Osteofibrous Dysplasia

Osteofibrous dysplasia (OFD) is a rare benign fibro-osseous lesion that affects the anterior tibial cortex in 85-90% of cases, with its peak in the first decade of life. Radiographs show an intracortical lucent lesion with sclerotic borders causing anterior tibial bowing. Most cases are managed conservatively with observation, as spontaneous resolution commonly occurs after skeletal maturity, and surgery is reserved for pathological fracture or progressive deformity.

Who. Peak incidence is at 4-8 years, with a mean age of 6, and 90% of lesions are diagnosed before the age of 10. OFD is rare in adults, where it usually represents progression from childhood. There is no gender predilection (the male to female ratio is equal) and no known geographic predilection. Referral series report older mean ages, 13.4 years in Sweet's AFIP series and 13.5 years in Dala-Ali's, so the first-decade picture describes the typical child rather than every patient.

Where. The tibia dominates:

• Tibia - 85-90%, in the anterior cortex of the diaphysis or metaphysis
• Fibula - 8-10%
• Radius and ulna - under 2%, extremely rare
• Other bones - case reports only

How rare. OFD is extremely rare: fewer than 200 cases have been reported in the literature, and it represents less than 0.2% of benign bone tumours.

Proposed mechanisms. Four have been put forward:

1. Developmental - a defect in cortical bone maturation
2. Reactive - a response to microtrauma in growing bone
3. Mesenchymal - abnormal differentiation of osteoblastic cells
4. Genetic - no consistent chromosomal abnormality has been identified

Separate from fibrous dysplasia. OFD lacks the GNAS (Gsα) Arg201 mutation present in fibrous dysplasia, and the molecular distinction confirms that the two are separate entities.

A debated spectrum with adamantinoma. This is a genuine controversy. Three lesions are placed along it:

• OFD - benign
• OFD-like (differentiated) adamantinoma - scattered keratin-positive epithelial cells; indolent, paediatric
• Classic adamantinoma - frank epithelial islands; low-grade malignant, with metastatic potential

Does OFD become adamantinoma? Direct malignant transformation of pure OFD has not been demonstrated in the large pooled paediatric series, and most reported "transformations" reflect under-sampling of an OFD-like adamantinoma at first biopsy. Malignant transformation is extremely rare (less than 1%), known from case reports only, and usually represents a misdiagnosis of adamantinoma, so it requires careful histological review. Fewer than 5% of patients develop adamantinoma, which is likely to have been pre-existing.

Macroscopic. The lesion is well-circumscribed fibrous tissue within the cortex, sharply demarcated from the surrounding bone. Osteoid formation gives it a gritty texture, and there is no cystic change or haemorrhage.

Microscopic. A fibrous stroma of bland, uniform spindle cells with moderate cellularity, no cellular atypia and rare to absent mitoses. Irregular osteoid seams surround the fibrous tissue.

The zoning phenomenon is the defining microscopic feature: a centre-to-periphery maturation gradient, in place of the haphazard pattern of fibrous dysplasia.

• Centre - the most immature and cellular zone, a loose, mildly myxoid fibrous stroma containing irregular woven-bone trabeculae
• Periphery - bone that becomes progressively more mature and lamellar, the outer trabeculae remodelling and blending into the surrounding host cortex
• Throughout - woven-bone trabeculae rimmed by a prominent layer of plump, active osteoblasts. This osteoblastic rimming is the single most useful discriminator

How zoning separates the fibro-osseous lesions. Fibrous dysplasia has curvilinear "alphabet-soup" or "Chinese-character" woven bone that arises directly from the stroma, without osteoblastic rimming and without a maturation gradient; its metaplastic bone is uniform and haphazard throughout. Adamantinoma and OFD-like adamantinoma share the fibro-osseous, osteoblast-rimmed background of OFD, so zonation alone does not exclude them. Zoning says "fibro-osseous, OFD family"; the epithelial islands say "adamantinoma".

View larger

Immunohistochemistry. The panel in OFD:

• Cytokeratin (AE1/AE3) - scattered positive stromal cells
• Epithelial membrane antigen (EMA) - negative
• S100 - variable positivity in the fibrous component
• CD34 - may be positive in stromal cells
• Ki-67 index - low, under 5%

The keratin trap. OFD is not simply "cytokeratin negative". Sweet's AFIP series found isolated cytokeratin-positive stromal cells in 93% of OFD lesions and none in any fibro-osseous control, while hyperchromatic epithelial islands were absent in all 30 OFD cases. Keratin therefore separates OFD from fibrous dysplasia, which is uniformly negative, and not from adamantinoma: what defines adamantinoma is the presence of the islands.

