1017

Fibrous Dysplasia vs
Adamantinoma of the Tibia:
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

Differentiation Based on Discriminant

Analysis of Clinical and Plain Film Findings

Johan L. Bloem1 Differentiation between benign fibrous dysplasia and malignant adamantinoma of the
Roy 0. van der HeuI2 tibia is challenging because of the impact the diagnosis has on the choice of treatment
Herman M. Schuttevaer1 (none or extensive surgery). The histologic and pathologic similarities of the lesions and
Dirkjan Kuipers1 the controversial relationship between fibrous dysplasia, osteofibrous dysplasia, and
adamantinoma complicate the matter. We found a large overlap of histologic features
in lesions considered either fibrous dysplasia or osteofibrous dysplasia on the basis of
the radiologic findings. The purpose of this study was to determine the value of the
plain radiograph of the lower leg in combination with clinical findings to differentiate the
benign from the malignant condition The clinical symptoms, radiographs, and histologic
slides of 46 patients with fibrous dysplasia and 22 with adamantinoma in the tibia were
reviewed retrospectively. In only one of 12 patients with radiologic or histologic char-
acteristics of osteofibrous dysplasia were both radiologic and histologic criteria for the
diagnosis present. A linear discriminant analysis was performed on six clinical (age,
spontaneous pain, pain after trauma, swelling only, pain and swelling, and bowing
deformity) and 25 radiologic signs. Fibrous dysplasia and its variant osteofibrous
dysplasia could be identified correctly in 87% (40 of 46 patients) and adamantinoma in
95% (21 of 22 patients) by using the patient’s age and four radiologic signs. When
results from the discnminant analysis of a randomized subgroup of patients (32) were
used on the other subgroup (36 patients), fibrous dysplasia was correctly identified in
84% (21 of 25) and adamantinoma in 82% (nine of 11). Fibrous dysplasia is more
prevalent than adamantinoma in a young patient, when radiographs show a ground-
glass appearance and anterior bowing and when there is no multilayered periosteal
reaction and moth-eaten destruction.
When radiologic signs and the patient’s age are combined, fibrous dysplasia and
adamantinoma can be discriminated in a high percentage of patients.

AJR 156:1017-1023, May 1991

The tibia is one of the preferential sites of fibrous dysplasia, a benign skeletal
developmental anomaly of bone-forming mesenchyma [1 -4]. A related condition
called osteofibrous dysplasia or ossifying fibroma is found exclusively in the tibia
or fibula [5-7]. Adamantinoma is a rare malignant neoplasm of bone that is typically
found in the tibia [8-i 1].
Differentiation between these three lesions is important in planning therapy.
Because fibrous dysplasia and adamantinoma both have large amounts of fibrous
stroma, radiologic and histologic differentiation between the two is difficult [1 4, ,
Received March 30, 1990: accepted after revi-
sion November 20, i990. 9-i 3]. The matter is further complicated by the fact that the relationship between
I Department of Radiology, Bldg. I, C2-S, Uni- fibrous dysplasia, osteofibrous dysplasia, and adamantmnoma is controversial [5-
versity Hospital Leiden, Rijnburgerweg 10, 2333 7, 1 2, 1 3]. The purpose of this study is to determine if clinical or radiologic features
AA Leiden, the Netherlands. Address reprint re can be used to differentiate the two benign disorders from adamantinoma.
quests to J. L. Bloom.
2 Department of Pathology, University Hospital
Dijkzicht Rotterdam, Dr. Molewaterplein 4, 3015
Materials and Methods
GD Rotterdam, the Netherlands.
036i-803x/91/1565-1017 We reviewed the files of the Registry of the Netherlands Committee on Bone Tumors,
0 American Roentgen Ray Society which contained 7500 files on patients at the time of the study. Although most new patients
 1018 BLOEM ET AL. AJR:156, May 1991

Fig. 2.-iO-year-old girl with ada-

mantinoma and painful swelling. Lat-
eral radiograph shows anterior bow-
ing. Tumor is multilobulated, contains
intralesional opacifications, and cx-
tends into cortex. Tumor is confined
within newly formed cortex (balloon-
ing). This is one of four patients with
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

adamantinoma in whom extensive fi-

brous stroma was found.

