are generally central radio-opacities
within radiolucencies. They have a small diameter, which is not much larger than the width of the teeth they affect. There is little buccolin-
gual expansion. These are OD lesions, have been traditionally called periapical cemental dysplasia, although they arise within the alveolar bone
and not the cementum. Although these teeth are vital, each of the central incisors display irregular pulp outlines in at least one of the coronal
sections. These suggest root resorption and should be evaluated further by a dentist.

Figure 8 CBCT of a case of FD affecting the ramus of the mandible. The margin is poorly defined. There is a little buccolingual expansion and the
dysplastic bone has a ground-glass appearance.

The better spatial resolution of conventional dental conventional radiography in contrast to 100% on CT.29 This
radiography permits more patterns of dysplastic bone to be was confirmed in another consecutive case series of FD
seen and not only the “ground glass” pattern.6 The “ground- investigated by CT.30 Nevertheless, CT can more readily
glass” pattern was only apparent in 38% of FD cases on display the full extent of the lesion, particularly within the
 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
Figure 9 CBCT of FD affecting the right maxilla in a 60-year-old
woman. There were no presenting complaints. (a) The panoramic
reconstruction displays an ill-defined ground-glass lesion in the
posterior sextant enveloping the apices of subjacent teeth. It has
displaced the floor of the maxillary sinus upwards. (b) The coronal
section is from the above case and is through the nasopalatine duct
(arrow). Although the spatial resolution is poorer in CBCT in contrast
to conventional radiography the margin between the dysplastic bone
and the normal adjacent bone is still not easy to define.
more anatomically complex maxilla (Figs 9 and 10).29 The
maxillary sinus is completely or partially obturated in
nearly every case of FD (Figs 9 and 10). All maxilla-only
cases extended back to the pterygoid process, but did not
displace it.29 CBCT can now display these features with a
similar if not better spatial resolution and certainly with a
reduced radiation dose.
All cases displayed expansion, which was fusiform in the
mandible (Fig 8) and an enlargement of the normal contour
in the maxilla (MacDonald-Jankowski’s Fig 81).6,26,29 All
cases reported buccolingual expansion and displacement
and/or thinning of the lower border of the mandible.6,26,29
Teeth were displaced in 35% of cases.6 This tooth
displacement was observed significantly more in Western
than in East Asian reports.6 It contributes to a malocclusion,
which may require orthodontics. Although osseo-
integrated implants into the dysplastic bone have so far
survived 5 years, their longer-term survival is uncertain.31
Although, root resorption is rarely observed,6 the loss of
the lamina dura within the dysplastic lesion can be used
confirm a radiological diagnosis of FD.32
Although, the risk of pathological fracture is highest in
polyostotic cases, especially in MAS, de Mattos et al.33 re-
ported that even 50% of monostotic cases fracture.33
Nevertheless, this risk does not appear to extend to FD
affecting the jaw as only one case of apparent pathological
31
fracture and non-union has been reported for FD affecting
the mandible (MacDonald-Jankowski’s Fig 4.).34 Therefore,
although biopsy may result in a fracture risk in FD affecting
the extragnathic skeleton,18 this may not be so for FD in the
jaws.
If the lesion is monostotic and confined to either jaw, then
no treatment is required except to address disfiguring
deformity. Surgery is generally indicated if there is a threat
to vision, which occurs particularly when the FD reduces the
diameter of the optic canal.31 Although blindness is a real
risk for FD primarily involving the skull base (optic canal),
there appears to be no report of blindness caused directly by
monostotic FD arising from the jaws. Although CT is
invaluable for measurement of the optic canal’s diameter in
cases of craniofacial FD, MRI may be more appropriate for
children.18 The fact that not one case of proptosis was re-
ported within a total of over 336 maxillary cases of FD in a
SR6 indicates that this is not a frequent finding of cases of FD
arising primarily in the jaws. This is particularly surprising
considering the substantial increase in the vertical dimen-
sion of the maxilla observed in one CT series.29 Nevertheless,
at least one case of proptosis has occurred in a case of
monostotic FD affecting the maxilla alone.29
There have been very few case series that have been
followed up for a long period. The longest and largest case
series of FD of the jaws is that of the Hong Kong Chinese, in
which 17 patients were followed-up for a mean of 9 years.26
Follow-up is important because not only does FD generally
not involute to lamellar bone, perhaps except in one case,35
but it may undergo reactivation in adult life by a precipi-
tating event such as pregnancy. A woman was blinded in
one eye by the reactivation of FD during her pregnancy.36
This reactivation in the pregnant female may be due to
the elevation in sex hormone receptors in the mutated
cells.18 Reactivation is not just confirmed to females with
FD; males have also occasionally exhibited reactivation.26
Due to previous subjective reports of reactivation of FD
by jaw surgery, some dentists are uncomfortable about
treating FD patients.37 Similarly, they may express concern
that bisphosphonate treatment, albeit low dose, for FD may
provoke bisphosphonate osteonecrosis of the jaws
(BRONJ).37
Despite the paucity in follow-up, 18% of cases in an SR
exhibited reactivation.6 This happens significantly more in
the sub-Saharan African global group than in the Western
and the East Asian global groups.6 In addition to reac-
tivation, follow-up is also recommended for other lesions
arising secondarily within the FD lesion. The two most
important are aneurysmal bone cysts (ABC) and
osteosarcomas.