Sampling. Small biopsies can miss both the maturation gradient and rare epithelial islands, which is why radiology-pathology correlation and generous sampling matter.

The classic case. A 6-year-old with painless anterior tibial swelling, or a lesion found incidentally on radiographs taken during a trauma work-up. The ways OFD presents:

• Asymptomatic - 40-50%, an incidental radiographic finding
• Painless swelling - 30-40%, an anterior tibial prominence
• Activity-related discomfort - 20-30%, mild aching with exercise
• Pathological fracture - 10-15%, with acute pain and deformity

Examination. There is a firm, non-tender mass over the anterior tibial cortex, and the overlying skin is normal, without warmth or erythema. An anterior tibial bow may be present with a long-standing lesion. Neurovascular status is always normal, and the adjacent joints have a full and painless range of motion.

Deformity. Anterior tibial bowing occurs in 15-20% of cases and progresses with growth while the lesion is active. It is typically mild, under 15° of angulation, with minimal functional impact in most cases, and rarely requires corrective osteotomy; progressive deformity requiring intervention occurs in 5-10%.

Natural history. Activity and histology change together as the child grows:

• Childhood (active phase) - slow expansion with growth; high cellularity and prominent osteoid production
• Adolescence (quiescent phase) - stabilisation in most cases; reduced cellularity and increased bone formation
• After skeletal maturity (resolution phase) - spontaneous resolution is common; fibrous tissue replacement with mature bone
• Adult follow-up - residual cortical thickening, with no active lesion

View larger

View larger

Radiographs come first, alongside clinical examination. The features of a classic lesion:

• An eccentric intracortical lesion of the anterior tibial cortex
• A well-defined lucency with sclerotic margins
• A geographic pattern of bone destruction, Lodwick grade IA
• Thinning, expansion and scalloping of the inner cortex
• Minimal or absent periosteal reaction in uncomplicated cases

View larger

CT delineates cortical involvement precisely, shows no calcification or ossification pattern in the matrix, and confirms the absence of an extraosseous component. 3D reconstruction is useful for surgical planning if needed.

View larger

MRI is done to exclude a soft-tissue mass, which is the adamantinoma concern. The signal characteristics of OFD:

• T1 - low to intermediate
• T2 - variable, depending on the ratio of fibrous to osseous tissue
• STIR - high signal in active lesions
• Contrast - moderate enhancement in active lesions

MRI showing no soft-tissue component strongly supports the diagnosis of OFD. If there is any soft-tissue mass, you must suspect adamantinoma and proceed to biopsy.

View larger

Nuclear medicine. A bone scan shows increased uptake in active lesions but has limited clinical utility and is not routinely required. PET is not indicated for a benign lesion.

Adamantinoma. It is defined histologically by hyperchromatic epithelial islands; OFD shows only scattered cytokeratin-positive stromal cells, so keratin-positive cells alone do not make the diagnosis. Adamantinoma has malignant potential and requires wide resection.

View larger

View larger

View larger

View larger

Fibrous dysplasia. Beyond its medullary, ground-glass appearance in the table above, it is typically seen in adolescents and young adults, its woven bone lacks osteoblastic rimming, and unlike OFD it does not resolve spontaneously.

View larger

View larger

The remaining differentials:

• Ossifying fibroma - primarily mandible and maxilla, rare in long bones; well-defined, radiolucent to radiopaque; spherical ossicles (psammomatoid bodies) on histology; young adults
• Stress fracture - activity-related pain of acute onset; a linear lucency with periosteal reaction; marrow oedema and a cortical fracture line on MRI; heals within 6-8 weeks
• Non-ossifying fibroma - metaphyseal, eccentric and medullary; sclerotic scalloped margins with cortical thinning; adolescents aged 10-15; storiform pattern, foam cells and haemosiderin on histology

(For the full behaviour and molecular basis of the two closest mimics, see the dedicated Fibrous Dysplasia, Adamantinoma and Non-Ossifying Fibroma topics; only their OFD-discriminating features are summarised here.)

The principle. OFD is benign and self-limiting, so consider observation until skeletal maturity in most cases. Surgery targets symptomatic deformity, not the lesion.