2
E
z

Fig. 1.-Graph shows age distribution of 46 patients with fibrous dys-

plasla and 22 patients with adamantinoma.

with bone tumors in the Netherlands are referred to this committee

for diagnosis, the true prevalence is not reflected in the files because
less troublesome cases are treated without consultation. Before
osteofibrous dysplasia was described as a separate entity [5-7], this
diagnosis was included in the cases of fibrous dysplasia. Distinction
between the two was made at the subsequent review for this study.
Seventy-two patients were identified with a histologic diagnosis of Characteristic findings indicative of fibrous dysplasia included fibrous
either adamantinoma (22 patients) or fibrous dysplasia (50 patients) stroma consisting of myxofibrous tissue and woven bone [1 -4]. The
in the tibia. Four of the patients with fibrous dysplasia were excluded presence of woven bone trabeculae rimmed by cuboidal active os-
because of extensive polyostotic disease (three patients) or the teoblasts (graded as high osteoblast activity on our three-point scale)
presence of McCune-Albright syndrome (one patient). Six patients and small bands of lamellar bone in the periphery (zonal architecture)
were included with lesions in both the ipsilateral fibula and tibia. was considered to be consistent with osteofibrous dysplasia [5-7].
Twenty-six males and 20 females had fibrous dysplasia of the tibia. The histologic diagnosis of adamantinoma was made when epithe-
The only patient with a lesion manifesting both the histologic and liallike cells arranged in pallisading nests and strands of cells were
radiologic criteria of osteofibrous dysplasia was, for analysis, included identified [9, 11-14].
among the fibrous dysplasia group. The adamantinoma group had In the 46 patients with fibrous dysplasia, histologic studies showed
eight males and 14 females. The mean age of patients with fibrous a marked range of remodeling activity. This varied from a near
dysplasia was 12 years (range, 1 -37 years), and of patients with absence of osteoblasts and osteoclasts (low osteoblast activity) to
adamantinoma, 23 years (range, 1 0-71 years) (Fig. 1). The clinical the presence of a large number of osteoblasts and osteoclasts (high
presentation was not recorded in two patients with fibrous dysplasia. activity), and from a pure woven bone pattern to a combination of
Patients with fibrous dysplasia presented with spontaneous pain (12 woven and lamellar bone. In four patients with adamantinoma, large
patients), pain after trauma (1 6 patients, nine of whom had a fracture), fields with fibrous stroma were encountered (Fig. 2).
swelling only (nine patients), pain and swelling (six patients), or a Follow-up data were available on 29 patients (17 with fibrous
bowing deformity (one patient). The following distribution of clinical dysplasia and 12 with adamantinoma). Follow-ups ranged from 5
signs was found in patients with adamantinoma: pain only in seven months to 22 years. The mean follow-up time was 4.5 years for
patients, pain after trauma in four patients (three of whom had a patients with fibrous dysplasia and 9.6 years for those with adaman-
fracture), swelling only in six patients, and pain and swelling in five tinoma.
patients. The average duration of symptoms was 6 months (range, Clinical characteristics analyzed included age and presenting
0-78 months) in patients with fibrous dysplasia (n = 44) and 31 symptoms (spontaneous pain, pain after trauma, swelling only, pain
months (range, 0-200 months) in patients with adamantinoma (n = and swelling, and bowing deformity). Anteroposterior and lateral
22). radiographs of the lower leg were reviewed by two radiologists in
The diagnoses were made by consensus by the Netherlands conference. Although the observers were aware that the patients had
Committee on Bone Tumors, consisting of at least two pathologists, either fibrous dysplasia or adamantinoma, the final diagnosis was not
two radiologists, and two orthopedic surgeons. All members of this available to them. The radiologic features analyzed are listed in Table
committee have extensive experience in the field of bone tumors. In 1 . Type of bone destruction was classified according to Lodwick
each case, the diagnosis was based on histopathology, plain film [1 5]. In order to discriminate anterior bowing from an expansive lesion
radiographs, and clinical evaluation. The histologic material was ob- involving the anterior cortex, bowing was defined as a deformity of
tamed by surgery in 38 patients with fibrous dysplasia and by biopsy the posterior cortex of the tibia on a lateral radiograph.
in eight patients. Pathologic material after resection was available in A linear discriminant analysis using all the clinical and radiologic
all 22 patients with adamantinoma. The histologic material was re- signs was performed to determine the feasibility of discriminating
viewed again by one of the authors. The osteoblast activity was fibrous dysplasia from adamantinoma. We also performed a discrim-
graded arbitrarily on a three-point scale as low, intermediate, or high. inant analysis based on selected signs. This subset of signs, derived
 AJR:156, May 1991 TIBIAL FIBROUS DYSPLASIA VS ADAMANTINOMA 1019

