You are in: eMedicine Specialties > Orthopedic Surgery > NEOPLASMS Osteofibrous DysplasiaArticle Last Updated: Feb 16, 2005AUTHOR AND EDITOR INFORMATIONSection 1 of 12
  Author: Robert Mervyn Letts, MD, FRCS(C), FACSC, Chief, Department of Surgery, Division of Pediatric Orthopedics, Children's Hospital of Eastern Ontario, University of Ottawa Coauthor(s): Darin Davidson, MD, Resident, Department of Orthopedics, University of British Columbia Editors: Lynn A Crosby, MD, FACS, Chief of Shoulder Division, Professor, Department of Orthopedic Surgery, Wright State University School of Medicine; Francisco Talavera, PharmD, PhD, Senior Pharmacy Editor, eMedicine; Sean P Scully, MD, PhD, Professor, Department of Orthopedics, University of Miami; Dinesh Patel, MD, FACS, Associate Clinical Professor of Orthopedic Surgery, Harvard Medical School; Chief of Arthroscopic Surgery, Department of Orthopedic Surgery, Massachusetts General Hospital; Harris Gellman, MD, Consulting Surgeon, Broward Hand Center, Voluntary Clinical Professor of Orthopedic Surgery and Plastic Surgery, Departments of Orthopedic Surgery and Surgery, University of Miami School of Medicine Author and Editor Disclosure Synonyms and related keywords: ossifying fibroma, congenital fibrous dysplasia, congenital fibrous defect of the tibia, intracortical fibrous dysplasia, infantile pseudarthrosis of the tibia, localized osteitis fibrosa, congenital kyphoscoliotic tibia, congenital pseudarthrosis, Campanacci syndrome, Campanacci's syndrome INTRODUCTIONSection 2 of 12
  Osteofibrous dysplasia is a rare condition that most commonly affects the tibia (see Image 1). Osteofibrous dysplasia is a nonneoplastic lesion that is frequently asymptomatic and has an unknown etiology. Most lesions of osteofibrous dysplasia affect the cortex of the tibia, predominantly the middle third of the diaphysis. The cortex often is expanded and thinned with multiple radiolucencies mixed with intervening areas of sclerosis. The second most common site of involvement is the fibula. The literature contains numerous case reports and small series of osteofibrous dysplasia affecting the tibia. Sweet et al reported 30 cases. The tibia was affected in each case, and the ipsilateral fibula was involved in 5 cases (17%). In a study of 10 children, the lesion was found to affect the tibia in each case, with one case (10%) having ipsilateral fibular involvement. Campanacci and Laus reported 35 cases. The tibia was affected in each case, with ipsilateral involvement of the fibula in 4 cases (11%). They further reported that 22 of 35 lesions (63%) affected the middle third of the tibial diaphysis. Ishida et al encountered 11 of 12 lesions (92%) in the tibia, with one lesion in the ulna. They reported that most tibial lesions affected the proximal diaphysis. Bilateral involvement is rare. However, in a study of 5 children reported by Ozaki et al, one child presented with bilateral lesions of both ulnae and tibiae. In the remaining 4 children, the tibia was affected in each, with one having ipsilateral fibular involvement. Wang et al reported one case of a lesion affecting the radius, and Schlitter reported the case of a lesion in the humerus. Osteofibrous dysplasia of the mandible, which occurs exclusively in adults, commonly is referred to as ossifying fibroma. History of the ProcedureFrangenheim first described the lesion in 1921 and reported it as a congenital osteitis fibrosa. Subsequently, Kempson reported 2 cases affecting the tibia of young children and named the lesion ossifying fibroma. In 1981, Campanacci and Laus studied 35 cases and coined the term osteofibrous dysplasia of the tibia and fibula. They proposed this term to replace the use of ossifying fibroma because of the supposed congenital origin of the condition, the histologic resemblance to fibrous dysplasia, and the apparent exclusive involvement of the tibia and fibula. It is occasionally referred to as Campanacci syndrome. FrequencyOsteofibrous dysplasia usually is first diagnosed in children younger than 10 years, with a peak incidence occurring in children aged 1-5 years. Several occurrences in newborns also have been reported. Reports of adults being diagnosed with de novo osteofibrous dysplasia have been reported, the oldest patient being diagnosed at age 39 years, as reported by Sweet et al. The age range at the time of diagnosis has been variable in the literature. Studies reported by Sweet et al and