The association of the ABC with FD of the long bones and
spine is well-known. Although ABCs have occasionally been
found in FD of the base of the skull,38 no detailed report of
ABC arising in FD of the jaws has so far been recorded.
The likelihood of FD to undergo sarcomatous change was
not reflected in any case series of monostotic FD of the jaws
until recently.6 Cheng et al.39 reported three cases of
sarcomatous transformation in the 324 cases of craniofacial
FD that presented to them over 17 years.39 All three cases
32 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36

Figure 10 These coronal computed tomographs (bone window) display FD affecting the mandible, the maxilla, the zygoma and the sphenoid. (a)
FD affecting the hemimaxilla and concomitant zygoma. The maxillary antrum has been completely obturated. The predominant pattern of
radiodensity is “ground glass”. Although the affected hemimaxilla (and also the zygoma) displays substantial buccal expansion and appreciable
medially-directed expansion of the lateral wall of the nasal cavity, upward expansion of the floor of the orbit is minimal. The affected alveolus
displays buccolingual expansion. The lack of a cortex on the buccal aspect of the alveolus should provoke consideration that this may have been
biopsied (arrow). (b) Medially-directed expansion of the lateral wall of the orbital cavity. The ethmoid air-cells are being obturated by the FD
affecting the sphenoid. The buccal cortex is intact. (c) FD affecting the ipsilateral hemi-mandible and the greater wing and body of the sphenoid.
The ethmoid air-cells have been completely obturated. (d) The ethmoid air-cells have been completely obturated. The superior and inferior
orbital fissures, although narrower than the contralateral unaffected side are nevertheless still patent. The affected hemimandible exhibits
substantial buccolingual expansion.

were polyostotic. Cheng et al.39 suggested that sarcomatous
transformation occurs spontaneously with MAS more
frequently than for the monostotic form. A periosteal re-
action was only apparent in one of eight cases of sarcoma-
tous transformation in FD investigated by CT, but
mineralization of the tumour matrix was observed in six.40
Furthermore, although soft-tissue extension was observed
in six cases, contrast enhancement was observed only in
one.40 Although the lamina dura of the periodontal liga-
ment is replaced by dysplastic bone, the space is not only
still patent, but subjectively it appears to be narrower than
normal,32 therefore any widening should be viewed as
suspicious. Although, “the majority of high-grade central
osteosarcomas do not pose diagnostic difficulties, especially
when histological features are correlated with radiological
and clinical findings”,41 this is not true for many low-grade
osteosarcomas.42 In such cases, the markers for osteosar-
coma, cyclin-dependent kinase 4 (CDK4) and murine
double-minute type 2 (MDM2), may assist.41,42
Ossifying Fibroma
Slootweg and El Mofty2 defined the OFs as “a well-
demarcated lesion composed of fibrocellular tissue and
 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
mineralised material of varying appearances.” The majority
present in adulthood in the teeth-bearing jaws, whereas a
few present in younger individuals as juvenile trabecular
ossifying fibroma (JTOF) and juvenile psammamatoid ossi-
fying fibroma (JPOF).”2 JTOF arises in the maxilla of children
of 8e12 years old (Fig 2), whereas JPOF arises in the bony
walls of the paranasal sinuses (Slootweg’s Fig 6.72) of older
teenagers or young adults.2 In addition to these features,
conventional OF can be further distinguished from JTOF and
JPOF. Conventional OF has an observable capsule radiolog-
ically and histologically (Fig 3), whereas JTOF and JPOF do
not (Fig 2).43,44 The appearance of this capsule radiologi-
cally clearly depends upon the spatial resolution of the
imaging technology used. Fig 3a displays a definite, but thin
capsule (appearing as a radiolucent line) on a conventional
intraoral sensor, which is not visible on the CBCT images
(Fig 3b and c).