TABLE 1: Radiologic Findings in Fibrous Dysplasia and Adamantinoma of the Tibia

No. (%)
Finding
Fibrous Dysplasia Adamantinoma
(n=46) (n=22)

Fracture 9 (20) 3(14)

Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

Anterior bowinga 11 (24) 1 (5)

Additional lesion(s) in the fibula 6 (13) 1 (5)
Multifocal lesions in tibia 5 (1 1 ) 4 (18)
Lesion(s) in diaphysis only 43 (93) 20 (91)
Lesion(s) in diaphysis and metaphysis 2 (4) 2 (9)
Lesion(s) in metaphysis and epiphysis 1 (2) 0
Eccentric localization 23 (50) 12 (55)
Central localization 23 (50) 10 (45)
Geographic destructiona 46 (1 00) 20 (91)
Local moth-eaten destructiona o 2 (9)
Predominantly osteolysis 3 (7) 2 (9)
Osteolysis with ossification or septationa 10 (22) 17 (77)
Ground-glass patterna 14 (30) 0
Ground glass with ossification or septationa 19 (41) 3 (14)
Smooth margin 22 (48) 9 (41)
Lobulated margin 24 (52) 1 3 (59)
Sclerotic margin 27 (59) 9 (41)
Regular cortical destruction 46 (1 00) 21 (95)
Irregular cortical destruction 0 1 (5)
Intracortical localization 6 (13) 2 (9)
No peniosteal new bone formation 3 (7) 4 (18)
Single-layered lamellar periosteal reaction 10 (22) 2 (9)
Multilayered lamellar periosteal reactiona 1 (2) 5 (23)
Ballooning 32 (70) 11 (50)
a Significantly different for fibrous dysplasia and adama ntinoma
(p < .05).

from the literature, included pain after trauma, age, ground-glass

appearance with and without densities, multilayered periosteal reac-
tion, moth-eaten destruction, anterior bowing, skip lesions in tibia,
and additional lesions in fibula [1 , 2, 4, 8-1 1 , 16-1 8]. The results
subsequently were tested by randomizing the patients into two
subsets and applying the discriminating functions derived from one
group to the other.