Ishida et al described an average age older than 10 years. In contrast, Komiya and Inoue, Ozaki et al, and Campanacci and Laus each reported an average age younger than 10 years. No significant sex preponderance consistently has been reported in association with osteofibrous dysplasia, although several studies have reported a slight male predilection. Sweet et al reported a series of 30 patients, 16 of whom were male. Campanacci and Laus noted that 21 of 35 patients (60%) in their series were male. This represents the largest reported sex preponderance. In a series of 80 patients, Park et al failed to report a sex preponderance, as there were 38 males and 42 females. EtiologyThe etiology of osteofibrous dysplasia, as well as the cell of origin, is unknown. Only one description of familial osteofibrous dysplasia has been reported by Hunter and Jarvis. Osteofibrous dysplasia has been postulated to arise from a fibrovascular abnormality. Johnson proposed a relationship between osteofibrous dysplasia and adamantinoma on the basis of a common causative factor, namely a fibrovascular defect. According to this theory, osteofibrous dysplasia results from an abnormality in the haversian canals, whereas adamantinoma develops secondary to a defect of intramedullary vasculature. Komiya and Inoue reported similar findings and suggested a deficiency in blood flow within the periosteum as the etiologic factor in osteofibrous dysplasia. Bridge et al investigated the cytogenetic aspects of osteofibrous dysplasia and reported trisomy 12 in 2 distinct specimens from the lesion of an 11-year-old boy and trisomy 7, 8, and 22 in another boy. Studies of adamantinoma have revealed trisomy 7 and 12, possibly suggesting a relationship between osteofibrous dysplasia and adamantinoma. Sherman et al reported the coexistence of adamantinoma and osteofibrous dysplasia in the same patient, adding additional evidence of the relationship between these 2 lesions. Several additional abnormalities have been discovered within the lesion. Consequently, these chromosomal anomalies may not be pathogenetic. On the other hand, Sakamoto et al have shown mutations at the Arg 201 codon in persons with fibrous dysplasia but not in persons with osteofibrous dysplasia, suggesting a different pathogenesis between the 2 lesions. ClinicalClassically, this lesion has been described as painless, with a localized firm swelling of the tibia as the presenting complaint. The tibia is frequently bowed either anteriorly or anterolaterally. In a study of 80 patients reported by Park et al, 25% of patients complained of pain, 12.5% of patients had a pathologic fracture, 8.8% of patients presented because of tibial swelling, and 6.2% of patients presented secondary to deformity. In another large study of 30 cases, Sweet et al reported 18 patients (60%) with complaints of pain, 13 patients (43%) with swelling, and 4 patients (13%) with deformity. One lesion was an incidental finding. Komiya and Inoue reported similar presenting complaints in a series of 10 cases. Ishida et al reported the duration of symptoms in 11 of 12 patients to range from 2 months to 5 years, with an average of 14 months; one lesion was asymptomatic. Of 3 newborns with osteofibrous dysplasia of the tibia, 2 had swelling and 1 had pathologic fracture. Differential diagnosis The predominant differential diagnosis of osteofibrous dysplasia is monostotic fibrous dysplasia, nonossifying fibroma, and adamantinoma. Fibrous dysplasia can be differentiated on the basis of several characteristic factors. Generally, it occurs in patients older than 10 years, more commonly affects the femur and ribs, and does not resolve spontaneously. Radiographically, fibrous dysplasia appears as an intramedullary lesion with a ground glass appearance. On histologic examination, fibrous dysplasia is not bordered by active osteoblasts and is cytokeratin-negative. Cytogenetically, fibrous dysplasia is related to anomalies affecting chromosomes 3 and 5. Sakamoto et al recognized that immunoreactivity for osteonectin in bone matrix was more commonly recognized in osteofibrous dysplasia. Nonossifying fibroma can be distinguished from osteofibrous dysplasia by several typical features. Nonossifying fibroma predominantly is a metaphyseal lesion. Histologically, it is characterized by a storiform pattern of spindle cells with scattered multinucleated giant cells with no bordering by active osteoblasts, and it is