Brannon and Fowler,4 determined that all OFs need to be
enucleated completely to prevent recurrence.4 Neverthe-
less, OFs do recur after careful surgery; 12% recurred overall
with no difference between global groups.7 There is some
circumstantial evidence that the onset of menopause may
initiate, re-activate, or accelerate growth. Those cases first
presenting during menopause were significantly larger than
those first presenting between attainment of the peak bone
mass and menopause.45
A hitherto under-considered aspect of OF is a possible
association with familial hyperparathyroidism, hyper-
parathyroidismejaw tumour syndrome (HPT-JT). This is
autosomal dominant.46 A secreting carcinoma is the cause
in 10e15%.46 The loss of staining for parafibromin, caused
by its inactivation by the HRPT2 gene,47 can be used to
predict an increased risk of death or recurrence.48
Furthermore, its course is more aggressive causing more
severe hypercalcaemia, which may actually precipitate a
hypercalcaemic crisis.46
OF has a predilection for females (71%) which is similar
for all global groups. Their mean age at first presentation is
31 years of age. The East Asians have significantly older age
(35 years) at first presentation than sub-Saharan Africans
(19 years).7 Sub-Saharan Africans present in the second
decade, whereas the other three global groups present in
the fourth decade.7 Males nearly account for a half of cases
in the second decade, but only for 15e20% in the subse-
quent three decades. The mean period of the patient’s prior
awareness for OF is 1.7 years.7
Thirty-one percent are detected as incidental findings,
whereas 66% first present with swelling and 16% with pain.
Swelling present significantly more frequently in East Asian
than in the Western global group and vice versa for those
discovered as incidental findings. The mandible is affected
in 75% of cases, except for sub-Saharan Africans among
whom both jaws are affected equally. This was particularly
significant in comparison to the 5:1 mandible: maxilla ratio
in East Asians. Mandibular cases of OF are equally distrib-
uted between the anterior and posterior sextants only in
the Western global group. The other global groups display a
predilection for the posterior sextant, particularly in East
Asians.7
33
Fifty-eight percent present with the classical presenta-
tion of a radiopacity-within-a-radiolucency. Twenty-six
percent are radiolucent and 16% are completely opaque.
Sub-Saharan Africans display significantly more radiolucent
lesions,7 which is likely to reflect their younger age at first
presentation. All radiolucent OFs in an East Asian report
presented in adolescents and young adults.45 The OFs in
that report are equally divided between those that have a
round or oval shape.45 The oval-shaped OFs are significantly
larger. Half of them present in females of 45 years of age or
over.45
Eighty-four percent present with buccolingual expan-
sion.7 The lower border of the mandible presents with
erosion and/or displacement (MacDonald-Jankowski’s
Fig 41) in a half of mandibular cases.7 The maxillary sinus is
involved by 90% of cases of OF arising subjacent to it.7 Teeth
are displaced in 27% of cases (Figs 2 and 3) and roots are
resorbed in 20% of cases.7
The original review1 cited sources that revealed the MRI
features of FD and OF. Since then there appear to be no
further publications on this topic. This may reflect not only
on the adequacy of these sources, but also that conventional
radiography augmented by CT is adequate in distinguishing
FD of the jaws from OF in almost every case.
Osseous Dysplasia
As can be seen by reference to MacDonald-Jankowski’s
Fig 11, of all lesions affecting the face and jaws, this group of
lesions, the ODs have perhaps seen most changes in their
nomenclature and classification since the WHO’s first edi-
tion.49 This change is still taking place. This FOL has now
been divided into two broad categories, florid osseous
dysplasia (FOD) and focal osseous dysplasia (FocOD).
Although they are not neoplasms, they have exceptionally
been included in the recent 2005 edition of the WHO
classification of odontogenic neoplasms,3 which excluded
almost all the other non-neoplasms, with the exception of
FD and a few others. Although they are generally not
considered to be odontogenic lesions, almost all appear
above the mandibular canal or below the junction of the
hard palate and thus are confined to the alveolar process.
This suggests, at least, some odontogenic influence upon
their genesis.