Results

Eight of the 68 lesions were located in the ventral cortex

of the tibia. The average age of these patients was 1 3 years
(range, 4-1 7 years). Because of localization and age, a radio-
logic diagnosis of osteofibrous dysplasia instead of fibrous
dysplasia was considered. However, the osteoblastic activity
in these patients was graded high in only one. In two of the
eight patients 1 4 and 1 5 years old, a histologic diagnosis of
adamantinoma was made easily. One of these patients had
low osteoblastic activity, the other had intermediate osteo- Fig. 3.-is-year-old girl with intracortical fibrous dysplasia and pain of
blastic activity. The osteoblastic activity in the remaining five 1 month’s duration. Lateral radiograph shows a small lesion with well-
defined margins in anterior cortex. No periosteal reaction was found.
patients was intermediate in two and low in three (Fig. 3). In Histologic examination showed low osteoblastic activity.
four other patients, high osteoblastic activity, such as occurs
in osteofibrous dysplasia, was found. One of these was the
patient with infantile pseudarthrosis. In the other three pa-
tients, a radiologic diagnosis of fibrous dysplasia instead of sia were both the histologic and radiologic criteria in accord-
osteofibrous dysplasia was made because the lesions were ance with the diagnosis of osteofibrous dysplasia. This patient
located within the marrow cavity and cortex instead of the was included in the analysis of patients with fibrous dysplasia.
anterior tibial cortex. The ages of these three patients were The radiologic findings are listed in Table 1 and are illus-
3, 1 3, and 24 years. Thus, in only one of these patients with trated in Figs. 2-1 0. The patients who had fractures were
radiologic or histologic characteristics of osteofibrous dyspla- part of the group of patients with pain after trauma (Fig. 4).
 1020 BLOEM ET AL. AJR:156, May 1991
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

Fig. 4.-9-year-old boy with fibrous dysplasia who had a fracture after minor trauma. Lateral radiograph shows a lobulated, osteolytic lesion with an
oblique fracture, located centrally in diaphysis.
Fig. 5.-7-year-old girl with fibrous dysplasia and acute onset of pain in lower leg. Lateral radiograph shows two lesions in tibia and one in fibula.
Lesion in proximal part of tibia is central; lesion in distal part of tibia is eccentric. All three lesions are osteolytic with marked septation. Lesions are
expansive and involve cortical bone. Anterior bowing of tibia is present.
Fig. 6.-i-year-old infant with pseudarthrosis and bowing deformity of lower leg. Lateral radiograph shows a defect in fibula with narrowing of marrow
canal and tapering of both ends. Lesion in tibia is cystic and is characterized by anterior bowing and valgus deformity (not shown). Diameter of tibia is
enlarged by callus formation. Tibial fracture is consolidated but still visible. Histologic diagnosis was fibrous dysplasia.
Fig. 7-19-year-old woman with adamantinoma who had pain after trauma. Lateral radiograph shows multiple lobulated osteolytic lesions in both tibia
and fibula. Lesion in proximal part of tibia is eccentric; lesions in midshaft and distal part of tibia are central. Cortical bone is thinned at endosteal side.
Penosteal bone formation resulted in an intact neocortex, which confines lesions.

One of the patients with a bowing deformity (Figs. 2 and 5) peniosteal reaction, moth-eaten destruction, anterior bowing,
had infantile pseudarthrosis (Fig. 6). Neither this patient nor skip lesions in tibia, and additional lesions in fibula) were used,
his family had stigmata of neurofibromatosis. No tumor, but fibrous dysplasia was identified correctly in 87% (40 of 46)
cellular fibrous tissue with woven bone rimmed by osteo- and adamantinoma in 95% (21 of 22). Six patients with fibrous
blasts, was found. Lamellar bone was present in the peniph- dysplasia and one with adamantinoma were not identified
ery. The osteoblastic activity was graded as high. correctly. Use of the results from the discriminant analysis of
Patients with fibrous dysplasia could be identified in 89% one of the randomized patient groups (32 patients) on the
(41 of 46 patients) and patients with adamantinoma could be other (36 patients) yielded the following results: fibrous dys-
identified in 95% (21 of 22) of cases when age, presenting plasia was correctly identified in 84% (21 of 25) and adaman-
symptoms, and all 25 radiologic signs (Table 1) were sub- tinoma in 81 % (nine of 1 1).
jected to a discniminant analysis. Five patients with fibrous The five signs with the greatest impact on making a diag-
dysplasia and one with adamantinoma were not identified nosis were, in order of decreasing importance, age, ground-
correctly. The one patient with histologic and radiologic signs glass appearance with or without ossification or septation (in
of osteofibrous dysplasia was identified as having fibrous favor of fibrous dysplasia), multilayered lamellar peniosteal
dysplasia and not adamantinoma. Fibrous dysplasia was iden- reaction (in favor of adamantinoma), moth-eaten destruction
tified correctly in 84% of cases (21 of 25) and adamantinoma (in favor of adamantinoma), and anterior bowing (in favor of
in 73% (eight of 1 1) when all radiologic and clinical signs used fibrous dysplasia). The other radiologic and clinical signs were
in a discniminant analysis on a randomized subset of patients not important in differentiating fibrous dysplasia from ada-
(32 patient) subsequently were evaluated on the other subset mantinoma.
(36 patients). None of the patients with adamantinoma had distant me-
When our selected signs (pain after trauma, age, ground- tastases on conventional chest radiographs at the time of
glass appearance with and without densities, mutilayered presentation. Later, metastases developed in the lung in three
 AJR:156, May 1991 TIBIAL FIBROUS DYSPLASIA VS ADAMANTINOMA 1021
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