cytokeratin-negative. More challenging is the distinction between osteofibrous dysplasia and adamantinoma. Accurate differentiation between these 2 lesions is essential for correct diagnosis and appropriate treatment. Adamantinoma has a similar predilection for the cortex of long bones, in particular the tibia, and may have similar radiologic and histologic findings as those of osteofibrous dysplasia. Distinction between these 2 lesions can be made on the basis that adamantinoma is associated with soft tissue extension, intramedullary involvement, periosteal reaction in the absence of pathologic fracture, and the histologic finding of hyperchromatic epithelial islands. Adamantinoma typically manifests with a larger more painful lesion, and it is usually found in patients older than 10 years. However, as suggested by Kuruvilla and Steiner, it is most likely that osteofibrous dysplasia is part of the morphologic spectrum of adamantinoma. Recently, Kanamori et al, using molecular cytogenic analysis, found recurrent extra copies of chromosomes 7, 8, 12, 19, and 21 in adamantinoma. Observation of these aneuploidies may be useful in differentiating adamantinoma from osteofibrous dysplasia. INDICATIONSSection 3 of 12
Due to the high recurrence rate, numerous studies have advocated nonoperative treatment of the lesion until after skeletal maturity is reached, at which time marginal resection and bone grafting may be performed without increased risk of recurrence. For patients of any age, surgical correction of associated deformities may be required. Surgery may be indicated if the lesion is aggressive, or if the patient experiences multiple pathological fractures. Resection of large portions of the lesion usually is not necessary and only increases susceptibility to recurrent fractures. RELEVANT ANATOMYSection 4 of 12
The tibia is a tubular long bone with a triangular shape in cross section. The bone is surrounded by 4 fascial compartments. The anteromedial surface lies subcutaneously and therefore has no soft tissue protection. The primary center of ossification appears at 7 weeks of gestation. The proximal ossific nucleus appears soon after birth and fuses with the metaphysis at approximately age 16 years. The distal ossific nucleus appears at age 2 years and fuses at age 15 years. In some, separate centers of ossification exist for the medial malleolus and tibial tubercle. The vascular supply to the tibia is provided predominantly by the posterior tibial artery, from which the nutrient artery enters the tibia at the origin of the soleus muscle along the oblique line of the tibia. The nutrient artery of the tibia has 3 ascending branches and one descending branch. The distal aspect of the tibia is supplied by periosteal anastomoses that enter the bone adjacent to the ankle joint. CONTRAINDICATIONSSection 5 of 12
Though nonoperative management is recommended in patients who are skeletally immature, there are no absolute contraindications to surgical intervention in children, with the exception of any underlying medical or anesthetic issues. Operative management is not recommended in patients who are skeletally immature because of the high recurrence rate following resection and curettage and because of the predisposition to fracturing after the bone has been weakened by biopsy. Pathologic fracture does not necessarily require surgical management since plaster immobilization frequently results in good fracture healing. WORKUPSection 6 of 12
Lab Studies
Imaging Studies
Diagnostic Procedures
Histologic FindingsDespite the characteristic radiographic appearance, Wang et al suggested that diagnosis should be based upon biopsy and pathologic examination (see Image 3). At the time of surgery, inspection reveals an intact periosteum. The cortex is thinned and may be perforated. The lesion itself is comprised of soft granular tissue that is whitish yellow–in color. Histologic characteristics of osteofibrous dysplasia have been well described in the literature. The overall appearance is that of zonal architecture. The lesional tissue is fibrous at its center with immature woven bone trabeculae. Vascular channels have been described within the lesion (see Image 4). At the periphery, a prominent border of active osteoblasts rimming the bony trabeculae exists. The presence of such a border is a differentiating factor between osteofibrous dysplasia and fibrous dysplasia, in which there is no border of active osteoblasts (see Image 5). As examination proceeds from the center of the lesion to the periphery, the bone trabeculae become larger and more lamellar in appearance. Fibroblasts involved in the lesion have been noted to be well-differentiated. Several studies have reported that cytokeratin-positive elements have been elucidated with immunohistochemical staining. Occasional hemorrhagic zones, cysts, or foci of cartilaginous differentiation have been reported. Multinucleated giant cells also have been encountered. TREATMENTSection 7 of 12