Florid Osseous Dysplasia
The SR-included reports of FODs50 are associated with
middle-to-old-aged women of sub-Saharan African and of
East Asian origins. The mean number of cases is highest for
the Western global group but surprisingly least for the sub-
Saharan African global group. This may reflect the greater
ethnic diversity within the former, particularly the United
States of America where those of non-European origin ac-
count for nearly a third of the population, and the lower life
expectancy in the latter. In the sub-Saharan African global
group, potential OD (FOD and FocOD) victims are more
likely to die earlier in life of other causes before they could
acquire the OD lesions.8
34 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
FOD overwhelming affects females (97%). The mean age
at first presentation is 49 years of age. The age range is
21e83 years of age. Half are discovered as incidental find-
ings; 48% first present with pain, 31% with swelling, and
30% with a discharge or a fistula.50 Comparison of two case
series derived from the same community showed that the
series of cases observed as incidental findings on radio-
graphs were significantly older than the series of cases that
presented with symptoms.50 More than one sextant needs
to be affected to fulfil a diagnosis of FOD. The presentation
on a panoramic radiograph or on a full-mouth survey (using
intraoral film or digital detectors) is usually bilateral. The
mandible is affected in 100% of cases and the maxilla in
67%.50
The radiological presentation of the lesions ranges from
radiolucencies (Fig 5) to one with one or more small central
radio-opacities (MacDonald-Jankowski’s Fig 101) onto sub-
stantial radio-opacities with a radiolucent periphery (Fig 4)
to complete radio-opacities that abut directly onto adjacent
normal bone (MacDonald-Jankowski’s Fig 111)
Diagnosis is readily achieved by conventional radiog-
raphy. Although there is little need for CT in the uncom-
plicated case, CT (both helical and cone-beam) has revealed
some more detail not appreciated by conventional radiog-
raphy (Fig 4), such as the central positioning of the osseous
dysplastic tissue within the lesion. The CT number of OD is
772e1587 HU, which is equivalent to cementum or cortical
bone.51
Focal Osseous Dysplasia
The mean number of cases of FocOD per year globally
was 4.9, which contrasts with 1.2 for FOD and 1.6 for FD and
1.7 for OF.9 Therefore, FocOD is the most frequent of all FOLs.
As already observed for FOD, the frequency was highest for
the Western global group and least for the sub-Saharan
African global group. The explanation is the same as that
for FOD.8
FocOD is confined to a single sextant (Figs 6 and 7).8 It
may present as a single lesion or as a group of juxtaposed
lesions.8 Prior to its recognition as a discrete entity, its
frequent solitary clinical presentation led to its inclusion
among OFs, which had already been discussed in reference
1. The radiological challenge in the diagnosis of FOLs is
greatest when attempting to distinguish between a OF and a
single FocOD lesion, particularly when the lesion is small
(Fig 6). Although OFs are associated with root resorption,
this is not a feature of most OFs especially the small ones.
This creates a dilemma for the dentist: should the lesion be
removed, which may be an OD and therefore better left
alone, or risk leaving the lesion, which may subsequently
prove to be an OF requiring even more extensive surgery?
The latter needs to be considered, particularly if complex
restorative treatment is being contemplated. Later surgery
to remove the now-enlarged OF may compromise the suc-
cess of that treatment.
Periapical cemental dysplasia (PCD (Fig 7); this should
really be periapical osseous dysplasia to be consistent with
Brannon and Fowler’s4 recommendations discussed in the
introduction) first appeared in the WHO’s first edition.49
This lesion appears initially as periapical radiolucencies
associated with non-caries lower incisors (Fig 7). It is not an
infrequent experience for the dental specialist to observe
root-treated lower incisors, whose apices are now associ-
ated with mature OD lesions. Avoidance of unnecessary
treatment can be achieved by pulp-vitality testing of all
periapical lesions, especially radiolucencies.
FocOD overwhelmingly affects females (88%8). The mean
age is 41 years of age. Sixty-four percent present as inci-
dental findings, 25% with swelling, 28% with pain, and 17%
with numbness. The last was reported only in East Asians.