Fig. 8.-6i-year-old man with adamantinoma and pain in tibia of 2 month’s duration. Lateral radiograph shows intramedullary osteolytic lesion and
endosteal thinning of cortex. Although bone destruction is partly geographic, especially in proximal and most distal parts, a local area with moth-eaten
destruction can be seen (arrows).
Fig. 9.-37-year-old man with fibrous dysplasia and pain in lower leg for 18 months. Anteroposterior radiograph shows typical ground-glass appearance.
Endosteal scalloping is present. Newly formed cortex (ballooning) confines lesion.
Fig. 10-34-year-old woman with an adamantinoma and pain in lower leg for 3 months.
A, Lateral radiograph shows a large expansive intracortical lesion and scalloping of endosteal margin of cortex. Intralesional calcifications can be seen.
B, Radiograph 8 months later shows enlargement of tumor and soft-tissue extension.

patients, and osseous and lymph node metastases developed relation of fibrous dysplasia to osteofibrous dysplasia, also
in another one. In one patient, a solitary liver metastasis was called ossifying
fibroma, and adamantinoma is controversial
found. In these five patients, at least one metastasis was [5-8, 13, 1 9]. The name osteofibrous dysplasia [1 6] is pre-
proved histologically. The disease-free interval in these five ferred to ossifying fibroma [7] because it better reflects the
patients ranged from 6 to 1 3 years. relation with fibrous dysplasia and does not cause confusion
with ossifying fibroma of the mandible and maxilla, which is a
nonrelated entity [20, 21 ]. Osteofibrous dysplasia has been
Discussion described as a bone-forming lesion located in the ventral,
intracortical area of the tibial shaft with histology different
Differentiation between fibrous dysplasia and adamanti- from fibrous dysplasia [5, 6, 19]. Contrary tofibrous dysplasia,
noma may be difficult when the two conditions are confined the formed, woven bone trabeculae in osteofibrous dysplasia
to the tibia because their radiologic appearance is similar. are rimmed by cuboidal osteoblasts.
Histologic findings for the two entities may be similar and, In our patients, radiologic signs of osteofibrous dysplasia
according to some authors, may even be related [1 , 4, 9-1 1, or ossifying fibroma (eight patients) and histologic evidence
16]. of high osteoblastic activity of osteofibrous dysplasia (five
Fibrous dysplasia may be monostotic (70-80%) or polyos- patients) were combined in only one patient. In lesions con-
totic (20-30%); only 2-3% of polyostotic cases are associ- sidered radiologically to be fibrous dysplasia or osteofibrous
ated with endocninopathy such as the McCune-Albnight syn- dysplasia, a marked remodeling activity was found; overlap
drome [1 -3]. Any bone may be affected, but the tibia is one was large and no basic difference was seen. Although osteo-
of the most frequently involved bones. Others are the rib, blastic rimming is not a feature of classic fibrous dysplasia,
femur, craniofacial bones, and humerus [1 4]. The true prey- , plump, elongated cells rimming woven bone and layering of
alence of fibrous dysplasia is not known because the lesion lamellar bone can be found in fibrous dysplasia [22]. We think
may be asymptomatic and thus may escape detection. The that ossifying fibroma or osteofibrous dysplasia is not a well-
 1022 BLOEM ET AL. AJR:156, May 1991