Medical therapyNonoperative treatment often is recommended until skeletal maturity is reached. Recurrent pathologic fractures may be an ongoing problem in some active children. Using a tibial brace similar to those used for congenital pseudarthrosis of the tibia may minimize recurrent pathologic fractures. A lace-up leather support from just below the knee to the ankle may be used. Fractures usually are undisplaced and can be treated in a walking patellar tendon-bearing cast. Healing is slower than normal but will occur, although refracture is not uncommon. If pathologic fracture occurs, treatment with plaster immobilization is sufficient for fracture healing. Surgical therapyA characteristic feature of this lesion is a high recurrence rate following resection and curettage. Due to the high recurrence rate, numerous studies have advocated nonoperative treatment of the lesion until after skeletal maturity is reached, at which time marginal resection and bone grafting may be performed without increased risk of recurrence. For patients of any age, surgical correction of associated deformities may be required. Campanacci and Laus suggested that wide resection with extensive bone grafting be performed in children who are skeletally immature if the lesion is aggressive with resulting marked expansion and bone destruction or multiple pathologic fractures. Intramedullary prophylactic rodding of the tibia also may be an option in children who frequently present with fractures; this approach is similar to the approach utilized in osteogenesis imperfecta. Resection of large portions of the lesion usually is not necessary and only increases susceptibility to recurrent fractures. COMPLICATIONSSection 8 of 12
A characteristic feature of this lesion is a high recurrence rate following resection and curettage. The recurrence rate has been reported at 64-100%. Goergen et al reported multiple recurrences in a 3-year-old boy and a 6-month-old boy following attempts at resection. Wang et al also reported multiple recurrences following surgical intervention. Campanacci and Laus indicated that recurrence did not occur in patients older than 10 years. Malignant transformation of the lesion, although very rare, has been reported. Ben Arush et al reported the only case of malignant transformation of osteofibrous dysplasia to synovial sarcoma. They described the course of a 14-year-old boy who was diagnosed with osteofibrous dysplasia of the tibia at age 4 years. Subsequently, at age 14 years, he presented with synovial sarcoma of the peroneal muscles of the ipsilateral leg. At the time of diagnosis, CT scan confirmed multiple pulmonary metastases. Malignant transformation to soft tissue sarcoma previously has been reported in cases of fibrous dysplasia, more commonly in the polyostotic variation. However, the case reported by Ben Arush et al is the only report of sarcomatous degeneration of osteofibrous dysplasia. OUTCOME AND PROGNOSISSection 9 of 12
The natural history of osteofibrous dysplasia is unpredictable. Growth rate of this lesion can vary from slow to rapid, and spontaneous resolution is possible. Campanacci and Laus reported 3 common clinical courses. First, moderate progression, particularly during the first 5-10 years of life, may occur. Second, a course of aggressive growth with resulting marked deformity may occur; and third, a course of spontaneous resolution is possible. Most commonly, there is continued growth of the lesion until skeletal maturity is reached, with the most rapid period of growth occurring in patients younger than 10 years. In most cases, the first course of moderate progression occurs early in life, followed by gradual improvement once skeletal maturity is attained. FUTURE AND CONTROVERSIESSection 10 of 12