Eighty-five percent affect the mandible. The posterior sex-
tants of the mandible and of maxilla are affected in 80% and
74%, respectively.8
Just over half (53%) of cases were well-defined. Of these
40% display a sclerotic periphery. Forty-nine percent of the
cases appear in dentate (teeth were still present) areas of
the jaws and the other 51% in edentulous (teeth were
extracted) areas. In the dentate areas, no tooth displace-
ment or root resorption is observed.8
The predominate radiological patterns are radiolucency
(31%), “central radiopacity within a radiolucency” (37%),
and “complete radiopacity” (32%). East Asian reports
display significantly more complete radio-opacities and
fewer radiolucencies than those of Western communities.8
This may reflect the older mean age of the East Asian
global group at first presentation (47 years) in comparison
to that of the Western global group (39 years).8
OD has been reported in association with neurofibro-
matosis type 1 (NF1), an autosomal dominant disease, with
some malignant transformation potential.52 Missing and
unerupted teeth, overgrowth of the alveolus, and dilated
mandibular canals and mandibular foramina are well
recognized dental manifestations of NF1. The radiolucent-
staged lesions appeared as periapical radiolucencies,
although suggestive of inflammatory disease, responded
positively to pulp-vitality testing. Although a few presented
with a “central radiopacity within a radiolucency,” the
reader should be cautious as not one case in the report of
Visnapuu et al.52 had been confirmed by histopathological
examination. Although these features presented in 35% of
women in that report,52 the authors’ later report on another
group of NF1 patients, did not reveal a single case.53 No
explanation was given for this difference between these
Finnish reports.
Expansive osseous dysplasia
Expansive osseous dysplasia (EOD) has a similar histo-
pathological presentation of conventional OD, but differs in
its clinical and radiological presentations and behaviour.
Young et al.54 coined the term familial gigantiform
cementoma (FGC) to differentiate them from the FODs, as
these lesions are usually multiple. Their expansion is so
extensive that, unlike FODs, they also occupy the basal as
well as the alveolar processes. They exhibit substantial
buccolingual expansion that may prompt excision.
Although this term was affirmed by Waldron,13 its overall
 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
appropriateness was recently challenged and the more
descriptive EOD offered as an alternative.55 Although no
oncogenes have been reported so far, many of these lesions
generally exhibit neoplastic behaviour. They are autosomal
dominant,54 have a family history, and affect the young of
both genders equally. Initially EODs were reported in
Caucasian kindreds,13 but have since been reported in other
global communities, and most recently, East Asians.56 There
have also been cases that present as EOD, but have no family
history. Such cases have been reportedly so far in adolescent
males (of East Asian57,58 and of Indian origin,59) and
American females.60 All displayed very aggressive behav-
iour and achieved substantial dimensions.
Three out of four males of an Iranian kindred, presenting
with familial EOD, also had multiple long bone fractures.61
When patients present with extensive bilateral, almost
symmetrical involvement by multiple radio-opacities of
both the alveolar and basal processes of both jaws, Gard-
ner’s syndrome should also be considered. Lee et al.62 re-
ported such a case in a middle-aged woman.
FOD and FocOD (including PCD) are best reviewed.
Although surgery should be avoided unless the lesions
produce symptoms, this option may no longer be tenable if
the edentulous site is required for an osseointegrated
implant.8 In such a case, in the absence of published evi-
dence currently, it is advised to surgically ablate the lesion
by lateral trepanation and curettage, then allowing it to heal
first. This should minimize a real risk of failure of the
implant by insertion into abnormal tissue. Only EOD require
routine surgical ablation because of their frequently
aggressive behaviour.
Conclusions and imaging strategy
The role of the radiologist is central to the definitive
diagnosis of a particular FOL, because their histopathologies
are similar. FOLs affecting the face and jaws differ from their
clinical, radiological, and histopathological (no cartilage in
maxillofacial cases) presentations elsewhere in the body.
Although, some, such as ODs, are very prevalent,
particularly in middle-to-old aged females of East Asian or
sub-Saharan African origin, others, such as FD and OF,
though less frequent have more significant management
implications for the patient. Their clinical presentations
differ. OD are most frequently observed as incidental find-
ings on panoramic radiographs, whereas FDs and OF are
most frequently discovered when the patient presents with
swelling.
Although the primary diagnosis is performed in almost
all cases from the clinical findings and conventional radio-
graphs, cross-sectional imaging is frequently required of
lesions affecting the anatomically complex maxillary sinus.
As most monostotic FDs first present in adolescents and
young adults, use of CBCT, rather than CT, may be better
indicated if such cross-sectional imaging is necessary, as it
imparts a reduced radiation dose. For the same reason CBCT
would better assist in the investigation of a reactivated FD of
the face and/or jaws. Polyostotic FD would better benefit
35
from MRI, especially in the young patient, particularly if
vision is threatened by FD of the base of the skull. Although
most ODs are readily diagnosed solely by their conventional
radiological features, some, particularly the solitary lesions,
have a more extensive differential diagnosis as already
established in MacDonald-Jankowski’s Fig 121; this has
been updated and developed further in a recent textbook.9
Recently malignant transformation in FD of the face and
jaws has been reported in polyostotic cases.