defined entity [7, 8, 19] but an active variant of fibrous the patients with an adamantinoma (Fig. 2) warrants some
dysplasia, an opinion shared by other authors [5, 6, 13]. constraint in using this criterion.
The prevalence of adamantinoma is estimated to be 0.1% The presence of infantile or congential pseudarthrosis,
of primary bone tumors [8]. A less frequently used synonym which was present in one of our patients (Fig. 6), allows the
is malignant angioblastoma. Typically, adamantinoma is found exclusion of adamantinoma from the list of differential diag-
in the tibia, but the tumors may occasionally be found at other noses [4, 1 7]. As was the case in our patient, on histologic
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

sites [9]. Although the histology of adamantinoma is reminis- examination, the area of pseudarthrosis may not contain
cent of that of ameloblastoma of the jaw, sometimes called tumorous tissue but only fibrous tissue with increased osteo-
adamantinoma of the jaw, no evidence has been reported blastic activity.
that suggests a relationship between adamantinoma of the Although patients with fibrous dysplasia more often had
tibia and ameloblastoma of the jaw [8-i 1]. Usually fibrous additional lesions in the fibula (1 3%, Fig. 5) than did patients
dysplasia and adamantinoma have clearly different histologic with adamantinoma (5%, Fig. 7), these additional lesions were
features. However, large fields of fibrous stroma (like those a relatively insignificant sign in differentiating between the two
that occur in fibrous dysplasia or osteofibrous dysplasia) may lesions. In their review of the literature, Moon and Mori [9]
mask adamantinomous tissue [1 2, 13]. We encountered this also found simultaneous localizations of adamantinoma in the
problem in four of our patients. The diagnosis of adamanti- tibia and fibula in 5% of their 200 reviewed patients. Solitary
noma could be made with confidence in these patients be- adamantinoma of the fibula may occur [18].
cause we included in this study only patients with ample Entirely intracortical localizations were seen in both fibrous
material available for histologic examination. Immunohisto- dysplasia (Fig. 3) and in adamantinoma (Fig. 1 0). This sign is
chemistry and electron microscopic demonstration of cyto- also not important in differentiating fibrous dysplasia from
keratins is helpful in identifying adamantinoma, especially in adamantinoma.
equivocal cases [14]. No metastases were found on chest radiographs of our
Depending on the signs used, fibrous dysplasia could be patients at presentation. Because metastases develop late in
identified correctly in 84% and adamantinoma in 73-82% of the course of adamantinoma, the absence of metastases
our patients. We used discriminant analysis and evaluated does not assist in making a differential diagnosis [1 1 1 2, 23, ,

the classification on an independent sample. These results 24].