Osteofibrous dysplasia and adamantinoma have similar clinical presentations, as well as radiologic and pathologic findings. Although adamantinoma can sometimes have the appearance of a low-grade osteogenic sarcoma, osteofibrous dysplasia does not exhibit histologic characteristics of malignancy. There may be a gradation in pathological appearance between benign osteofibrous dysplasia, benign adamantinoma, and the malignant appearance of more aggressive adamantinoma, which usually is encountered in adults. In such instances, osteoid production with cellular mitoses may give the appearance of an osteogenic sarcoma and indeed may progress to frank malignancy. Because the clinical course and radiologic appearance of osteofibrous dysplasia is diagnostic in children, biopsy is seldom indicated and should be avoided, if possible. In patients presenting at skeletal maturity, where the incidence of adamantinoma is higher, biopsy of the mid portion of the lesion may be necessary for diagnosis. Complete resection of the entire lesion of osteofibrous dysplasia is neither recommended nor necessary. Several studies have investigated a possible relationship between adamantinoma and osteofibrous dysplasia. Dockerty and Meyerding first reported a relationship between benign fibro-osseous lesions and adamantinoma. Markel was the first to thoroughly investigate this relationship. Subsequently, 3 cases of tibial adamantinoma that mimicked osteofibrous dysplasia were reported, 2 of which occurred in children younger than 10 years. Several studies have proposed that either osteofibrous dysplasia represents a benign form of adamantinoma or that it is the result of a resolved adamantinoma. Czerniak et al described a lesion, which they termed differentiated adamantinoma, as an intracortical lesion with pathologic findings similar to those of osteofibrous dysplasia; furthermore, they described differentiated adamantinoma as affecting individuals younger than those affected with classic adamantinoma. Czerniak et al and Springfield et al reported that differentiated or osteofibrous dysplasialike adamantinoma can progress to adamantinoma. Furthermore, it has been suggested that these lesions represent a continuum from osteofibrous dysplasia to adamantinoma, through osteofibrous dysplasialike adamantinoma as an intermediate. Hazelbag et al reported several findings to support this relationship. First, they noted continuity from epithelial cells in osteofibrous dysplasia to primary epithelioid tumour, as in adamantinoma. Second, the mean age at diagnosis of osteofibrous dysplasia and osteofibrous dysplasialike adamantinoma is younger than the age at diagnosis of adamantinoma. Third, there are similar radiographic findings. Fourth, 2 patients in their study demonstrated progression from osteofibrous dysplasia to adamantinoma at the time of local recurrence. With respect to this relationship between osteofibrous dysplasia and adamantinoma, it has been suggested that the sequence may occur in reverse, such that an adamantinoma may regress to osteofibrous dysplasia. In contrast, Springfield et al refuted this claim and indicated that such regression was not likely. Study results implicating a relationship are in conflict with the investigation by Park et al. Park et al reported no progression from osteofibrous dysplasia to adamantinoma and advocated that osteofibrous dysplasia is distinct from adamantinoma. They did, however, suggest that osteofibrous dysplasia might be related to fibrous dysplasia because 2 cases in their series transformed from osteofibrous dysplasia to monostotic fibrous dysplasia. There have been several reports of pathologic examination of adamantinoma that have demonstrated areas of the lesion similar in appearance to osteofibrous dysplasia. This finding leads to the potential for misdiagnosis secondary to inadequate biopsy. The potential for misdiagnosis on the basis of inadequate biopsy sample may explain reports of progression of osteofibrous dysplasia to adamantinoma, a transformation refuted by Park et al. Springfield et al suggested that histologic diagnosis of osteofibrous dysplasia should be regarded with caution due to the possibility of a nonrepresentative biopsy specimen. Hazelbag et al advocated biopsy of the center of the lesion in order to prevent such an error, while Sweet et al suggested examination of the entire specimen in an attempt to identify areas consistent with adamantinoma. MULTIMEDIASection 11 of 12
REFERENCESSection 12 of 12
Osteofibrous Dysplasia excerpt Article Last Updated: Feb 16, 2005 | ||||||||||||||||||||||||||||||||||||||||