The clinician should be aware of the association between
OF and HPT-JT that may be caused by a secreting carcinoma,
It may also provoke a hypercalcaemic crisis. Blood chemis-
try and nuclear medicine may be indicated.
Acknowledgements
Figs 2, Dr. Bastidas, Oral and Maxillofacial Surgeon,
Montefiore Medical Centre, the Bronx, New York. Fig 3, Dr
Kenneth Chow, Oral and Maxillofacial Surgeon, Vancouver,
Canada. Fig 4, Dr Raman Sumanth, Melaka Manipal Medical
College, Malaysia. Fig 5, Dr Thomas Li, Oral and Maxillofacial
Radiologist, Hong Kong. Fig 6, Dr Jason Chen, Oral and
Maxillofacial Surgeon, Vancouver, Canada. Fig 7, Dr Michael
Matwychuk, Okanagen Endodontic Specialists, Canada.
Fig 8, Dr David Gane, CEO and President LED Medical Di-
agnostics. Fig 9, Dr Allan Abuabara, Specialist in Dental
Radiology and Imaging, Joinville, Brasil. Fig 10, Wiley-
Blackwell, Fig 11.35 in MacDonald D. Oral and Maxillofacial
Radiology: a Diagnostic Approach. 2011.
References
1. MacDonald-Jankowski DS. Fibro-osseous lesions of the face and jaws.
Clin Radiol 2004;59:11e25.
2. Slootweg PJ, Mofty SK, Ossifying fibroma. Barnes L, et al., editors. WHO
classification of tumours, pathology and genetics of tumours of the head
and neck. Lyon: International Agency for Research on Cancer (IARC);
2005. p. 319e20.
3. Slootweg PJ, Osseous dysplasia. Barnes L, et al., editors. WHO classifica-
tion of tumours, pathology and genetics of tumours of the head and neck.
Lyon: International Agency for Research on Cancer (IARC); 2005. p. 323.
4. Brannon RB, Fowler CB. Benign fibro-osseous lesions: a review of cur-
rent concepts. Adv Anat Pathol 2001;8:126e43.
5. Schajowicz FM. Histological typing of bone tumours. In: International
histological classification of tumours. 2nd ed. London: Springer-Verlag;
1993. p. 39e40.
6. MacDonald-Jankowski DS. Fibrous dysplasia: a systematic review. Den-
tomaxillofac Radiol 2009;38:198e215.
7. MacDonald-Jankowski DS. Ossifying fibroma: systematic review. Den-
tomaxillofac Radiol 2009;38:495e513.
8. MacDonald-Jankowski DS. Focal cemento-osseous dysplasia: a system-
atic review. Dentomaxillofac Radiol 2008;37:350e60.
9. MacDonald D. Oral and maxillofacial radiology; a diagnostic approach.
Wiley Blackwell; 2011. 33,34, 164e80, 208e10, 215e19, 270, 290,
300e2.
10. Cavalli-Sforza LL, Menozzi P, Piazza A. The history and geography of hu-
man genes. Abridged paperback edition. New Jersey: Princeton University
Press; 1996. See figures 2.3.2.A and B.
11. Tang H, Quertermous T, Rodriguez B, et al. Genetic structure, self-
identified race/ethnicity, and confounding in caseecontrol association
studies. Am J Hum Genet 2005;76:268e75.
12. MacDonald D. Oral and maxillofacial radiology; a diagnostic approach.
Wiley Blackwell; 2011. p. 59e66. 255e60.
36 D.S. MacDonald / Clinical Radiology 70 (2015) 25e36
13. Waldron CA. Fibro-osseous lesions of the jaws. J Oral Maxillofac Surg
1993;51:828e35.
14. Eisenberg E, Eisenbud L. Benign fibro-osseous diseases: current con-
cepts in historical perspective. Clin Nor Am 1997;9:551e62.
15. Jundt G, Fibrous dysplasia. Barnes L, et al., editors. WHO classification of
tumours, pathology and genetics of tumours of the head and neck. Lyon:
International Agency for Research on Cancer (IARC); 2005. pp; 321e2.
16. Shi RR, Li XF, Zhang R, et al. GNAS mutational analysis in differentiating
fibrous dysplasia and ossifying fibroma of the jaw. Mod Pathol
2013;26:1023e31.