were obtained by combining the clinical and radiologic signs. The role of CT and MR imaging in differentiating fibrous
The most important signs that suggest a diagnosis of fibrous dysplasia from adamantinoma is probably limited, considering
dysplasia over adamantinoma are, in order of decreasing the histology of both lesions. Involvement of cortical bone is
importance, young age, presence of ground-glass appear- demonstrated exquisitely well with both cross-sectional tech-
ance with or without additional opacifications, absence of niques [25-27]. The signal intensities of fibrous dysplasia are
multilayered peniosteal reaction and moth-eaten destruction, low on Ti -weighted images and variable on T2-weighted
and presence of anterior bowing. images [26, 28]. We found in four patients that the signal
Our patients who had fibrous dysplasia were usually intensity of adamantinoma is similar to that of other tumors
younger than those who had adamantinoma; all patients with and not specific (unpublished data).
fibrous dysplasia were less than 38 years old, and 46% of When no endocrine symptoms or obvious polyostotic fi-
them were 10 years old or younger. This is in accordance brous dysplasia is found, the radiograph of the lower leg is
with the age distribution reported by Gibson and Middlemiss used to assist in differentiating fibrous dysplasia from ada-
[1 6]. The peak occurrence of fibrous dysplasia is reported to mantinoma. When the selected signs described in this study
be between the ages of 1 0 and 1 5 years, and patients with are combined, the two disorders can be correctly identified in
polyostotic fibrous dysplasia or osteofibrous dysplasia are a high percentage of patients. However, the most important
even younger [1 4, 8, 1 9]. The ages of our patients with
, discriminating signs can occur in both lesions; therefore, a
adamantinoma were more or less equally distributed over the large biopsy to allow a histologic diagnosis often will be
second to fifth decades. The peak occurrence of adamanti- necessary. A histologic diagnosis of adamantinoma, with or
noma is reported to be between ages 1 1 and 30, but the without radiologic support, solves the diagnostic dilemma.
reported range in the literature is between 3 and 74 years [9, The impact of a combined radiologic and histologic diagnosis
13]. of fibrous dysplasia on management depends on the level of
A ground-glass appearance with or without intralesional confidence and clinical symptoms. When a histologic diagno-
opacifications frequently was found in fibrous dysplasia (72%, sis of fibrous dysplasia cannot be supported by the radiologic
Fig. 9), but it also was encountered in adamantinoma (14%). diagnosis, further action, such as obtaining follow-up studies
Conversely, a mostly osteolytic appearance, with or without or performing a new biopsy or sometimes even a resection,
intralesional opacifications, was definitely present more often will be necessary.
in adamantmnoma (86%) than in fibrous dysplasia (28%).
Reports on anterior bowing in patients with fibrous dyspla-
ACKNOWLEDGMENTS
sia are, in contrast to those on bowing deformity of the femur,
rather sparse [1 6]. In our patients, anterior bowing was found We thank the Netherlands Committee on Bone Tumors and the
in about one quarter of patients with fibrous dysplasia (24%, referring physicians for their cooperation. Statistic analysis was done
Fig. 5). However, the presence of anterior bowing in one of by Roland Brand.
 AJR:156, May 1991 TIBIAL FIBROUS DYSPLASIA VS ADAMANTINOMA 1023

REFERENCES the tibia: ultrastructural and immunohistochemic study with reference to

histogenesis. Clin Orthop i984:1 90:299-310
1 . Feldman F. Tuberous sclerosis, neurofibromatosis, and fibrous dysplasia. 15. Lodwick GD. Radiologic concepts. In: Hodes PJ, Lodwick GS, eds. Atlas
In: Resnick D, Niwayama G, eds. Diagnosis of bone and joint disorders, of tumor radiology: the bone and joints. Chicago: Year Book Medical,
2nd ed. Philadelphia: Saunders, 1988:4033-4073 1973:1-83
2. Harris WH, Dudley HR. Barry RJ. The natural history of fibrous dysplasia. 16. Gibson MJ, Middlemiss JH. Fibrous dysplasia of bone. Br J Radio!
J Bone Joint Surg (Am] 1962:44-A:207--233 1971:44:1-13
Downloaded from www.ajronline.org by 66.96.225.122 on 03/07/21 from IP address 66.96.225.122. Copyright ARRS. For personal use only; all rights reserved