The decision. Age, radiology and what the lesion is doing choose the pathway:

1. Classic presentation (age under 10, typical radiographs) - observe with serial imaging; no biopsy.
2. Atypical features (age over 10, soft-tissue mass, aggressive appearance) - MRI to assess the soft-tissue component, and biopsy is mandatory to exclude adamantinoma. Consider en bloc resection if there are adamantinoma-like features.
3. Pathological fracture - conservative fracture management first. Consider curettage if there is non-union after 6 months, with bone grafting to augment healing.
4. Progressive deformity - observe until skeletal maturity if mild (under 15°); corrective osteotomy after maturity if significant (over 20°). Address the lesion separately if it is still active.

ObservationBracingSurgery

Who. Children under 15 with an asymptomatic lesion and a stable radiographic appearance.

Surveillance. Radiographs on a schedule that lengthens with time:

• First year - radiographs every 3-6 months
• Years 2-5 - every 6-12 months
• Until skeletal maturity - annual radiographs
• After maturity - discharge if stable

Activity. An asymptomatic lesion needs no specific restriction. Avoid high-impact sport if cortical thinning is severe (over 50%), and return to activity once a pathological fracture has healed.

What to tell the family. With observation, 40-60% resolve spontaneously after skeletal maturity and 30-40% stabilise without further progression. Progression occurs in 10-20%, and these may require intervention.

Rationale. The lesion weakens the anterior tibial cortex, so the two ways an observed lesion can misbehave are a pathological fracture and progressive apex-anterior bowing. An external orthosis cannot heal or shrink the lesion, but it can share load across the diaphysis, limit the bending moment on the weakened cortex, and buy time until the lesion becomes quiescent at skeletal maturity.

When. Consider it for a symptomatic or radiographically extensive lesion with significant cortical thinning, for activity-related pain, or for a healing or healed pathological fracture while the child remobilises. It is an adjunct to observation, not a substitute for surveillance.

What. A well-moulded clamshell (thermoplastic) tibial orthosis or a patellar-tendon-bearing (PTB)-style ankle-foot orthosis offloads the diaphysis; a simple below-knee brace during high-risk activity is a lower-burden option. The choice is pragmatic, since there is no lesion-specific device and no high-level evidence favouring one design.

Limits. Bracing does not prevent the underlying deformity from progressing if the lesion is biologically active, and it does not replace the definitive answer to established bowing, which is corrective osteotomy near maturity. Prolonged rigid immobilisation risks stiffness and disuse osteopenia, so protection is graded to activity rather than continuous.

When a fracture occurs anyway. The great majority unite in a standard cast, and bracing is the step-down as the child returns to weight-bearing. Non-union (rare), not the brace, is what triggers surgery.

Clinical Pearl

Bracing in osteofibrous dysplasia is protective, not curative. If the examiner presses on "how does the brace change the natural history?", the honest answer is that surveillance and, ultimately, deformity-directed osteotomy - not the orthosis - determine the outcome.

Indications. Surgery is considered for:

1. Pathological fracture not healing with conservative treatment
2. Progressive deformity, over 20° of angulation
3. Diagnostic uncertainty, when adamantinoma cannot be excluded
4. Persistent pain affecting quality of life
5. Patient or family preference after counselling

Curettage and bone grafting. Through a cortical window the lesion is curetted thoroughly and grafted with autograft or allograft; it is used for small lesions and pathological fractures. Recurrence is 50-60%, highest in children under 10, and multiple recurrences are common, so it often requires repeat procedures and may ultimately require en bloc resection.

En bloc resection. Wide resection with intercalary reconstruction is the definitive treatment, with recurrence under 5%, but the morbidity is significant and reconstruction is challenging. It is reserved for concern about adamantinoma, or for multiple recurrences after failed curettage.

Corrective osteotomy. A closing wedge or dome osteotomy for established deformity over 20° once the lesion is inactive, preferably after skeletal maturity. Correction is good, but the osteotomy does not address the underlying lesion.

Send the tissue. Always send tissue for histology to exclude adamantinoma, remembering that it is excluded by the absence of epithelial islands, not by cytokeratin staining.

View larger

View larger

Pathological fracture usually follows minor trauma. About 90% heal with conservative management in a cast; non-union is rare (under 5%) and may require surgery, and refracture is possible if the lesion remains active.

Prognosis is excellent. OFD is a benign disease with the potential for spontaneous resolution, no metastatic potential and a normal life expectancy, and the vast majority have minimal functional impairment.