17. Patel MM, Wilkey JF, Abdelsayed R, et al. Analysis of GNAS mutations in
cemento-ossifying fibromas and cemento-osseous dysplasias of the
jaws. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010;109:739e43.
18. Chapurlat RD, Orcel P. Fibrous dysplasia of bone and McCuneeAlbright
syndrome. Best Pract Res Clin Rheumatol 2008;22:55e69.
19. Collins MT, Singer FR, Eugster E. McCuneeAlbright syndrome and the
extraskeletal manifestations of fibrous dysplasia. Orphanet J Rare Dis
2012;7(Suppl. 1):S4.
20. Ferreira EC, Brito CC, Domingues RC, et al. Whole-body MR imaging for
the evaluation of McCune-albright syndrome. J Magn Reson Imaging
2010;31:706e10.
21. Gaume tou E, Tomeno B, Anract P. Mazabraud’s syndrome. A case with
multiple myxomas. Orthop Traumatol Surg Res 2012;98:455e60.
22. Cohen Jr MM. The new bone biology: pathologic, molecular, and clinical
correlates. Am J Med Genet A 2006;140:2646e706.
23. Vo € lkl TM, Do € rr HG. McCuneeAlbright syndrome: clinical picture and
natural history in children and adolescents. J Pediatr Endocrinol Metab
2006;19(Suppl. 2):551e9.
24. Huening MA, Reddy S, Dodd LG. Fine-needle aspiration of fibrous
dysplasia of bone: a worthwhile endeavor or not? Diagn Cytopathol
2008;36:325e30.
25. Wootton-Gorges SL. MR imaging of primary bone tumors and tumor-
like conditions in children. Magn Reson Imaging Clin N Am
2009;17:469e87.
26. MacDonald-Jankowski DS, Li TK. Fibrous dysplasia in a Hong Kong
community; the clinical and radiological presentations and the out-
comes of treatment. Dentomaxillofac Radiol 2009;38:63e72.
27. Slootweg PJ, Mu € ller H. Differential diagnosis of fibro-osseous jaw le-
sions. A histological investigation on 30 cases. J Craniomaxillofac Surg
1990;18:210e4.
28. Sperber GH. Craniofacial development. Hamilton, Canada: BC Decker Inc;
2001. p. 67.
29. MacDonald-Jankowski DS, Yeung R, Li TK, et al. Computed tomography
of fibrous dysplasia. Dentomaxillofac Radiol 2004;33:114e8.
30. Sontakke SA, Karjodkar FR, Umarji HR. Computed tomographic features
of fibrous dysplasia of maxillofacial region. Imaging Sci Dent
2011;41:23e8.
31. Ricalde P, Magliocca KR, Lee JS. Craniofacial fibrous dysplasia. Oral
Maxillofac Surg Clin North Am 2012;24:427e41.
32. Petrikowski CG, Pharoah MJ, Lee L, et al. Radiographic differentiation of
osteogenic sarcoma, osteomyelitis, and fibrous dysplasia of the jaws.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;80:744e50.
33. De Mattos CB, Binitie O, Dormans JP. Pathological fractures in children.
Bone Joint Res 2012;1:272e80.
34. MacDonald-Jankowski DS. Fibrous dysplasia in the jaws of a Hong Kong
population: radiographic presentation and systematic review. Dento-
maxillofac Radiol 1999;28:195e202.
35. Alvares LC, Capelozza AL, Cardoso CL, et al. Monostotic fibrous dysplasia:
a 23-year follow-up of a patient with spontaneous bone remodeling.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009;107:229e34.
36. Daly BD, Chow CC, Cockram CS. Unusual manifestations of craniofacial
fibrous dysplasia: clinical, endocrinological and computed tomographic
features. Postgrad Med J 1994;70:10e6.
37. Akintoye SO, Boyce AM, Collins MT. Dental perspectives in fibrous
dysplasia and McCuneeAlbright syndrome. Oral Surg Oral Med Oral
Pathol Oral Radiol 2013;116:e149e55.
38. Salmasi V, Blitz AM, Ishii M, et al. Expanded endonasal endoscopic
approach for resection of a large skull base aneurysmal bone cyst in a
pediatric patient with extensive cranial fibrous dysplasia. Childs Nerv
Syst 2011;27:649e56.
39. Cheng J, Yu H, Wang D, et al. Spontaneous malignant transformation in
craniomaxillofacial fibrous dysplasia. J Craniofac Surg 2013;24:141e5.
