There has been considerable controversy concerning the nosology of benign fibro-osseous lesions, due in part to the varied histomorphologic patterns of stroma and bone in these lesions and the fact that similar or even identical microscopic features can be shared among two or more different entities. We have classified these diseases according to pathogenetic mechanisms yet the most practical, utilitarian approach is to consider BFOL in the context of combined clinical, radiologic and pathologic characteristics (Table ). It is axiomatic that a definitive diagnosis can rarely be rendered on the basis of microscopic features alone; although there are subtle changes that may lead the pathologist to favor one entitiy over others. Age sex and race are variables that carry weight in the formulation of a final diagnosis. The lesional boundaries as seen in radiographic imaging studies are also of utmost importance since some BFOL are associated with teeth; some show diffuse ill-defined borders while others are well delineated. And finally, there are indeed certain histologic patterns that may be restricted to only one disease process.
Clinical, radiographic and microscopic parameters that distinguish among the benign fibro-osseous lesions
The diagnostic criteria that are most controversial, at least as gleaned from the earlier literature are those that distinguish (1) fibrous dysplasia from ossifying fibromas and chronic sclerosing osteomyelitis, (2) florid cementoosseous dysplasia from diffuse sclerosing osteomyelitis and gigantiform cementoma and (3) focal cementoosseous dysplasia from ossifying fibromas.
The microscopic features outlined by Reed [199
] Zimmerman et al. [200
] and vanHorn et al. [201
] for fibrous dysplasia refer to the presence of “metaplastic” bone, characterized by trabeculae that are morphed from simple mesenchymal connective tissue in the absence of osteoblasts and also represented a maturation arrest with deposition of woven bone only. In the craniofacial bones these criteria do not apply. Documented followup with periodic rebiopsy discloses that early formative lesions of fibrous dysplasia lay down woven bone trabeculae with extentive osteoblastic rimming of trabeculae and marked stromal hypercellularity whereas “older, mature” lesions show both woven and lamellar bone with trabeculae that lack juxtaposed osteoblasts. The socalled “chinese or Hebrew” letter figures remain a classis feature, yet these atypical trabeculae may be seen in other BFOL. The major criteria for the diagnosis of fibrous dysplasia in the craniofacial complex are (1) a fibroosseous histology with trabecular rather than cemental bone, (2) onset in childhood or teenage years, (3) unilateral distribution with polyostotic forms being uncommon, (4) a ground glass radiographic pattern in the maxilla and facial bones or a mottled mixed radiolucent/radioopaque pattern in the mandible, all with poorly defined lesional boundaries.
While fibrous dysplasia and ossifying fibromas may share microscopic features, the clinicoradiologic differences are now widely accepted. Recently, Toyosawa et al. [164
] demonstrated that the two entities can be distinguished from one another on the basis of molecular detection of Gs-alpha mutations in fibrous dysplasia of the jaws while ossifying fibromas are found lacking.
The confusion between fibrous dysplasia and chronic sclerosing osteomyelitis is common. Both show osseous expansion, occur in youngsters and manifest a ground glass radiographic pattern, clinico radiologic findings that may be almost identical. The major differentiating factors include symptomatology and histology. FD is painless while CSO patients complain of episodic dull pain and it is always the mandible that is affected in CSO whereas the maxilla is more often involved than the mandible in FD. Histologically, FD never shows islands of dense cortical bone while CSO does along with concomitant fibro-osseous features. Lastly, bone biopsy culture will often reveal low virulence anaerobic bacteria in CSO and proliferative periostitis is a frequent accompaniment.
The second areas of controversy involve confusion over diagnostic features for florid cementoosseous dysplasia, diffuse sclerosing osteomyelitis and gigantiform cementoma. This confusion began with a paper by Shafer [202
], in which a series of cases were published under the rubric of CSO when in fact, it was later realized that the lesions he was referring to were, in reality, classic examples of florid cemento-osseous dysplasia, an entity that was clearly delineated 19 years after Shafer’s publication, by Melrose et al. [111
]. In defense of Shafer’s original interpretation, many cases of FlCOD become secondarily infected, fistulated and sequestrated, signs of bone fide
osteomyelitis. Yet later in the course of events, Scandanavian papers appeared that presented cases of sclerosing bone lesions in children that could be attributed to low virulence bacterial organisms, many of which are anaerobic. [130
]. The characteristic microscopic features outlined by Melrose et al. [111
] and by Su et al. [105
] clearly define an entity that differs from sclerosing osteomyelitis with cementicle formation, “ginger root” trabeculae and large osteosclerotic islands. Furthermore, FlCOD is more often reported among African descent subjects, tends to be multifocal and at surgery, empty bone cavities are encountered.
Another point of controversy involves confusing FlCOD with gigantiform cementoma. There are many papers in the older literature that report cases of gigantiform cementoma that are in reality instances of FlCOD. Yet other reports have been referred to as multiple enostoses. Gigantiform cementoma as originally reported by Agazzi and Beloni [203
] and more recently by Young et al. [185
] and Abdulsayed et al. [186
] exhibits the classic clinicoradiologic/pathologic features that have already been detailed. One caveat, in familial cases of gigantiform cementoma, there are family members who present with radiologic changes that are mild (incomplete penetrance and/or espressivity) and do in fact resemble FlCOD.
Lastly, focal cementoosseous dysplasia is often confused with ossifying fibromas. Importantly, the former is an endosseous nonneoplastic process that occurs around the roots of mandibular teeth and fails to expand bone. Alternatively, ossifying fibromas are potentially aggressive lesions that cause cortical expansion and often cause divergence of contiguous teeth. Both lesions may show similar histologic features with trabecular bone and cementifying areas. Older lesions of FCOD may show dense corticated bone islands, a finding that is not present in ossifying fibromas.
It is noteworthy that molecular studies are finally being performed and well defined DNA lesions are being uncovered in osteitis deformans, fibrous dysplasia and ossifying fibromas. Surely BFOLs of the jaws with familial or ethnic features such as gigantiform cementoma, florid cementoosseous dysplaisia, periapical cemental dysplasia and focal cementoosseous dysplasia will be subjected to rigorous analysis in the search for an undercurrent of molecular disease.