3. Lichtenstein L, Jaffe HL. Fibrous dysplasia of bone. Arch Patho! Lab Med 17. Resnick D. Additional congenital or heritable anomalies and syndromes.
1942:3:777-816 In: Resnick D, Niwayama G, eds. Diagnosis of bone and joint disorders,
4. Wilner D. Fibrous dysplasia of bone. In: Wilner D, ed. Radiology of bone 2nd ed. Philadelphia: Saunders, 1988:3540-3595
tumors and allied disorders. Philadelphia: Saunders, 1982:1443-1581 18. Sowa DT, Doriman HD. Unusual localization of adamantinoma of long
5. Campbell CJ, Hawk T. A variant of fibrous dysplasia (osteofibrous dyspla- bones. J Bone Joint Surg (Am] i986:68-A:293-296
sia). J Bone Joint Surg (Am] i982;64-A:231-236 19. Campanacci M, Laus M. Osteofibrous dysplasia of the tibia and fibula. J
6. Nakashima Y, Yamamuro T, Fujiwara Y, Kotoura Y, Mori E, Hamashima Bone Joint Surg (Am] i981;63-A:367-375
Y. Osteofibrous dysplasia (ossifying flbroma of long bones). Cancer 20. Waldron CA. Fibro-osseous lesions of the jaws. J Oral Maxll!ofac Surg
1983:52:909-914 1985:43:249-262
7. Kempson RL. Ossifying fibroma of the long bones. Arch Patho! Lab Med 21 . Eversole LR, Leider AS, Nelson K. Ossifying fibroma: a clinicopathologic
1966:82:218-233 study of sixtyfour cases. Oral Surg Oral Med Oral Patho! 1985:60:
8. Mm-a JM. Adamantinoma and osteofibrous dysplasia. In: Mirra JM, ed. 505-511
Bone tumors. Philadelphia: Lea & Febiger, 1989: 1203-1 233 22. Kransdorf MJ, Moser RP, Gilkey FW. From the archives of the AFIP:
9. Moon NF, Med H. Adamantinoma of the appendicular skeleton-updated. fibrous dysplasia.
RadioGraphics 1990:10:519-537
C!in Orthop 1986:204:215-237 23. Schneider H, Enderle R. Zur Differentialdiagnose eines metastasierenden
1 0. Resnick D, Kyriakos M, Greenway GD. Tumors and tumor-like lesions of Adamantinoms der Tibia und Fibula. Arch Orthop Trauma Surg
bone: imaging and pathology of specific lesions. In: Resnick D, Niwayama i979;94: 143-1 49
G, eds. Diagnosis of bone and joint disorders, 2nd ed. Philadelphia: 24. Cohn BT, Brahms MA, Froimson A. Metastasis of adamantinoma sixteen
Saunders, 1988:3616-3889 years after knee disarticulation. J Bone Joint Surg (Am] 1986:68-A:
1 1 . Wilner D. Adamantinoma of long bones. In: Wilner D, ed. Radiology of 772-776
bone tumors and allied disorders. Philadelphia: Saunders, 1982: 25. Daffner RH, Kirks DR, Gehweiler JA, Heaston DK. Computed tomography
2387-2410 of fibrous dysplasia. AJR 1982:139:943-948
1 2. Weiss SW, Dorfman HD. Adamantinoma of long bone: an analysis of nine 26. Dominguez R, Saucedo J, Fenstermacher M. MRI findings in osteofibrous
new cases with emphasis on metastasing lesions and fibrous dysplasia- dysplasia. Magn Reson Imaging 1989:7:567-570
like changes. Hum Patho! i977;8:141-153 27. Zeanah WR, Hudson TM, Springfield DS. Computed tomography of ossi-
1 3. Schajowicz F, Santini-Araujo E. Adamantinoma of the tibia masked by tying fibroma of the tibia. J Comput Assist Tomogr i983:7(4):688-691
fibrous dysplasia. C!in Orthop 1989:238:294-301 28. Utz JA, Kransdorf MJ, Jelinek JS, Moser RP, Berrey BH. MR appearance
1 4. Mori H, Yamamoto 5, Hiramatsu K, Miura T, Moon NF. Adamantinoma of of fibrous dysplasia. J Comput Assist Tomogr 1989:13(5):845-851