40. Sun TT, Tao XF, Shi HM. Spontaneous osteosarcoma in craniomax-
illofacial fibrous dysplasia: clinical and computed tomography imaging
features in 8 cases. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
2014;118:e24e31.
41. Puls F, Niblett AJ, Mangham DC. Molecular pathology of bone tumours:
diagnostic implications. Histopathology 2014;64:461e76.
42. Dujardin F, Binh MB, Bouvier C, et al. MDM2 and CDK4 immunohisto-
chemistry is a valuable tool in the differential diagnosis of low-grade
osteosarcomas and other primary fibro-osseous lesions of the bone.
Mod Pathol 2011;24:624e37.
43. Hall G. Fibro-osseous lesions of the head and neck. Diag Histopathol
2012;18:149e58. http://dx.doi.org/10.1016/j.mpdhp.2012.01.005.
44. Urs AB, Kumar P, Arora S, et al. Clinicopathologic and radiologic corre-
lation of ossifying fibroma and juvenile ossifying fibromadan institu-
tional study of 22 cases. Ann Diagn Pathol 2013;17:198e203.
45. MacDonald-Jankowski DS, Li TK. Ossifying fibroma in a Hong Kong
community; the clinical and radiological presentations and the out-
comes of treatment. Dentomaxillofac Radiol 2009;38:514e23.
46. Chen JD, Morrison C, Zhang C, et al. Hyperparathyroidismejaw tumour
syndrome. J Intern Med 2003;253:634e42.
47. Aldred MJ, Talacko AA, Savarirayan R, et al. Dental findings in a family
with hyperparathyroidismejaw tumor syndrome and a novel HRPT2
gene mutation. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
2006;101:212e8.
48. Gill AJ. Understanding the genetic basis of parathyroid carcinoma.
Endocr Pathol 2014;25:30e4.
49. Pindborg JJ, Kramer IRH, Torloni H. Histological typing of odontogenic
tumours, jaw cysts and allied lesions. WHO International histological
classifications of tumours. No.5. Geneva: WHO; 1971. p. 31e4.
50. MacDonald-Jankowski DS. Florid cemento-osseous dysplasia: a sys-
tematic review. Dentomaxillofac Radiol 2003;32:141e9.
51. Ariji Y, Ariji E, Higuchi Y, et al. Florid cemento-osseous dysplasia.
Radiographic study with special emphasis on computed tomography.
Oral Surg Oral Med Oral Pathol 1994;78:391e6.
52. Visnapuu V, Peltonen S, Ellil€ a T, et al. HRPT2. Peri-apical cemental
dysplasia is common in women with NF1. Eur J Med Genet
2007;50:274e80.
53. Visnapuu V, Peltonen S, Tammisalo T, et al. Radiographic findings in the
jaws of patients with neurofibromatosis 1. J Oral Maxillofac Surg
2012;70:1351e7.
54. Young SK, Markowitz NR, Sullivan S, et al. Familial gigantiform
cementoma: classification and presentation of a large pedigree. Oral
Surg Oral Med Oral Pathol 1989;68:740e7.
55. Noffke CE, Raubenheimer EJ, MacDonald D. Fibro-osseous disease:
harmonizing terminology with biology. Oral Surg Oral Med Oral Pathol
Oral Radiol 2012;114:388e92.
56. Shah S, Huh KH, Yi WJ, et al. Follow-up CT findings of recurrent familial
gigantiform cementoma of a female child. Skeletal Radiol 2012;41:341e6.
57. Miyake M, Nagahata S. Florid cemento-osseous dysplasia. Report of a
case. Int J Oral Maxillofac Surg 1999;28:56e7.
58. Ong ST, Siar CH. Florid cemento-osseous dysplasia in a young Chinese
man. Case report. Aust Dent J 1997;42:404e8.
59. Kumar VV, Ebenezer S, Narayan TV, et al. Clinicopathologic conference:
multiquadrant expansile fibro-osseous lesion in a juvenile. Oral Surg
Oral Med Oral Pathol Oral Radiol 2012;113:286e92.
60. Abdelsayed RA, Eversole LR, Singh BS, et al. Gigantiform cementoma:
clinicopathologic presentation of 3 cases. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2001;91:438e44.
61. Moshref M, Khojasteh A, Kazemi B, et al. Autosomal dominant giganti-
form cementoma associated with bone fractures. Am J Med Genet A
2008;146A:644e8.
62. Lee BD, Lee W, Oh SH, et al. A case report of Gardner syndrome with
hereditary widespread osteomatous jaw lesions. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2009;107:e68e72.