Long-term outcomes. By the age of 20, 70-80% of lesions are stable or resolved. Residual asymptomatic cortical thickening remains in 10-20%, and 5-10% require surgical intervention for complications.

Favourable prognostic indicators. Features that favour a good outcome:

• Classic radiology: well defined, with a sclerotic rim
• Early diagnosis, before the age of 10
• A single lesion rather than multicentric disease
• Observation accepted, with no rush to surgery
• Typical location in the anterior tibia
• Small size, under 5 cm
• Excellent bone quality surrounding the lesion

Global Epidemiology

• OFD is rare worldwide - well under 0.2% of primary bone tumours - with consistent demographics across published series from Europe, North America and Asia: tibial predominance, anterior cortex of the middle third, and presentation in childhood/adolescence (mean age 7-13 years depending on series).
• No geographic or ethnic predilection and no sex predilection are reported. The largest single dataset is the 101-lesion international series pooling UK and Australian tertiary centres (Dala-Ali, 2022).

Side-by-Side Guidance

There is no single dedicated society guideline for this rare lesion; practice derives from bone-tumour society principles and the major case series. The recommendations below are broadly concordant rather than conflicting.

Registry Notes

• OFD is too rare for dedicated registry survival data. National bone-tumour and sarcoma registries (e.g. the UK NSSG/regional bone-tumour services, European EURACAN reference network, Scandinavian sarcoma group databases) capture it mainly to flag and track the OFD-like adamantinoma it can mimic, where long-term recurrence (beyond 15 years, Houdek 2018) justifies registry surveillance.

High- vs Limited-Resource Practice Variation

• Well-resourced settings: ready access to MRI, immunohistochemistry (cytokeratin/EMA) and specialist sarcoma MDTs allows confident observation of classic lesions and early detection of atypical ones; deformity is managed with planned osteotomy and growth modulation.
• Limited-resource settings: plain radiography remains the mainstay; the key practice point is to AVOID intralesional curettage of a presumed benign lesion that could be an under-sampled adamantinoma, and to refer atypical, painful or progressive cortical tibial lesions to a centre with histopathology and limb-salvage capability rather than treating locally.

Why This Matters in Exams

Globally, examiners test the OFD-adamantinoma-fibrous dysplasia triad: location (cortical vs medullary), cytokeratin nuance, GNAS molecular distinction, the observation-first natural history, and the danger of mislabelling an atypical lesion.

Related pages: Adamantinoma for the low-grade malignancy at the other end of this spectrum — and note the discriminator is hyperchromatic epithelial islands, not cytokeratin, which 93% of OFD lesions also express; Fibrous Dysplasia for the lesion cytokeratin genuinely does separate OFD from, and for the GNAS/Gsα mutation that is present there and absent here; Congenital Pseudarthrosis of the Tibia for the other cause of anterior tibial bowing in a young child, which must not be curetted; and Describing a Bone Tumour Radiograph and Predicting Bone Tumour by Age for the systematic approach that reaches an intracortical anterior tibial diaphyseal lesion in the first decade.

• Is OFD a precursor to adamantinoma? Unresolved. Large paediatric series (Park, Dala-Ali) report NO transformation, arguing OFD is a self-limiting benign lesion. Pathology series (Hazelbag) document OFD-like adamantinomas recurring as classic adamantinoma, supporting a precursor or sampling-bias model. The pragmatic consensus: pure OFD does not need to be treated as pre-malignant, but an atypical or relapsing lesion warrants thorough sampling to exclude OFD-like adamantinoma.
• OFD vs OFD-like adamantinoma on a small biopsy. Scattered keratin-positive cells occur in both; the malignant label requires hyperchromatic epithelial islands. Limited biopsies under-sample, so radiology-pathology correlation and, where feasible, generous sampling are essential.
• Role and timing of biopsy. Classic radiology in a young child can be observed without biopsy. Biopsy is reserved for atypical age (over the second decade), a soft-tissue mass, aggressive radiology, or progression - precisely because needling a benign lesion risks recurrence and offers little when the picture is typical.
• What to operate on. Modern thinking (Dala-Ali) shifts the surgical target from the lesion to the deformity: intralesional curettage has high recurrence and rarely changes natural history, while wide excision carries substantial morbidity. Corrective osteotomy for established angular deformity, ideally near maturity, is increasingly favoured over lesion-directed surgery.
• Surveillance metric. Anterior angular deformity on full-length tibial radiographs (not lesion size alone) is the most useful parameter to follow, with bowing under 10 degrees predicting successful conservative management.